FDA Approves First Adjuvant Immunotherapy for Colon Cancer

Tecentriq (atezolizumab)What to know

30-second read

For Oncology · Gastroenterology · Pathology

Who

Adults and children (2+) with resected, node-positive (stage III) colon cancer whose tumor is mismatch-repair deficient (dMMR) or microsatellite instability-high (MSI-H).

What’s new

The first FDA-approved adjuvant immunotherapy for colon cancer: atezolizumab (Tecentriq, IV, or Tecentriq Hybreza, subcutaneous) added to oxaliplatin-based chemotherapy.

Evidence

ATOMIC (712 patients): 3-year disease-free survival 86.3% vs. 76.2% with chemo alone — HR 0.50 (95% CI 0.35–0.73), p<0.001.

Big advantage

Roughly a 50% relative cut in three-year recurrence or death risk, in a tumor subtype that responds poorly to chemotherapy alone.

Key consideration

Grade 3/4 adverse events rose to 84.1% vs. 71.9%; 2 treatment-related deaths in the atezolizumab arm. Overall survival is not yet proven.

Availability

Approved 8 October 2026. Both formulations use existing J-codes (J9022 IV, J9024 SC); dMMR/MSI-H testing is already standard at colon cancer diagnosis.

Practice impactDiscuss with eligible patients

A genuine advance for a chemo-resistant tumor subtype, but it adds real toxicity, including two fatal events, to patients who would otherwise be cancer-free after surgery. Confirm dMMR/MSI-H status before offering it; survival benefit is not yet proven.

Read the full physician analysis

Full analysis · 9 min readFDA approval notice, 8 Oct 2026 · ATOMIC trial (A021502/NCT02912559) · Genentech, 9 Oct 2026

The U.S. Food and Drug Administration approved atezolizumab (Tecentriq, and its subcutaneous co-formulation Tecentriq Hybreza) in combination with oxaliplatin-based chemotherapy on 8 October 2026, for adjuvant treatment of stage III colon cancer that is mismatch-repair deficient (dMMR) or microsatellite instability-high (MSI-H). Genentech, which announced the approval the following day, calls it Tecentriq's 12th U.S. indication — and the first time any checkpoint inhibitor has been approved for colon cancer outside the metastatic setting.

Immunotherapy Moves Into the Adjuvant Setting

Checkpoint inhibitors have had a foothold in mismatch-repair-deficient colorectal cancer since 2017, when nivolumab won accelerated approval for refractory metastatic disease. Pembrolizumab followed in June 2020 with first-line approval for unresectable or metastatic MSI-H/dMMR colorectal cancer, based on KEYNOTE-177's 16.5-versus-8.2-month progression-free survival advantage over chemotherapy. Nivolumab plus ipilimumab reached the same first-line metastatic setting in April 2025. In each case, immunotherapy was answering a question about disease that had already spread. This approval is different: it reaches patients after curative-intent surgery, aiming to prevent recurrence in a tumor subtype that, on its own biology, tends to resist standard chemotherapy.

Who the New Label Covers

The approval is for patients with completely resected, node-positive (stage III) colon cancer whose tumor has been confirmed dMMR or MSI-H. Adults can receive IV atezolizumab at 840 mg every two weeks, 1,200 mg every three weeks, or 1,680 mg every four weeks, paired with chemotherapy for six months and then continued alone for six months, or until recurrence or unacceptable toxicity. Children two years and older can receive weight-based IV dosing (10 mg/kg, maximum 840 mg, every two weeks, or 15 mg/kg, maximum 1,200 mg, every three weeks). Tecentriq Hybreza, the subcutaneous formulation co-formulated with hyaluronidase-tqjs, is approved for patients 12 years and older weighing at least 40 kg, dosed as one 15 mL injection every three weeks on the same combination-then-monotherapy structure; dosing below that age or weight threshold is not established.

The trial behind the approval, ATOMIC, tested only atezolizumab added to mFOLFOX6 — oxaliplatin, leucovorin and 5-fluorouracil. The approved label's own wording is broader, authorizing atezolizumab with “a fluoropyrimidine and oxaliplatin” rather than naming FOLFOX specifically, which leaves room to read CAPOX (capecitabine plus oxaliplatin) as an alternative backbone. No trial evidence for that substitution was found for this story. Prescribers reaching for CAPOX on the strength of label wording alone should know the evidence base is FOLFOX, not CAPOX.

