In light of recent evidence from the ATOMIC trial, balancing adjuvant and neoadjuvant immunotherapy approaches continues to challenge clinical decision-making.
The recently-published results from the ATOMIC trial, demonstrating that the addition of atezolizumab to modified FOLFOX6 (fluorouracil, oxaliplatin, and leucovorin) regimen (mFOLFOX6) significantly improved disease-free survival among patients with stage III mismatch repair–deficient (dMMR) colon cancer, represents a major step forward in the treatment of this molecularly-defined subgroup (N Engl J Med. 2026 Mar 26;394(12):1155-1166). However, while an effect on a time-to-event endpoint, as reported in the study, typically leads to quick implementation of the respective regimen into both guidelines and clinical practice, the ATOMIC story presents more complex. The critical perspective on the trial can be summarised on two levels: a “conduct-question”, asking whether the trial strategy meets today’s requirements and standards, and a “strategic question” asking whether the overall trial strategy was appropriate – i.e. primarily focusing on whether the respective patient population should undergo upfront surgery (as required in an ATOMIC-based algorithm) or whether a neoadjuvant treatment approach offers better chances for long-term survival.
Starting with the first question, the choice of backbone chemotherapy - FOLFOX only in ATOMIC versus a more widespread use of capecitabine and oxaliplatin (CAPOX) in daily clinical practice - and the duration of chemotherapy in ATOMIC - six months instead of the three months suggested for the majority of patients based on the results of the IDEA-trial (Lancet Oncol. 2020 Dec;21(12):1620-1629) – are common points of criticism. Acknowledging that these aspects are unlikely to be explored prospectively and that there is no clear evidence of a negative interaction between different chemotherapy backbones and atezolizumab, a pragmatic interpretation of the study findings may be that the clinical use-case could include three months of CAPOX as an alternative to the six months of FOLFOX as used in ATOMIC.
The question of overall strategy is more complex to address. While ATOMIC is a positive phase III trial providing robust evidence of clinical benefit, neoadjuvant treatment strategies have not been evaluated on a similar scale. Their recognition as valuable treatment options is based primarily on the NICHE-2 trial, which was a moderately sized, single arm, phase II trial reporting extraordinary short- and long-term efficacy with just two doses of nivolumab and one dose of ipilimumab in advanced microsatellite-high instability (MSI-H)/dMMR colon cancer prior to surgery. It is important to note that a neoadjuvant strategy may, in principle, be disadvantaged compared to a classical adjuvant strategy because of selection bias, since the neoadjuvant cohort inevitably includes patients with complications of surgery whereas the adjuvant cohort excludes patients who are not fit to undergo postoperative treatment. Based on these considerations, it is a strong hypothesis that a neoadjuvant strategy, as suggested by the NICHE-2 trial (N Engl J Med. 2024 Jun 6;390(21):1949-1958), is a superior strategy as compared to the ATOMIC-based strategy. This may remain true despite the justified criticism that such a neoadjuvant approach could overtreat some patients and be curative itself, thereby introducing the question of organ preservation into the field of colon cancer.
Some of these aspects will be addressed by ongoing trials. While a head-to-head comparison of ATOMIC versus NICHE-2 is not yet available, the current data seem to support the frequent use of a NICHE-2-based regimen, at least in those patients who encounter complicated or mutilating surgery.
