Preliminary Response

Paper No. __ Filed: July 28, 2015

04841.00006/7042666.1

UNITED STATES PATENT AND TRADEMARK OFFICE

________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD

________________

COALITION FOR AFFORDABLE DRUGS VI LLC Petitioner,

v.

CELGENE CORPORATION Patent Owner

________________

Case IPR2015-01103 Patent 6,315,720

________________

PATENT OWNER PRELIMINARY RESPONSE

PURSUANT TO 35 U.S.C. § 313 AND 37 C.F.R. § 42.107

PROTECTIVE ORDER MATERIAL

Patent Owner Preliminary Response IPR2015-01103 Patent 6,045,501

04841.00006/7042666.1 - i -

TABLE OF CONTENTS

Page

I. INTRODUCTION ........................................................................................... 1

II. BACKGROUND ............................................................................................. 3

A. The Challenge of Protecting A Fetus From A Teratogenic Drug While Allowing A Patient Access to Its Efficacy........................ 5

B. Previous Attempts to Control Access to Other Drugs Were Unsuccessful ........................................ 11

C. The ’720 Patent ................................................................................... 13

III. ARGUMENT ................................................................................................. 19

A. The Petition Should Be Denied Because CFAD Has Failed To Show A Reasonable Likelihood That The Claims Are Obvious .......................................... 20

1. CFAD’s Expert Declaration Is Entitled To “little or no weight” ............................................... 20

(a) Dr. Fudin Is Not A POSA ............................................... 20

(b) Dr. Fudin’s Opinions Are Unsupported, Verbatim Recitations of CFAD’s Conclusory Arguments .................................... 21

2. The Petition Fails To Address The Redundancy With The 1096, 1102 And 1103 Petitions ................................ 22

3. Ground 1: The Claimed Inventions Would Not Have Been Obvious Over Mitchell, Dishman, and Cunningham ....................................... 24

(a) CFAD Fails To Address Both the Claims and the Prior Art “As a Whole” ......................... 25

(b) CFAD Has Not Provided A Motivation To Combine Mitchell With Dishman or Cunningham ........ 26



04841.00006/7042666.1 - ii -

(c) The Ground 1 References Fail To Disclose, Teach, or Suggest Key Elements of the Claimed Inventions ....................... 32

i. Claim 1 Would Not Have Been Obvious Over Mitchell, Dishman, And Cunningham ........ 32

1) Element 1(c) ............................................... 32

2) Element 1(d) ............................................... 35

3) Element 1(e) ............................................... 35

ii. Dependent Claims 2–27 Would Not Have Been Obvious Over Mitchell In View Of Dishman And The Knowledge Of A POSA ........ 38

1) Claim 5 ....................................................... 39

2) Claim 6 ....................................................... 40

3) Claim 9 ....................................................... 41

4) Claim 10 ..................................................... 42

5) Claims 13, 14, 23-25 .................................. 42

6) Claim 17 ..................................................... 44

7) Claims 18 and 19 ....................................... 44

iii. Independent Claim 28 And Dependent Claims 29–32 Would Not Have Been Obvious Over Mitchell In View Of Dishman In Further View Of Cunningham And Knowledge Of A POSA ....................................... 46

1) Claim 28 ..................................................... 47

2) Claims 29-32 .............................................. 47

4. CFAD Fails To Address The Objective Evidence of Nonobviousness Regarding the ’720 Patent ............................. 49



04841.00006/7042666.1 - iii -

(a) Long-felt Need Further Supports The Nonobviousness Of The Claimed Inventions ................. 49

(b) Commercial Success Further Supports The Nonobviousness Of The Claimed Inventions ................. 51

(c) Third-Party Praise And Awards Further Supports The Nonobviousness Of The Claimed Inventions ......... 52

(d) Licensing by Others Further Supports The Nonobviousness Of The Claimed Inventions ................. 52

(e) Unexpected Results Further Supports The Nonobviousness Of The Claimed Inventions ................. 53

B. The Petition Should Be Denied Under 35 U.S.C. §§ 314(a) and 316(b) ............................................... 54

C. The Petition Should Be Denied For Failing To Name All Real Parties-In-Interest ..................................... 57

IV. CONCLUSION .............................................................................................. 60



04841.00006/7042666.1 - 1 -

I. INTRODUCTION

Pursuant to 35 U.S.C. § 313 and 37 C.F.R. § 42.107(a), Patent Owner

Celgene Corporation (“Celgene”) submits this Preliminary Response to Coalition

For Affordable Drugs VI LLC’s (“CFAD”) Petition for Inter Partes Review (the

“Petition”) of U.S. Patent No. 6,315,720 (the “’720 patent”).

The ’720 patent describes and claims improved methods for delivering a

potentially dangerous drug to a patient (including teratogenic drugs such as

thalidomide) while avoiding the occurrence of adverse side effects (such as birth

defects of the type associated with thalidomide). The inventions were conceived as

part of Celgene’s efforts to significantly improve its existing program for

controlling patient access to thalidomide, which was known as the System for

Thalidomide Education and Prescribing Safety, or S.T.E.P.S.® The improved

program, which Celgene called Enhanced S.T.E.P.S.®, is an embodiment of the

’720 patent and has been used in connection with thalidomide and other potentially

teratogenic pharmaceutical products since 2001. During that time it has

successfully prevented 100% of drug-related birth defects. In fact, the inventions

of the ’720 patent were so successful and innovative that the FDA required other

drug manufacturers to copy Celgene’s patented methods if they wanted to keep

their products on the market, resulting in licenses to several of Celgene’s patents,

including the ’720 patent.



04841.00006/7042666.1 - 2 -

Notwithstanding Celgene’s significant innovation, CFAD has filed the

present Petition as part of a hedge fund investment strategy developed by the real

parties-in-interest (“RPI”). CFAD’s Petition, however, has several fatal defects.

First, CFAD relies heavily on an expert declaration that is entitled to little or

no weight because the declarant is not a person of ordinary skill in the art (POSA),

and because the declaration merely reiterates CFAD’s conclusory arguments.

Second, CFAD’s Ground 1 obviousness argument fails on the merits because

none of the cited references disclose, teach, or suggest all elements of the

challenged claims. CFAD’s proposed modifications to, and combinations of, the

alleged prior art are driven entirely by hindsight and supported only by conclusory

assertions by CFAD’s declarant, who merely parrots the arguments in the Petition.

There is also no rational basis for combining CFAD’s references, as they are

directed to different endeavors. For example, one of CFAD’s cited references,

Cunningham, is not related in any way to restricted drug distribution, assessing

risks associated with pharmaceuticals, or preventing the occurrence of adverse

events, let alone to teratogenic side effects. Ground 1 does not warrant an IPR trial

of the ’720 patent.

Third, CFAD’s Petition is an improper use of the IPR proceedings.

Specifically, CFAD and the RPI are abusing and misusing the IPR process in an

attempt to effectuate changes in the stock prices of the targeted innovator



04841.00006/7042666.1 - 3 -

pharmaceutical companies. The self-serving actions of the RPI create unwarranted

burdens for both patent owners and the Board. The Board should exercise its

discretion under 35 U.S.C. §§ 314(a) and 316(b) and deny the Petition.1

Fourth and finally, the Petition should also be denied because it fails to

name all RPI—a threshold requirement for an IPR. Specifically omitted from the

RPI identification are the investors in the Hayman funds responsible for filing the

Petition. Having failed to name all RPI, the Petition cannot be considered.

II. BACKGROUND

Celgene is a biopharmaceutical company that is committed to improving the

lives of patients through research and development of drug products that treat

cancers and other devastating conditions. Three drug products developed by

Celgene are relevant to this IPR: Thalomid®, Revlimid®, and Pomalyst®.

Thalomid® is approved for the treatment of (1) erythema nodosum leprosum

(“ENL”)—an inflammatory condition associated with leprosy; and (2) newly

diagnosed multiple myeloma (“MM”) (in combination with dexamethasone). Ex.

2001 at 1. The active ingredient in Thalomid® is thalidomide, which is well known

for its teratogenicity (or ability to cause severe birth defects). Unfortunately, as

described in more detail below, the devastating effects of thalidomide were felt

1 With the Board’s permission, Celgene has separately moved to dismiss the

Petition pursuant to 37 C.F.R. § 42.12.



04841.00006/7042666.1 - 4 -

worldwide during the thalidomide tragedy of the 1950s and 1960s, and continue

today for the surviving victims. See, e.g., Ex. 2002 at 1; Ex. 1001 at 1:40-45.

Revlimid® is approved for the treatment of (1) transfusion-dependent anemia

due to low- or intermediate-1-risk myelodysplastic syndromes associated with a

deletion 5q abnormality with or without additional cytogenetic abnormalities;

(2) MM (in combination with dexamethasone); and (3) mantle cell lymphoma in

certain patients whose disease has relapsed or progressed after two prior therapies.

Ex. 2003 at 1. Pomalyst® (in combination with dexamethasone) is approved for

the treatment of MM in patients who have received at least two prior therapies and

whose disease has progressed. Ex. 2004 at 1. The active ingredient in Revlimid®

is lenalidomide, and the active ingredient in Pomalyst® is pomalidomide. Ex. 2003

at 1; Ex. 2004 at 1. According to the FDA-approved package inserts, lenalidomide

and pomalidomide are “thalidomide analogue[s],” and if these drugs are “used

during pregnancy, [they] may cause birth defects or embryo-fetal death.” Ex. 2003

at 1, 3, 7, 22-23; Ex. 2004 at 1, 2, 4, 14, 24. As described herein, due in large

measure to the inventions claimed in the ’720 patent (and other of Celgene’s

patents), Celgene’s FDA-approved systems for preventing fetal exposure to the

active ingredients in Thalomid®, Revlimid®, and Pomalyst® have been 100%

successful in preventing drug-related birth defects.



04841.00006/7042666.1 - 5 -

A. The Challenge of Protecting A Fetus From A Teratogenic

Drug While Allowing A Patient Access to Its Efficacy

Beginning in 1958, thalidomide was marketed in Europe as a sedative and a

treatment for pregnant women with morning sickness. See Ex. 1001 at 1:40-45;

Ex. 2002 at 1. Shortly after entering the European market, it was discovered that

thalidomide caused deformities in children born to mothers who had taken the drug

during pregnancy. Id. As a result, by 1962, thalidomide had been withdrawn from

all markets. Id. By then, the damage had been done. Thalidomide had been

linked to more than 10,000 birth defects in at least 46 countries. Id. The birth

defects were severe. Some children were born with missing or abnormal limbs,

feet, or hands, a condition known as phocomelia (from the Greek for “seal”

“limb”). Id. Many other deformities and complications were linked to

thalidomide, including abnormal or absent ears, and heart and kidney problems.

Id. As a result of this tragedy, drug regulatory authorities worldwide, including the

FDA, revised their regulations to ensure that new drugs were screened for safety in

addition to efficacy, and were specifically investigated for their potential to cause

harm to a developing fetus. Id.; Ex. 2005.

Years later, it became clear that despite its teratogenicity, thalidomide had

the power to benefit certain patient populations. Ex. 1001 at 1:46-62.

