Between 1968 and 1974, 348 patients with lung cancer were primarily treated with radiation therapy. There were 66 such patients (19%) who survived a minimum of 18 months and are the subject of this report. Of this group, 30 patients have no evidence of disease from 18-96 months, with a median follow-up of 38 months. Thirty-three patients are dead of disease. The five-year actuarial survival of the total group of 348 patients was 5.6%. There were 14 stage I and II patients who survived a minimum of 18 months, of whom 11 had no evidence of disease. Of the 42 Stage III patients, 18 presently show no evidence of disease. There were 13 patients who failed with locally recurrent disease; in this group a dose-response relationship was demonstrated. A local failure rate of 50% (4/8) was observed for patients who received fewer than 5000 rad, 22% (6/27) for patients receiving 5000-5500 rad, 18% (2/11) in patients receiving 5500-5900 rad, and 5% (1/20) for patients who received more than 5900 rad. Radiotherapeutic technique was a significant variable in local failure. Forty-six percent (6/13) of those patient failures may have been eliminated with the use of careful treatment planning with simulation. A statistically significant difference in dose was noted for patients with central recurrence, mean dose 4725 rad, 1561 RET, 81 TDF, when compared with patients with control of gross disease with radiation, mean dose 5740, 1880 RET, 108 TDF. There were four patients with marked early complications (6%) and eight patients with late complications (12%). There were no deaths attributable to radiation. Although most patients with advanced lung carcinoma die of distant disease, a significant number of patients can achieve long-term survival when radically treated with high-dose radiation therapy.