Personalizing gene therapy in gastric cancer

Copyright © 2006 Prous Science. CCC: 0214-0934/2006. DOI: 10.1358/dnp.2006.19.9.1050425 533

by P. Vogiatzi,M. Cassoneand P.P. Claudio

Despite a gradual decline inincidence in many Westerncountries, gastric cancer

remains a common lethal malig-nancy.1 More than 50% of gastric can-cer patients are in metastatic stage atthe time of diagnosis and, despitetherapeutic advances in the past 15years, overall prognosis remains verypoor. Conventional treatment ap-proaches such as chemotherapy,radiotherapy, hormone therapy andreductive surgery have been used tocombat the disease, but often withpoor results. In response to a trulyurgent need, researchers are trying toprovide more effective therapies tosignificantly improve survival time.Gene therapy refers to a large spec-trum of treatment options for manydisorders not limited to cancer. Asexpected, paradigms for gene therapyare mainly derived from experiencewith treatment of inherited geneticdisorders, and for diseases such ascystic fibrosis (CF), Duchenne’s mus-cular dystrophy, hemophilia, β-tha-

lassemias and severe combinedimmune deficiency (SCID),2–6 realis-tic options for treatment have beenestablished.

Gene therapy has been proposed asa “definitive solution” in cancer caredue to its potential ability to terminatetumor cells by reverting their very firstmalignant features. This research fieldis witnessing a growing number oftechnical advancements and newapplications, though results obtainedso far are prominent in preclinicalstages only, with little therapeutic ben-efits in clinical trials. Even whenapplied to the cure of a single disease

such as gastric cancer, gene therapyencompasses a broad range of poten-tial new therapeutics, and needs to beexplained and discussed with a criticaland prudent approach, presenting bothits promises and pitfalls. Specifically,in this review article, we summarizedevelopments in the overall picture ofgene therapy that have paved the wayto clinical trials in the gastric cancerfield and some of the challenges andpossible limitations in this cancer type.Using such knowledge we discussnovel engineering techniques as wellas targeting strategies and possiblecombinatory regimens towards im-provement of the therapeutic efficacy.

LOOKING AHEAD

On the road to personalizing gene therapy in gastric cancer:proposals for transforming dreams into reality

Personalizing Gene Therapyin Gastric Cancer

SummaryGene therapy was proposed many decades ago as a more straightforward and defin-itive way of curing human diseases, but only recently technical advancements andimproved knowledge have allowed its active development as a broad and promisingresearch field. After the first successes in the cure of genetic and infectious diseases,it has been actively investigated as a means to decrease the burden and sufferinggenerated by cancer. The field of gastric cancer is witnessing an impressive flourish-ing of studies testing the possibilities and actual efficacy of the many different strate-gies employed in gene therapy, and overall results seem to be two-sided: while orig-inal ideas and innovative protocols are providing extremely interesting contributionswith great potential, more advanced-phase studies concluded so far have fallen shortof expectations regarding efficacy, although invariably demonstrating little or no toxi-city. An overview of the major efforts in this field is provided here, and a critical dis-cussion is presented on the single strategies undertaken and on the overall balancebetween potentiality and pitfalls. © 2006 Prous Science. All rights reserved.

Drug News Perspect 19(9), November 2006 Available on the web at: www.prous.com/journals

Gene therapy can be classifiedaccording to its different end results.In “gene replacement,” the malignan-cy harboring a defective tumor sup-pressor gene is treated by the transferof a normal or super-active version ofthe gene. “Gene silencing” encom-passes a series of different strategiesused to impair the expression of onco-genes. Among these, “antisense thera-py” is obtained by introducing com-plementary oligonucleotides whichbind to the mRNA or DNA and pre-vent translation or transcription,respectively. “Cytotoxic gene thera-py” fosters the tumor cells to producean enzyme capable of converting anontoxic compound or “prodrug” intoa physiologically active agent orcytotoxic agent. “Immunotherapy”improves the host’s immune responseto antigens beared by a particulartumor, while in “drug resistance rever-sal” a cascade of events leading tooverexpression of resistance proteins,such as drug efflux pumps, is modu-lated and brought to normal levels.

Historical backgroundFerrari and colleagues7 used a

retroviral vector to generate peripher-al blood lymphocytes capable of trans-ducing adenosine deaminase (ADA),an important enzyme in the purinecatabolic pathway. Those lympho-cytes, obtained from patients affectedby ADA-negative SCID, were injectedinto SCID mice, restoring immunecompetence. The experiments demon-strated that gene transfer is necessaryand sufficient for development of spe-cific immune functions in vivo and hastherapeutic potential. Blaese et al.8reported results of the first gene thera-py clinical trial for ADA-deficientSCID, concluding that gene therapywas a safe and effective option for thetreatment of some patients with thissevere immunodeficiency disease. Infact, ex vivo retroviral-mediated trans-fer of the ADA gene was performed onthe T cells of two children with mildSCID: a 4-year-old and a 9-year-oldstarted gene therapy September 14,1990 and January 1991, and receiveda total of 11 and 12 infusions, respec-tively. After 2 years of gene treatment

both patients presented normalizedblood T lymphocytes counts, and inte-grated vector and ADA gene expres-sion in T cells persisted.

