Factors influencing survival after resection of pulmonary metastases from colorectal cancer

Background: Selection of patients for resection of lung metastases from colorectal cancer is problematic. The aim of this study was to evaluate clinically relevant prognostic factors and to define a subgroup of patients who would most benefit from such surgery.

Patients: Seventy-five patients (median age 58 (range 33-82) years) with pulmonary metastases from colorectal cancer underwent 104 R0 lung resections. Median follow-up was 33 (range 4-116) months. Patients who had no evidence of recurrent extrathoracic disease, no more than three metastases on either side, lobectomy as the maximal surgical procedure, and adequate cardiorespiratory function were eligible for surgery. Univariate and multivariate Cox regression, and classification and regression tree subgroup analyses were performed.

Results: Overall median survival was 33 months, with 3- and 5-year survival rates of 47 and 27 per cent respectively. Size of metastases (relative risk (RR) 2.6) and extent of resection (RR 0.4) were identified as independent prognostic factors. Primary tumour stage was significant in univariate analysis. Subgroup analysis defined two statistically relevant prognostic groups: patients with a maximum metastasis size of 3.75 cm or less with a disease-free interval of more than 10 months and patients with larger metastases and a shorter disease-free interval. Median survival and 5-year survival were 45 months and 39 per cent in the former group, and 24 months and less than 11 per cent in the latter.

Conclusion: Subgroup analysis provided criteria for the selection of patients for R0 resection of lung metastases from colorectal cancer and differentiated between those at high or low risk of early tumour progression; the latter patients would benefit most from surgery.