Search

Brain metastases (BM) affect up to 50% of patients with NSCLC and are associated with poor prognosis in patients with non-targetable genomic alterations. Radiation forms the typical mainstay of intracranial management in such patients given the historically guarded intracranial efficacy of most systemic options. Immunotherapy-based approaches have shown promise but the extent that patients with BM may display extended survival with combination radiotherapeutic and immunotherapy-based approaches remains unclear. We identified 160 patients harboring non-squamous NSCLC with 486 newly-diagnosed BM between 2015-2023 managed with brain-directed radiation (either stereotactic-based, SRT or whole brain radiation therapy, WBRT) and immune checkpoint inhibition within 3 months of radiation at Dana-Farber Cancer Institute/Brigham and Women's Hospital (Boston, MA). PD-L1 tumor expression was categorized as <1%, 1-49%, 50-89%, and ≥90% based on immunohistochemistry. TMB was assessed via targeted next-generation sequencing. Cox-proportional hazards regression was used to assess all-cause mortality. Mixed-effects Fine-Gray models were used for per-metastasis outcomes, including rates of local control (LC) and radiation necrosis (RN). The median number of BM was 3 (IQR 1-7). Neurosurgical resection in advance of radiation was performed in 51/160 (32%) pts. SRT was used in 130/160 (81%) versus 30/160 (19%) patients treated with WBRT, respectively. Median follow-up time was 30.1 months (mo). Higher PD-L1 expression was associated with significantly longer overall survival (OS) with a median survival in patients with PD-L1<1%, 1-49%, 50-89% and ≥90% being 11.8, 14.4, 29.5 and 33.1 mo, respectively (Table). Two-year LC rates were excellent and similar across PD-L1 categories. Two-year rates of RN were numerically higher in patients with higher PD-L1 (Table). Higher TMB (≥90th vs. <90th percentile) was not associated with significantly longer OS, or higher LC or RN rates (Table). Higher PD-L1 tumor expression was associated with improved all-cause mortality. A significant percentage of patients with a very high PD-L1 displayed long-term survival, highlighting the potential importance of multimodality therapy in such patients, and demonstrating that clinicians should be cognizant of treatment-associated long-term toxicities in this population. [ABSTRACT FROM AUTHOR]

We aim to implement an immune cell score model in routine clinical practice for resected non-small-cell lung cancer (NSCLC) patients (NCT03299478). Molecular and genomic features associated with immune phenotypes in NSCLC have not been explored in detail. We developed a machine learning (ML)-based model to classify tumors into one of three categories: inflamed, altered, and desert, based on the spatial distribution of CD8+ T cells in two prospective (n = 453; TNM-I trial) and retrospective (n = 481) stage I-IIIA NSCLC surgical cohorts. NanoString assays and targeted gene panel sequencing were used to evaluate the association of gene expression and mutations with immune phenotypes. Among the total of 934 patients, 24.4% of tumors were classified as inflamed, 51.3% as altered, and 24.3% as desert. There were significant associations between ML-derived immune phenotypes and adaptive immunity gene expression signatures. We identified a strong association of the nuclear factor-κB pathway and CD8+ T-cell exclusion through a positive enrichment in the desert phenotype. KEAP1 [odds ratio (OR) 0.27, Q = 0.02] and STK11 (OR 0.39, Q = 0.04) were significantly co-mutated in non-inflamed lung adenocarcinoma (LUAD) compared to the inflamed phenotype. In the retrospective cohort, the inflamed phenotype was an independent prognostic factor for prolonged disease-specific survival and time to recurrence (hazard ratio 0.61, P = 0.01 and 0.65, P = 0.02, respectively). ML-based immune phenotyping by spatial distribution of T cells in resected NSCLC is able to identify patients at greater risk of disease recurrence after surgical resection. LUADs with concurrent KEAP1 and STK11 mutations are enriched for altered and desert immune phenotypes. • A novel ML model was devised and validated for in situ immune phenotyping. • The ML-driven immune phenotypes served as an independent prognostic indicator. • LUADs featuring concurrent KEAP1 and STK11 mutations predominantly display altered or desert-like immune phenotypes. • In the desert phenotype, the enrichment of nuclear factor-κB signaling correlates with the exclusion of cytotoxic T cells. [ABSTRACT FROM AUTHOR]