Only two DNA-based pharmaceu-ticals have received approval fromregulatory agencies: an antisenseoligonucleotide formulation, calledVitravene (USA, 1998) and an aden-oviral gene therapy product, calledGendicine (China, 2004). Fomivirsensodium, or Vitravene produced by IsisPharmaceuticals, an inhibitor ofimmediate early region 2 (IE2) ofhuman cytomegalovirus (CMV), isactive against both sensitive and gan-ciclovir or foscarnet-resistant mutantsof CMV,9 and is used for the localtreatment of CMV retinitis in AIDSpatients. Gendicine, manufactured bySiBiono Genetech, contains an aden-ovirus expressing p53, known to trig-ger apoptosis.10 The approval wasbased on a large, randomized clinicaltrial in patients with late-stage headand neck squamous cell carcinoma(HSNCC) and showed a three-foldincrease in complete responses whenan Ad-p53 injection was combinedwith chemo- and radiotherapy.

To date, there have been more than31,000 publications in the PubMeddatabase regarding the field of genetherapy. Overall, 65% of clinical trialswere performed in the United Statesand only 29% in Europe, of which 12%took place in the United Kingdom, and6.5% in Germany. Gene therapy hasbeen applied more frequently in can-cer diseases (67%), and less frequent-ly in monogenic diseases (8.7%), vas-cular diseases (8.7%) and infectiousdiseases (6.6%). The main virusesused as potential vectors for transduc-ing genes into cancer cells are aden-oviruses and retroviruses (25% and24% of clinical trials, respectively).The most widely studied nonviral vec-tors are liposomes (8.3% of clinicaltrials). Another approach is the use ofdirect injection of plasmid DNA (17%of clinical trials), but this techniquecan only transfect cells immediatelyadjacent to the injection site, so only asmall number of cells can be treated.

Latest data released by the GeneMedicine Journal Website in 2006(http://www.wiley.co.uk/genetherapy)show that only 1% of all clinical trialsare in phase II/III, and 2.1% are inphase III.

Gene therapy applicationsin stomach cancer

Gene replacementMany approaches other than

surgery, radiations and conventionalchemotherapy have been suggested aspotential candidates for the futuretreatment of gastric cancer. Amonggene therapies, introducing tumor sup-pressors that may be inactivated intumors would be a straightforwardoption.

TP53 or p53 protein is a notablekey molecular node, which involvesseveral pathways including cell cycle,apoptosis, DNA repair, differentiation,angiogenesis, the cytoskeleton andcell motion, the mitochondrial respira-tion and cellular senescence.11,12 Thep53 gene is critical for the suppressionof tumorigenesis, and this property hasproven useful in gene therapy applica-tions. Introduction of the p53 tumorsuppressor gene via a recombinantadenovirus inhibits the growth of gas-tric cancer cells in vitro and invivo,13,14 Very recently, Takimoto andcolleagues,15 investigated the combi-nation effect of adenoviral vector car-rying wild type p53 (Ad-p53) with his-tone deacetylase inhibitors (HDACI)and sodium butyrate, on xenograftedhuman gastric cancer cells (KATO-III)and hepatocellular carcinoma cells(HuH7) in nude mice. They confirmedan increased expression of Coxsackieadenovirus receptors with an associat-ed increment of transgene (X-gal)expression with sodium butyrate treat-ment in KATO-III cells, and highlight-ed the role of sodium butyrate as apowerful enhancer of p53 gene thera-py for cancer.

p16INK4A is a known cell cycle reg-ulator and a tumor suppressor gene,frequently mutated in various human

534 P. Vogiatzi et al. pp. 533–540

LOOKING AHEAD Drug News Perspect 19(9), November 2006

malignancies, including gastric can-cer, with a frequency exceeded only bythe p53 gene. Jeong and collaboratorsin 200316 demonstrated that thereplacement of exogenous wild-typep16INK4A delays gastric cell prolifera-tion and promotes chemosensitivity.

The tumor suppressor gene PTENis mutated in a variety of human can-cers. The growth regulatory functionsof PTEN are primarily mediated via itslipid phosphatase activity, whichspecifically reduces the cellular levelsof phosphatidylinositol 3,4,5-trisphos-phate. Genetic approaches haverevealed a surprising diversity of glob-al and cell type-specific PTEN-regu-lated functions that appear to be pri-marily controlled by modulation of asingle phosphoinositide.17 PTEN isdescribed as “mutated in multipleadvanced cancers 1” in Online Mende-lian Inheritance in Man (OMIM601728) database, and Hang et al. in200518 demonstrated for the first timeits regulatory role in human gastricgrowth. They showed that adenovirus-mediated transfer of PTEN inhibitedcell growth and induced apoptosis ingastric cell lines and in human gastrictumor xenografts.

The proapoptotic BCL2-associatedX protein (BAX) induces cell death byacting on mitochondria. It has beenshown that adenovirus mediated trans-fer of proapoptotic Bax gene was suc-cessful in vitro and in vivo, and mayprove effective in gene therapy ofstomach cancers.19

The caspases are key effector com-ponents of apoptosis. A cascade ofprotease reactions is believed to beresponsible for the apoptotic changesobserved in mammalian cells under-going programmed cell death. Thiscascade involves members of theaspartate-specific cysteine proteasesof the ICE/CED3 family, also knownas the caspase family. The adenovirusmediated gene transfer of caspase-8 ingastric cancer cell lines induces apop-tosis in detached carcinoma cells andshows potential activity against dis-semination of gastric and possibly

other carcinoma cells.20 Fu et al., in2003,21 showed that the recombinantexpression of caspase-3, which worksin a common pathway, can induceapoptosis on the SGC7901 gastric cellline.

