Different non-cross-resistant

agents have been used as a maintenance strategy after a defined number of induction cycles with a platinum-based regimen in several randomized clinical trials (Table 2). Table 2 Studies with switch to a different agent after a platinum-based induction First Author (N of randomized pts to maintenance) Maintenance Schema Primary End Point Median PFS (mo) P value Median OS (months) P value References Fidias P. (309) Immediate vs delayed docetaxel OS 5.7 vs 2.7 0.0001 12.3 vs 9.7 0.08 [26] Ciuleanu T. (663) Pemetrexed vs placebo PFS 4.3 vs 2.6 0.0001 13.4 vs 10.6 0.012 [27] Cappuzzo F. (889) Erlotinib vs placebo PFS 12.3 vs 11.1 0.0001 12 vs 11 0.063 [31] Perol M. (464) Gemcitabine vs erlotinib vs placebo PFS Akt inhibitor 3.7 vs 2.8 vs 2.1 nr HR 0.86 vs 0.81 na [21] Kabbinavar F.* (768) Selleckchem SB203580 Bevacizumab ± Erlotinib PFS 4.8 vs 3.7 0.006 Na na [32] Gaafar RM (173) Gefitinib vs placebo OS 4.1 vs 2.9 0.0015 Na na [33] *In this trial bevacizumab was already present in the induction therapy nr: not reported, na: not available Vinorelbine

versus placebo Westeel et al. designed a trial testing vinorelbine maintenance in stage IIIB and IV NSCLC after induction with mitomycin, ifosfamide and cisplatin (MIC). Nearly 600 patients were recruited and 181 were randomized to receive vinorelbine maintenance or BSC for up to 6 months. Mean duration of therapy was 13.8 months and 23% of patients completed 6 months of vinorelbine: in the majority of cases treatment interruption was due to disease progression (38%) or treatment toxicity (21%). The HR for OS, after adjusting

for stage, was 1.08 (95% CI = 0.79 to 1.47; p = .65) and median OS was 12.3 months in both arms. Clomifene One- and 2-year survival rates were 42.2% and 20.1% in the vinorelbine arm and 50.6% and 20.2% in the BSC arm respectively (log-rank P = .48). No difference in PFS was observed (HR = 0.77, 95% CI = 0.56 to 1.07; p = .11; median PFS 5 months with vinorelbine and 3 months in the BSC arm) [25]. Immediate versus delayed docetaxel Fidias and coll. conducted a phase III trial randomly assigning patients with objective response or stable disease after four cycles of gemcitabine/carboplatin first-line chemotherapy to immediate (‘maintenance’) docetaxel or a “”delayed”" second-line docetaxel, initiated at the time of disease progression. A total of 566 patients were enrolled and 309 patients with non-progressive disease were randomized. Among 153 patients assigned to immediate docetaxel, 145 (94.8%) received at least one treatment cycle and among 154 patients assigned to the to “”delayed docetaxel”", 98 (62.8%) patients initiated therapy. Reasons for not initiating the planned second-line included toxicity from previous treatment, decline in PS, and investigator’s decision. The median number of docetaxel cycles administered in both arms was 4.4.