Accordingly, Celgene believed it would be beneficial if the drug were made

available to those patient populations. Due to its known teratogenicity, however,



04841.00006/7042666.1 - 6 -

Celgene realized that to market thalidomide, it needed to develop a system that

would allow patients in need of thalidomide to access it while ensuring that no

thalidomide-related birth defects would occur. Ex. 1001 at 1:59-64.

In other words, Celgene was faced with a great challenge—find a way to

effectively avoid the teratogenic side effects of thalidomide while still making the

drug available to patients in need. Indeed, Celgene recognized that it would need

to create a system that was so effective at preventing birth defects of the type

associated with teratogenic drugs that it would convince skeptical FDA regulators,

and understandably vocal and concerned thalidomide victims around the world,

that Thalomid® could be safely marketed. In other words, the new system would

need to convince the FDA to lift its nearly 40-year ban on thalidomide.

Celgene rose to meet this challenge, first developing the methods claimed in

U.S. Patent No. 6,045,501 (the “Elsayed ’501 patent”) (Ex. 1003) (described in

Celgene’s Preliminary Response to IPR2015-01092). Celgene then improved upon

its own work by inventing the methods claimed in the ’720 patent (described

below). In other words, Celgene’s inventions for controlling access to thalidomide

(and other teratogens) came in two phases. Celgene first developed S.T.E.P.S.®,

which is claimed in the Elsayed ’501 patent and was first implemented in the

original July 1998 Thalomid® package insert (“July 1998 Thalomid PI,” Ex. 2032).

The initial implementation of S.T.E.P.S.® only retrospectively identified patient



04841.00006/7042666.1 - 7 -

behaviors that posed a risk of fetal exposure to thalidomide. See Ex. 2006 at 328.

While this patented system was 100% successful in preventing birth defects of the

type associated with thalidomide in patients taking Thalomid®, Celgene (and the

inventors of the ’720 patent) still saw room for significant improvement.

Thus, the ’720 patent’s inventors developed changes that transformed

S.T.E.P.S.® from a retrospective program into a prospective program that they

called Enhanced S.T.E.P.S.® See id. Specifically, from October 1999 to

September 2001, the inventors focused on implementing changes that were

“designed to identify patients exhibiting behaviours providing risk for fetal

exposure and to remedy those behaviours prior to dispensing thalidomide.” Id.

(emphasis added).



04841.00006/7042666.1 - 8 -

As explained in the ’720 patent, the inventors

came up with the authorization process of Enhanced S.T.E.P.S.® that is

conducted by the registered pharmacy consulting the computer readable

medium to retrieve a prescription approval code. Ex. 1001 at 13:43-45.

Enhanced S.T.E.P.S.® was implemented for the first time with the September

2001 Thalomid® package insert (“September 2001 Thalomid PI”), which is not

prior art to the ’720 patent. See Ex. 2008. Specifically, the September 2001

Thalomid PI included the significant improvements claimed in the ’720 patent.

Those improvements are reflected in the revisions to the “Storage and Dispensing”

section of the July 1998 Thalomid PI, which CFAD relies upon in IPR2015-01096:

DRUG MUST ONLY BE DISPENSED IN NO MORE THAN A 1

MONTH SUPPLY AND ONLY ON PRESENTATION OF A NEW

PRESCRIPTION WRITTEN WITHIN THE PREVIOUS 7 DAYS.

SPECIFIC INFORMED CONSENT (copy attached as part of this

package insert) AND COMPLIANCE WITH THE MANDATORY

PATIENT REGISTRY AND SURVEY ARE REQUIRED FOR ALL



04841.00006/7042666.1 - 9 -

PATIENTS (MALE AND FEMALE) PRIOR TO DISPENSING BY

THE PHARMACIST.

BEFORE DISPENSING THALOMID® (thalidomide), YOU MUST

ACTIVATE THE AUTHORIZATION NUMBER ON EVERY

PRESCRIPTION BY CALLING THE CELGENE CUSTOMER

CARE CENTER AT 1-888-4-CELGENE (1-888-423-5436) AND

OBTAINING A CONFIRMATION NUMBER. YOU MUST ALSO

WRITE THE CONFIRMATION NUMBER ON THE

PRESCRIPTION. YOU SHOULD ACCEPT A PRESCRIPTION

ONLY IF IT HAS BEEN ISSUED WITHIN THE PREVIOUS 7

DAYS (TELEPHONE PRESCRIPTIONS ARE NOT PERMITTED);

DISPENSE NO MORE THAN A 4-WEEK (28- DAY) SUPPLY,

WITH NO AUTOMATIC REFILLS; DISPENSE BLISTER PACKS

INTACT (CAPSULES CANNOT BE REPACKAGED); DISPENSE

SUBSEQUENT PRESCRIPTIONS ONLY IF FEWER THAN 7

DAYS OF THERAPY REMAIN ON THE PREVIOUS

PRESCRIPTION; AND EDUCATE ALL STAFF PHARMACISTS

ABOUT THE DISPENSING PROCEDURE FOR THALOMID®

(thalidomide).

Ex. 2009 at 4; see Ex. 2008 at 20.

Today, Enhanced S.T.E.P.S.® is known as the Thalomid REMS, which is an

FDA-approved Risk Evaluation and Mitigation Strategy (“REMS”). Enhanced

S.T.E.P.S.® is claimed in Jepson format in the ’720 patent as an improvement over

S.T.E.P.S.®, which was originally claimed in the ’501 patent. Celgene’s continued

success in preventing birth defects due to Thalomid®, Revlimid®, and Pomalyst® is



04841.00006/7042666.1 - 10 -

in large measure due to the inventions of the ’720 patent and other Celgene patents

(including the ’501 patent). Had Celgene not developed the claimed methods for

preventing fetal exposure to teratogenic (or potentially teratogenic) drugs,

Thalomid®, Revlimid®, and Pomalyst® would not be available to patients today.

The FDA first approved Thalomid® for the treatment of ENL in July 1998

subject to 21 CFR § 314.520 (“Subpart H”), which allowed the FDA to approve

drugs that were shown to be effective, but that could only be used safely under

restricted conditions. Ex. 1031 at 0002. Specifically, the approval letter stated that

Thalomid® was being approved because Celgene presented adequate information

to demonstrate that the drug would be safe and effective for use when marketed

under S.T.E.P.S.®, which is an embodiment of the methods described in Celgene’s

patents. Id. The FDA subsequently approved Celgene’s labeling changes that

incorporated the improvements of Enhanced S.T.E.P.S.®, which is an embodiment

of the methods described in the ’720 patent. See Ex. 2009 at 4; Ex. 2008 at 20.

Then, when Thalomid® was approved for the treatment of newly diagnosed MM in

2006, the FDA maintained the same restrictions on distribution of the drug. Ex.

2010 at 1. Similarly, the FDA conditioned its approval of both Revlimid® and

Pomalyst® (for all indications), on Celgene’s use of the same restrictions applied to

Thalomid®. Ex. 2011 at 1; Ex. 2012 at 7-10.



04841.00006/7042666.1 - 11 -

B. Previous Attempts to Control Access to

Other Drugs Were Unsuccessful

Others attempted to control access to the drugs Accutane® (isotretinoin) and

Clozaril® (clozapine) before Celgene invented S.T.E.P.S.®, but as described below,

those attempts failed to meet their goals. Both failed to provide a workable

solution for dealing with a drug like thalidomide, where even a single drug-related

birth defect was unacceptable.

Accutane® contains isotretinoin, a form of vitamin A (a vitamin A analogue)

that has been used as a treatment for severe cystic acne since 1982. Ex. 1010 at

101. Isotretinoin can and has caused birth defects in children whose mothers are

taking the drug. Id. Because of the teratogenic effects of isotretinoin, and instead

of removing the drug from the market, the manufacturer of isotretinoin

implemented the Pregnancy Prevention Program (“PPP”).

The PPP, however, was not effective in preventing women who were taking

Accutane® from becoming pregnant or preventing birth defects. In fact, the

Mitchell reference that CFAD relies upon (Ex. 1010), discloses that there were

over 400 pregnancies that occurred during isotretinoin treatment, at least six of

which resulted in live born infants with at least minor anomalies, and at least one

with major anomalies. Id. at 103-04. Mitchell further reports that in the first

several months of the PPP, there was incomplete compliance with several parts of

the program, such as failures to ensure negative pregnancy tests before beginning



04841.00006/7042666.1 - 12 -

treatment, failures to wait until menses begins before treatment, and failures to use

effective birth control before, during, and after treatment. Id. at 104. Compliance

remained incomplete even after the manufacturer changed the medication package

to highlight the most important parts of the PPP. Id.

Publications in prominent scientific journals also disparaged the PPP. For

instance, an article in the New England Journal of Medicine reported that “[s]ince

the [PPP] program was implemented in 1989, a substantial number of fetuses have

been exposed to the drug. As many as 30 percent of the women with exposed

fetuses did not use any mode of contraception, even though they were cognizant of

the high fetal risk.” Ex. 2013 at 1130.

Clozaril® contains clozapine, and is used to treat schizophrenia. Ex. 1007 at

899. Its release into the market was permitted only with certain prescribing and

distribution restrictions implemented by the manufacturer. Id. These restrictions

were put in place because the use of clozapine is associated with a high frequency

of agranulocytosis, a potentially fatal blood disorder. Ex. 2014 at 52. Clozapine is

not a teratogenic drug—it does not cause birth defects—and the system employed

to monitor its use was not tailored to restrict or prevent birth defects. Instead, the

manufacturer developed a program called the Clozaril National Registry to

“enhance patient safety by facilitating early detection of potentially dangerous

white blood cell suppression.” Id. at 52-53. But like the PPP, the clozapine system



04841.00006/7042666.1 - 13 -

was also a failure. Specifically, during its first five years, 382 patients developed

agranulocytosis, and 12 of those patients died as a result. Id. at 55.

When Celgene invented S.T.E.P.S.® and the methods claimed in the ’501

patent, the programs in place for both the Accutane® and Clozaril® were

ineffective. The ’501 patent was a significant innovation over those systems and,

as noted above, the ’720 patent was a significant improvement over the ’501

patent. To date, Celgene’s FDA-approved systems for preventing fetal exposure to

the active ingredients in Thalomid®, Revlimid®, or Pomalyst®—all of which are

covered by the claims of the ’720 patent—have been 100% successful in

preventing birth defects of the type associated with thalidomide. Celgene’s

patented methods were so successful that, in 2006, it became clear that the PPP

(and later attempts at managing the distribution of isotretinoin) were ineffective by

comparison. Consequently, the FDA required the manufacturers of isotretinoin to

use Celgene’s patented methods if they wanted to keep their products on the

market. This resulted in licenses to several of Celgene’s patents, including the

’720 patent, in connection with the distribution of isotretinoin under a program

known as iPledge. See Ex. 2015 at 4, 8-9; Ex. 2016 at 1; and Ex. 2017 at 1.