Recently, it has been shown theintroduction of the tumor suppressorgene Fhit, which is frequently inacti-vated in gastric carcinomas, decreasesthe sensitivity to carcinogens andinduces apoptosis in stomach tumorsin vivo,22 confirming that the increas-ing number of genes and mechanismstargeted by gene replacement ap-proach is undoubtedly good news inthis young research field.

Gene silencing approachesDouble-stranded RNA (dsRNA)-

depending posttranscriptional double-stranded RNA, better known as RNAinterference (RNAi), is a very promis-ing gene-silencing technique suggestedby a physiological regulatory pheno-menon first recognized in Caenor-habditis elegans.23 RNAi technique hasbeen used to knock out genes in manyembryo and animal cell lines such asHela, HEK293 and P19.24,25 In RNAi,dsRNA degrades homologous mRNAand hence blocks the expression of thecorresponding gene. Although themechanism of RNAi has not been fullyelucidated, RNAi shows great value infunctional genomics studies and genetherapy as a simple and effective geneknockout tool.

The tumor suppressor gene E-cad-herin or CDH1 is a specific calciumion-dependent cell adhesion molecule.Germline CDH1 mutations are obser-ved in siblings with an inherited sus-ceptibility to diffuse gastric cancer(OMIM 192090). Zheng et al., 2005,26

suppressed CDH1 expression andtumor invasion in MKN45 gastric cellline by RNAi technique, demonstrat-ing the metastatic ability of CDH1 andthe potent role of this geneticapproach.

A novel approach for amelioratingchemotherapy of gastric carcinomathrough the X-linked inhibitor of

apoptosis (XIAP) was recently evalu-ated in vitro. The downregulation ofXIAP via antisense RNA led to apop-tosis of gastric cancer lines, correlat-ing with cellular p53 status and acti-vation of caspase-3.27

The RNAi technology, and smallRNAs (guide RNAs or siRNAs)design is a hot spot for research, butvariability in laboratory designschemes is a major drawback thatneeds to be addressed. Fire et al., in1998,23 showed that the dsRNA target-ing intron and promoter sequence hadno interference effect in C. elegans.Elbashir et al., 2002,28 indicated that 5′and 3′ UTRs should be avoided indesigning siRNA, taking into consid-eration the copious protein-bindingregions. These protein and translationinitiation complexes will affect thecombination between siRNA endonu-clease complex and mRNA, henceyielding no detectable interference. Inplants, dsRNA targeted promotersequence showed specific inhibitionof gene expression and induction ofmethylation of sequence of interest.29

The siRNAs, which usually rangefrom 21 to 25 nucleotides, dependingon the species of origin, have the sig-nificant disadvantage that their effectsare transient persisting approximately1 week. The long (~500 nucleotides)dsRNAs could produce stable silenc-ing in embryonic mammalian cells,but their utility is limited because ofthe restriction of cells that lackendogenous, nonspecific responses todsRNA. Paddison et al., in 2002,30

proposed short hairpin RNAs(stRNAs) as valid experimental toolsproduced exogenously or in vivo fromRNA polymerase III promoters. ThestRNAs permit the creation of contin-uous cell lines in which the target genesuppressed is stably maintained byRNAi and may be useful for the con-struction of transgenic animals. Inthis direction, Silva and colleagues, in2005,31 generated large-scale arrayed,sequence-verified libraries comprisedof more than 140,000 second-genera-tion short hairpin RNA expressionplasmids, targeting the most knownand predicted genes in human and

P. Vogiatzi et al. pp. 533–540 535

Drug News Perspect 19(9), November 2006 LOOKING AHEAD

mouse genomes available to the scien-tific community for investigation ofindividual gene functions and genom-ic approaches.

Other gene silencing approachesinclude the use of dominant negativemutant alleles, which bind a proteincomplex inhibiting protein function,the antisense oligonucleotides (ASO),and the ribozymes. Min et al., 2005showed the blockade by adenovirus-mediated expression of a truncateddominant negative insulin-like growthfactor (IGF) I receptor which sensi-tized the gastric cancer cells tochemotherapy and suppressed theirperitoneal dissemination in vivo.32

The idea of using an ASO arosethree decades ago;33 however, it maystill make “sense” in therapeuticstrategies. Kim and colleagues in2004, observed that administeringbcl-2 ASOs induced a downregulationof the antiapoptotic protein bcl-2 andthus increased significantly the sensi-tivity of stomach cancer to chemother-apeutics in vivo.34 Unfortunately, onlya small number of ASOs are currentlyused in clinical trials, including gastriccancer, because of some toxicityobserved in patients, such as impair-ment of complement and coagulationcascades, thrombocytopenia, hyper-glycemia and hypotension, attribut-able to the chemistry of the ASOs.35

Bi et al. in 2001 demonstrated thatreversion of the malignant gastric phe-notype can be achieved through inhi-bition of the oncogene c-erbB2/neu-encoded protein p185 by the specificribozyme RZerb2 very efficiently bothin vitro and in vivo.36 The finding thatribozymes have RNA catalytic activi-ty is undoubtedly another importantstep in gene therapy. In theory,ribozymes have a specific advantageover ASOs; since each ribozyme’senzymatic activity should result incleavage of multiple copies of themRNA target, while ASOs might beexpected to interact with only onemolecule of mRNA target. However,attempts to introduce modificationsthat improve the stability of ribozymes

also increased the affinity for substratemRNA (i.e., hybridization strength),thus causing a strong reduction of thecatalytic activity.37 This may explainthe few reports of ribozymes use inclinical trials; however, alternativemodifications or new RNA-basedenzymes yet to be described maychange the destiny of ribozymes intherapeutic applications.

