Polyvalent Tumor Vaccines and Malignant Melanoma
By Perry Shen, MD, and Donald Morton, MD
In recent years, increased understanding of the immunobiology of tumors has renewed interest in an immunotherapeutic approach to malignancies, especially melanoma. Melanoma is considered an immunogenic tumor; the well-documented phenomenon of spontaneous regression in cutaneous malignant melanoma is thought to be due to anti-melanoma antibodies against tumor antigens. Laboratory investigators have found a significantly higher serum titer of anti-melanoma antibodies in patients with localized disease and in those who have experienced spontaneous regression than in patients with metastatic melanoma. Additionally, cytotoxic T-lymphocytes, which have demonstrated in-vitro tumor cell killing, have been isolated from melanoma patients. These findings seem to indicate that melanoma tumors express tumor-associated antigens (TAAs) that can serve as targets for immunotherapy.1 In addition, unlike other solid tumors such as those found in the breast or colon, there are no standard forms of adjuvant therapy that have proved effective against melanoma, thus providing more impetus for alternative treatments.
Theoretical Advantages of Polyvalent Vaccines
Polyvalent vaccines can be allogeneic or autologous. Some of the problems with preparations developed from the patient’s own (autologous) cells are the limited number of TAAs, tolerance of the patient to these TAAs, and the logistical difficulty and expense of harvesting tumor cells and preparing a customized vaccine for each patient. By contrast, an allogeneic vaccine is not patient-specific. Instead, it can be produced from selected cells lines that are known to express multiple TAAs and a broad range of human leukocyte antigen (HLA) expression. Because polyvalent vaccines possess foreign HLA antigens, they induce a stronger immune response to cross-reacting TAAs than do autologous cells.2-4 Moreover, the same polyvalent vaccine can be used to treat different individuals; thus, its manufacturing methods and quality control can be standardized, reducing its cost.
Some allogeneic vaccines may consist only of a purified TAA. Unfortunately, preparations with a limited profile of purified TAA also have a limited number of potential immune targets. Because of the variable expression of each tumor antigen in melanoma, each patient’s melanoma and HLA type must be matched to antigens in the vaccine. This is complicated by the fact that a cancer cell is genetically unstable and becomes heterogenous in antigen expression over time. Thus, tumor clones can escape an immune attack that targets only a few, specific TAAs. By contrast, a broad-spectrum polyvalent vaccine will generate a cross-reacting immune response with nearly all melanomas.2,5,6
Whole Cell Vaccine
Animal studies have shown that whole cells induce a stronger cytotoxic cellular and humoral immune response than do tumor cell lysates or shed antigens. Whole cells may also be able to directly present their tumor cells to host T-cells if there is a defect in the patient’s antigen processing and presentation system.1 In 1978, Morton and associates reported a randomized, clinical trial of an allogeneic melanoma vaccine in patients with metastatic melanoma.7 Their efforts eventually led to the development of CancerVax (C-Vax), a polyvalent "antigen-enriched" whole-cell melanoma vaccine. C-Vax is a live-cell preparation of three allogeneic melanoma cell lines chosen for their high expression of immunogenic TAAs and melanoma-associated antigens. Initial phase II trials of C-Vax in patients with distant metastatic melanoma (American Joint Committee on Cancer [AJCC] Stage IV) demonstrated a five-year overall survival rate of 25% for 157 patients treated with vaccine, compared with only 6% for 1,521 historic controls treated with nonvaccine therapies (P = 0.0001). The median survival of the vaccine patients and historical controls was 23 months and 7.5 months, respectively.8 A subsequent matched-pair analysis was undertaken to compare the outcome of AJCC Stage IV patients receiving C-Vax vs. non-C-Vax adjuvant therapy after complete metastasectomy. Rates of five-year overall survival were 35.6% and 17.1% for C-Vax and non-C-Vax groups, respectively, and the median overall survival was 37.2 and 14.3 months, respectively (P = 0.0005).9
Phase II studies in patients with melanoma metastatic to the regional lymph nodes (AJCC Stage III) also showed a significant difference (P < 0.0002) in 5-year overall survival rates between 283 C-Vax patients and 1474 historical controls: 53% and 39%, respectively.10 Most recently, a retrospective matched-pair analysis demonstrated a significantly (P = 0.0071) higher rate of five-year overall survival in 165 patients receiving C-Vax than in 165 patients undergoing observation after resection of intermediate/thick primary (AJCC Stage II) melanoma (80% vs 68%, respectively).11 Based on these findings, the John Wayne Cancer Institute began multicenter phase III studies of C-Vax in patients undergoing complete resection of AJCC Stage III and IV melanoma.
