Economic Implications of Plant-made Pharmaceutical Production in North Carolina

Economic Implications of Plant-made Pharmaceutical Production in North Carolina

Christopher F. Dumas

Associate Professor, University of North Carolina Wilmington

Troy G. Schmitz

Associate Professor, Arizona State University

Christopher R. Giese

Graduate Research Assistant, Arizona State University

Michael Sligh

Rural Advancement Foundation International – USA

The Rural Advancement Foundation International - USA cultivates markets,

policies and communities that support thriving, socially just and

environmentally sound family farms.

While focusing on North Carolina and the southeastern United States, we also

work nationally and internationally. RAFI is creating a movement among farm,

environmental and consumer groups to ensure that:

• family farmers have the power to earn a fair and dependable income;

• everyone who labors in agriculture is respected, protected, and valued by

society;

• air, water and soil are preserved for future generations;

• the land yields healthy and abundant food and fiber that is accessible to

all members of society;

• the full diversity of seeds and breeds, the building blocks of agriculture, are reinvigorated and publicly protected.

2008 RAFI-USA. All rights reserved.

Rural Advancement Foundation International - USA

PO Box 640

Pittsboro, NC 27312

www.rafiusa.org

919-542-1396

i

Preface

For more than 10,000 years, farmers have worked with the environment to create new plants, fiber, and food to sustain life all over the earth. As we lose farmers, we lose diversity. As we lose diversity, we lose farmers. The social, economic, and technological changes converging on our rural communities are rapidly changing how food is produced and what comes to our tables. RAFI-USA believes that farmers and consumers must be informed, involved with each other, and active in protecting and directing the use of natural and human agricultural resources. RAFI-USA approaches all agricultural policy, practice and technology options with the same basic questions:

o Who will benefit? o Who will be harmed? o Who will pay, if something goes wrong? o Who will decide?

These are fundamental questions and deserve our attention. In the best cases, these questions should be answered prior to adoption of new agricultural initiatives, and should be addressed in a fully open and transparent process – especially those initiatives which can have profound and/or long-term impacts. RAFI-USA also uses the “triple-bottom-line” assessment when evaluating new agricultural initiatives:

o Is it economically viable for the farmers? Will they receive a fair price and reasonable return on their investment?

o Is it environmentally sound? What are the risks to the environment, local communities, biodiversity and the ecosystem?

o Is it socially just? Do farmers, workers and others participating in this initiative have full rights and ownership of the technology? Are the contracts fair? Are the farmers in control of the management decisions of this initiative?

These two sets of tools, benefit assessment and the “triple-bottom line” analysis, guide our evaluations of any potential new agricultural initiatives. It is in this spirit that we have commissioned this report. We hope our recommendations can help shape a full and meaningful dialogue regarding the future of pharmaceutical crops in North Carolina agriculture, and the real opportunities to achieve the “triple-bottom” line. Michael Sligh January 2008

ii

Executive Summary

Over the last twenty years, agriculture has seen the introduction and rapid deployment of genetically modified (GM), or “transgenic,” crops for food (i.e., corn and soybeans) and fiber (i.e., cotton). Plant-made pharmaceuticals (PMPs) are a class of GM crop not intended for use as food or feed. Rather, PMPs are intended for use as therapeutic drugs for humans or livestock, or as materials for research and industry (e.g., cell culture media). PMP plants are used as factories to produce the PMP product, the product is extracted from the plant, and the plant remains are discarded. Scientists and industry groups typically cite two reasons for pursuing PMP production methods. First, lower cost: production of high-quality pharmaceutical components (proteins and antibodies) is presently done using cell cultures inside bioreactors, which is very costly (US$105-175 per gram) and limits the size of the consumer market. Second, growing demand: by the end of the decade, there could be more than 80 antibody-dependent products with an estimated value of US$20-90 billion, provided adequate production capacity can be developed. Proponents of PMP crops claim that PMP production will increase the range of available drug products, reduce the time required to bring new drugs to market, lower the cost of drug production, and provide additional markets for farmers. Opponents of GM and PMP crops cite potential food safety risks from cross-contamination of food crops, consumer skepticism of genetically engineered products, potential environmental hazards, and past regulatory mistakes as reasons for their opposition.

The regulatory history of PMPs grown outdoors as field crops is not encouraging. Although PMPs have been grown by several companies in experimental field trials regulated by the U.S. Department of Agriculture (USDA) since the early 1990s, none has been grown in commercial quantities (although one just received a permit to grow at commercial scale in 2007), and no PMP drug products have as yet been approved by the U.S. Food and Drug Administration (FDA). (Some PMPs are being sold in small quantities for use as research materials.) Escape of PMP plants from USDA-regulated field trials has been followed by regulatory reform at USDA, but PMP plants have continued to escape from field trials following the reform effort. If PMP plants escape from their designated areas and become mixed with plants that are intended for use as food, and the mixture enters the food supply, large disruptions of the food industry can occur.

This report will review information on the potential economic benefits, environmental impacts, and externalized costs of GM crops in general, and PMP crops in particular, for North Carolina. Special attention will be devoted to PMP rice developed by Ventria Bioscience. Ventria’s PMP rice is currently undergoing field trials in North Carolina. At present, Ventria’s PMP rice is the only field-grown PMP crop in the state. As of 2007, Ventria’s

iii

three PMP rice products, the pharmaceuticals lactoferrin, lysozyme, and serum albumin have not been approved by the FDA for drug, food, or animal feed uses. The products have been marketed as research and bioprocessing materials, but it is not clear that Ventria has received substantial revenues from these uses. Ventria plans to market the products as anti-diarrheal additives for infant oral rehydration solutions and as nutritional supplements in yogurt, granola bars, performance drinks and other products. Ventria has also mentioned adding lysozyme to animal feed as a substitute for antibiotics. Ventria claims a potential market for these products of more than $2 billion annually. The company’s estimates of potential profitability and economic impacts should be considered with caution. Even if eventually approved by FDA, Ventria’s products may not be profitable as anti-diarrheal additives for infant formulas marketed in developing countries without subsidies, and the profitability of these products in sports drinks, granola bars, etc., is speculative.

Although Ventria is conducting field trials in North Carolina, it currently plans to grow and process PMP rice at commercial scale in Kansas. Ventria projects 30,000 acres of PMP rice production per year in Kansas upon full scale commercialization. Assuming this speculative acreage forecast is correct, with an average farm size of approximately 700 acres in Kansas, perhaps 43 farmers would benefit from PMP rice production. At Ventria’s estimate of $150 to $600 in additional returns per acre relative to corn, PMP rice may bring Kansas farmers an additional $4.5 to $18 million per year. Perhaps 50 more people would be employed in Ventria’s proposed PMP rice processing facility in Kansas. Using typical economic multiplier numbers, perhaps 150 additional jobs would be supported in Kansas due to economic multiplier effects. Including economic multiplier effects, Ventria estimates that $45 million annually in economic impact would be generated by PMP rice production activities in Kansas. For comparison, in 2006, Kansas agriculture produced over $11 billion in crop, animal, and related agricultural output, with a total economic impact of $28 billion. Ventria’s estimated economic impact of $45 million per year is small relative to the scale of Kansas agriculture.

For those farmers considering PMP crop production, several factors should be considered in addition to potentially higher revenues per acre. Ventria is implementing the field trials using independent grower contracts. At this early stage, Ventria covers all costs for the North Carolina farmers growing PMPs on subcontract. In the future, independent growers will be expected to provide a seed-to-harvest package deal for the firm’s PMP production. This will involve significant investment in PMP-specific training and dedicated farm equipment. The USDA requires each PMP grower to have dedicated land area, dedicated equipment for planting and harvesting, and separate areas for cleaning PMP equipment and processing PMP crops. Employee training is also required as part of compliance with new FDA and USDA regulatory statues for molecular farming. This raises the possibility that molecular farming contracting for field-grown PMP crops will require

iv

such costly investments in infrastructure and compliance that only the largest, wealthiest growers would be able to participate and profit. Furthermore, use of PMP crops by some farmers may impose “spillover” costs on other farmers who do not grow PMP crops. Farmers who do not grow PMP crops may have to spend money to certify that their crops are “PMP-free” if grown in the same region as PMP crops. This is an especially important issue for organic farmers.

In addition to the potential costs of PMP production to the farm sector, there may also be environmental costs if field-grown PMP products have a detrimental effect on fish, wildlife, insects (e.g., bees), or wild plants. While much work has been done on the environmental impacts of GM plants used for food, relatively little work has been done on the potential environmental impacts of PMP plants. At this point, the most that can be said is that the potential environmental impacts of PMP field crop production are unknown. For PMP products grown using familiar field crops, the environmental impacts may be small, assuming that the PMP product itself within the plant is not harmful, but again, information is very incomplete and no firm conclusions can be drawn. Ongoing work in bioconfinement methods may reduce the environmental risk of PMP plants.

Detrimental human health effects are another potential cost of PMP production. While detrimental human health effects of products intended for pharmaceutical use are certainly possible, these products would need approval by FDA for use as drugs or food, and any non-accidental effects would likely be small, assuming conscientious review by FDA. In contrast, the issue of accidental, detrimental human health effects looms large in the PMP debate. If PMP products not intended for use as food somehow enter the food supply and become ingested by humans, the effects could be significant, as these products may not have undergone food safety testing by FDA. Again, the brief history of PMP crop field trials indicates that it is very difficult to prevent co-mingling of PMP and non-PMP crops, implying that the potential for accidental contamination of the food supply is an important issue.

Neither food plants nor farmers’ fields are necessary for the production of PMPs. PMPs can be grown using non-food plants in contained systems instead of agricultural fields. Some alternative PMP containment systems utilizing non-food plants include duckweed (Lemna spp.), tobacco (Nicotiana spp.), algae (Chlamydomonas reinhardtii) and moss (Physcomitrella patens), and fungi (Aspergillus niger). Yet another option is to produce PMPs using food crops grown inside greenhouses, such as potatoes grown hydroponically. Advantages of growing PMPs in containment systems include better uniformity of product, lack of residues from herbicides, pesticides and fungicides, and greatly reduced risk of contaminating the food supply. Two disadvantages of producing proteins in containment systems are that it is thought to cost more, and it is thought to take longer to bring the product to market. However, recent advances in closed-system

v

technology have eliminated some of the cost difference between field grown PMPs and contained systems. In North Carolina, the availability of highly-skilled biotech labor and innovations in the use of contained production systems that attain high product purity are catalyzing market expansion of contained PMP production. Contained PMP production currently co-exists with profitable organic and local food suppliers in the state.

At the present time, PMP production via food crops in the field should not be considered a cornerstone of future agricultural policy or rural economic development policy in North Carolina or elsewhere in the United States. Given past difficulties in securing FDA approval for PMP products, the benefits of PMP production are too speculative. Furthermore, given past difficulties in preventing the escape of PMP products in the field, the risks and potential costs of future containment loss events are too great.

Contents

1 Introduction ................................................................................. 1

2 GM and PMP Regulation............................................................... 2

2.1 Regulatory Framework and Experience .................................. 2

2.2 Ventria Bioscience -- Regulatory History .............................. 14

3 Potential Benefits of PMPs .......................................................... 20

3.1 Overview .............................................................................. 20

3.2 The Case of Ventria Bioscience............................................. 22

4 Potential Costs of PMPs .............................................................. 29

4.1 Farm Costs and Potential Grower Profitability ...................... 29

4.2 Government Subsidies to Biotech and PMP .......................... 38

4.3 PMP Health Risks in Intended Uses...................................... 42

4.4 Containment Loss and Potential PMP Liability Costs ............ 44

4.4.1 Consumer Reaction to GM and PMP Products in Food ...........................................................................................................45

4.4.2 Food Market Reaction to GM/PMP Containment Loss 48

4.4.3 NRC Recommendations to Reduce GM/PMP Liability Costs ...........................................................................................................54

4.5 Externality “Spillover” Costs Affecting Non-GM and Organic Farmers............................................................................... 56

4.6 Externality “Spillover” Costs of Environmental Hazards........ 58

5 Alternatives to Food Crop PMPs.................................................. 66

6 Conclusions ............................................................................... 67

7 RAFI Recommendations ............................................................. 74

8 References.................................................................................. 77

9 Tables ........................................................................................ 89

1

1 Introduction

Over the last twenty years, agriculture has seen the introduction and rapid deployment of genetically modified (GM), or “transgenic,” crops. While crop changes produced by traditional breeding technologies such as hybrid corn and Green Revolution rice and wheat have had critics, opposition to the production of GM crops has developed more quickly and publicly. A new type of GM crop, plant-made pharmaceuticals (PMPs), has been undergoing field trials and is on the verge of commercial-scale production. PMPs are therapeutic drugs or medical products produced inside genetically modified plants. The debate concerning GM and PMP crops involves three primary issues: the benefits and costs of the technology and its products, regulatory measures to preserve human and environmental safety, and the appropriate legal framework to encourage innovation, promote competition, and preserve intellectual property (Nelson 2001). The potential benefits of GM crops include higher crop yields, enhanced nutritional characteristics, and reduced production costs through lower pesticide or fertilizer requirements. Proponents of PMP crops also claim that PMP production will increase the range of available drug products, reduce the time required to bring new drugs to market, lower the cost of drug production, and provide additional markets for farmers (BIO 2002a, 2002b, 2006). Critics of GM and PMP crops cite potential food safety risks from cross-contamination of food crops, consumer skepticism of genetically engineered products, potential environmental hazards, past regulatory mistakes, and increasing corporate control of agriculture as reasons for their opposition (e.g., Freese 2007, 2002).

The U.S. Department of Agriculture (USDA) began regulating GM crops in 1986 (USDA 2005a). Since that time, USDA has approved over 10,600 applications for field-testing GM crops at more than 49,300 sites. Although GM crops have been in use commercially since China introduced virus-resistant tobacco and soybeans in the early 1990s, the first commercial use of GM crops in Western countries was the Flavr Savr tomato, a delayed-ripening tomato introduced by Calgene in the US in 1994 (Nelson 2001). The global volume of GM crop production expanded rapidly over the next ten years. While most GM crops are grown in North America, large quantities are also produced in Argentina, Mexico, and South Africa. GM crops in widespread use include corn, soybeans, cotton, potatoes and canola. From the beginning of commercialization in 1994, the global area planted in GM crops grew at an annual rate of 13%, reaching 102 million hectares (252

2

million acres) by 2006 (ISAAA 2006). Soybeans, corn (maize), and cotton are the leading GM crops in terms of acreage. In 2006, an estimated 10.3 million farmers worldwide grew GM crops in 22 countries (Table 1). The United States is the world leader in GM crop area, with 54.6 million hectares under cultivation, while Spain is the leading European producer with 60 thousand hectares. The eleven developing countries planting GM crops account for forty percent of GM crop acreage, a percentage that has been increasing steadily.

Considerable controversy surrounds the use and adoption of GM crops. Some consumer advocacy groups believe that genetically engineered foods hold health and environmental dangers. Anti-biotech activists have labeled GM products as “Franken Foods” and have raised long-term health concerns regarding the consumption of GM products. Several of these groups have been successful in launching anti-GM campaigns that have influenced the rate of GM adoption. Some examples include: (1) the decisions of the United States and Canada to forego the adoption of GM wheat varieties; (2) the decision of Aventis to terminate the production of Starlink corn (see p. 38 below); and (3) the decisions of California and Missouri to ban the production of certain pharmaceutical rice varieties. In the United States, the rate of adoption of GM crop varieties slowed considerably during the early 2000s. The United States Food and Drug Administration (FDA) approved on average 9.4 GM-food varieties a year between 1995 and 1999. This approval rate fell to 3.0 GM-food varieties a year between 2000 and 2004 (Weise, 2005). Similarly, the United States Department of Agriculture (USDA) approved on average 8.2 GM-crop varieties per year from 1994 to 1999, but only 2.6 GM-crop varieties per year from 2000 to 2004.

The regulatory situation affecting GM crop production has changed over time, with some observers complaining that regulations have become too burdensome, stifling innovation and application of new technologies, while others claim that regulators are too lenient and allow too much risk. All agree that the regulatory process is complex and varies greatly from country to country, complicating trade. The approval procedures and labeling regulations covering GM foods differ among countries. In general, biotech regulations are less stringent in the United States (US) than in the European Union (EU), which in part explains why GM products are more widespread in the US. In fact, several international biotech and pharmaceutical companies based in the EU conduct field trials in the United States, because their products have not been approved for production in the EU (Moss et al.

3

2006). Finally, the legal framework that protects intellectual property embodied in GM crops affects the ownership and control of new GM projects and the distribution of associated profits.

It is within this industry and policy setting that we consider the economic and environmental implications of a specific type of GM crop, plant-made pharmaceuticals, or PMPs, for North Carolina farmers (Table 2). PMPs are pharmaceutical products produced and extracted from genetically modified plants; the plant is used as a factory to produce the PMP product, the product is extracted, and the plant remains are discarded. PMP plants can be grown inside laboratories or greenhouses, or they can be grown outside in fields like agricultural crops. The goal of both traditional plant breeding and new GM technologies like PMPs is to identify desirable genetic traits and combine them in a crop variety that can be grown profitably. Desirable traits are divided into two classes—agronomic characteristics that reduce the costs of cultivation, and product characteristics that increase value to consumers and the price consumers are willing to pay. PMPs are a type of GM crop that offers a new product characteristic—the ability to produce pharmaceutical products. PMPs are referred to as “Generation 3” GM plants. Generation 1 GM plants featured genetic modifications that reduced the costs of crop production by reducing the need for pesticides, making the plant more drought-tolerant, etc. RoundUp Ready™ soybeans are an example of a Generation 1 GM plant. Generation 2 GM plants improved the nutritional qualities of food plants. For example, “Golden Rice” has enhanced levels of vitamin A. Generation 3 GM plants, PMP plants, differ from Generation 1 and 2 crops in that PMPs are not intended for use as food or animal feed. Instead, the pharmaceutical product is extracted from the PMP plant, and the plant is discarded. However, either food plants (e.g., corn or rice) or non-food plants (e.g., tobacco or algae) can be used to produce PMP products. When food plants are used, the plants are discarded after the PMP product is extracted. However, when food plants are used, there is a risk that PMP plants may become mixed with non-PMP plants grown for food or feed during the planting, pollination, harvesting, transportation, storage, or processing phases of production.

PMP field trials began in the US (using corn/maize) in 1992, peaked in 1998, and declined beginning in 2001 (Smyth et al. 2004). While corn accounts for 47 percent of all PMP field trial permits, since 2001 there has been increasing interest in PMP safflower, rice and especially tobacco (Freese and Caplan 2006). Field trials in Canada (using canola) also peaked in 1996-1998 and

4

declined following 2000 but are climbing again based on safflower. Other countries report a small number of PMP field trials between 1995 and 2002. To date, PMP crops have not been grown in the field on a commercial scale in the United States, and no PMP products have been approved by the U.S. Food and Drug Administration (Freese 2007). However, firms currently engaged in field trials will presumably wish to grow successful products at commercial scale in the future.

PMP crops and products present new opportunities and risks for North Carolina farmers. Because PMP agriculture is in its infancy, relatively little information is available on PMP crops and products. This report will review information on the potential economic benefits, environmental impacts, and externalized costs of GM crops in general, and PMP crops in particular, for North Carolina. Special attention will be devoted to PMP rice developed by Ventria Bioscience. Ventria’s PMP rice is currently undergoing field trials in North Carolina. At present, Ventria’s PMP rice is the only field-grown PMP crop in the state. Ventria has PMP rice field trials in Missouri as well and grows PMP rice experimentally in greenhouses in California (Sacramento Bee 2006). In May 2007, Ventria received approval from the USDA to plant up to 3,200 acres of PMP rice in Kansas and has begun work on a PMP rice processing facility in Junction City, Kansas (Ventria Bioscience 2006). It appears that any PMP rice grown in North Carolina will be transported to Kansas for processing. Because Ventria’s PMP rice is the first PMP crop to be grown in the field uncontained at commercial scale in the North Carolina, decisions concerning its production, processing, transportation, marketing and regulation are potentially precedent-setting.

2

2 GM and PMP Regulation

2.1 Regulatory Framework and Experience

A fundamental lesson of economic theory and practical experience is that the “invisible hand” of private markets cannot be relied upon to correct externality “spillover” costs precisely because the financial incentives that drive the invisible hand are distorted. In such situations, society often turns to government action to coordinate and regulate private market actions for the public good. Given the potential externality “spillover” costs associated with GM and PMP crops (see sections 3.5 and 3.6), society has chosen to regulate them. Industries often request government regulation to prevent “bad apple” firms from ruining industry reputations and alienating consumers. For example, the Biotechnology Industry Organization, the leading GM and PMP industry trade association, supports “strong regulatory oversight for all products of crop biotechnology” (BIO 2007).

The basic institutional structure for regulating all biotechnology products in the United States is the “Coordinated Framework for Regulation of Biotechnology” established in 1986 (see, e.g., Pew Initiative 2004). In general, this framework involves three federal agencies: the USDA’s Animal and Plant Health Inspection Service (APHIS), which regulates the importation, interstate movement, and field testing of GM plants; the FDA, which regulates food and feed additives, human drugs, and medical devices; and the Environmental Protection Agency, which regulates the use of all pesticides, including those expressed in GM plants.

Because USDA/APHIS is authorized to regulate potential plant pests under the Federal Plant Protection Act, and since all GM plants have the potential to be plant pests, all GM plants are considered “regulated articles” by USDA/APHIS. Use of such articles outside a contained facility (e.g., in a field test) requires authorization from USDA/APHIS through either a “notification” procedure or a permit procedure. In 1993, the USDA promulgated new regulations governing field tests of genetically engineered plants, removing permit requirements for most GM plants but retaining them for PMPs. GM plants that do not require a permit are authorized through the notification process.

3

Under the notification process, GM plants (but not PMP plants or PMIP, plant-made industrial proteins, plants) can be grown in field trials with simple notification of the USDA. For GM plants intended for use as food or feed, the GM plant developer also initiates a “consultation” with the FDA, during which the plant typically undergoes a voluntary food safety review. For GM plants modified to have pesticidal properties, the EPA requires an additional experimental use permit under the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA). Upon successful completion of the field trials, GM plant developers can apply for deregulated status from USDA/APHIS. If deregulated status is granted, a GM crop can then be freely commercialized with no further oversight by USDA/APHIS, and this is in fact the route that has been used for all the major commercial GM crops currently on the market. If the plant has pesticidal properties, it must still register with EPA prior to marketing.

Since 1993, PMP field trials have been regulated under the USDA’s permit procedure rather than the notification procedure. In theory, the permit procedure was supposed to be stricter than the notification procedure. However, A National Research Council report (2002) on the environmental effects of transgenic (GM) plants found that “the only practical trigger used by APHIS [was] the presence of a previously identified plant pest or genes from a plant pest in the transformed plant. Other operational triggers are needed for transgenic plants that may have associated risks but lack the above characteristics.” The NRC report also found that APHIS assessments of potential environmental effects of transgenic plants are largely based on environmental effects considered at small spatial scales. Potential effects from “scale-up” associated with commercialization are rarely considered. The report recommended that post-commercialization validation testing be used to assess the adequacy of pre-commercialization environmental testing and that this testing should be conducted at spatial scales appropriate for evaluating environmental changes in both agricultural and adjacent, unmanaged ecosystems. The NRC report also found that the APHIS process should be made significantly more transparent and rigorous by enhanced scientific peer review, solicitation of public input, and development of determination documents with more explicit presentation of data, methods, analyses, and interpretations. In the committee’s review of public participation in the review process it was apparent that the number of comments on Federal Register notices had declined almost to zero. Committee discussions with representatives of public interest groups indicated that this decline in responses to

4

APHIS Federal Register notices was at least in part due to a perception that APHIS was only superficially responsive to comments. The committee found that there was a need for APHIS to actively involve more groups of interested and affected parties in the risk analysis process while maintaining a scientific basis for decisions. Furthermore, the NRC committee found that the extent of “confidential business information” in registrant documents sent to APHIS hampered external review and transparency of the decision-making process.

