Conditional Survival Analysis of Metastatic Colorectal Cancer Patients Living ≥24 Months: A Single Institutional Study.

OBJECTIVES: The survival of patients with metastatic colorectal cancer (CRC) has been increasing over recent decades due to improvements in chemotherapy and surgery. There is a need to refine prognostic information to more accurately predict survival as patients survive for any given length of time to assist multidisciplinary cancer management teams in treatment decisions. MATERIALS AND METHODS: We performed a single center retrospective analysis of patients treated with metastatic CRC (unresectable and resectable) who survived >24 months between 2005 and 2015 (N=155). Patient tumor and treatment related variables were collected. Overall survival (OS) estimates conditional on surviving >24 months were compared with actuarial survival estimates of a cohort of patients (33,104 resected, 39,382 unresected) from the National Cancer Database (NCDB). RESULTS: With a median follow-up of 44.2 months, the median OS of resected patients (n=86) was not reached. The median OS of unresected patients was 75.9 months. The conditional survival probabilities of living 1, 2, or 3 years longer after 24 months of survival are 92%, 72%, and 52%, respectively, in unresectable patients and 98%, 92%, and 89% in patients who were resected. The corresponding NCDB 1, 2, and 3 year actuarial survival was 38%, 20%, and 11% for unresected patients and 68%, 46%, and 32% for resected. CONCLUSIONS: These results indicate that CRC patients who survive 24 months with metastatic colorectal cancer have an excellent prognosis and surgery may be appropriate in a subset of patients initially deemed unresectable.