Background: Nivolumab significantly improved OS versus docetaxel in patients with previously treated advanced non-squamous NSCLC (CheckMate 057; NCT01673867). KaplanÀMeier OS curves for nivolumab and docetaxel crossed at w7 months, suggesting nonproportional hazards between arms.

8001 Background: Erlotinib (E), a small-molecule EGFR TKI, is proven to extend survival versus placebo (P) in 2nd/3rd-line advanced NSCLC. The phase III SATURN study (BO18192) was initiated to evaluate E as maintenance therapy after standard 1st-line platinum-based chemotherapy (CT) in advanced NSCLC. Methods: Patients with no evidence of disease progression after 4 cycles of CT were randomized to receive either E 150 mg/day or P until progression or unacceptable toxicity. The primary endpoint was progression-free survival (PFS) in all patients and the co-primary endpoint was PFS in EGFR immunohistochemistry-positive (IHC+) patients. Results: A total of 1,949 patients entered the CT phase, of whom 889 were randomized to E (n=438) or P (n=451). Median age was 60 years for both arms. Baseline characteristics for E and P arms (%): male/female: 73/27 and 75/25; adenocarcinoma + BAC/squamous-cell/other: 47/38/15 and 44/43/13; stage IIIB/IV: 26/74 and 24/76; Caucasian/Asian/other: 84/14/2 and 83/15/2; ECOG PS 0/1: 31/69 and 32/68; current/former/never smoker: 55/28/18 and 56/27/17. PFS (by investigator assessment; confirmed by independent review) was significantly prolonged with E versus P in all patients (HR 0.71 [95% CI 0.62–0.82]; p<.0001) and in EGFR IHC+ patients (HR 0.69 [95% CI 0.58–0.82]; p<.0001). Subgroup analyses will be reported. Response rate was 12% with E versus 5% with P. Disease control rate (complete response + partial response + stable disease >12 wks) was 40.8% with E versus 27.4% with P (p<.0001). OS data are not yet mature. E was well tolerated: the majority of treatment-related adverse events (AEs) were grade 1/2. AEs reported in ≥10% of all patients were rash (60% with E versus 9% with P) and diarrhea (20% with E versus 5% with P); again, most were grade 1/2. Only 2.3% of patients receiving E had a serious treatment-related AE and 2.8% withdrew due to a treatment-related AE. Conclusions: The SATURN study met its primary and co-primary endpoints with high statistical significance. Erlotinib in the 1st-line maintenance setting is well tolerated, and significantly improves disease control and delays progression versus placebo across patient subgroups. [Table: see text]

No abstract

8020 Background: The SATURN (BO18192) study investigated whether erlotinib maintenance therapy improved PFS in patients (pts) with advanced NSCLC who had obtained clinical benefit from 1st-line chemotherapy. This study included a prospective analysis of the prognostic/predictive value of several molecular markers. Methods: 889 pts with advanced NSCLC whose disease had not progressed following 4 cycles of 1st-line platinum-doublet chemotherapy were randomized to erlotinib 150 mg/day or placebo. Mandatory tumor specimens were collected at baseline and tested for EGFR protein expression using immunohistochemistry (IHC), EGFR gene copy number using fluorescent in-situ hybridization (FISH), and EGFR and KRAS somatic mutations using DNA sequencing. Pts were stratified according to EGFR IHC status (any membranous staining in ≥10% tumor cells used as cut-off); the co-primary endpoint was PFS in EGFR IHC+ pts. Baseline whole blood samples were obtained for genotyping of EGFR (intron 1 CA-repeat polymorphisms). Results: In the overall population, erlotinib significantly prolonged PFS vs placebo (HR 0.71, p<.0001; primary endpoint). The co-primary endpoint was also met, with erlotinib significantly improving PFS in the EGFR IHC+ group (HR 0.69, p<.0001). Many tumor samples were assessable for molecular marker status (see table). Biomarker data suggest that patients derived a PFS benefit with erlotinib irrespective of EGFR FISH or EGFR intron 1 CA-repeat status. The magnitude of benefit with erlotinib was similar in both KRAS-mutant and KRAS wild-type pts. Conclusions: This is the largest biomarker analysis performed for erlotinib in a randomized, placebo-controlled setting, and answers key scientific questions regarding the prognostic and predictive value of potential biomarkers of efficacy. Full data will be presented. [Table: see text] [Table: see text]