Studies conducted by the Spanish Lung Cancer Group indicate that cisplatin- or carboplatin-based chemotherapy can yield a 25% response rate, 9-month median survival time, and 30% 1-year survival rate in patients with stage III and IV non-small cell lung cancer. Phase II trials of single-agent paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ) have an almost 30% response rate in non-small cell lung cancer. Based on these results, we decided to examine whether the sequence-dependent effects of paclitaxel/etoposide influence treatment outcome (antitumor response) and toxicity. In vitro data show a paradoxical antagonist rather than additive effect. In the first part of our study (part A), paclitaxel and etoposide were administered at the same time. In the second part (part B), etoposide preceded paclitaxel. In both parts, patients with previously untreated stage IIIB or IV non-small cell lung cancer with good performance status were eligible. In part A, etoposide (fixed dose, 100 mg/m2) on days 1, 2, and 3 was administered by 30-minute infusion; paclitaxel (175 mg/m2) was given by a 3-hour infusion on day 1. In part B, the etoposide dose and schedule were the same, but paclitaxel (same dose) was administered on day 4. Treatment in both parts was repeated every 21 days for a maximum of 10 cycles. In part A, 18 patients were entered and no objective responses were observed. In part B, 21 patients were accrued, 17 of whom had sufficient follow-up for response assessment. Seven objective responses were achieved (two complete and five partial responses, for an objective response rate of 41%). Seven patients had no change and three had progressive disease. Frequency and severity of side effects were not significantly different in either part of the study. However, grade 4 neutropenia was observed in 10 (59%) patients and one (5%) patient in parts A and B of the trial, respectively. Nonhematologic toxicity was slight. In conclusion, paclitaxel cytotoxicity is abrogated when it is given concurrently with etoposide. When etoposide precedes paclitaxel, a more effective paclitaxel/etoposide schedule is attained.