Abstract
The phase III ATOMIC trial recently showed that adding atezolizumab to standard adjuvant chemotherapy improved disease-free survival (DFS) in patients with mismatch-repair-deficient (dMMR) stage III colon cancer; however, its efficacy in mismatch-repair-proficient (pMMR) disease remains unknown. To address this issue, we reconstructed individual patient-level survival data (IPD) for the centrally confirmed dMMR subgroup, as well as those without centrally confirmed dMMR status (an approximation of the pMMR subgroup) in ATOMIC. The analysis of reconstructed IPD faithfully reproduced the trial's reported DFS gain by adding atezolizumab to adjuvant chemotherapy in the dMMR subgroup (3-year DFS, 86.2% vs 77.0%; HR, 0.52 (95% CI 0.36 to 0.77)). Intriguingly, DFS was also in favor of the chemo-immunotherapy arm in the approximated pMMR subgroup (3-year DFS, 87.1% vs 77.4%; HR, 0.45 (95% CI 0.16 to 1.27)). The interaction test further demonstrated that treatment effect did not differ by MMR status (P(interaction)=0.808). These findings raise the possibility that adjuvant chemo-immunotherapy may confer benefit even in pMMR stage III colon cancer, which underscores the need for prospective validation in this population.