Atezolizumab is cleared for adjuvant dMMR colon cancer after surgery

The Food and Drug Administration approved atezolizumab, sold as Tecentriq, on 8 October for use in combination with a fluoropyrimidine and oxaliplatin as adjuvant treatment of adults and children aged 2 and older with Stage III mismatch repair deficient (dMMR) colon cancer [s1][s2]. On the same day the agency approved the subcutaneous formulation, Tecentriq Hybreza, for the same indication in patients 12 and older who weigh at least 40 kg [s2][s3].

Atezolizumab is a PD-L1 checkpoint inhibitor. The adjuvant setting here means treatment given after surgery, in patients who have no visible remaining disease, to lower the chance that the cancer comes back. The qualifier that matters is "mismatch repair deficient": a tumour that has lost the machinery for correcting DNA copying errors, which makes it unusually visible to the immune system. That biology is the reason a checkpoint inhibitor was tested in this group at all.

What the label says

For colon cancer the regimen is atezolizumab 840 mg given intravenously every 2 weeks together with mFOLFOX6 chemotherapy for 6 months (12 cycles), followed by atezolizumab on its own for a further 6 months (13 cycles) — a total of up to one year of the antibody [s1]. mFOLFOX6 is a standard combination of oxaliplatin, leucovorin and fluorouracil [s1]. The single paediatric patient in the trial received weight-based dosing of 10 mg/kg every 2 weeks up to the 840 mg adult dose [s1].

The evidence

Approval rests on ATOMIC (ML39057; NCT02912559), a multicentre, randomised, open-label trial in 711 adults and one paediatric patient with Stage III dMMR colon cancer, all of whom had undergone a complete (R0) resection with no residual nodal or metastatic disease [s1]. Patients with rectal tumours were excluded, as were those with active autoimmune disease [s1]. Participants were randomised 1:1 to atezolizumab plus mFOLFOX6 followed by atezolizumab monotherapy, or to mFOLFOX6 alone [s1].

The primary endpoint was disease-free survival, assessed by the investigator [s1]. At the analysis, 46 of 355 patients (13%) in the atezolizumab arm had a disease-free survival event — a recurrence, a new primary tumour, or death from any cause — against 81 of 357 (23%) on chemotherapy alone [s1]. The hazard ratio was 0.50 (95% CI 0.35 to 0.73), with a one-sided p-value of 0.0001 [s1]. Median disease-free survival was not reached in either arm, which is expected in an early-stage population where most patients remain event-free [s1].

The trial population skews the way this disease does: median age 64 (range 13 to 91), 55% female, 85% white [s1]. Tumours were right-sided in 84% of patients, and 36% had N2 nodal status, meaning four or more involved nodes [s1]. Overall survival, a separate and more demanding question, was not the measure this approval was built on.

Safety, as far as it goes

In the ATOMIC safety population, serious adverse reactions occurred in 52% of patients receiving atezolizumab with mFOLFOX6; the most frequent (at 3% or more) were diarrhoea (8%), nausea (6%), abdominal pain (4.9%), peripheral sensory neuropathy (4.6%) and fever (3.8%) [s1]. Treatment with atezolizumab was permanently stopped in 17% of patients, most often for peripheral neuropathy [s1].

Much of that burden comes from the chemotherapy backbone rather than the antibody. But checkpoint inhibitors carry their own class risk — immune-mediated inflammation of the lung, bowel, liver, skin and endocrine glands — and the label's warnings on those reactions apply here as they do across atezolizumab's other uses [s1].

What is and is not established

What the evidence supports is that, in resected Stage III colon cancer with a mismatch repair deficient tumour, adding a year of atezolizumab to standard chemotherapy halves the rate of disease-free survival events over the follow-up so far [s1]. That is a large relative effect in a biomarker-defined group.

What it does not yet establish is a survival benefit, which the current data cannot show, or anything about patients outside the dMMR subset — the entire rationale rests on that specific tumour biology [s1]. Longer follow-up will determine whether the early separation in disease-free survival translates into fewer deaths.

Red flags: anyone who has had surgery for colon cancer and develops a persistent cough, breathlessness, severe diarrhoea, or yellowing of the skin or eyes while on immunotherapy should contact their oncology team promptly, because these can signal immune-mediated reactions. This article describes regulatory and research news and is not medical advice. Decisions about adjuvant treatment are for patients and their treating clinicians.