Cytotoxic gene therapyIn cancer research, the transfection

of genes capable of sensitizing malig-nant cells to a drug, or activating pro-drugs into their effective form, isexpected to greatly increase the effica-cy and selectivity of therapy. The mostfrequently used suicide gene/prodrugsystem is the herpes simplex virus(HSV) thymidine kinase (HSV-tk)/ganciclovir (GCV) system that canconvert the prodrug GCV (nontoxic)into phosphorylated GCV (activeagent). The phosphorylated GCVinhibits the synthesis of DNA in thecells and enhances the destruction ofcancer cells via apoptotic and non-apoptotic mechanisms. Indeed, Kwonet al. in 2003 showed that the bfl-1antiapoptotic gene could be used incombination with HSV-tk/GCV toenhance host immune responseagainst colon cancer.38 Chung-Faye etal. in 2001 proposed the recombinantexpression of the bacterial enzymenitroimidazole reductase gene with theprodrug CB1954 in a phase I study asa valid treatment of gastric cancer.39

Ueda et al. in 2001 showed that dou-ble transduction by the recombinantadeno vector-expressing carcinoem-bryonic antigen and cytosine deami-nase renders gastric cancer more sen-sitive to 5-fluocytosine in vitro andin vivo than the single infection.40

Zhang et al. in 2006 demonstrated thatthe recombinant retroviral expressionof both HSVtk and TNFα genesenhanced the antitumor effect in vivo,though it did not produce a significantdifference in cell survival rate invitro.41 Although a few reports havebeen published so far, this area of genetherapy is promising and it should beexplored more since it can take advan-tage of knowledge and expertisealready developed with conventional

drugs using the broad variety of phys-iological mechanisms that can beexploited to create new active thera-peutic interactions.

Viral therapyAdenoviral vectors used in clinical

protocols mainly target cancer dis-eases, gastric cancer included, becausethey have the ability to infect alsopostmitotic tissues and are producedat titers high enough to transduceefficiently cells. New recombinantadenoviral vectors address the majordisadvantages of those of the first-generation also because of theimproved safety and capacity toaccommodate larger DNAs, and forthe reduced inflammatory response.42

The oncolytic viral strategy is exploit-ed using lytic viruses genetically engi-neered such as adenoviruses, whichkill the host cells during their lyticreplication cycle and are differentfrom the “classic” gene therapyviruses, which work as gene deliveryagents and do not replicate.43,44

Following a success in preclinicalstudies, they are now under evaluationin clinical trials.

ONYX-015 is a fusion of aden-oviruses 2 and 5 and it is to date themost frequently used in clinical trials.Two phase I/II clinical trials showedthe efficacy of ONYX-015 in metasta-tic gastrointestinal cancer.45,46

Heideman et al. in 2002 examinedretargeting an adenovector lackingnative binding activity to gastric can-cer-specific epithelial cell adhesionmolecule (EpCAM), by coupling it toa bispecific single-chain antibodydirected at EpCAM; this indirect tar-geting strategy resulted in higherselectivity for primary cancer cells,with conserved activity.47

Other experimental approaches ofviral therapy targeting cell to cellinteraction molecules in gastric cancerhave been recently reported. The genetransfer of CD80, a ligand of CD28,into gastric cancer cells using an ade-noviral vector in vitro and in vivo hasshown potential efficacy in vivo.48

536 P. Vogiatzi et al. pp. 533–540

LOOKING AHEAD Drug News Perspect 19(9), November 2006

Adenoviral-mediated gene trans-duction of NK4, an antagonist of hepa-tocyte growth factor (HGF), inhibitsboth peritoneal metastasis and intra-tumor vessels in gastric cancer in vivo,regardless of the level of cmet/HGFreceptor expression in the tumor cells,and especially in the early stages ofperitoneal metastasis.49

More recently, in vitro and in vivouse of cyclooxygenase-2 tumor-spe-cific promoter-driven conditionallyreplicating adenoviruses (COX-2CRAds) with 5/3 chimeric fiber mod-ification has been shown to be a poten-tially valuable tool for viral therapy ofgastric cancer.50

ImmunotherapyImmunomodulatory gene therapy

consists of the induction of cellularimmune responses to metastaticlesions. Immunotherapy is usuallyperformed by injecting into the skin ofthe patient a suspension of irradiatedtumor cells that have been transducedwith a cytokine to stimulate a systemicimmune response against tumor-spe-cific antigens. Unfortunately, the factthat only a few tumor-specific anti-gens act as recognition targets, theactivity limited to low tumor burden,and the high financial and labor cost,are at present important obstacles toimmunotherapeutic applications.

Tanaka et al. in 2002 evidencedthat transfection of the adhesion mol-ecule ICAM-1 gene to cancer cells invitro and in vivo can be effectiveagainst the peritoneal metastasis ofgastric carcinoma.51 In a later report,Tanaka et al. in 2004, constructedan adenoviral vector, AdICAM-2,that encodes the full-length humanICAM-2 gene under the control of thecytomegalovirus promoter, and inves-tigated its antitumor effects in vitroand in vivo.52 They concluded thattheir gene therapy approach might beadvantageous for the cure of humanscirrhous gastric carcinoma, whichdevelops peritoneal disseminationwith high frequency.

