269 - Clinical and Genomic Predictors of Local Failure in NSCLC Patients Treated with Lung SBRT

05:10pm - 05:20pm PT

Room 155/157

Presenter(s)

Roshal Patel, MD - Memorial Sloan Kettering Cancer Center, New York, NY

R. R. Patel¹, V. C. Ng¹, N. Shaverdian¹, D. Gelblum¹, P. Iyengar¹, C. B. Simone II², D. R. Gomez¹, A. Rimner³, and A. J. Wu¹; ¹Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, ²New York Proton Center, New York, NY, ³Department of Radiation Oncology, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, German Cancer Consortium (DKTK), partner site DKTK-Freiburg, Freiburg, Germany

Purpose/Objective(s): Stereotactic body radiation therapy (SBRT) offers effective local control (LC) for primary, metastatic, and localized node-negative recurrent lung tumors in patients with non-small cell lung cancer (NSCLC). Local failure post SBRT can lead to morbidity and mortality. We aim to identify clinical and genomic predictors of LC to better optimize treatment regimens.

Materials/Methods: We compiled a retrospective database of NSCLC patients who received lung SBRT (defined as treatment with a biologically effective dose of =80Gy), and who underwent next generation sequencing (NGS). The NGS panel includes at least 341 genes and utilizes tumor and matched normal DNA samples to detect protein-coding alterations, copy number alterations, and structural rearrangements. Mutations considered to be “oncogenic” or “likely oncogenic” were included in this analysis. The primary endpoint of this study was LC. Local recurrence was defined by tumor recurrence overlapping with the PTV margin and assessed using CT, PET-CT, and/or biopsy data. Cox proportional hazard regression was used for statistical testing.

Results: We identified 366 consecutive patients with NSCLC treated with lung SBRT between 2015-2023 who had NGS. Median follow-up was 25.7 months. At the time of treatment, patients were early-stage (33%), advanced/metastatic (62%), and recurrent (11%). Median size of treated tumor was 2.3cm (IQR 1.6-3.4cm). The most common dose regimens were 50Gy/5 (45%), 48Gy/4 (24%), and 54Gy/3 (13%).

Most patients received a BED =100 Gy (89%, range 80-151.2 Gy). Cumulative LC rates at 12- and 24-months were 92% and 78%, respectively. On univariable analysis, larger tumor size (p<0.001, HR 1.39, 95% CI 1.2-1.6) and lower BED (p=0.033, HR 0.97, 95% CI 0.97-0.99) were associated with worse LC; however, stage at treatment was not (p=0.21). We identified 61 genes enriched in at least 5% of tumor samples (>18 occurrences). Of these, seven altered genes were associated with decreased risk of LC: CREBBP (7%, p=0.046, HR 2.08, 95% CI 1.0-4.3), KMT2B (8%, p<0.001, HR 2.87, 95% CI 1.5-5.3), KMT2C (7%, p=0.024, HR 2.16, 95% CI 1.1-4.2), KMT2D (11%, p=0.044, HR 1.94, 95% CI 1.0-3.7), NFE2L2 (6%, p=0.048, HR 2.1, 95% CI 1.0-4.4), NKX2-1 (13%, p=0.014, HR 2.04, 95% CI 1.2-3.6), and SDHA (7%, p=0.012, HR 2.5, 95% CI 1.2-5.0). On multivariable analysis, larger tumor size (p<0.001, HR 1.3, 95% CI 1.2-1.6) and KMT2B alterations (p=0.01, HR 2.4, 95% CI 1.2-4.6) were independently associated with decreased LC. KMT2B alterations were found in both early (39%) and advanced/recurrent (61%) stage patients. The 12- and 24-month cumulative LC rates for KMT2B altered tumors were 74% and 33%, respectively.

Conclusion: This analysis identified independent factors associated with LC post-SBRT, including tumor size and the presence of KMT2B gene alterations. KMT2 family genes play a role in histone methylation and, therefore, can impact radiation sensitivity. Further studies may validate the role of dose intensification in these patients.