The watch-and-wait (w&w) strategy is associated with frequent local regrowth (LR). Distant metastases (DM) occur more often in the patients with LR than in those without. However, it is unknown whether omitting immediate surgery results in the additional risk of DM. A systematic review and meta-analysis were performed to determine the maximum risk of additional DM. To estimate this, we used data showing the proportions of DM in patients with and without LR, assuming that the excess DM in patients with LR may develop in two ways: from subclinical DM already present at baseline and due to seeding from the uncontrolled primary tumor, and that the incidence of subclinical DM at baseline in the LR subgroup is at least not lower than in the non-LR subgroup. Based on the calculated rate of excess DM in the LR subgroup we have obtained the rate for the whole group of patients undergoing w&w. The maximum estimated risk of additional DM was 3.0% (95% CI: 1.2-4.9%) in the total group. After correction for short follow-up, the maximum risk at 5 years was 6.5%. Thus, the risk of excess DM is between 0% and 6.5%. Other evidence from a systematic review and the conservative assumptions taken for the calculation of the correction suggest that this maximum risk may be overestimated. The additional risk of DM seems to be low. However, the high probability of bias, heterogeneity of the patients' population and low quality of evidence make our estimation uncertain.

Background: We aimed to evaluate whether total neoadjuvant therapy (TNT) results in long-term overall survival (OS) or quality of life (QoL) benefit compared with chemoradiation if all patients are being considered for radical resection, and whether the ATRESS phenomenon (i.e., reduction in post-metastatic survival) impacts OS after TNT. Methods: Systematic review of randomised trials comparing TNT with neoadjuvant (chemo)radiation. Results: Six trials were identified. Follow-ups were too short to resolve whether TNT improves long-term OS. QoL analysis in one trial showed worse long-term neurotoxicity-related QoL (any neurotoxicity: 14% vs. 3%), higher rate of grade III+ acute toxicity (48% vs. 25%), longer duration of neoadjuvant treatment (19 vs. 6 weeks) and higher rate of locoregional failure (10% vs. 7%) in TNT vs. chemoradiation. This should be weighed against an absolute 8% reduction in the incidence of distant metastases (DM) after TNT. ATRESS could explain a discrepancy between reduction of DM and the absence of OS improvement after TNT in one trial. Conclusion: In the light of unproven OS benefit, the gain of TNT (reduction of DM) does not seem to outweigh the harm (excess of toxicity). ATRESS can be a reason for the absence of the OS benefit despite the reduction in DM.