Reply to S. Buti and S. Culine

Abstract

The comments Buti1 and Culine2 regarding our article3 about the role of cetuximab with or without paclitaxel in patients with advanced urothelial cancer. Their comments highlight some of the challenges in conducting clinical trials in this area. Buti proposes an additional analysis comparing the study's overall survival to the expected survival on the basis of the distribution of patient characteristics using the Bellmunt criteria published in 2010.4 Using these criteria, seven patients in the combination arm of our study had zero risk factors, 13 patients had one risk factor, seven patients had two risk factors, and one patient had three risk factors. The small size of these subsets precludes a meaningful statistical comparison of survival. We did report the progression-free and overall survival of patients who received chemotherapy for metastatic disease and those who had visceral metastases. However, because of the small sample size, we believe these results should be considered hypothesis generating only, and we discuss this limitation in our article.3 We agree with Culine that time from prior chemotherapy may be an important prognostic factor. We have contributed data from this study to a pooled analysis that recently demonstrated that a shorter time from prior chemotherapy is associated with worse response.5 This information may be used to design additional studies. KRAS mutation status was not known to be predictive of response to cetuximab in advanced colorectal cancer when this study began enrolling patients in 2006. However, we would not recommend testing for KRAS mutational status as an entry criteria for future studies of this combination on the basis of the low probability of finding a KRAS mutation in a patient with metastatic urothelial cancer. Juanpere et al6 found just one KRAS mutation among the 33 patients within the cohort of high-grade invasive bladder cancer. A higher but still low frequency of KRAS mutations were found in superficial tumor samples, which have a lower probability for metastasis. Other investigators have also found a low frequency of KRAS mutations in patients with urothelial carcinoma.7,8 This study was designed as a noncomparative phase II study; therefore, each arm was investigated separately. We hope to test this combination in a randomized comparative fashion using paclitaxel as a control arm. Vinflunine, although available in Europe, is not available in the United States and is not a practical comparator arm. Our study met its prespecified primary end point, which demonstrated that the combination of cetuximab and paclitaxel extended progression-free survival compared with historical controls, providing the rationale for future testing of this combination in a larger randomized trial.