Re-Evaluating Duration of Therapy in Advanced Non–Small-Cell Lung Cancer: Is It Really Duration or Is It More About Timing and Exposure?

Abstract

The optimal therapeutic approach for the individual patient diagnosed with advanced stage IV non–small-cell lung cancer (NSCLC) in 2009 is becoming increasingly difficult to define. Platinum-based doublets have been a mainstay of therapy in the first-line setting since a landmark meta-analysis clearly demonstrated a significant, yet modest, survival benefit compared to best supportive care. The development of several active drugs in patients previously treated with platinum-based doublets has lead to the approval of four agents in the secondand third-line setting: docetaxel, pemetrexed, gefitinib, and erlotinib. The validation of both the epidermal growth factor receptor (EGFR) and vascular endothelial growth factor (VEGF) pathways as therapeutic targets in advanced NSCLC has been an important step in modestly improving outcomes. Scagliotti et al recently showed a therapeutic advantage for pemetrexed and cisplatin compared to gemcitabine and cisplatin in nonsquamous NSCLC. This finding shattered the concept that all platinum-based doublets are created equal. Histology now matters both for the safety of bevacizumab and the efficacy of pemetrexed. Two major questions have been addressed over the past decade or so. One has been the optimal duration of first-line platinum-based therapy in advanced NSCLC. The initial trials addressing this issue were performed because of observations suggesting that responses occurred early (after the first two to three cycles) and because of cumulative toxicity to prolonged therapy. The message from these trials was that survival was not improved by prolonged therapy with platinum-based doublets led to most major guidelines recommending a brief duration (four cycles) in the first-line setting. These trials were done before the validation of both the EGFR and VEGF pathways with cetuximab and bevacizumab where the targeted agent was continued until disease progression. Whether this should be called prolonged therapy or maintenance therapy is a matter of debate. A second question addressed the continuation of therapy with a single agent (which had known activity in advanced, refractory NSCLC) following a defined duration of first-line platinum-based therapy. Many of these trials evaluated agents approved in the secondor third-line setting. Again, whether this is prolonged duration, sequential therapy, maintenance therapy, or early administration of second-line therapy remains a matter of debate. Interestingly, these trials seem to have had a much greater impact on both progression-free survival (PFS) as well as overall survival (OS) compared to the extended use of platinumbased doublets. In practice, first-line therapy with platinum-based doublets is typically delivered for four (maybe six in some cases) cycles. If bevacizumab or cetuximab was part of the original regimen, it would typically be continued until disease progression based on the design of the pivotal trials. There are virtually no data regarding the appropriate surveillance strategy for patients once they have completed first-line therapy (ie, what tests should be done and when, how often should patients be evaluated). There is also no information on the value of a treatment break. Many patients who have significant cumulative toxicities during the first four cycles of therapy often resolve these toxicities with a short break or time off therapy. Second-line therapy administered at the time of disease progression (presumably radiographic or symptomatic) has been demonstrated to improve survival and palliate symptoms. However, many patients never receive such therapy due to rapid disease progression or disease-related events which significantly compromises their performance status. Strategies which would identify disease progression earlier or routine administration of effective second-line therapy at a defined time point may increase the number of patients receiving subsequent lines of therapy. In this issue of Journal of Clinical Oncology, Soon et al provided a very timely and insightful meta-analysis addressing the duration of therapy in advanced NSCLC. They identified 13 randomized clinical trials (comprising 3,027 patients) of varying designs, none of which included bevacizumab or other targeted agents. The authors concluded that extending therapy, particularly with a third generation agent or regimen, substantially improved PFS (hazard ratio [HR], 0.75; 95% CI, 0.69 to 0.81; P .00001) but had a far less dramatic, but still statistically significant, effect on OS (HR, 0.92; 95% CI, 0.86 to 0.99; P .03). Soon et al should be commended for completing this analysis, as it helps us to refocus the issue of treatment strategies in the era of multiple lines of therapy. Several issues with this meta-analysis deserve further consideration. Five of the 13 trials (1,369 patients or 45% of all the patients included in the meta-analysis) simply evaluated the duration of first-line therapy with platinum-based doublets comparing three to four cycles with six or more cycles. None of these trials were individually positive for OS. In almost all trials comparing JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 27 NUMBER 20 JULY 1