A randomized phase III trial of four chemotherapy regimens in advanced non-small cell lung cancer (NSCLC)
Presenter: Joan H. Schulter
Since the initial randomized trials in the early 1990's which showed improved median and overall survival by adding chemotherapy to radiation therapy (RT) in NSCLC patients, several newer agents have been introduced. Some of these agents are used off protocol based on phase II data, known activity in other sites, and results of small phase III trials. This trial tests 3 "experimental" chemotherapy arms against a single "reference" arm.
- All arms contained either cisplatin (P), or carboplatin (Cb). Arm A, the reference arm, used P + paclitaxel (T). Arm B used Gemcitabine (G) and P. Arm C used docetaxel (Tx) and P. Arm D used T and Cb.
- Dosages were: arm A - P 75 mg/m2 on day 1, T 175 mg/m2/24 hours; arm B - G 1000 mg/m2 on days 1, 8, 15 and P 100 mg/m2 on day 1; arm C - Tx and P both 75 mg/m2 on day 1; arm D - T 225 mg/m2/3 hours on day 1 and Cb AUC of 6 on day 1.
- Arms A,C, and D repeated q21 days, arm D q28 days.
- Eligible patients were required to have measurable or evaluable disease, ECOG performance status (PS) of 0-2 (changed to 0-1 in 10/97 due to excess toxicity), stage IIIB (because of malignant pleural or pericardial effusion) or stage IV disease. Patients were stratified by brain metastasis, weight loss, and age.
- Patients were registered between 10/96 and 5/99.
- 1207 patients were registered on study, 1146 were eligible for analysis. Of the eligible patients, 1083 had performance status (PS) 0-1, and 968 were stage IV. 994 patients had died as of May 1, 2000.
- Median follow-up was 7.7 months. Median survival was 7.8 months.
- In general, the four arms were found to have comparable toxicities. Less fever and neutropenia was seen in arms B and C. Less nausea and fewer grade 4-5 toxicities were seen in arm D, but patients in arm D experienced worse peripheral neuropathy.
- Median survival in arms A, B, C, and D were 7.1 months, 8.1 months, 7.4 months, 8.2 months, respectively. One year actuarial survival for these arms was 31%, 36%, 31%, and 34%, respectively. Overall response was 21%, 21%, 17%, and 15%, respectively. None of these differences were statistically significant. An increased time to progression was seen in arm B. None of these data were available at the time the abstract was printed.
- Among these four treatment modalities, there was no difference in overall survival between arms. Treatment with the gemcytabine arm showed a statistically significant increase in time to progression.
- The chemotherapy regimens offer a modest but significant improvement over observation alone.
- The pursuit of more effective treatment for locally advanced NSCLC continues. Currently, cure rates are low and there is little hope in the near future for improvement. Although there have been increased response rates using newer drugs as single agents, combination chemotherapy does not appear to yield significant gains.