Increasing the Interval Between Neoadjuvant Chemoradiotherapy and Surgery in Rectal Cancer.

To the Editor: We read with great interest the metaanalysis by Petrelli et al reporting the impact of lengthening the waiting time between the end of a chemoradiotherapy (CRT) and the surgical resection for adenocarcinoma of the rectum. The authors concluded, after inclusion of 13 studies, that a longer waiting time was associated with an increased rate of pathological complete response (pCR) defined as ypT0N0. Beside the inherent drawbacks of every meta-analysis, we would like to temper the conclusions of the authors. As it is often the case, the majority of the included studies (69%) were retrospective with inevitable bias. Tran et al reported in their article that the selection of the waiting period was done by the attending surgeon and probably influenced by the tumor size or response to CRT. In 2 studies, patients with a clinical complete response after CRT were excluded before the analysis of the influence of lengthening the interval before surgery. This point is of interest as these patients probably had a sterilized tumor. At the very least, the rate of pCR is less relevant in such selected population. The therapeutic management also differed in some studies. In 5 studies, different types of chemotherapy were used (biotherapy, irinotecan, or oxaliplatin) in addition to the usual regimen of 5-FU. The authors raised that point in their meta-analysis and performed a new analysis after exclusion of these 5 studies. However, Garcia-Aguilar et al added 2 cycles of mFOLFOX-6 after completion of CRT for every patient with a longer interval, where Sloothaak et al more frequently (P < 0.001) gave induction or interval systemic chemotherapy for some patients with a long delay before surgery. These 2 studies make more complex the interpretation of the higher rate of pCR after an increased interval. Is the higher rate due to a longer waiting interval or the addition