Elizabeth Dudnik1, Elias Bshara2, Ahuva Grubstein3, Ludmila Fridel4, Tzippy Shochat5, Laila C Roisman6, Maya Ilouze7, Anna Belilovski Rozenblum2, Smadar Geva2, Alona Zer7, Ofer Rotem7, Aaron M Allen8, Nir Peled9. 1. Thoracic Cancer Service, Davidoff Cancer Center, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel. Electronic address: elizabetadu@clalit.org.il. 2. Sackler Faculty of Medicine, Tel Aviv University, POB 39040 Ramat Aviv, Tel Aviv 69978, Israel. 3. Department of Diagnostic Radiology, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel. 4. Department of Pathology, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel. 5. Statistical Consulting Unit, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel. 6. Cancer Institute, Soroka University Medical Center, Beer-Sheva 84101, Israel. 7. Thoracic Cancer Service, Davidoff Cancer Center, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel. 8. Thoracic Cancer Service, Davidoff Cancer Center, Rabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel; Sackler Faculty of Medicine, Tel Aviv University, POB 39040 Ramat Aviv, Tel Aviv 69978, Israel. 9. Cancer Institute, Soroka University Medical Center, Beer-Sheva 84101, Israel; Ben Gurion University of Negev, POB 653 Beer-Sheva 8410501, Israel.
Abstract
OBJECTIVES: Efficacy of immune check-point inhibitors (ICPi) in NSCLC with rare targetable drivers (RTDs) is largely unknown. MATERIALS AND METHODS: Consecutive patients with NSCLC and RTDs (non-EGFR/ALK, n-82) were selected from the Davidoff Cancer Center database. ORR, PFS, OS with ICPi, OS since advanced disease diagnosis, TMB, MSI, and PD-L1 expression were analyzed; uni- and multivariate PFS and OS analyses were done. OS with ICPi was compared between the RTD cohort and the non-selected NSCLC cohort (n-278). RESULTS: Of 50 tumors tested, 32%, 38%, and 30% were associated with ≥50%, 1-49% and <1% PD-L1 expression, respectively. Median TMB (n-48) comprised 4 muts/Mb (0-57); TMB ≥ 10 muts/Mb was seen in 19% of tumors. Both TMB and PD-L1 expression varied across different RTDs. All the 47 tumors were MSI stable. ORR with ICPi (n-44) was 16%, median PFS was 3.2 months (95% CI, 2.6-5.0), median OS was 16.2 months (95% CI, 8.4-NR). No correlation was seen between OS with ICPi and PD-L1 expression (p > 0.4), TMB (p > 0.8), or RTD type (p > 0.3). In the multivariate analysis, ECOG PS (p-0.005), targeted agents exposure (p-0.005), and ICPi exposure (p-0.04) were the only variables which correlated with OS since advanced disease diagnosis. Median OS since advanced disease diagnosis comprised 32 months (95% CI, 19.9-44.9) and 13 months (95% CI, 6.6-15.9) for patients who were and were not exposed to ICPi, respectively (log-rank test-6.3; p-0.01). In the inter-cohort comparison, for patients matched for ECOG PS (0/1), median OS with ICPi comprised 17.5 months (95% CI, 8.1-NR) and 8.6 months (95% CI, 6.7-NR) for RTD and non-selected patients, respectively (log-rank test-2.4, p-0.1). CONCLUSION: In NSCLC with RTD, ICPi have favorable efficacy and independent impact on OS. NSCLC with RTD is associated with MSI stable status and variable levels of PD-L1 expression and TMB; their predictive value remains to be determined.
OBJECTIVES: Efficacy of immune check-point inhibitors (ICPi) in NSCLC with rare targetable drivers (RTDs) is largely unknown. MATERIALS AND METHODS: Consecutive patients with NSCLC and RTDs (non-EGFR/ALK, n-82) were selected from the Davidoff Cancer Center database. ORR, PFS, OS with ICPi, OS since advanced disease diagnosis, TMB, MSI, and PD-L1 expression were analyzed; uni- and multivariate PFS and OS analyses were done. OS with ICPi was compared between the RTD cohort and the non-selected NSCLC cohort (n-278). RESULTS: Of 50 tumors tested, 32%, 38%, and 30% were associated with ≥50%, 1-49% and <1% PD-L1 expression, respectively. Median TMB (n-48) comprised 4 muts/Mb (0-57); TMB ≥ 10 muts/Mb was seen in 19% of tumors. Both TMB and PD-L1 expression varied across different RTDs. All the 47 tumors were MSI stable. ORR with ICPi (n-44) was 16%, median PFS was 3.2 months (95% CI, 2.6-5.0), median OS was 16.2 months (95% CI, 8.4-NR). No correlation was seen between OS with ICPi and PD-L1 expression (p > 0.4), TMB (p > 0.8), or RTD type (p > 0.3). In the multivariate analysis, ECOG PS (p-0.005), targeted agents exposure (p-0.005), and ICPi exposure (p-0.04) were the only variables which correlated with OS since advanced disease diagnosis. Median OS since advanced disease diagnosis comprised 32 months (95% CI, 19.9-44.9) and 13 months (95% CI, 6.6-15.9) for patients who were and were not exposed to ICPi, respectively (log-rank test-6.3; p-0.01). In the inter-cohort comparison, for patients matched for ECOG PS (0/1), median OS with ICPi comprised 17.5 months (95% CI, 8.1-NR) and 8.6 months (95% CI, 6.7-NR) for RTD and non-selected patients, respectively (log-rank test-2.4, p-0.1). CONCLUSION: In NSCLC with RTD, ICPi have favorable efficacy and independent impact on OS. NSCLC with RTD is associated with MSI stable status and variable levels of PD-L1 expression and TMB; their predictive value remains to be determined.