Progress in Small-Cell Lung Cancer: The Lowest Common Denominator

Abstract

Progress in the therapy of small-cell lung cancer (SCLC), commonly viewed at initial presentation as one of the most highly chemotherapy-sensitive solid tumors, has been painfully slow. In fact, it could be argued that little to no therapeutic advances have been made for extensive stage patients in more than 20 years. In a disease uniquely positioned from its causal biology and growth kinetics to be vulnerable to both new chemotherapeutic drug classes and the wealth of targeted therapeutic agents which have entered clinical development within the last decade, why has so little progress been made? In this issue of Journal of Clinical Oncology, Hermes et al present a randomized trial in extensive SCLC comparing oral etoposide plus carboplatin (EC) to a newer chemotherapeutic regimen of considerable interest: irinotecan and carboplatin (IC). The authors conclude that IC prolongs survival over EC, without compromising quality of life, supporting the use of this regimen as a new standard in SCLC. Viewed from a North American perspective, the Hermes et al trial clearly raises as many questions as it answers. Are the conclusions warranted? Does the regimen share a common denominator with other recent reports in the literature, or is it a one-of-a-kind experience? Finally, does this study truly represent progress in SCLC? To address these questions, we first must consider current treatment perspectives for extensive-stage SCLC. Although regional differences exist, in North America and worldwide, first-line chemotherapy for extensive stage SCLC most commonly consists of a platinum agent (cisplatin or carboplatin) combined with multiday intravenous etoposide. Attempting to build on this platform, several randomized phase III trials, all published in JCO, have disappointingly shown no survival advantage with the addition of topotecan consolidation, paclitaxel, Bec-2 vaccination, or thalidomide to a platinum-etoposide backbone. However, in 2002, the Japan Clinical Oncology Group published the striking results of J9511, a small phase III trial demonstrating the superiority of irinotecan and cisplatin (IP) over etoposide and cisplatin (EP) in response rate, progression-free survival (PFS) and survival, both overall and at 1 and 2 years. Was this the advance we had been waiting for? In view of pharmacogenomic considerations for irinotecan in Japanese populations and the limited patient sample size in J9511 due to Data Safety Monitoring Board–mandated early closure, confirmatory studies were launched. Using a modified dose schedule of IP, Hanna et al were unable to demonstrate superiority over EP. Most recently, at the 2008 Annual Meeting of the American Society of Clinical Oncology, the Southwest Oncology Group reported results of S0124, a large phase III study comparing IP to EP, in which the treatment regimens utilized in J9511 were exactly duplicated. While there were no significant efficacy differences observed between IP and EP in this North American population, a patient-level toxicity comparison of S0124 and J9511 confirmed greater toxicity in the Japanese study. Table 1 compares key parameters regarding patient demographics and end points of the four phase III trials comparing irinotecan and platinum to etoposide and platinum regimens. Clearly, the study by Hermes et al stands out in several ways: eligibility criteria (performance status [PS] 0 to 4 and brain metastases), use of oral etoposide, dose of carboplatin (area under the curve, 4), and age-mandated cycle 1 dose reductions, to name some of the most obvious. These differences in study design and patient characteristics may explain differences in survival outcomes; the Hermes et al trial reports the shortest overall survival times; response rates and PFS were not reported. Moreover, arbitrary dose reductions for patients older than age 70 may also have adversely affected survival times in the Hermes et al study. More than half of all new lung cancer patients are older than 70 years of age. The rationale for empiric age-related dose reductions in the Hermes et al study is unclear because older patients appear to tolerate chemotherapy as equally well as younger patients with similar PS and equal degrees of comorbidities. Potential pharmacogenomic variability in irinotecan metabolism due to genetic polymorphisms of uridine diphosphateglucuronosyltransferase is another dimension worth considering. Both interindividual and population-related differences are well recognized. While this factor was not assessed in the Hermes et al trial, other recent studies suggest it is of clinical significance in predicting the therapeutic index of this agent, including a preliminary report from the S0124 study. In addition, the treatment regimens selected for the Hermes et al trial may have contributed to reduced efficacy in both arms of the study. For example, despite initial enthusiasm, use of oral etoposide has fallen substantially in the United States. This is in part due to interpatient differences in bioavailability and pharmacodynamics, JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 26 NUMBER 26 SEPTEMBER 1