C. The ’720 Patent

As discussed above, the ’720 patent describes and claims improved methods

for delivering a potentially dangerous drug to a patient (including teratogenic drugs



04841.00006/7042666.1 - 14 -

such as thalidomide) while avoiding the occurrence of adverse side effects (such as

birth defects). See Ex. 1001 at Abstract; 1:13-16. The claims are set forth in

Jepson format, with a preamble comprising a general description of the known

elements of the claimed invention (which are claimed in the Elsayed ’501 patent,

and embodied in S.T.E.P.S.® and the July 1998 Thalomid PI), and the novel,

improved elements (which are embodied in Enhanced S.T.E.P.S.® and the

September 2001 Thalomid PI) set forth after the “improvement comprising”

clause. See 37 C.F.R. § 1.75(e) (describing the preferred format for claims to

improvements). For example, independent Claim 1 of the ’720 patent recites:

In a method for delivering a drug to a patient in need of the drug,

while avoiding the occurrence of an adverse side effect known or

suspected of being caused by said drug, wherein said method is of the

type in which prescriptions for said drug are filled only after a

computer readable storage medium has been consulted to assure that

the prescriber is registered in said medium and qualified to prescribe

said drug, that the pharmacy is registered in said medium and

qualified to fill the prescription for said drug, and the patient is

registered in said medium and approved to receive said drug, the

improvement comprising:

a. defining a plurality of patient risk groups based upon a predefined

set of risk parameters for said drug;

b. defining a set of information to be obtained from said patient,

which information is probative of the risk that said adverse side

effect is likely to occur if said drug is taken by said patient;



04841.00006/7042666.1 - 15 -

c. in response to said information set, assigning said patient to at least

one of said risk groups and entering said risk group assignment in

said medium;

d. based upon said information and said risk group assignment,

determining whether the risk that said adverse side effect is likely

to occur is acceptable; and

e. upon a determination that said risk is acceptable, generating a

prescription approval code to be retrieved by said pharmacy before

said prescription is filled.

Ex. 1001 at Claim 1. The only other independent claim, Claim 28, includes all of

the elements of Claim 1 and adds the following further limitation: “wherein said

adverse side effect is likely to arise in patients who take said drug in combination

with at least one other drug.” Id. at Claim 28. The dependent claims further limit

and define the controls to avoid the occurrence of adverse side effects. See id. at

Claims 2-27, 29-32.

During prosecution, the Examiner extensively considered the most

significant prior art, namely the Elsayed ’501 patent, and concluded that it did not

disclose, teach, or suggest the ’720 patent’s inventions. Specifically, the Examiner

found that “Elsayed et al. teaches a method for delivering a drug to a patient while

preventing the exposure to a fetus or to other individuals with contraindications to

the drug that includes registering prescribers, pharmacists, and patients in a

computer readable storage medium for authorizing and monitoring distribution of



04841.00006/7042666.1 - 16 -

the drug,” but noted that Elsayed “does not teach a step of generating of a

prescription number or code, which can be retrieved by a pharmacy before said

prescription is filled.” Ex. 1002 at 0091 (emphasis added). In other words, the

Examiner concluded that the Elsayed ’501 patent disclosed only what was in the

preamble of the ’720 patent’s Jepson claims, but did not disclose, teach, or suggest

the claimed improvements, which were not implemented or publicly disclosed until

the September 2001 Thalomid PI.

With respect to the “approval code” limitation, the Examiner originally

rejected the claims over U.S. Patent No. 6,202,923 to Boyer (Ex. 1005), stating

that “Boyer et al, teaches a method of an automated pharmacy system (see

abstract) that improves prescription processing,” and that it would have been

obvious to “use an automated pharmacy arrangement as taught by Boyer et al.

which includes a step for generating a prescription number or code associated with

said prescription by a computer workstation.” Ex. 1002 at 0092. The ’720 patent’s

inventors overcame this rejection.

Specifically, the inventors explained that the automated prescription code in

Boyer was a number that was merely associated with every prescription, whereas

the prescription approval code of the ’720 patent was based on a risk assessment of

information collected from the patient, and was generated only if and when the risk

was deemed acceptable. The inventors noted that the claims required “an



04841.00006/7042666.1 - 17 -

affirmative decision” to be made before the prescription was filled, and Boyer did

not require such an approval code, nor did it require an approval code to be

generated before the prescription may be filled. Id. at 0106-07. In response to this

argument, the Examiner allowed all the claims of the ’720 patent.2

Here, CFAD relies on three references in its Petition to argue that “Claims 1-

32 are unpatentable under 35. U.S.C. § 103”: Mitchell (Ex. 1010), Dishman (Ex.

1007) and Cunningham (Ex. 1008). Pet. at 11. Mitchell and Cunningham do not

contain any new material, information, or disclosures that were not previously

considered by the Examiner during prosecution of the ’720 patent, and Dishman

does not disclose anything that is relevant to the claimed inventions.

Mitchell is a 1995 article that “report[s] the results of an ongoing survey

designed to assess compliance with” Accutane’s PPP, which was implemented in

an attempt to minimize pregnancies among women exposed to the drug

2 The Examiner’s first Office Action also indicated that Claims 28-32 were

allowable if written in independent form because “none of the prior art, either

alone or in combination, teaches or suggests a method for delivering a drug to a

patient while preventing exposure to a contraindicated individual that includes

screening for possible interaction between the drug and a second drug that has been

administrated to the subject.” Ex. 1002 at 0059.



04841.00006/7042666.1 - 18 -

isotretinoin. Ex. 1010 at 101. Dr. Allen Mitchell, who designed the survey, was a

physician with the Slone Epidemiology Unit of Boston University. Id.

The examiner did not cite Mitchell during prosecution, but the Background

of the Invention section of the ’720 patent states that:

“[m]ethods for monitoring and educating patients to whom a drug is

distributed have been developed in connection with Accutane … [a]

pregnancy prevention program was developed, and the Slone

Epidemiology Unit of Boston University designed and implemented a

survey to evaluate these efforts … [t]he survey identified relatively

low rates of pregnancy during Accutane treatment, which suggests

that such a program can be effective … [e]nrollment in the Accutane

survey is voluntary … assessing the representativeness of the women

who have been enrolled in the survey has been problematic, and it has

been difficult to determine whether the survey results can be

generalized to all female Accutane users.”

Ex. 1001 at 2:13-35. Accordingly, the Examiner considered the relevant

disclosures from Mitchell regarding Accutane’s PPP and the Slone survey

evaluating its effectiveness.

With respect to Cunningham, the Examiner considered a divisional of that

patent, U.S. Patent No. 6,055,507 (the “’507 patent”), which shares the same

specification as Cunningham and is listed on the face of the ’720 patent. See Ex.

1001 at References Cited. CFAD does not cite any disclosures from Cunningham

that were not already before the Examiner.



04841.00006/7042666.1 - 19 -

Moreover, CFAD uses Cunningham to make the same arguments that were

before the Examiner based on Boyer; in other words, that it would have been

obvious to use the approval code disclosed in Boyer (or Cunningham) to arrive at

the claimed inventions. See Pet. at 23-24. CFAD ignores that this Office has

already considered substantially the same art and arguments CFAD presents, and

concluded that the ’720 patent’s inventions are not disclosed, taught, or suggested

by the prior art.

With respect to Dishman, the Examiner would not have found any

disclosures or information to be relevant to the claimed inventions because the

clozapine distribution system, as a whole, taught away from using such a system to

prevent adverse side effects. Specifically, the Examiner would have considered the

disclosures in Dishman to be irrelevant because the clozapine system was

unsuccessful after 382 patients developed agranulocytosis in the first five years,

and 12 of those patients died as a result. Ex. 2014 at 55.

III. ARGUMENT

The Petition should be denied due to several fatal defects. First, CFAD

relies heavily on an expert declaration that is entitled to little or no weight because

the declarant is not a POSA, and because the declaration merely parrots CFAD’s

conclusory arguments. Second, CFAD’s Ground 1 obviousness argument fails to

establish a reasonable likelihood that the challenged claims are unpatentable.



04841.00006/7042666.1 - 20 -

Third, CFAD’s Petition is an improper use of the IPR process, and should be

denied under 35 U.S.C. §§ 314 and 316. Fourth, CFAD’s Petition fails to name all

RPI as required by 35 U.S.C. § 312(a)(2).

A. The Petition Should Be Denied Because CFAD Has Failed To

Show A Reasonable Likelihood That The Claims Are Obvious

1. CFAD’s Expert Declaration Is

Entitled To “little or no weight”

As discussed below, CFAD’s Petition relies heavily on the declaration of Dr.

Jeffrey Fudin. See generally Pet. (citing Ex. 1027). Dr. Fudin’s opinions,

however, are entitled to little or no weight, including because: (1) Dr. Fudin’s

experience does not qualify him as a POSA; and (2) his opinions are unsupported,

verbatim recitations of CFAD’s conclusory arguments.

(a) Dr. Fudin Is Not A POSA

Dr. Fudin does not have the knowledge of a POSA or any experience with

the problem that is solved by the ’720 patent. CFAD defines a POSA as a

pharmacist. Pet. at 17. In particular, a POSA under CFAD’s definition “would

typically have either a Pharm.D. or a BS in pharmacy with approximately 5-10

years of related experience and a license to practice as a registered pharmacist in

any one or more of the United States.” Id. This definition improperly ignores the

true field of the claimed inventions, which is avoidance of adverse events

associated with drug products (i.e., pharmaceutical drug product risk management).

As such, Celgene disagrees with CFAD’s definition of a POSA, and proposes that



04841.00006/7042666.1 - 21 -

a POSA would have had at least a bachelor’s degree and at least 2 years of

experience in risk management relating to pharmaceutical drug products, or a B.S.

or M.S. in pharmaceutical drug product risk management or a related field.

Celgene’s definition of a POSA is supported by the claims and specification

of the ’720 patent. See generally Ex. 1001. CFAD’s POSA, on the other hand, is

intentionally shaped to mirror the skills of Dr. Fudin and the alleged prior-art

references on which CFAD relies. While Dr. Fudin “work[ed] in close

collaboration with medical staff members in the management of various acute and

chronic pain disease states,” and has “experience with . . . computerized billing and

patient record systems” through his work as a pharmacist (Ex. 1027 at ¶¶ 5, 8), he

does not have the knowledge of a POSA or any experience with the problem that is

solved by the ’720 patent. For example, a POSA would have experience in

designing and implementing systems for controlling access to pharmaceutical drug

products that have the potential for life threatening adverse events—the subject

matter of the ’720 patent. Dr. Fudin’s opinions regarding patentability of the ’720

patent claims are not from the view of a POSA and are thus entitled to no weight.

(b) Dr. Fudin’s Opinions Are Unsupported, Verbatim

Recitations of CFAD’s Conclusory Arguments

Dr. Fudin’s opinions are also entitled to little or no weight because they are

unsupported and largely verbatim recitations of CFAD’s conclusory arguments.

See infra. To the extent that Dr. Fudin fails to cite any underlying factual bases for



04841.00006/7042666.1 - 22 -

his opinions, they are entitled to little or no weight. See 37 C.F.R. § 42.65(a);

Ashland Oil, Inc. v. Delta Resins & Refractories, Inc., 776 F.2d 281, 294 (Fed. Cir.

1985) (stating a lack of factual support for expert opinion “may render the

testimony of little probative value in a validity determination”).

