Background
Phase III non-inferiority randomized trial. 599 patients with locally advanced rectal cancer (cT3–4 or cN+) at 20 centers in China. Compared short-course RT followed by sequential CAPOX chemotherapy (a TNT approach) to standard long-course concurrent CRT.
Interventions and follow up
Arm A (TNT): Short-course RT (25 Gy/5 fx) → CAPOX × 4 cycles → TME at 1–2 weeks after last cycle
Arm B (CRT): Long-course CRT (50 Gy/25 fx + concurrent capecitabine) → TME at 6–8 weeks
Primary endpoint: 3-year disease-free survival (DFS); non-inferiority margin HR <1.2
Median follow-up: 62.4 months
Arm B (CRT): Long-course CRT (50 Gy/25 fx + concurrent capecitabine) → TME at 6–8 weeks
Primary endpoint: 3-year disease-free survival (DFS); non-inferiority margin HR <1.2
Median follow-up: 62.4 months
Results
3-yr DFS: 64.5% vs 62.3%, HR 0.883 (95% CI 0.691–1.127), P<.001 for non-inferiority
pCR: 22.9% vs 12.5%, P=.002
R0 resection: 91.8% vs 90.7%, P=.73
3-yr OS: 86.5% vs 85.8%, HR 0.935, P=.66
Sphincter preservation: 72.9% vs 69.2%, P=.36
pCR: 22.9% vs 12.5%, P=.002
R0 resection: 91.8% vs 90.7%, P=.73
3-yr OS: 86.5% vs 85.8%, HR 0.935, P=.66
Sphincter preservation: 72.9% vs 69.2%, P=.36
Adverse events
Overall grade ≥3 acute: 42.9% (SCRT+CAPOX) vs 32.8% (LCRT), P=.013
Hematologic/GI: leukopenia 8.0% vs 4.7%, diarrhea 7.0% vs 5.4%
Neurologic: peripheral neuropathy any grade 15.4% vs 1.0%; postoperative complications similar
Hematologic/GI: leukopenia 8.0% vs 4.7%, diarrhea 7.0% vs 5.4%
Neurologic: peripheral neuropathy any grade 15.4% vs 1.0%; postoperative complications similar
Conclusions
SCRT followed by sequential CAPOX (a TNT approach) was non-inferior to long-course CRT for 3-year DFS in LARC (HR 0.883). SCRT+CAPOX produced significantly higher pCR rates (22.9% vs 12.5%) with comparable sphincter preservation and OS, supporting SCRT-based TNT as an alternative standard.
Key Limitations
Conducted exclusively in China, limiting generalizability to Western populations and surgical practice standards. Non-inferiority design precludes superiority conclusions for DFS. Higher acute grade ≥3 toxicity (43% vs 33%) and clinically meaningful peripheral neuropathy (15.4% any grade) in the SCRT+CAPOX arm are important considerations. CAPOX rather than FOLFOX was used, so results may differ from RAPIDO.
Clinical Context
STELLAR joins RAPIDO in establishing SCRT-based TNT as a guideline-endorsed option for LARC. The higher pCR (22.9%) with SCRT+CAPOX supports its use when organ preservation or maximum tumor downstaging is a priority. ESMO includes SCRT-based TNT as an evidence-based option for high-risk LARC; STELLAR is particularly relevant for Asian populations and settings where FOLFOX delivery is challenging.
References