{"doi":"10.1101/2025.08.14.670178","title":"Genomic Characterization of Lung Cancer in Never-Smokers Using Deep Learning","abstract":"Abstract Despite promising results in using deep learning to infer genetic features from histological whole-slide images (WSIs), no prior studies have specifically applied these methods to lung adenocarcinomas from subjects who have never smoked tobacco (NS-LUAD) – a molecularly and histologically distinct subset of lung cancer. Existing models have focused on LUAD from predominantly smoker populations, with limited molecular scope and variable performance. Here, we propose a customized deep convolutional neural network based on ResNet50 architecture, optimized for multilabel classification for NS-LUAD, enabling simultaneous prediction of 16 molecular alterations from a single H&E-stained WSI. Key architectural modifications included a simplified two-layer residual block without bottleneck layers, selective shortcut connections, and a sigmoid-based classification head for independent prediction of each alteration, designed to reduce computational complexity while maintaining predictive accuracy. The model was trained and evaluated on 495 WSIs from the Sherlock- Lung study (70% training with 10% internal test set for 10-fold cross-validation, and 30% held-out validation set for final evaluation). For the held-out validation data, our model achieved high areas under the receiver operating characteristic curve [AUROC] values =0.84-0.93 for detecting 11 features: EGFR, KRAS, TP53, RBM10 mutations, MDM2 amplification, kataegis, CDKN2A deletion, ALK fusion, whole-genome doubling, and EGFR hotspot mutations (p.L858R and p.E746_A750del). Performance was low to moderate for tumor mutational burden (AUROC=0.67), APOBEC mutational signature (AUROC=0.57), and KRAS hotspot mutations (p.G12C: AUROC=0.74, p.G12V: AUROC=0.55, p.G12D: AUROC=0.43). Compared to results from established architectures such as Inception-v3 on the same WSIs, our model demonstrated significantly improved performance for most features. With further optimization, our model could support triaging for molecular testing and inform precision treatment strategies for NS-LUAD patients.","journal":"bioRxiv (Cold Spring Harbor Laboratory)","year":2025,"id":572773,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":0,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9508,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1376885,"name":"Thi‐Van‐Trinh Tran","orcid":"0000-0003-3132-9055","position":1,"is_corresponding":false},{"id":622412,"name":"Praphulla Bhawsar","orcid":null,"position":2,"is_corresponding":false},{"id":240367,"name":"Tongwu Zhang","orcid":"0000-0003-2124-2706","position":3,"is_corresponding":false},{"id":1328507,"name":"Wei Zhao","orcid":"0000-0002-1830-338X","position":4,"is_corresponding":false},{"id":621123,"name":"Phuc H. Hoang","orcid":"0000-0002-9265-8525","position":5,"is_corresponding":false},{"id":291444,"name":"Karun Mutreja","orcid":"0009-0007-0712-7257","position":6,"is_corresponding":false},{"id":291443,"name":"Scott M. Lawrence","orcid":"0000-0002-5297-7978","position":7,"is_corresponding":false},{"id":360915,"name":"Nathaniel Rothman","orcid":"0000-0002-0866-9943","position":8,"is_corresponding":false},{"id":1141558,"name":"Qing Lan","orcid":"0000-0001-7791-5020","position":9,"is_corresponding":false},{"id":41685,"name":"Robert Homer","orcid":"0000-0002-2055-5885","position":10,"is_corresponding":false},{"id":616303,"name":"Marina K. Baine","orcid":"0000-0003-4939-6491","position":11,"is_corresponding":false},{"id":74587,"name":"Lynette M. Sholl","orcid":"0000-0002-9532-9735","position":12,"is_corresponding":false},{"id":353891,"name":"Philippe Joubert","orcid":"0000-0001-7784-6387","position":13,"is_corresponding":false},{"id":683989,"name":"Charles Leduc","orcid":"0000-0001-9477-1479","position":14,"is_corresponding":false},{"id":15536,"name":"William D. Travis","orcid":"0000-0003-3160-6729","position":15,"is_corresponding":false},{"id":6431,"name":"Stephen J. Chanock","orcid":"0000-0002-2324-3393","position":16,"is_corresponding":false},{"id":5285,"name":"Jianxin Shi","orcid":"0000-0001-8606-4707","position":17,"is_corresponding":false},{"id":693260,"name":"Soo-Ryum Yang","orcid":null,"position":18,"is_corresponding":false},{"id":272280,"name":"Jonas S. Almeida","orcid":"0000-0002-7883-7922","position":19,"is_corresponding":false},{"id":240362,"name":"Maria Teresa Landi","orcid":"0000-0003-4507-329X","position":20,"is_corresponding":false},{"id":341710,"name":"Monjoy Saha","orcid":"0000-0002-6470-3544","position":0,"is_corresponding":true}],"reference_count":38,"raw_metadata":{"citation_network_status":"fetched"},"created_at":"2026-07-19T02:57:27.876396Z","pmid":"40894597","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}