Abstract 2609
Background
Immunotherapy is a new standard of care for metastatic NSCLC patients. Approval of this therapy was based on randomized phase 3 trials. Clinical trials employees’ strict patient selection criteria, and this may not represent the ‘real-world’ population.
Methods
From January 2011 to December 2017, all patients with metastatic NSCLC referred for first oncological evaluation at 4 Hospitals in Brazil were identified by electronic database and included in the analysis. Main eligibility criteria used in first-line phase 3 immunotherapy trials were selected to be evaluated. OS was estimated by Kaplan-Meier curves. Cox proportional hazards model was performed to identify factors associated with survival. All statistical analyses were performed using SAS 9.4.
Results
537 patients were included in this analysis. Mean age was 62.73 ± 10.47 years, 57.3% male and 67.0% had adenocarcinoma. 332 (61.8%) patients didn't meet one or more eligible criteria. Patients with ECOG PS ≥ 2 and/or active brain metastasis accounted alone for 78.3% of non-eligibility cases. The median OS after the diagnosis of metastatic disease was 7.56 (95% CI: 6.37 to 9.59) months in the non-eligible group and 14.55 (95% CI: 12.16 to 18.23) in the eligible group. Logrank test detected a statistically significant difference between the survival curves in both groups (p = 0.0001). The hazard ratio (HR) of 1.778 (95% CI: 1.425 - 2.217) to mortality reflects worse prognostic features in non-eligible group. Also, Logrank test detected a statistically significant difference between the survival curves of ECOG 0-1 and ECOG 2-4 (HR 2.313 95% CI: 1.839 – 2.909 p < 0.0001) and histology, with a HR of 1.479 (95% CI: 1.135 – 1.927 p = 0.0036) in favor of adenocarcinoma. Median OS in ECOG 0-1 group was 13.17 months (95% CI: 11.89 – 15.05) and in ECOG 2-4 was 6.05 months (95% CI: 4.67 – 6.77). Median OS in Adenocarcinoma group was 12.48 months (95% CI: 9.63 – 13.83) and in Squamous cell was 6.51 months (95% CI: 5.29 – 11.17).
Conclusions
A significant part of real life Brazilian NSCLC population doesn’t fit the strict selection criteria specified by clinical trials. As soon as the experience and safety with this treatment improves, is desirable that future trials admits patiets more representative of real world NSCLC population.
Clinical trial identification
Legal entity responsible for the study
J. Cé Coelho.
Funding
Has not received any funding.
Editorial Acknowledgement
Disclosure
All authors have declared no conflicts of interest.