Commentary (Sarosdy): BCG Immunotherapy for Transitional-Cell Carcinoma in Situ of the Bladder

Dr. Lamm's review of bacillus Calmette-Guérin (BCG) in the treatment of carcinoma in situ (CIS) of the bladder reflects one of the largest personal experiences of any investigator worldwide, and he should be congratulated for being one of those most responsible for the improved quality of life of many people due to the widespread use of BCG in bladder cancer. It is clear that BCG therapy allows a majority of patients with CIS to avoid cystectomy, and that it prevents progression to muscle involvement in many high-risk patients with papillary tumors.

Unanswered Questions

Despite many years of study and despite the Southwest Oncology Group (SWOG) trial of maintenance BCG, two basic questions remain unanswered: What constitutes the optimal course of therapy? How do you identify patients who must receive more than what initially appeared to be adequate therapy?

The SWOG data have not been completely analyzed or subjected to peer review, and Dr. Lamm's interpretation of these data may not represent a consensus of his co-investigators. The trial results he cites are from an "interim" data analysis of May 1991 (Lamm's reference 35). The overall survival benefit referred to as "significant" has lost its statistical significance on more recent analysis, and no disease-specific analysis has been calculated (personal communication, B. Blumenstein, PhD).

The fact that responses of CIS dramatically improved with the first maintenance course of three weekly treatments argues more for the need for more than a single 6-week course than it does for continued semiannual maintenance treatments for 2.5 additional years. While the development of late recurrences after 3 years of maintenance BCG may support the need for continued therapy, those very recurrences are likely to make the recurrence-free curves look similar to the recurrences seen with intravesical chemotherapy after maintenance chemotherapy is stopped [1].

Finally, the toxicity of the maintenance arm in the SWOG trial increased dramatically. Grade 3 toxicity occurred in only 9% of 586 patients receiving the initial six treatments but in 26% of 247 patients given maintenance therapy (P < 00001); at least 10% of the latter group failed to complete therapy due to toxicity (personal communication, B.Blumenstein, PhD).

Comparisons of Results Difficult

An additional salient point not readily apparent from Dr. Lamm's discussion is that entry criteria for the trials he describes have varied with regard to a critical factor, and that such variability makes it difficult to compare the results of these trials. No BCG trial to date has required that patients have evidence of the continued presence of CIS after resection and biopsy, such as by positive post-transurethral resection (TUR) cytology. Many patients with histologic evidence of CIS have had all disease resected; our experience indicates that at least 30% have a negative cytology after TUR alone.

In contrast, the interferon trial Dr. Lamm refers to (his reference 46) and the bropirimine studies mentioned all required a positive post-biopsy cytology [2,3]. Thus, published response rates of 45% to 50% for interferon and bropirimine may be equal to those for BCG if the latter could be corrected to assess responses only in those with positive cytology prior to treatment. Combining patients with disease and those in whom BCG is used in prophylactic fashion against recurrence falsely increases the apparent "response" rate.

Megadose Vitamin Data Unconvincing

The use of megadose vitamins as an adjunct to BCG is intriguing, but the data are unconvincing. Chemoprevention is a process generally thought to require more time than therapy, and the use of BCG or vitamins to treat CIS is therapeutic, not prophylactic. Furthermore, close analysis of the time-to-recurrence prophylaxis curves for the patients with papillary tumors described by Dr. Lamm (his reference 41) reveals that 60% of patients receiving BCG and megadose vitamins remained recurrence-free, as compared with only 20% of patients receiving BCG and RDA vitamins. Why were the results with BCG and RDA vitamins so much worse than those achieved with BCG alone in other reports summarized by Dr. Lamm in his review? Rather than ascribe a benefit to the combination of BCG and megavitamins in the prevention of tumor recurrence, an explanation is needed as to why the BCG and RDA vitamin group did so poorly compared to BCG alone in other studies.

BCG Appropriate for Many Patients

Despite these caveats, I recommend aggressive use of BCG immunotherapy in many patients [4,5]. Those with nuisance recurrences of low-risk tumors are well-suited for intravesical chemotherapy. Dr. Lamm's reference to the potential carcinogenicity of such long-used agents as thiotepa (Thiotepa) and mitomycin (Mutamycin) is completely unsupported by clinical data or experience. For patients with papillary tumors who fail to respond to such prophylaxis, or those with high-risk tumors or CIS, therapy with BCG is clearly indicated. I do not recommend megavitamins for any patient, and I recommend maintenance BCG treatments only on an individual basis.

For initial therapy, I prescribe two 6-week courses during weeks 1 to 6 and weeks 13 to 18. For patients given BCG after failure of chemoprophylaxis for low-grade nuisance tumor recurrences, I do not recommend maintenance BCG after the initial two 6-week courses. For patients who presented with high-risk disease, such as CIS alone or grade III/T-1 tumors, I use maintenance BCG only if they have tolerated the two 6-week courses well, with minimal symptoms. In patients who experience grade 3 toxicity, I suggest BCG maintenance with adjunctive isoniazid as an option, but allow patients to decide whether they want to receive this treatment. For high-risk patients who experience severe symptoms during initial therapy, I recommend continued BCG only if warranted by the clinical situation, ie, continued dysplasia or visible erythema that is inflammatory on biopsy.

Summary

Immunotherapy of superficial bladder cancer and CIS is clearly a valuable adjunct to local surgical resection, and has revolutionized the care of patients with this disease. Continued efforts are needed to produce an increased and more durable response rate with less toxicity, but patients clearly enjoy better bladder preservation and quality of life due to the novel introduction and evaluation of BCG beginning in the 1970s.

References:

1. Prout GR, Koontz WW Jr., Coombs LJ, et al for National Bladder Collaborative Group A: Long-term fate of 90 patients with superficial bladder cancer randomly assigned to receive or not to receive thiotepa. J Urol 130:678, 1983.

2. Sarosdy MF, Lamm DL, Williams RD, et al: Phase I trial of oral bropirimine in superficial bladder cancer. J Urol 147:31, 1992.

3. Sarosdy MF, Lowe BA, Schellhammer PF, et al: Bropirimine immunotherapy of bladder CIS: Positive phase II results of an oral interferon inducer. J Urol 13:304A, 1994.

4. Sarosdy MF: Principles of intravesical chemotherapy and immunotherapy. Urol Clin North Am 19:509, 1992.

5. Sarosdy MF: Immunotherapy of superficial bladder carcinoma. AUA Update Series 14:233-240, 1995.