Indeed, the Board has been clear that it will exercise its “well-established

discretion to give little weight to conclusory, unsupported expert testimony,” such

as testimony that “d[oes] not elaborate on [the Petitioner’s] position because it

simply repeat[s the Petitioner’s] conclusory statements verbatim.” TRW

Automotive US LLC v. Magna Elecs., Inc., IPR2014-00258, Paper 18 at 10-11

(Aug. 27, 2014); see also Apple Inc. v. Smartflash LLC, CBM2014-00111, Paper 7

at 21 (Sept. 30, 2014) (denying institution where the expert declaration “simply

reiterates [the Petitioner’s] contentions and conclusory reasoning”). Celgene

submits that the Board should do the same here.

2. The Petition Fails To Address The Redundancy

With The 1096, 1102 And 1103 Petitions

As an initial matter, the grounds and references in this Petition are redundant

to those made by Petitioner in IPR2015-01096 and IPR2015-01102. In Ground 2

of IPR2015-01096, CFAD relies on the July 1998 Thalomid PI, in combination

with Cunningham, as a secondary reference with respect to the limitations of

claims 1-32. See IPR2015-01096 Pet. at 11, 51-60. In Ground 1 of IPR2015-

01102, CFAD relies on Powell and Dishman, in combination with Cunningham,



04841.00006/7042666.1 - 23 -

with respect to the limitations of claims 1-32. See IPR2015-01102 Pet. at 11, 17-

44. Here, CFAD relies on Mitchell, in combination with Dishman and

Cunningham, with respect to the limitations of claims 1-32. Pet. at 11, 17-45.

While CFAD relies on different primary references, CFAD’s explanations of the

alleged disclosures of the references in this Petition are substantially the same as

CFAD’s explanations of the alleged disclosures of the references in Ground 2 of

IPR2015-01096 and in Ground 1 of IPR2015-01102.

The use of redundant references and arguments in multiple IPR petitions

contradicts regulatory and statutory mandates, and the Board has indicated that it

will not consider such grounds. See, e.g., Liberty Mut. Ins. Co. v. Progressive Cas.

Ins. Co., CBM2012-00003, Paper 7 at 1-2 (Oct. 25, 2012). Redundant grounds

impose a significant burden on the Board and the patent owner, and cause

unnecessary delay that jeopardizes completion by the statutory deadline. Id. at 2.

Because “[t]he Board seeks to streamline and converge issues at all phases

of the proceeding . . . at [the] time of institution the Board analyzes the petition on

a claim-by-claim, ground-by-ground basis, to eliminate redundant grounds.” Idle

Free Sys. v. Bergstrom, Inc., IPR2012-00027, Paper 26 at 4-5 (June 11, 2013).

The redundancy inquiry does not focus on “whether the applied prior art

disclosures have differences, for it is rarely the case that the disclosures of different

prior art references, will be literally identical.” EMC Corp. v. Personal Web



04841.00006/7042666.1 - 24 -

Techs., LLC, IPR2013-00087, Paper 25 at 3 (June 5, 2013). Instead, the

redundancy inquiry focuses on “whether the petitioner articulated a meaningful

distinction in terms of relative strengths and weaknesses with respect to application

of the prior art disclosures to one or more claim limitations.” Id. at 3-4. The

burden is on the petitioner to articulate such a “meaningful distinction.” ScentAir

Techs., Inc. v. Prolitec, Inc., IPR2013-00180, Paper 18 at 3 (Aug. 26, 2013).

CFAD does not articulate any relative strengths or weaknesses of its

multiple grounds, shed light on how any of the proposed obviousness combinations

improves on any of the other combinations, or explain why one primary reference

is better or worse than another. CFAD has thus failed to identify a “meaningful

distinction” among the alleged prior art combinations as applied to the claims. See

LaRose Indus., LLC v. Capriola Corp., IPR2013-00120, Paper 20 at 4 (July 22,

2013). Thus, the Board should not consider more than one alleged ground for each

of the challenged claims in the event that a trial is instituted.

3. Ground 1: The Claimed Inventions Would Not Have Been

Obvious Over Mitchell, Dishman, and Cunningham

CFAD’s Ground 1 lacks merit because CFAD: (1) fails to address both the

claims and the prior art as a whole; (2) does not identify a colorable motivation to

combine or modify the cited art; and (3) cannot show that the prior art disclosed,

taught, or suggested each and every element of the challenged claims.



04841.00006/7042666.1 - 25 -

(a) CFAD Fails To Address Both the

Claims and the Prior Art “As a Whole”

CFAD’s obviousness analysis is improper because it fails to address both the

claims and prior art as a whole. “[A] patent composed of several elements is not

proved obvious merely by demonstrating that each of its elements was,

independently, known in the prior art.” KSR Int’l Co. v. Teleflex Inc., 550 U.S.

398, 418 (2007). Indeed, “[i]f identification of each claimed element in the prior

art were sufficient to negate patentability, very few patents would ever issue.” In

re Rouffet, 149 F.3d 1350, 1357 (Fed. Cir. 1998). CFAD ignores this well settled

principal of patent law. Rather, with hindsight, CFAD parses each individual

element of the challenged claims for the purpose of identifying corresponding

portions of the prior art that allegedly suggest that element to a POSA. Not once

does CFAD address the combination of claim elements as a whole. For this reason

alone, its obviousness ground should be denied.

The law also requires that CFAD consider the prior art as a whole, including

teachings that point away from the claimed inventions. See Leo Pharm. Prods. v.

Rea, 726 F.3d 1346, 1355 (Fed. Cir. 2013). As described below, CFAD also fails

to conduct this critical analysis. Instead, CFAD picks and chooses only those

portions of the prior art that allegedly support its position. For this additional

reason, CFAD’s obviousness analysis is legally flawed and should be rejected.



04841.00006/7042666.1 - 26 -

(b) CFAD Has Not Provided A Motivation To

Combine Mitchell With Dishman or Cunningham

CFAD’s Petition should also be denied because it does not explain how the

teachings of the references would be arranged or combined by a POSA, or why a

POSA would have made such a combination without the benefit of hindsight. A

“Petitioner must show some reason why a [POSA] would have thought to combine

particular available elements of knowledge, as evidenced by the prior art, to reach

the claimed invention.” Heart Failure Techs., LLC v. Cardiokinetix, Inc.,

IPR2013-00183, Paper 12 at 9 (July 31, 2013). Indeed, obviousness “cannot be

sustained by mere conclusory statements; instead, there must be some articulated

reasoning with some rational underpinning to support the legal conclusion of

obviousness.” In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006).

The Board has repeatedly denied institution where a petitioner fails to

demonstrate that it would have been obvious to combine or modify references,

holding that conclusory or general statements regarding the reasons to combine or

modify references are insufficient. See, e.g., TRW Auto. US LLC v. Magna Elecs.

Inc., IPR2014-00293, Paper 19 at 17 (July 1, 2014); Moses Lake Indus. v. Enthone,

Inc., IPR2014-00243, Paper 6 at 20 (June 18, 2014); Biodelivery Sci. Int’l, Inc. v.

Monosol Rx, LLC, IPR2015-00167, Paper 6 at 26-27 (May 20, 2015). Absent an

articulated reasoning with a rational underpinning to support the legal conclusion

of obviousness, CFAD fails to establish a prima facie case of obviousness. See



04841.00006/7042666.1 - 27 -

TRW Auto. US LLC v. Magna Elecs., IPR2014-00257, Paper 16 at 7, 12 (June 26,

2014) (denying petition where Petitioner failed to set forth motivation to combine).

Here, CFAD has not provided any analysis that supports a motivation to

combine Mitchell, Dishman, and Cunningham. Its arguments should be rejected

because they: (1) are rooted in hindsight, using the ’720 patent’s disclosures and

the inventors’ own path as their basis to combine the references; (2) fail to account

for the fact that the references are directed to entirely different endeavors and also

teach away from the claimed inventions; and (3) ignore that there can be no

motivation to combine because the prior art does not disclose each and every

element of the claimed inventions.

First, CFAD improperly uses the ’720 patent’s disclosures and the

inventors’ own path as its basis to combine the references. Specifically, CFAD

alleges that a POSA: (a) would look to “Mitchell for guidance on ‘an approach

that seeks to keep the drug available while minimizing the . . . hazard,’ and would

garner from its recommendations for ‘delivering a drug to a patient in need of the

drug, while avoiding the occurrence of an adverse side effect known or suspected

of being caused by said drug’” (Pet. at 18 (quoting Ex. 1027 at ¶ 76)) (emphasis

added); and then (b) once “[a]rmed with the disclosure of Mitchell . . . ‘would have

been motivated to look to the system disclosed in Dishman to further implement a

computerized registry for “delivering a drug to a patient in need of the drug,



04841.00006/7042666.1 - 28 -

while avoiding the occurrence of an adverse side effect known or suspected of

being caused by said drug.”’” Pet. at 21 (quoting Ex. 1001 at 18:16-18) (emphasis

added).

CFAD admits, however, that the emphasized language is from the ’720

patent’s preamble. Id.; Ex. 1001 at 18:17-19. Thus, CFAD “defin[es] the nature of

the problem to be solved as the specific problem solved by the

invention . . . [CFAD] has relied on impermissible hindsight to supply the reason

to combine.” Purdue Pharma L.P. v. Depomed, Inc., IPR2014-00379, Paper 72 at

28 (July 8, 2015); see also Texas Instruments Inc. v. Vantage Point Tech. Inc.,

IPR2014-01105, Paper 8 at 17 (Jan. 5, 2015) (finding it impermissible to use the

challenged patent’s description of the invention as a roadmap to piece together the

prior art). As such, its argument lacks merit.3

CFAD’s argument also fails because it relies on the inventors’ own path.

Specifically, CFAD argues that a POSA “did look to the same isotretinoin system

3 Moreover, the only support for CFAD’s conclusory argument is its expert’s

recitation of the same conclusory statement without the “rational underpinning”

required to establish obviousness. See Pet. at 21 (citing Ex. 1027 at ¶ 88.) Dr.

Fudin’s opinion should be given little to no weight because it “d[oes] not elaborate

on [CFAD’s] position because it simply repeat[s CFAD’s] conclusory statements

verbatim.” TRW Auto., IPR2014-00258, Paper 18 at 10.



04841.00006/7042666.1 - 29 -

described in Mitchell with respect to the ’720 Patent” based on statements made by

one of the ’720 patent’s inventors. Pet at 21. But CFAD has not presented any

evidence that the ’720 patent’s inventors are a POSA under CFAD’s definition or

otherwise. In fact, the inventors are not pharmacists and, therefore, would not fall

under CFAD’s definition of a POSA. Further, it is well settled that the thought

process of the inventor is not evidence of what would have been known or done by

a POSA. See Interconnect Planning Corp. v. Feil, 774 F.2d 1132, 1138 (Fed. Cir.

1985). Thus, CFAD’s hindsight-driven motivation argument should be rejected.

Second, CFAD fails to account for the fact that the prior-art references are

directed to completely different endeavors and also teach away from the claimed

inventions. Specifically, CFAD ignores that a POSA would not have been

motivated to combine Dishman with Mitchell because Dishman is not directed to

the same endeavor as Mitchell—a system designed to prevent exposure of a drug

to a fetus and thus, fetal birth abnormalities. Further, a POSA looking to develop a

system for universal use would not find a distribution system isolated to the

Department of Veteran Affairs (as Dishman is) to be particularly instructive. In

that regard, the risks associated with commercial pharmacy distribution of a

teratogenic drug are far more complex, and require different management, than

distribution to a small group of individuals at the Department of Veterans Affairs.



