Background: There had been no phase III trial to compare the efficacy between trifluridine/tipiracil (TAS-102) plus bevacizumab (BEV) with TAS-102 monotherapy in refractory metastatic colorectal cancer (mCRC). Our study sought to investigate the efficacy and safety of TAS-102 plus BEV versus TAS-102 monotherapy in these patients. Methods: We retrospectively analyzed the patients with refractory mCRC who received TAS-102 (35 mg/m2 twice daily; day 1–5 and 8–12; every 28 days) with or without BEV (5 mg/kg; day 1,15; every 28 days) between November 2020 and October 2022 in Hunan Cancer Hospital. All patients had previously received at least second-line treatment. The primary endpoint was overall survival (OS). Results: A total of 58 patients received TAS-102 plus BEV, and another 32 patients received TAS-102 monotherapy. With a median follow-up of 4.60 months (range, 0.20–22.80), the median overall survival (mOS) was 10.83 months [95% confidence interval (CI), 10.10–NA] for TAS-102 plus BEV, 7.43months (95% CI, 5.17–NA) for TAS-102 monotherapy. The hazard ratio (HR) was 0.90 (95% CI, 0.42–1.92), P= 0.78. Median progression-free survival (mPFS) was 4.67 months (95% CI, 3.57–7.37) for TAS-102 plus BEV, and 4.30 months (95% CI, 3.03–NA) for TAS-102 monotherapy. The HR was 1.04 (95% CI, 0.54–1.99), P= 0.91. The Cox multivariate regression analysis demonstrated that the patients who received treatment after TAS-102 with or without BEV(HR:0.25;95% CI, 0.09-0.71) and previously undergone Irinotecan (HR:0.23;95% CI, 0.06-0.95) were independent risk factors for OS and the patients who previously received Cetuximab was confirmed to have a significant better PFS(HR:0.17;95% CI, 0.03-0.91). Of the 70 patients who were evaluated, the TAS-102 plus BEV showed higher ORR and DCR than TAS-102 monotherapy, but the difference was not significant (ORR: 12.2% versus 0%; p= 0.16; and DCR: 77.5% versus 61.9%; p= 0.29). The incidence of platelet count decrease (grade ≥3) was significantly higher in the TAS-102 plus BEV group than the TAS-102 monotherapy group (3.00% vs. 0.00%, p = 0.03), whereas other adverse events were comparable between the two groups. Conclusions: Compared with TAS-102 monotherapy, TAS-102 plus BEV seemed to have comparable safety but improved disease control and prognosis in patients with refractory mCRC, although the difference was not significant. Further prospective studies are still needed to confirm our conclusions.