Shi et al. in 2005 constructed ananovaccine coencapsulated with thegastric cancer-specific antigen MG7mimotope peptide and adjuvant CpGoligodeoxynucleotides (CpG ODN1645) using new nanotechnology asnanoemulsion.53 They evaluated itsimmunocompetence in vivo, conclud-ing that this vaccinal approach mayhave important implications for gas-tric cancer therapy. The same scientif-ic group more recently suggested thatvaccines based on MG7-Ag mimotopeare antigenic and protective againstgastric cancer, and that the heterolo-gous primer-boost strategy (both theoral DNA vaccine and the adenovirusvaccine) increases the efficacy ofMG7-Ag mimitope DNA vaccines.54

Drug resistance transferMultidrug resistance (MDR) is

characterized by structural and func-tional resistance of the cells to unre-lated drugs and is a serious impedi-ment in chemotherapeutic manage-ment of cancer. In a recent study ongastric and pancreatic carcinoma cells,Zhou et al. (2006) demonstrated thatadenovirus-mediated enhancement ofthe c-Jun NH2-terminal kinase (JNK)reduces the level of P-glycoprotein ina dose- and time-dependent manner.55

The MDR phenotype is often causedby drug efflux pumps in the plasmaticmembrane of cancer cells. P-glycopro-tein, encoded by mdr1 gene, is thebest-characterized drug efflux pump.The authors suggest that adenoviralJNK increases the activator P-glyco-protein binding activity in the MDRcells, and that the decrease of theP-glycoprotein expression level isassociated with drug accumulation inthe cells, enhancing the sensitivity ofthe MDR cells to chemotherapeuticagents.

Antiangiogenesisgene therapy

Angiogenesis plays a pivotal rolein facilitating growth and metastasis ofsolid tumors. The vascular endothelialgrowth factor (VEGF) is a specific andcritical regulator of angiogenesis, seri-ously considered in angiogenesis-based treatments. Stoeltzing and col-

leagues in 2004 demonstrated that thedirect suppression of Hif-1α de-creased the VEGF expression inhibit-ing gastric tumor growth in vivo.56

Antiangiogenic therapy can be basedon transduction in the tissues sur-rounding the tumor to create an anti-angiogenic environment, rather thanon transduction of target genes intocancer cells. Sako et al. in 2004, foundthat the soluble VEGF receptor sFlt-1transduced by an adenovector in peri-toneal mesothelial cells is able toinhibit the peritoneal dissemination ofgastric cancer in vivo and to increasethe survival of treated animals.57

Clinical perspectivesand future challengesof gene therapy

The first two SCID patients curedby gene therapy are now leading nor-mal lives. Although gene therapy hasgreat promise, many scientific obsta-cles remain before it becomes a prac-tical form of therapy for most cancers.In the case of gastric cancer, advancesin current treatment modalities havebeen recorded, but the overall clinicaloutcome so far remains dismal. Someresearchers feel it is too early to wit-ness an invasion of the cancer marketwith products based on gene therapy.The small number of marketed genetherapy products is not necessarily anegative sign, since this is due in partto a severe licensing system. Large-scale production and administration ofthese agents will require the develop-ment of new technology and addi-tional training for physicians andparamedical staff. Coordinated, multi-disciplinary efforts engaging clini-cians in surgical, medical and radia-tion oncology, basic scientists andmedical statisticians is necessary.Clinicians will have the important taskof enrolling patients in phase I and IItrials to test the safety and potentialantitumor efficacy of new agents or,more likely, of the combination ofagents in the optimal temporal rela-tionship to the core surgical operation.Bench research on antisense therapy,monoclonal antibodies, immunothera-py, and biological agents such as theinterferons, interleukins and vaccines

P. Vogiatzi et al. pp. 533–540 537

Drug News Perspect 19(9), November 2006 LOOKING AHEAD

must go on in order to provide newpotential agents to be tested.

Regarding the state of the art ofantisense therapy as a genetic tool,synthetic siRNAs have been widelytested in vitro, but demonstrated a lotof limitations in vivo. They have lowtransduction efficiency, a short half-life, and preferentially target the liverafter systemic application; for this rea-son they are of little use for systemiccancer gene therapy.37 These obstaclesmay however be overcome by theexpression of siRNAs from targetedviral vectors.58

The minor toxicities observed inclinical trials of ASOs have been read-ily reversed. This suggests that at leastsome gene therapy medications can besafely administered to cancer patients.New generations of modified ASOscould further reduce or eliminate theundesirable effects.

Future modifications to the basicimmunotherapy may include com-bination with cytokines or other sti-mulatory molecules to increasetumor vaccine efficiency and induceantitumor activity both in situ andsystemically.

Future work must be directedtoward parallel developments in thesophistication of ribozymes structure,ASOs, and vector designs. At present,formivirsen is the first and only U.S.FDA-approved ASO drug. Phase Itrial in humans using AVI-4126, athird-generation ASO, which targetsc-myc mRNA, administered intra-venously, demonstrated no toxicityand to be a promising new and safetherapeutic strategy for prostate can-cer.59

Some physicians and scientistshave an a priori negative point of viewregarding gene therapy, but we wouldlike to suggest a more careful exami-nation of all the advantages and disad-vantages, considering the recent histo-ry of gene therapy and the fact that thenegative results of early clinical trialsreflect in large part the still embryon-

ic nature of the field. There is no short-age of ideas and applications for genetherapy, but there are important limita-tions which include low efficiency ofgene transfer, poor specificity ofresponse, lack of truly tumor-specifictargets, and of course, our incompleteunderstanding of transcription con-trol.60 The results of some early clini-cal trials should not be viewed nega-tively, but instead as reflecting ahealthy scientific process and thushelping to define the hurdles that wemust overcome in order to succeedwith gene therapy. The latest progressin developing new systems of in vivogene delivery is promising for thetreatment of a number of malignan-cies, including gastric cancer. Immu-notherapy52 or the latest adenoviralvectors47 may allow individualizedgastric therapy based on the histologyor the site of the gastric tumor.

