To evaluate the response rate for intravesical gemcitabine given in three different schedules to patients with recurrent multiple carcinoma of the urinary bladder Stage Ta, grade 1-2, in whom all but one marker lesion was removed. Furthermore, we sought to define the safety profile.

This was a multicenter, open-label, randomized, Phase II study in which gemcitabine 2000 mg in 100 mL of unbuffered saline was instilled as a single dose (n = 11), two doses per week for 3 weeks (n = 11), or once weekly for 6 weeks (n = 10). Efficacy was evaluated using cystoscopy after 9 weeks. Toxicity was evaluated by assessing the liver, kidney, bone marrow, and coagulation function at defined intervals and by questionnaire.

A total of 32 patients were included, 2 of whom were subsequently excluded because of protocol violations. The overall complete remission rate was 31%. The respective subgroup response rate was 10% in the single-dose group, 44% in the once-weekly group, and 40% in the twice-weekly group. The most common side effect was nausea. One patient withdrew because of nausea and fever, and an additional 2 patients had reversible hematologic toxicity (mild thrombocytopenia causing delayed instillation and mild anemia). The side effects were generally in the multiple-dose groups, with an overrepresentation in women. Ten patients were unable to retain the drug intravesically for the full hour.

The results of our study have shown that gemcitabine has a tumor ablative effect when given intravesically for bladder cancer. A single dose seemed ineffective, and the multiple dosing regimens seemed effective. The side effects were generally mild.