Overall survival (OS) in a u.s. Asian population with stage IV NSCLC with EGFR mutations treated with first-line (1L) osimertinib (Osi) compared to earlier generation (Gen) of TKIs or sequential treatment (Tx).

Abstract

e20597 Background: Osi, a 3rd-gen EGFR tyrosine kinase inhibitor (TKI), was approved as 1L tx for metastatic non-small-cell lung cancer (NSCLC) with EGFR mutations (ex19del or L858R) in 2018. This was based on the FLAURA trial, showing improved progression-free survival and OS of 1L osi compared to 1st-gen TKIs. Despite this, subgroup analyses revealed no OS difference in Asian or L858R patients (pts). Studies such as GioTag and UpSwinG showed better median OS in Asians with ex19del mutations on sequential afatinib and osi tx, prompting the question if sequencing of tx may also benefit L858R pts. We conducted this analysis of real-world outcomes in U.S. Asian pts with EGFR mutated NSCLC treated with 1L osi, 1st- or 2nd- gen TKIs (prior-TKIs-only), or sequential 1st- or 2nd- gen TKI followed by osi (sequential-TKI-osi). Methods: We conducted a single-institution IRB-approved retrospective study at NYU, including Asian pts diagnosed with stage IV NSCLC with EGFR ex19del or L858R, who received any gen of TKIs from 1/1/2010 to 1/13/2023. We compared 3-year (3y) OS based on tx and mutation subtypes, excluding pts with less than 3 years of follow-up. The analysis utilized the Cox proportional-hazards model to calculate hazard ratios (HR) and P values of OS. Survival curves were estimated using the Kaplan-Meier method. The log-rank tests were used to assess survival differences between tx groups. Additionally, a multivariable Cox proportional-hazards model explored the association between pts' survival time and multiple predictor variables. Results: We analyzed 139 cases: 103 (74%) females, 68 (49%) L858R, 113 (81%) never smokers, and age 31-96 years. The 3y OS analysis revealed no significant difference in the OS in L858R pts who received prior-TKIs-only (p = 0.83) or sequential-TKI-osi (p = 0.90), compared to 1L osi (median OS not reached in all three groups). However, in ex19del pts, prior-TKIs-only (HR 0.31, 95% CI 0.13-0.76) and sequential-TKI-osi (HR 0.30, 95% CI 0.11-0.85) txs showed superior OS compared to 1L osi, with a median OS of 24.7 mo in 1L osi and not reached in other two txs. Among all predictor variables (tx, age, sex, EGFR subtypes, and smoking status), only age had a significant but small impact on OS (HR 1.03, 95% CI 1.01-1.06). Conclusions: Our real-world analysis revealed that Asian stage IV NSCLC pts with L858R treated with 1L osi had a similar 3y OS compared to those treated with prior-TKIs-only and sequential-TKI-osi. Whereas, 1L osi was inferior to both txs in OS for ex19del pts. This finding may be unique and significant in Asian ex19del pts, though our analysis is limited by the small cohort of a retrospective study. Larger cohorts from retrospective studies or prospective trials are needed to investigate the response of Asian ex19del pts to 1L osi, while underling the need for novel tx options for L858R pts.