Recurrent superficial transitional cell carcinoma of the bladder: adjuvant topical chemotherapy versus immunotherapy. A prospective randomized trial.

A multicenter, prospective, randomized controlled study was begun in 1985 on the effect of ethoglucid and keyhole-limpet hemocyanin in the prevention of recurrent superficial transitional cell carcinoma of the bladder (stages pTa to pT1, grades 1 to 3 according to the recommendation of the International Union Against Cancer and the World Health Organization). The study was performed on a selected group of patients at high risk for further recurrences. All of these patients were pre-treated with different chemotherapeutic agents (doxorubicin or mitomycin C) and still had recurrent superficial transitional cell carcinoma. All tumors were removed by transurethral resection and all patients were presumed to be free of tumor at initiation of the prophylactic instillations. Patients in the ethoglucid group received 0.565 gm. (solution of 1%) ethoglucid weekly for 6 weeks and then monthly for 1 year. Patients in the keyhole-limpet hemocyanin group were immunized with 1 mg. keyhole-limpet hemocyanin intracutaneously, and then weekly bladder instillations of 30 mg. were given for 6 weeks and then monthly for 1 year. The percentage of recurrences, recurrence rate, interval free of disease, tumor progression and effect on downstaging were evaluated for both therapeutic arms. The percentage of recurrences (60.9% in the ethoglucid group versus 55.3% in the keyhole-limpet hemocyanin group) and the comparison of interval to recurrence for all patients showed no statistical significant difference (p = 0.808, Mantel-Cox test). A comparison of the interval to recurrence in patients with recurrent tumors only showed a mean interval free of disease of 8.8 months for patients given ethoglucid versus 5.5 months for those given keyhole-limpet hemocyanin (p = 0.006, Wilcoxon test). Recurrence rate (4.8 versus 6.5, respectively) and tumor progression rate (21.7 versus 21.1%, respectively) showed no statistically significant difference (p greater than 0.1).