Deep learning using histological images for gene mutation prediction in lung cancer: a multicentre retrospective study - 03/01/25

Doi : 10.1016/S1470-2045(24)00599-0

Yu Zhao, PhD ^a,^b,^{*
Joint first authors, contributed equally}, Shan Xiong, MD ^a,^c,^{*
Joint first authors, contributed equally}, Qin Ren, MS ^b,^{*
Joint first authors, contributed equally}, Jun Wang, MS ^b,^{*
Joint first authors, contributed equally}, Min Li, MD ^d,^{*
Joint first authors, contributed equally}, Lin Yang, MD ^e,^{*
Joint first authors, contributed equally}, Di Wu, MD ^f, Kejing Tang, MD ^g, Xiaojie Pan, MD ^h, Fengxia Chen, MD ⁱ, Wenxiang Wang, MD ^j, Shi Jin, MD ^l, Xianling Liu, MD ^a, Gen Lin, MD ⁿ, Wenxiu Yao, MD ^o, Linbo Cai, MD ^p, Yi Yang, MD ^q, Jixian Liu, MD ^r, Jingxun Wu, MD ^s, Wenfan Fu, MD ^t, Kai Sun, MS ^b, Feng Li, MD ^a, Bo Cheng, MD ^a, Shuting Zhan, MM ^a, Haixuan Wang, MM ^a, Ziwen Yu, MD ^a, Xiwen Liu, MD ^m, Ran Zhong, MD ^a, Huiting Wang, MD ^a, Ping He, MD ^u, Yongmei Zheng, MM ^a, Peng Liang, PhD ^a, Longfei Chen, MD ^v, Ting Hou, MD ^v, Junzhou Huang, PhD ^b, Bing He, PhD ^b, Jiangning Song, PhD ^w, Lin Wu, MD ^k,^{^{†
Joint senior authors, contributed equally}}, Chengping Hu, PhD ^d,^{^{†
Joint senior authors, contributed equally}}, Jianxing He, ProfPhD ^a,^{^{†
Joint senior authors, contributed equally}}, Jianhua Yao, PhD ^b,^{^{†
Joint senior authors, contributed equally}}, Wenhua Liang, ProfPhD ^a,^c,^{^{†
Joint senior authors, contributed equally},}⁎

^a Department of Thoracic Oncology and Surgery, The First Affiliated Hospital of Guangzhou Medical University, State Key Laboratory of Respiratory Disease & National Clinical Research Center for Respiratory Disease, Guangzhou, China

^b AI Lab, Tencent, Shenzhen, China

^c Department of Thoracic Oncology and Surgery, Hengqin Hospital, The First Affiliated Hospital of Guangzhou Medical University, Hengqin, China

^d Department of Respiratory Medicine, National Key Clinical Specialty, Branch of National Clinical Research Center for Respiratory Disease, Xiangya Hospital, Central South University, Changsha, China

^e Department of Thoracic Surgery, Shenzhen People’s Hospital, 2nd Clinical Medical College of Jinan University, Shenzhen, China

^f Department of Respiratory Medicine, Shenzhen People’s Hospital, 2nd Clinical Medical College of Jinan University, Shenzhen, China

^g Division of Pulmonary and Critical Care Medicine, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China

^h Department of Thoracic Surgery, Shengli Clinical Medical College of Fujian Medical University, Fujian Provincial Hospital, Fuzhou, China

ⁱ Department of Thoracic Surgery, Hainan General Hospital, Haikou, China

^j Thoracic Surgery Department 2, Hunan Cancer Hospital and The Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University, Changsha, China

^k Department of Thoracic Medical Oncology, Hunan Cancer Hospital and The Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University, Changsha, China

^l Department of Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen, China

^m Department of Oncology, The Second Xiangya Hospital of Central South University, Changsha, China

ⁿ Department of Thoracic Oncology, Clinical Oncology School of Fujian Medical University, Fujian Cancer Hospital, Fujian Key Laboratory of Advanced Technology for Cancer Screening and Early Diagnosis, Fuzhou, China

