From September 1980 to March 1983, 30 cases were registered in a Southwest Oncology Group Study. Twenty-four cases were evaluable and constitute the basis for this report. Patients were diagnosed with adenocarcinoma or large cell lung carcinoma. Tumors were considered inoperable but able to be encompassed in a single radiotherapy (RT) port. Seventy-two percent of measured tumors were 4 cm in diameter or bigger (range 2 cm to 10 cm). RT was given initially to the primary lung tumor and ipsilateral hilar, mediastinal, and supraclavicular nodes, at 2 Gy per day; total dose was 44 Gy. The areas involved by tumor were boosted with 10 Gy more. Prophylactic cranial irradiation (PCI) was started at the same time with 15 treatments of 2.75 Gy. A 2-week rest period was instituted after the first 11 treatments. Chemotherapy (CT) was given from day 1 which consisted of 5-Flourouracil, 500 mg./M2, (bolus day 1 and 8) Vincristine, 1 mg./M2, and Mitomycin C, 5 mg./M2 both given on day 1. Cycles were repeated at 28 day intervals for 3 cycles and at 6 week intervals for 5 more cycles, or until progression, with persistent disease. Eight cases (33%) achieved complete response (CR), and 5 (21%) partial response (PR). Overall median survival was 37 weeks and 2 years survival was 8%. CR patients had the best chance for long-term survival. Relapses were evenly distributed between extra and intrathoracic sites, with the latter even between the inside and outside the RT field. No patient died with clinical evidence of metastasis to the brain (MB), although one was found to have MB at autopsy. Toxicity was severe in 7 cases (29%) and 2 deaths are considered toxicity related. When comparing these results to those from the literature, we found this protocol has achieved a slightly higher CR rate than what is expected with RT alone, without survival improvement. As CR patients have the best prognosis, simultaneous CT-RT might offer some promise, but at the expense of increased toxicity. PCI was effective in preventing or delaying MB, and thus deserves further investigation. We should caution that the study of possible long-term effects of PCI could not be assessed because of the short median survival of the patients. It is possible that a less aggressive time-dose fractionation to the brain might be as effective as the one used in this protocol.