In addition to the 2002 NRC report, several incidents in 2002 involving PMP crop contamination of food products caused USDA to reevaluate its PMP permitting process. In September 2002, ProdiGene, Inc. was ordered by USDA to burn 155 acres of food crop corn in Iowa to ensure that it was not pollinated by a nearby field of ProdiGene’s PMP corn (New York Times 2002). In November 2002, ProdiGene was fined US$250,000 in a second incident for allowing experimental PMP corn grown in Aurora, Nebraska, in the preceding year to contaminate a soybean crop grown in the same field in 2002. The contamination was discovered by USDA APHIS inspectors, but only after the soybeans had been harvested and stored with other soybeans in a commercial grain silo, contaminating 500,000 bushels of soybeans. ProdiGene bought the contaminated soybeans and had them destroyed at a cost of US$3.5 million. ProdiGene was also forced to post a $1 million bond to cover potential damages from any future contamination episode. The US government made an interest-free loan to ProdiGene, because the small biotech company had insufficient funds to pay (Washington Post 2003). This can create an incentive problem for the bio-pharma industry as a whole, as the small firms typical of the industry would not have the funds to pay such fines. The problem is that if firms know that the government will provide loans or loan guarantees to pay fines resulting from regulatory violations, then firms do not have the financial incentive to maintain containment of pharmaceutical crops (Smyth et al. 2004).

In mid-December 2002, Dow AgroSciences was fined for failing to meet permit conditions to prevent gene transfer from an experimental transgenic maize variety undergoing field trials at Molokai, Hawaii (Smyth et al. 2004). That same month, Pioneer Hi-Bred was fined for planting experimental transgenic maize in an unapproved location that was too close to other experimental maize plantings in Kauai, Hawaii. In April 2003, Dow was again fined for violating an EPA permit, this time in Kauai. The fine resulted from the detection of 12 transgenic maize plants that

5

contained an unapproved gene that is suspected of coming from the pollen of another experimental plot located nearby. Although Dow officials discovered the plants, Dow failed to notify EPA promptly, and EPA officials expressed disappointment over the delay.

In 2003, on the heels of the regulatory violations occurring between 2001 and 2003, including the high-profile 2002 Prodigene incidents, USDA permit regulation of PMP field trials was strengthened (USDA 2006A). Crop-specific measures were stipulated to ensure containment, including isolation distance of test plots (for maize, for example, the distance is one mile, eight times the distance required for the production of foundation seeds), planting of buffer borders of non-GM crops was mandated, and perimeter fallow zones were required. In addition, the use of dedicated equipment was mandated, there were post-harvest restrictions on land use, and APHIS was to perform a specified number of inspections during the field test growing season.

Also in 2003, the USDA introduced a new category of regulated products, “value added protein for human consumption.” As of October 2006 (UCS 2006b), the only two compounds classified as value added proteins are lactoferrin and lysozyme, two of the products grown by Ventria Biosciences in North Carolina. Significantly, the USDA allows value added proteins to be regulated under the notification process rather than requiring permits. However, Ventria voluntarily submitted requests for permits to grow its PMP crops.

USDA oversight of PMP crop field trials under the notification / voluntary permit process depends to a great extent on company reports filed with the USDA at the end of the field trial, or annually for multi-year permits. Such reports are required to include any adverse impacts of the experimental crop. Batie and Ervin (2001) point out that because firms receive no financial benefit from discovering adverse impacts, they have little incentive to investigate them. Freese et al. (2004) goes further and suggests that a clear conflict of interest exists. Because self-reporting of adverse impacts to the USDA could entail revocation or non-renewal of the permit, and thus loss of profits, the company’s duty to report such adverse effects is clearly in conflict with its financial interest. Dalton (2002) reports that Pioneer Hi-Bred and Dow AgroEvo denied access to proprietary materials required by independent scientists to conduct biosafety analysis of Bt sunflower after the firms initially cooperated with scientists and

6

the scientists’ preliminary findings indicated potential biosafety risks.

In 2005, the USDA APHIS regulatory program was criticized by its own Inspector General audit for failing to properly regulate and track GM and PMP crop field tests, even after the USDA strengthened regulations in 2003 (USDA 2005a). The audit found: “To evaluate the Animal and Plant Health Inspection Service’s (APHIS) controls over releases and movements of regulated genetically engineered plants, we visited 91 field test sites in 22 States that were either planted or harvested. We inspected the sites for compliance with APHIS’ requirements for the growing or post-harvest season. We found that APHIS, the USDA agency that oversees biotechnology regulatory functions for the Department, needs to strengthen its accountability for field tests of genetically engineered crops. In fact, at various stages of the field test process—from approval of applications to inspection of fields—weaknesses in APHIS regulations and internal management controls increase the risk that regulated genetically engineered organisms will inadvertently persist in the environment before they are deemed safe to grow without regulation.”

In particular, the 2005 USDA audit of APHIS found:

(1) The precise locations of all genetically engineered field test sites planted in the United States are not always known. After authorizing field tests, APHIS does not follow up with all permit and notification holders to find out exactly where the fields have been planted or if they have been planted at all.

(2) Before approving field tests, APHIS does not review notification applicants’ containment protocols, which describe how the applicant plans to contain the genetically engineered crop within the field test site and prevent it from persisting in the environment.

(3) At the conclusion of the field test, APHIS does not require permit holders to report on the final disposition of genetically engineered pharmaceutical and industrial harvests, which are modified for nonfood purposes and may pose a threat to the food supply if unintentionally released. As a result, we found that two large harvests of genetically engineered pharmaceutical crops remained in storage at the field test sites for over a year without APHIS’ knowledge or approval of the storage facility.

(4) APHIS does not specify when genetically engineered crops must be destroyed, or “devitalized,” following the field test. Approved applicants sometimes allow harvested crops to lie in

7

the field test site for months at a time, their seeds exposed to animals and the elements. Also, because APHIS has not specifically addressed the need to physically restrict edible genetically engineered crops from public access, we found a regulated edible genetically engineered crop, which had not gone through the Food and Drug Administration’s regulatory process for approval for human consumption, growing where they could easily be taken and eaten by passersby.

(5) Field inspectors “did not inspect all pharmaceutical and industrial field test sites five times during the 2003 growing season, as APHIS has announced to the public. APHIS has also stated publicly that pharmaceutical and industrial field test sites would be inspected twice during the postharvest period, or the year following the end of the field test, during which the field must be monitored for regrowth of the genetically engineered crop. In one case, a violation at a pharmaceutical field test site in our sample went undetected because PPQ [APHIS Plant Protection and Quarantine] did not perform the required inspections at that site during the 2003 postharvest monitoring period” (USDA 2005a).

Despite USDA’s assurances that it would address the issues raised in the 2005 audit, new containment breach incidents in 2006 raised questions about the ability of even USDA’s new, strengthened regulations to contain GM crops. Twice in 2006, current regulations did not prevent GM rice from contaminating non-GM commercial rice supplies, halting exports of US rice to some countries and causing substantial economic losses for US rice farmers (Washington Post 2006, Bennett 2007). In January 2006, GM Liberty Link (LL601) rice (not approved for human consumption) was found in rice processed by Riceland Foods in Stuttgart, Arkansas (Fortune 2007). Arkansas produces about 45 percent of U.S. rice, and Stuttgart is home to America’s two largest rice mills. The rice was then found in commercial rice supplies in Texas, Louisiana, Mississippi and Missouri, as well. The Liberty Link rice may have come from a rice research station in Crowley, LA, operated by Louisiana State University. Although Bayer CropScience had dropped plans to produce LL601 in 2001 and did not pursue USDA approval for commercial production, the rice had been grown in several test locations, including Louisiana State University’s rice research station near Crowley, LA, from 1999 to 2001 (Washington Post 2006). It was later determined that at least one variety of rice (Cheniere) grown at the research station was

8

contaminated with LL601 since at least 2003, even though the closest Cheniere plot was 160 feet from the LL601 plot (16 times the then current USDA standard). It is unknown whether the grains from the two plots were mixed before or after cultivation, or whether the LL601 plants fertilized some of the Cheniere plants. However, it was not until July 31, 2006, that Bayer CropScience notified USDA and the U.S. Food and Drug Administration that the company had detected trace amounts of regulated LL601 in commercial long-grain rice (USDA 2007b). On August 18, 2006, Bayer CropScience applied to USDA for deregulation of LL601, the same day that USDA announced the LL601 contamination (Washington Post 2006). The Center for Food Safety claimed that this was merely an effort by Bayer CropScience to avoid legal liability, as Bayer CropScience had no intention of bringing the LL601 rice to market. In November 2006, APHIS announced that 2003 Cheniere variety was the only foundation seed that tested positive for regulated genetically engineered LL601, and farmers were advised not to plant it. APHIS also announced that a sample of the 2003 Cheniere variety indicated the presence of trace levels of unregulated LL62. LL62, LL06 and LL601 are rice varieties engineered by Bayer CropScience to be tolerant to herbicides marketed under the brand name LibertyLink. APHIS had deregulated LL62 and LL06 in 1999. On November 24, 2006, USDA-APHIS retroactively deregulated Liberty Link LL601 rice, declaring it safe for human consumption. Later tests found contamination by two additional strains of unapproved Liberty Link rice in another type of foundation seed rice, Clearfield 131, which farmers were also advised not to plant. Table 3 provides the USA Rice Federation’s estimates of the impacts of the LL601 rice incident on U.S. rice export markets. Many importing nations increased testing, labeling and certification requirements, and some stopped U.S. rice imports altogether. It is estimated that 63 percent of U.S. rice exports were affected.

In 2006, the USDA consolidated its regulations and policies into a single document: “Draft Guidance for APHIS Permits for Field Testing or Movement of Organisms with Pharmaceutical or Industrial Intent” (USDA 2006b). Under the 2006 consolidated regulations, PMP crops are defined as those genetically engineered crops produced with pharmaceutical intent. Under the PMP permit process, PMP developers must submit detailed explanations of the genetic engineering process, the purpose and design of the proposed production, and the methods to be used to ensure confinement. Upon approval, the USDA issues a permit specifying conditions that must be met before, during and after production.

9

The conditions include: separating of PMP crops from crops intended for food or feed, cleaning production equipment, allowing government inspection of the site, and post-harvest monitoring and land use restrictions. In contrast to GM products intended for use as food or feed, under the permit process PMP crops are not deregulated at the end of field trials; instead, PMP crops remain regulated under permit.

The FDA has authority to regulate the manufacture of pharmaceuticals under the Federal Food, Drug, and Cosmetic Act (FFDCA) but has decided to rely on the USDA to oversee PMP crop production (FDA 2002). An exception is the category of “indirect food additives,” which includes substances that become components of food indirectly. The PMPs in PMP crops would be considered indirect food additives unless classified by FDA as “Generally Regarded as Safe,” or “GRAS.” Substances can be classified as GRAS if (1) they were in food prior to 1958 and were safe, or (2) they are generally recognized, among qualified experts, as having been shown to be safe food additives through scientific procedures. Since most PMPs are not intended for use as food, most do not have scientific evidence for their safety, and hence, would not be considered GRAS, and, therefore, would be regulated by FDA as indirect food additives. As food additives, the developers would have to submit documentation to the FDA demonstrating that the products are safe in food. Without FDA approval, such non-GRAS food additive products would be considered “adulterated,” could not legally participate in interstate commerce, and would typically trigger recall actions. As of October 2006, the FDA had not indicated whether it planned to classify PMPs as indirect food additives (UCS 2006b). However, the FDA (2002) has said that the presence of PMP materials in food could render it adulterated under the FFDCA. This effectively establishes a “zero tolerance” level for PMPs and PMIPs in food or feed products. Meeting a zero-tolerance level is difficult and essentially impossible to achieve with absolute certainty. This is a conundrum, but one that exists under current regulations in the United States as well as abroad. Because it is widely accepted that 100% purity is not attainable, a zero-tolerance standard raises the question of what should happen in those (inevitable) events when it is violated. Costly recalls of adulterated food may be necessary, firms may be exposed to consumer and public backlash, and liability issues would inevitably arise (Moschini 2006). While some have called for relaxing the zero-tolerance policy for PMP contaminants in the food supply and would instead allow some small, positive tolerance levels, presumably to minimize the

10

financial liability of small loss of containment events, the USDA and FDA have maintained the zero-tolerance standard (Freese and Caplan 2006). The food industry has opposed relaxing the zero-tolerance standard, fearful of consumer and export market rejection of food if even low levels of PMPs appear in the food supply (National Food Products Association 2003). Perhaps it is not surprising that the Grocery Manufactures of America and the National Food Processors Association have taken positions against the use of food/feed crops for pharmaceuticals (USA Today 2006, Freese and Caplan 2006). In 2003, the former CEO of Kraft Foods singled out the issue of PMP contamination of foods as a threat to her company and the food industry as a whole (Chicago Sun Times 2003).

As of late 2006, the USDA-APHIS had never denied a petition for a new GM crop, although about a third of all petitions are withdrawn when APHIS challenges company claims on petition supporting documentation (National Public Radio 2006).

On February 28, 2007, the USDA announced yet another incident involving loss of containment--rice seed in Arkansas were contaminated with GM rice variety LL62. In March 2007, the USA Rice Federation (2007a) expressed doubt that current USDA regulations can prevent GM contamination of the U.S. non-GM commercial rice supply: “The USA Rice Federation supports the USDA action in March 2007 to prevent the planting and distribution of Clearfield 131 (CL131) rice seed that could contain trace levels of genetic material unapproved for commercialization. . . . By the same token, we are increasingly frustrated with the apparent lack of ability on the part of private companies and federal regulators to control research and maintain accountability of the resulting products. The current approach to research, development and management in the biotechnology industry must be replaced with more conservative methodologies. . . . The USA Rice Federation has a long established policy that there must be market acceptance and regulatory approval prior to the production of genetically engineered rice in the United States.”

The North American Millers’ Association’s Statement on the Use of Food and Feed Crops for the Production of Plant-made Pharmaceuticals and Industrial Products (NAMA 2007) states: “NAMA has significant concern that current confinement systems for controlling the seed, pollen and output of plant-made pharmaceuticals and industrial products cannot control 100 percent of the genetic material of the newly developed organism or prevent deliberate evasion of the security protocol. . . . NAMA

11

believes the risk of adulteration from genetic material not approved for food and feed entering the food chain is unacceptable. NAMA believes that preventing such adulteration is the responsibility of the technology developer and the U.S. government because the prevention of such adulteration is totally within their control.”

On May 4, 2007, a federal judge in San Francisco ordered farmers to stop planting Monsanto’s GM Roundup Ready alfalfa seed because of the risk that it could contaminate nearby non-GM, organic alfalfa fields (Sacramento Bee 2007). This ruling is significant in that it was the first time that GM crop planting was stopped due to the potential for, rather than actual, containment loss. Nationwide, about 200,000 acres of Roundup Ready alfalfa have been planted since the seed was approved for commercial use in June 2005. The judge criticized USDA for failing to adequately assess potential problems with cross-pollination before approving the alfalfa seed for commercial planting. The judge ruled that contamination of an organic alfalfa field with the Roundup Ready gene could effectively destroy the organic farmer’s crop.

In 2007, the USDA (2007c) conducted an investigation of the LibertyLink rice incidents and released findings in October 2007. On August 1, 2006, USDA’s Animal and Plant Health Inspection Service (APHIS) initiated an investigation after Bayer CropScience reported that regulated genetically modified LLRICE601 (Cocodrie variety rice) had been detected in the long-grain rice variety Cheniere. Investigators determined that genetically modified LLRICE601 and Cheniere variety rice were grown at the same location and at the same time at the Rice Research Center North Farm in Crowley, Louisiana, in 1999, 2000, and 2001 under a Bayer CropScience contract. The varieties were separated during those three years by distances of 210 feet, 3,000 feet, and 165 feet respectively. Cheniere was never planted on a location that had been previously occupied by LLRICE601, according to the records provided. Affidavits stated that equipment cleaning had been accomplished by the parties involved at the Rice Research Center North Farm in Crowley, Louisiana, for all planting, harvesting, and cleaning operations during this period. Because rice seed for the period 1999-2002 was no longer available, the exact mechanism for incursion of the LLRICE601 gene into the Cheniere variety, such as gene flow or mechanical mixture, was not determined.

On February 16, 2007, USDA (2007c) expanded the LibertyLink rice investigation to include the discovery of regulated genetic material, later identified as LLRICE604, in the long-grain rice variety Clearfield 131 (CL131). The Arkansas State Plant Board

12

reported that up to 30 percent of the samples of CL131—a long-grain variety of rice developed by LSU that was to be sold as certified rice seed in the spring of 2006—had tested positive for the 35SBar gene i n LLRICE604. The variety Cocodrie containing LLRICE604 was developed by Bayer CropScience (formerly Dow AgroEvo) and was tested at various locations, including the LSU Rice Research Station North Farm in Crowley, Louisiana, between 1998 and 2000. Because the development of these two varieties did not overlap in location and time, the most likely entry point for LLRICE604 into CL131 was through a means other than direct crosspollination. Because LLRICE604 was not detected in representative samples of breeding lines at LSU, the exact time period and means of incursion of the LLRICE604 gene into the CL131 variety was not determined.

USDA is currently exploring revisions to its biotechnology regulations in Title 7, Part 340 of the Code of Federal Regulations (CFR). In July 2007, APHIS published a draft environmental impact statement (http://www.aphis.usda.gov/newsroom/ content/2007/07/content/printable/complete_eis.pdf) that evaluates potential options for revising the biotechnology regulatory program. As a result of this review, APHIS has compiled a list of lessons learned (USDA 2007d) and considerations to enhance its regulatory framework. The lessons learned were:

1. Records are sometimes not easily obtainable because they

are not retained by the permit and notification holders. USDA is exploring whether to require the creation and retention of additional records to inform potential investigations.

2. Efforts to test seed samples during the investigation were hampered by the unavailability of seed samples. USDA is considering (a) revisions to the Plant Protection Act that would provide the agency with authority to subpoena seed samples and (b) revising regulations to require sample retention by permit and notification holders for a specified period of time.

3. In some instances, researchers and developers were unclear about their responsibilities in the event of an unauthorized release of genetically-modified material. USDA is considering revising regulations to require that permit applicants submit contingency plans that address unauthorized releases., have testing procedures to identify released genes, and retain samples of genetically modified materials for test purposes.

13

4. Efforts by USDA offices to work together to collect, test, and track samples were complicated by lack of prior interoffice links and agreements. The USDA is examining options for interoffice memoranda of understanding and agreements to improve collaboration.

5. In some cases, formal, contractual relationships between researchers, developers and other parties did not exist or had expired. This hampered the investigation. USDA is exploring revisions to regulations that would require certain business agreements among technology researchers, developers and other parties.

6. The sufficiency of isolation distances between experimental crops and nearby field crops to ensure confinement was unclear due to advances in scientific understanding. USDA is exploring revising policy to ensure that the latest science is incorporated into isolation distance recommendations.

7. Appropriate quality management systems were not consistently found throughout the biotechnology industry, increasing the likelihood of compliance problems. The USDA is launching a new outreach program to improve quality management systems in the industry.

8. Difficulties in retrieving information delayed inspections and investigations. USDA plans to use its “ePermits” electronic permit system to improve information access and retrieval.

In terms of the potential effects of international biotech regulations on U.S. farmers, in 2004, the European Union adopted a new Directive on Environmental Liability (2004/36/CE) that established the “polluter pays” principle with respect to adverse effects of new organisms, such that producers and biotechnology companies may be accountable for any uncontrolled release of GM materials (Belcher et al. 2005). The European food market is for the most part closed to trade in North American corn, soybeans and canola (Brassica sp.) at least partly because of the extensive adoption of GM varieties in the US and Canada, combined with the lack of effective identity preservation mechanisms to deliver quality assured non-GM produce for the EU market. However, in 2006 the World Trade Organization ruled in favor of the United States and GM food producers when it decided that the European Union had breached international rules by restricting imports of GM crops and foods made from them (New York Times 2006a). In fact, the WTO ruling simply claimed that Europe had failed to follow its own procedures, resulting in undue delays, rather than faulting the

14

European regulatory process for GM crops. If a PMP product produced by a U.S. farmer somehow contaminated a shipment of food to the EU (perhaps organic food), it is uncertain at this time whether the farmer or the biotech company would be liable.

2.2 Ventria Bioscience -- Regulatory History

Currently, Ventria Bioscience is the only firm with PMP field trials in North Carolina, and no PMP products are grown in the field uncontained at commercial scale in the state. Ventria has conducted field trials of rice genetically engineered to produce human milk proteins in North Carolina since 2005. Table 4 provides an overview of Ventria’s regulatory history as described in this section of the report. Ventria Bioscience was founded in 1993 by Dr. Ray Rodrequez, currently a professor of molecular and cellular biology at the University of California, Davis (Ventria Biosciences web site, http://www.ventria.com/, accessed July 20, 2007). In 1997, Ventria developed a proprietary production technology, ExpressTec, that uses rice and barley plants to produce proteins. As of 2007, Ventria had produced three potential protein products, the pharmaceuticals lactoferrin, lysozyme, and serum albumin. These products have not been approved by the FDA for drug, food, or animal feed uses. The products have been marketed as limited research and industrial bioprocessing materials (for cell culture and cell lysis applications) under the brand names Lacromin (lactoferrin, since 2005), Lysobac (lysozyme, since 2006) and Cellastim (serum albumin, since 2006). Ventria plans to market the extracted milk proteins as an anti-diarrheal additive for infant oral rehydration solutions (Bethell 2006) and as nutritional supplements in yogurt, granola bars, performance drinks and other products. Ventria has also mentioned adding rice-based lysozyme to animal feed as a substitute for the antibiotics added to feed (San Francisco Chronicle 2002).

Lactoferrin and lysozyme possess antimicrobial properties and several of Ventria’s proposed uses for its recombinant proteins are explicitly medical in nature. Therefore, the permits initially provided by the USDA for Ventria’s rice production were specifically for rice engineered to produce pharmaceuticals and industrial chemicals. Ventria has made several attempts to change the USDA designation for its rice. In 2003, USDA changed the designation of Ventria’s products from “pharmaceutical proteins produced” to “value added protein for human consumption.” This

15

reclassification of Ventria’s recombinant proteins could potentially pose a number of potential health risks that have not been adequately investigated. In addition, Ventria initiated a voluntary consultation with the FDA so that its rice could be considered as a genetically engineered crop intended for general food use. Finally, Ventria is seeking Generally Recognized as Safe (GRAS) status from the FDA, which would exempt it from the food additive review process.

The FDA considers PMPs to be indirect food additives unless classified as GRAS. Ventria’s products do not have GRAS status. Therefore, Ventria’s products would be regulated by FDA as indirect food additives. As food additives, Ventria must submit documentation to the FDA demonstrating that the products are safe in food. Without FDA approval, food containing non-GRAS food additives would be considered “adulterated,” could not legally participate in interstate commerce, and would typically trigger recall actions. This effectively establishes a “zero tolerance” level for Ventria’s PMPs in food or feed products. The potential for contamination of food-grade rice with Ventria’s PMPs raises the question of unintended exposure. However, the FDA plays virtually no role in pharma crop regulation unless a company reaches the clinical trial stage, typically after 5 to 10 years of outdoor field trials. The FDA does not regulate Ventria’s pharma rice at the field trial stage, and will not regulate it at any stage if the intended use of the rice is production of a research chemical, a “medical food” (which is different from the regulatory category “food”), or for export. Although FDA may ultimately review lactoferrin and/or lysozyme produced from Ventria’s pharma rice if Ventria attempts to market them as food or feed, it will not consider the potential human health impacts of these pharmaceuticals as accidental contaminants in the food supply if Ventria markets the products for research use, as “medical foods,” or produces them for export.

The EPA has authority to regulate products intended for use as pesticides. The EPA has not reviewed Ventria’s PMP rice despite evidence that its pharmaceutical proteins possess pesticidal properties and could harm beneficial organisms, create more aggressive weeds, or disrupt soil ecology, because the PMP rice products are not intended for use as pesticides. Although a scientific advisory panel to the EPA has recommended full length amino acid sequencing of plant-produced recombinant proteins, Ventria has only tested a subset of its amino acid sequences.

In 2004, the USDA granted Ventria Bioscience field trial release permits to grow PMP rice on 120 acres in California (USDA APHIS

16

Permit No. 03-365-01r); however, Ventria was blocked from growing its rice in California (Silber 2004) by opposition from California rice growers.

On June 28, 2005, the USDA announced a “Finding of No Significant Impact” (FONSI) and the availability of an Environmental Assessment (EA) for the proposed field release of Ventria’s PMP rice in Missouri and North Carolina (Federal Register 2005a, 2005b). Based on the EA, USDA/APHIS concluded that the Missouri and North Carolina field releases will not present a risk of introducing or disseminating a plant pest and will not have a significant impact on the quality of the human environment. The USDA granted Ventria field trial release permits to grow PMP rice in 2005 on 200 acres in Scott County, Missouri, (USDA APHIS Permit No’s. 04-302-01r, 04-309-01r, 05-004-01r) and on 70 acres in Washington County, North Carolina, (USDA APHIS Permit No’s. 05-073-01r, 05-117-01r, 05-117-02r) (USDA 2007a).