^o Department of Oncology, University of Electronic Science and Technology of China, Sichuan Cancer Hospital and Institute & Cancer, The Second People’s Hospital of Sichuan Province, Chengdu, China

^p Department of Oncology, Guangdong Sanjiu Brain Hospital, Guangzhou, China

^q Department of Thoracic Surgery, Chengdu Third People’s Hospital, Affiliated Hospital of Southwest Jiaotong University, Chengdu, China

^r Department of Thoracic Surgery, Peking University Shenzhen Hospital, Shenzhen Peking University-The Hong Kong University of Science and Technology Medical Center, Shenzhen, China

^s Department of Medical Oncology, The First Affiliated Hospital of Xiamen University, Xiamen, China

^t Department of Chest Surgery, Affiliated Cancer Hospital & Institute of Guangzhou Medical University, Guangzhou, China

^u Department of Pathology, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China

^v Burning Rock Biotech, Guangzhou, China

^w Biomedicine Discovery Institute and Monash Data Futures Institute, Monash University, Melbourne, VIC, Australia

^* Correspondence to: Prof Wenhua Liang, Department of Thoracic Oncology and Surgery, The First Affiliated Hospital of Guangzhou Medical University, State Key Laboratory of Respiratory Disease & National Clinical Research Center for Respiratory Disease, 510120, Guangzhou, China Department of Thoracic Oncology and Surgery The First Affiliated Hospital of Guangzhou Medical University State Key Laboratory of Respiratory Disease & National Clinical Research Center for Respiratory Disease Guangzhou 510120 China

Summary

Background

Accurate detection of driver gene mutations is crucial for treatment planning and predicting prognosis for patients with lung cancer. Conventional genomic testing requires high-quality tissue samples and is time-consuming and resource-consuming, and as a result, is not available for most patients, especially those in low-resource settings. We aimed to develop an annotation-free Deep learning-enabled artificial intelligence method to predict GEne Mutations (DeepGEM) from routinely acquired histological slides.

Methods

In this multicentre retrospective study, we collected data for patients with lung cancer who had a biopsy and multigene next-generation sequencing done at 16 hospitals in China (with no restrictions on age, sex, or histology type), to form a large multicentre dataset comprising paired pathological image and multiple gene mutation information. We also included patients from The Cancer Genome Atlas (TCGA) publicly available dataset. Our developed model is an instance-level and bag-level co-supervised multiple instance learning method with label disambiguation design. We trained and initially tested the DeepGEM model on the internal dataset (patients from the First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China), and further evaluated it on the external dataset (patients from the remaining 15 centres) and the public TCGA dataset. Additionally, a dataset of patients from the same medical centre as the internal dataset, but without overlap, was used to evaluate the model’s generalisation ability to biopsy samples from lymph node metastases. The primary objective was the performance of the DeepGEM model in predicting gene mutations (area under the curve [AUC] and accuracy) in the four prespecified groups (ie, the hold-out internal test set, multicentre external test set, TCGA set, and lymph node metastases set).

