Abstract
BACKGROUND: The effectiveness of third - line treatment for metastatic colorectal cancer (mCRC) still needs enhancing. In real-world clinical practice, third-line treatment options are complex and varied. However, real-world data on third-line treatment for mCRC remains limited. Further investigation is needed into treatment patterns, the efficacy of different regimens, and their influencing factors. METHODS: This study retrospectively analyzed 229 mCRC patients receiving third-line therapy (2013-2023). Kaplan-Meier method was used to draw the overall survival (OS) and progression-free survival (PFS) curves. With the chemotherapy group serving as the control group, the efficacy differences between third-line regimens were compared using the Log rank test. Univariate analysis was first conducted to evaluate prognostic factors, the indicators with statistical significance were analyzed by multivariate analysis using the Cox proportional hazards model. RESULTS: Fruquintinib plus immunotherapy achieved a median PFS of 8.8 months (95% CI: 8.6-9.0), which was significantly longer than the 3.6 months (95% CI: 3.1-4.1) in the chemotherapy group (p = 0.034). All 229 patients had proficient mismatch repair(pMMR) tumors or genetic testing suggestive of MSI-L/microsatellite stability(MSS). The third-line median PFS in the setting was 3.8 months, and the median OS was 13.3 months. Univariate analysis identified metastasis local treatment, RAS/BRAF mutation status, metastasis-to-PD2(defined as the second progression following first-line treatment initiation) interval, baseline CEA, and baseline CA199 as significant prognostic factors. Multivariate analysis confirmed metastasis-to-PD2 interval and baseline CA199 as independent prognostic indicators. CONCLUSION: In the real-world setting, fruquintinib combined with immunotherapy was associated with benefits in third-line treatment of patients with pMMR mCRC. Shorter metastasis-to-PD2 interval and elevated baseline CA19-9 levels at the start of third-line therapy were independent poor prognostic factors.