To the Editor: Dr Oettle and colleagues1 reported the results of a multicenter randomized controlled trial of adjuvant chemotherapy with gemcitabine vs observation in patients with resectable pancreatic cancer. They provided the results of a survival analysis based on all eligible patients (intent-to-treat); they also presented a qualified survival analysis based on the population of patients who had received at least 1 complete cycle of gemcitabine, intended to more closely reflect the “true” therapeutic potential of the drug. Besides a large and significant advantage of gemcitabine in delaying the development of recurrent disease, there was a statistically nonsignificant improved survival in the treated group in the intent-to-treat analysis (log-rank P = .06), which reached statistical significance in the qualified analysis (P = .02). However, the potential survival advantage was only seen 2 years after treatment and was more prominent after 5 years, with twice as many survivors in the treated group compared with the observation group.