A controlled study of postoperative radiotherapy for patients with completely resected nonsmall cell lung carcinoma

BACKGROUND. Postoperative radiotherapy is commonly used to treat patients w ith completely resected nonsmall cell lung carcinoma, but its effect on ove rall survival has not been established. METHODS. After undergoing complete surgical resection, 728 patients with no nsmall cell lung carcinoma (221 Stage I, 180 Stage II, and 327 Stage III) w ere randomized to receive either postoperative radiotherapy at a total dose of 60 gray or observation only. The main end point was overall survival. RESULTS. At the reference date, 218 of 355 patients in the control group ha d died and 262 of 373 in the radiotherapy group had died. Five-year overall survival was 43% for the control group and 30% for the radiotherapy group (P = 0.002, log rank test; relative risk [RR]: 1.33; 95% confidence interva l [CI]: 1.11-1.59). This result was not modified by adjustment for potentia l prognostic factors. The excess mortality rate for the radiotherapy group was due to an excess of intercurrent deaths (P = 0.0001; RR: 3.47; the 5-ye ar intercurrent death rate was 8% for the control group and 31% for the rad iotherapy group). Radiotherapy had no significant effect on local recurrenc e (RR: 0.85; 95% CI: 0.64-1.14) and no effect on metastasis (RR: 1.06; 95% CI: 0.85-1.31). The rate of non-cancer-related death increased with the dos e per fraction delivered. CONCLUSIONS. Postoperative radiotherapy increased the rate of incidence of intercurrent deaths and reduced overall survival for patients with complete ly resected nonsmall cell lung carcinoma. [See editorial counterpoint on pa ges 195-6, this issue.] Cancer 1999;86:265-73. (C) 1999 American Cancer Soc iety.