Paclitaxel and carboplatin: a phase I study in advanced non-small cell lung cancer.

In the treatment of non-small cell lung cancer paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ) has significant activity and carboplatin has single-agent activity comparable with that of cisplatin, with less pronounced nonhematologic toxicity. The optimal doses of paclitaxel and carboplatin in combination have not been determined. We designed a phase I study combining a fixed paclitaxel dose of 175 mg/m2, administered either by 3- or 1-hour infusion, with escalating doses of carboplatin given every 4 weeks. The starting carboplatin dose was 175 mg/m2, with planned dose increases in increments of 25 mg/m2. The primary study objective was to find the maximum tolerated dose of the combination. Secondary objectives were to determine the toxicity profile, response rate, and feasibility of a 1-hour paclitaxel infusion with steroid premedication delivered only 1 hour before the paclitaxel infusion. Eligibility criteria included age 18 to 75 years, no prior chemotherapy, stage IIIB to IV disease, Eastern Cooperative Oncology Group performance status 0 to 2, no second tumors, measurable or evaluable disease, and informed consent. We achieved a carboplatin dose level of 300 mg/m2 without reaching the maximum tolerated dose. The dose-limiting toxicity was granulocytopenia. However, only one patient had a neutrophil count less than 500/microL during the first cycle. Other toxicities during the first and remaining 73 delivered cycles were mild to moderate. Only one patient had treatment delayed, and no dose reductions were necessary. Of 22 patients entered, 19 were evaluable for response (two were not evaluable and one was too early to evaluate). Six partial responses (31%; 95% confidence interval, 13% to 57%), five (26%) stable diseases, and eight (42%) disease progressions were observed. No additional side effects were observed with the 1-hour paclitaxel infusion and single-dose steroid premedication 1 hour before chemotherapy. The study will continue until the paclitaxel/carboplatin maximum tolerated dose is reached.

Mitomycin C, vinblastine, and carboplatin regimen in patients with nonsmall cell lung cancer. A phase II trial.

BACKGROUND: The mitomycin C, vinblastine, and cisplatin (MVP) combination is one of the most frequently used in the palliative setting, but it produces considerable toxicity. Carboplatin and cisplatin have different patterns of toxicity. The goal of this study was to evaluate a combination similar to MVP, using carboplatin instead of cisplatin to render it more feasible in an outpatient setting. METHODS: Inclusion criteria for this study included: inoperable patients or patients relapsing after previous surgery, with nonsmall cell lung carcinoma (NSCLC), a performance status (PS) > 50%, and no previous chemotherapy. The chemotherapy regimen included carboplatin, 300 mg/m2 on Day 1; mitomycin, 8 mg/m2 on Day 1; and vinblastine, 4 mg/m2 on Days 1, 8, and 15 (on Day 15 vinblastine was delivered only in the first cycle) (MVC) every 3 weeks for at least 3 cycles. RESULTS: From August 1991 until August 1994, 70 patients entered the trial. All were evaluable for toxicity and response. The median age was 62 years (range, 40-73 years). The male/female ratio was 60:10 (86%:14%); the ratio of Stage III to Stage IV disease was 26:44 (37%:63%); and the ratio of PS > 70 to < or = 70 was 49:21. A total of 296 cycles (median, 4 [range, 1-6 cycles] per patient) were delivered, 280 of 296 (95%) in an outpatient setting with only 4 patients requiring hospitalization for treatment delivery. Overall response rate (RR) was 38.6% (95% confidence interval [CI], 27-51%) (1 complete response, 1.5%; 26 partial responses, 37.1%). Median duration of response was 9.8 months (range, 2-27 months). In Stage III patients the RR was 42% and in Stage IV patients it was 34%. Overall median survival was 9.5 months (95% CI, 6.8-15.3 months). Survival at 1 year was 39% (standard error [SE] 3.6%) and was 11% at 2 years (SE 3.6%). In Stage III patients median survival was 13 months and the 1-year survival rate was 54% (SE 10%); Stage IV patients had a median survival of 7.4 months and a 1-year survival rate of 28% (SE 7%). Delivered dose intensity was: carboplatin, 71%; vinblastine, 60%; and mitomycin C, 77% of the planned dose intensity. The back calculation of carboplatin area under the curve (AUC) with Calvert's formula and with the Cockcroft-Gault glomerular filtration rate estimation, showed a median AUC value of 4 (range, 2-8). Using the more precise Chatelut formula, AUC was again 4 (range, 2-7). Hematologic toxicity was the major side effect; Grades 3 and 4 leukopenia were observed in 34% and 6% of patients, respectively, and Grades 3 and 4 thrombocytopenia in 25% and 4% of patients, respectively. Grade 2 infection occurred in 10% of patients, with only 1 case of sepsis; severe constipation and Grade 2 alopecia occurred in only 1 patient; and no case of higher than Grade 1 nephrotoxicity was observed. No pulmonary toxicity was observed. Compliance with treatment was good with only one patient refusal after the first cycle. CONCLUSIONS: Chemotherapy for advanced NSCLS is still controversial, because effectiveness in terms of RR and symptom control must be weighed against treatment toxicity and costs. From our study it appears that MVC is easy to deliver in an outpatient setting, and has good patient compliance, low toxicity profile, and promising RR and response duration. The substitution of carboplatin for cisplatin in regimens for advanced NSCLC should be considered.