A meta-analysis of randomized controlled trials (Br Med J 311:899-909, 1995) has shown that the use of cisplatin (Platinol)-based combination chemotherapy in patients with good performance status leads to a modest improvement in median survival and an absolute increase in 1-year survival proportion of 10%. There are several different platinum-based regimens approved by the Food and Drug Administration for use in advanced non–small-cell lung cancer. Whether any one regimen is superior is unclear. A recent randomized controlled trial found no difference in median survival and quality of life between carboplatin (Paraplatin)/paclitaxel (Taxol)—the most commonly used regimen in the United States—and cisplatin/vinorelbine (Navelbine)—a regimen more popular in Europe and Canada (Proc Am Soc Clin Oncol 18: 461a [abstract 1777], 1999). The newer agents gemcitabine (Gemzar) and docetaxel (Taxotere) are among the most active single agents in non–small-cell lung cancer, and use of either in combination with cisplatin has shown promise.

In abstract #2, Schiller et al report the results of an important phase III trial performed by the Eastern Cooperative Oncology Group (ECOG) that compares four different regimens in patients with advanced non–small-cell lung cancer. The control arm of cisplatin/paclitaxel (24-hour infusion) was compared to three experimental arms: cisplatin/gemcitabine, cisplatin/docetaxel, and carboplatin/paclitaxel. The study enrolled a total of 1,207 previously untreated patients, of whom the vast majority had stage IV disease (vs stage IIIB). Due to significant toxicity seen in the first 63 patients enrolled with performance status 2 (Proc Am Soc Clin Oncol 18:461a [abstract 1779], 1999), the protocol was amended to restrict entry to those with performance status 0 or 1. Patients were stratified on the basis of stage, performance status, weight loss, and the presence or absence of brain metastases.

It was found that the regimens had comparable overall degrees of toxicity, although there were slight variations in the toxicity spectrum associated with each regimen, as can be seen in the abstract. Disappointingly, there were no differences in median or 1-year overall survival between the control arm and any of the experimental arms. Similarly, there was no superiority of any arm in terms of response rate. Cisplatin/gemcitabine was associated with a modest prolongation of time to progression when compared to the control arm: 4.5 vs 3.5 months, P = .002. One can conclude from this study that at the present time there is no one regimen that is distinctly superior in terms of survival, response rate, or even degree of toxicity. Choice of regimen should be based therefore on other factors, such as comorbid illness, patient preference, convenience, and cost. The toxicity seen in those with performance status 2 in the ECOG trial helps reinforce that cisplatin-based combination chemotherapy should not be used in these patients.

In those who have failed cisplatin-based chemotherapy, single-agent docetaxel, 75 mg/m² every 3 weeks, has been shown to offer a survival advantage (J Clin Oncol 18: 2095-2103, 2000). In an attempt to improve outcomes in the second-line setting, Spiridonidis et al (abstract #1968) performed a phase II trial of the combination of docetaxel and gemcitabine given on a 4-week schedule. In a group of 40 previously treated patients with good performance status, responses were seen in 12 patients (32%). No further efficacy data are available at this time. Hematologic toxicity was significant, with febrile neutropenia occurring in 10% of patients and grade 3/4 thrombocytopenia in 20%; grade 3 asthenia was the most common nonhematologic toxicity, occurring in 33% of patients.

In the absence of survival data, it is difficult to comment on this trial. Although the response rate of 32% is greater than that seen in phase II trials of single-agent docetaxel in the second-line setting, increases in response rates do not reliably translate into improvements in survival. At the present time, the standard of care for the second-line treatment of medically suitable patients with advanced non–small-cell lung cancer should remain single-agent docetaxel at the dose and schedule mentioned. Future studies of docetaxel combinations should consider using weekly docetaxel, which causes less myelosuppression and might be more tolerable.