04841.00006/7042666.1 - 30 -

Furthermore, even if a POSA would look to the system disclosed in

Dishman, they would not ignore the prior art as a whole, which teaches away from

the Clozaril system discussed in Dishman because the system, as a whole, was a

failure. See supra at II.B. As such, a POSA would have been taught away from

combining Dishman with other references, including Mitchell or Cunningham. A

reference that teaches away cannot serve as a basis for obviousness. See In re

Gurley, 27 F.3d 551, 553 (Fed. Cir. 1994). “A reference may be said to teach

away when a [POSA], upon reading the reference, would be discouraged from

following the path set out in the reference, or would be led in a direction divergent

from the path that was taken by the applicant.” Id. That is exactly what Dishman

does.

A POSA also would understand that Cunningham is directed to a completely

different endeavor than Mitchell and Dishman. Specifically, Cunningham is

directed to reducing monetary costs associated with providing free pharmaceutical

drug samples to patients for marketing purposes, and making sure pharmacies are

credited and replenished when they provide the free samples. Cunningham is not

related in any way to restricted distribution of pharmaceutical drugs, does not

address assessing risks associated with pharmaceuticals, and is not related to

preventing the occurrence of adverse events, let alone teratogenic side effects. A

POSA would have no reason to look to Cunningham.



04841.00006/7042666.1 - 31 -

Further, patient information is not collected for the system disclosed in

Cunningham, nor is the patient registered in that system. The only reason that

CFAD cites Cunningham is because it uses the same words—“approval code”—as

the claimed inventions. In fact, CFAD concedes that a POSA would only look for

the approval code required by the challenged claims “once armed” with the

disclosures of the ’720 patent. Pet. at 22-23. This is hindsight, and demonstrates

that there would have been absolutely no motivation for a POSA to combine the

cited references. See Texas Instruments, IPR2014-01105, Paper 8 at 17.

Finally, a POSA would have been taught away from combining the

Accutane system described in Mitchell with any other prior art because it too was a

failure. Specifically, Mitchell discloses that compliance with the program was not

complete and, as a result, there were hundreds of pregnancies during isotretinoin

treatment, at least six of which resulted in liveborn infants with fetal birth defects.

See supra at II.B. CFAD ignores these additional reasons that a POSA would not

have been motivated to combine Mitchell, Dishman, and Cunningham.

Third, CFAD ignores that there can be no motivation to combine the

references because the prior art does not disclose each and every element of the

claimed inventions. Specifically, the Federal Circuit has instructed that motivation

to combine is only a consideration “if all the elements of an invention are found in

a combination of prior art references.” Medichem, S.A. v. Rolabo, S.L., 437 F.3d



04841.00006/7042666.1 - 32 -

1157, 1164 (Fed. Cir. 2006). As discussed below, the prior art does not disclose all

of the elements of the challenged claims. CFAD has failed to meet its burden of

showing a motivation to combine for this additional reason.

For the foregoing reasons, CFAD has failed to set forth a motivation to

combine the prior art as set forth in Ground 1. Its Petition should be rejected.

(c) The Ground 1 References Fail To Disclose, Teach, or

Suggest Key Elements of the Claimed Inventions

CFAD relies on Mitchell, Dishman, and Cunningham as teaching the

elements of the ’720 patent. None of these references, however, disclose, teach, or

suggest all of the elements of the claimed subject matter, and certain elements are

missing from every reference. CFAD relies only on attorney argument, its expert’s

unsupported and conclusory statements, and impermissible hindsight to attempt to

correct the deficiencies in the prior art. CFAD falls far short of showing a

reasonable likelihood that the challenged claims are obvious.

i. Claim 1 Would Not Have Been Obvious

Over Mitchell, Dishman, And Cunningham

1) Element 1(c)

CFAD incorrectly argues that Mitchell and Dishman disclose “in response to

said information set, assigning said patient to at least one of said risk groups and

entering said risk group assignment in said medium.” See Pet. at 19-22.

First, CFAD argues that Mitchell “teaches a ‘patient-qualification checklist’

for assigning patients to [] risk groups . . . includ[ing], for example, ‘women of



04841.00006/7042666.1 - 33 -

childbearing age’ and ‘women who were at high risk of becoming pregnant.’” Pet.

at 19-20. Not so. Mitchell merely references the PPP’s “patient-qualification

checklist,” without elaborating on what it entailed. See Ex. 1010 at 101. Further,

while Mitchell teaches that the PPP included women of child bearing age and who

were at a high risk of becoming pregnant, it does not teach that these women were

assigned to “risk groups” in order to conduct a prospective risk analysis as required

by claim 1. In fact, Mitchell admits that this prospective risk analysis did not occur

and that there were high levels of noncompliance and many pregnancies in the

voluntary PPP. See, e.g. Ex. 1010 at 104. Thus, CFAD’s and Dr. Fudin’s

conclusory argument that a POSA “would understand that a prescriber would

assign a patient to [Mitchell’s] various risk groups, as in the first portion of Claim

1(c)” (Pet. at 20; Ex. 1027 at ¶¶ 82-83), is unsupported, and entitled to little or no

weight. See TRW Auto., IPR2014-00258, Paper 18 at 10.; 37 C.F.R. §42.65(a).

Second, CFAD admits that Mitchell does not disclose “entering said risk

group assignment in said [computer readable storage] medium.” See Pet. at 20.

Yet it argues that “because Mitchell published the aggregated patient risk group

information collected, it would have been obvious to [a POSA] that Mitchell

inherently ‘entered the risk group information in a computer readable storage

medium.’” Id. (quoting Ex. 1027 at ¶ 86). As an initial matter, the only alleged

support for CFAD’s conclusory “inheren[cy]” argument is its expert’s



04841.00006/7042666.1 - 34 -

unsupported, nearly verbatim opinion. CFAD’s argument is entitled to little to no

weight for this reason alone. See TRW Auto., IPR2014-00258, Paper 18 at 10; 37

C.F.R. § 42.65(a). Further, as explained above, Mitchell does not disclose any

“risk group assignment[s].” And, as CFAD correctly notes, Mitchell’s records

were kept for a “survey” to determine the effectiveness of the PPP. These records

had nothing to do with assigning risk groups to conduct a prospective risk analysis

with the computer readable storage medium as set forth in claim 1. CFAD’s

argument is based on impermissible hindsight.

Third, CFAD argues that Dishman discloses “storage of the patient’s

‘clinical and demographic information’ on a computer readable storage medium”

and that a POSA “would have understood . . . that this computerized system must

include a patient’s risk group assignment data in order to determine which

prescriptions should be ‘locked out.’” Pet. at 22 (quoting Ex. 1027 at ¶ 91).

Again, the only support asserted for CFAD’s conclusory argument is its expert’s

verbatim, unsupported opinion. CFAD’s argument is entitled to little to no weight

for this reason alone. See TRW Auto., IPR2014-00258, Paper 18 at 10; 37 C.F.R. §

42.65(a). Moreover, Dishman does not disclose, teach, or suggest any “risk

groups,” let alone “lock[ing] out” prescriptions based on “risk groups.” Instead,

the “lock out system” in Dishman is designed to monitor if there are “three



04841.00006/7042666.1 - 35 -

consecutive drops in the WBC count.” Ex. 1007 at 900. This has nothing to do

with risk group assignments. CFAD’s argument is entirely hindsight driven.

Accordingly, for the foregoing reasons, CFAD has failed to show that its

cited references disclose, teach, or suggest Element 1(c).

2) Element 1(d)

CFAD argues that Mitchell discloses Element 1(d): “based upon said

information and said risk group assignment, determining whether the risk that said

adverse side effect is likely to occur is acceptable.” Pet. at 19. CFAD’s argument,

however, is based on the assertion that “Mitchell teaches a ‘patient-qualification

checklist’ for assigning patients to [] risk groups” (see Pet. at 19-20 (citing Ex.

1027 at ¶¶ 79,94), which as explained above, is incorrect.

Moreover, even if Mitchell did disclose risk group assignments (it does not),

CFAD’s argument does not provide any discussion or explanation as to how

Mitchell’s risk groups disclose, teach, or suggest the affirmative step of

“determining whether the risk that said adverse effect is likely to occur is

acceptable.” CFAD’s argument fails for this additional reason. CFAD has failed

to show that its cited references disclose, teach, or suggest Element 1(d).

3) Element 1(e)

CFAD makes three arguments regarding why Dishman and Cunningham

allegedly teach Element 1(e): “upon a determination that said risk is acceptable,



04841.00006/7042666.1 - 36 -

generating a prescription approval code to be retrieved by said pharmacy before

said prescription is filled.” See Pet. at 22-25. Each argument fails.

First, CFAD argues that Dishman’s disclosure of clozapine dispensing only

upon the pharmacist’s verification that a patient’s white blood cell count is within

an acceptable limit would have rendered the concept of an “approval code”

obvious. Id. at 22-23. Not so. Dishman does not disclose, teach, or suggest a

pharmacy retrieving an approval code to prospectively control the risk of an

unacceptable side effect. CFAD does not provide any support for its conclusory

assertion, and its reliance on Dr. Fudin’s declaration (id. (citing Ex. 1027 at ¶¶ 95-

96)) is entitled to little or no weight because Dr. Fudin simply repeats CFAD’s

conclusory argument. See TRW Auto., IPR2014-00258, Paper 18 at 10.

Second, CFAD argues that the ’720 patent’s disclosure that “[s]uitable

computer readable storage media . . . will be apparent to one of ordinary skill in the

art, once armed with the teachings of the present application,” would have made

it obvious to a POSA to use an approval code to implement Dishman’s lockout

system. Pet. at 23 (citing Ex. 1027 at ¶ 97). CFAD’s argument here is backwards;

it is using the ’720 patent’s disclosure to argue that Dishman’s lockout system is

obvious. Whether or not a POSA would be motivated to use the ’720 patent’s

approval code “to implement” Dishman is irrelevant to whether or not the ’720

patent’s approval code to prospectively control the risk of an adverse side effect is



04841.00006/7042666.1 - 37 -

obvious in light of Dishman’s lockout system. The ’720 patent’s approval code

was not known before the ’720 patent’s priority date, and CFAD’s reliance on the

’720 patent’s disclosure is impermissible hindsight. See InTouch Techs., Inc. v.

VGo Commc’ns, Inc., 751 F.3d 1327, 1351-52 (Fed. Cir. 2014). Moreover, CFAD

has not identified any “design need or market pressure,” or “a finite number of

identified, predictable solutions,” to support its argument that a POSA would have

been motivated to use the ’720 patent’s approval code to implement Dishman’s

lockout system. See Pet. at 23 (citing Ex. 1027 at ¶ 92 and Bayer Schering Pharma

AG v. Barr Labs., Inc., 575 F.3d 1341, 1347 (Fed. Cir. 2009)).

Third, recognizing that Dishman does not disclose the claimed approval

code, CFAD turns to Cunningham. Specifically, CFAD argues that it would have

been obvious to “utilize the claim 1(e) approval code to implement Dishman’s

lockout system in light of Cunningham’s disclosure.” Pet. 23-24. Again, CFAD

makes the irrelevant argument that the ’720 patent’s approval code renders

Dishman obvious. This does not render Element 1(e) obvious.

