In the current issue of the Journal of Thoracic Oncology, Noble and colleagues provide an excellent comprehensive review of systemic treatment for patients with advanced non-small cell lung cancer (NSCLC) previously treated with chemotherapy. The authors conclude that sufficient evidence exists for the routine use of either single-agent pemetrexed or docetaxel or erlotinib and that insufficient evidence exists for the routine use of other agents or combinations. I concur with these conclusions. However, whereas these agents result in prolonged survival and improved quality of life for some patients, the overall gains are modest at best. In fact, most patients still receive either no benefit or minimal benefits from treatment, and, for those who benefit, the duration of benefit is weeks to months, not months to years. We can and must do better, but how? An ideal starting point would be to enroll all eligible, willing patients into clinical trials that test new agents, combinations, and strategies. As with the first-line treatment of NSCLC with cytotoxics, a plateau has been reached with chemotherapy in the secondand third-line settings. We must take a cue from other recent successes in oncology in which molecularly targeted agents are given to carefully selected patient populations based on disease and tumor characteristics. In the last year, several agents have offered hope that these advances are forthcoming. I will mention just four, although there are dozens of other promising agents in the same class or others with differing mechanisms of action currently under investigation. Angiogenesis plays a vital role in tumor survival, and bevacizumab has already been proven effective in patients with NSCLC in the first-line setting. However, what is the role of continuing anti-angiogenic therapy beyond first line? Sorafenib is a multi-kinase inhibitor whose targets include the Ras signaling pathway and the vascular endothelial growth factor (VEGF) receptors. A phase II study (n 52) of sorafenib in patients with previously treated NSCLC was recently reported by Gatzemeier et al.1 Of the patients, 59% were reported to have stable disease, including 31% who had objective regression of disease. The authors reported that patients with higher VEGF levels had worse outcomes and that VEGF levels declined with treatment. Sunitinib is also a multi-kinase inhibitor that targets the angiogenesis receptors of VEGF and PDGF. A phase II study (n 63) by Socinski et al.2 included patients with advanced NSCLC who had received at least one prior chemotherapy regimen, including 57% who were receiving sunitinib as at least third-line therapy. Overall, 9.5% achieved a partial response, and an additional 42.9% had stable disease. Remarkably, 45 of 56 patients (80.4%) with disease evaluation reported had tumor regression at some point during treatment. These trials excluded patients treated with bevacizumab in the first-line setting. With the forthcoming, expected broad use of bevacizumab in the first-line setting, one must logically ask whether inhibition of the VEGF receptor intracellular tyrosine kinase will be effective in someone who has previously received bevacizumab. ZD6474 is a dual kinase inhibitor that targets both the epidermal growth factor (EGFR) and VEGFR. Natale et al.3 reported the results of a randomized phase II study (n 168) of gefitinib versus ZD6474. At the time of disease progression, crossover to the other agent was allowed. The hazard rate for progression-free survival (the primary