Background
Phase III randomized double-blind placebo-controlled trial. 760 patients with heavily pretreated metastatic colorectal cancer (≥3 prior lines including fluoropyrimidine, oxaliplatin, irinotecan, and bevacizumab; and cetuximab or panitumumab if KRAS wild-type). ECOG PS 0–1. Conducted at 114 centers across 16 countries. Evaluated regorafenib — a multikinase inhibitor targeting VEGFR, TIE2, PDGFR, FGFR, KIT, RET, and RAF — as a late-line option.
Interventions and follow up
Arm A: Regorafenib 160 mg PO daily on days 1–21 of each 28-day cycle (3 weeks on, 1 week off) + best supportive care until disease progression or unacceptable toxicity
Arm B: Placebo + best supportive care
Primary endpoint: Overall survival (OS)
mFollow up: 6.4 month
Arm B: Placebo + best supportive care
Primary endpoint: Overall survival (OS)
mFollow up: 6.4 month
Results
mOS: 6.4 vs 5.0 months, HR 0.77 (95% CI 0.64–0.94), P=.0052
mPFS: 1.9 vs 1.7 months, HR 0.49 (95% CI 0.42–0.58), P<.0001
DCR: 41% vs 15%, P<.0001
ORR: 1% vs 0.4%
mPFS: 1.9 vs 1.7 months, HR 0.49 (95% CI 0.42–0.58), P<.0001
DCR: 41% vs 15%, P<.0001
ORR: 1% vs 0.4%
Adverse events
Overall: Grade ≥3 AEs 54% (regorafenib) vs 14% (placebo); dose reductions required in 38%
Dermatologic: Hand-foot skin reaction (grade ≥3) 17%
Constitutional/GI: Fatigue (grade ≥3) 10%, diarrhea 7%
Vascular/hepatic: Hypertension (grade ≥3) 7%, elevated ALT/AST 5%
Deaths: Drug-related deaths 4 (regorafenib) vs 0 (placebo)
Dermatologic: Hand-foot skin reaction (grade ≥3) 17%
Constitutional/GI: Fatigue (grade ≥3) 10%, diarrhea 7%
Vascular/hepatic: Hypertension (grade ≥3) 7%, elevated ALT/AST 5%
Deaths: Drug-related deaths 4 (regorafenib) vs 0 (placebo)
Conclusions
Regorafenib produced a statistically significant OS benefit vs placebo in heavily pretreated mCRC (mOS 6.4 vs 5.0 months, HR 0.77, P=.0052), providing proof-of-concept that antiangiogenic multikinase inhibition retains activity in chemotherapy-refractory mCRC. The absolute benefit is modest (∼1.4 months median OS), and toxicity is substantial.
Key Limitations
Key Limitations: Modest absolute OS gain (1.4 months) of uncertain quality-of-life impact given substantial toxicity (54% grade ≥3 AEs). ORR is near zero (<1%), limiting utility for patients needing tumor response. No predictive biomarker identified — benefit appears homogeneous across unselected mCRC. Subsequent head-to-head comparison with TAS-102 (SUNLIGHT) showed TAS-102 + bevacizumab is superior to regorafenib for OS. Starting dose of 160 mg is often not tolerated; dose-escalation strategies (60→120→160 mg) may improve tolerability (ReDOS trial).
Clinical Context
CORRECT established regorafenib as an FDA-approved (September 2012) late-line option for mCRC. However, in the era of SUNLIGHT (TAS-102 + bevacizumab, mOS 10.6 months) and with approved KRAS/HER2-targeted agents, regorafenib is increasingly used as a later option or in patients without actionable biomarkers. TAS-102 + bev is generally preferred over regorafenib based on SUNLIGHT data. ESMO-MCBS score: 2.