Findings

Assessable pathological images and multigene testing information were available for 3697 patients who had biopsy and multigene next-generation sequencing done between Jan 1, 2018, and March 31, 2022, at the 16 centres. We excluded 60 patients with low-quality images. We included 3767 images from 3637 consecutive patients (1978 [54·4%] men, 1514 [41·6%] women, 145 [4·0%] unknown; median age 60 years [IQR 52–67]), with 1716 patients in the internal dataset, 1718 patients in the external dataset, and 203 patients in the lymph node metastases dataset. The DeepGEM model showed robust performance in the internal dataset: for excisional biopsy samples, AUC values for gene mutation prediction ranged from 0·90 (95% CI 0·77–1·00) to 0·97 (0·93–1·00) and accuracy values ranged from 0·91 (0·85–0·98) to 0·97 (0·93–1·00); for aspiration biopsy samples, AUC values ranged from 0·85 (0·80–0·91) to 0·95 (0·86–1·00) and accuracy values ranged from 0·79 (0·74–0·85) to 0·99 (0·98–1·00). In the multicentre external dataset, for excisional biopsy samples, AUC values ranged from 0·80 (95% CI 0·75–0·85) to 0·91 (0·88–1·00) and accuracy values ranged from 0·79 (0·76–0·82) to 0·95 (0·93–0·96); for aspiration biopsy samples, AUC values ranged from 0·76 (0·70–0·83) to 0·87 (0·80–0·94) and accuracy values ranged from 0·76 (0·74–0·79) to 0·97 (0·96–0·98). The model also showed strong performance on the TCGA dataset (473 patients; 535 slides; AUC values ranged from 0·82 [95% CI 0·71–0·93] to 0·96 [0·91–1·00], accuracy values ranged from 0·79 [0·70–0·88] to 0·95 [0·90–1·00]). The DeepGEM model, trained on primary region biopsy samples, could be generalised to biopsy samples from lymph node metastases, with AUC values of 0·91 (95% CI 0·88–0·94) for EGFR and 0·88 (0·82–0·93) for KRAS and accuracy values of 0·85 (0·80–0·88) for EGFR and 0·95 (0·92–0·96) for KRAS and showed potential for prognostic prediction of targeted therapy. The model generated spatial gene mutation maps, indicating gene mutation spatial distribution.

Interpretation

We developed an AI-based method that can provide an accurate, timely, and economical prediction of gene mutation and mutation spatial distribution. The method showed substantial potential as an assistive tool for guiding the clinical treatment of patients with lung cancer.

Funding

National Natural Science Foundation of China, the Science and Technology Planning Project of Guangzhou, and the National Key Research and Development Program of China.

Translation

For the Chinese translation of the abstract see Supplementary Materials section.

Le texte complet de cet article est disponible en PDF.

Plan

Introduction

Methods

Study design and participants

Procedures

Outcomes

Statistical analysis

Role of the funding source

Export

Vol 26 - N° 1

P. 136-146 - janvier 2025 Retour au numéro

Article précédent

The landscape of primary mismatch repair deficient gliomas in children, adolescents, and young adults: a multi-cohort study
Logine Negm, Jiil Chung, Liana Nobre, Julie Bennett, Nicholas R Fernandez, Nuno Miguel Nunes, Zhihui Amy Liu, Martin Komosa, Melyssa Aronson, Cindy Zhang, Lucie Stengs, Vanessa Bianchi, Melissa Edwards, Sheradan Doherty, Ayse Bahar Ercan, Maria F Cardenas, Michael Macias, Matthew R Lueder, Michelle Ku, Monique Johnson, Yuan Chang, Jose Rafael Dimayacyac, Adam A Kraya, Yiran Guo, Stav Naky, Julia Keith, Andrew F Gao, David G Munoz, Lananh Nguyen, Derek S Tsang, Mary Jane Lim-Fat, Sunit Das, Adam Shlien, Vijay Ramaswamy, Annie Huang, David Malkin, Anita Villani, Birgit Ertl-Wagner, Adrian Levine, Giles W Robinson, Brad H Pollock, Logan G Spector, Shizuko Sei, Peter B Dirks, Gad Getz, Kim E Nichols, Adam C Resnick, David A Wheeler, Anirban Das, Yosef E Maruvka, Cynthia Hawkins, Uri Tabori

| Article suivant

Pulmonary neuroendocrine neoplasms: the molecular landscape, therapeutic challenges, and diagnosis and management strategies
Triparna Sen, Yosuke Dotsu, Virginia Corbett, Sonam Puri, Utsav Sen, Theresa A Boyle, Phil Mack, Fred Hirsch, Raid Aljumaily, Abdul Rafeh Naqash, Vineeth Sukrithan, Nagla Abdel Karim

Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.

Déjà abonné à cette revue ?

connectez-vous ou créez un compte

Deep learning using histological images for gene mutation prediction in lung cancer: a multicentre retrospective study - 03/01/25

Summary

Background

Methods

Findings

Interpretation

Funding

Translation

Plan

Export citations

Fichier

Contenu

Accès rapides

Mon compte

Aide & support

Plateformes Elsevier Masson

Déclaration CNIL