Progress and Prospects in Vaccine Therapy for Gynecologic Cancers

Despite the development of chemotherapeutic agents, radiation techniques, and improved surgical procedures, many women with gynecologic malignancies will die from recurrent disease. In this broad review, Gurski and Steller examine potential vaccine strategies to improve disease control. The use of vaccines in both prophylactic and therapeutic settings is discussed, and a general overview of vaccines directed against both viral and nonviral tumor-associated antigens is presented.

Clinical Application of Vaccine Technology

The authors first look at the relationship of human papillomavirus (HPV) to cervical cancer and discuss the opportunity to build on prior vaccine experience in the prophylaxis of infections. The E6 and E7 proteins of papillomavirus are important in malignant transformation, and peptide sequences derived from them have been found to be immunogenic in animals. Furthermore, in vitro production of cytotoxic T-lymphocytes capable of tumor recognition and lysis of human cervical cancer cells have been demonstrated.

A phase I study using recombinant vaccinia encoding for E6/E7 genes generated antibodies to E7 in three of eight patients, while a cytotoxic T-lymphocyte response was demonstrated in only one patient.[1] In addition, the potential dangers of using full-length genes that encode for potentially oncogenic viral or cellular proteins are appropriately highlighted.

More complex and novel alternative strategies are discussed, such as using recombinant vaccinia to encode for the minimal determinant peptide of the desired tumor-specific antigen, or using lysosomal membrane-associated proteins (LAMP-1) to concentrate E7 into the major histocompatibility complex (MHC) class II processing pathway. This discussion serves to emphasize that even when much is known about the pathways of viral oncogenesis, and in vitro data suggest therapeutic targets at several levels, the clinical application of this information is less straightforward.

Upon reviewing the section that describes possible mechanisms for generating a response by cytotoxic T lymphocytes using various vaccines, it is striking to remember the rapid increase in specificity that has occurred in a short time as technology has developed. In the 1980s, initial attempts at nonspecific immunopotentiation gave way to autologous or allogeneic cancer cell preparations used largely in patients with melanoma. These original preparations simply used cultured autologous or allogeneic melanoma cells selected for high expression of an uncontrolled variety of ganglioside, protein, and glycoprotein antigens.

End Points of Treatment

Recently, the availability of methods to identify, synthesize, and chemically modify these antigens to improve antigenicity has revolutionized the ability to construct vaccines with single tumor-specific or shared antigens.[2,3] While we agree with the authors that generation of a cytotoxic T-lymphocyte response is important, it should be stated more clearly that, at present, the consistent demonstration of this response in humans remains controversial. The problem may be one of insensitive assays, ineffective vaccines, or both. Only antibody responses to target antigens and delayed-type hypersensitivity skin tests can be regularly augmented by currently available tumor vaccines, and it is this consistent immunogenicity that remains the focus of selecting vaccines for further testing. These are also the only two surrogate markers that correlate with disease-free and overall survival.[4] The clinical relevance of achieving increased cytotoxic T-lymphocyte levels remains unproven.

Timing of Vaccines

The discussion of prophylactic vs therapeutic vaccine strategies raises the important issue of when to administer a vaccine in a disease course. Patients with heavily pretreated advanced disease may not be ideally suited for determining vaccine efficacy. Ovarian cancer patients, however, are unique in this regard. Those with marker-only disease following maximal surgical cytoreduction can be categorized as having (1) minimal disease burden, (2) a nearly certain chance of relapse, and (3) a serum surrogate marker to assess response.

 The targets for ovarian cancer listed in the authors' table 1 could be expanded to include a variety of cancer cell- surface antigens consistently over-expressed, such as the blood group-related carbohydrate antigen Lewis-Y and the mucin backbone peptide antigen MUC-1. Other described cell-surface antigens on ovarian cancer include GM-2, GLOBO-H, sTn, and TF, which are all reasonable targets for investigation.[2,3] At Memorial Sloan-Kettering Cancer Center, we are currently conjugating MUC-1 and Lewis-Y with keyhole limpet hydrocyanin (KLH) in the presence of the immune adjuvant QS-21. Preliminary phase I trials with MUC-1 have shown antibody responses, and the Lewis-Y vaccine program will soon begin to accrue patients with ovarian cancer.

Summary

Gurski and Steller highlight well the advances in vaccine technology that allow the more consistent production of antibodies through the use of a variety of antigens and techniques. The phase I trials discussed illustrate the point that phase I testing in vaccine trials often seeks to find a maximum biologic effect, a less well-defined outcome than the maximum tolerated dose (MTD). Trials to date are largely phase I or II whether in the advanced or surgically adjuvant setting, and comparisons to historical controls are difficult. The stage is set to continue the systematic search for vaccines that have consistent immunogenicity with sustained antibody responses. Using the paradigm of polyvalent vaccine development for infectious diseases, those most immunogenic vaccines will be combined, and the combination vaccine with a suggestion of clinical benefit should then be rapidly tested in randomized, prospective phase III trials. The questions of timing, patient selection, and possible combination with other standard adjuvant cytotoxic therapies remain unanswered. Vaccines have the potential to be an exciting, tolerable, and powerful addition to available therapies, and prospective randomized documentation of any benefit will ultimately lead to their rational application.

References:

1. Irvine KR, McCabe BJ, Rosenberg SA, et al: Synthetic oligonucleotide expressed by a recombinant vaccinia virus elicits therapeutic CTL. J Immunol 154: 4651-4657, 1995.

2. Zhang, S, Cordon-Cardo C, Zhang H, et al: Selection of tumor antigens as targets for immune attack using immunohistochemistry. I. Focus on gangliosides. Int J Cancer, 1997 (in press).

3. Zhang S, Zhang, H, Cordon-Cardo, et al: Selection of tumor antigens as targets for immune attack using immunohistochemistry. II. Blood group related antigens. Int J Cancer, 1997 (in press).

4. Berd D, Maguire HC, Schuchter LM, et al: Autologous hapten modified melanoma vaccine as post surgical adjuvant treatment after resection of nodal metastasis. J Clin Oncol 15:2359-2370, 1997.