In 2005, Ventria was blocked from growing its rice in Missouri (Bennett 2005) by farmers and food companies concerned about contamination of their food crops with Ventria’s PMP crops containing proteins that have not been approved by FDA.

In comments filed on June 2, 2005 with the USDA, the Food Products Association (2005) expressed its “concerns with the Ventria lysozyme and lactoferrin applications, as well as other non-food proteins expressed in food crops, center on the clear possibility and consequences of adulteration of food/feed supplies due to contamination by food crops that have been genetically engineered to produce pharmaceuticals or industrial compounds unapproved for food/feed use.”

In June 2005, Ventria planted approximately 60 acres of PMP rice in North Carolina (New York Times 2005c). Planting went forward in North Carolina in 2005 despite objections from researchers at the North Carolina Department of Agriculture and Consumer Services’ Tidewater Research Station (http://www.ncagr.com/Research/trs.htm ), located in Plymouth, NC, where rice varieties from around the world are tested before introduction into U.S. rice breeding programs. Ventria’s field trial location is about a half-mile from the research station (UCS 2006a). According to USDA scientist Dr. David Marshall, who is based at North Carolina State University: “The potential exists for stray rice pollen to be carried via air currents from the Ventria Biosciences fields to the Nursery and pollinating the introduced germplasm. If this were to occur, genes from the rice expressing human lactoferrin could be introduced into the rice germplasm of

17

the National Plant Germplasm System, and thus be disseminated throughout the U.S.” (Center for Food Safety 2005). In comments on Ventria’s North Carolina field test proposal, Dr. Karen Moldenhauer, the Chair of the Rice Crop Germplasm Committee (CGC) and Professor at the University of Arkansas, said: “CGC is concerned about the perception of a grow out this close to the quarantine nursery and hope that they consider moving this grow out to a location farther away (at least 15 miles) from the Tidewater Research Station of NCDA & CS at Plymouth, NC” (Center for Food Safety 2005). The USDA subsequently moved the station to Beltsville, MD (USA Today 2006).

In January 2006, the Union of Concerned Scientists (UCS) filed a Freedom of Information Act request for information on USDA-APHIS inspections and company compliance with federal permit requirements at the Ventria field test site in North Carolina for the 2005 growing season. The USDA provided information detailing how often the USDA inspected the site, what the USDA found, and how well Ventria followed permit requirements. The USDA records showed that (1) the USDA failed to inspect the Ventria site during planting and harvest, two of the most critical times with respect to ensuring containment, (2) Ventria submitted only one of nine required notification/planting reports to USDA, (3) the USDA completed only three of five required inspections at the Ventria site, and (4) the USDA did not communicate with Ventria about the effects of Hurricane Ophelia, which passed close by the site in September 2005 (UCS 2006a). A UCS report concluded that the USDA was apparently failing to adequately monitor and inspect the Ventria test site.

Ventria withdrew USDA permits for PMP rice field trials in MO in February 2006 (USDA APHIS Permit No’s. 05-336-01r, 05-336-02r).

North Carolina field trials were subsequently approved by USDA in November 2005 and went forward in 2006 (USDA APHIS Permit No’s. 05-293-01r, 05-332-01r, 05-332-02r). In March 2006, Ventria received approval from USDA to expand its field trials in Washington County, NC, from 70 to 335 acres.

North Carolina field trials for 2007 were also approved by USDA in November and December 2006 (USDA APHIS Permit No’s. 06-305-04r, 06-285-01r).

On February 28, 2007, the USDA released a draft environmental impact statement concluding that Ventria’s PMP rice could be grown in Kansas with no undue risks (Ironically, on the same day

18

the USDA announced that rice seed in Arkansas were contaminated with GM rice LL62.) (Washington Post 2007).

Despite the containment breaches involving Liberty Link rice in 2006 and 2007, in May 2007 the USDA granted Ventria release permits to grow 3,200 acres of commercial PMP rice in Geary County, Kansas (USDA APHIS Permit No’s. 06-285-02r, 06-278-01r, 06-278-02r, Fortune 2007), which would be the world’s largest PMP planting to date (Weiss 2007, Freese 2007). On May 16, 2007, the USDA announced a “Finding of No Significant Impact” (FONSI) and the availability of an Environmental Assessment (EA) for the proposed field release of Ventria’s PMP rice in Kansas (Federal Register 2007). Based on the EA, USDA/APHIS concluded that the Kansas field releases will not present a risk of introducing or disseminating a plant pest and will not have a significant impact on the quality of the human environment. APHIS stated in the ruling that “The combination of isolation distance, production practices, and rice biology make it extremely unlikely that this rice would impact the U.S. commercial rice supply.” However, these are the same factors that have failed to prevent containment breaches in the past.

The Union of Concerned Scientists (UCS 2007) criticized the USDA’s decision on Ventria’s Kansas application based on the following grounds:

(1) Ventria did not supply enough information on the acres to be planted (3,200 acres are implied in other USDA documents)

(2) the procedures and safeguards to be used by Ventria to ensure that none of the PMP rice escapes containment or persists in the environment after harvest, as described in the permit application and the Ventria’s standard operating procedures (SOPs), were not made public in USDA’s environmental assessment documents, the documents on which USDA made its permit approval decision

(3) the analysis made public by USDA does not consider three potential routes of containment loss: production, shipment and storage of PMP seed prior to planting, post-harvest transport of PMP rice to processing facilities, unintentional dissemination of PMP rice in the field by extreme weather events, such as floods and tornados (the proposed Kansas sites are within 4 miles of the Kansas River and one mile of the Smoky Hill River tributary, both of which flooded in 1993 according to the National Oceanic and Atmospheric Administration; Kansas ranks third among states in tornado frequency, with an average of 47 tornados per year), and

19

(4) the containment breaches involving GM Liberty Link rice that have occurred under current USDA regulations.

It should be noted that the USA Rice Federation (2007b) filed comments with USDA on March 29, 2007, strongly recommending that APHIS deny Ventria permission to grow [PMP] rice:

“The USA Rice Federation today expressed its disappointment with USDA APHIS’ approval of the Ventria Bioscience request to grow rice containing human proteins in Geary County, Kansas. . . . The USA Rice Federation is disappointed with the APHIS decision and hopes Ventria and regulators will carefully ensure that sound and enforced protocols will prevent contamination of the commercial rice supply—an event that would be devastating to the rice industry. . . . The U.S. rice industry is still reeling from the release of BayerCropScience’s genetically engineered Liberty Link rice into the U.S. Delta-region rice fields. We are living with the effect of unintended events and consequences. This decision will not generate any comfort among U.S. commercial rice growers.”

Ventria received permits to produce value-added proteins using PMP rice field trials in KS in May 2007 (USDA APHIS Permit No’s. 06-278-01r, 06-278-02r, 06-285-02r). Ventria received permits to produce pharmaceutical products using PMP rice in KS in February 2008 (USDA APHIS Permit No. 07-342-102r).

Ventria received permits to produce pharmaceutical proteins using PMP rice field trials in NC in March 2008 and permits to produce pharmaceutical products using PMP rice field trials in NC in April 2008 (USDA APHIS Permit No’s. 07-341-103r, 08-093-108r).

Again, it should be emphasized that because Ventria’s PMP rice will be the first PMP crop to be grown in the field uncontained at commercial scale in the United States, decisions concerning its regulation are potentially precedent-setting.

20

3 Potential Benefits of PMPs

3.1 Overview

For millennia, farmers have used selective breeding to produce crops with desirable characteristics. The novel aspect of GM technology is the ability to move genes and associated characteristics between organisms that are not sexually compatible, creating organisms with previously unavailable bundles of characteristics. GM technology has been used to increase crop yield, drought tolerance, herbicide tolerance, disease/insect resistance, and product quality. Most recently, GM technology has been used to produce PMP substances within crop plants. Many of the PMP products under development are proteins--antibodies, enzymes, vaccines and other therapeutic agents--due to an increasing number of protein-based drug discoveries by pharmaceutical companies. In 2005 alone, 38 new protein-based drugs were approved and more are in the FDA pipeline (Williams 2006, 2007). The pharmaceutical industry seeks low-cost production methods for these new drug products. Producing drugs inside green plants, PMPs, is one of several alternatives.

Scientists and industry typically cite two reasons for pursuing plant made pharmaceuticals (PMPs) (Smyth et al. 2004). First, production of high-quality pharmaceutical components (proteins and antibodies) is presently done using cell cultures inside bioreactors, which is very costly (US$105-175 per gram) and limits consumer affordability. Cell culture bioreactors take an average of three to seven years to build and cost on average US$450-$600 million to complete. Second, there is insufficient bioreactor capacity to meet current production needs, let alone expected future needs over the next decade (BIO 2002b). Antibodies produced in bioreactors using mammalian cell cultures are expensive, difficult to scale up, and pose safety concerns due to potential contamination with pathogenic organisms or oncogenic DNA sequences (BIO 2002b). As of 2002, production of just four pharmaceutical products required 75% of global bioreactor capacity (BIO 2002a). By the end of the decade, there could be more than 80 antibody-dependent products with an estimated value of US$20 billion, provided adequate production capacity can be developed (Smyth et al. 2004). The Biotechnology Industry

21

Organization, an industry trade group, reports that a January 2005 study by Frost & Sullivan, a market research firm, found that the PMP market could realize total cumulative revenues of US$98.2 billion by 2011 (BIO 2006). The potential size of the market drives investigation of alternative production methods, including PMP production. Compared with other production methods, the costs of producing and storing plant-produced pharmaceuticals are relatively low, plants may be able to produce the product for extended periods of time, product quality is relatively high, and risk of contamination by pathogens is low (Table 5). The leading PMP plants have been corn/maize, canola/rapeseed, safflower, tobacco and rice.

In July 2006, Calgary-based SemBioSys announced that it can produce over one kilogram of insulin per acre of PMP safflower (BIO 2006). This is enough to supply 2,500 patients for one year of treatment each. With insulin demand projected to be 16,000 kilograms per year by 2012, SemBioSys’ GM safflower provides a way to supply insulin to a growing diabetic patient population. It is claimed that producing insulin in PMP safflower can reduce capital costs by 70 percent and product costs by 40 percent, compared to existing insulin manufacturing. In February 2007, the USDA announced a preliminary decision to allow SemBioSys to plant 1000 acres of PMP safflower in Washington state, although this initial planting would produce a drug to treat diseases in farmed shrimp and promote fish growth rather than insulin. (The USDA’s decision to allow SemBioSys to plant PMP safflower on a commercial scale has been criticized (UCS/CU 2007) based on the fact that the USDA review did not assess the potential risks of escaped PMP safflower in the environment, including the risk of becoming an agricultural plant pest, but rather assumed that SemBioSys’ proposed containment measures would be 100 percent effective.) Other PMP products under development in 2006 included: cystic fibrosis treatment from GM corn (Meristem); treatment for ovarian cancer from GM tobacco (Chlorogen); GM tobacco to address dental caries, as well as the common cold, and hair loss (Planet Biotechnology); monoclonal antibodies from GM duckweed (Biolex), and human milk proteins from rice (Ventria Bioscience) (BIO 2006).

Although Ventria’s recently proposed PMP rice processing facility in Kansas may promote economic development in the region (assuming project financing and construction proceed as projected, and Ventria is able to secure necessary approvals to market its products), the history of PMP product development to date indicates that caution is warranted when projecting the economic

22

development benefits of PMP production. Many PMP companies have either gone bankrupt or have ceased pursuit of PMP production, or switched to non-food crop PMP production (Freese and Caplan 2006). ProdiGene was saved from bankruptcy after its PMP corn contaminated non-GM soybeans in 2001 and it was forced to pay for the cleanup by a USDA no-interest loan; ProdiGene was subsequently taken over by Stine Seed. CropTech went bankrupt in 2003 after pursuing PMP production in tobacco. Meristem Therapeutics stopped PMP corn trials in Colorado in 2003 due to farmer-led opposition. Monsanto ceased development of PMP corn and soybeans in 2003 even though it had received 44 field trial permits from USDA. Epicyte Pharmaceutical, once a leader in PMP corn development, went bankrupt and was taken over by Biolex in April 2004; Biolex now produces PMPs using the non-food plant duckweed inside controlled bioprocessing facilities. LargeScale Biology went bankrupt in 2005 after pursing PMP production in viral-vectored tobacco. Ventria Bioscience dropped field trial plans in California in 2004 and Missouri in 2005 due to farmer opposition.

3.2 The Case of Ventria Bioscience

In this section, we consider in detail the potential benefits associated with Ventria Bioscience’s PMP rice development and production, as Ventria’s PMP rice may be the first PMP crop to be produced in the field uncontained at commercial scale in the U.S. Given that Ventria is a private company developing a new product in the very competitive biotech industry, the firm does not provide estimates of the potential benefits to the firm itself associated with the eventual production and marketing of its PMP products. In terms of current employment supported by the firm’s activities, the Sacramento Bee (2006) reports that Ventria had 18 employees in its Sacramento headquarters in 2006.

In 1997, Ventria developed a proprietary production technology, ExpressTec, that uses rice and barley plants to produce proteins. As of fall 2007, Ventria has only three potential products, the pharmaceuticals lactoferrin, lysozyme, and serum albumin that have not been approved by the FDA for drug, food, or animal feed uses. These products have been marketed as research and bioprocessing materials (for cell culture and cell lysis applications) under the brand names Lacromin (lactoferrin, since 2005), Lysobac (lysozyme, since 2006) and Cellastim (serum albumin, since 2006) by Ventria directly, and by firms InVitria (http://www.invitria.com/index.html) and Sigma-Aldrich

23

(http://www.sigmaaldrich.com/catalog/search/TablePage/15552187), but it is not clear that Ventria has received substantial revenues from these uses. As of fall 2007, Ventria appears to be supported financially primarily by venture capital and with some indirect subsidies from state (Kansas) economic development agencies. For example, the Kansas Bioscience Authority gave $1 million to Junction City, KS, to support the attraction of Ventria Bioscience (http://www.kansasbioauthority.org/projects_funded/). (There appear to be no subsidies to date from North Carolina state government.) Ventria plans to market the extracted milk proteins as an anti-diarrheal additive for infant oral rehydration solutions (Bethell 2006) and as nutritional supplements in yogurt, granola bars, performance drinks, and other products. Ventria has also mentioned adding rice-based lysozyme to animal feed as a substitute for the antibiotics added to feed (San Francisco Chronicle 2002). Ventria claims a potential market for these products of more than $2 billion annually. Ventria claims the following economic and societal benefits associated with its PMP products (Ventria Bioscience 2007):

• Potentially save hundreds of thousands of lives globally by reducing childhood diarrhea in developing countries;

• Reduce duration of childhood diarrhea by 4 million days annually in the US and help these children get back to school sooner; Help parents return to work sooner with an economic impact of $1.6 billion over five years in the US alone;

• A $50 million positive economic impact over five years from direct employment in Ventria’s bioprocessing operations in Junction City, Kansas;

• A $228 million positive economic impact over five years to farmers and rural communities from Ventria’s field production activities in Kansas;

• $37.5 million in savings to the US Government and American taxpayers when compared to government subsidized rice production;

• Successful introduction of these first products may lead to additional products being developed using plants as a biological factory. This multiplies the benefits to society and the US economy.

24

In support of the first claim, Ventria sponsored a study in Peru to assess the efficacy of rice-based oral rehydration solution containing recombinant human lactoferrin and lysozyme in Peruvian children with acute diarrhea (Zavaleta 2007, Bethell 2006). Ventria’s interpretation of the study results is that Ventria’s products helped to reduce the duration of acute diarrhea by 30%, or a day and a half. (Average duration: 5.21 days for control vs. 3.67 days for Ventria’s products). In addition, Ventria claims that the study shows that children receiving Ventria’s product more likely to recover from their diarrhea and were less likely to relapse into another episode of diarrhea. Freeze (2007) disputes the study findings on several grounds related to alleged problems with the study methodology.

Even if the Zavaleta (2007) study results are scientifically sound, the potential profitability of Ventria’s oral rehydration supplement products to the firm itself may be limited by the inability of consumers in the target market, low income households in developing countries, to pay. Ventria’s CEO Scott Deeter has said that financial support from foundations might be necessary to make oral rehydration solutions containing his company’s proteins widely available (USDA 2003, Freese 2007).

With respect to Ventria’s claim of potential benefits to consumers in the United States, where consumers have a greater ability to pay for the product, Ventria applies it’s interpretation of the results from the Peruvian study to the number of childhood diarrhea cases in the United States and the number of working parents and the average daily wage in the U.S (Ventria 2007). The application of the Peruvian study results to the United States may overstate potential benefits in the U.S. if children in the U.S. have better overall nutrition, sanitary conditions, and hygiene, relative to Peruvian children, reducing the relative benefit of Ventria’s products. A controlled study of Ventria’s products on children in the U.S. would appear to be necessary to verify this benefit claim.

Another potential hurdle to realizing consumer benefits in the United States is that, despite the results from the Peru study, Ventria has failed to gain “Generally Recognized as Safe” (GRAS) status from the U.S. Food and Drug Administration for its rice-derived pharmaceutical proteins in four petitions since 2003 (Table 4). Ventria has applied to the FDA to approve its PMP proteins as a “medical food” rather than a drug (USA Today 2006). As a medical food, Ventria would not need to conduct long and costly human tests. Instead, Ventria submitted data from scientists in support of “generally regarded as safe,” or GRAS, status. If Ventria

25

wins approval to add its PMP proteins to infant formulas, there is no requirement to label any food products in the U.S. as containing genetically engineered ingredients.

Part of the reason why Ventria has yet to be granted GRAS status for its PMP rice may be that a 2004 National Academy of Sciences report (NAS 2004) recommended more stringent testing for new ingredients in infant formulas. To date, Ventria has chosen not to submit its proteins for review by FDA as new drugs, a more rigorous review process. Concurrently, however, another company (Agennix, based in Houston, TX) has been developing recombinant human lactoferrin under FDA’s new drug review process for use as an anti-cancer drug since 1996 (Freese 2007). The material is being produced in genetically modified fungus in a contained manufacturing facility, not in field crops. That lactoferrin is being considered as a potent anti-cancer drug raises concern about Ventria’s attempt to gain approval for the material under the less stringent food additive regulations. Production of lactoferrin in fungus also presents a potential competitor for PMP rice lactoferrin, depending on regulatory approvals and relative production costs.

If Ventria’s products are eventually certified as safe, the net benefits of Ventria’s products to potential consumers, economically speaking, are defined as the incremental benefits beyond those provided by the next-best substitute product. Even if Ventria’s products are completely safe and effective, the benefits to the ultimate consumers, infants at risk for diarrhea, should be measured relative to the benefits provided by the next-best substitute product. Freese (2007) makes the case that improved sanitation facilities, clean drinking water supplies, improved hygienic practices, use of disinfectants, and better breastfeeding practices, in combination with existing oral rehydration therapy, provide a good substitute for rice-derived proteins in terms of reducing the incidence of diarrhea, perhaps at lower cost, in developing countries. In the U.S., the benefits of the next-best substitute treatment for childhood diarrhea would need to be compared with the benefits of Ventria’s products to determine the potential net benefits of Ventria’s products to U.S. consumers.

Potential consumer benefits in the U.S. may be reduced if the patent holder, Ventria, can exert monopoly power and raise consumer prices. Although there are no estimates of Ventria’s ability to exert monopoly power in the market for transgenic rice products, Kostandini et al. (2006) estimated the potential size and distribution of economic gains from biopharming transgenic

26

tobacco as a source of human serum albumin using an economic surplus model under imperfect competition. Kostandini et al. determined that the development of transgenic tobacco would generate annual profit flows of between $25 million and $49 million for the patent holder. Because the patent holder can exert monopoly power in the output market, consumer prices are higher, and consumer benefits lower, than would be the case in a competitive market. However, should both rice and tobacco prove successful as sources of serum albumin, some degree of competition between the two would presumably lower prices, reduce profits, and benefit consumers.

In addition to consumers, Ventria’s products may provide economic benefits to farmers, crop transportation, processing and distribution workers, and others who receive benefits due to economic multiplier effects. With respect to benefits claimed by Ventria for farmers, PMP processing workers, and the local rural community near Junction City, Kansas, the site of Ventria’s planned PMP processing facility, see Section 4.1 of this report below.

With respect to the estimated benefits that may accrue to the Junction City, Kansas, community due to the economic multiplier effects of Ventria-related farmer and processor activity, Ventria estimates that “with a projected 30,000 acres of production per year upon full scale commercialization of Ventria’s products, we estimate the resulting economic benefit to be $18 million per year in direct economic benefit for farmers and the rural community of Junction City, Kansas.” Ventria’s proposed PMP rice processing facility in Junction City “is a $6 million capital improvement project and is expected to employ 10 people within the first year of operation. Employment will expand as the demand for Ventria’s products grows. It is estimated that an employment of 50 people in Junction City, Kansas will be required for full-scale production.” Ventria assumes an economic multiplier of 2.54 [based on the economic multiplier used by Junction City/Geary County Economic Development Commission], to develop an estimate of the total economic benefit (direct benefits plus economic multiplier effects) for farmers and rural communities from Ventria’s products of $45 million per year over the first five years of full-scale production. For comparison, in 2006, Kansas agriculture produced over $11 billion in crop, animal, and related agricultural output, with over $3 billion in wage, rent, interest, and profit income (USDA 2007e). Using a 2.54 economic multiplier, the total economic impacts of the $11 billion in direct impact would be on the order of $28 billion. Ventria’s estimated economic impact of

27

$45 million per year is small relative to the$28 billion impact of Kansas agriculture.

Ventria’s claim of “$37.5 million in savings to the US Government and American taxpayers when compared to government subsidized rice production” is not valid, as PMP-rice is not grown for food and so will not substitute for the rice grown for food that receives the rice subsidy. However, if Ventria’s PMP-rice replaces subsidized corn, then Ventria would potentially be able to claim a reduction in corn subsidies as savings to U.S. taxpayers. Forty-four percent of farms in the Kansas region received government payments in 2005, with an average payment of $17,000 per farm (USDA 2007e), or $18,000-$20,000 per farm in Geary county, Kansas (KFMA 2006). In 2007, Ventria estimated “. . . a projected 30,000 acres of production per year upon full scale commercialization of Ventria’s products” (Ventria Biosciences 2007). With an average farm size of approximately 700 acres in Kansas (http://www.ers.usda.gov/StateFacts/KS.HTM), perhaps 43 farmers would participate in Ventria’s PMP rice production in Kansas. If we assume that 43 farms growing Ventria’s rice would have otherwise participated in farm programs in which they would have received $17,000 each in government payments, then substituting Ventria rice for corn in Kansas could save taxpayers on the order of $731,000. This number is very small relative to the almost $286 million in net government payments made to Kansas agriculture in 2006 (USDA 2007e).

The economic benefits of Ventria’s PMP rice field test activity in North Carolina are difficult to determine, as Ventria will not reveal information on the numbers of farmers or researchers actively at work in the state (Sargent 2007). However, given the low acreage involved, it is likely that only a very few farmers are participating in the field tests. It is known that Ventria project researchers, including professors from North Carolina State University, are based at the Tidewater Research Station in Plymouth, NC (Washington Daily News 2006a). NCSU professor John Van Duyn, is reportedly doing research for the Ventria project in Washington County.

Although Ventria looked for a place to process its rice that would be “within 50 miles” of its PMP rice field test site in Washington county, NC, and Dr. Scott Deeter of Ventria said that the company was considering placing a processing facility in Washington County, Greenville or Wilson, NC, Ventria said in December 2006 that it planned to maintain operation of 200 acres of PMP rice in Washington County, NC, but that it would expand rice production

28

and establish a rice processing plant in Kansas instead of North Carolina (Washington Daily News 2006b).

29

4 Potential Costs of PMPs

The potential benefits of PMPs must be weighed against the potential costs, including: (1) the costs to the farmer of specialized, dedicated equipment, training, administration and liability to the GM/PMP-producing farmers, (2) any government subsidies to Ventria or farmers, (3) the costs of any harm to human health in intended uses (e.g., allergies), (4) liability costs associated with the potential loss of containment of PMP products and subsequent contamination of the food supply, (5) externality “spillover” costs affecting non-GM producing farmers, including organic farmers, and (6) externality “spillover” costs affecting the environment. The first four cost categories are considered in this section of the report, while the two types of externality costs are covered in following sections.

