Reduced fruquintinib dose showed no significant survival association in colorectal cancer cohort

In a multicenter retrospective cohort of 69 patients with refractory metastatic colorectal cancer, an initial fruquintinib dose below 5 mg was not significantly associated with progression-free survival or overall survival compared with the standard 5 mg dose. Median progression-free survival was 3.68 months in the reduced-dose group and 3.98 months in the 5 mg group, while median overall survival was 5.88 and 7.43 months, respectively. The results were not statistically significant and should not be interpreted as evidence that the two strategies are equivalent.

The investigators described the findings as hypothesis-generating because the sample was modest, the groups were not randomized, and residual confounding could not be excluded. Fruquintinib is used as third-line or later therapy for metastatic colorectal cancer that has progressed after standard treatments. Treatment can be complicated by hypertension, hand-foot syndrome and proteinuria, which may require dose modification or discontinuation in clinical practice.

Investigators therefore examined whether the initial dose was associated with survival, dose changes, proteinuria and early treatment exposure. The study included 44 patients who started at 5 mg and 25 who started below 5 mg. Patients receiving reduced initial doses generally had worse performance status than those receiving 5 mg, an important baseline difference that could influence both treatment decisions and subsequent outcomes.

For progression-free survival, the reported hazard ratio was 1.26, with a 95% confidence interval of 0.59 to 2.64 and a p-value of 0.54. For overall survival, the hazard ratio was 1.40, with a 95% confidence interval of 0.60 to 3.14 and a p-value of 0.42. Neither confidence interval excluded 1, and neither association reached the study’s threshold for statistical significance.

Initial dose also showed no significant association with either survival outcome in multivariable Cox proportional hazards models or propensity score-adjusted Cox analyses. The latter analyses were intended to address potential selection bias, but the broad confidence intervals reflect substantial uncertainty in the estimates. The data therefore do not demonstrate that lowering the starting dose preserves survival, nor do they show that it worsens survival.

Dose modification and proteinuria measures differed between the groups. Dose reduction was reported in 45.5% of patients in the 5 mg group and 20.0% of those in the reduced-dose group, a difference associated with a p-value of 0.0406. The median worst post-treatment urine protein-to-creatinine ratio was also higher in the standard-dose group: 0.91 compared with 0.27 in the reduced-dose group, with a p-value of 0.0374.

These are observed associations from a nonrandomized cohort and do not establish that the initial dose caused the differences. Median relative dose intensity during the first eight weeks was 66.67% in the 5 mg group and 53.81% in the reduced-dose group, with a p-value of 0.0246. Taken together, the findings provide real-world evidence that reduced initial doses of fruquintinib were used in selected patients without a statistically detectable survival association in this cohort.

They also show that a lower prescription did not necessarily translate into greater overall early treatment exposure. However, the study cannot determine whether individualized starting doses are safe, equivalent or preferable across broader populations. Its scope was limited to 69 patients receiving third-line or later therapy, and poorer performance status among many reduced-dose patients may have introduced confounding.

The wide confidence intervals, retrospective design and baseline imbalance mean that clinically meaningful survival differences cannot be ruled out.

Subject of Research: Association between initial fruquintinib dose, survival and tolerability in refractory metastatic colorectal cancer

Article Title: Initial dose of fruquintinib and clinical outcomes in patients with refractory metastatic colorectal cancer: a multicenter real-world study

Article References: Hirakawa, M., Okagawa, Y., Kawamura, K., Yokoyama, R., Suzuki, N., Omori, G., Ameda, S., Oiwa, S., Ito, R., Usami, M., Sato, M., Doi, A., Kawano, Y., Sagawa, T., Fujikawa, K., Tsuji, Y., & Takada, K. (2026). Initial dose of fruquintinib and clinical outcomes in patients with refractory metastatic colorectal cancer: a multicenter real-world study. BMC Cancer. https://doi.org/10.1186/s12885-026-17085-w

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17085-w

Keywords: fruquintinib, metastatic colorectal cancer, dose optimization, real-world evidence, progression-free survival, overall survival, proteinuria, treatment tolerability, retrospective cohort, Initial, dose, clinical