What the ATOMIC Trial Showed

ATOMIC (Alliance A021502, NCT02912559) was an NCI-sponsored, Alliance for Clinical Trials in Oncology-led study run in partnership with Genentech and the German AIO cooperative group. It enrolled 712 patients with resected stage III dMMR colon cancer between September 2017 and January 2023, randomizing them 1:1 to atezolizumab plus mFOLFOX6 for 12 cycles followed by atezolizumab alone, or to mFOLFOX6 alone for 12 cycles. The primary endpoint was disease-free survival, and the results were published in the New England Journal of Medicine in early 2026.

ATOMIC at a glance

86.3% vs. 76.2%Three-year disease-free survival, atezolizumab-chemo vs. chemo alone

HR 0.50Hazard ratio for recurrence or death (95% CI, 0.35–0.73; p<0.001)

712Patients randomized 1:1; median follow-up 40.9 months

84.1% vs. 71.9%Grade 3–4 adverse events, atezolizumab-chemo vs. chemo alone

Figures as reported in the ATOMIC trial's published results and cited in FDA's and Genentech's approval announcements. Median disease-free survival was not reached in either arm.

The effect size — a hazard ratio of 0.50, meaning roughly half the risk of recurrence or death at three years — is large by the standards of adjuvant colon cancer trials, where incremental gains of a few percentage points are the norm. Overall survival, a secondary endpoint, remains immature: no statistically significant difference between arms has emerged yet, and longer follow-up is needed before that question is settled. That matters for how this is framed to patients: the proven benefit so far is fewer recurrences, not yet a demonstrated survival advantage.

The Safety Trade-off in a Curative Setting

Grade 3 or 4 adverse events of any cause occurred in 84.1% of patients on atezolizumab plus chemotherapy, against 71.9% on chemotherapy alone. Grade 3/4 hematologic toxicity ran 46.8% versus 38.6%; nonhematologic toxicity 69.4% versus 54.5%, including grade 3/4 fatigue in 10.1% versus 3.3%. Immune-related adverse events — hypothyroidism, hyperglycemia, colitis, maculopapular rash — occurred at rates at least five percentage points higher with atezolizumab, though the proportion reaching grade 3 or 4 was not significantly different between arms. Six patients died in the atezolizumab arm against two on chemotherapy alone; two of those deaths (sudden death, sepsis) were adjudicated treatment-related.

Atezolizumab carries no boxed warning — standard for a PD-L1 inhibitor — but its label's warnings and precautions cover the full range of immune-mediated reactions that can affect any organ system: pneumonitis, colitis, hepatitis, endocrinopathies (thyroid, adrenal, pituitary, type 1 diabetes), severe skin reactions, nephritis, infusion reactions, and solid-organ transplant rejection in the rare patient with a prior transplant. None of that is new to atezolizumab. What is new is the population it is now reaching: patients who have already had their tumor surgically removed, and who would, on the old standard of care, complete chemotherapy and return to surveillance without six to twelve more months of an agent carrying these risks. The ATOMIC safety data say that trade-off produced a net benefit in this trial population — but it is a trade-off, not a free addition, and worth saying so explicitly when discussing the regimen with a patient who is already cancer-free after surgery.

Where This Leaves the Adjuvant Standard

Oxaliplatin-based chemotherapy — FOLFOX or CAPOX — has been the backbone of adjuvant treatment for resected stage III colon cancer for roughly two decades, since trials established oxaliplatin's benefit over fluoropyrimidine alone in the early 2000s. Duration is already individualized by risk: NCCN guidance generally allows three months of CAPOX or three to six months of FOLFOX for lower-risk (T1–3, N1) disease, and longer courses for higher-risk (T4 or N2) disease. Atezolizumab's addition does not replace that framework; it sits on top of it, for the dMMR/MSI-H minority of patients whose tumors already behave differently. Most stage III colon cancer is mismatch-repair proficient (pMMR), and ATOMIC says nothing about those patients — they remain on chemotherapy alone.

Checkpoint inhibitors in colorectal cancer, by setting

	Atezolizumab + chemo (new)	Pembrolizumab	Nivolumab + ipilimumab
Setting	Adjuvant, resected stage III dMMR/MSI-H colon cancer	First-line, unresectable or metastatic MSI-H/dMMR colorectal cancer	First-line, unresectable or metastatic MSI-H/dMMR colorectal cancer
U.S. approval	8 October 2026	29 June 2020	8 April 2025
Pivotal trial	ATOMIC: DFS 86.3% vs. 76.2% at 3 years, HR 0.50	KEYNOTE-177: PFS 16.5 vs. 8.2 months, HR 0.60	CheckMate 8HW: PFS not reached vs. 5.8 months, HR 0.21
Disease status	Curative intent — no detectable disease after surgery	Advanced/metastatic — measurable disease	Advanced/metastatic — measurable disease

Comparator trial figures as reported in each drug's own approval record; this table does not imply a head-to-head comparison between regimens in different disease settings. Prescribe from current labels, not this table.

