Commentary on Abstracts #986 and #1015

Currently there are a number of available agents that are moderately active in non–small-cell lung cancer (NSCLC). These include cisplatin (Platinol), gemcitabine (Gemzar), vinorelbine (Navelbine), paclitaxel (Taxol), docetaxel (Taxotere), and irinotecan (Camptosar). How best to combine them, maximizing survival while minimizing toxicity, is the subject of intense investigation.

In an effort to decrease toxicity without compromising response and survival rates, a major effort has been undertaken to find a two-drug combination that does not contain cisplatin. The Greek Cooperative Group (abstract #986) used a regimen with promising results in a pilot phase II trial-docetaxel plus gemcitabine-and subsequently compared it to docetaxel plus cisplatin in a randomized phase II trial powered to detect differences in toxicities between the arms, but not necessarily modest response rate or survival differences.

Granulocyte colony-stimulating factor (G-CSF, filgrastim [Neupogen]) was administered prophylactically in both arms. The toxicities were similar, aside from diarrhea, which was more common in the cisplatin group. The response rates, duration of response, and survival were similar. Interestingly, the response rates for adenocarcinoma were better with the nonplatinum regimen, while the cisplatin regimen was better for the nonadenocarcinomas. This unusual and unexpected observation needs confirmation in phase III trials before it is widely accepted, and is used to guide future clinical practice. This trial suggests that the results with non–platinum-containing regimens are similar to those with cisplatin, but whether toxicity or quality of life can be significantly improved remains in question.

The addition of a third drug to an active two-drug regimen in NSCLC, while occasionally increasing response rates, generally has not improved survival and almost always increases toxicity (Proc Am Soc Clin Oncol 13:347 [abstract 1161], 1994). One study reexamined this strategy. Frasci et al (abstract #1015) present the results of a phase II study of the three-drug regimen cisplatin/gemcitabine/paclitaxel in a cohort of chemotherapy-naive patients with excellent performance status. A high objective response rate was observed (68%), with a similar proportion of patients showing an improvement in quality-of-life scores. Projected 1-year survival was an impressive 70%, but no confidence intervals are stated, and the follow-up time is short.

The regimen was not without toxicity, however, and one would like more details about the eight episodes of “severe nonhematologic toxicity.” The results of this trial are interesting, but one must be wary of drawing too many conclusions from small phase II studies, which have highly selected patients, and the results of which often lead to overly enthusiastic interpretation. The phase III trial involving this three-drug regimen will help to truly determine its place in the management of advanced NSCLC.