A prospective, interventional, single center study to correlate outcomes using OncoPredikt, an (AI-enabled platform) to predict genomic signatures from histopathological images in lung adenocarcinoma patients.

This study focuses on the application of the OncoPredikt AI platform for predicting and correlating genomic signatures in lung cancer from histopathological lung biopsy slides. Lung cancer, the most commonly reported cancer worldwide, accounted for approximately 2.2 million new cases and 1.8 million deaths in 2020. Molecular classification of lung cancer, particularly the detection of specific mutations in genes like EGFR, BRAF, KRAS, and others, has advanced targeted therapeutic strategies. The study design is a prospective, interventional, single-center approach with a sample size of 52 cases and having study duration for patient recruitement is  12 months. It includes patients with confirmed lung cancer diagnoses and available Next-Generation Sequencing (NGS) reports, specifically for those with or without mutations in a set of key genes. The exclusion criterion focuses on patients without lung adenocarcinoma with a biomarker. The study will record various data parameters, including demographic details, cancer stage at diagnosis, mutation status, treatment details (chemotherapy, surgery, targeted therapy, radiotherapy, and immunotherapy), and the interpretation from the OncoPredikt AI model. The primary objective is to evaluate the accuracy of OncoPredikt in predicting lung panel gene signature mutations using diagnostic H&E slides, potentially offering a quicker and more cost-effective screening tool, even when tissue availability is limited.