Suggestions for regulatory agency approval of second-line systemic therapy for metastatic transitional cell carcinoma.

TO THE EDITOR: Bellmunt and colleagues are to be congratulated for conducting a large, randomized, phase III trial of vinflunine compared with placebo as second-line therapy for advanced transitional cell carcinoma (TCC) of the urothelium. As a result of this study, vinflunine has been approved by the European Medicine Agency and is the first agent approved by a regulatory agency for this indication. However, it is unlikely that vinflunine will be approved by the US Food and Drug Administration. Although an extension of survival, the primary end point was disappointingly not demonstrable with vinflunine compared with placebo by an intention-to-treat (ITT) analysis; there was a statistical improvement in response rate (RR) and progression-free survival (PFS) in the ITT population. The lack of approved second-line drugs in the United States for this disease represents a significant unmet medical need. While numerous chemotherapy and targeted agents demonstrate poor or no activity in small phase II trails a handful of second-line agents yield modest RRs of 10% to 20%, median PFS of 2 to 3 months, and a median survival of 6 to 9 months. We write to offer suggestions for study designs to address this gap and hopefully expedite approval of agents for this indication. Eligibility criteria for these phase II trials of second-line therapy have been variable and confusing. Some trials have considered all prior perioperative chemotherapy as first-line chemotherapy, while others have only included perioperative therapy administered within the prior 6 to 12 months before relapse as first-line therapy. Others required first-line therapy for metastatic disease. Some trials have enrolled a mixture of patients who had received previous perioperative therapy and recurred after an unspecified duration of time (a more favorable population) along with patients who experienced treatment failure with first-line therapy for metastatic disease. The trial by Bellmunt and colleagues permitted patients progressing at any time after first-line platinum-containing chemotherapy for metastatic disease and those who had received prior perioperative chemotherapy were excluded. More than 80% of those enrolled had progressed within 6 months after prior chemotherapy. Progression within 6 months of prior chemotherapy in the perioperative setting constitutes a poor prognostic group of patients, but may not be identical to patients who progress with pre-existent metastatic disease. Enrollment of both categories of such chemotherapy-resistant patients progressing within 6 months of prior chemotherapy in phase II trials will facilitate comparison across trials, but the patients should be stratified and analyzed separately. The recognized prognostic factors of visceral metastasis and performance status lower than 80 (demonstrated in the first-line setting) may confound outcomes, but probably have a more modest, but unproven, impact in the salvage setting. Despite the heterogeneity of the populations enrolled, the outcomes have been remarkably and uniformly dismal. More than 70% of patients treated by Bellmunt et al had visceral involvement, although patients with poor Karnofsky performance status lower than 80 were not enrolled. Such differences should be required to be reported in all trials and further study and identification of prognostic factors in the salvage setting is also necessary. Systemic agents for unmet needs in the salvage setting have been approved for a number of malignancies other than TCC, on the basis of modest activity (RR 15% to 20%) demonstrated in large phase II trials (eg, topotecan for sensitive, recurrent small-cell lung cancer, third-line capecitabine for recurrent breast cancer, and pralatrexate for recurrent peripheral T-cell lymphomas). The second-line setting of progressive advanced TCC within 6 months of prior chemotherapy deserves to belong in the same category. Placebo-controlled randomized trials have been conducted with new agents in salvage unmet need settings (eg, sunitinib for imatinib-resistant/intolerant gastrointestinal stromal tumor, docetaxel for non–small-cell lung cancer after prior chemotherapy). A phase III trial is being conducted to compare amrubicin with topotecan for recurrent small-cell lung cancer, although phase II evidence of substantial activity for amrubicin in this setting exists. However, we believe this design constitutes poor utilization of resources when available agents have negligible activity, and a more efficient paradigm of large phase II trials for the approval of new agents should be considered. Since new agents have been approved in other diseases in areas of unmet medical need based on modest activity in phase II studies, we believe that secondline chemotherapy-resistant TCC represents a similar unmet medical need and should be subjected to the same standards. Patients with metastatic TCC and relatively chemosensitive relapse beyond 6 months of first-line chemotherapy may continue to require randomized phase III trials to demonstrate improved outcomes with new regimens. In earlier phase II trials, vinflunine appeared to demonstrate modest activity as second-line therapy for advanced TCC. One study recruited 51 patients after an unspecified interval from prior therapy and demonstrated an 18% RR and a median duration of response of 9.1 months. Another large phase II trial of 175 patients with platinum-resistant disease (progression after prior platinum-based regimen as perioperative or first-line treatment for advanced disease within 12 months) confirmed these results (RR, 15%; median duration of response, 6 months). It is difficult to accrue patients to trials of TCC in general. This is an elderly population often with a rapidly progressive course of disease. A poor performance status and compromised renal function are relatively common, which precludes participation in many trials. It is unclear to us whether the activity of vinflunine seen in the prior large phase II studies required validation in this phase III trial and its attendant utilization of resources. We propose a large phase II trial design to evaluate new agents for the secondline setting of chemotherapy-resistant metastatic TCC. Patients with progression within 6 months of first-line chemotherapy (stratifying for perioperative or in the metastatic setting) would be eligible for such JOURNAL OF CLINICAL ONCOLOGY C O R R E S P O N D E N C E VOLUME 28 NUMBER 13 MAY 1 2010