The immunologic basis for the improved survival of melanoma patients treated with C-Vax was analyzed in a study of 77 AJCC Stage IV patients who received C-Vax after complete resection of all disease.12 Patients (n = 29) who exhibited elevated levels of IgM antibody against a glycoprotein tumor-associated antigen (TA90) and a strong delayed-type-hypersensitivity (DTH) skin response had a five-year overall survival rate of 75%, while those (n = 35) who exhibited only one of those parameters had a five-year overall survival of 36%, and patients (n = 13) who had neither elevated antibody levels nor a strong DTH response had a five-year overall survival of 8% (P < 0.001). This study provides evidence that humoral and cellular immune responses strongly correlate with clinical outcome in melanoma.
Cell Lysate Vaccines
Lysate vaccines are based on the idea that immunogenic anti-tumor responses can be generated by attaching a foreign component, such as a bacterial or viral derivative, to membrane TAA. Mitchell and associates developed a lysate of membrane-associated tumor antigens from two melanoma cell lines. This allogeneic homogenate was combined with a nonspecific adjuvant DETOX (a combination of detoxified bacterial endotoxin and mycobacterial cell wall skeletons; Ribi ImmunoChem Research, Hamilton, MT), to form a vaccine (Melacine, Ribi ImmunoChem Research) that possessed both TAAs and melanoma-specific antigens.13
Phase II studies have been performed using Melacine in AJCC Stage IV, III, and II melanoma. One hundred six patients receiving Melacine for treatment of metastatic melanoma (AJCC Stage IV) had a 20% response rate, with 5% complete responses and 15% partial responses. The median survival time of the entire group was 12.2 months. Eight percent of patients had a long-term response which resulted in a median overall survival of 46 months.14 In another phase II study, Melacine was administered with low-dose cyclophosphamide (to reduce suppressor T-cell activity) to 23 patients with resected AJCC Stage II disease and 44 patients with resected AJCC Stage III disease. Although clinical response strongly correlated with an increase in cytotoxic T-lymphocytes, only 30% of those with increased T-cells had a remission of more than one year.13
One phase III trial with Melacine vs. standard chemotherapy in AJCC Stage IV melanoma produced no significant difference in overall survival.14 However, a significant percentage of patients who failed to respond to a least one treatment cycle of Melacine developed major clinical responses to salvage therapy with interferon-alfa-2b.15 Two other phase III trials have been initiated to compare Melacine plus interferon-alfa vs. interferon-alfa alone in AJCC Stage III and IV melanoma. Also, the Southwest Oncology Group (SWOG) has completed a prospective, randomized trial evaluating Melacine vs. observation in intermediate thickness (1.5-4.0 mm) melanomas. The final results await further observation and data analysis.13
The first multicenter phase III trial of postoperative adjuvant active specific immunotherapy for AJCC Stage III melanoma was performed by Wallack and associates in 1988. They used a polyvalent vaccine derived from four human allogeneic melanoma cell lines, which were then infected with a lytic vaccinia virus. Viral-induced oncolysis can induce strong antitumor immune responses by combining immunogenic viral antigens with weak TAAs. The combined lysates of all four melanoma cell lines formed the vaccinia melanoma oncosylate (VMO) vaccine.13
In his phase III trial, 250 patients with surgically resected AJCC Stage III melanoma were randomized to treatment with VMO or vaccinia virus alone as a control. With a median follow-up of 46.3 months, there was no statistically significant increase in disease-free or overall survival in patients treated with VMO compared with vaccinia virus alone. A retrospective subset analysis showed a statistically significant improvement in survival of 21.3% at five years for male patients aged 44-57 years (n = 20) with one to five positive nodes after treatment with VMO compared to the vaccinia virus alone group (P = 0.046). None of the melanoma cell lines used in the vaccine expressed tumor peptide antigens associated with HLA-A2. It is possible the VMO vaccine may not have been able to induce a proper cellular immune response to HLA-A2-restricted tumor antigens. Approximately 50% of individuals in the North American population are HLA-A2-positive.16
Hersey and associates have also developed a polyvalent vaccine using a vaccinia virus melanoma cell lysate (VMCL). The preparation method is similar to that used by Wallack et al, except this vaccine only uses one allogeneic melanoma cell line. In a phase II study,17 80 patients with AJCC Stage III melanoma were treated with VMCL after surgical resection of disease. The five-year overall survival of the VMCL group (60%) was superior to that of a historical control group of 151 patients (34%) and a concurrent, nonrandomized control group of 55 patients (35%). A prospective, randomized phase III trial of postoperative adjuvant therapy with VMCL vs. observation was initiated in patients with thick primary melanomas (T4 lesions) and regional nodal melanoma.18 The trial has accrued 569 patients so far, with a median survival of 84 and 65 months in the VMCL and observation groups, respectively. The results have not yet reached statistical significance, but the trial is planning to accrue patients for another three years.
