In a third study, a team led by Dr. Zeinab Alawadi, a surgeon and postdoctoral fellow at the University of Texas MD Anderson Cancer Center, Houston, analyzed data from 14,399 patients in the National Cancer Data Base. They had been diagnosed with stage IV colon cancer between 2003 and 2005. The researchers excluded patients who had nonelective resection or surgery at other sites, such as metastasectomy.
The primary tumor was resected in 55% of all patients studied and in 74% of patients included in a 1-year landmark analysis done to account for early deaths related to comorbidity or disease burden, reported Dr. Alawadi.
In the entire cohort, primary resection conferred a significant survival benefit after standard multivariate adjustment (hazard ratio, 0.39) that persisted after propensity score weighting to account for treatment selection bias (hazard ratio, 0.41). The benefit was also significant, but much attenuated, in an instrumental variable analysis, another method for accounting for treatment selection bias (relative mortality rate, 0.88).
In the 1-year landmark population, primary resection conferred a smaller significant survival benefit after standard multivariate adjustment (hazard ratio, 0.60) that persisted after propensity score weighting (hazard ratio, 0.59). But there was no longer a significant benefit in the instrumental variable analysis here.
“Among the entire cohort of patients with stage 4 colon cancer, primary tumor resection offered no survival benefit over systemic chemotherapy alone when the [instrumental variable] method was applied at the 1 year landmark,” the investigators write.
“Subject to selection and survivor treatment bias, standard regression analysis may overestimate the benefit of [primary tumor resection],” they concluded.