In our previous study, FCCC 93-024, paclitaxel by 24-h infusion combined with carboplatin yielded a response rate of 62% and median survival of 54 weeks in advanced non-small cell lung cancer (NSCLC). Myelosuppression proved dose-limiting, requiring the routine use of granulocyte-colony stimulating factor (G-CSF). Based on the reported activity of 1-h paclitaxel infusion in NSCLC and minimal myelosuppression at doses of 135 and 200 mg/m2 every 3 weeks and the suggestion of a dose-response relationship, we launched an intrapatient dose escalation trial of combination carboplatin and 1-h paclitaxel. Chemotherapy-naïve patients with advanced NSCLC received paclitaxel 175 mg/m2 1-h and carboplatin dosed to a fixed targeted area under the concentration-time curve (AUC) of 7.5 at three weekly intervals for six cycles. In the absence of grade 4 myelosuppression, paclitaxel was escalated by 35 mg/m2/cycle on an intrapatient basis to a maximum dose of 280 mg/m2 by cycle 4. G-CSF was not routinely used. 57 patients (pts) were accrued from November 1994 through to April 1996. 44 pts (77%) had Eastern Cooperative Oncology Group (ECOG) performance status 1. Median age was 64 (range: 34-80) years. Cumulative peripheral sensory neuropathy proved dose-limiting and prohibitive in the first 20 evaluable patients (cohort A): grade > or = 1 in 15 patients (75%), grade 3 in 6 (30%), generally occurring at paclitaxel doses > or = 215 mg/m2 and obligating 3 patients to have treatment halted in the absence of disease progression. The protocol, therefore, was revised and the initial paclitaxel dose reduced to 135 mg/m2 with intrapatient dose escalation of 40 mg/m2/cycle to a maximum dose of 215 mg/m2, recapitulating the original dosing schema used in FCCC 93-024. 35 patients were enrolled in this second cohort (B); 33 proved evaluable. Whilst 17 (52%) experienced peripheral sensory neuropathy, grade 3 neurotoxicity developed in only 3 (9%). Myelosuppression also was less pronounced, with 42% exhibiting grade 4 granulocytopenia and 30% grade > or = 3 thrombocytopenia in cohort B compared with 70% and 50%, respectively in cohort A. Of the first 22 patients accrued to cohort A, 12 (55%) had major objective responses. Median survival was 48.5 weeks, 1-year survival rate 45% and 2-year survival rate 18%. Of 33 evaluable patients in cohort B, 9 (27%) had major objective responses. Median survival was 46 weeks, 1-year survival rate 47% and 2-year survival rate 12%. Combination paclitaxel by 1-h infusion and carboplatin at a fixed targeted AUC of 7.5 is active in advanced NSCLC. Neurotoxicity, not myelosuppression, proved dose-limiting at paclitaxel doses exceeding 215 mg/m2. Lower doses may be associated with lower response rates, but do not appear to compromise survival.