ATOMIC
Randomized Trial of Standard Chemotherapy Alone or Combined With Atezolizumab as Adjuvant Therapy for Patients With Stage III Colon Cancer and Deficient DNA Mismatch Repair
Phase 3N = 712NCT02912559
- Population
- Patients aged 12 or older with completely resected stage III colon adenocarcinoma and dMMR status; 712 randomised from September 2017 to January 2023.
- Design
- International multicentre phase 3 trial (Alliance A021502) in patients with resected stage III dMMR colon cancer, randomised 1:1 to atezolizumab plus mFOLFOX6 for 6 months with atezolizumab continued as monotherapy (12 months in total) or to mFOLFOX6 alone for 6 months. Stratified by N stage, T stage and tumour sidedness. Primary endpoint disease-free survival; secondary endpoints overall survival and adverse events. The second prespecified interim analysis (data cutoff 4 February 2025) met the primary endpoint.
- Primary endpoint
- DFS
- Arms
- Atezolizumab plus mFOLFOX6 (experimental, n = 355): Atezolizumab 840 mg every 2 weeks for 12 cycles with mFOLFOX6 for 6 months, then atezolizumab monotherapy (12 months in total)
- mFOLFOX6 alone (control, n = 357): mFOLFOX6 for 12 cycles (6 months)
- People
- Sinicrope FA, Alliance for Clinical Trials in Oncology
DFSHR 0.50 (95% CI 0.35–0.73)Δ 36-month DFS+10.1 pointsOSHR 0.90 (95% CI 0.55–1.47)Δ 60-month OS+1.8 pointsΔ Grade 3 or 4 adverse events+12.2 pointsDFSHR 0.41 (95% CI 0.27–0.64)
Show the numbers
| Endpoint | Analysis | Atezolizumab plus mFOLFOX6 | mFOLFOX6 alone | Effect |
|---|---|---|---|---|
| DFS | primarymedian follow-up 40.9 monthsSinicrope 2026 | 86.3% at 3 y (81.8–89.8%) | 76.2% at 3 y (70.9–80.6%) | HR 0.50 (95% CI 0.35–0.73)P<0.001127 DFS events (46 with atezolizumab plus mFOLFOX6 and 81 with mFOLFOX6) per the full text. The stratified log-rank P value in the full text is 0.0002. |
| OS | interimmedian follow-up 45.8 monthsSinicrope 2026 | 89.7% at 5 y (85.2–92.9%) | 87.9% at 5 y (83.1–91.4%) | HR 0.90 (95% CI 0.55–1.47)P=0.6787Secondary endpoint, first analysis, immature: 31 deaths with atezolizumab plus mFOLFOX6 and 33 with mFOLFOX6 alone; no statistically significant difference. |
| Grade 3 or 4 adverse eventsTreated patients (346 atezolizumab-mFOLFOX6; 334 mFOLFOX6 alone, from the full text Table 2) | primarySinicrope 2026 | 84.1% | 71.9% | Regardless of attribution. Per the full text: non-haematological grade 3 or 4 events 69.4 percent versus 54.5 percent; grade 5 events in six patients with atezolizumab plus mFOLFOX6 (two judged treatment-related: sudden death and sepsis) and two with mFOLFOX6 alone (both judged unrelated). |
| DFSPatients who received more than six cycles of mFOLFOX6 (subgroup) | exploratorySinicrope 2026 | — | — | HR 0.41 (95% CI 0.27–0.64)Described as a prespecified subgroup analysis and, in the Discussion, as exploratory findings that suggest but do not prove that benefit depends on adequate chemotherapy exposure. |
| DFSPatients who received six or fewer cycles of mFOLFOX6 (subgroup) | exploratorySinicrope 2026 | — | — | HR 0.97 (95% CI 0.44–2.11)Same exploratory subgroup analysis as the preceding result. |
| Endpoint | Analysis | Atezolizumab plus mFOLFOX6 | mFOLFOX6 alone | Effect |
|---|---|---|---|---|
| DFS | primarymedian follow-up 40.9 monthsSinicrope 2026 | 86.3% at 3 y (81.8–89.8%) | 76.2% at 3 y (70.9–80.6%) | HR 0.50 (95% CI 0.35–0.73)P<0.001127 DFS events (46 with atezolizumab plus mFOLFOX6 and 81 with mFOLFOX6) per the full text. The stratified log-rank P value in the full text is 0.0002. |
| OS | interimmedian follow-up 45.8 monthsSinicrope 2026 | 89.7% at 5 y (85.2–92.9%) | 87.9% at 5 y (83.1–91.4%) | HR 0.90 (95% CI 0.55–1.47)P=0.6787Secondary endpoint, first analysis, immature: 31 deaths with atezolizumab plus mFOLFOX6 and 33 with mFOLFOX6 alone; no statistically significant difference. |
| Grade 3 or 4 adverse eventsTreated patients (346 atezolizumab-mFOLFOX6; 334 mFOLFOX6 alone, from the full text Table 2) | primarySinicrope 2026 | 84.1% | 71.9% | Regardless of attribution. Per the full text: non-haematological grade 3 or 4 events 69.4 percent versus 54.5 percent; grade 5 events in six patients with atezolizumab plus mFOLFOX6 (two judged treatment-related: sudden death and sepsis) and two with mFOLFOX6 alone (both judged unrelated). |
| DFSPatients who received more than six cycles of mFOLFOX6 (subgroup) | exploratorySinicrope 2026 | — | — | HR 0.41 (95% CI 0.27–0.64)Described as a prespecified subgroup analysis and, in the Discussion, as exploratory findings that suggest but do not prove that benefit depends on adequate chemotherapy exposure. |
| DFSPatients who received six or fewer cycles of mFOLFOX6 (subgroup) | exploratorySinicrope 2026 | — | — | HR 0.97 (95% CI 0.44–2.11)Same exploratory subgroup analysis as the preceding result. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| DFS | 0.5 | 0.35–0.73 |
| OS | 0.9 | 0.55–1.47 |
| DFS | 0.41 | 0.27–0.64 |
| DFS | 0.97 | 0.44–2.11 |