[en] An analysis of intrathoracic tumor control was carried out in 378 patients with histologically proven unresectable non-oat cell carcinoma of the lung treated with definitive radiotherapy, randomized to one of four treatment regimens: 4000 rad split course (2000 rad in five fractions in one week, two weeks rest and additional 2000 rad in five fractions in one week) or 4000, 5000 or 6000 rad continuous courses, five fractions per week. Between 85 and 101 patients are analyzed in each treatment group. The complete plus partial response was 46-51% in the 4000 rad groups in contrast to 61-66% in the 5000 to 6000 rad groups (P = 0.008). The overall two year survival rate was 10-11% for the patients treated with 4000 rad split or continuous course, and 19% in the patients treated with 5000 to 6000 rad. The complete response in patients with tumors 3 cm or less in diameter was 16% when treated with 4000 rad in contrast to 20-31% in those treated with 5000-6000 rad. In the patients with lesions from 4 to 6 cm in diameter, complete and partial tumor regression was 48% in the 4000 rad group, 67% with 5000 rad, and 71% with 6000 rad. Strict quality assurance criteria in radiotherapy are necessary to achieve optimal treatment results and a careful program to evaluate techniques of irradiation and protocol compliance should be maintained in cooperative group studies in order to enhance the validity of clinical trials