The phase III E1505 trial was designed to investigate the addition of bevacizumab to adjuvant chemotherapy in patients with early-stage, completely resected NSCLC. It was based on the rationale that the benefit of adjuvant cisplatin-based chemotherapy is modest in this population. E1505 included 1,501 patients with completely resected, stage IB NSCLC. They were randomised to either 4 cycles of cisplatin-based chemotherapy only or the same chemotherapy plus bevacizumab for up to one year. Four chemotherapy regimens per investigator choice were allowed: cisplatin/ vinorelbine, cisplatin/ docetaxel, cisplatin/ gemcitabine, and cisplatin/ pemetrexed.
E1505 was powered for the primary endpoint of OS only and was stopped early for futility. The updated results presented at the ASCO Congress confirm the lack of difference between the two treatment arms with regard to OS and DFS; the hazard ratio was 0.99 for both endpoints.
This analysis also focussed on outcomes based on chemotherapy subsets. Patients were pooled with respect to the regimen used regardless of treatment arm (with or without bevacizumab) and divided into non-squamous and squamous cohorts to account for the restriction of pemetrexed administration to patients with non-squamous histology. DFS and OS were calculated for each chemotherapy group.
This post-hoc, non-randomised subset analysis yielded no differences for OS and DFS across all four adjuvant cisplatin- based chemotherapy regimens in both squamous and non-squamous tumours. Moreover, hazard ratios were calculated using vinorelbine as a reference, because cisplatin/ vinorelbine had been the regimen used in prior adjuvant trials. Again, no significant differences were noted for patients with both histologies.