Abstract 5632: Evaluating a novel molecular biomarker of angioinvasive lung adenocarcinoma with spatial transcriptomics

Abstract Microscopic vascular invasion (VI) is predictive of recurrence in stage I lung adenocarcinoma (LUAD) but is difficult to assess in resection specimens and cannot be accurately predicted prior to surgery. Thus, new biomarkers are needed to identify this aggressive subset of stage I LUAD tumors. To assess molecular and tumor microenvironment (TME) features associated with angioinvasive LUAD we profiled 171 resected stage I tumors with and without VI by RNA-seq, including 16 tumors by high-resolution spatial transcriptomics (10x Genomics Visium). Visium capture areas were selected by an experienced thoracic pathologist to include invasive foci, tumor regions distal to foci, and tumors without invasive foci. We identified a robust molecular signature of VI from the RNA-seq data containing subclusters of genes involved in hallmark programs of tumor suppression, EMT, angiogenesis, growth and metabolism. This VI-associated signature increases across the spectrum of indolent to aggressive stage I LUAD histopathology and is predictive of recurrence-free survival. Analysis of 43,421 Visium spots across 16 tumors revealed high inter-tumor patient heterogeneity, with most spots clustering by tumor identity, suggesting tumor-intrinsic properties. Scoring the Visium data for VI signature expression revealed spatial variability around VI foci, with distinct spatial distributions of different signature subclusters among tumor and stromal compartments within and adjacent to regions of tumor intravasation. We found increased expression of the bulk VI signature in invasive tumors, regardless of whether the capture area contained the invasive focus. We independently verified this finding by performing laser-capture microdissection and RNA-seq from invaded vessel regions and uninvaded parenchyma of 8 additional tumors with VI and tumor regions of non-VI tumors. Given increased signature expression in regions of invasive tumors at a distance from VI foci, we leveraged our bulk RNA-seq dataset to develop a transcriptomic predictor of VI. We applied a nested cross-validation approach within a training cohort to select a machine learning model. We then evaluated the performance of our model in an independent test set. Finally, we generated over 200 pseudo-bulked in silico biopsies similar in size to standard transthoracic needle biopsies using the spatial data. The scores from these in silico biopsies were similar to predictions from matched bulk RNA-seq data, suggesting robustness to intra-tumor heterogeneity. Our combined bulk and spatial transcriptomics analysis suggests that VI-associated gene expression extends far from the site of intravasation and can be used to predict the presence of VI. This may enable the prediction of angioinvasive LUAD from small biopsy specimens, allowing for more tailored treatment prior to surgery. Citation Format: Dylan Steiner, Lila Sultan, Travis Sullivan, Emily Green, Hanqiao Liu, Xiaohui Xiao, Gang Liu, Avrum Spira, Sarah Mazzilli, Kimberly Rieger-Christ, Eric Burks, Jennifer Beane, Marc Lenburg. Evaluating a novel molecular biomarker of angioinvasive lung adenocarcinoma with spatial transcriptomics. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5632.