4.1 Farm Costs and Potential Grower Profitability

Some GM crop technologies and products are developed by public institutions (such as public universities and federal research laboratories) financed by tax dollars, while others are developed by private, profit-seeking firms. The intellectual property developed by public institutions is typically financed by tax dollars and distributed to users without charge, for example, though publication in publically-available academic journals, whereas the intellectual property developed by private firms is typically owned by the inventor, who tries to recoup his development costs and make a profit by, for example, increasing the price of GM crop seed, charging a technology fee, or requiring that the crop be sold back to the firm.

Regardless of the source of innovation, farmers must somehow gain from a new technology in order to adopt it. Typically, new technology must provide increased financial returns to the farmer by some combination of raising crop yields, lowering input costs, enhancing crop quality (thereby increasing the price consumers are willing to pay), or reducing farm management effort. A rough estimate of gains to farmers from PMP crop production could be made by estimating increases in net returns (benefits minus costs) per acre and multiplying the per acre gains by the number of affected acres. When Ventria was considering locating its PMP rice

30

processing facility in northwestern Missouri, the firm reportedly agreed to pay PMP-growing farmers in the region about twice what they would typically earn growing their next-most profitable crop (New York Times 2006b). Ventria (Ventria Biosciences 2007) estimated that farmers located near the site of its planned PMP rice processing facility in Junction City, Kansas, will “earn approximately $150 in additional profit per acre plus additional economic impact from more intensive management required of Ventria’s production, requiring an additional $300 per acre. For example, a corn farmer that is currently generating $587 per acre from corn production would generate an economic impact of $1,037 per acre, or an increase of $450 per acre if they switch to Ventria’s production.” These estimates are based on analysis by Daniel O’Brien, Associate Professor and Extension Agricultural Economist, Kansas State University. In addition, Ventria makes the claim that:

“. . . [farmers] are able to receive a more consistent revenue stream versus their alternatives because they do not shoulder losses caused by poor yields, weather damage, disease or insect damage, or other negative impacts typically faced by farmers today. Third, the farmers are trained in new value-added farming practices, quality control, and regulatory requirements. Finally, farmers are able to enter multi-year agreements which provide more certainty about future cash flow, thereby improving their financial outlook. Based on the above, we estimate an economic benefit to farmers of $600 per acre in positive economic impact compared to their alternative with corn” (Ventria Biosciences 2007).

Per acre impacts in North Carolina would likely be different from those in Kansas, due to differences in crops grown, their production costs, and market prices. As discussed by Wisner (2005), PMP firms such as Ventria will be the sole suppliers of their PMP products and may choose to let farmers compete with one another for PMP production contracts, inevitably lowering the contract prices paid to farmers, and reducing farmer benefits from PMP production.

Turning to estimates of the eventual number of acres that may be devoted to Ventria’s PMP rice production, in a presentation to a USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003), Dr. Scott Deeter, president and CEO of Ventria Bioscience, described the likely acreage involved as approximately 10,000

31

acres and the number of farmers involved as “not very many people;” furthermore, it is not clear that all of this acreage would be rice grown in North Carolina:

“In 2008, well I'm making the prediction that the second decade here is where we're going to hit the mainstream or going to really begin producing products with these -- with these platforms. I'm saying 10,000 acres. Depending on your efficiency and your yield, that would be an enormous amount of pharmaceutical product. An enormous amount, okay? So this is success for us, okay? Ten thousand acres. I look at that and I put my pharmaceutical hat on, and I say, holy mackerel, is that really that much volume in the pharmaceutical business? Okay? And the answer is yes, because many of the new products that are being developed in the pharmaceutical industry require chronic dosage at high levels. And we haven't been able to go after those products in the past because we didn't have a system to do it. I look at that from my agricultural hat, and I say, that's nothing. How many farmers is that? Maybe, you know, [a few] good-sized farmers, that's not very many people. So that's just two perspectives here that I think are kind of interesting as you think about this.”

When a member of the Advisory Committee said: “I have to tell you that my constituents, or who I'm representing here today, are wheat producers. And to a larger extent, I would suppose just production agriculture. And when you put the numbers on the table of 10,000 to 100,000 acres, frankly, that's very, very small in the scope of U. S. agriculture,” Dr. Deeter’s response implied agreement, and that Ventria’s efforts would not significantly affect the rural agricultural economy in the U.S.:

“But the big benefit here, in my mind, is human health. That's the problem [Ventria is] working on. We're not -- we're not working on rural development.”

In 2007, Ventria estimated “. . . a projected 30,000 acres of production per year upon full scale commercialization of Ventria’s products” (Ventria Biosciences 2007). With an average farm size of approximately 700 acres in Kansas (http://www.ers.usda.gov/StateFacts/KS.HTM), perhaps 43 farmers would benefit from PMP rice production in Kansas, but the number would probably be lower, as a smaller number of larger

32

farms would reduce costs, based on economies of scale in the use of specialized farm equipment and farmer education required to produce PMP crops. Multiplying this acreage estimate by Ventria’s estimates of $150 to $600 per acre in additional returns to farmers results in a ballpark estimate of $4.5 to $18 million for farmers. The number of farmers supported and the amount of farm income earned in Kansas would be lower if Ventria decides to produce some of its rice in a second location, which it reportedly is considering. In 2006, Kansas had approximately 64,000 farms that produced over $11 billion in crop, animal, and related agricultural output, with over $3 billion in wage, rent, interest, and profit income (USDA 2007e). A direct economic impact of $4.5 to $18 million is very small relative to the size of Kansas agriculture.

In any event, Ventria’s Dr. Deeter implied in his statement before the USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003) that most of the new opportunities in PMP would be in the areas of processing and support, rather than in production agriculture:

“Those new skills are going to bring new opportunities, especially, I think Michael said it well, especially in the ancillary areas. All the support, all the facilities where you process these. These are biotechnology production facilities. One of the challenges right now in the biopharmaceutical industry, even traditional biopharmaceuticals, is talent. Not enough people know how to run these kinds of facilities. So, as we develop that in a state like say Iowa, then there's all these other services that come around it that really are significant.”

However, in September 2006 Ventria announced plans to locate a PMP rice processing facility in Junction City, Kansas, instead of Iowa or North Carolina (Ventria Bioscience 2006). Significantly for North Carolina economic development, the facility will not be located in North Carolina. Nonetheless, it is possible that North Carolina farmers could benefit by growing PMP rice and transporting it to Kansas for processing. Ventria reportedly wants to grow rice at commercial scale in at least two locations in the United States in order to have a diverse supply base (it is also reportedly searching for a growing area in the Southern Hemisphere to be able to produce year-round) (New York Times 2005c). In the Junction City announcement, Kansas Agriculture Secretary Adrian Polansky said that farmers are expected to be among the project’s major beneficiaries, as those who grow the rice that supplies the facility can earn a premium compared to their

33

next most lucrative crop. “This is as an important development for Kansas farmers, who stand to benefit from the additional income,” he said. However, North Carolina farmers were not mentioned, and it is unclear whether Ventria plans to grow PMP rice in North Carolina.

Given Ventria’s investment in the Junction City, Kansas, processing facility, it is likely that the total acreage of Ventria’s PMP rice farmed in North Carolina will be small. Nonetheless, Ventria secured USDA permits for field trials involving PMP rice in North Carolina for 2005 and March 2006. The 2006 planting in North Carolina is 335 acres of two plant-made industrial enzymes, lactoferrin and lysozyme, using genetically modified rice (Oryza sativa). (Ventria received field trial permits to produce value-added protein using PMP rice in NC in November and December 2006. Ventria withdrew permits to produce selectable markers and increased protein levels in PMP rice in NC in December 2007. The firm received a permit to produce pharmaceutical protein in PMP rice in NC in March 2008. Ventria also received a permit to produce pharmaceutical product in PMP rice in NC in April 2008.) Ventria is implementing the field trials using independent grower contracts. At this early stage, Ventria pays all costs for the North Carolina farmer growing their PMPs on subcontract. In the future, independent growers will be expected to provide a seed-to-harvest package deal for the firm’s recombinant protein production (Williams 2006, 2007). As statements made before the USDA Biotechnology Advisory Committed quoted above indicate, this will likely involve significant, specific investment in PMP-related training and dedicated farm equipment. Since 2003, each PMP grower is now required to have dedicated land area, dedicated equipment for planting and harvesting in addition to separate areas for cleaning and processing. Employee training is also required as part of compliance with new FDA and USDA regulatory statues for molecular farming (Stewart and Knight 2005). This raises concerns that molecular farming contracting for field-grown PMP plants will require such costly investments in infrastructure and compliance that only the largest, wealthy growers in North Carolina can profit.

Despite Ventria’s claims that farmers will profit from growing PMP crops, farm profits may be small, as the PMP patent holder, the biotech firm supplying the seeds, can exert monopsony (a “monopsony” is a buyer’s monopoly) power over farmers. That is, competition among farmers for the limited number of PMP production contracts may lower the price that farmers receive for the PMP product when sold back to the PMP developer/processor.

34

For example, Kostandini et al. (2006) estimated the potential size and distribution of economic welfare gains from biopharming transgenic tobacco as a source of human serum albumin using an economic surplus model under imperfect competition. Because the patent holder can exert monopsony power in the input market (over farmers), it was found that production of transgenic tobacco was unlikely to significantly benefit tobacco farmers.

Ventria is currently planning to produce serum albumin using PMP rice (Washington Post 2007). If both sources (transgenic tobacco and transgenic rice) of serum albumin are approved, any monopoly power in the product market that PMP rice developers and farmers might have enjoyed would presumably be reduced, which would lower the price of serum albumin, in turn lowering profits for the biotech patent holders and farmers (but benefiting final consumers).

For those farmers choosing to grow Ventria’s PMP crops, significant training, dedicated equipment, extra paperwork, and adherence to strict operating procedures to ensure containment (the violation of which could have tremendous financial consequences) would appear to be required. In a presentation to a USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003), Dr. Scott Deeter, president and CEO of Ventria Bioscience, addressed these issues:

“Right. So our -- our Director of Field Production carries out our training. The first part of the training is actually what is the -- what is the risk of noncompliance, which is pretty severe. I mean, essentially, it's the company. It's make or break. And so we want people to be very clear that this isn't something that's even thought of as optional. That's number one. Number two, the training consists of specifically of the operating personnel is going to be on a piece of equipment. We might have a standard operating procedure for cleaning the combine, okay? And that cleaning procedure would have probably 30 different steps, you know, estimating. And each of those steps, we would go through on the equipment with the -- with the field personnel. And then when they actually carry out that activity, then they would sign the forms that said they're aware of the cleaning procedure, they carry it out according to this procedure, et cetera. So they go through -- there's a lot of paperwork in this process. And they would go through that for each standard operating procedure. The -- we've actually trained, at a broad level, our entire company. We

35

can do that, okay? We're not that big of a company. But we've trained everyone because we believe it's a -- it's a culture. You've got to build a culture of here's the regulation, but then here's our operating procedures. Because, in most cases, our operating procedures are significantly advanced, even ahead of the regulation. I mean that's just our approach from a quality standpoint as well as our own goods. Did that answer your question?”

Asked about the special operating procedures, combine cleaning, and dedicated equipment necessary to grow PMPs, a farmer (and former president of the Iowa Corn Growers Association) growing corn PMPs in Iowa for the French company Meristem Therapeutics reported in a USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003):

“So when we started this three years ago, we weren't required to do so, but it looked to us like the only way that you could really guarantee the kinds of identity preservation that are necessary is to have dedicated equipment for that plot of production. If any of you have ever tried to clean out a combine, you know how difficult it would be to get every broken kernel, all the dust, everything out of a conventional combine. Someday we'll have combines that will be self-cleaning, but we don't have them today. So we started with dedicated equipment. We've argued to APHIS for -- well, we argued for two years that everyone ought to be using dedicated equipment in the 250 permits that they have issued around the country. APHIS told us no, no, you can't require people to have dedicated equipment because that would shut down, basically, all the land-grant institutions around the country that are doing work, because they simply can't afford dedicated equipment for each plot. And our argument was, well, we didn't think that they could afford not to have dedicated equipment. So we were very pleased for the new rules for this year [2003], that there is dedicated equipment required now.”

“. . . Each protein would have its own building, its own equipment, its own system. I think that's the only safe way to do it, at least at this point, with the technology we have today. Now, I personally don't believe it's possible to get every protein out of a combine. That's why we use the dedicated equipment. And we are starting, in our SOPs now, we haven't completed it yet, but we're starting to write the SOPs for decommissioning equipment. In other words, some

36

day in the future, that combine is going to become obsolete for whatever reason, and -- and we're not going to use it any more. So we have to decide if there is a procedure, if you can auto clean that combine, if there is a detergent or enzyme you can use to clean the combine, or should you simply set it on fire and bury it? Whatever it takes, that's what will happen. Now, there was comment made earlier about the cost of this production system. . . . we put together a used set of equipment for this project for about $20,000. And the size of equipment we have for these small plots, 6, 8, 10, 12 farmers could all use the same equipment. You could put it on a truck, take it from farm to farm, and use the same equipment. So if you split up the $20,000 cost over -- over 8 or 9 farms, you can see it's not that significant. Hopefully, if Scott, this is worked right, the margins for us will be significant enough that we can afford to do that. But I don't see any problem with the equipment, commissioning it or decommissioning it. Those are all things that are done routinely in pharmaceutical production.”

“. . . And then we have designed this system where we have a sort of traditional wagon here that the corn is put into, but we have installed a drying floor in the bottom of the wagon, and we have installed a drying fan here. And so this corn is harvested, transported, dried and stored in this -- in this wagon. Before it leaves the field, this wagon is weighed. We have portable scales. And we use a mass balance system. So we know exactly how much product we take from the plot to the storage building you saw in the earlier slide. And so then, as we process, we clean and separate and have the cobs and broken pieces of stalk that are separated out and cleaned out. Everything is measured at the end so we know exactly how much mass we have divided out at the end. It has to match exactly in balance with the mass we took out of the field. And there again, it's another system that we're not required to do, but we do. And this is my last slide. This is -- this is double-wall containers. The product is being shipped back to France from the farm.”

The farmer was asked about the issues involved in transport of that equipment not only to and from the plot at his facility, but also about the potential of sharing equipment among multiple producers. The farmer replied:

“So what our SOPs [standard operating procedures] require is that the planter, for instance, the planting units in

37

each box are disassembled. When they leave the building, you can easily see that there's no seed in the planter. The planter's taken to the plot. The units are reassembled in the plot. Seed is put in the planter. The plot is planted. Then the seed is taken out, the units are disassembled again to show that there's no seed in the units. And the planter's transported back. Now, the combine, of course, you can't do that. You can only clean as well as -- I mean you can crawl in there, and you can get to a lot of places, but you can't get every place. And we use a vacuum to vacuum out everything that we can possibly reach. The main -- the main concern when transporting the combine is something falling out of the combine. But if you can get in there and clean everything that you can reach, there isn't a chance of something falling out then, because it can't get to an edge and fall out. Did I answer the question? Okay.”

The farmer’s statements above appear to indicate that additional, special procedures are necessary when using PMP farm equipment, and that some equipment, such as combines, cannot be cleaned completely for transportation between farms. If a spill occurred during transport between farms, resulting in a loss of containment, it is possible that APHIS would disallow sharing equipment across farms. If this occurred, the costs of maintaining specialized equipment for PMP could not be shared by multiple farmers, as the farmer quoted above describes.

In his presentation to the USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003), Dr. Scott Deeter, president and CEO of Ventria Bioscience, also described the special skills that might be needed to run PMP confinement operations and cope with the potential liability of confinement loss, skills that most farmers may not currently possess:

“. . . this is a new way to do business, and it's very different. And maybe even easier to come into this without a perspective of agriculture than to come in with a perspective. It's almost easier to train someone that's not run a combine before than it is someone who has run a combine. I, my own opinion, it takes a very sophisticated level of understanding to manage this kind of a production facility or operation. And Bill, I think, is leading the pack in this area, so he's the right person to talk to. But I would, guessing, he's going to say it's very different than what he does traditionally. So it's more of a new paradigm and it requires new skills.”

38

The Iowa corn farmer echoed Dr. Deeter’s assessment of the skills needed to write the standard operating procedures (SOPs) necessary for PMP crop production. Even with highly educated members, the farmer’s production team found writing good SOPs difficult:

“One of the fellows in our group is a process engineer from a large company here in the U. S. Another is a registered pharmacist, who actually -- he actually spent most of his time developing some GMP for animal vaccines. But anyway, we have a group of people like that. We sat down and wrote our standard operating procedures. And we thought it would be very easy. You know, I've tried growing crops for 29 years now. And I thought, well, this will be easy. I can write SOPs for field production. And, of course, as most of you know, it's very difficult to write SOPs.”

4.2 Government Subsidies to GM Crops and PMP

In order to arrive at a measure of the net benefits of PMPs to the citizens of a state such as North Carolina, the benefits to farmers, the PMP industry and consumers must be sufficient to outweigh any subsidies received by the industry. We will provide a few examples of financial subsidies offered to PMP firms producing corn and tobacco before focusing on the incentives offered by Missouri and Kansas to PMP rice-producer Ventria Biosciences. We conclude the section with a more in depth review of the potential incentives offered to PMP firms in North Carolina.

An Iowa state economic development fund gave the Prodigene Corporation $6 million in subsidies for PMP corn production, a wasted investment, given Prodigene’s PMP contamination incident in 2001-2002 (Des Moines Register 2003).

CropTech was a private Virginia corporation developing biopharm tobacco crops. In its 10-year existence, it received over $12 million in state and federal subsidies (Roanoke Times 2000a, 2000b). When it couldn’t stay afloat in Virginia, it sought financing from North and South Carolina, but filed for bankruptcy before it could accept South Carolina’s incentive package.

Ventria Biosciences and Northwest Missouri State University (located in Maryville, MO) signed an agreement in November 2004 calling for the university to build and equip a $30 million plant-sciences center in Maryville that would house Ventria and perhaps other biotech companies (St. Louis Post Dispatch 2005). At the

39

time, Ventria was considering similar offers from universities in Georgia, Louisiana and North Carolina (New York Times 2005c). Ventria planned to extract the proteins at a plant under construction at Northwest Missouri State University. Ventria agreed to move its headquarters from Sacramento, CA, to Maryville, MO. The university agreed to spend about $10 million on the site, and the state government agreed to contribute an additional $10 million. Ventria agreed to pay the university $500 per acre for crops grown on university land. Ventria agreed to grow 70% of its rice in Missouri, and Ventria agreed to pay farmers double what they would make on their next-most profitable crop. The university received a 4 percent share in Ventria (New York Times 2006b). University president Dean Hubbard helped raise $5 million in venture capital from private sources, money received by Ventria by the spring of 2005. (Hubbard later became a Ventria board member.) Although the growing season is generally considered too short for growing conventional rice in northern Missouri, Ventria planted four field test plots of several varieties of its PMP rice in the region in 2005. Twelve of the fourteen PMP rice varieties planted in field tests were successful, indicating that PMP rice could be grown in the region (New York Times 2006b). Ground was broken for the plant in the summer of 2005. However, delays in the approval of funding by the Missouri state legislature led Ventria to terminate its agreement. Ventria stated that it needed “processing facilities in place sooner than possible” for the Northwest Missouri State University to design and build the site. Legislative concern over the deal delayed a $10 million state contribution to the $23 million, 60,000 square foot plant (New York Times 2006b). The university also discussed with farmers the possibility of forming a separate company or co-op to own and operate the plant.

Reportedly, a Topeka, KS, economic development agency proposed $2.25 million in incentives to encourage Ventria to build its $10 million rice processing facility in Kansas (Sacramento Bee 2006). Private investors and a Kansas state biotechnology agency may add to the offer, according to the Topeka Chamber of Commerce.

Currently, the state of North Carolina supports many direct and indirect financial incentive programs to attract biotech industries, such as the PMP industry, to the state. While North Carolina is certainly not alone among states in providing financial incentives for economic development, and while there can be sound policy reasons for providing such incentives, it is nevertheless the case

40

that a proper accounting of net benefits must subtract the value of any subsidies from the gross benefits of industry economic activity.

Economic development subsidies can be classified as direct and indirect. Direct subsidies include grants and loans made by the state to a firm as incentive to locate in the state. Direct subsidies also include worker training paid for by the state. Indirect subsidies include state funding for university research, business incubator facilities, research institutes, and regional business park development (such as funding for utility extension to industrial parks).

In North Carolina, firms in the Biotech industry, including those in the PMP industry, have access to economic development programs available to all industries, such as worker training programs offered by the state Community College System, and grant and loan programs offered through the North Carolina Department of Commerce (NCDC), Commerce Finance Center. In addition, biotech firms have access to programs sponsored by the state-funded North Carolina Biotechnology Center and indirect support through biotechnology-related research funded by the state at two new biotechnology research centers located on university campuses, the Biomanufacturing Training and Education Center (BTEC) located at North Carolina State University and the Biomanufacturing Research Institute and Technology Enterprise (BRITE) located at North Carolina Central University. The biotech industry also benefits from financial support to BTEC and BRITE through the Golden Leaf Foundation, a foundation established to manage North Carolina’s share of the settlement funds from state lawsuits against the tobacco industry. The Foundation has provided funds for two key biotechnology education and training facilities located at NC State University and NC Central University as part of the North Carolina Biotechnology Center (see below). The Foundation also provided start-up money for the NC Community College System’s (NCCCS) BioNetwork(see below). The Foundation provided the following list of recent biotech GM crop and pharma-related grants:

Golden Leaf Grants:

1) Idealiance - $250,000 assistance with Targacept (Winston-Salem) to find uses of tobacco proteins in health related applications

2) East Carolina University - $250,000 GMP Training for Analytical Laboratory Workers in the Biopharma Industry

3) Pitt County Community College - $161,000 Associate Degree Program in Biotechnology targeting DSM Pharmaceutical expansion

41

4) Pitt County Development Commission - $311,000 assist with incubator upfit for the location of three Biopharma companies

5) Cleveland County Community College - $170,000 Bioinformatics and Specialized Biotechnology Training targeting job creation in healthcare and agribio related industries

6) Martin County Economic Development Commission - $400,000 incubator upfit to lease space for Quintiles (A contract research company providing a widerange of clinical research services for biotech and pharmaceutical clients.)

7) Pitt County Community College - $50,000 Industrial System/Biotechnology Training Clean room Project

8) Town of Holly Springs - $2,050,000 public infrastructure improvements necessary to locate Novartis (350 jobs)

In addition, the foundation has invested in NC venture capital organizations that make equity investments in the biotech and agri-pharma industries. The foundation was not able to reveal the names of businesses that have benefited from venture capital funding, only that Golden Leaf has made a significant investment in early and late stage financing to assist with capitalization needs.

Golden Leaf Foundation investments in Venture Capital Organizations:

o Hatteras BioCapital $30 million

o Aurora Ventures V &VI $10 million

o Carousel Capital III $6 million

A few biotech firms have benefited from NCCCS worker training programs (Meyer 2007). Bayer CropScience, which develops GM products, has participated in the NEIT program (see Appendix A for NEIT program description). Durham Tech Community College helped train workers in core work skills, including safety, management, supervisory, presentations, and team building at Bayer CropScience’s facility in Research Triangle Park, NC. Embrex, Maxton (Scotland Co.), makes an in-egg (in ovo) vaccine for the poultry industry. Embrex participated in the FIT program at Richmond County Community College, training workers in fork lift safety, stainless steel welding (for sterile environment) and a filtration and separation class (class involved actual process for making vaccine). However, it is not clear that either the Bayer CropScience or the Embrex training directly involves PMP development or production.

Ventria Bioscience is the only PMP firm currently operating in North Carolina. As far as can be determined from consultations

42

with NCDC, BTEC, BRITE and Golden Leaf foundation staff, and direct inquiries with Ventria BioScience, Ventria has received no direct subsidies to date. However, Ventria may benefit in the future from worker training program subsidies, should it decide to expand its agricultural production programs in the state, as agricultural workers would likely require substantial training to successfully implement the involved operation procedures required for PMP production. In addition, Ventria may benefit from subsidized university research and industrial park infrastructure should it decide to expand processing operations in the state.

Additional, detailed information on North Carolina’s economic development subsidy programs is provided in Appendix A.

