Conditional survival following first-line treatment of MSS metastatic colorectal cancer: A pooled analysis of 12 randomized controlled trials.

Abstract

e15565 Background: Metastatic colorectal cancer (mCRC) has a poor prognosis and is a major burden to patients and healthcare systems. In the last decades, the therapeutic landscape of mCRC has evolved, improving overall survival (OS). Conditional survival estimates prognosis among patients who already survived a defined period, which can guide individual risk assessment and treatment decisions in the course of disease. Analysis of conditional OS and progression-free survival (PFS) in a large cohort of mCRC patients receiving first-line systemic treatment is lacking. Methods: We analyzed data of mCRC patients from the safety analysis set of 12 randomized controlled trials. Patients with incomplete data, MSI-H patients and patients who received immune checkpoint inhibitors in first line treatment were excluded. OS was calculated from time of randomization or registration until death. PFS was calculated from time of randomization or registration until progression or death, whichever occurred first. Conditional survival was estimated among patients alive and/or progression-free at predefined landmarks (12, 24, and 36 months). Survival was assessed using the Kaplan-Meier method and Cox proportional hazard modeling. The Benjamini-Hochberg method was used to adjust for multiple testing. Results: 3,507 patients from the initial cohort met the inclusion criteria and were analyzed. Mean age was 61.5 years and 64.1% were male. Median follow-up was 50.7 months. Details of conditional survival estimates of patients in our cohort are summarized in table 1. Interestingly, the “remaining” OS ranged from 20-27 months, irrespective of the landmark analyzed. In patients with PFS > 24 months, only history of local or surgical treatment of metastases in the first 24 months (HR 0.31, p < 0.001) and only one organ initially involved with metastases (HR 0.54, p = 0.022) were significantly associated with ongoing conditional OS. Conclusions: Conditional OS of mCRC patients alive at predefined OS landmarks remained stable with increasing OS time. Furthermore, conditional OS and PFS improved substantially in patients with PFS > 24 months after first-line treatment. In this small, but significant fraction of patients, initial local or surgical treatment might be the key factor of long survival. OS/PFS landmark Proportion surviving (OS) [%] Proportion surviving (PFS) [%] median cOS (OS-conditioned) median cOS (PFS-conditioned) median cPFS (PFS-conditioned) 0 100.0 100.0 27.29(26.16 - 28.03) 27.29(26.16 - 28.03) 10.53(10.26 - 10.81) 12 83.2 42.4 19.68(18.77 - 21.16) 32.29(30.6 - 34.97) 6.11(5.65 - 6.71) 24 55.6 13.4 19.65(18.29 - 21.29) 81.38(47.1 - NE) 18.97(15.35 - 25.53) 36 36.2 7.9 20.97(19.19 - 23.93) 69.38(NE - NE) 32.06(31.3 - NE) cOS, conditional OS; cPFS, conditional PFS; NE, not estimable. Time is reported in months. 95% confidence intervals are indicated in brackets.