Moreover, as discussed above, Cunningham is directed to reducing costs

associated with providing free pharmaceutical samples, and Cunningham’s

“approval code” is in no way tied to a prospective determination that the risk of

adverse events is acceptable, as required by claim 1. Specifically, Cunningham

does not consider any patient information, nor has CFAD even attempted to



04841.00006/7042666.1 - 38 -

explain how patient information could be incorporated into Cunningham’s system.

Cunningham’s process is unrelated to the claimed method.

Further, CFAD ignores that the Examiner of the ’720 patent considered a

divisional application with the same specification as Cunningham, and similar

“approval code” arguments based on Boyer (Ex. 1005), and found they did not

disclose, teach, or suggest Element 1(e). See generally Ex. 1002; see supra at II.C.

CFAD presents no basis to make a contrary finding here.

For the foregoing reasons, CFAD has failed to show that there is a

reasonable likelihood that claim 1 is obvious.

ii. Dependent Claims 2–27 Would Not Have Been

Obvious Over Mitchell In View Of Dishman

And The Knowledge Of a POSA

Because CFAD has failed to show that there is a reasonable likelihood that

claim 1 is obvious, it also has failed to show that there is a reasonable likelihood

that any of claims 2-27, which depend from claim 1, are obvious. See In re Fine,

837 F.2d 1071, 1076 (Fed. Cir. 1988) (holding that a dependent claim that depends

from a non-obvious independent claim is also nonobvious). CFAD has also failed

to show that there is a reasonable likelihood that claims 5, 6, 9, 10, 13, 14, 17-19,

and 23-25 are obvious for further reasons.



04841.00006/7042666.1 - 39 -

1) Claim 5

CFAD argues that Mitchell and Dishman teach claim 5’s additional

limitation that “said risk group assignment and informed consent is verified by said

prescriber at the time that said patient is registered in said computer readable

storage medium.” Its arguments lack merit.

First, CFAD argues that because Mitchell discloses the screening of

“enrollment forms,” it would have been obvious that a doctor would have verified

a patient’s informed consent and risk group assignment when registering the

patient in the database. Pet. at 27. This argument makes no sense. The

“enrollment forms” disclosed in Mitchell related to patient enrollment into the

voluntary survey designed “to assess compliance with [Accutane’s distribution]

program,” which is the subject of Mitchell, not the enrollment forms for the

distribution program itself. See Ex. 1010 at 101. Mitchell does not disclose, teach,

or suggest risk group assignments, registering a patient in a computer readable

storage medium, or a prescriber verifying the risk group assignment and informed

consent at the time of registration. CFAD’s bare assertion that a doctor should

verify the requirements for treatment when registering the patient because “the

enrollment forms were screened on receipt to exclude enrollments that were

apparently fraudulent, men, and previously enrolled women” (see Pet. at 27, citing



04841.00006/7042666.1 - 40 -

nearly verbatim the Fudin Decl., Ex. 1027 at ¶¶ 107-08), is conclusory and

incorrect.

Second, CFAD argues that Dishman’s disclosure of patient screening

teaches claim 5’s additional limitation. As an initial matter, CFAD provides no

explanation for why patient “screening” equates to verifying a patient’s informed

consent and risk group assignment. Regardless, Dishman does not disclose, teach,

or suggest risk group assignments or a prescriber verifying the risk group

assignment at the time of registration. And even if “screening” patients did equate

to “verifying the risk group assignment and informed consent,” Dishman teaches

that the pharmacist, not the prescriber, screens the patient to specifically “ensure[]

that the physician does not waste time evaluating patients.” See Pet. at 27 (quoting

Ex. 1007 at 900). As such, it would teach away from the claimed invention.

CFAD cannot meet its burden on claim 5 for these additional reasons.

2) Claim 6

CFAD argues that claim 6’s additional limitation that the “said risk group

assignment and said informed consent is transmitted to said computer readable

storage medium by facsimile and interpreted by optical character recognition

software” is obvious based on: (1) Dishman’s teaching of faxing patient

evaluations to the NCCC; and (2) a POSA’s knowledge to “transfer paper data into



04841.00006/7042666.1 - 41 -

a computer database by fax, which is then ‘interpreted by optical character

recognition [OCR] software.” Pet. at 28 (quoting Ex. 1027 at ¶ 114).

The only support for CFAD’s conclusory argument is its expert’s

unsupported, nearly verbatim opinion. See id. CFAD’s argument is entitled to

little or no weight (see 37 C.F.R. § 42.65(a)), and cannot establish obviousness,

especially here, where the claimed elements are clearly not found in the prior art.

3) Claim 9

CFAD argues that claim 9’s additional limitation that “diagnostic testing is

probative of the concentration of said drug in a tissue of said patient” is obvious

based on “the extensive diagnostic testing taught by Mitchell and Dishman.” Pet.

at 29. Mitchell and Dishman, however, do not disclose, teach, or suggest such

testing. See generally Ex. 1007; Ex. 1010; see also Pet. at 29 (describing testing).

Recognizing this deficiency, CFAD next argues, relying on only the Fudin

Declaration, that: (1) it is obvious to conduct testing of a patient’s tissue to

determine the concentration of a drug in any patient who received that drug; and

(2) because many drugs are preferably absorbed in tissue, as opposed to uniformly

in blood or serum, a POSA would understand the importance of performing testing

“of the concentration of such a non-uniformly distributed drug in the drug’s target

tissues.” Pet. at 30 (citing Ex. 1027 at ¶¶ 122-123). Dr. Fudin’s opinions are

entirely unsupported and entitled to little or no weight. See 37 C.F.R. §42.65.



04841.00006/7042666.1 - 42 -

CFAD’s argument is based on nothing more than its attorneys’ and expert’s

say-so, and should be rejected. The additional limitation of claim 9 is not

disclosed in the prior art, and CFAD therefore cannot prove obviousness. See 37

C.F.R. § 42.104(b)(4); see also Toshiba Corp. v. Optical Devices, LLC, IPR2014-

01446, Paper 7 at 16 (Mar. 10, 2015) (finding that Petitioner failed to show a

reasonable likelihood of proving a challenged claim obvious when it failed to

supply a prior art reference demonstrating a claimed feature).

4) Claim 10

Similar to claim 9, CFAD fails to demonstrate where Dishman or Mitchell

disclose, teach, or suggest the additional limitation of claim 10 that “said

diagnostic testing comprises genetic testing.” Instead, CFAD once again merely

cites to Dr. Fudin’s conclusions that it would have been obvious to a POSA to

include genetic testing. See Pet. at 30. Mitchell and Dishman do not disclose,

teach, or suggest genetic testing, and Dr. Fudin’s unsupported opinions are entitled

to little or no weight. See 37 C.F.R. § 42.65(a). The additional limitation of claim

10 is not disclosed in the prior art, and CFAD cannot prove obviousness. See 37

C.F.R. § 42.104(b)(4); Toshiba , IPR2014-01446, Paper 7 at 16.

5) Claims 13, 14, 23-25

Each of claim 13, 14, and 23-25 claims avoiding a side effect arising in a

recipient, or a fetus carried by a recipient, of the bodily fluid of the drug-receiving



04841.00006/7042666.1 - 43 -

patient. CFAD does not argue that any of its cited references disclose, teach, or

suggest such a situation. Instead, CFAD improperly relies on two external

references (Exs. 1018-1019) to argue that it was obvious that thalidomide “was

likely to cause side effects either in a sexual partner of a male being treated with

the drug as in Claims 14, 24, and 25, or in the fetus of this sexual partner as in

Claims 13 and 23.” Pet. at 33.

As an initial matter, Exs. 1018-1019 are not asserted in CFAD’s ground of

unpatentability and its argument based on these references should be rejected for

this reason alone. See Boehringer Ingelheim Int’l v. Biogen, Inc., IPR2015-00418,

Paper 14 at 17 (July 13, 2015) (citing 37 C.F.R. § 42.104(b)(2)). In addition,

CFAD has failed to articulate any reason to look to these two additional references

or any motivation to combine these references with the references asserted in

Ground 1. CFAD’s arguments fail for this additional reason. See Norman Int’l,

Inc. v. Hunter Douglas Inc., IPR2014-00276, Paper 11 at 20 (June 20, 2014)

(denying institution and stating that the motivation to combine analysis “should be

made explicit”).

Moreover, CFAD ignores that Powell (which CFAD relies on in IPR2015-

01102) teaches away from controlling side effects of a drug that arise in a

recipient, or a fetus carried by a recipient, of the bodily fluid of the drug-receiving

patient. Specifically, contrary to Exs. 1018 and 1019, Powell teaches that, with



04841.00006/7042666.1 - 44 -

respect to thalidomide, “[n]o effects on male sperm are recognized.” Ex. 1006 at

903. This disclosure teaches away from claims 13, 14, and 23-25.

6) Claim 17

Claim 17 requires that the survey regarding a patient’s behavior and

compliance with the risk avoidance measures “is conducted telephonically using an

integrated voice response system.” CFAD does not argue that any of its Ground 1

references disclose, teach, or suggest claim 17’s additional limitation. Instead,

CFAD cites, but does not discuss, the Mundt reference (see Pet. at 38 (citing Ex.

1017)) to argue that conducting surveys by telephone was well known to a POSA.

This is improper (see Boehringer, IPR2015-00418, Paper 14 at 17; 37 C.F.R.

§ 42.104(b)(2); 37 C.F.R. § 42.6(a)(3)), and does not correct the deficiencies in the

prior art. CFAD does not provide any explanation for why a POSA would have

looked to Mundt or combined Mundt with Mitchell and Dishman, or why using an

integrated voice response system would have otherwise been obvious over Mundt

or any other prior art reference. CFAD cannot meet its burden on claim 17 for

these additional reasons.

7) Claims 18 and 19

Claims 18 depends from claim 15 and requires that the second set of

information, to be collected and entered on a periodic basis before the patient is

approved to receive the drug, “comprises the results of a pregnancy test.” Claim



04841.00006/7042666.1 - 45 -

19 requires that the periodic basis is about 28 days. CFAD asserts Mitchell and

Dishman against these claims, but admits that “neither reference teaches requiring

periodic pregnancy tests before prescribing the next cycle of drugs.” Pet. at 40.

Notwithstanding this deficiency in the prior art, CFAD argues that it would have

been obvious to a POSA to apply Mitchell’s requirement for excluding pregnancy

to Dishman’s weekly WBC tests to arrive at the claimed invention. Id. Each of

CFAD’s specific arguments is flawed.

CFAD first contends that Mitchell “teaches obtaining periodic pregnancy

test results” because the “the survey covered the treatment period and the

subsequent six months,” and “the authors asked about the occurrence of pregnancy

during or after treatment.” Pet. at 39. CFAD is incorrect for several reasons.

First, collecting data on the occurrence of pregnancy during or after treatment is

not the same as obtaining periodic pregnancy test results while a patient is in

treatment. Second, the survey in Mitchell did not require periodic pregnancy test

results to be obtained from patients. Third, the PPP described in Mitchell did not

require periodic pregnancy testing. Mitchell does not support CFAD’s argument.

