1191PRETROSPECTIVE COMMUNITY BASED EXPLORATIVE STUDY ON CISPLATIN-BASED ADJUVANT CHEMOTHERAPY VS. SURGERY ONLY IN COMPLETELY RESECTED STAGE IB NON-SMALL CELL LUNG CANCER (NSCLC)

Aim: The benefit of adjuvant chemotherapy (ACT) following complete surgery in NSCLC stage IB is not fully clarified. ACT in this situation is currently considered optional. Ouraim wasto examine aconsecutive group of unselected completely resected NSCLC stage IB patients (pts) who either received ACTor solelysurgical resection, and compare overall survival (OS) and disease free survival (DFS) between the two groups. Methods: All pts underwent complete surgery at the National University Hospital in Copenhagen during 2005-2012 and were considered fit for ACT with cisplatin 75 mg/ m 2 i.v. day 1 and vinorelbine 30 mg/m 2 i.v. day 1 + 8 q 3 wks for 4 cycles. Some pts decided to receive ACT (proband group (PG)) while other patients declined (control group (CG)). 63 variables were collected from medical and surgical records. Co-morbidity was scored according to Charlson’s co-morbidity score (CCS). Kaplan-Meier plots on 5-year overall survival and disease free survival were calculated and compared by log-rank analysis. A multi-variate Cox regression analysis was performed to explore on predictors for outcome. Results: Totally 184 pts were reviewed, with 74 pts in the proband group and 110 pts in the control group. The patients in CG were older (p < 0.001), had higher CCS (p < 0.001), and had more pts with pleural invasion (p = 0.028) than PG group. 5-year OS was 80 % for PG vs. 59 % in CG, p = 0.03, while 5-year DFS in PG was 73 % and 54 % in CG, p = 0.022. The Cox regression analysis revealed that ACT (p = 0.02), low CCS (p = 0.003) and low performance status (p = 0.035) were independent significant prognostic factors regarding better OS. Conclusions: Patients who decided to receive ACTwere younger and had lower CCS. ACT was an independent prognostic factor and patients receiving ACT benefitted significantly with respect to both DFS and OS when compared to controls without ACT. Disclosure: All authors have declared no conflicts of interest.