SURVIVAL OUTCOMES AND COST-EFFECTIVENESS OF COMPREHENSIVE GENOMIC PROFILING (CGP) VERSUS SEQUENTIAL SINGLE GENE TESTING (SGT) APPROACH FOR PATIENTS WITH METASTATIC NON-SQUAMOUS NON-SMALL CELL LUNG CANCER (NSNSCLC) TREATED IN THE REAL-WORLD SETTING

Background: In nsNSCLC, the number and complexity of genomic biomarkers for targeted therapies has increased in recent years, which supported a recommendation for adopting CGP upfront over SGT by many international societies. However, there is no evidence that routine use of CGP improves overall survival (OS) in patients with nsNSCLC. Given the lack of randomized trials in this scenario, real-world data (RWD) and cost-effectiveness models can help payers in the decision to finance large NGS panels over non-NGS tests. Methods: Retrospective study from Oncoclínicas Data Lake comparing two historical periods (SGT in 2018-2019 versus CGP in 20202021) in selected patients with metastatic nsNSCLC treated with first-line palliative therapy in Minas Gerais or Rio de Janeiro and with minimum follow-up of 6 months. The primary endpoints of this study were OS rate at 12 months (OS12m) and the Number Needed to Treat (NNT) to prevent one death event within 12 months. The secondary endpoints were incremental cost-effectiveness ratio (ICER) of CGP per true positive case detected and the Cost of Preventing an Event (COPE) according to the RWD model and a standard-of-care (SoC) synthetic model (whereby patients follow strict guidelines for SoC treatment based on test results). Results: Clinical and molecular test data were available for 80 patients in CGP cohort and 78 in the SGT cohort. OS12m was 86.5% (95%CI 76%-98%) in patients undergoing CGP and 68.5% (95%CI 58%-80%) for SGT population. First-line therapy with a matched targeted agent (EGFR, ALK/ROS1 and BRAF inhibitors) increased from 15% with SGT to 19% with CGP, and use of immunotherapy agents also increased from 8% in 20182019 to 43% in 2020-2021 (without significant impact in OS12m). The NNT with CGP was 6 (95%CI: 4 – 9), ICER of CGP was R$ 34.834,00. The COPE according to RWD model was R$491.963. On the other hand, if SoC is followed (increased use of targeted agents and reduction in immunotherapy), the NGS was a cost-saving strategy in the first year of follow-up, with savings of R$ 22.618 to prevent one event. Conclusions: Although the sample size and follow-up do not allow comparative analyses, we believe that these data reinforce the value of early and facilitated access to CGP in selected nsNSCLC populations. It also confirms our ability to collect clinico-genomics RWD and generate evidence that is valuable to the scientific community and healthcare systems, improving decision-making of stakeholders.