In this issue of the Journal of Clinical Oncology, Ramlau et al describe the results of a phase III trial comparing oral topotecan versus intravenous docetaxel in patients with previously treated advanced stage non–small-cell lung cancer (NSCLC). The study was large (more than 800 patients), employed a noninferiority design, and was performed by an experienced international group of lung cancer investigators. The authors report that oral topotecan is active in this clinical setting, and that the study met predefined criteria for noninferiority with regard to survival. What do these results mean for the practicing oncologist, and will they alter our second-line management of NSCLC patients? To put these data into perspective, we must first ask how this study stacks up against previously reported trials of second-line therapy. In the current study, for example, efficacy parameters of response rate and overall survival were not statistically different between docetaxel and topotecan (5% in each arm, and 31.7 v 27.9 weeks, respectively; P .057), whereas time to progression was superior in the docetaxel arm (13.1 v 11.3 weeks, respectively; P .02). Although both agents were reported to be relatively well tolerated, the hematologic toxicity profile favored docetaxel, as did the overall quality-of-life assessment (P .0001). In fact, despite a significant expansion of the therapeutic armamentarium, there appears to be no clear winner among the agents tested in randomized trials of second-line therapy, at least in terms of efficacy parameters (Table 1). This conclusion seems particularly well justified when one looks at the results from study to study with docetaxel at 75 mg/m, which served as a study arm in five of the seven trials listed. Most likely, variability in eligibility criteria and patient characteristics account for differences between studies. In addition to measures of efficacy and toxicity, docetaxel is reported to convey quality-of-life benefits in this clinical setting, and to be cost effective. So where do we go from here? With the exception of the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) erlotinib and gefitinib, for which patient characteristics such as sex (female), histology (adenocarcinoma or bronchioloalveolar carcinoma), smoking status (never-smoking), or tumor biomarkers (EGFR gene copy number or mutation status) portend a greater likelihood of benefit, there are no clear-cut clinical features, and as yet no generally accepted predictive biomarkers, to guide physicians in selecting from among the remaining available options for second-line therapy for the majority of patients with NSCLC. An obvious question is whether combinations of these second-line therapies are likely to be better than single agents in achieving the palliative goals of second-line therapy. Once again, there are no definitive results to suggest that combinations are superior, although many clinical trials are ongoing investigating the possibilities. Two potential considerations worthy of discussion are combinations of the chemotherapeutic agents listed above with EGFR TKIs or with antiangiogenic agents such as bevacizumab. With regard to the former, the possibility of antagonism between concurrently administered chemotherapy and EGFR TKIs has led to a series of studies evaluating the concept of pharmacodynamic separation, such as intermittent dosing of the EGFR TKI together with chemotherapy. At the same time, combinations of bevacizumab with chemotherapy, recently shown to improve survival in first-line therapy, are also of interest in secondline therapy, as exemplified by a joint North Central Cancer Treatment Group/Southwest Oncology Group trial investigating pemetrexed plus bevacizumab. Table 1. Results of Randomized Phase III Trials in Second-Line Treatment of NSCLC