Chemo Doublets Should Be Standard Treatment for Advanced NSCLC

ASCO—A new study from the Cancer and Leukemia Group B (CALGB 9730) shows that chemotherapy doublets should be the standard treatment for advanced non-small-cell lung cancer (NSCLC), Rogerio C. Lilenbaum, MD, reported at the 38th Annual Meeting of the American Society of Clinical Oncology (abstract 2).

"Not only did the results show an increase in response rate but, more importantly, an increase in survival time in patients who received carboplatin [Paraplatin] and paclitaxel [Taxol], compared with paclitaxel alone," said Dr. Lilenbaum, director of the Thoracic Oncology Program, Mount Sinai Comprehensive Cancer Center, Miami Beach.

The phase III trial randomized 584 patients (561 evaluable) with stage IIIB/IV NSCLC to receive paclitaxel 225 mg/m² given over 3 hours on day 1 with or without carboplatin to AUC 6 every 3 weeks for up to six cycles. The patients had received no prior chemotherapy.

The median patient age was 63.5 years; 155 (27%) were 70 years of age or older, and 99 (18%) were performance status (PS) 2 on the ECOG scale. Quality of life was assessed at baseline, 2, 6, 9, and 12 months, using the EORTC QLQ-C30 and QLQ-LC13 questionnaires.

The objective response rate was 29% in patients on combination therapy vs 17% in those on paclitaxel alone (P < .0001), Dr. Lilenbaum said.

With a median follow-up of 19.7 months, median overall survival was 8.8 months with the combination vs 6.7 months with single-agent therapy, a 24% relative increase in survival for those receiving carboplatin/paclitaxel. Failure-free survival was also significantly longer with combination therapy—4.6 months vs 2.5 months for single-agent therapy.

One-year survival, at 37% for the combination vs 33% for the single agent, was not significantly different. This may be due, in part, to a higher use of second-line therapy in the single-agent arm (39% vs 32%) and of platinum-based combinations in particular (14% vs 7%), Dr. Lilenbaum said.

Failure-free survival was significant in favor of the combination arm by both log-rank and Wilcoxon analyses. The overall survival curves, he noted, split at the beginning and converged at about 12 months. When analyzed by log-rank, according to the original study design, the overall survival difference between the two arms was not significant.

However, he said, "the proportional hazards assumption that underlies the use of the log-rank method assumes that treatment-related hazards, or deaths in our case, are independent of time, an assumption that clearly was not satisfied in our study as shown by the curves."

Therefore, Dr. Lilenbaum said, the CALGB statisticians felt justified in using the Wilcoxon method (P = .0125) in addition to log-rank (P = .2022) to analyze overall survival.

Toxicity Levels

Although toxicity levels were slightly higher with combination therapy, the patients’ quality of life was not compromised, Dr. Lilenbaum said, as quality of life scores did not differ significantly between the two groups in any of the 25 quality of life parameters tested. In addition, there was no significant difference between the two arms in quality of life in the elderly patients (age 70 and over) or in the PS 2 patients .

Grade 3-4 neutropenia was significantly more common in the combination group (62% vs 32%) but febrile neutropenia rates were similar (8% with the combination vs 6% with paclitaxel alone). "There was only one toxic death per arm in the study," he said.

An unexpected finding, Dr. Lilenbaum said, was that resource utilization, a cost-effectiveness surrogate, was comparable in both arms. Resource utilization refers to the costs associated with treatment, including length of hospitalization and use of supplemental therapies, but not the actual costs of the chemotherapy agents.

"This is one of the most comprehensive prospective analyses of resource utilization in oncology," he said, adding that collection of cost data and cost-effectiveness analyses are currently underway.

Subset Analyses

Patients age 70 and over who received combination therapy experienced a survival benefit similar to those under age 70 (median overall survival, 8 months vs 5.8 months for single-agent therapy).

The survival curves did not reach statistical significance, "most likely due the small number of patients in the subset and the lack of statistical power at this level," Dr. Lilenbaum commented.

Although PS 0-1 patients had significantly longer median survival than PS 2 patients (8.8 months vs 3.0 months) with either therapy, when analyzed by treatment arm, median survival was significantly longer for the PS 2 patients with combination therapy (4.7 months vs 2.4 months with single-agent therapy).

"In this subset, despite small numbers, difference in survival was significant in favor of the combination both by log-rank and Wilcoxon analysis," he said. In addition, the PS 2 patients tolerated combination therapy well, he said, noting that there were no toxic deaths in this subset.

Conclusions

Dr. Lilenbaum concluded that combination therapy produces a higher response rate in patients with advanced NSCLC than single-agent therapy, and significantly improves failure-free and overall survival. It is associated with greater toxicity, mainly hematologic, "but the clinical implications of these toxicities were mild," he said.

He recommended that patients with advanced NSCLC and good performance status should be treated with a platinum-based doublet. "It is the template upon which new treatment strategies should be developed," he said. Further, he said, elderly patients can be treated safely with a carboplatin doublet and should be offered such therapy.

As for PS 2 patients, he said, "the benefit of chemotherapy remains to be conclusively demonstrated, but selected patients who choose active therapy may be offered combination therapy."