Coding, Cost and What Is Still Unknown

Both formulations bill under existing, permanent J-codes — J9022 for IV atezolizumab and J9024 for the subcutaneous Tecentriq Hybreza — so no new coding infrastructure is required for this indication specifically. No atezolizumab biosimilar has reached the U.S. market as of this approval. What remains genuinely open: a population-based estimate of how many stage III colon cancer patients are actually dMMR/MSI-H was not identified with enough independent corroboration to state here with confidence (published figures for earlier-stage colorectal cancer commonly cluster in the 10–15% range, but a number specific to stage III disease and this approval was not confirmed); whether a specific companion diagnostic is named in the approval; and how quickly payers will treat this as routine versus requiring prior authorization for an adjuvant regimen with a DFS, not yet an OS, endpoint behind it.

• Confirm biomarker status before the conversation. dMMR/MSI-H testing is already standard at diagnosis for colorectal cancer; make sure the result is in hand and unambiguous before raising this regimen with a patient.
• Set expectations on survival, not just recurrence. The approval rests on a disease-free survival benefit; overall survival data are still immature. Say so plainly rather than implying a proven survival advantage.
• Counsel on toxicity in a curative context. A patient who is already free of detectable disease is taking on real immune-related risk, including the toxicity profile seen in ATOMIC's two treatment-related deaths, for a reduction in recurrence risk rather than treatment of active cancer.
• Use the evidence-based backbone. ATOMIC tested FOLFOX, not CAPOX; the label's broader wording is not itself evidence that the two are interchangeable here.
• Referral and sequencing. This is a decision for medical oncology following surgical resection and pathology confirmation of dMMR/MSI-H status — a three-specialty handoff that benefits from being flagged early, not after chemotherapy has already started.

The honest summary is that this is a real advance for a biologically distinct, chemotherapy-resistant minority of colon cancer, not a change to how most colon cancer is treated. For the patients it covers, it is the first time a checkpoint inhibitor has been offered with curative rather than disease-control intent in this cancer — and the first time oncologists have had to weigh immunotherapy's toxicity against a patient's chance of staying cancer-free, rather than against active, measurable disease. Every story on this desk is filed by specialty on the For Physicians front.

How this approval got here, and what is still open

1. Sep 2017

   ATOMIC begins enrolling

   712 patients with resected stage III dMMR colon cancer, randomized to atezolizumab plus mFOLFOX6 or mFOLFOX6 alone.

2. Jan 2023

   Enrollment completes

   The trial closes to new patients after roughly five years of accrual.

3. Early 2026

   Results published

   The New England Journal of Medicine publishes ATOMIC's disease-free survival results.

4. 8 Oct 2026

   FDA approves atezolizumab plus chemo

   Adjuvant treatment of stage III dMMR/MSI-H colon cancer; Genentech announces the approval the next day.

5. Ahead

   Overall survival maturation

   Longer follow-up is needed before a survival benefit, as opposed to a recurrence benefit, can be confirmed.

Frequently asked questions

Does this approval mean every stage III colon cancer patient should get atezolizumab?

No. It applies only to the dMMR/MSI-H subset, roughly a minority of stage III cases (estimates in the literature commonly run in the 10–15% range, though a precise population-based figure specific to stage III disease was not independently confirmed for this story). Patients whose tumors are mismatch-repair proficient (pMMR) — the majority — are not included in this approval and should continue on standard oxaliplatin-based chemotherapy; the ATOMIC trial did not study them.

Does this change how colon cancer is tested at diagnosis?

Not procedurally. Universal mismatch-repair/microsatellite-instability testing at diagnosis is already standard practice for newly diagnosed colorectal cancer, primarily to screen for Lynch syndrome. This approval raises the stakes of a test most patients already receive, rather than adding a new one — though confirming the result is now a prerequisite for a specific treatment decision, not just a screening data point.

Is the chemotherapy backbone in this approval specifically FOLFOX, or could CAPOX be substituted?

The ATOMIC trial tested only atezolizumab plus mFOLFOX6. The approved label's wording is broader — a fluoropyrimidine and oxaliplatin, not FOLFOX by name — but no trial evidence for an atezolizumab-CAPOX combination was identified for this story. Treat FOLFOX as the evidence-based backbone until a label or guideline explicitly states otherwise.