Chemotherapy Doublet Produces Higher Response Rates and Modest Gain in Survival Over Single-Agent Therapy

MIAMI BEACH—Chemotherapy doublets outperform single-agent chemotherapy in advanced non-small-cell lung cancer (NSCLC), producing higher response rates and a modest gain in overall survival, according to results of a study from the Cancer and Leukemia Group B (CALGB), presented at the 38th Annual Meeting of the American Society of Clinical Oncology (ASCO).

"We believe combination chemotherapy or a chemotherapy doublet should be the standard of care for patients with advanced non-small-cell lung cancer," stated Rogerio C. Lilenbaum, MD, director of the Thoracic Oncology Program, Mount Sinai Comprehensive Cancer Center in Miami Beach (ASCO abstract 2).

Although "combination chemotherapy is associated with greater toxicity, mainly hematologic, the clinical implications of these toxicities were minor," Dr. Lilenbaum noted. Quality-of-life measurements did not differ between those patients receiving combination or single-agent chemotherapy. In addition, subset analysis revealed that elderly patients can be treated safely with combination chemotherapy.

"This study and others confirm that elderly patients should not only be offered treatment but offered the same treatment as others," commented Paul A. Bunn, Jr, MD, director of the University of Colorado Cancer Center in Denver and President-Elect of ASCO.

‘Is It Worth It?’

A systematic review of the literature found 25 trials conducted between 1974 and 1996 that compared single-agent to combination therapy in advanced NSCLC. The cumulative results showed a "twofold increase in response rate" for the combination chemotherapy, but it was "associated with a threefold increase in serious adverse effects," Dr. Lilenbaum reported. "We wanted to answer in a scientific manner a question that physicians and patients ask themselves all the time: ‘Is combination chemotherapy worth it?’"

The phase III trial initially included 584 patients with stage IIIB/IV NSCLC, although 23 patients (4%) were ineligible or never received treatment and were excluded from analysis. The median age was 63.5 years, and 26% of patients were over 70. Most were male (399). Eighteen percent had performance status (PS) 2. Quality of life was assessed at baseline and at 2, 6, 9, and 12 months.

Patients were randomized to receive paclitaxel at 225 mg/m² administered over 3 hours on day 1, with or without carboplatin (Paraplatin) to AUC 6 every 3 weeks, for up to 6 cycles. "We were able to demonstrate that a more aggressive regimen of two drugs in combination indeed prolonged survival in the disease, compared to only one drug," Dr. Lilenbaum stated. "This survival benefit extended to elderly patients and also to patients who are very ill at the time of presentation. These findings were somewhat unexpected."

Analyzing Responses

The objective response rate was 29% of patients receiving the combination therapy vs 17% for those receiving paclitaxel alone, "a difference that was highly statistically significant," according to Dr. Lilenbaum.

At a median follow-up of 19.7 months, median overall survival was 8.8 months for the combination chemotherapy arm vs 6.7 months with single-agent therapy. Failure-free survival figures were 4.6 vs 2.5 months. "The difference in failure-free survival was statistically significant in favor of combination chemotherapy, both long-range and proximate," Dr. Lilenbaum said.

The differences in overall survival, however, were not significant when analyzed by log-rank, as called for in the study design. The curves split at the beginning and then converge at approximately 12 months, producing 1-year survival rates of 37% for the combination arm vs 33% for the paclitaxel-only arm.

"The proportional hazards assumption that underlines the use of 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," Dr. Lilenbaum said. The CALGB statisticians therefore felt justified in using the Wilcoxon method (P = .0125) in addition to log-rank (P = .2022) to analyze overall survival. By the Wilcoxon method, the difference in survival data is significant, he noted.

Hematologic toxicities, including grade 3/4 neutropenia and thrombocytopenia, "were significantly more pronounced in the combination arm." Dr. Lilenbaum said (62% vs 32%). However the rates of febrile neutropenia were similar (8% vs 6%), and in each arm of the study there was only one toxic death.’’ Toxicity was manageable in both arms, without major clinical implications," he said.

Subset Analyses

Subset analyses revealed that patients over 70 fared as well as younger patients. Complete clinical response in the elderly subset was 36% in the combination arm and 21% in the single-agent arm.

Median survival rates were 8 vs 5.8 months, and 1-year survival was 31% vs 35%. The survival differences did not reach statistical significance, which Dr. Lilenbaum said was most likely a result of the "small number of patients and the lack of power of the study."

The outcome for patients with PS 2 was "significantly worse," Dr. Lilenbaum reported, with a median survival of 3 months for PS 2 vs 8.8 months for PS 0-1. At 10% for PS 2 patients and 18% for PS 0-1 patients, 1-year survival rates were "clearly different." Within the subset, however, "despite the small number, the difference in survival was significant in favor of combination chemotherapy, both by log-rank and Wilcoxon analysis," Dr. Lilenbaum reported. Median survival was 4.7 months for PS 2 patients on combination therapy vs 2.4 months for those on single-agent therapy—"almost double"—Dr. Lilenbaum said.

Quality-of-life scores did not differ between the two treatment arms in any of the 25 parameters tested. Nor were there any significant differences in quality-of-life issues for the elderly or patients with poor performance status.

Data on 542 patients "from enrollment to death" found no difference in several measurements of resource utilization—costs associated with treatment, Dr. Lilenbaum reported. These include days in hospital and intensive care unit, outpatient, emergency and home health visits, and tests such as x-ray and magnetic resonance imaging.

Resulting Recommendations

Dr. Lilenbaum listed the CALGB’s recommendations, based on the group’s own data:

• "Patients with advanced non-small-cell lung cancer should be treated with a platinum-based doublet (a carboplatin doublet in our case). It is the template upon which new treatment strategies should be developed."

• "Elderly patients can be treated similarly and safely with a carboplatin doublet."

• "The benefit of chemotherapy in PS 2 patients remains to be conclusively demonstrated." These patients should be entered in randomized trials testing best supportive care vs one drug or one drug vs two drugs.

• Single-agent chemotherapy used as second-line therapy may improve survival and quality of life.

• Clinical trials should also address combinations and targeted therapies, such as epidermal growth factor receptor (EGFR) inhibitors, which may improve survival and quality of life in third-line therapy. Two-drug and targeted therapy should also be studied at earlier stages.

Dr. Bunn echoed the call for additional trials. "We know what the standard treatment is—it’s two chemotherapeutic agents. The inability of triplets or alternating chemotherapy to improve survival results makes the development of novel targeted treatments more imperative."