Peer review is essential to evaluatethe scientific and ethical basis of genetherapy studies and clinical trials. It isequally important that scientists pre-sent and publish the results of suchstudies in a balanced way and temperour enthusiasm with practical reality.Despite some pitfalls, the promise ofgene therapy is intact and will likelybecome part of the overall multi-modality approach to treating cancer.The low levels of toxicity observed todate in cancer gene therapy researchindicate that combination with con-ventional treatments can be obtainedwithout increasing treatment-relatedmorbidities. This said, should weabandon the hope for a more effectivetherapy in cancer obtained using mul-timodal approaches?

AcknowledgmentsThe authors are also thankful to Caitlin

Logan for editorial assistance. We regret thatmany excellent papers relevant to this reviewcould not be cited due to space limitation.Paraskevi Vogiatzi acknowledges the Ph.D. pro-gram: “Oncological Genetics” of the Universityof Siena, Italy.

References1. Jemal, A., Siegel, R., Ward, E., Murray, T.

et al. Cancer statistics, 2006. CA Cancer JClin 2006, 56: 106–30.

2. Griesenbach, U., Geddes, D.M. and Alton,E.W. Gene therapy progress and prospects:Cystic fibrosis. Gene Ther 2006, 13:1061–7.

3. Chakkalakal, J.V., Thompson, J., Parks, R.J.and Jasmin, B.J. Molecular, cellular, andpharmacological therapies for Duchenne/Becker muscular dystrophies. FASEB J2005, 19: 880–91.

4. Mannucci, P.M. and Tuddenham, E.G. Thehemophilias—from royal genes to genetherapy. N Engl J Med 2001, 344: 1773–9.

5. Sadelain, M. Recent advances in globingene transfer for the treatment of beta-tha-lassemia and sickle cell anemia. Curr OpinHematol 2006, 13: 142–8.

6. Kohn, D.B., Sadelain, M. and Glorioso, J.C.Occurrence of leukaemia following genetherapy of X-linked SCID. Nat Rev Cancer2003, 3: 477–88.

7. Ferrari, G., Rossini, S., Giavazzi, R. et al.An in vivo model of somatic cell gene thera-py for human severe combined immunodefi-ciency. Science 1991, 251: 1363–6.

8. Blaese, R.M., Culver, K.W., Miller, A.D. etal. T lymphocyte-directed gene therapy forADA-SCID: Initial trial results after 4years. Science 1995, 270: 475–80.

9. Geary, R.S. Ueda, Henry, S.P. and Grillone,L.R. Fomivirsen: Clinical pharmacologyand potential drug interactions. ClinPharmacokinet 2002, 41: 255–60.

10. Peng, Z. Current status of gendicine inChina: Recombinant human Ad-p53 agentfor treatment of cancers. Hum Gene Ther2005, 16: 1016–27.

11. Vogelstein, B., Lane, D. and Levine, A.J.Surfing the p53 network. Nature 2000, 408:307–10.

12. Laptenko, O. and Prives, C. Transcriptionalregulation by p53: One protein, many possi-bilities. Cell Death Differ 2006, 13: 951–61.

13. Tatebe, S., Matsuura, T., Endo, K. et al.Adenovirus-mediated transfer of wild-typep53 gene results in apoptosis or growtharrest in human cultured gastric carcinomacells. Int J Oncol 1999, 15: 229–35.

14. Ohashi, M., Kanai, F., Ueno, H. et al.Adenovirus mediated p53 tumour suppres-sor gene therapy for human gastric cancercells in vitro and in vivo. Gut 1999, 44:366–71.

15. Takimoto, R., Kato, J., Terui, T. et al.Augmentation of antitumor effects of p53gene therapy by combination with HDACinhibitor. Cancer Biol Ther 2005, 4: 421–8.

16. Jeong, Y.W., Kim, K.S., Oh, J.Y. et al.Exogenous wild-type p16INK4A geneinduces delayed cell proliferation and pro-motes chemosensitivity through decreasedpRB and increased E2F-1 expressions. Int JMol Med 2003, 12: 61–5.

17. Goberdhan, D.C. and Wilson, C. PTEN:Tumour suppressor, multifunctional growthregulator and more. Hum Mol Genet 2003,12: R239–48.

538 P. Vogiatzi et al. pp. 533–540

LOOKING AHEAD Drug News Perspect 19(9), November 2006

18. Hang, Y., Zheng, Y.C., Cao, Y., Li, Q.S. andSui, Y.J. Suppression of gastric cancergrowth by adenovirus-mediated transfer ofthe PTEN gene. World J Gastroenterol2005, 11: 2224–9.

19. Tsunemitsu, Y., Kagawa, S., Tokunaga, N.et al. Molecular therapy for peritoneal dis-semination of xenotransplanted humanMKN-45 gastric cancer cells with aden-ovirus mediated Bax gene transfer. Gut2004, 53: 554–60.

