A randomized controlled study of immunotherapy with interleukin 2 (IL-2) and lymphokine-activated killer (LAK) cells was conducted in 105 patients after noncurative resection of primary lung cancer. Half the patients received only the standard postoperative radiation therapy or chemotherapy (control group). The other half received immunotherapy with IL-2 and LAK cells in addition to the standard therapy (immunotherapy group). The primary endpoint was survival. The 7-year survival rate was greater in the immunotherapy group than in the control group (39.1% vs. 12.7%, P < 0.01). Among patients with squamous cell carcinoma, there was no statistical difference in outcome. In contrast, the 7-year survival rate among patients with adenocarcinoma in the control group was only 5.2% but for those in the immunotherapy group it was 38.9% (P < 0.05). If resection was noncurative because of pulmonary metastasis, residual cancer or incomplete resection of lymph nodes, then immunotherapy was effective. If resection was noncurative because of residual cancer in the chest wall or diaphragm, or because of carcinomatous pleuritis or pleural dissemination, then there was no statistical difference in survival between the control group and the immunotherapy group.