Polyvalent Shed Antigen Vaccine
A polyvalent preparation of purified antigens lacks HLA antigens and thus does not induce anti-HLA antibodies, which can make it difficult to measure the cellular immune response to vaccine treatment.19 Bystryn and associates developed a polyvalent vaccine consisting of partially purified shed antigens prepared from four melanoma cell lines (3 human and 1 hamster). Ninety-four AJCC Stage III patients treated with this vaccine after surgical resection demonstrated a five-year overall survival of 50%, vs. 33% for historical controls. A strong correlation was found between immune response and prognosis. Of the vaccine-treated patients, those who developed a cellular immune response (as measured by DTH skin reaction) to the vaccine had a median disease-free survival 4.7 years longer and an overall survival 3.7 years longer than those who did not (p £ 0.02). Among vaccine-treated patients, the five-year overall survival of antibody responders and nonresponders was 71% and 44%, respectively (p £ 0.01). Based on these results, a phase III, double-blinded, randomized trial was initiated in resected AJCC Stage III melanoma. Unfortunately, the trial was interrupted early because of slow patient accrual, and only 38 patients were randomized to treatment with the shed polyvalent vaccine or a placebo. Median follow-up was 39 months, and the median disease-free survival of the vaccine (n = 24) and placebo (n = 14) groups was 18.6 and 7.1 months, respectively. Mortality at two years was 23% for the vaccine group and 40% for the placebo group. These results suggest efficacy, but the number of patients is too small for definitive conclusions.20-22
Commentary and Future Directions
This brief review of polyvalent vaccines for melanoma offers cause for both optimism and caution. Polyvalent active specific immunotherapy, one of the older vaccine technologies, finally is starting to produce objective clinical results. The Table summarizes clinical trials using the polyvalent vaccines presented in this article. Phase II trials of CancerVax (Morton’s group) and VMCL (Hersey’s group) have shown significant efficacy, and these vaccines are currently being evaluated in prospective, multicenter, randomized phase III studies. On the other hand, completed phase III trials of vaccine formulations developed by Mitchell’s, Bystryn’s, and Wallack’s groups have not demonstrated a conclusive clinical benefit in patients with regional and distant metastatic melanoma.
Polyvalent allogeneic cell-based vaccines are but one of myriad approaches to immunotherapy in melanoma. Two new advances in vaccine development are the in-situ modification of autologous tumor cells with immunostimulatory genes and the use of dendritic cells to stimulate cytotoxic T-lymphocyte responses. The concept of genetically modifying tumor cells to be more immunogenic by introducing genes that encode immunomodulatory proteins has stimulated much interest. Also, the priming of cytotoxic T-cells is a critical component of the immune response to tumors and is dependent on the presentation of the relevant antigen by professional antigen-presenting cells. Dendritic cells are considered to be the most potent of all antigen-presenting cells and have been proposed as ideal candidates for the induction of anti-tumor immunity in the vaccine treatment paradigm. Both of these immunotherapeutic modalities are still in the early phases of study, but there have been encouraging clinical trials in patients with metastatic melanoma.13
With the failure of the most recent Eastern Cooperative Oncology Group (ECOG) trial to demonstrate any significant improvement in overall survival using interferon-alfa-2b in high risk primary and regionally metastatic melanoma,23 it can be stated that there is currently no effective standard adjuvant therapy available for patients with high-risk melanoma. Any patients with advanced disease should be enrolled in a clinical trial testing investigational therapy. Active specific immunotherapy, based on our increased understanding of tumor-host immune interactions and whose clinical efficacy has been successfully tested in prospective randomized studies, eventually will significantly improve the quality and duration of life for patients with malignant melanoma. (Dr. Shen is a Senior Fellow, Training Program in Surgical Oncology; Dr. Morton is Medical Director and Surgeon-in-Chief at the John Wayne Cancer Institute, Saint John’s Health Center, Santa Monica, CA.)
