2030 - Clinical Outcomes of Early-Stage NSCLC Treated with Definitive SBRT by KRAS and EGFR Status and PD-L1 Expression

E. M. Davies¹, C. Billena², G. M. Videtic³, K. Yang⁴, and K. L. Stephans³; ¹Cleveland Clinic, Cleveland, NC, ²Memorial Sloan Kettering Cancer Center, Manhatten, NY, ³Department of Radiation Oncology, Cleveland Clinic, Cleveland, OH, ⁴Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, Cleveland, OH

Purpose/Objective(s): EGFR and KRAS mutational status and PD-L1 expression of non-small cell lung cancer (NSCLC) guide choice of systemic therapy. Previous analyses of KRAS-mutant tumors suggest that these mutations confer radioresistance to stereotactic body radiotherapy (SBRT), while the radiosensitivity of EGFR-mutant tumors is unknown. Other series have supported varying responses to definitive SBRT for early-stage NSCLC tumors according to PD-L1 expression. We review the clinical outcomes of stage I NSCLC tumors by KRAS, EGFR, and PD-L1 status treated with definitive SBRT. Materials/

Methods: Our institution implemented molecular testing of NSCLC biopsy specimens in 2016, and PD-L1 testing in 2018. We abstracted patient data, tumor characteristics, and clinical outcomes for all patients treated with definitive SBRT for stage I NSCLC from 1/2015-12/2023. None of these patients received any adjuvant or concurrent therapy. Datasets were stratified by mutational status and PD-L1 expression. Local recurrence (LR), disease recurrence (DR), and overall survival (OS) were determined for each case. Cumulative incidence (CIN) rates of LR and DR were compared using Gray’s test. OS was compared using Cox proportional hazard ratios and Kaplan-Meier analysis.
Results: Mutational status was available for 279 patients in the early-stage NSCLC SBRT database. 120 patients (43.0%) and 24 patients (8.6%) possessed KRAS- and EGFR-mutant tumors, respectively. PD-L1 expression was known for 153 patients and was treated as a categorical variable (0% or <1%, 63 pts; 1-49%, 53; 50-100%, 37). 49.8% of patients were alive at median follow up of 23 months. By Gray’s test assessing EGFR mutation status, there was no difference in LR (2-year CIN of 0% vs 1.4%, p=0.12) or DR (2-year CIN of 15% vs 30%, p=0.60) for EGFR-mutant tumors versus wild type; patients with EGFR mutant tumors exhibited superior OS (likelihood ratio 5.29, p=0.02). KRAS mutation status was not associated with differences in LR (3-year CIN 0.9% vs 3.3%, p=0.30), DR (3-year CIN 42% vs 33%, p=0.30), or OS (5-year survival 37.9% vs 44.7%, not significant (NS)). Patients with PD-L1 status 1-49% tended to have a lower LR (3-year CIN 8.4%, 2.7%, 0%, respectively, NS) which held for DR (3-year CIN 47%, 36%, 49%, NS). While on Cox proportional hazard analysis there was no difference in OS between the groups (p=0.09), Kaplan-Meier univariate analysis demonstrated superior survival among the PD-L1 1-49% group (p<0.01), with 5-year OS rates of 33.9% among patients with 0% or <1% PD-L1 expression, 68.4% among 1-49%, and 21.8% among 50-100%.
Conclusion: In spite of previous data suggesting relative radioresistance of EGFR-mutant tumors, this series suggests superior OS among this subgroup. These data provide no evidence of differential LR, DR, or OS among KRAS-mutant tumors treated with definitive SBRT. There may be superior LR, DR, and OS associated with PD-L1 status 1-49% that warrants further study.