20. Nishimura, S., Adachi, M., Ishida, T. et al.Adenovirus-mediated transfection of cas-pase-8 augments anoikis and inhibits peri-toneal dissemination of human gastric car-cinoma cells. Cancer Res 2001, 61:7009–14.

21. Fu, Y.G., Qu, Y.J., Wu, K.C., Zhai, H.H.,Liu, Z.G. and Fan, D.M. Apoptosis-inducingeffect of recombinant Caspase-3 expressedby constructed eukaryotic vector on gastriccancer cell line SGC7901. World J Gastro-enterol 2003, 9: 1935-9.

22. Ishii, H., Zanesi, N., Vecchione, A. et al.Regression of upper gastric cancer in miceby FHIT gene delivery. FASEB J 2003, 17:1768–70.

23. Fire, A., Xu, S., Montgomery, M.K., Kostas,S.A., Driver, S.E. and Mello, C.C. Potentand specific genetic interference by double-stranded RNA in Caenorhabditis elegans.Nature 1998, 391: 806–11.

24. Harborth, J., Elbashir, S.M., Bechert, K.,Tuschl, T. and Weber, K. Identification ofessential genes in cultured mammalian cellsusing small interfering RNAs. J Cell Sci2001, 114: 4557–65.

25. Paddison, P.J., Caudy, A.A. and Hannon,G.J. Stable suppression of gene expressionby RNAi in mammalian cells. Proc NatlAcad Sci U S A 2002, 99: 1443–8.

26. Zheng, Z.H., Sun, X.J., Zhou, H.T., Shang,C., Ji, H. and Sun, K.L. Analysis of metas-tasis suppressing function of E-cadherin ingastric cancer cells by RNAi. World JGastroenterol 2005, 11: 2000–3.

27. Tong, Q.S., Zheng, L.D., Wang, L. et al.Downregulation of XIAP expression inducesapoptosis and enhances chemotherapeuticsensitivity in human gastric cancer cells.Cancer Gene Ther 2005, 12: 509–14.

28. Elbashir, S.M., Harborth, J., Weber, K. andTuschl, T. Analysis of gene function insomatic mammalian cells using small inter-fering RNAs. Methods 2002, 26: 199–213.

29. Stokes, T. DNA-RNA-protein gang togetherin silence. Trends Plant Sci 2003, 8: 53–5.

30. Paddison, P.J., Caudy, A.A., Bernstein, E.,Hannon, G.J. and Conklin, D.S. Short hair-pin RNAs (shRNAs) induce sequence-speci-fic silencing in mammalian cells. GenesDev 2002, 16: 948–58.

31. Silva, J.M., Li, M.Z., Chang, K. et al.Second-generation shRNA libraries cover-ing the mouse and human genomes. NatGenet 2005, 37: 1281–8.

32. Min, Y., Adachi, Y., Yamamoto, H. et al.Insulin-like growth factor I receptor block-ade enhances chemotherapy and radiationresponses and inhibits tumour growth inhuman gastric cancer xenografts. Gut 2005,54: 591–600.

33. Zamecnik, P.C. and Stephenson, M.L.Inhibition of Rous sarcoma virus replicationand cell transformation by a specific oligo-deoxynucleotide. Proc Natl Acad Sci U S A1978, 75: 280–4.

34. Kim, R., Emi, M., Tanabe, K. and Toge, T.Therapeutic potential of antisense Bcl-2 asa chemosensitizer for cancer therapy.Cancer 2004, 101: 2491–502.

35. Gleave, M.E. and Monia, B.P. Antisensetherapy for cancer. Nat Rev Cancer 2005, 5:468–79.

36. Bi, F., Fan, D., Hui, H., Wang, C. andZhang, X. Reversion of the malignant phe-notype of gastric cancer cell SGC7901 by c-erbB-2-specific hammerhead ribozyme.Cancer Gene Ther 2001, 8: 835–42.

37. Jason, T.L., Koropatnick, J. and Berg, R.W.Toxicology of antisense therapeutics.Toxicol Appl Pharmacol 2004, 201: 66–83.

38. Kwon, G.Y., Jeong, J., Woo, J.K. et al. Co-expression of bfl-1 enhances host responsein the herpes simplex virus-thymidinekinase/ganciclovir gene therapy system.Biochem Biophys Res Commun 2003, 303:756–63.

39. Chung-Faye, G., Palmer, D., Anderson, D.et al. Virus-directed, enzyme prodrug thera-py with nitroimidazole reductase: A phase Iand pharmacokinetic study of its prodrug,CB1954. Clin Cancer Res 2001, 7: 2662–8.

40. Ueda, K., Iwahashi, M., Nakamori, M. et al.Carcinoembryonic antigen-specific suicidegene therapy of cytosine deaminase/5-fluo-rocytosine enhanced by the cre/loxP systemin the orthotopic gastric carcinoma model.Cancer Res 2001, 61: 6158–62.

41. Zhang, J.H., Wan, M.X., Pan, B.R. and Yu,B. Cytotoxicity of HSVtk and hrTNF-alphafusion genes with IRES in treatment of gas-tric cancer. Cancer Lett 2006, 235:191–201.

42. Kamen, A. and Henry, O. Development andoptimization of an adenovirus productionprocess. J Gene Med 2004, 6: 184–92.

43. Ko, D., Hawkins, L. and Yu, D.C.Development of transcriptionally regulatedoncolytic adenoviruses. Oncogene 2005,24: 7763–74.

