Primary Tumor Resection for Stage IV Colorectal Cancer in the Era of Targeted Chemotherapy.

BACKGROUND

In the current era of targeted therapies, the benefits of resection of primary tumors in patients with unresectable stage IV colorectal cancer, specifically with regard to overall survival, are unknown.

METHODS

Our study population comprised 208 consecutive patients with unresectable stage IV colorectal cancer who received chemotherapy containing at least one molecular target agent, bevacizumab, cetuximab, and panitumumab, at the National Cancer Center Hospital from 2006 to 2013. To lessen the effects of confounding factors between two treatment groups (resection versus non-resection) such as performance status, presence of severe symptoms, M subcategories (M1a versus M1b, M1c) according to the TNM classification, primary tumor site, and CEA value, we conducted three different propensity score analyses (regression adjustment, stratification, and matching).

RESULTS

Of the 208 patients, 108 (52%) underwent resection of the primary tumor, while 100 (48%) did not. Regression adjustment revealed that resection was not associated with longer overall survival (hazard ratio of 0.70 (95% CI [0.49-1.00]; p = 0.051)). Stratification analysis of five strata revealed inconsistent results (hazard ratios ranged from 0.50 to 1.58); specifically, resection was associated with longer overall survival in four strata, but with shorter survival in one stratum. The propensity score-matched cohort (64 matched pairs) yielded a hazard ratio of 0.76 (95% CI [0.51-1.15]; p = 0.197).

CONCLUSIONS

All three analyses revealed that, in the current era of chemotherapy with target agents, primary tumor resection was only marginally influential and did not significantly improve overall survival over chemotherapy alone.