The Goldie-Coldman model explaining the kinetics of tumor cell kill and drug resistance has a potential application in designing chemotherapy regimes. In this Southwest Oncology Group (SWOG) trial we tested the alternation of two potentially noncrossresistant drug combinations with a concurrent drug combination in patients with limited small-cell lung cancer. The concurrent drug combination consisted of etoposide (VP-16), 75 mg/m2/intravenously (IV), days 1, 2, and 3; vincristine, 1.0 mg/m2/IV, days 1 and 8; Adriamycin (doxorubicin; Adria Laboratories, Columbus, OH), 40 mg/m2/IV, day 1; and cyclophosphamide, 750 mg/m2/IV, day 1 (EVAC). The alternating combination consisted of VP-16, 100 mg/m2/IV, days 1, 2, and 3; and cisplatin (CDDP), 100 mg/m2/IV, day 1, alternating with vincristine, 1.0 mg/m2/IV, days 1 and 8; Adriamycin, 50 mg/m2/IV, day 1; and cyclophosphamide, 750 mg/m2/IV, day 1 (VP-16/CDDP-VAC). Chemotherapy was administered at 3-week intervals for six cycles both before and after chest (5,000 rads/5 weeks) and whole brain radiotherapy (3,000 rads/2 weeks). One hundred ninety-nine patients received EVAC and 201 received the alternating combination. There was no significant difference in the response rate to the initial six cycles of treatment with EVAC (CR, 40%) versus the alternating combination (CR, 38%). There was no significant difference between the best response, EVAC (CR, 48%) and VP-16/CDDP-VAC (CR, 51%). Median survival for all randomized patients on EVAC is 15.1 months versus 16.5 months on the alternating combination (P = .58). Toxicities consisted primarily of bone marrow suppression, anorexia, nausea and vomiting, peripheral neuropathies, and alopecia. As in previous trials, the chest was the most common site of relapse (33%). There were no differences in the incidence and sites of relapse between the two treatment arms. These treatments appear equally effective at inducing remission and prolonging survival in patients with small-cell lung cancer.