CFAD next cites Dishman’s disclosure that “contraindications to clozapine

therapy include … pregnancy” sin support of its allegation that it would have been

obvious to a POSA “to also require a mid-treatment pregnancy test for Dishman’s

clozapine patients upon a missed period.” Pet. at 39-40 (citing the nearly verbatim



04841.00006/7042666.1 - 46 -

Fudin Decl., Ex. 1027 at ¶ 167). However, CFAD and its expert do not provide

any reason or support for this argument. Thus, CFAD’s expert’s statement is

entitled to little or no weight (see 37 C.F.R. § 42.65(a); Ashland Oil, Inc., 776 F.2d

at 294), and CFAD’s related argument should be rejected.

Further, CFAD ignores that a POSA would not be motivated to combine

Dishman and Mitchell because a pregnancy test and WBC count are completely

different tests given with different frequency. Likewise, measuring WBC counts is

completely unrelated to attempting to prevent teratogenic side effects or birth

defects. Further, the prescription in Dishman can still be filled unless the computer

notices three consecutive drops in WBC count, and any subsequent lockout can be

overridden. Ex. 1007 at 900. This teaches away from the invention claimed in the

’720 patent, which seeks to prevent even one exposure that could lead to life-

threatening adverse effects and birth defects. CFAD cannot meet its burden on

claims 18 and 19 for these additional reasons.

iii. Independent Claim 28 And Dependent Claims

29–32 Would Not Have Been Obvious Over

Mitchell In View Of Dishman In Further View

Of Cunningham And Knowledge Of a POSA

Claim 28 is identical to claim 1 but for the additional limitation: “wherein

said adverse side effect is likely to arise in patients who take said drug in

combination with at least one other drug.” Accordingly, for the reasons set forth

above with regard to claim 1, CFAD has failed to show there is a reasonable



04841.00006/7042666.1 - 47 -

likelihood that claims 28 and 29-32 are obvious. CFAD has further failed to meet

its burden on these claims for the reasons discussed below.

1) Claim 28

CFAD’s sole argument with regard to claim 28 is the conclusory assertion

that it would have been obvious to a POSA that the adverse effects described in

Mitchell and Dishman would also arise in patients taking medications in

combination with isotretinoin or clozapine. Pet. at 42 (citing Ex. 1027 at ¶ 175).

Neither CFAD nor Dr. Fudin provides any explanation or objective support for this

statement, and neither Mitchell nor Dishman mention the use of other drugs.

CFAD’s argument is entirely baseless, rooted in hindsight, and should be rejected

for this additional reason.

2) Claims 29-32

Claims 29-32 depend from claim 28 and are thus not obvious for the reasons

set forth above regarding claims 1 and 28. Moreover, neither Mitchell nor

Dishman mention the use of other drugs in combination with isotretinoin or

clozapine, or the use of diagnostic testing for evidence of drug use. Thus, claim

29, which requires that the set of information is probative of the likelihood that a

patient may take the drug and another drug in combination, and claim 31, which

requires diagnostic testing for the use of other drugs, are not obvious over Mitchell

and Dishman for this additional reason.



04841.00006/7042666.1 - 48 -

Attempting to fill in the holes, CFAD improperly relies on seven external

references that discuss drug-drug interactions (Exs. 1020-1022), or that relate to

substance abuse in general or among schizophrenics (Exs. 1023-1026). Pet. at 42-

44. These references are not asserted in CFAD’s ground of unpatentability and

they should not be considered. See Boehringer, IPR2015-00418, Paper 14 at 17;

37 C.F.R. § 42.104(b)(2). In addition, CFAD has failed to articulate any reason to

look to these references or any alleged motivation to combine these references with

the references asserted in Ground 1. CFAD’s obviousness arguments regarding

these dependent claims fail for this additional reason. See Norman, IPR2014-

00276, Paper 11 at 20.

Moreover, CFAD’s attempt to use substance abuse in the schizophrenic

population as a reason why it would have been obvious to “test for the regimen’s

contraindicated substance abuse by screening for the presence of alcohol and other

drugs” (Pet. at 44) fails for at least two reasons.

First, the contraindications disclosed in Dishman do not include the use of

other drugs; the NCCC only “recommends” that clozapine not be used in patients

who, because of substance abuse, cannot be relied upon to keep follow-up

appointments. Ex. 1007 at 900. Second, despite the fact that Dishman recognized

the potential for substance abuse in the schizophrenic population, Dishman did not

disclose, teach, or suggest the use of diagnostic testing for evidence of other drugs.



04841.00006/7042666.1 - 49 -

The additional limitations of claims 29-32 are not disclosed in the prior art, and

CFAD cannot prove obviousness. See 37 C.F.R. § 42.104(b)(4); Toshiba ,

IPR2014-01446, Paper 7 at 16.

For the foregoing reasons, CFAD has failed to show that there is a

reasonable likelihood that claims 28-32 are obvious.

4. CFAD Fails To Address The Objective Evidence of

Nonobviousness Regarding the ’720 Patent

Objective indicia of nonobvious are “not just a cumulative or confirmatory

part of the obviousness calculus but constitute[] independent evidence of

nonobviousness.” Ortho-McNeil Pharm., Inc. v. Mylan Labs., Inc., 520 F.3d 1358,

1365 (Fed. Cir. 2008). In fact, the Federal Circuit has observed that objective

indicia “can be the most probative evidence of nonobviousness in the record, and

enable[] the . . . court to avert the trap of hindsight.” Crocs, Inc. v. Int’l Trade

Comm’n, 598 F.3d 1294, 1310 (Fed. Cir. 2010). Here, CFAD does not address or

“challenge the merits of Patent Owner’s secondary consideration evidence,” and

thus its petition fails for this additional reason. Omron Oilfield & Marine, Inc. v.

MD/Totco, IPR2013-00265, Paper 11 at 16 (Oct. 31, 2013).

(a) Long-felt Need Further Supports The

Nonobviousness Of The Claimed Inventions

The existence of a long-felt and unsolved need for the patented inventions

supports a finding of nonobviousness. See Georgia-Pacific Corp. v. U.S. Gypsum,



04841.00006/7042666.1 - 50 -

195 F.3d 1322, 1330 (Fed. Cir. 1999). Prior to the development of the patented

inventions, there was a need for a prospective system that controlled the risk of

adverse side effects (such as birth defects of the type associated with thalidomide)

before the patient received the drug. Celgene developed Enhanced S.T.E.P.S.®,

which is covered by the methods claimed in the ’720 patent, to address that need.

Without the inventions of the ’720 patent, Thalomid®, Revlimid®, and Pomalyst®

would not be subject to strict patient access controls before the patient receives the

drug. This prospective risk analysis improves upon Celgene’s earlier inventions

(described in the ’501 patent), which only retrospectively checked for risks

associated with harmful drugs like thalidomide. Accordingly, the inventions

claimed in the ’720 patent provided a way for patients in need to receive valuable,

life-altering treatments, including cancer treatments.

The need for the therapies provided by Thalomid®, Revlimid®, and

Pomalyst® is evidenced by the fact that the FDA granted Orphan Drug Exclusivity

to each product for meeting an unmet need felt by relatively small, untreated

patient populations. Ex. 2018; Ex. 2019; Ex. 2020. Moreover, all three drugs were

approved under accelerated approval regulations during the FDA approval process

so that they could reach patients in need faster. See, e.g., Ex. 2010 at 1-2; Ex. 2011

at 1; Ex. 2012 at 1.



04841.00006/7042666.1 - 51 -

Of course, as noted above, none of these approvals would have been

possible without the inventions claimed in the ’720 patent, which ultimately were

required as conditions for approval by the FDA. See Ex. 1031 at 0002; Ex. 2010 at

1; Ex. 2011 at 1; Ex. 2012 at 7-10 (discussing the drugs’ approvals subject to

Subpart H). The problems associated with safe access to teratogenic drugs

addressed by the claimed inventions were not solved by the prior-art Clozaril and

Accutane systems. Instead, those problems were solved by Celgene’s invention of

S.T.E.P.S.® (described in the ’501 patent), and the improvements provided by

Enhanced S.T.E.P.S.® (described in the ’720 patent) further safeguarded the

public’s access to teratogenic drugs. The FDA’s grant of ODE and accelerated

approval for these drugs, which contain actual or potential teratogens, also

demonstrates the faith that the FDA placed in the inventions claimed in the ’720

patent, which represent the gold standards in pharmaceutical risk management.

(b) Commercial Success Further Supports The


Commercial success is “usually shown by significant sales in a relevant

market.” J.T. Eaton & Co. v. Atl. Paste & Glue Co., 106 F.3d 1563, 1571 (Fed.

Cir. 1997); see also Neupak, Inc. v. Ideal Mfg. & Sales Corp., 41 F. App’x 435,

440 (Fed. Cir. 2002) (“[S]ales figures alone can provide satisfactory evidence of

commercial success.”).



04841.00006/7042666.1 - 52 -

Here, commercial success of the patented inventions is evidenced by the

sales of Thalomid®, Revlimid®, and Pomalyst®, which together have annual U.S.

revenues of billions of dollars. See Ex. 2021 at 62. Each sale of these drugs is

made pursuant to, and is only possible as a result of, a restricted distribution or

REMS program that is covered by one or more claims of the ’720 patent. See Ex.

1031 at 0002; Ex. 2010 at 1; Ex. 2011 at 1; Ex. 2012 at 7-10 (discussing the drugs’

approvals subject to Subpart H). In other words, nexus is present because, but for

the methods claimed in the ’720 patent, no sales of Thalomid®, Revlimid®, and

Pomalyst® would have occurred.

(c) Third-Party Praise And Awards Further Supports

The Nonobviousness Of The Claimed Inventions

Praise in the industry “tends to indicate that the invention [is] not obvious.”

Kinetic Concepts, Inc. v. Smith & Nephew, Inc., 688 F.3d 1342, 1370 (Fed. Cir.

2012). Here, industry recognition for the patented inventions also supports a

finding of nonobviousness. Ex. 2006 at 329 (“The S.T.E.P.S.® programme has

been successful in preventing fetal exposure to thalidomide.”)

(d) Licensing by Others Further Supports The


Objective evidence of nonobviousness may also include “licenses showing

industry respect.” WMS Gaming, Inc. v. Int'l Game Tech., 184 F.3d 1339, 1359

(Fed. Cir. 1999). Here, licensing by others is evidenced by the fact that third



04841.00006/7042666.1 - 53 -

parties have licensed the ’720 patent. For example, at the FDA’s urging, several

manufacturers of isotretinoin (the active ingredient in Accutane®) licensed several

patents, including the ’720 patent, in connection with the restricted-distribution

program known as iPledge after it became clear that the PPP (and later attempts at

managing the distribution of isotretinoin) were ineffective. See Ex. 2015 at 4, 8-9;

Ex. 2016 at 1; and Ex. 2017 at 1.

(e) Unexpected Results Further Supports The


Unexpected results are important indicia of nonobviousness. See United

States v. Adams, 383 U.S. 39, 51-52 (1966). “[T]hat which would have been

surprising to a person of ordinary skill in a particular art would not have been

obvious.” In re Soni, 54 F.3d 746, 750 (Fed. Cir. 1995).

