학술논문
Multimodal integration of radiology, pathology and genomics for prediction of response to PD-(L)1 blockade in patients with non-small cell lung cancer
Document Type
Original Paper
Author
Vanguri, Rami S.; Luo, Jia; Aukerman, Andrew T.; Egger, Jacklynn V.; Fong, Christopher J.; Horvat, Natally; Pagano, Andrew; Araujo-Filho, Jose de Arimateia Batista; Geneslaw, Luke; Rizvi, Hira; Sosa, Ramon; Boehm, Kevin M.; Yang, Soo-Ryum; Bodd, Francis M.; Ventura, Katia; Hollmann, Travis J.; Ginsberg, Michelle S.; Gao, Jianjiong; Hellmann, Matthew D.; Sauter, Jennifer L.; Shah, Sohrab P.
Source
Nature Cancer. 3(10):1151-1164
Subject
Language
English
ISSN
2662-1347
Abstract
Immunotherapy is used to treat almost all patients with advanced non-small cell lung cancer (NSCLC); however, identifying robust predictive biomarkers remains challenging. Here we show the predictive capacity of integrating medical imaging, histopathologic and genomic features to predict immunotherapy response using a cohort of 247 patients with advanced NSCLC with multimodal baseline data obtained during diagnostic clinical workup, including computed tomography scan images, digitized programmed death ligand-1 immunohistochemistry slides and known outcomes to immunotherapy. Using domain expert annotations, we developed a computational workflow to extract patient-level features and used a machine-learning approach to integrate multimodal features into a risk prediction model. Our multimodal model (area under the curve (AUC) = 0.80, 95% confidence interval (CI) 0.74–0.86) outperformed unimodal measures, including tumor mutational burden (AUC = 0.61, 95% CI 0.52–0.70) and programmed death ligand-1 immunohistochemistry score (AUC = 0.73, 95% CI 0.65–0.81). Our study therefore provides a quantitative rationale for using multimodal features to improve prediction of immunotherapy response in patients with NSCLC using expert-guided machine learning.
Shah and colleagues used machine learning to integrate features from computed tomography scan images, programmed death ligand-1 (PD-L1) immunohistochemistry and genomics into a multimodal predictor of response to anti-PD(L)1 in patients with NSCLC that outperformed single features such as tumor mutational burden.
Shah and colleagues used machine learning to integrate features from computed tomography scan images, programmed death ligand-1 (PD-L1) immunohistochemistry and genomics into a multimodal predictor of response to anti-PD(L)1 in patients with NSCLC that outperformed single features such as tumor mutational burden.