4.3 PMP Health Risks in Intended Uses

Ventria Bioscience is currently conducting field trials of PMP rice in North Carolina, from which altered versions of the human milk proteins lactoferrin, lysozyme, and alpha-1-antitrypsin can be readily extracted. Ventria also produces human serum albumin, a blood protein used in medical therapies and cell culture, using PMP rice (Washington Post 2007), but it is unclear whether this product is grown in North Carolina. This review will focus on the potential human health risks of lactoferrin and lysozyme. There are potential human health risks associated with these altered proteins (Freese et al. 2004). These risks include: (1) An adverse response of certain bacteria which feed on the iron in human lactoferrin; (2) The potential for altered versions of lactoferrin and lysozyme to trigger allergic reactions and auto-immune responses in certain individuals; and (3) The potential for recombinant lysozyme to contain a mutation which causes hereditary systemic amyloidosis - a rare disease marked by slowly progressive renal impairment that can take decades to reach end-stage. Each of these potential risks will be discussed below. To the extent that these risks lead to actual injury or sickness with associated medical costs, these costs should be considered when assessing the net benefits of PMP products.

Lactoferrin is an important component of infection-fighting white blood cells that circulate in the bloodstream. Lactoferrin binds to free iron at infection sites, inhibiting the ability of microbes to feed on iron, so that they essentially starve to death. The potential uses for Ventria’s rice-derived lactoferrin include treatments for bacterial-induced diarrhea, topical infections, and antibiotics in poultry feed. It could also be used as an additive in baby formula

43

for treatment of severe diarrhea in children. There are a number of bacteria that feed on the iron in lactoferrin. This category of bacteria includes those which cause ulcers and stomach cancers, meningitis, whooping cough, legionnaires’ disease, gonorrhea, and other genital disease. Therapeutic use of human lactoferrin could stimulate growth of such pathogens, resulting in an adverse response. For example, while Weinberg (2001) believes that human lactoferrin has therapeutic potential, he also believes that it could stimulate growth of the gut bacterium helicobacter pyloris, which is implicated in causing ulcers, chronic gastritis, and certain forms of stomach cancer.

The recombinant protein lysozyme produced from pharma rice may contain the mutation which causes systemic amyloidosis. Because the sequencing of Ventria’s recombinant lysozyme is incomplete, this risk is yet to be determined. However, Ventria’s proteins may have the capacity to cause immune system dysfunction. There is a growing body of evidence demonstrating unexpected immunologic responses to biopharmaceuticals produced in genetically engineered cell cultures. In these cell culture production systems, a human gene encoding a medically useful protein such as insulin is spliced into bacteria or mammalian cells, which then produce a recombinant version of the protein, known as a biopharmaceutical. While the immune system does not normally attack a bodily protein because it is recognized as “self,” it may respond to the corresponding biopharmaceutical due to subtle differences that cause the body to recognize it as foreign. In some cases, the administered biopharmaceutical merely elicits an immune system response that reduces or eliminates the drug’s potency.

Altered lactoferrin and lysozyme are not exactly identical to the human versions of these proteins. The result of these differences is that they can trigger allergic reactions in certain individuals. A recent National Academy of Sciences report on the safety of new ingredients in infant formula notes that “. . . the commercial production of milk proteins using recombinant technologies may produce unintended side effects,” such as allergic reactions (NAS 2004). They can also lead to auto-immune responses in which the immune system responds to the body’s own proteins as if they are antigens, thus destroying or damaging normal body tissue. Lonnerdal (2002) states that recombinant rice-expressed lactoferrin and lysozyme are stable to digestion and heat, two properties widely regarded as characteristics of food allergens. Digestive stability is of particular concern in infants. Infants and young children are 3-4 times more likely to have food allergies than

44

adults, and sick infants are even more vulnerable (USEPA 2000). Ventria’s lactoferrin also has a “significant amino acid sequence homology to a known human allergen, bovine lactoferrin, an allergen found in cow’s milk.” (Lonnerdal 2002). In studies listed on Ventria’s website, only 11 amino acids at the N-terminals of the native and recombinant versions were sequenced and compared. While 10 of 11 of these amino acids were demonstrated to be identical, the other 119 amino acids were apparently not tested.

In 2002, a Ventria collaborator stated that rice-derived lactoferrin should be tested on rats and infant rhesus monkeys before testing on humans (Lonnerdal 2002). However, Freese (2007) finds that no tests on animals to assess potential effects on humans have been conducted to date. The methodology used in Ventria’s field study in Peru on the efficacy of its PMP proteins in treating diarrhea in infants has been criticized by Freese (2007).

Given the potential human health risks and lack of full information (e.g., trials on animals, clinical trials, drug interaction trials, etc.), it is not surprising that consumers hesitate to purchase, and consumer advocacy groups hesitate to endorse, PMP products.

4.4 Containment Loss and Potential PMP Liability Costs

In addition to the potential health risks of PMP products in their intended uses, the spread of GM and PMP crop cultivation creates the potential for new liabilities in the agricultural and food system (Smyth et al. 2004). PMP products intended for pharmaceutical or industrial applications may be unintentionally released from “containment” areas and contaminate non-GM crops or food. Containment areas are locations where PMP crops or products are isolated from non-GM crops or food. Loss of containment can occur during PMP field production, transportation, processing, or storage. The legal liability inherent in GM / PMP crop production is emerging as an important consideration in the potential profitability and adoption of these crops (Belcher et al. 2005).

45

4.4.1 Consumer Reaction to GM and PMP Products in Food

Because PMPs have not been commercialized, there has been little research to date on consumers’ willingness to accept PMP products and the effects of PMP-origin labeling on consumers’ choice of pharmaceutical products, or consumer reaction of PMP contamination of food products. In contrast, there has been a plethora of economic research on consumers’ reaction to GM food products and the effects of GM labeling on consumers’ choice of food. Since a major concern regarding PMP products is loss of containment, subsequent contamination of food crops or the food supply, and consumer rejection of contaminated food, the studies of consumer attitudes toward GM food shed some light on potential consumer reaction to PMP contamination of the food supply. Since PMPs are not intended as food and do not undergo the same pre-market food safety scrutiny as GM products intended as food, consumers may be more wary of PMP contamination than they are of GM food products; this an area for future research. Here we review the available research on consumer reaction to GM food.

Consumer attitudes tend to vary greatly from country to country, and even from region to region as well. In general, GMO food labeling has a large effect on whether or not consumers choose to buy GM foods. Foods that are labeled as “genetically modified,” are less desirable by consumers, in spite of potential benefits these foods may confer to their consumers. In the US, the labeling of GM foods is voluntary; no foods have been labeled as GM foods in the market, even though many food products do indeed contain GM ingredients. In other countries, such as those in the European Union (EU) and Japan, the labeling of GM foods for many products is mandatory. Under this regulatory environment, most if not all food manufacturers and retailers would not market any GM foods for fear of consumer resistance. Because GM foods labeled as such cannot be found in the marketplace, the extent of consumer acceptance of GM foods cannot be easily assessed. Thus, other methods, such as consumer surveys, are used.

Chern et al. (2002) conducted student surveys in Norway, Japan, Taiwan, and the US and they also conducted two pilot national telephone surveys in Norway and the US. These survey results reveal that American and Taiwanese students are more favorable to GM foods than Norwegian and Japanese students. Furthermore, the majority of students in all four countries supported a mandatory labeling of GM foods. The estimated

46

percentages of the willingness-to-pay for a premium of non-GM vegetable oil are 55-69% for Norwegian students, 50-62% for American students, 33-40% for Japanese students, and 17-21% for Taiwanese students. These results imply substantial premiums that consumers in all of these countries are willing to pay in order to avoid GM foods.

The pilot telephone surveys conducted in Norway and the US not only reinforce the findings obtained from the student surveys, but also provide more consistent data for a cross-country comparison. The surveys show that the Norwegian consumers were more concerned about GM foods than the American consumers. However, consumers in both countries showed strong support for mandatory labeling of GM foods and, in the case of salmon, were willing to pay for substantial premiums to avoid both GM-fed and GM salmon. However, the additional amount that students were willing to pay for non-GM salmon were considerably higher in Norway than in the US.

Fulton and Giannakas (2004) show that consumer aversion to GM products has important economic implications for GM crop introduction. This willingness of consumers to pay more for non-GM food affects food markets in the United States in a number of ways. Individual consumers in the United States, as well as food processing corporations such as Frito Lay and Taco Bell, are willing to pay a premium for non-GM products. Therefore, producers of non-GM products must pay part of the segregation costs, which cut into their producer surplus. Because the market for GM imports in Europe is virtually non-existent, U.S. exporters of non-GM crops must take great care to ensure that their products are not contaminated.

Onyango et al. (2005) showed that the use of choice modeling experiments provides a way of valuing non monetary attributes associated with consumption of GM food products and a way of identifying consumer preferences. The results indicate how product attributes of price, product benefits, and technology influence consumer demand for genetically modified food products. The results show how a consumer makes tradeoffs between the product attributes. The results suggest that health, environmental and production-related benefits have a positive effect on choice. Also, the results generally show that genetic modification is viewed negatively.

An indirect measure of consumer acceptance of GM products is the reaction of food importers to GM contamination events in the food supply. Presumably, food importers would stop purchasing

47

GM-contaminated food if they believed that their customers would not purchase it. By this indirect measure, foreign consumer acceptance of GM products is low. In 2004, Japanese customers said that they would not buy California rice if Ventria were allowed to produce PMP rice in the state (USA Today 2006). As a recent example, when it was announced in August 2006 that the U.S. rice supply was contaminated with GM rice, Japan banned U.S. rice imports, stores in Germany, Switzerland and France pulled American rice off the shelves and a ship transporting U.S. rice was quarantined in Rotterdam, Netherlands (Washington Post 2006).

In addition to direct studies of consumers and the indirect actions of importers, consumer advocacy groups also provide a gauge of potential consumer reaction to PMP products. While many advocacy groups do not oppose properly regulated and approved PMP products, many do oppose production of PMPs outdoors in food crop plants, including the American Consumers Union , the Public Interest Research Group, and the Center for Food Safety in the U.S., and GeneWatch in the U.K. (Daily Mail 2007).

The actions of some food processors and food industry groups are also an indication of potential consumer reaction to the incorporation of GM products in processed foods. For example, in March 2004, the North American Millers’ Association sent a letter to the USDA urging more stringent oversight of PMP crops, warning that “the risk of adulteration from genetic material” modified for pharmaceutical or industrial uses entering the food chain was, in its view, “unacceptable” (Los Angeles Times 2004). In 2005, Anheuser-Busch, the largest beer producer in the U.S. and the number 1 buyer of rice, threatened to stop buying rice grown in Missouri if some farmers were allowed to grow genetically modified rice there in field tests. The company reached a compromise after the state pushed the farmers to grow the GM rice 120 miles from other rice fields in the state (New York Times 2005b). More than half of Missouri’s rice is exported to European and Caribbean countries, where consumers are concerned about GM food. The vice president of Riceland Foods, one of the largest rice mills in the U.S. that markets more than half of Missouri’s rice, said “We are still having to make statements to our customers that the rice we export is not genetically modified. . . . We are concerned longer term that if Ventria and others get involved that will get harder to say” (New York Times 2005c). In comments filed on June 2, 2005 with the USDA, the Food Products Association (2005) expressed its strong opposition to the use of food crops to produce PMPs in the “absence of controls and procedures that

48

ensure essentially 100% protection of the food supply. . . .The use of food crops to produce materials not intended to be in the food supply must only proceed under systems proven to prevent any contamination or adulteration of the food supply . . . We cannot overemphasize the importance of the agency and development companies making an aggressive proactive effort to developing ‘fail safe’ systems that anticipate, rather than just react to mistakes. To date, effective control programs have not been demonstrated to our satisfaction.” The Grocery Manufacturers Association of America also opposes bio-pharming (USA Today 2006).

Some farm industry groups also oppose PMP production in food crop plants based on negative consumer reaction to potential loss of containment and mixing of GM and non-GM food crops.. With respect to Ventria’s PMP rice, Bob Papanos of the U.S. Rice Producers Association said in the spring of 2006 during the Liberty Link rice contamination incidents, “We just want [Ventria] to go away. This little company could cause major problems” (USA Today 2006). According to the USA Rice Federation (2007a), the announcement by USDA in August 2006 that trace amounts of Bayer CropScience Liberty Link rice entered the U.S. commercial rice supply resulted in “the closure of the European Union as a destination for U.S. rice.” In a spring 2007 a public comment to the U.S. Department of Agriculture regarding the agency’s consideration of Ventria’s application for approval to grow PMP rice in Kansas, the U.S.A. Rice Federation wrote: “If Ventria’s pharmaceutical rice were to escape into the commercial rice supply, the financial devastation to the U.S. rice industry would likely be absolute. There is no tolerance, either regulatory or in public perception, for a human gene-based pharmaceutical to end up in the world’s food supply” (New York Times 2007).

4.4.2 Food Market Reaction to GM/PMP Containment Loss

In the United States, persons who believe that their crop or property suffered damage from an activity involving a GM crop (by pollen flow, for example) could bring a tort liability suit under various legal presumptions, including trespass, negligence, or private nuisance (Kershen 2004). Tort claims based on strict liability (i.e., liability without fault) may also be possible, perhaps more so for PMPs and PMIPs than for first-generation GM products. A recent lawsuit in the Canadian province of Saskatchewan (Monsanto v. Schmeiser) involved a farmer who

49

grew and marketed a patented crop without paying the Technology Use Agreement (TUA) fee to a biotechnology firm (Belcher et al. 2005). The farmer claimed that GM canola arrived in his field through random cross-pollination and/or seed translocation, and thus he should not be liable for the TUA fee. In the end, the case was decided in favor of the biotechnology firm (SCC 2004). These legal developments show that farmers who produce GM crops could face a risk of litigation from neighboring farmers and biotechnology companies, as well as public institutions. Williams (2006, 2007) reports that in North Carolina, contract growers have not fully addressed the question of liability for food or feed mixups. Liability risk will likely be highest in the case of GM food plants used for PMPs which are grown adjacent to the same food plants without PMPs, which likely explains Ventria’s decision to avoid field trials in California, despite the California Rice Commission’s approval of Ventria’s plan for commercial-scale planting of PMP rice in the state (Los Angeles Times 2004). (Greenpeace also picketed a Ventria rice field in Sutter County, California, in the summer of 2001, San Francisco Chronicle 2002).

The financial liability to the farmer associated with losing containment and losing the investment in PMP-specific farm equipment is significant. When asked about the importance of maintaining containment to the liability of farmers producing PMPs in the United State, a farmer (and former president of the Iowa Corn Growers Association) growing corn PMPs in Iowa for the French company Meristem Therapeutics reported in a USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003):

“So it's absolutely critical that we have perfect containment. We have no -- no tolerance for mistakes in our system. That's why almost all the requirements that we have from the regulatory folks, we double or triple, because our family farm, my family, our business, is in jeopardy here if we screw up. And it's not like if I work at a biotech company, whereas if there's a mistake, I can just go to the next company and get a job. If I have a mistake, I'm done.”

“. . . the group that we work with is a hand-picked group of producers. It's 72 members in Iowa, and right now, my brother and I are the only ones producing [pharma crops], but the other 71 would certainly like to at some point in the future. But you need to know exactly who you're working

50

with, and you have to trust them. Because all it takes is one mistake by one individual and you've collapsed the whole system.”

Although rice is not grown for food in North Carolina, future PMP products grown in food crops such as corn, soybeans or tobacco could create these kinds of liabilities if the PMP crop planted by a North Carolina farmer was proven to cause an adverse effect on his neighbor’s crop. If this were to occur, the PMP producer could be personally liable for damages. If PMP rice is grown in North Carolina for Ventria Biosciences, and the rice is transported to Ventria’s proposed facility in Junction City, Kansas (Ventria Biosciences 2006), then there is the additional risk of loss of containment during transportation along hundreds of miles of highway or rail line. Which party--farmer, transportation company, biotech company, etc.—would be responsible for the financial liability arising from loss of containment could be a complex and expensive legal issue, involving complex contracts and significant legal fees. Liability insurance could be a significant barrier to entry for farmers seeking to grow PMP crops, yet this remains an unanswered question for independent growers.

Financial liability and risk is exacerbated by international trade in GM and potentially PMP crops. Export market access may be restricted when a GM trait is found in the U.S. supply (even if present in trace levels). In such a case, individuals could bring liability suits based on the concept of public nuisance. Episodes of accidental GM contamination support this conclusion, including the high-profile StarLink and Prodigene cases. In the StarLink case farmers claimed economic damages arising because the introduction of GM crops affected export market access and market price for their non-transgenic crop (Moschini 2006). In the late 1990s, Aventis CropScience, a multinational French-based corporation, introduced StarLink corn into the United States even though the EPA had not approved StarLink corn for human consumption. On September 18, 2000, the Cry9C Starlink gene was found in sample of Taco Bell taco shells. Kraft Foods, Inc, the producer of Taco Bell taco shells, recalled the product after further testing by the FDA. Many other food products were also recalled because of the presence of StarLink corn, including other corn-based taco shells, tostada shells, tortillas, tortilla chips, and chili seasonings kit. Due to the recalls, Japan and South Korea

51

temporarily halted US corn imports until testing procedures were firmly established (Schmitz et al. 2004).

StarLink corn is toxic to European corn borers and certain other insect pests. By 2000, StarLink corn was grown on approximately 362,000 acres, roughly 40% of which were in Iowa. StarLink corn became commingled with non-StarLink corn in the US grain-handling system. According to Lin et al. (2003), commingling of StarLink corn with other corn varieties was exacerbated by three factors: (1) some of the corn grown on the buffer zone was probably cross-pollinated with StarLink corn; (2) a portion of the StarLink corn, including that grown on the buffer zone, had entered the marketplace prior to the effort to contain StarLink-commingled corn; and (3) some elevators did not know they were receiving StarLink-commingled corn. In order for US corn to be sold for food purposes both in the United States and in major importing countries, it now had to be segregated and tested.

Once implemented, StarLink testing became highly stringent—the tolerance level ranged from one kernel in a sample of 400 in the US to as much as one kernel in three samples of 800 in Japan. Several class-action lawsuits ensued, Fingers et al. v. Kraft Foods North America, Inc., et al. was one such case. The plaintiffs claimed they had allergic reactions to food containing Cry9C. The Centers for Disease Control tested the 17 people who claimed StarLink had made them sick and found that none of them had antibodies consistent with allergic reactions to StarLink. Despite these results, a federal judge approved a $9 million settlement in March 2002. Mulholland et al. v. Aventis Crop Science USA Holding, Inc. was another such case. The plaintiffs, who were non-StarLink corn growers in seven Midwestern states, claimed property damage and corn loss claims. Property damage claimants were compensated for lost market value, transportation, and storage costs resulting from actual contamination of their crops, fields, equipment, and property. Corn loss claimants were compensated for the alleged reduction in the general price of corn due to the presence of StarLink corn in the US corn supply. A settlement for $110 million was reached in February 2003.

The latter case is important since it is applicable to ongoing debates over the acceptance and impacts of genetically engineered crops both within the United States and abroad. The StarLink incident illustrates the complexity of isolating crop varieties within the existing grain marketing system and preventing unwanted commingling. Schmitz et al. (2005) performed an analysis of the impact of StarLink corn on US producers and found that US

52

producers lost between $26 million and $288 million in the first year following the StarLink incident. However, if it were not for US corn subsidies, the full economic damages would have been in the range of between $298 million and $964 million (Schmitz et al. 2005).

In 2002, Monsanto applied for environmental safety approval of its Roundup Ready wheat for use in Canada. However, the Canadian Wheat Board threatened to sue Monsanto if its genetically engineered wheat passed the crop varietal registration process in Canada. As a result, Monsanto withdrew from its GM wheat program in North America in May 2004. Two economic studies were conducted to analyze the effect that decision had on producers and consumers in various countries. Berwald et al. (2006) concluded that if the United States and Canada would have allowed Monsanto’s wheat to be grown, consumers in both countries would have benefited. On the other hand, Furtan et al. (2005) concluded that, under the same circumstances, wheat producers would actually lose money, while both consumers and the biotech company would benefit. In July 2004, the USDA determined that Monsanto’s GM wheat was safe for human and livestock consumption. However, this approval came after Monsanto decided to put an end to its GM wheat research program.

Ventria’s Vice President Delia Bethell noted in a 2003 petition to the FDA that Ventria’s PMP rice could inadvertently enter the commercial food rice supply (Ventria 2003). GM pest-resistant rice grown experimentally in China allegedly made its way illegally into Chinese seed markets, sold by a Chinese agricultural university specializing in GM rice research (New York Times 2005a, 2005b). It could be argued that the U.S. regulatory system is significantly different from China’s, but the StarLink corn and Liberty Link rice experiences (see Section 4 of this report) would appear to indicate that the U.S. system also has difficulty keeping GM products out of the food supply.

In 2004, the National Research Council released a report on the biological confinement of genetically engineered organisms (NRC 2004a). The NRC found that an organism that is typically grown to produce a common and widespread food product, such as Ventria’s rice, probably would be a poor choice as a precursor for an industrial compound unless that organism were to be grown under stringent conditions of confinement. This is an important issue for any novel compound or GM organism for which zero tolerance of bioconfinement failure is needed. Engineering organisms that are

53

not otherwise used for food or feed could be an effective way to prevent a transgenic compound from entering the human food chain. The NRC report recommended that “Alternative nonfood host organisms should be sought for genes that code for transgenic products that need to be kept out of the food supply.”

In a presentation to a USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003), Dr. Scott Deeter, president and CEO of Ventria Bioscience, described why Ventria is using rice, instead of a non-food crop, to produce PMPs:

“One of the questions, why do we work with food crops? Well, there's a lot of reasons, actually. It's not something that we decided up front and said, well let's use food crops. In fact it was because of the biology of food crops and the science and the platform there. So grains can store -- have a natural protein storage mechanism. They also are free of infectious contaminants such as prions. So they're not animal source. I don't know if you remember last year, last summer, where we dealt with the West Nile virus and the blood supply. Well, that's because the virus was essentially transmitted to new -- new individuals through the blood. Through the blood supply. Because that blood comes from an animal origin, from a human origin, in fact. Grains don't carry the viral vectors that humans do. So that provides a benefit in terms of manufacturing. These crops are generally regarded as safe. What does that mean? Well, a lot of the production systems that are in use today by the biopharmaceutical industry, if, for instance, one is E. coli, and we don't hear a lot about that, and there's probably good reason for why we don't hear about it. Nobody really wants to think of their drugs being produced in E. coli. However, grains -- so if we produce in E. coli we've got to do enormous amounts of processing and purification to get the active ingredient out. If we produce in grains, we have a host that is basically safe. And if the only difference here is the active ingredient, and we show that the active ingredient is processed in a way that provides safety and efficacy through the clinical process, we can envision a time where we don't need to go through the purification processing to the extent that we have to today in the biotechnology world. And that is a big chunk of the total cost of pharmaceuticals. So there's another benefit there, especially for orally or topically delivered products.”

Ventria’s potential liability costs for losing containment of its PMP products could be very large. Consider the example of Aventis Crop Science and Bayer CropScience. In the spring of 2001, Aventis Crop Science destroyed GM “Liberty Link” rice that was

54

being grown experimentally in Texas after its GM StarLink corn contaminated the human food supply, resulting in Aventis paying a multi-million dollar settlement to affected parties (Fortune 2007). When Aventis’ crop science unit was later sold to Bayer CropScience, Bayer CropScience dropped plans to bring Liberty Link rice to market. These firms presumably dropped plans to produce Liberty Link rice due to the potential risk of loss of containment and the possibility it could enter the human food supply. Nonetheless, in 2006 Liberty Link rice somehow made its way into the U.S. commercial food rice supply, perhaps from a research test plot at Louisiana State University. According to Jeffrey Barach, a vice president of the Food Products Association, “Once [GM rice] is in the pipeline, it’s very hard to get it out” (Washington Post 2006). Four hundred rice growers have filed a federal lawsuit against Bayer CropScience. Ten rice seed dealers from Arkansas, Missouri and Louisiana have also sued Bayer CropScience, alleging that the company’s carelessness ruined their seed. Tilda, a British rice importer, has sued Bayer CropScience, Riceland Foods and Producer’s Rice Mill, saying it had to destroy or send back Arkansas rice. The North American Millers’ Association has called for “Mandatory liability insurance coverage to indemnify all downstream traders, handlers, processors and food manufacturers for the full cost of recall, destruction and brand degradation as a result of gene flow or other release of genetic material into the food or feed industries” (NAMA 2007).

