European Organisation for Research and Treatment of Cancer (EORTC) trial 22921 compared adjuvant fluorouracil-based chemotherapy (CT) to no adjuvant treatment in a 2 x 2 factorial trial with randomization for preoperative (chemo)radiotherapy in patients with resectable T3-4 rectal cancer. The results showed no significant impact of adjuvant CT on progression-free or overall survival, although a difference seemed to emerge at approximately, respectively, 2 and 5 years after the start of preoperative treatment. We further explored the data with the aim of refining our understanding of the long-term results. Data of 785 of the 1,011 randomly assigned patients who whose disease was M0 at curative surgery were used. Using meta-analytic methods, we investigated the homogeneity of the effect of adjuvant CT on the time to relapse or death after surgery (disease-free survival [DFS]) and survival in patient subgroups. Although there was no statistically significant impact of adjuvant CT on DFS for the whole group (P > .5), the treatment effect differed significantly between the ypT0-2 and the ypT3-4 patients (heterogeneity P = .009): only the ypT0-2 patients seemed to benefit from adjuvant CT (P = .011). The same pattern was observed for overall survival. Exploratory analyses suggest that only good-prognosis patients (ypT0-2) benefit from adjuvant CT. This could explain why, in the whole group, the progression-free and overall survival diverged only after the poor-prognosis patients (ypT3-4) had experienced treatment failure. Patients in whom no downstaging was achieved did not benefit. This also suggests that the same prognostic factors may drive both tumor sensitivity for the primary treatment and long-term clinical benefit from further adjuvant CT.
