Background
Phase III randomized open-label trial. 492 patients with metastatic colorectal cancer who had received ≥2 prior regimens including fluoropyrimidine, oxaliplatin, irinotecan, and bevacizumab-based therapy. ECOG PS 0–1. Evaluated whether adding bevacizumab to TAS-102 improves outcomes over TAS-102 alone in the refractory setting, based on evidence that bevacizumab retains anti-angiogenic activity even after progression on prior bevacizumab-containing regimens (beyond-progression strategy).
Interventions and follow up
Arm A: TAS-102 35 mg/m² PO bid on days 1–5 and 8–12 of each 28-day cycle + bevacizumab 5 mg/kg IV on day 1 of each 14-day cycle, continued until disease progression or unacceptable toxicity
Arm B: TAS-102 35 mg/m² PO bid on days 1–5 and 8–12 of each 28-day cycle alone (monotherapy), until disease progression or unacceptable toxicity
Primary endpoint: Overall survival (OS)
mFollow up: 14.5 month
Arm B: TAS-102 35 mg/m² PO bid on days 1–5 and 8–12 of each 28-day cycle alone (monotherapy), until disease progression or unacceptable toxicity
Primary endpoint: Overall survival (OS)
mFollow up: 14.5 month
Results
mOS: 10.6 vs 7.5 months, HR 0.61 (95% CI 0.49–0.77), P<.001
mPFS: 5.6 vs 2.4 months, HR 0.44 (95% CI 0.36–0.54), P<.001
ORR: 5.5% vs 2.0%
DCR: 72.5% vs 56.5%
12-month OS rate: 43% vs 30%
mPFS: 5.6 vs 2.4 months, HR 0.44 (95% CI 0.36–0.54), P<.001
ORR: 5.5% vs 2.0%
DCR: 72.5% vs 56.5%
12-month OS rate: 43% vs 30%
Adverse events
Overall: Grade ≥3 AEs 72.4% (TAS-102 + bev) vs 69.5% (TAS-102 alone); discontinuation due to AEs 10.6% vs 6.7%
Hematologic (grade ≥3): Neutropenia 43.8% vs 38.3%, anaemia 12.5% vs 15.5%, febrile neutropenia 7% vs 5%
Bevacizumab-related (grade ≥3): Hypertension 4.5% vs 0.8%, GI perforation 1% (bev arm) vs 0%
Hematologic (grade ≥3): Neutropenia 43.8% vs 38.3%, anaemia 12.5% vs 15.5%, febrile neutropenia 7% vs 5%
Bevacizumab-related (grade ≥3): Hypertension 4.5% vs 0.8%, GI perforation 1% (bev arm) vs 0%
Conclusions
Adding bevacizumab to TAS-102 produced a clinically meaningful improvement in OS (mOS 10.6 vs 7.5 months, HR 0.61) and PFS (mPFS 5.6 vs 2.4 months, HR 0.44) in refractory mCRC. The 3.1-month absolute OS gain and consistent benefit across subgroups established TAS-102 + bevacizumab as the new standard of care in the refractory setting.
Key Limitations
Key Limitations: Open-label design (not blinded). ORR remains very low (5.5%), limiting utility for patients requiring objective response. Grade ≥3 neutropenia remains high (44%), requiring monitoring and G-CSF support strategies. Bevacizumab adds cost and IV administration burden. No direct head-to-head comparison with regorafenib; SUNLIGHT enrolled after RECOURSE, making cross-trial comparisons complicated by patient selection. Benefit appears independent of prior bevacizumab exposure but subgroup analyses are limited.
Clinical Context
SUNLIGHT led to FDA approval of TAS-102 + bevacizumab for previously treated mCRC in August 2023, largely supplanting TAS-102 monotherapy and positioning the combination as the preferred refractory regimen over regorafenib in indirect comparisons (mOS 10.6 vs ∼6 months with regorafenib). ESMO-MCBS score: 4. SUNLIGHT is arguably the most impactful late-line mCRC trial since RECOURSE, nearly doubling mOS vs RECOURSE monotherapy control arms.
References
References: Prager GW et al, NEJM 2023 (primary analysis)