4.4.3 NRC Recommendations to Reduce GM/PMP Liability Costs

The National Research Council recently evaluated the safety of

genetically engineered foods (NRC 2004). The NRC report noted that because most crops can produce allergens, toxins, or anti-nutritional substances, both conventional breeding methods as well as genetic engineering have the potential to produce unexpected and unintended changes in a food crop. Hence, the issue of food safety with respect to GM and PMP crops is a matter of degree rather than kind. Furthermore, the NRC found that no adverse health effects in the human population attributed to genetic engineering had been documented. However, the NRC found that potential food safety risks exist, and the recent advent of genetic engineering science and GM and PMP crops likely mean that consumers are less familiar with these risks. The NRC report considered ways to reduce safety risks before product

55

commercialization (pre-market) and after commercialization (post-market).

Current safety assessments in the pre-market period focus on comparing the GE food with its conventional counterpart to identify uniquely different components. The NRC found that despite technological advances in analytical chemistry, our ability to interpret the consequences to human health of changes in food composition is limited due to the complexity of food composition. Although current analytical methods can provide a detailed assessment of food composition, limitations exist in interpreting their biological significance. That is, we have limited abilities to interpret the analytical results and predict health effects. The NRC report concludes that “The knowledge and understanding needed to relate such compositional information to potential unintended health effects is far from complete . . . Furthermore, currently available bioinformatics and predictive tools are inadequate for correlating compositional analyses with biological effects.” The NRC recommended that “genetic modification in food, including genetic engineering, undergo an appropriate safety assessment. . . . prior to commercialization.” The NRC report also concludes “Although the array of analytical and epidemiological techniques available has increased, there remain sizeable gaps in our ability to identify compositional changes that result from genetic modification of organisms intended for food; to determine the biological relevance of such changes to human health; and to devise appropriate scientific methods to predict and assess unintended adverse effects on human health.”

Post-commercialization or post-market evaluation tools, including tracking and epidemiological studies, are important components of the overall assessment of food safety. Post-market surveillance is a common procedure, for example, with new pharmaceuticals and has been beneficial in the identification of harmful and unexpected side-effects. The NRC recommended that “When warranted by changes such as altered levels of naturally occurring components above those found in the product’s unmodified counterpart . . . [post-market surveillance] Improve[s] the ability to identify populations that are susceptible to food allergens and develop databases relevant to tracking the prevalence of food allergies and intolerances in the general population, and in susceptible population subgroups.” This [is] especially pertinent to GE foods because of the unique ability of GE processes to introduce gene sequences that generate novel products that are incorporated into organisms intended for use as food and especially in situations where the novel products are

56

introduced at levels that have the potential to alter dietary intake patterns (e.g., elevated levels of key nutrients). The NRC recommended that we “Develop a database of unique genetic sequences (DNA, polymerase chain reaction sequences) from GE foods entering the marketplace to enable their identification in post-market surveillance activities.” Although post-market surveillance has not been used to evaluate any of the GE crops that are currently on the market and there are challenges to its use, this approach holds promise in monitoring the potential effects, both anticipated and unanticipated, of GE foods.

4.5 Externality “Spillover” Costs Affecting Non-GM and Organic Farmers

As with any new product, the introduction of plant-made pharmaceuticals has the potential for unintended, negative consequences. Any such negative side effects are generally termed ‘externality,” or “spillover,” costs by economists, because the costs affect someone other than the producer and consumer of the product in question—the costs “spillover” onto someone “external” to the transaction between the product’s producer and consumer (Nelson and De Pinto 2001). With respect to PMPs, potential externality costs fall into two broad categories, costs imposed by PMP production on non-PMP and organic farmers, and environmental damage costs.

PMP production imposes externality costs on non-GM farmers, such as organic farmers. These “spillover” costs include the increased costs of product identification and certification necessary to gain entry into markets banning some or all GM products. Exporters of non-GM crops need to assure the importing country that their product meets importing country regulations. Testing each unit of product as it crosses international borders or at the point of sale to the final consumer is either prohibitively expensive or not technically feasible. Recently, firms have begun to develop new quality-assured supply chains using various segregation, identity-preserving production and marketing (IPPM), and traceability systems to bridge the gap between heterogeneous country regulations and the traditional agricultural food production systems that produce and deliver pooled, homogeneous commodities (Smyth et al. 2004). For example, maize and soybean producers have tried to develop and deliver quality assured GM-free grains to Europe and Japan (Lin 2002), beef producers are

57

developing quality assurance and trace-back systems (Spriggs and Isaac 2001), and canola producers and processors in Canada have developed IPPM systems to manage more than 39 new varieties (Smyth and Phillips 2003). However, even if these systems achieve their goals, they will be costly, and it is the existence of the GM and PMP products that creates the need for these systems. If a farmer chooses to grow GM-free products and sell them to GM-free markets, he must bear part of the costs of certifying that his products are GM-free. These costs can include the extra costs of segregating and tracking the crop during storage, transportation and handling, additional testing and marketing costs, and opportunity costs of missing the best times to sell the product due to processing and transportation delays and bottlenecks (Table 6). These costs are necessitated by the developers and growers of GM products, including PMP products. Hence, the “spillover” costs that GM-free farmers must incur to certify that their products are GM-free should be counted as a cost of GM crop production (Moss, Schmitz and Schmitz 2007, 2004). Most existing studies of IPPM costs are based on niche markets with low volume, where estimated costs are C$30-40/tonne, or 15% to 20% above the cost of conventional supply systems (Smyth et al. 2004) (Table 7). Some believe that costs would fall with larger production volumes, while others believe that there are too many constraints (e.g., storage, trucking) in the system for IPPM to be feasible at larger scales. Lin and Johnson (2004) estimate the cost of segregating non-GM maize and soybeans at 12% of the average farm price.

The prospect of PMP products raises a number of issues germane to the debate on the coexistence of GM and non-GM agriculture, including organic agriculture. Because many of these products are being pursued with host plants that are also used in food and feed production (such as maize and rice), the possibility of adventitious presence of PMP and PMIP traits in the food and feed system becomes a real possibility. The two major avenues by which this could happen are pollen and seed dispersal. Pollen is dispersed mainly by wind and insects. Seeds are dispersed by wind, animals and human activities associated with seed production, harvest, transportation, storage, and processing.

Preliminary research in plant biology has indicated that it will likely take longer than a year for a non-GM or organic grower to be able to meet the existing standards for non-GM markets on a field contaminated by a GM version of the crop (Gilligan et al. 2003, Norris et al. 1999). The length of time necessary for crops to meet more rigorous standards, such as organic certification, after a field has been contaminated, will likely be even longer. Hence, the costs

58

to non-GM farmers of GM contamination of their fields could be at least a year’s worth of production, if not more. Indeed, seeds from GM crops have produced volunteer populations for up to 3 years (Norris et al., 1999). The potential loss of containment imposes identity preservation costs on other farmers not growing PMP crops. For example, certified organic farmers in North Carolina pay for a quality-control process in which the food is routinely tested for the presence of genetically modified DNA (Williams 2006, 2007).

As discussed in Section 2.1 above, on May 4, 2007, a federal judge in San Francisco ordered farmers to stop planting Monsanto’s GM Roundup Ready alfalfa seed because of the risk that it will contaminate nearby non-GM, organic alfalfa fields (Sacramento Bee 2007). This ruling is significant in that it was the first time that GM crop planting was stopped due to the potential for, rather than actual, containment loss. The judge criticized USDA for failing to adequately assess potential problems with cross-pollination before approving the alfalfa seed for commercial planting. The judge ruled that contamination of an organic alfalfa field with the Roundup Ready gene could effectively destroy the organic farmer’s crop.

4.6 Externality “Spillover” Costs of Environmental Hazards

Since GM organisms were introduced into the environment nearly 20 years ago, questions have been raised about the consequences of the escape of those organisms and their engineered genetic material––transgenes––into natural and managed ecosystems. A National Research Council report (2002) on the environmental effects of transgenic plants found that although the transgenic process presents no new categories of risk compared to conventional methods of crop improvement, specific traits introduced by both approaches can pose unique risks. The report found that both transgenic and conventional approaches (for example, hybridization, mutagenesis) for adding genetic variation to crops can cause changes in the plant genome that result in unintended effects. Genetic improvement of crops by both approaches typically involves the addition of genetic variation to existing varieties, followed by screening for individuals that have only desirable traits. The screening component will remove many

59

but not all of the unanticipated physical and ecological traits that could adversely affect the environment.

More specifically, the NRC report found that (1) small and large genetic changes have had substantial environmental consequences; (2) the consequences of biological novelty depend strongly on the specific environment, including the genomic, physical, and biological environments into which they are introduced; (3) the significance of the consequences of biological novelty depend on societal values; and (4) introduction of biological novelty can have unintended and unpredicted effects on the recipient community and ecosystem.

A growing body of research has considered the environmental impacts of introducing GM crops into an agricultural landscape (Batie and Ervin 2001, Barton and Dracup 2000). The NRC (2002) report on the environmental effects of transgenic plants identified four primary categories of potential hazards from the release of transgenic crop plants: (1) hazards associated with the movement of the transgene itself with subsequent expression in a different organism or species, (2) hazards associated directly or indirectly with the transgenic plant as a whole, (3) non-target hazards associated with the transgene product outside the plant, and (4) resistance evolution in the targeted pest populations. The potential indirect effects of transgenic crop plants on human health as mediated by the environment constitute a fifth category of hazard discussed in the NRC report. In addition to these categories, the EU recognizes altered genetic diversity as a separate category of environmental hazard in its modified directive 90/220 (European Commission 2001). The NRC does not recognize altered genetic diversity per se as an environmental hazard because the NRC considers the effects of altered genetic diversity, such as increased extinction rate, a compromised genetic resource, inbreeding depression, or increased vulnerability to environmental stresses, to be the actual environmental hazard. In the NRC classification scheme, the effects of altered genetic diversity are addressed under transgene movement effects. The EU recognizes altered genetic diversity as a distinct category of potential effect as a precautionary measure, because the effects of transgene movement are uncertain and are presently incompletely characterized. By recognizing the more easily measured, intermediate effects on genetic diversity as a potential hazard, the EU believes its risk analysis will address and manage all of the effects the committee lists under movement of transgenes without having to assess them specifically.

60

Environmental hazards can be divided into two categories—short and long-run (Nelson and De Pinto 2001). Short-run hazards, such as effects on non-target organisms (such as the monarch butterfly, see Losey et al. 1999), can usually be stopped or reversed by ending use of the GM crop. In contrast, long-run costs, such as the development of pest resistance or enhanced survivability of the crop, making it a weed, usually develop slowly and are not reversible.

Short-run environmental costs typically consist of negative impacts on non-target organisms. For example, although the target species of Bt corn are a few types of lepidopteran insects, other related, non-target species are killed as well, such as the monarch butterfly (Losey et al. 1999). In addition, other species that rely on target or non-target affected species as a food source may be negatively impacted as well.

Long-run environmental costs include the development of resistance in insect and weed pest species, increased weed characteristics of the crop itself due to enhanced survivability, and negative impacts on non-target species due to gene flow (sexual transmission of genetic material from one species to another) from the GM crop to other species (Nelson and De Pinto 2001). For example, increased resistance to Bt pest control toxin is a potential problem in insect pests of GM crop Bt corn such as the European corn borer (Huang et al. 1999). Farmers have attempted to slow the development of insect resistance by planting refugia (crop areas without GM crops), but farmer participation in refugia programs has been low due to the added management burden (Nelson and De Pinto 2001). In addition to insect resistance to pesticides, some weeds have developed resistance to herbicides as a result of planting herbicide-resistant GM plants such as glyphosate-resistant soybeans in combination with high herbicide dosage (Nelson and De Pinto 2001).

Increased weediness of the crop itself appears to be problem so far only for canola (Nelson and De Pinto 2001) and creeping bentgrass (National Public Radio 2006). GM crop Roundup Ready Creeping Bentgrass escaped from test plots in central Oregon and was discovered growing in the wild 21 kilometers from its Oregon test plot in 2004. This was the first know occurrence of a GM-plant escaping and surviving in the wild. The wind-pollinated, perennial crop has weedy relatives and was found to have already hybridized with wild plants. Grass seed farmers in the Willamette Valley region of eastern Oregon are worried that contamination of their fields could jeopardize their $374 million business.

61

Concerns related to gene flow include movement of herbicide and virus-resistance genes from GM crops to wild relatives and flow of antibiotic resistance marker genes to microorganisms in the stomachs of animals that eat GM crops. If a herbicide- or virus-resistant GM crop passes resistance to wild relatives, weed problems can be exacerbated. Canola, for example, is grown in regions where weeds with some degree of sexual compatibility exist (Nelson and De Pinto 2001). Other crops with some potential for gene flow include sugar beets, cucumbers and squashes.

Antibiotic resistance marker genes are used to indicate the successful insertion of novel genetic material into GM plants. The marker gene is linked to another gene that produces the desired GM trait, and both genes are inserted into the potential GM plant. If GM plant cells grown in the laboratory are not killed by certain antibiotics, it is an indication that the gene insertion was successful. Two potential problems related to the used of antibiotic resistance marker genes are the inadvertent interference with therapeutic antibiotics and genetic flow to microorganisms (Nelson and De Pinto 2001). Antibiotic resistance genes are in the food supply now. For example, in 1993, Calgene asked the Food and Drug Administration (FDA) to evaluate the product of the kanamycin resistance gene in the FlavrSavr tomato as a food additive. The FDA decided that the gene product was safe as an additive in both food and feed because the frequency of gene transfer was very low, and the natural incidence of resistance genes in the environment is much higher.

A major concern regarding the co-existence of GM and non-GM crops in agricultural landscapes is gene transfer through cross-pollination of non-GM crops by nearby GM crops (Belcher et al. 2005, Haslberger 2001) and the biocontamination of non-GM crops through seed translocation. There already exists some research into the potential for gene transfer to occur (Rieger et al. 2002, Gilligan et al. 2003, Hucl and Matus-Cadiz 2001). This research indicates that gene transfer can occur in most crops, and that transfer is influenced by factors such as pollination mode, genetic stability, fitness and fertility of hybrids, and proximity and size of crop populations (Barton and Dracup 2000).

The USDA (2005b) discussed the potential for rice cross-pollination in the Environmental Assessment of Ventria’s proposed PMP rice plantings in Missouri and North Carolina:

“Rice is not sexually compatible with plant species outside of the Oryza genus. There are no sexually compatible species of Oryza other than Oryza sativa growing in the United States.

62

Rice is primarily self-pollinating and out-crossing rates usually occur at a very low rate (generally less than 1%). The floral structure of O. sativa and the short viability of its pollen present biological barriers to cross-pollination (Gealy et al. 2003). A rice floret opens only once for a short period of time, usually for a little over an hour or less, during which fertilization can occur. Pollen viability is for no longer than five to ten minutes, but the stigma can remain viable for two to four days and can be fertilized by foreign pollen if for some reason it is not fertilized by its own pollen (Gealy et al. 2003). Due to the high self-pollinating characteristic of rice, the Association of Official Seed Certifying Agencies (AOSCA) certified seed regulations for foundation seed require a minimum isolation distance from other rice varieties of at least ten feet when ground drilled and 50 feet if ground broadcast (AOSCA 2003). With proper isolation distances maintained between Ventria’s rice and other cultivars of rice, gene escape via crosspollination would be highly unlikely. Temporal isolation can further reduce the likelihood of effective pollination and fertilization. In addition, another mechanism for gene escape would be out-crossing with weedy/red rice. The establishment of a weedy rice population next to the field site could offer a means of escape of the gene from the production area. Since red rice seeds often have dormancy and shatter easily, the gene could be harbored in a weedy population for a number of years.”

Hence, although the likelihood of cross-pollination / outcrossing is low, nearby populations of weedy (uncultivated, volunteer) populations of red rice could become contaminated and serve as a harbor for an escaped gene for a number of years.

Unintended gene flow occurred in some varieties of Mexican maize in the fall of 2001 (Quist and Chapela 2001). In the summer of 2002, Snow et al. (2003) reported evidence suggesting that a trait from transgene insertions may be able to move to other plants, creating the conditions for potential “superweeds.” However, gene flow is much less likely in rice, due to its self-pollinating characteristics.

In a presentation to a USDA Biotechnology Advisory Committee meeting in 2003 (USDA 2003), Dr. Scott Deeter, president and CEO of Ventria Bioscience, described the methods Ventria would use to limit gene flow:

“Now, there are several technologies that are being developed from male sterility of the pollen to switches to

63

other technologies that are being employed to manage what I'll generally call as gene flow, okay? And this could be gene flow either direction. It could be unwanted gene flow into your pharmaceutical production crop, or gene flow outside of your pharmaceutical production crop. The food industry is concerned about the latter. The pharmaceutical industry is concerned about both. So there, in Ventria's perspective, this is a critical question for this industry. And this is why we've chosen two self-pollinating crops. So self- pollinating crops mean that the male and female reproductive systems are contained within the plant. They aren't designed for pollination in the wind or through insects. A plant is designed that way. So we've -- we've developed our system around plants that have self- pollinating features. Like rice, barley and wheat are three that do. We use rice and barley. The reason for this is -- has to do with the -- there's several studies that go back. We know a lot about the biology of rice. We know a lot about biology of barley. These studies, the rice bio-safety study, the U. C. Davis study, the bottom line of those studies is beyond 30 feet, they seem to let no out-crossing. The out-crossing rate of beyond six feet i s, in most studies, as close to zero as you can get. So we use a 100-foot setback, which isn't more than 3 times what we see in any of these studies.

“. . . So we grow our pharmaceutical crops under a permit issued by USDA. We directly manage using standard operating procedures. Our field personnel are trained, or we maintain ownership of the seed and grain, as I said. We use a hazard analysis, critical control point philosophy, so we manage each step of the way for quality of our own product, but also for environmental stewardship purposes. And we process these products before we then go on to sell the active ingredient. We do 100 percent internal audit of our standard operating procedures to make sure that our field personnel follow the procedures. We do that through interviews as well as through random audit. The -- we have a dedicated field production. We have dedicated field production equipment. Anything that comes into contact with plant material is dedicated for pharmaceutical production. It's not used for any other type of production. We have -- that includes harvesting, storage and processing. Initial processing up to the stage where it gets converted into a flour form. We use double-contained transportation, which is a requirement by

64

USDA. And we have -- we have third parties. In addition to USDA, we use the California Crop Improvement Association to audit our -- that our field production practices achieve the result that we're after, which is no out-crossing. That's what -- this is what we do. And I know that many of the companies that are involved in our industry follow very similar approach.”

It is unclear whether Ventria uses third parties to audit its field production practices in North Carolina. The National Research Council report (2002) on the environmental effects of transgenic plants recommends “that two different types of general ecological monitoring be used to assess unanticipated or long-term, incremental environmental impacts of transgenic crops. One type of monitoring involves use of a network of trained observers to detect unusual changes in the biotic and abiotic components of agricultural and nonagricultural ecosystems. The second involves establishment of a long-term monitoring program that examines the planting patterns of transgenic plants, and uses a subset of species and abiotic parameters as indicators of long-term shifts in an ecosystem.” It is not clear whether Ventria has established such monitoring programs for its North Carolina PMP rice.

An additional route for containment loss is the possibility of spills or other mishaps during long distance transportation of PMP rice from fields to processing facilities. The PMP rice currently grown in North Carolina must be taken to Iowa or Missouri for protein extraction (Washington Daily News 2006a). With the announcement (Ventria Biosciences 2006)in 2006 that Ventria plans to locate its PMP rice processing facility in Junction City, Kansas, any PMP rice grown in North Carolina for Ventria Biosciences would presumably be shipped to Kansas for processing, presenting an additional risk of loss of containment during transportation along hundreds of miles of highway or rail line.

A recent National Research Council report NRC (2004a) discussed the possible use of bioconfinement methods to confine certain GM organisms and their transgenes to specifically designated release settings. All bioconfinement methods can be conceptually divided into three general categories: those that reduce the spread or persistence of GM organisms; those that reduce unintended gene flow from GM organisms into other organisms; and those that limit expression of transgenes. Most of the bioconfinement methods discussed in the NRC report are

65

equivalent to natural mechanisms of reproductive isolation that act to maintain species barriers. However, the NRC noted that, in plants, the leakiness of those species boundaries is well known. Within species, distinctive breeding systems such as dioecy (male or female plants) and self-incompatibility also are known to be leaky. Moreover, experience suggests that sterility is rarely absolute. Thus, the NRC found that “in most circumstances, single-method efforts at bioconfinement are likely to be less than 100% effective in preventing the escape of transgenes, especially if large numbers of plants are involved.” Although the efficacy of some of the approaches is known, most are untested. Failures in the bioconfinement of GM organisms have not been documented to date, in part because so few methods have been implemented. However, the NRC found that “given the imperfections of methods under development and those of methods that have been applied to nonengineered species, it is likely that failure will occur.” The NRC found that “Current methods for detecting and culling individual GM organisms after a bioconfinement failure are very limited, and they depend on the organism and scale of the original release of the GM organism.” Currently, monitoring is difficult because it involves searching for what often will be a rare event over a potentially large area. Ideally, monitoring methods would be developed that could identify escapes with remote sensing.

With respect to any potential impacts of U.S.-grown PMP crops on the environment outside U.S. territory, the National Research Council report (2002) on the environmental effects of transgenic plants found that “APHIS’ jurisdiction and the focus of its Environmental Assessments are confined to the United States, but some APHIS assessments discuss potential environmental effects of specific transgenic plants outside the United States.” In effect, any potential environmental hazards outside the U.S. due to U.S. PMP production are not considered by U.S. regulators.

66

5 Alternatives to Food Crop PMPs

Neither food plants nor farmers’ fields are necessary for the production of PMPs (Williams 2006, 2007). PMPs can be grown in sterile containment systems instead of agricultural fields. At this time, pharmaceutical firms place a premium on uniformity and purity of recombinant protein to such an extent that containment systems are preferred. A major drawback of field crop PMPs is that protein content is variable from crop to crop. Recombinant proteins produced in containment systems are more uniform. In addition, proteins produced in containment systems are free of residues from herbicides, pesticides and fungicides. Contamination risks to food supplies are greatly reduced. Proteins exuded via roots of genetically modified plants and harvested from the container’s aqueous media offer some processing advantages. Two disadvantages of producing proteins in containment systems are that it is thought to cost more, and it is thought to take longer to bring the product to market. However, Agres (2006) reports that recent advances in closed-system technology have eliminated some of the cost difference between PMPs and contained cell culture systems.

The range of molecular farming systems using contained, non-field systems continues to expand. Some commercial systems for containment include non-food plants such as duckweed (Lemna spp.) (Cox et al. 2006), tobacco (Nicotiana spp.) (Somerville and Bonetta 2001, Poirier et al. 1992), algae (Chlamydomonas reinhardtii) (Franklin and Mayfield 2004) and moss (Physcomitrella patens) provide good choices for containment systems (Streatfield 2005, Fischer et al. 2004, Gasdaska et al. 2003). Agennix is producing lactoferrin in a closed, fermentation system using Aspergillus niger (a filamentous fungi) for treatment of lung cancer and diabetic food ulcers (http://www.agennix.com). See also Freese (2002, Appendix 5) for additional information on alternatives to PMP food crop production. Another option is to produce PMPs using food crops inside greenhouses, such as the potatoes grown hydroponically in greenhouses by AltaGen Bioscience (San Francisco Chronicle 2002).

At present, biotechnology firms based in North Carolina produce PMPs using containment systems based on non-food plants rather than food crop plants in un-contained farm fields. High product purity and access to highly-skilled labor are catalyzing market expansion of PMP production in contained systems in the state.

67

Contained PMP production currently co-exists with profitable organic and local food farmers. North Carolina may be able to exploit containment system technology to gain the benefits of PMP production while avoiding the risks of production methods that use food crops in farm fields.

6 Conclusions

Over the last twenty years, agriculture has seen the introduction and rapid deployment of genetically modified (GM) crops for food (i.e., corn and soybeans) and fiber (i.e., cotton). Plant-made pharmaceuticals (PMPs) are a class of GM crop not intended for use as food or feed. Rather, PMPs are intended for use as therapeutic drugs for humans or livestock, or as materials for research and industry (e.g., cell culture media). PMP plants are used as factories to produce the PMP product, the product is extracted from the plant, and the plant remains are discarded. PMP plants can be grown inside laboratories or greenhouses, or they can be grown outside in fields as agricultural crops. When grown as crops, they can be food plants (e.g., corn, safflower, or rice) or non-food plants (e.g., tobacco).

