FOLFOX for Stage II Colon Cancer? A Commentary on the Recent FDA Approval of Oxaliplatin for Adjuvant Therapy of Stage III Colon Cancer Axel Grothey, Mayo Clinic College of Medicine, Departments of Medical Oncology, Rochester, MN Daniel J. Sargent, Mayo Clinic College of Medicine, Health Science Research, Rochester, MN For almost 15 years, adjuvant chemotherapy has been known to improve disease-free survival (DFS) and overall survival (OS) in colon cancer. The pivotal study, 1 published by Charles Moertel in 1990, demonstrated improved OS and DFS for 12 months of treatment with bolus fluorouracil (FU) and levamisole, and led to the first National Cancer Institute (NCI) consensus recommendation for stage III colon cancer. 2 Subsequent studies conducted in the 1990s established 6 to 8 months of adjuvant therapy with bolus FU plus leucovorin (LV) as standard of care. The positive re- sults of the international MOSAIC (Multi-center Interna- tional Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer) trial, 3 which enrolled both stage II and III patients, led to US Food and Drug Administration approval of oxaliplatin plus FU/LV (FOLFOX) for patients with stage III colon cancer in No- vember 2004, 4 which followed the European approval as “adjuvant treatment of stage III (Dukes C) colon cancer after complete resection of the primary tumor” in Septem- ber 2004. The US Food and Drug Administration approval was based on the demonstration of the statistical superiority of FOLFOX to infusional plus bolus FU/LV (LV5FU2 regi- men) in 3- and 4-year DFS in the stage III subgroup of patients in the MOSAIC trial. 3,5 Whereas the approval of FOLFOX in the adjuvant setting is a welcome and clinically important addition to improving outcome in patients with locally advanced colon cancer, the restriction of the labeled approval (in the United States and Europe) to stage III disease alone warrants further discussion. The MOSAIC trial was designed to evaluate the efficacy of FOLFOX4 compared with LV5FU2 in patients with both stage II and stage III colon cancer. 3 The study design as- sumed that 40% of patients enrolled would have stage II disease, with stage III disease in the remaining 60%. Ran- domization was stratified by stage, and this projected ratio was achieved exactly when the accrual goal of 2,248 patients was met. The protocol-specified primary end point for the trial was 3-year DFS for the entire study population, and although an analysis stratified by stage was planned, subset analyses comparing the efficacy of treatment within each stage were not. Thus, the approval of FOLFOX as adjuvant treatment solely for stage III colon cancer is based on an analysis that was neither protocol specified nor powered for this analysis. This practice is statistically questionable, and to our knowledge is unusual in the history of the US Food and Drug Administration. Accepting subgroup analysis of clinical trials as a basis for drug approval may open the door for potential future approaches to “rescue” an otherwise negative trial by dem- onstrating a statistically significant result in a specific sub- group of patients enrolled in the trial. Alternatively, a trial could have a positive (significant) overall result, but due to power considerations, fail to be positive in any specific subgroup. Previously, the US Food and Drug Administra- tion has (rightfully) rejected subgroup analyses as a poten- tial basis for drug approval. Therefore, it is of great interest to note that in the case of the MOSAIC trial, the US Food and Drug Administration departed from this well- established practice. The precedent of approval in a subset of patients might be understood if there was clear and compelling evidence of an interaction, in this case of treat- ment efficacy by stage, or if the subset of patients excluded JOURNAL OF CLINICAL ONCOLOGY COMMENTS AND CONTROVERSIES VOLUME 23  NUMBER 15  MAY 20 2005 3311 Journal of Clinical Oncology, Vol 23, No 15 (May 20), 2005: pp 3311-3313 DOI: 10.1200/JCO.2005.11.691 Downloaded from jco.ascopubs.org on July 27, 2016. For personal use only. No other uses without permission. Copyright © 2005 American Society of Clinical Oncology. All rights reserved.