Author reply

Abstract

I thank Munoz et al. for their interest in our study.1 I fully agree that some important questions regarding the daily management of patients with carcinoma of an unknown primary site remain unresolved. Indeed, the true impact of chemotherapy is unknown and to our knowledge no standard regimen has been established to date. A stratified-by-risk randomized trial comparing chemotherapy with best supportive care is supposed to be the best evidence-based option. However, I believe that such a trial is not feasible in daily practice. I find it very difficult to offer an abstention policy to patients who have an expected survival of approximately 1 year. It is the main reason why a prognostic model was developed and validated to select patients according to simple clinical and biologic features.2 In poor-risk patients for whom the expected median survival is no greater than 6 months, the question of the impact of chemotherapy is actually crucial and the chemotherapy regimen should be minimally toxic. Conversely, to our knowledge, the optimal chemotherapy regimen for patients considered to be at “good” risk (for whom the expected median survival is approximately 12 months) has yet to be determined. Because there is no standard, it was not obvious how best to determine the reference arm. Cisplatin alone was chosen to assess the effect of gemcitabine on overall survival in accordance with the previous experience of the French Group.3