Although the safety of combination chemotherapy without primary tumor resection (PTR) in patients with stage IV colon cancer has been established, questions remain regarding a potential survival benefit with PTR. The objective of this study was to compare mortality rates in patients who had colon cancer with unresectable metastases who did and did not undergo PTR. An observational cohort study was conducted among patients with unresectable metastatic colon cancer identified from the National Cancer Data Base (2003-2005). Multivariate Cox regression analyses with and without propensity score weighting (PSW) were performed to compare survival outcomes. Instrumental variable analysis, using the annual hospital-level PTR rate as the instrument, was used to account for treatment selection bias. To account for survivor treatment bias, in situations in which patients might die soon after diagnosis from different reasons, a landmark method was used. In the total cohort, 8641 of 15,154 patients (57%) underwent PTR, and 73.8% of those procedures (4972 of 6735) were at landmark. PTR was associated with a significant reduction in mortality using Cox regression (hazard ratio [HR], 0.45; 95% confidence interval [CI], 0.44-0.47) or PSW (HR, 0.46; 95% CI, 0. 44-0.49). However, instrumental variable analysis revealed a much smaller effect (relative mortality rate, 0.91; 95% CI, 0.87-0.96). Although a smaller benefit was observed with the landmark method using Cox regression (HR, 0.6; 95% CI, 0.55-0.64) and PSW (HR, 0.59; 95% CI, 0.54-0.64), instrumental variable analysis revealed no survival benefit (relative mortality rate, 0.97; 95% CI, 0.87-1.06). Among patients with unresectable metastatic colon cancer, after adjustment for confounder effects, PTR was not associated with improved survival compared with systemic chemotherapy; therefore, routine noncurative PTR is not recommended. Cancer 2017;123:1124-1133. © 2016 American Cancer Society.
Read full abstract