| Table | |||||
Polyvalent Melanoma Vaccines |
|||||
| Vaccine (Investigator) | Preparation | Number of cell lines | Completed clinical trials | ||
| AJCC Stages tested | Eligible Patients | Results | |||
| CancerVax (Morton) | Whole irradiated | 3 human | AJCC Stage IV (phase II) | 157 | OS: P = 0.0001 |
| live cells | AJCC Stage IV (phase II) | 116 surgically resected | DFS: P = 0.005 | ||
| OS: P = 0.0005 | |||||
| AJCC Stage III (phase II) | 283 surgically resected | OS: P £ 0.0002 | |||
| AJCC Stage II (phase II) | 330 surgically resected | OS: P = 0.0071 | |||
| Melacine (Mitchell) | Mechanically disrupted lysate | 2 human | AJCC Stage IV (phase II) | 106 | Median OS 12.2 months |
| AJCC Stage III (phase II) | 44 surgically resected | 5-year OS 66% | |||
| AJCC Stage II (phase II) | 23 surgically resected | 5-year OS 78% | |||
| AJCC Stage IV (phase III) | 140 | OS: NS | |||
| Polyvalent Shed | Partially purified shed | 3 human and | AJCC Stage III (phase II) | 94 surgically resected | OS: 5-year OS 50% |
| Antigen (Bystryn) | antigens | 1 hamster | AJCC Stage III (phase III) | 38 surgically resected | 2-year OS 77% |
| Vaccinia Melanoma | Viral lysate | 4 human | AJCC Stage III (phase III) | 217 surgically resected | DFS: NS; OS: NS |
| Oncosylate (Wallack) | |||||
| Vaccinia Virus | Viral lysate | 1 human | AJCC Stage III (phase II) | 182 surgically resected | 5-year OS 50-60% |
| Melanoma Lysate | |||||
| (Hersey) | |||||
| AJCC = American Joint Committee on Cancer; DFS = disease-free survival; OS= overall survival; NS = not significant | |||||
References
1. Chan AD, Morton DL. Semin Oncol 1998;25:611-622.
2. Ollila DW, Kelley MC, Gammon G, et al. Semin Surg Oncol 1998;14:328-336.
3. Toes RE, Blom RFF, van der Voort E, et al. Cancer Res 1996;56:3782-3787.
4. Knight BC, Souberbielle BE, Rizzardi GP, et al. Melanoma Res 1996;4:299-306.
5. Morton DL, Foshag LJ, Hoon DSB, et al. Ann Surg 1992;216:463-482.
6. Albino AP, Lloyd KO, Houghton AN, et al. J Exp Med 1981;154:1764-1778.
7. Morton DL, Eilber FR, Holmes EC, et al. Aust NZ J Surg 1978;48:49-52.
8. Morton DL, Barth A. CA Cancer J Clin 1996;46:225-244.
9. Hsueh EC, Nizze A, Essner R, et al. Proc Am Soc Clin Oncol 1997;16:492.
10. Barth A, Hoon DSB, Foshag L, et al. Cancer Res 1994;54:3342-3345.
11. Shen P, Foshag LJ, Essner R, et al. Am Soc Clin Oncol Annual Mtg (1999); poster #2059.
12. Hsueh EC, Gupta RK, Qi K, et al. J Clin Oncol 1998;16:2913-2920.
13. Hemmila MR, Chang AE. J Surg Oncol 1999;70:263-274.
14. Mitchell MS. Semin Oncol 1998;25:623-635.
15. Mitchell MS, Jakowatz J, Harel W, et al. J Clin Oncol 1994;12:402-411.
16. Wallack MK, Sivanandham M, Balch CM, et al. J Am Coll Surg 1998;187:69-77.
17. Hersey P. World J Surg 1992;16:251-260.
18. Hersey P, Coates A, Tyndall L. Mel Res 1997;7(S):S22-S23.
19. Ravindranath MH, Morton DL. Cancer Medicine. Philadelphia: Lea & Febiger, 1998.
20. Bystryn JC, Oratz R, Henn M, et al. Cancer 1992;69:1157-1164.
21. Miller K, Abeles G, Oratz R, et al. Cancer 1995;75:495-502.
22. Bystryn JC, Oratz R, Shkapiro RL, et al. Proc Am Soc Clin Oncol 1998;17:434a(abstr).
23. Kirkwood JM, Ibrahim J, Sondak V, et al. Proc Am Soc Clin Oncol 1999;537a(abstr).
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content