Abstract / Summary
Backgrounds: Fruquintinib is a selective vascular endothelial growth factor receptor (VEGFR)-1/2/3 inhibitor approved for previously treated metastatic colorectal cancer. As its use expands across gastrointestinal (GI) malignancies and combination regimens, randomized evidence is needed to define the toxicity profile most relevant to clinical monitoring, particularly hypertension, dermatologic toxicity, renal toxicity, bleeding, and thrombotic events. The objective of this study was to synthesize randomized controlled trial evidence to quantify the incidence and relative risk of key toxicities associated with fruquintinib in gastrointestinal malignancies. Methods: MEDLINE, EMBASE, and Cochrane CENTRAL were searched from inception through 1 January 2026 for phase II-III randomized controlled trials of fruquintinib in GI cancers reporting hypertension, proteinuria, hemorrhage, venous thromboembolism (VTE), and hand-foot skin reaction or palmar-plantar erythrodysesthesia (HFSR/PPE). Trial-reported CTCAE adverse events were pooled as risk ratios (RRs) with 95% confidence intervals (CIs) using Mantel-Haenszel random-effects models. Exploratory subgroup analyses were performed in metastatic colorectal cancer (mCRC). Results: Four randomized trials (n = 1872) were included. Fruquintinib significantly increased grade ≥3 hypertension (RR 9.01, 95% CI 4.67-17.40) and grade ≥3 HFSR/PPE (RR 26.00, 95% CI 6.42-105.29). Any-grade proteinuria was also increased (RR 1.89, 95% CI 1.30-2.74). Grade ≥3 hemorrhage occurred at low absolute rates and was numerically higher with fruquintinib, but the pooled estimate did not reach statistical significance (RR 1.82, 95% CI 0.92-3.61). VTE estimates were imprecise because of sparse events (any-grade VTE: RR 1.23, 95% CI 0.53-2.88; grade ≥3 VTE: RR 1.96, 95% CI 0.52-7.36). Conclusions: In randomized evidence across GI malignancies, the principal safety signals associated with fruquintinib are grade ≥3 hypertension and grade ≥3 HFSR/PPE, together with increased any-grade proteinuria. Grade ≥3 hemorrhage was uncommon and numerically higher, whereas VTE estimates remained imprecise. These findings support early blood pressure optimization, proactive dermatologic management, routine urinalysis monitoring, and individualized assessment of bleeding and thrombotic risk when initiating therapy.