Applying multi-omics to detect indolent and aggressive early-stage lung adenocarcinoma
Embargo Date
2027-09-04
OA Version
Citation
Abstract
Histologic grading of lung adenocarcinoma (LUAD) is a strong predictor of outcome but can only be completed after surgical resection. Biomarkers that are predictive of indolent and aggressive LUAD grade have the potential to improve preoperative management of early-stage disease. Here, we leveraged two approaches, radiomics and transcriptomics, to enhance prognostication and treatment planning for stage I LUAD. First, we validated an existing radiomic biomarker, the score indicative of lung cancer aggression (SILA), in surgically resected stage I LUAD (n=161) histologically graded as indolent low malignant potential (LMP), intermediate, or aggressive vascular invasive (VI) subtypes. The SILA scoring of preoperative computed tomography (CT) scans predicted post-surgical recurrence and resected pathologic grade. Second, to assess molecular and tumor microenvironment features associated with the VI subtype we analyzed resected stage I tumors (n=163) with and without VI by RNA sequencing, including 15 samples by high-resolution spatial transcriptomics. Interestingly, gene expression alterations associated with VI were detectable beyond the invaded vessels, shedding light on changes in the broader tumor microenvironment associated with VI. Moreover, we developed a transcriptomic predictor of VI that was not sensitive to intra-tumor heterogeneity and accurately predicted VI-positive tumors and patient recurrence-free survival. Our findings demonstrate that radiomic and transcriptomic approaches can enhance our ability to predict tumor behavior and may eventually guide preoperative treatment decisions in early-stage LUAD. This may include neoadjuvant therapy and more tailored surgical interventions.
Description
2024