Many of the PMP products under development are proteins--antibodies, enzymes, vaccines and other therapeutic agents--due to an increasing number of protein-based drug discoveries by pharmaceutical companies. In 2005 alone, 38 new protein-based drugs were approved and more are in the FDA pipeline. The pharmaceutical industry seeks low-cost production methods for these new drug products. Producing drugs inside green plants, PMPs, is one of several available production methods. Scientists and industry groups typically cite two reasons for pursuing the PMP production method. First, production of high-quality pharmaceutical components (proteins and antibodies) is presently done using cell cultures inside bioreactors, which is very costly (US$105-175 per gram) and limits the size of the consumer market. Cell culture bioreactors take an average of three to seven years to build and cost on average US$450-$600 million to complete. Second, there is insufficient bioreactor capacity to meet current production needs, let alone expected future needs over the next decade. As of 2002, production of just four pharmaceutical products required 75% of global bioreactor capacity.

By the end of the decade, there could be more than 80 antibody-dependent products with an estimated value of US$20-90 billion, provided adequate production capacity can be developed. The

68

potential size of the market drives investigation of alternative production methods, including PMP production. Proponents of PMP crops claim also that PMP production will increase the range of available drug products, reduce the time required to bring new drugs to market, lower the cost of drug production, and provide additional markets for farmers. Opponents of GM and PMP crops cite potential food safety risks from cross-contamination of food crops, consumer skepticism of genetically engineered products, potential environmental hazards, past regulatory mistakes, and increasing corporate control of agriculture as reasons for their opposition.

The regulatory history of PMPs grown outdoors as field crops is not encouraging. Although PMPs have been grown by several companies in experimental field trials regulated by the U.S. Department of Agriculture since the early 1990s, none has been grown in commercial quantities (although one just received a permit to grow at commercial scale in 2007), and no PMP drug products have as yet been approved by the U.S. Food and Drug Administration. (Some PMPs are being sold in small quantities for use as research materials.) Escape of PMP plants from USDA-regulated field trials has been followed by regulatory reform by USDA, but PMP plants have continued to escape from field trials following the reform effort. Because many PMPs are grown in food plants (even though the PMP plants are not intended for use as food), if the PMP plants escape from their designated areas and become mixed with plants that are intended for use as food, and the mixture enters the food supply, large disruptions of the food industry can occur, as the mixture cannot be used for food, because the PMPs have not been approved by the FDA for use as drugs (much less for use as food).

The lack of commercial scale data limits the ability to assess the (potential) benefits and costs of PMP products. However, a preliminary assessment can be made based on data available for other (non-PMP) GM crops and on current plans for PMP crop production and processing facilities. As of 2007, the one PMP product with planned commercial scale production as a field crop is PMP rice to be grown and processed by Ventria Bioscience in Kansas. At present, Ventria appears to be supported financially by venture capital and government subsidies, as it has only three potential products, the pharmaceuticals lactoferrin, lysozyme, and serum albumin that have not been approved by the FDA for drug, food, or animal feed uses. The products have been marketed as research and bioprocessing materials (for cell culture and cell lysis applications) by InVitria, Sigma-Aldrich, and Ventria itself, but it is

69

not clear that Ventria has received substantial revenues from these uses. Ventria plans to market the extracted milk proteins as an anti-diarrheal additive for infant oral rehydration solutions and as nutritional supplements in yogurt, granola bars, performance drinks and other products. Ventria has also mentioned adding rice-based lysozyme to animal feed as a substitute for the antibiotics added to feed. Ventria claims a potential market for these products of more than $2 billion annually. Ventria estimates that an additional $45 million annually in economic impact will be generated by PMP rice production activities on farms and economic multiplier effects. For comparison, in 2006, Kansas agriculture produced over $11 billion in crop, animal, and related agricultural output, with over $3 billion in wage, rent, interest, and profit income (USDA 2007e). Using a 2.54 economic multiplier, the total economic impacts of the $11 billion in direct impact would be on the order of $28 billion. Ventria’s estimated economic impact of $45 million per year is small relative to the$28 billion impact of Kansas agriculture. In addition, some small (less than $1 million) savings to taxpayers may also result if farmers forego growing government subsidized crops to grow unsubsidized PMP rice.

How much of these estimated economic impacts will likely benefit farmers? Ventria estimates “. . . a projected 30,000 acres of production per year upon full scale commercialization of Ventria’s products.” With an average farm size of approximately 700 acres in Kansas, perhaps 43 farmers would benefit from PMP rice production in Kansas, but the number would probably be lower, as a smaller number of larger farms would reduce costs, based on economies of scale in the use of specialized farm equipment and farmer education required to produce PMP crops. Multiplying this acreage estimate by Ventria’s estimate of $150 to $600 per acre in additional returns to farmers results in a ballpark estimate of $4.5 to $18 million for farmers. Again, a relatively small number compared to the size of the Kansas agriculture industry. While some of these farmers will undoubtedly be located in Kansas, Ventria is reportedly looking for a second field production site. As Ventria has PMP rice field test sites in North Carolina, the state is certainly a contender. In addition to the direct employment of farmers, perhaps 50 people would be employed in Ventria’s proposed PMP rice processing facility in Kansas, and Ventria had 18 employees in its Sacramento headquarters as of 2006. Using typical economic multiplier numbers, perhaps 150 additional jobs would be supported in Kansas due to the economic multiplier effects of PMP rice production and processing in the state.

70

Ventria’s estimates of potential profitability and economic impacts should be considered with caution. Ventria has not obtained FDA approval for its pharmaceutical rice products despite four petitions to FDA since 2003, and the firm recently withdrew its petition to FDA for GRAS status for its product lactoferrin. Even if eventually approved by FDA, Ventria’s products may not be profitable as anti-diarrheal additives for infant formulas marketed in developing countries without subsidies from philanthropic foundations (according to Ventria’s own statements), and the profitability of these products in other uses (in sports drinks, granola bars, etc.) is speculative. Even if infant formula additives were profitable for Ventria after philanthropic subsidies, philanthropies themselves may not choose to subsidize these products if other means of reducing infant mortality (e.g., improved sanitation, hygiene, and breastfeeding practices) are more cost-effective for the philanthropies.

For those farmers considering PMP crop production, several factors should be considered in addition to potentially higher revenues per acre. Ventria is implementing the field trials using independent grower contracts. At this early stage, Ventria covers all costs for North Carolina farmers growing PMPs on subcontract. In the future, independent growers will be expected to provide a seed-to-harvest package deal for the firm’s PMP production. This will involve significant investment in PMP-specific training and dedicated farm equipment. Since 2003, each PMP grower is now required to have dedicated land area, dedicated equipment for planting and harvesting, and separate areas for cleaning PMP equipment and processing PMP crops. Employee training is also required as part of compliance with new FDA and USDA regulatory statues for molecular farming. This raises the possibility that molecular farming contracting for field-grown PMP crops will require such costly investments in infrastructure and compliance that only the largest, wealthy growers would be able to participate and profit. That only some farmers may profit from growing a new crop is not unusual, but what is different about PMP crops is that their use by some farmers is likely to impose “spillover” costs on other farmers who do not grow PMP crops. Farmers who do not grow PMP crops may have to spend money to certify that their crops are “PMP-free” if grown in the same region as PMP crops. This is an especially important issue for organic farmers. In addition, even if a farmer is not located in a PMP-growing region, if the food supply is contaminated by PMP crops grown in other region, all growers of that crop may be hurt by industry-wide consumer boycotts and export restrictions on the product. Given

71

the short history and limited extent of PMP crop field testing, loss of containment and food supply contamination incidents have been relatively frequent, so the potential costs of temporary market loss to non-PMP farmers should not be discounted.

In addition to the potential costs of PMP production to the farm sector, there may also be environmental costs if field grown PMP products have a detrimental effect on fish, wildlife, insects (e.g., bees), or wild plants. While much work has been done on the environmental impacts of GM plants used for food, relatively little work has been done on the potential environmental impacts of PMP plants. At this point all that can reasonably be said is that the potential environmental impacts of PMP field crop production are unknown. For PMP products grown using familiar field crops, the environmental impacts may be small, assuming that the PMP product itself within the plant is not harmful, but again, information is very incomplete and no firm conclusions can be drawn. Ongoing work in bioconfinement methods may reduce the environmental risk of PMP plants.

Detrimental human health effects are another potential cost of PMP production. While detrimental human health effects of products intended for pharmaceutical use are certainly possible, these products would need approval by FDA for use as drugs or food, and any non-accidental effects would likely be small, assuming conscientious review by FDA. One exception might be the creation of new product categories by USDA for regulatory expediency that might allow PMP products to enter medicines or foods as nutritional supplements without FDA review. Such “redefining products to avoid regulation” should be prevented to avoid altogether the potential problem of non-accidental detrimental human health effects.

In contrast, the issue of accidental, detrimental human health effects looms large in the PMP debate. If PMP products not intended for use as food somehow enter the food supply and become ingested by humans, the effects could be significant, as these products may not have undergone food safety testing by FDA, because they were not intended to be used as food. If any of the genetically modified proteins from Ventria’s rice were eventually discovered anywhere in the human food system, it could have disastrous financial consequences for farmers. Ventria would likely shut-down production, which would involve significant transitional costs to farmers as they would have to clean out all of their seeding, harvesting, transportation, and storage equipment. Farmers would also have to ensure that any residual rice plants,

72

seed, etc. were completely wiped out. Finally, farmers would incur significant costs in transforming their equipment for use in some other endeavor. Farmers in the rest of the country would sue Ventria for economic damages resulting from the reduction in price caused by the increased segregation and testing costs imposed on the U.S. crop storage system. Moreover, several countries that import rice from the United States would temporarily ban shipments, which would decrease demand for US rice, further depressing crop prices. Again, the brief history of PMP crop field trials indicates that it is very difficult to prevent co-mingling of PMP and non-PMP crops, implying that the potential for accidental contamination of the food supply is an important issue. Even if PMP-farmers (and non-PMP farmers) could insure themselves against liability for food supply contamination, they might still suffer financial costs if consumers continued to fear a contamination incident and avoided or boycotted the crop, depressing demand and crop prices. Furthermore, even if PMP crops could be contained with 100 percent reliability, sales of non-PMP crops might still suffer if consumers did not believe the containment reliability estimates, which might be a reasonable belief to hold, given the history of containment breaches during field trials. The National Research Council (2004) has proposed a number of recommendations related to pre-market and post-market safety assessments that could reduce both accidental and non-accidental human health risks associated with PMP production.

Neither food plants nor farmers’ fields are necessary for the production of PMPs. PMPs can be grown in sterile containment systems instead of agricultural fields. Advantages of growing PMPs in containment systems include better uniformity of product, lack of residues from herbicides, pesticides and fungicides, and greatly reduced risk of contaminating the food supply. Two disadvantages of producing proteins in containment systems are that it is thought to cost more, and it is thought to take longer to bring the product to market. However, recent advances in closed-system technology have eliminated some of the cost difference between PMPs and contained cell culture systems.

North Carolina has a competitive advantage in PMP production nationally without using food crops in farm fields, as its resident biotechnology firms use containment systems based on non-food plants. Contained production systems that attain high product purity and access to highly-skilled biotech labor are catalyzing market expansion for PMPs grown in contained systems in the state. Contained PMP production currently co-exists with

73

profitable organic and local food suppliers in the state. Choice among contained PMP production systems continues to expand. For example, some alternative PMP containment systems utilizing non-food plants include duckweed (Lemna spp.), tobacco (Nicotiana spp.), algae (Chlamydomonas reinhardtii) and moss (Physcomitrella patens), and fungi (Aspergillus niger). Yet another option is to produce PMPs using food crops grown inside greenhouses, such as potatoes grown hydroponically.

At the present time, PMP production via food crops in the field should not be considered a cornerstone of future agricultural policy or rural economic development policy in North Carolina or elsewhere in the United States. Given past difficulties in securing FDA approval for PMP products, the benefits of PMP production are too speculative. Furthermore, given past difficulties in containing PMP products in the field, the risks and potential costs of future escape events are too great.

74

7 RAFI Recommendations

We are grateful for the considerable expertise and the careful, fair and balanced approach that Dr. Dumas and his colleagues have taken in assessing this novel and controversial subject.

Our goals are to provide information and analysis for family farmers and policy leaders for making better-informed choices about our shared agricultural future. We support a much more transparent process, and consideration of a broader range of sound options, prior to committing our farmers and states to any potential rural economic initiatives. This is especially important in cases like this, which can place farmers or states into additional risk or jeopardy and can impact existing successful agricultural and pharmaceutical enterprises.

The scope of this report is focused on North Carolina policy and opinion leaders, farmers and civic organizations; however, its lessons and implications can be applied to other states as well. The outdoor development of novel genetically engineered pharmaceuticals utilizing food crops otherwise consumed as foods, such as rice, and the novel and very early stage of this development leads RAFI-USA to the following recommendations:

We agree with the USA Rice Federation, Grocery Manufacturers Association of America, the editors of the journal Nature Biotechnology, and others who have called for an end to the use of food crops for plant-made pharmaceuticals development, and especially not in the open field settings. This very novel technology poses considerable potential legal, environmental and health risks – ones to which the farmers involved will potentially be the most liable. States and farmers should avoid such unintended problems and contaminations by avoiding the outdoor planting of such crops.

We also remain quite concerned by the lack of meaningful and comprehensive regulatory oversight of these plant-made pharmaceuticals. The problems arising from this lack of regulatory rigor has been well documented in this report. RAFI-USA strongly urges immediate steps be taken to independently evaluate these failures and deficiencies and for the appropriate regulatory agencies to resolve these problems by committing more resources into

75

comprehensive assessments prior to approvals and into more on-going post- introduction tracking and oversight.

The overall farmer economic benefit of PMP rice production at this time seems very meager and does not constitute a meaningful rural economic engine for North Carolina or other state’s farmers.

This state and other states confronting this issue should take a case-by-case approach to accessing these novel pharmaceutical crop initiatives. We also urge consideration of the history of the PMP sector, which is dominated by small biotech start-ups, many of which have gone bankrupt, and none of which has yet produced a single FDA-approved drug despite field-testing that dates back to 1991. State incentives seem unjustified and premature for rice-based pharmaceutical development for several reasons: for one, the FDA’s continuing refusal to grant “generally recognized as safe” status to Ventria’s pharma-rice derived proteins; as well as the potentially high risks and the very low number of farmers that would be required to meet any potential market niche that may develop. Also, the low economic returns to such farmers as compared to other, lower-risk agricultural enterprises seems to warrant a more cautious approach.

The state of North Carolina and other states should also more carefully assess their own liability, risks and the potential impacts of their support for open field PMPs; as well as the impacts on other non-open field biotech industries in their states.

A much broader range of farmer and rural community options should be comprehensively compared and evaluated prior to considering support for PMP development.

It is our assessment that using rice, a major worldwide food crop, in open field development of genetically engineered pharmaceuticals does not meet our criteria of the “triple-bottom-line” wins of economic, environmental and social benefits for North Carolina farmers and citizens.

We strongly urge much greater public transparency, accountability, scrutiny and debate as well as a strong call of support for a much fuller dialogue about meaningful ways to revitalize rural economies that are desperately

76

needed to help farmers meet growing consumer demand for safe, local and nutritious foods, while preserving the environment and providing fair returns back to our farm families and communities.

77

8 References

Agres, T. 2006. LSBC closing signals pharming trouble. The Scientist. January 17, 2006. http://www.the-scientist.com/news/display/22969.

AOSCA. 2003. "Yellow Books" 2003 Operational Procedures, Crop Standards and Service. Association of Official Seed Certifying Agencies. Meridian, Idaho.

Programs. AOSCA, Meridian, Idaho

Barton, J.E., and M. Dracup. 2000. Genetically modified crops and the environment. Agronomy Journal. 92:797-803.

Batie, S.S., and D.E. Ervin. 2001. Transgenic crops and the environment: Missing markets and public roles. Environment and Development Economics. 6:435-457.

Belcher, K., J. Nolan, and P.W.B. Phillips. 2005. Genetically Modified Crops and Agricultural Landscapes: Spatial Patterns of Contamination. Ecological Economics. 53:387-4401.

Bennett, B. “Ventria BioScience finds Missouri producers unwelcoming.” Delta Farm Press. May 26, 2005. http://deltafarmpress.com/news/

Bennett, D. “Arkansas’ emergency session on CL 131 rice.” Delta Farm Press. March 1, 2007. http://deltafarmpress.com/news/

Bethell,D. 2006. “Lactiva and Lysomin: Helping to Save Lives By Improving Oral Rehydration Solution.” Ventria Bioscience. Sacramento, CA. May 1, 2006. http://www.ventria.com/news/ORS%20Article%2005-06.asp

Berwald, Derek, C.A. Carter, and Guillaume P. Gruere. 2006. “Rejecting New Technology: The Case of Genetically Modified Wheat.” American Journal of Agricultural Economics. 88(2):432-447.

Bajaj, Shavindra, and Amitabh Mohanty. 2005. Recent Advances in Rice Biotechnology—Towards Genetically Superior Transgenic Rice. Plant Biotechnology Journal. 3:275-307.

BIO. 2007. “Statement Regarding Plants That Produce Pharmaceutical and Industrial Products.” Biotechnology Industry Organization. Washington, D.C. http://www.bio.org/healthcare/pmp/statement.asp

78

BIO. 2006. “Industry Update on Plant-Made Pharmaceuticals.” Biotechnology Industry Organization. Washington, D.C. December 2006.

BIO. 2002a. “The Case for Plant-Made Pharmaceuticals.” Biotechnology Industry Organization. Washington, D.C. April/May 2002.

BIO. 2002b. “Advantages of Plants to Produce Therapeutic Proteins.” Biotechnology Industry Organization. Washington, D.C. April/May 2002.

Center for Food Safety. 2005. “USDA Approves Applications to Grow Rice with Human Genes on 270 Acres in North Carolina and Missouri.” Press Release, Center for Food Safety. Washington, DC. June 30, 2005.

Chern, Wen S., K. Rickertsen, N. Tsuboi, and T.T. Fu. 2002. “Consumer Acceptance and Willingness to Pay for Genetically Modified Vegetable Oil and Salmon: A Multiple-Country Assessment.” The Journal of Agrobiotechnology, Management, and Economics. ????

Chicago Sun-Times. 2003. “Kraft exec wants tougher rules on planting crops for drugs.” Chicago Sun-Times. April 4, 2003.

Cox, Kevin M., Jason D Sterling, Jeffrey T Regan, John R Gasdaska, Karen K Frantz, Charles G Peele, Amelia Black, David Passmore, Cristina Moldovan-Loomis, Mohan Srinivasan, Severino Cuison, Pina M Cardarelli & Lynn F Dickey. 2006. Glycan optimization of a human monoclonal antibody in the aquatic plant Lemna minor.” Nature Biotechnology. 24(12):1591-1597.

Daily Mail. 2007. “Human Genes in Your Food.” Daily Mail. London, United Kingdom. March 6, 2007.

Dalton, R. 2002. Superweed study falters as seed firms deny access to transgene. Nature. 419(6908), 655.

Des Moines Register. 2003. "Investments drain state money." Des Moines Register. April 27, 2003.

Dobson, C. 2002. Canada and identity preservation systems for non-GM canola. MSc Thesis, University of Saskatchewan. Saskatoon, Canada.

European Commission. 2001. Directive 2001/18/EC of the European Parliament and of the Council of 12 March 2001 on the deliberate release into the environment of genetically modified organisms and repealing Council Directive

79

90/220/EEC. Official Journal of the European Communities 17.4.2001: L 106:1–38

FDA. 2002. “Draft guidance for industry: drugs, biologics and medical devices derived from bioengineered plants for use in humans and animals” Federal Register 67:57828-29. September 12, 2002.

Federal Register. 2007. Vol. 72, No. 94, Docket No. APHIS-2007-0006. May 16, 2007. pp. 27539-27540.

Federal Register. 2005a. Vol. 70, No. 123, Docket No’s. 05-007-3. June 28, 2005. pp. 37077-37078.

Federal Register. 2005b. Vol. 70, No. 123, Docket No’s. 05-006-3. June 28, 2005. pp. 37079-37080.

Fischer R., Stoger E., Schillberg S., Christou P. and Twyman R.M.. 2004. Plant-based production of biopharmaceuticals. Current Opinion in Plant Biology. 7: 152-158.

Food Products Association. 2005. “FPA Questions Proposed Use of Food Crops to Produce Plant-Made Pharmaceuticals.” Food Products Association. Washington, DC. June 3, 2005. http://www.fpa-food.org/content/newsroom/article.asp?id=285

Franklin, S.E., and S.P. Mayfield. 2004. Prospects for molecular farming in the green alga Chiamydomonas reinhardtii. Current Opinion in Plant Biology. 7: 159-165.

Freese, B. 2007. “A Grain of Caution: A Critical Assessment of Pharmaceutical Rice.” Center for Food Safety. Washington, DC. April 2007.

Freese, B., and R. Caplan. 2006. “Plant-Made Pharmaceuticals: Financial Risk Profile.” Friends of the Earth and U.S. Public Interest Research Group. Washington, D.C.

Freese, B. 2005. “Pharmaceutical Rice in Missouri.” Friends of the Earth. Washington D.C.

Freese, B. 2002. “Manufacturing Drugs and Chemicals in Crops: Biopharming Poses New Threats to Consumers, Farmers, Food Companies and the Environment.” Genetically Engineered Food Alert. http://www.gefoodalert.org.

Freese, B., M. Hansen, and D. Gurian-Sherman. 2004. “Pharmaceutical Rice in California: Potential Risks to Consumers, the Environment, and the California Rice Industry.” Friends of the Earth. Washington D.C.

Fulton, Murray and K. Giannakas. 2004. “Inserting GM Products into the Food Chain: The Market and Welfare Effects of

80

Different Labeling and Regulatory Regimes.” American Journal of Agricultural Economics 86(1):42-60.

Furtan, W.H., R.S. Gray, and J.J. Holzman. 2005. “Regulatory Approval Decisions in the Presence of Market Externalities: The Case of Genetically Modified Wheat.” Journal of Agricultural and Resource Economics. 30(1):12-27.

Gasdaska J.R., Spencer D. and Dickey L.. 2003. Advantages of Therapeutic Protein Production in the Aquatic Plant Lemna. Bioprocessing Journal. 2: 49-56.

Gealy D.R., D.H. Mitten, and J.N. Rutger. 2003. Gene flow between red rice (Oryza sativa) and herbicide-resistant rice (O. sativa): implications for weed management. Weed Technology. 17: 627-645

Gilligan, C.A., D. Claessen, and F. van den Bosch. 2003. Spatial and temporal dynamics of gene movements arising from deployment of transgenic crops. Proceedings of the Frontis Workshop—Environmental Costs and Benefits of Transgenic Crops in Europe. Wageningen University.

Haslberger, A. 2001. GMO contamination of seeds. Nature Biotechnology. 19(7):613.

Huang, F., Buschman, L.L., Higgins, R.A., and McGaughey, W.H. 1999. “Inheritance to Bacillus thuringiensis Toxin (Dipel ES) in European Corn Borer.” Science 284:965-7.

Hucl, P., M. Matus-Cadiz. 2001. Isolation distances for minimizing out-crossing in spring wheat. Crop Science. 41:1348-1351.

ISAAA. 2006. “Global Status of Commercialized Biotech/GM Crops: 2006.” International Service for the Acquisition of Agri-Biotech Applications. Brief 35. Ithaca, New York. http://www.isaaa.org

Kershen, D.L. 2004. Legal Liability Issues in Agricultural Biotechnology. Crop Science. 44:456-463.

KFMA. 2007. 2006 County Report for Geary county, Kansas. Kansas Farm Management Association. http://www.kmar105.com/KFMA/Default.htm

Kostandini, Genti, B.F. Mills, and G.W. Norton. 2006. “The Potential Impact of Tobacco Biopharming: The Case of Human Serum Albumin.” American Journal of Agricultural Economics 88(3):671-679.

81

Leighton, P. “Study on infants in Peru sparks ethics inquiry.” Science and Development Network. July 18, 2006. http://www.scidev.net/news/

Lin, W.W., and D.D. Johnson. 2004. “Segregation of non-biotech corn and soybeans: Who bears the cost?” Chapter in: Santaniello, V., and R.E. Evenson. (eds). The Regulation of Agricultural Biotechnology. Wallingford, UN: CAB International.

Lin, William, G.K. Price, and E.W. Allen. 2003. “StarLink: Impacts on the U.S. Corn Market and World Trade.” Agribusiness. 19(4):473-488.

Lin, W.W. 2002. “Estimating the costs of segregation for non-biotech maize and soybeans.” Chapter in: Santaniello, V., Evenson, R.E., and Zilberman, D. (eds). Market Development for Genetically Modified Foods. Wallingford, UK: CAB International.

