Gene Therapy for Ovarian Cancer

September 1, 2001

Advances in molecular virology and biotechnology have led to the engineering of vectors that can efficiently transfer genes to target cells. Gene therapy strategies were developed along two lines: Cytotoxic approaches

Ovarian carcinoma is the fourth leading cause of cancer deathin the female population and the most fatal gynecologic malignancy. Dueprimarily to the lack of effective screening strategies and the deficiency ofsigns and symptoms in patients with early-stage disease, a high fatality ratepersists. Despite advances in surgical technique and modern chemotherapy,long-term survival for most patients with advanced ovarian carcinoma hasremained at approximately 15% to 30% over the past 20 years. Clearly, moreeffective treatment strategies are needed for this disease.

Gene therapy represents a novel investigational therapeutic approach for thetreatment of ovarian cancer. Drs. Coukos and Rubin have written a comprehensivereview of the current vectors available in the treatment of ovarian cancer andthe gene therapy strategies in which these vectors are employed. It is clearthat the major obstacles to successful gene therapy are the low efficacy andspecificity of gene delivery and the potential toxicity. The authors addressthese obstacles, in one fashion or another, with each of the approaches theydescribe. We would like to discuss our approach to these issues by way ofexample, focusing on adenovirus as the vector of choice.

Enhancing Infectivity of the Virus

Results from preliminary human clinical gene therapy trials for ovariancancer have uniformly demonstrated the extremely limited efficacy ofcurrent-generation vector systems in accomplishing tumor cell modification ordeath, as summarized in the article by Drs. Coukos and Rubin. With respect toadenovirus, this limitation is primarily due to the fact that ovarian cancerprimary tumor cells are relatively resistant to infection. This resistanceoccurs mostly as a result of low levels of the coxsackie/adenovirus receptor onthe surface of ovarian cancer cells. Therefore, enhancing infectivity of thevirus is of utmost importance.

As the authors mention, modifying the viral tropism by incorporating theArg-Gly-Asp (RGD) peptide into the HI loop of the knob has consistentlydemonstrated coxsackie/adenovirus receptor-independent gene transfer toprimary ovarian cancer cells that is two to three orders of magnitude higherthan that observed with an unmodified adenovirus vector. This modificationallows the adenovirus to use cell surface integrin receptors, in addition to thecommon adenovirus entry pathway via coxsackie/adenovirus receptors. Although theRGD-modified adenovirus vector exhibits preferential gene transfer to primaryovarian cancer cells as opposed to human mesothelial tissue, integrins areubiquitously expressed (albeit to differing degrees, depending on the celltype). Therefore, while this RGD motif efficiently enhances infectivity inovarian cancer cells, it may not allow for the most stringent control ofspecificity.

Tumor-Specific Promoters

As discussed in the article, various vector strategies that exploit differentaspects of tumor biology are being used in tumor-specific gene therapy. Theseapproaches to achieving tumor-specific gene expression have met with varyingdegrees of success. Our group has been working on developing vectors that usetumor-specific promoter elements to restrict gene expression or adenovirusreplication. In the latter instance, conditionally replicative adenovirus agentshave been created by placing an essential adenovirus replication gene under thecontrol of a tumor-specific promoter.

Among those agents tested in our laboratory, the cyclooxygenase-2M (cox-2M)promoter has a high relative index and the greatest fidelity of promoteractivity in ovarian cancer.[1] Furthermore, this promoter is capable ofmitigating liver toxicity in vivo when controlling expression of the herpessimplex virus-thymidine kinase gene. Therefore, the cox-2M promoter is ahighly attractive candidate for creating a conditionally replicative adenoviruswith which to treat ovarian cancer. Clearly, efficacy and specificity can beachieved in a variety of ways; however, in order to bring this advancedtechnology to the clinic, it is imperative that toxicity remains minimal.

Adenovirus Vectors

Due to the low transduction efficiency of vectors used to date, increasingdoses of viral vector are required to obtain effective tumor transduction. Withincreased doses necessary for efficient transduction, toxicity often becomes aproblem. A simultaneous advantage and disadvantage of using adenovirus for genetherapy is the fact that administration of adenovirus is followed by an immuneresponse. This may be advantageous for the purpose of enhanced tumor recognitionby the immune system. However, as Drs. Coukos and Rubin point out, neutralizingantibodies within the peritoneal fluid decrease the efficacy of adenovirusvectors in vitro. This becomes a serious problem, especially when repeatedadministrations are needed—not only secondary to decreased vector efficacy,but also because of an overall enhanced inflammatory response.

We have found that an adenovirus vector containing the RGD peptidemodification of the fiber knob domain directs efficient gene transfer to celllines and primary ovarian cancer cells in the presence of ascites containinghigh-titer neutralizing antiadenovirus antibodies.[2] These results, therefore,suggest that such modified adenovirus vectors can not only be utilized forincreasing transduction efficiency, but may also be beneficial in mitigatingtoxic side effects by evading the immune system in the peritoneal compartment.

Conclusions

Regardless of the gene therapy vehicle of choice, the obstacles to maximizingefficacy and specificity while minimizing toxicity remain the same for anyapproach. Obviously, none of the approaches used to date are perfect on theirown. Although we are using adenovirus to address these issues, it is clear thatadvances are being made with many vectors along similar avenues. Nevertheless,there continues to be a need for a multimodal approach to ovarian cancertreatment, as Drs. Coukos and Rubin comment. In addition, we suggest that amultimodal approach be taken with respect to vector development. In this regard,multiple targeting strategies should be incorporated into one vector to allowthe circumnavigation of difficulties that prevent the successful delivery ofgene therapy and the realization of this technique’s full potential as atherapeutic modality for ovarian cancer.

References:

1. Casado E, Gomez-Novarro J, Yamamsto M, et al: Strategies to accomplishtargeted expression of transgenes in ovarian cancer for molecular therapeuticapplications. Clin Cancer Res. In press.

2. Blackwell JL, Li H, Gomez-Navarro J, et al: Using a tropism-modifiedadenoviral vector to circumvent inhibitory factors in ascites fluid. Human GeneTher 11:1657-1669, 2000.