Here, unexpected results are demonstrated by the fact that the inventions

claimed in the ’720 patent have enabled the unexpected benefit of safely

administering teratogenic and potentially teratogenic drugs to patients who might

not otherwise have had access to them. Unexpected results are also demonstrated

by the fact that Enhanced S.T.E.P.S.®—which is covered by the claims of the ’720

patent—has been 100% successful in preventing birth defects of the type

associated with thalidomide. This stands in stark contrast to the birth defects

associated with the PPP and deaths associated with the Clozaril National Registry.



04841.00006/7042666.1 - 54 -

Celgene’s results are especially unexpected in the face of skepticism

regarding the alleged impossibility of zero risk of drug-related birth defects before

the patented inventions were marketed. See, e.g., Ex. 2022 at 235 (“We know that

when this drug is used by women of childbearing potential, the risk for causing

serious birth defects can never be lowered to zero.”); id. at 241 (“Unfortunately,

even with a stronger program for thalidomide, some affected infants will also be

born.”); Ex. 2023 at 28 (“It’s important to remember that a zero risk is not

achievable. There is no system that will prevent the single birth of a child with

phocomelia.”); Ex. 2024 at 33 (“[N]o matter how exemplary the conduct of the

manufacturers is, no matter how good a job of warning the physicians and the

pharmacists do, there will be children born with birth defects as a result of

thalidomide. It’s a fact. There will be.”); Ex. 2025 at 2 (“It is the view of the

Thalidomide Victims Association of Canada that babies will be born disabled. No

system is foolproof.”). These unexpected results further support a finding of

nonobviousness.

For the foregoing reasons, CFAD has failed to show a reasonable likelihood

that any claims of the ’720 patent would have been obvious.

B. The Petition Should Be Denied Under

35 U.S.C. §§ 314(a) and 316(b)

As explained above, CFAD’s petition should be denied because it fails on

the merits. The Petition should also be denied because it is being used for an



04841.00006/7042666.1 - 55 -

improper purpose. Specifically, the RPI are non-practicing entities who are

abusing and misusing the IPR process as part of an investment strategy aimed at

causing changes in the stock prices of targeted innovator pharmaceutical

companies. The law gives the Board discretion over which petitions to accept, and

permits the Board to reject petitions that hinder “the ability of the Office to timely

complete [IPR] proceedings.” 35 U.S.C. §§ 314(a) and 316(b). The Board should

use its discretion and deny this Petition.

The Petitioner, CFAD, is one of fifteen “Coalition For Affordable Drugs”

companies set up by one or more of the RPI. Ex. 2026; Ex. 2027 at 1-2. Each

CFAD entity is a wholly-owned subsidiary of Hayman Credes Master Fund, L.P.,

which, through a series of other “Hayman” investment firms, is controlled by RPI

J. Kyle Bass. Pet. at 1-2. RPI Erich Spangenberg and his notorious non-practicing

entity, IPNav, are paid consultants to the Hayman investment firms. Pet. at 1-2.

The RPI and CFAD are not pharmaceutical companies that seek to compete in the

space covered by the ’720 patent. In fact, neither the RPI nor CFAD have any

legitimate business interest in the targeted patent or the products it covers.4

4 Consequently, were the Board to deny CFAD’s petition, it would “not invade any

legal right conferred upon CFAD.” See Consumer Watchdog v. Wis. Alumni

Research Found., 753 F.3d 1258, 1262 (Fed. Cir. 2014) (holding that because

Consumer Watchdog did not have any connection to the challenged patent, “the



04841.00006/7042666.1 - 56 -

Instead, CFAD exists for the sole purpose of “filing and publicizing [IPR]

patent challenges against pharmaceutical companies while also betting against their

shares.” Ex. 2028 at 1. Indeed, Mr. Bass has publicly trumpeted his investment

strategy of attacking patents in the pharmaceutical industry in what he termed a

“short activist strategy.” Ex. 2029 at 1. Unsurprisingly, two of the RPI admittedly

exist for the “primary purpose of . . . generat[ing] superior risk-adjusted returns

through long or short positions with regard to selected companies, primarily in the

pharmaceuticals sector.” Ex. 2030 at 5. In fact, the RPI, through seven different

CFAD entities, have filed a combined sixteen IPRs against ten innovators,

including five against Celgene. See IPR2015-00720, -817, -988, -990, -1018, -

1076, -1086, -1092, -1093, -1096, -1102, -1103, -1136, -1169, -1241, -1344.

If the RPI are permitted to file IPRs for the sole purpose of profiting by

causing changes in public companies’ stock prices, then the Board will be over-

burdened with similar petitions. In fact, two other hedge funds have also recently

filed IPRs, following the RPI’s model. See IPR2015-00858, -1046, -1047. Such

improper petitions will divert the Board’s resources from other, properly filed

Board’s denial of Consumer Watchdog’s [inter partes reexamination] request did

not invade any legal right conferred upon Consumer Watchdog”).



04841.00006/7042666.1 - 57 -

petitions.5 Celgene respectfully submits that the Board should use its discretion

under §§ 314(a) and 316(b) to ensure that proper proceedings (those used as less

costly alternatives to litigation) are timely resolved; it should not be overburdened

by abusive petitions filed solely for the purpose of executing hedge fund

investment strategies. For this additional reason, the Petition should be denied.

C. The Petition Should Be Denied For Failing

To Name All Real Parties-In-Interest

The Petition should further be denied because it is procedurally defective.

An IPR petition “may be considered only if . . . the petition identifies all real

parties in interest.” 35 U.S.C. § 312(a)(2). Here, the Petition fails to meet that

threshold requirement. Specifically, the Petition fails to name each investor in the

Hayman funds responsible for filing the Petition. The unnamed investors are RPI

not only because they fund the Petition, but also because CFAD is merely a proxy

for the unnamed investors who stand to gain or lose financially from the Petition.

5 As Senator Kyl explained, Congress’s decision to provide the PTO with the

authority to govern the procedure before it under 35 U.S.C. § 316 “reflects a

legislative judgment that it is better that the [PTO] turn away some petitions . . .

than it is to allow the [PTO] to develop a backlog of instituted reviews that

precludes the [PTO] from timely completing all proceedings.” 157 Cong. Rec.

S1377 (daily ed. Mar. 8, 2011).



04841.00006/7042666.1 - 58 -

“[T]he ‘real party-in-interest’ is the party that desires review of the patent.

Thus, the ‘real party-in-interest’ may be the Petitioner itself, and/or it may be the

real party or parties at whose behest the petition has been filed.” Zoll Life Corp. v.

Philips Elecs. N.A., IPR2013-00606, Paper 13 at 9 (Mar. 20, 2014). A party who

funds a petition is likely to be an RPI. 77 Fed. Reg. at 48760. A petition fails to

name all RPI when it is filed by a proxy who is “at most, a ‘nominal plaintiff’ with

‘no substantial interest’ in the[] IPR challenge[] apart from those of its client[s].”

RPX Corp. v. VirnetX, Inc., IPR2014-00171, Paper 52 at 7-10 (June 23, 2014).

As explained above, Mr. Bass went on a public relations campaign, pitching

wealthy investors to contribute to his hedge fund to file the present IPR. Ex. 2028

at 4. “The fund require[d] a minimum $1 million investment.” Id. The RPI’s plan

was to file petitions on “a big-selling drug [to] rattle investors,” anticipating that

the filing of petitions would cause a change in public companies’ stock prices. Ex.

2031 at 2. Eighty percent of the profits resulting from the investments would be

returned to the unnamed investors. See Ex. 2028 at 4 (“Mr. Bass’s firm will keep

20% of all profits earned.”). The unnamed investors funded the petition.

Specifically, Hayman investment materials evidence that investors “will generally

bear costs” associated with their investments, including “outside professional fees

and expenses, including those of attorneys,” and “litigation expenses.” Ex. 2030 at

8. As such, the unnamed investors are RPI.



04841.00006/7042666.1 - 59 -

The unnamed investors are also RPI because CFAD is nothing more than a

proxy for the unnamed investors to execute their investment strategy. In fact,

CFAD has “‘no substantial interest’ in the[] IPR challenge[] apart from those of its

client[s]”—the unnamed investors. See RPX, IPR2014-00171, Paper 52 at 7-10.

As such, the Petition should be denied due to their omission.

As noted above, CFAD is a wholly owned subsidiary of Hayman Credes

Master Fund, L.P. (“Credes”). Pet. at 1-2. Credes is funded by a network of other

hedge funds, several of which are not named as RPI here, even though they invest

substantially all of their assets in, and conduct substantially all of their activities

through, one or more of the RPI. See, e.g., CFAD II LLC v. NPS Pharms., Inc.,

IPR2015-00990, Paper 19 at 30-41 (Jul. 24, 2015). The Petition also fails to name

any of the beneficial owners of any of the RPI funds. See id. at 34-35.

Additionally, general partners, managers, trustees, and directors have some control

and decision-making authority over the finances and activities of an organization,

especially hedge funds, but the Petition does not identify any partners, managers,

trustees, or directors of any of the RPI funds, other than Mr. Bass. See id. at 34-36.

The Petition should be denied due to the omission of the additional hedge funds,

beneficial owners, and partners, managers, trustees, and directors, all of whom are

RPI.



04841.00006/7042666.1 - 60 -

IV. CONCLUSION

For the foregoing reasons, the Board should deny institution of the Petition.

Date: July 28, 2015 Respectfully submitted,

By: /F. Dominic Cerrito (Reg. No. 38,100)/ F. Dominic Cerrito (Reg. No. 38,100) QUINN EMANUEL URQUHART & SULLIVAN, LLP 51 Madison Avenue, 22nd Floor New York, NY 10010 Tel: (212) 849-7000 Fax: (212) 849-7100 [email protected] Anthony M. Insogna (Reg. No. 35,203) JONES DAY 12265 El Camino Real Suite 200 San Diego, CA 92130 Tel: (858) 314-1200 Fax: (858) 314-1150 [email protected] Attorneys for Celgene Corporation


04841.00006/7042666.1

UNITED STATES PATENT AND TRADEMARK OFFICE

________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD

________________

COALITION FOR AFFORDABLE DRUGS VI LLC Petitioner,

v.

CELGENE CORPORATION Patent Owner

________________

Case IPR2015-01103 Patent 6,315,720

________________

CERTIFICATE OF SERVICE


04841.00006/7042666.1

CERTIFICATE OF SERVICE

Pursuant to 37 C.F.R. § 42.6(e), the undersigned hereby certify that

PATENT OWNER PRELIMINARY RESPONSE PURSUANT TO 35 U.S.C.

§ 313 AND 37 C.F.R. § 42.107 and accompany exhibits (Exs. 2001-2032) were

served on July 28, 2015 by filing these documents through the Patent Review

Processing System, as well as e-mailing copies to

[email protected], [email protected], and

[email protected].

Date: July 28, 2015 Respectfully submitted,

By: /F. Dominic Cerrito (Reg. No. 38,100)/ F. Dominic Cerrito (Reg. No. 38,100) QUINN EMANUEL URQUHART & SULLIVAN, LLP 51 Madison Avenue, 22nd Floor New York, NY 10010 Tel: (212) 849-7000 Fax: (212) 849-7100 [email protected] Anthony M. Insogna (Reg. No. 35,203) JONES DAY 12265 El Camino Real Suite 200 San Diego, CA 92130 Tel: (858) 314-1200 Fax: (858) 314-1150 [email protected] Attorneys for Celgene Corporation


Preliminary Response

Documents