Lonnerdal, Bo. 2002. “Expression of Human Milk Proteins in Plants.” Journal of the American College of Nutrition 21(3):219.

Los Angeles Times. 2004. “Pharmaceutical Rice Plan Advances in State: A commission backs Ventria’s proposal to plant genetically altered grain to make medicinal proteins.” Los Angeles Times. March 30, 2004.

Losey, J.E., Rayor, L.S., and Carter, M.E. 1999. “Transgenic Pollen Harm Monarch Larvae.” Nature. 399:214.

Lu, Bao-Rong, and Allison A. Snow. 2005. Gene Flow from Genetically Modified Rice and Its Environmental Consequences. Bioscience. 55(8):669-678.

Ma, J. K-C, P.M.W. Drake and P. Christou. 2003. The production of recombinant pharmaceutical proteins in plants. Nature Genetics. 4:794-805.

Marshall A. 2004. Nature Biotechnology. 22: 133.

Meyer, M. Personal communication. Director, NC BioNetwork, North Carolina Community College System July 25, 2007.

Moschini, Giancarlo. 2006. “Pharmaceutical and Industrial Traits in Genetically Modified Crops: Coexistence with Conventional Agriculture.” American Journal of Agricultural Economics 88(5):1184-1192.

Moss, C.B., A. Schmitz, and T.G. Schmitz. 2006. “First-Generation Genetically Modified Organisms in Agriculture.” Journal of Public Affairs 6:46-57.

82

Moss, C.B., A. Schmitz, and T.G. Schmitz. “Segregating Genetically Modified and Non-Genetically Modified Corn in a Marketing Channel.” Applied Economics. (forthcoming 2007).

Moss, C B., T.G. Schmitz, and A. Schmitz. 2004. “The Brave New World: Imperfect Information, Segregation Costs, and Genetically Modified Organisms.” German Journal of Agricultural Economics 53:303-308.

NAS. 2004. “Infant formula: Evaluating the Safety of New Ingredients.” Committee on the Evaluation of the Addition of Ingredients New to Infant Formula, National Academy of Sciences. http://www.nap.edu/catalog/10935.html

National Public Radio. 2006. “Super Grass Spreads Beyond Its Oregon Test Plot.” Day to Day, National Public Radio. November 20, 2006.

Nelson, Gerald C. 2001. Genetically Modified Organisms in Agriculture: Economics and Politics. Academic Press. San Diego, CA.

Nelson, Gerald C., and Alessandro De Pinto. 2001. “GMO Adoption and Nonmarket Effects.” Chapter in: Genetically Modified Organisms in Agriculture: Economics and Politics. Academic Press. San Diego, CA.

New York Times. 2007. “How to Confine the Plants of the Future?” New York Times. April 8, 2007.

New York Times. 2006b. “Biotech Plan in Missouri Suffers Setback.” New York Times. January 6, 2006.

New York Times. 2006a. “World Trade Agency Rules for U.S. in Biotech Dispute.” New York Times. February 8, 2006.

New York Times. 2005a. “Modified Rice May Benefit China Farms, Study Shows.” New York Times. May 3, 2005.

New York Times. 2005b. “China’s Problem with ‘Anti-Pest’ Rice.” New York Times. April 16, 2005.

New York Times. 2005c. “Can Gene-Altered Rice Help Rescue the Farm Belt?” New York Times. August 16, 2005.

New York Times. 2002. “Spread of Gene-Altered Pharmaceutical Corn Spurs $3 Million Fine.” New York Times. December 7, 2002.

Norris, C.E., E.C. Simpson, J.B. Sweet, and J.E. Thomas. 1999. Monitoring weediness and persistence of genetically modified

83

oilseed rape (Brassica napus) in the U.K. Chapter in: Lutman, P.J.W. (ed), Gene Flow and Agriculture: Relevance for Transgenic Crops. BCPC Symposium Proceedings, vol. 72.

NAMA. 2007. “Issue Brief: North American Millers' Association Statement on the Use of Food and Feed Crops for the Production of Plant-made Pharmaceuticals and Industrial Products.” North American Millers’ Association. Washington, DC. http://www.namamillers.org/is_briefs_biopharm.html

National Food Products Association. 2003. “NFPA Comment to USDA RE: Docket No. 03-031-1 . Field Testing of Plants Engineered to Produce Pharmaceuticals and Industrial Compounds. 68 Federal Register 11337, March 10, 2003.” Food Products Association. Washington, DC. May 12, 2003. http://www.fpa-food.org/content/newsroom/article.asp?id=43

NRC. 2004a. “Biological Confinement of Genetically Engineered Organisms.” Committee on the Biological Confinement of Genetically Engineered Organisms, National Research Council. 284 pp. National Academy Press. Washington, DC.

NRC. 2004b. “Safety of Genetically Engineered Foods: Approaches to Assessing Unintended Health Effects.” Committee on Identifying and Assessing Unintended Effects of Genetically Engineered Foods on Human Health, National Research Council. 256 pp. National Academy Press. Washington, D.C.

NRC. 2002. “Environmental Effects of Transgenic Plants: the Scope and Adequacy of Regulation.” Committee on Environmental Impacts Associated with Commercialization of Transgenic Plants, Board on Agriculture and Natural Resources, National Research Council. 342 pp. National Academy Press. Washington, DC.

Onyango, Benjamin and R. Govindasamy. 2005. “The Promise of Ag-Biotech.” CHOICES. 20(4).

Pew Initiative. “Issues in the Regulation of Genetically Engineered Plants and Animals.” Pew Initiative on Food and Biotechnology. Pew Charitable Trusts. Washington, DC. April 2004.

Poirier Y., Dennis D.E., Klomparens K. and Somerville C.. 1992. Polyhydroxybutyrate, A Biodegradable Thermoplastic, Produced in Transgenic Plants. Science. 256: 520-523.

Quist, D., and I. Chapela. 2001. Transgenic DNA introgressed into traditional maize landraces in Oaxaca, Mexico. Nature 414:541-543.

84

Rieger, M.A., M. Lamond, C. Preston, S.B. Powles, and R.T. Roush. 2002. Pollen-mediated movement of herbicide resistance between commercial canola fields. Science. 296:2386-2388.

Roanoke Times. 2000a. “Healthy hope for tobacco.” Roanoke Times. Roanoke, Virginia. June 5, 2000.

Roanoke Times. 2000b. “Biotech group receives funding.” Roanoke Times. Roanoke, Virginia. August 11, 2000.

Sacramento Bee. 2007. “Judge blocks altered alfalfa: Organic farmers say cross-pollination of genetically modified plants could risk crop contamination.” Sacramento Bee. May 4, 2007.

Sacramento Bee. 2006. “Kansas opens arms to Ventria’s biotech rice.” Sacramento Bee. June 13, 2006.

San Francisco Chronicle. “Future pharmers—Biotech firms using plants in attempt to produce proteins faster, cheaper.” San Francisco Chronicle. August 12, 2002.

Sargent, Brandy. Personal communication. Corporate Communications, Ventria Bioscience. Sacramento, CA. July 27, 2007.

Schmitz, Troy G., A. Schmitz, and C.B. Moss. 2005. “The Economic Impact of StarLink Corn.” Agribusiness. 21(3):391-407.

Schmitz, Troy G., A. Schmitz, C.B. Moss. 2004. “Did StarLink Reduce Import Demand for Corn?” Journal of Agriculture & Food Industrial Organization. 2(2).

Silber, J. “Permit for biotech rice is denied.” Contra Costa Times. April 9, 2004.

Smyth, Stuart, Peter W.B. Phillips, William A. Kerr and George G. Khachatourians. 2004. Regulating the Liabilities of Agricultural Biotechnology. CABI Publishing. Cambridge, MA.

Smyth, S., and P.W.B. Phillips. 2003. Product differentiation alternatives: Identity preservation, segregation and traceability. AgBioForum 5(2):30-42.

Snow, A., D. Pilson, L. Rieseberg, M. Paulsen, N. Pleskac, M. Reagon, D. Wolf and S. Selbo. 2003. A Bt transgene reduces herbivory and enhances fecundity in wild sunflowers. Ecological Applications 13(2):279-286.

Somerville CR and Bonetta D.. 2001. Plant Physiology 125: 168-171.

85

Spriggs, J., and G.E. Isaac. 2001. International Competitiveness and Food Safety: The Case of Beef. Wallingford, UK: CAB International.

SSC. 2004. Monsanto vs. Schmeiser. Supreme Court of Canada. SCC 34.

Stewart P.A. and A.J. Knight. 2005. Technology Forecasting and Social Change. 72: 521-534.

St. Louis Post Dispatch. 2005. “Firm races against clock to plant rice in Missouri.” St. Louis Post Dispatch. April 23, 2005.

Streatfield, S.J. 2005.Regulatory issues for plant-made pharmaceuticals and vaccines. Expert Reviews of Vaccines. 4: 591-601.

UCS. 2007. “UCS Comments to USDA on Kansas Pharma Rice.” Union of Concerned Scientists. Washington, DC. March 29, 2007. http://www.ucsusa.org/food_and_environment/

UCS. 2006a. “UCS Uncovers Lax USDA Oversight of Pharma Crops: New Evidence Points to Need for Ban on Pharma Food Crops.” Union of Concerned Scientists. Washington, DC. http://www.ucsusa.org/food_and_environment/

UCS. 2006b. “Position Paper: Pharmaceutical and Industrial Crops.” Union of Concerned Scientists. Washington, DC. http://www.ucsusa.org/food_and_environment/

UCS/CU. 2007. “Comments to USDA on Pharma Safflower.” Union of Concerned Scientists, Washington, DC, and Consumers Union, Yonkers, NY. March 7, 2007. http://www.ucsusa.org/food_and_environment/

USA Rice Federation. 2007a. “USA Rice Supports APHIS Action on Clearfield 131, but Questions Regulatory Practices.” USA Rice Federation News. March 6, 2007.

USA Rice Federation. 2007b. “USA Rice Disappointed with APHIS Approval of Pharmaceutical Rice Production in Kansas.” USA Rice Federation News. May 16, 2007.

USA Today. 2006. “Biotech firm raises furor with rice plan.” USA Today. May 14, 2006.

USDA. 2007a. “Release Permits for Pharmaceuticals, Industrials, Value-Added Proteins for Human Consumption, or for Phytoremediation Granted or Pending by APHIS as of July 6, 2007.” U.S. Department of Agriculture, APHIS Biotechnology

86

Regulatory Services. http://www.aphis.usda.gov/brs/ph_permits.html

USDA. 2007b. “USDA Provides Update for Farmers on Genetically Engineered Rice.” U.S. Department of Agriculture, APHIS Biotechnology Regulatory Services Program Announcement. February 2007.

USDA. 2007c. “Report of LibertyLink Rice Incidents.” U.S. Department of Agriculture, APHIS News Release. October 4, 2007. http://www.aphis.usda.gov/newsroom/content/2007/10/

USDA. 2007d. “Lessons Learned and Revisions under Consideration for APHIS’ Biotechnology Framework.” U.S. Department of Agriculture, APHIS News Release. October 4, 2007. http://www.aphis.usda.gov/newsroom/content/2007/10/

USDA. 2007e. “State Fact Sheets: Kansas – Farm Income and Value Added Data.” U.S. Department of Agriculture, Economics Research Service. http://www.ers.usda.gov/StateFacts/KS.htm

USDA. 2007f. “Commodity Costs and Returns: U.S. and Regional Cost and Return Data – Corn, Prairie Gateway Region.” U.S. Department of Agriculture, Economics Research Service. http://www.ers.usda.gov/Data/CostsAndReturns/

USDA. 2006a. “Permitting Genetically Engineered Plants that Produce Pharmaceutical Compounds.” U.S. Department of Agriculture, APHIS Biotechnology Regulatory Services, BRS Factsheet.. February 2006.

USDA. 2006b. “Draft Guidance for APHIS permits for field testing or movement of organisms with pharmaceutical or industrial intent.” U.S. Department of Agriculture, APHIS Biotechnology Regulatory Services. March 2006.

USDA. 2005a. “Audit Report: Animal and Plant Health Inspection Service Controls Over Issuance of Genetically Engineered Organism Release Permits.” U.S. Department of Agriculture, Office of Inspector General, Southwest Region, Audit 50601-8-Te. December 2005.

USDA. 2005b. “USDA/APHIS Environmental Assessment In response to permit applications (04-302-01r and 05-117-01r) received from Ventria Bioscience for field-testing of rice, Oryza sativa, genetically engineered to express human lactoferrin.” U.S. Department of Agriculture, Animal and Plant Health

87

Inspection Service, Biotechnology Regulatory Services. June 21, 2005.

USDA. 2003. Advisory Committee on Biotechnology and 21st Century Agriculture (AC21). Meeting transcript. U.S. Department of Agriculture. June 17, 2003. http://www.usda.gov/agencies/biotech/ac21/meetings/mtg_june03/ac21_mtg_june03.html

USEPA. “A Set of Scientific Issues Being Considered by the Environmental Protection Agency Regarding: Assessment of Scientific Information Concerning StarLink Corn.” FIFRA Scientific Advisory Panel, SAP Report No. 2000-06. December 1, 2000. http://www.epa.gov/scipoly/sap/meetings/2000/november/one.pdf

Ventria Bioscience. 2007. “Ventria Product Benefits and Economic Analysis.” Corporate Communications, Ventria Bioscience. Sacramento, CA. July 27, 2007.

Ventria Bioscience. 2006. Press release--“Kansas Officials Announce Agreement for Bioprocessing Facility in Junction City Facility for plant-made pharmaceuticals continues Kansas’ advancement in biotechnology.” Sacramento, CA. September 29, 2006. http://www.ventria.com/news/Press%20Release%209-29-06.asp

Ventria Bioscience. 2003. “Safety, Compositional and Nutritional Aspects of LF164 Rice Transformational Event.,” Biotechnology Notification File 082, voluntarily submitted to the U.S. Food and Drug Administration on November 24, 2003 by Ventria Bioscience. Obtained in a Freedom of Information Act request by Center for Food Safety, Washington, DC.

Washington Daily News. 2006a. “Biotech firm is shopping for a place to grow and process engineered rice.” Washington Daily News. Washington, NC. June 27, 2006.

Washington Daily News. 2006b. “Ventria growing larger crop in Washington County.” Washington Daily News. Washington, NC. December 10, 2006.

Washington Post. 2007. “USDA Backs Production of Rice with Human Genes.” Washington Post. March 2, 2007.

Washington Post. 2006. “Gene-Altered Profit-Killer.” Washington Post. September 21, 2006.

88

Washington Post. 2003. “U.S. Will Subsidize Cleanup of Altered Corn.” The Washington Post. March, 26, 2003.

Weinberg, E.D. 2001. “Human lactoferrin: a novel therapeutic with broad spectrum potential.” Journal of Pharmacy and Pharmacology. 53(10):1303-1310.

Wesseler, J.H.H. 2005. Environmental Costs and Benefits of Transgenic Crops. Springer. Norwell, MA.

Williams, C. “Finding the Fit Between Molecular Farming and Organic Farming Opportunities in North Carolina—Report on a Roundtable Discussion.” Department of Biology, Duke University. Durham, NC. April 3, 2006.

Williams, C. 2007. “The fit between organic and pharma crops in North Carolina.” Nature Biotechnology. 25(2):166-167.

Wisner, Robert. 2005. “The Economics of Pharmaceutical Crops: Potential Benefits and Risks for Farmers and Rural Communities.” Union of Concerned Scientists. UCS Publications. Two Brattle Square. Cambridge, MA.

Zavaleta N, Figueroa D, Rivera J, Sanchez J, Alfaro S, Lonnerdal B. 2007. Efficacy of rice-based oral rehydration solution containing recombinant human lactoferrin and lysozyme in Peruvian children with acute diarrhea. J. Pediatric Gastroenterology and Nutrition. 44(2): 258-64.

89

9 Tables

Table 1. Countries Growing 50,000+ Hectares of Genetically Modified (GM) Crops in 2006. (Source: ISAAA 2006)

Rank Country Land Area in GM Crops (million hectares)

GM Crops Grown

1 United States

54.6 Soybean, maize, cotton, canola, squash, papaya, alfalfa

2 Argentina 18.0 Soybean, maize, cotton 3 Brazil 11.5 Soybean, cotton 4 Canada 6.1 Canola, maize, soybean 5 India 3.8 Cotton 6 China 3.5 Cotton 7 Paraguay 2.0 Soybean 8 South

Africa 1.4 Soybean, maize, cotton

9 Uruguay 0.4 Soybean, maize 10 Philippines 0.2 Maize 11 Australia 0.2 Cotton 12 Romania 0.1 Soybean 13 Mexico 0.1 Cotton, soybeaen 14 Spain 0.1 Maize 15 Colombia <0.1 Cotton 16 France <0.1 Maize 17 Iran <0.1 Rice 18 Honduras <0.1 Maize 19 Czech

Republic <0.1 Maize

20 Portugal <0.1 Maize 21 Germany <0.1 Maize 22 Slovakia <0.1 Maize

90

Table 2. Technological “Generations” of Genetically Modified (GM) Plants and Plant-Made Pharmaceuticals (PMPs)

Technological Generation

Goal of Genetic Modification

Plant Use

Example Crop

Generation 1 Reduce production costs of food plants

Food

RoundUp-Ready™ soybeans resist pesticides

Generation 2 Increase value of food plants by improving nutritional content

Food

“Golden Rice” with enhanced vitamin A

Generation 3

Produce pharmaceutical or industrial products using food or non-food plants

Factory

Pharma-rice produces human milk enzymes

Table 3. U.S. Rice Export Markets Impacted by the Presence of LL601 Rice. (Source: USA Rice Federation web site, June 2007.)

2006 Ranking

Country

2006 Exports

($millions)

Importer Reaction to Presence of LL601 in U.S. Rice Supply

Trade

Impacted 1 Mexico $205 GM certification now required; trade disrupted X 2 Japan $169 Testing now required X 3 Iraq $145 Testing now required at 1% sensitivity X 4 Haiti $112 Trade continues 5 Canada $107 Testing now required at 0.5% sensitivity X 6 EU $69 Trade in long grain rice stopped X 7 Saudi

Arabia $42 Trade continues; labeling required for presence >

1%

8 Nicaragua $40 Trade continues 9 Cuba $40 Trade disrupted; situation uncertain X 10 Honduras $39 Trade continues ... 12 Korea $32 Testing now required X ... 16 Philippines $20 Trade stopped X ... 18 Taiwan $20 Testing now required X

Total U.S. Exports: $1,289 Share of Total U.S. Exports Impacted: 63% In addition, Russia has banned all U.S. rice imports, and the UAE importers are seeking a “GE free” certificate on U.S. rice.

Table 4. Ventria Bioscience – Regulatory History Overview

Date Event

1993 Ventria Bioscience founded by Dr. Ray Rodrequez, professor of molecular and cellular biology at University of California, Davis.

1997 Ventria develops ExperssTec, a proprietary technology that uses rice and barley plants to produce proteins

2003 USDA changes Ventria’s product designation from “pharmaceutical proteins” to “value added protein for human consumption”

2003

Ventria applies to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lactoferrin rice as a possible contaminant in food rice. Ventria sought approval of Lf rice as contaminant while publicly claiming Lf rice would not contaminate food. If approved, PMP rice would be exempt from the food additive review process.

2004 USDA grants Ventria field trial release permits to grow PMP rice on 120 acres in California

2004 Opposition from California rice farmers blocks field production of PMP rice in California

2004 Ventria applies to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lactoferrin rice as ingredient in foods, beverages, and medical foods.

2005 Ventria markets Lacromin (lactoferrin) for laboratory cell culture use

2005 USDA announces “Finding of No Significant Impact” (FONSI) and the availability of an Environmental Assessment for the proposed field trials of Ventria’s PMP rice in Missouri and North Carolina

2005 USDA grants Ventria field trial release permits to grow 200 acres of PMP rice in Missouri and 70 acres of PMP rice in North Carolina in 2005

2005 Ventria applies to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lysozyme rice as an antimicrobial agent and ingredient in various foods.

2005 Ventria begins PMP rice field trials in NC

2005 Field production of Ventria’s PMP rice in Missouri blocked by farmers and food processors

2005

Researchers at Tidewater Research Station, NC, a half-mile from the Ventria field trial site, complain that rice germplasm at the station, part of the National Plant Germplasm System, could be contaminated by the Ventria field trial plantings

2005 For unknown reasons, Ventria withdraws petition to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lysozyme rice as an antimicrobial agent and ingredient in various foods.

2006 Ventria applies to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lysozyme rice for use in infant formulas & pediatric oral rehydration solutions.

2006

Union of Concerned Scientists files Freedom of Information Act request for USDA-APHIS inspection and company compliance reports for Ventria field test site in NC; UCS concludes that USDA was not adequately

monitoring and inspecting the Ventria test site.

2006 Ventria markets Lysobac (lysozyme) for laboratory cell lysis use

2006 USDA expands Ventria field trial permits to grow up to 335 acres of PMP rice in NC in 2006

2006

Ventria withdraws petition to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lactoferrin rice as ingredient in foods, beverages, and medical foods because FDA indicated it would not approve lactoferrin as safe.

2006 Ventria markets Cellastim (serum albumin)

2006 USDA moves the National Plan Germplasm System operations from the Tidewater Research Station, NC, to a station in Beltsville, MD.

2006

Ventria withdraws petition to FDA for “Generally Recognized as Safe” (GRAS) status for PMP lactoferrin rice as ingredient in foods, beverages, and medical foods because FDA indicated it would not approve lactoferrin as safe.

2007 USDA approves Ventria field trial permits to grow PMP rice in NC in 2007

2007 USDA releases draft environmental impact statement concluding that Ventria’s PMP rice could be grown in Kansas with no undue risks

2007 USDA announces that food rice seed in Arkansas has been contaminated with genetically modified rice variety LL62

2007 USDA announces “Finding of No Significant Impact” (FONSI) and the availability of an Environmental Assessment for the proposed field production of Ventria’s PMP rice Kansas

2007 USDA grants Ventria field trial release permits to grow 3,200 acres of PMP rice in Kansas

2007 Union of Concerned Scientists criticizes USDA’s decision to allow field production of PMP rice in Kansas

2007 USA Rice Federation files comments with USDA opposing field production of PMP rice in Kansas

Table 5. Comparison of Production Systems for Recombinant Human Pharmaceutical Proteins. (Source: Source: Ma et al., 2003)

Production Systems Production

System

Feature Bacteria Yeast

Mammalian cell culture

Transgenic animals

Plant cell cultures

Transgenic plants

Overall cost Production timescale Scale-up capacity Product quality Glycosylation Contamination risks Storage cost

Low Short High Low None Endotoxins Moderate

Medium Medium High Medium Incorrect Low risk Moderate

High Long Very low Very high Correct Viruses, prions, and oncogenic DNA Expensive

High Very long Low Very high Correct Viruses, prions, and oncogenic DNA Expensive

Medium Medium Medium High Minor differences Low risk Moderate

Very low Long Very high High Minor differences Low risk Inexpensive

Table 6. Example of Identity Preserving Production and Marketing (IPPM) System Costs—

Canadian HT canola in 1996 (1996 Canadian dollars) (Source: Smyth et al. 2004)

Cost Category AgrEvo & Manitoba Pool Elevators ($/t)

Saskatchewan Wheat Pool ($/t)

On-farm costs Freight Inefficiency Dead Freight Processor Administration Opportunity cost Collective subsidy Total IPP Cost

$1 $5-6 $1.50-2 $3-4 $4 $20 ----- $34-37

$1 $7-10 $2-3 $3-5 $5 $10 $5-7 $33-41

t = metric tonne

Table 7. Summary of Identity Preserving Production and Marketing (IPPM) System Cost Studies. (Source: Smyth et al. 2004)

Commodity (terms) Year

IPPM Cost per Tonne

of Commodity Non-GM Canola (FOB Vancouver vessel; minimum) Non-GM Soybeans (FOB export position; minimum) Non-GM Maize (primary elevator to export elevator) Non-GM Soybeans (primary elevator to export elevator) Food Maize (FOB inland elevator) High Oil Maize (FOB inland elevator) Food Soybeans (FOB inland elevator) STS Soybeans (FOB inland elevator) GM canola (crushed domestically) Soybeans (container, in store Japan, no producer costs or testing)

2004 Est. 2004 Est. 2004 Est. 2004 Est. 1998 1998 1998 1998 1996 2000

C$25.90-$30.65 US$35.53-$40.92 US$12.47 US$28.54 US$43.22 US$14.01 US$91.58 US$17.99 C$33-$41 US$27.72

C$ is Canadian dollar.