44. Mathis, J.M., Stoff-Khalili, M.A. andCuriel, D.T. Oncolytic adenoviruses - selec-tive retargeting to tumor cells. Oncogene2005, 24: 7775–91.

45. Reid, T., Galanis, E., Abbruzzese, J. et al.Hepatic arterial infusion of a replication-selective oncolytic adenovirus (dl1520):Phase II viral, immunologic, and clinicalendpoints. Cancer Res 2002, 62: 6070–9.

46. Galanis, E., Okuno, S.H., Nascimento, A.G.et al. Phase I-II trial of ONYX-015 in com-bination with MAP chemotherapy inpatients with advanced sarcomas. GeneTher 2005, 12(5): 437–45.

47. Heideman, D.A., van Beusechem, V.W.,Offerhaus, G.J. et al. Selective gene transferinto primary human gastric tumors usingepithelial cell adhesion molecule-targetedadenoviral vectors with ablated native tro-pism. Hum Gene Ther 2002, 13: 1677–85.

48. Kosaka, K., Yashiro, M., Sakate, Y. et al.CD80 gene therapy for lymph node involve-ment by gastric carcinoma. Int J Oncol2004, 25: 1319–25.

49. Ueda, K., Iwahashi, M., Matsuura, I. et al.Adenoviral-mediated gene transduction ofthe hepatocyte growth factor (HGF) anta-gonist, NK4, suppresses peritoneal metas-tases of gastric cancer in nude mice. Eur JCancer 2004, 40: 2135–42.

50. Ono, H.A., Davydova, J.G., Adachi, Y. et al.Promoter-controlled infectivity enhancedconditionally replicative adenoviral vectorsfor the treatment of gastric cancer. Gastro-enterol 2005, 40: 31–42.

51. Tanaka, H., Yashiro, M., Sunami, T., Ohira,M., Hirakawa, Y.S. and Chung, K. Lipidmediated gene transfection of intercellularadhesion molecule-1 suppresses the peri-toneal metastasis of gastric carcinoma. Int JMol Med 2002, 10: 613–7.

52. Tanaka, H., Yashiro, M., Sunami, T., Sakate,Y., Kosaka, K. and Hirakawa, K. ICAM-2gene therapy for peritoneal disseminationof scirrhous gastric carcinoma. Clin CancerRes 2004, 10: 4885–92.

53. Shi, R., Hong, L., Wu, D. et al. Enhancedimmune response to gastric cancer specificantigen Peptide by coencapsulation withCpG oligodeoxynucleotides in nanoemul-sion. Cancer Biol Ther 2005, 4: 218–24.

54. Lin, T., Liang, S., Meng, F. et al. Enhancedimmunogenicity and antitumour effects withheterologous prime-boost regime using vac-cines based on MG7-Ag mimotope of gas-tric cancer. Clin Exp Immunol 2006, 144:319–25.

55. Zhou, J., Liu, M., Aneja, R., Chandra, R.,Lage, H. and Joshi, H.C. Reversal of P-gly-coprotein-mediated multidrug resistance incancer cells by the c-Jun NH2-terminalkinase. Cancer Res 2006, 66: 445–52.

56. Stoeltzing, O., McCarty, M.F., Wey, J.S. etal. Role of hypoxia-inducible factor 1alphain gastric cancer cell growth, angiogenesis,and vessel maturation. J Natl Cancer Inst2004, 96: 946–56.

P. Vogiatzi et al. pp. 533–540 539

Drug News Perspect 19(9), November 2006 LOOKING AHEAD

57. Sako, A., Kitayama, J., Koyama, H. et al.Transduction of soluble Flt-1 gene to peri-toneal mesothelial cells can effectively sup-press peritoneal metastasis of gastric can-cer. Cancer Res 2004, 64: 3624–8.

58. Howard, C.M., Fosberg, F., Minimo, C.,Liu, J.B., Merton D.A. and Claudio, P.P.Ultrasound guided site specific gene deliv-ery system using adenoviral vectors andcommercial ultrasound contrast agents. JCell Physiol 2006, 209: 413–21.

59. Iversen, P.L., Arora, V., Acker, A.J., Mason,D.H. and Devi, G.R. Efficacy of antisensemorpholino oligomer targeted to c-myc inprostate cancer xenograft murine modeland a Phase I safety study in humans. ClinCancer Res 2003, 9: 2510–9.

60. Berk, A.J. Recent lessons in gene expres-sion, cell cycle control, and cell biologyfrom adenovirus. Oncogene 2005, 24:7673–85.

Paraskevi Vogiatzi and Marco Cas-sone are Post Doctoral Fellows at theSbarro Institute for Cancer Researchand Molecular Medicine, College ofScience and Technology, TempleUniversity, Philadelphia, Pennsyl-vania, USA. Pier Paulo Claudio* iscurrently Associate Professor at theDept. of Biochemistry and MolecularBiology & Dept. of Surgery, Joan C.

Edwards School of Medicine,Marshall University, Huntington,West Virginia, USA. Dr. Vogiatzi isalso in the Department of MolecularBiology, Medical Genetics Unit,University of Siena, Siena, Italy.*Correspondence: Pier Paolo Clau-dio, M.D, Ph.D., Joan C. EdwardsSchool of Medicine, Marshall Univer-sity, Dept. of Biochemistry andMolecular Biology, & Dept. ofSurgery, 1700 3rd Ave., Huntington,WV 25704, USA. E-mail: [email protected].

540 P. Vogiatzi et al. pp. 533–540

LOOKING AHEAD Drug News Perspect 19(9), November 2006