Adding immunotherapy to chemo cuts relapse in mismatch-repair-deficient colon cancer

Adding the immunotherapy atezolizumab to standard chemotherapy after surgery sharply reduced the chance that stage III colon cancer with a specific genetic defect came back: in the phase 3 ATOMIC trial, three-year disease-free survival was 86.3% with atezolizumab plus chemotherapy against 76.2% with chemotherapy alone [s1]. The benefit was confined by design to tumours that are mismatch-repair-deficient, a subtype long known to respond unusually well to immune-checkpoint drugs [s1].

Mismatch-repair-deficient (dMMR) tumours cannot fix certain DNA copying errors, so they accumulate mutations that make them look conspicuous to the immune system — the biological reason checkpoint inhibitors work well in this group. Those drugs were already established in advanced dMMR bowel cancer; ATOMIC tested whether adding one to chemotherapy in the adjuvant setting, immediately after curative surgery, could stop relapses before they start [s1]. Standard adjuvant treatment for stage III colon cancer is a fluoropyrimidine drug combined with oxaliplatin, and the trial kept that backbone in both groups, isolating the effect of the immunotherapy added on top [s1].

What they did

ATOMIC randomly assigned 712 patients with resected stage III dMMR colon cancer in a 1:1 ratio: 355 to atezolizumab plus modified FOLFOX6 chemotherapy (fluorouracil, oxaliplatin and leucovorin) for six months, followed by atezolizumab alone to complete a year of treatment, and 357 to modified FOLFOX6 alone for six months [s1]. The trial was registered as ATOMIC and enrolled only patients whose resected tumours were confirmed to be mismatch-repair-deficient [s2]. The primary endpoint was disease-free survival — time without the cancer returning or the patient dying [s1]. The enrolled group skewed toward higher-risk disease: the median age was 64 years, 55.1% were women, and 53.9% had tumours classified as T4, N2, or both [s1].

What it showed

At a median follow-up of 40.9 months, three-year disease-free survival was 86.3% (95% confidence interval, 81.8 to 89.8) in the atezolizumab group and 76.2% (95% confidence interval, 70.9 to 80.6) with chemotherapy alone, a hazard ratio for recurrence or death of 0.50 (95% confidence interval, 0.35 to 0.73; P<0.001) [s1]. A hazard ratio of 0.50 corresponds to roughly halving the rate of relapse or death over the follow-up period — a large effect for an adjuvant therapy, where the goal is to convert more surgeries into cures rather than to shrink visible tumours. With a median follow-up approaching three and a half years, the estimate rests on reasonably mature data for the three-year time point it reports [s1].

The trial's design is a case study in the value of testing a treatment in the right patients. dMMR tumours make up only a minority of stage III colon cancers, and restricting enrolment to that subtype is what allowed a moderately sized trial to detect so clear an effect — the same logic behind matching a drug to a tumour's molecular profile rather than treating all colon cancers alike. Disease-free survival is the conventional yardstick for adjuvant colon-cancer trials because relapses cluster in the first few years after surgery, so a gap that is already this wide at three years is unlikely to be a fluke of short follow-up [s1].

The cost side

The benefit came with added toxicity. Adverse events of grade 3 or 4 occurred in 84.1% of patients who received atezolizumab plus chemotherapy and in 71.9% of those who received chemotherapy alone [s1]. Checkpoint inhibitors can trigger immune-mediated inflammation of the bowel, skin, liver and endocrine glands, and adding one to a year of treatment lengthens the exposure over which those effects can appear. Whether the extra six months of atezolizumab monotherapy is necessary, or whether the chemotherapy phase carries most of the benefit, is not resolved by this trial, which tested the combination as a single package against chemotherapy alone [s1]. Grade 3 or 4 events are those serious enough to need medical intervention or hospitalisation, so the roughly 12-percentage-point excess in the atezolizumab group is not a trivial one, even set against a relapse benefit of similar size.

What to watch

ATOMIC was funded by the National Cancer Institute — part of the U.S. National Institutes of Health — and by Genentech, which makes atezolizumab [s1]. Overall survival, a secondary endpoint, was not yet mature, so the disease-free-survival gain has not yet been shown to translate into longer life, though in adjuvant colon cancer disease-free survival at three years has historically tracked survival closely [s1]. The practical question is testing: the result only helps patients whose tumours are checked for mismatch-repair status, a test that is not universal, and it does not speak to the far larger group of colon cancers that are mismatch-repair-proficient. For those, and for the many cancers still found late, the older priorities hold — screening that catches disease before it spreads remains where most of the survival gain in colon cancer is won.