Non-invasive measurement of PD-L1 status and prediction of immunotherapy response using deep learning of PET/CT images

doi:10.1136/jitc-2020-002118

CORC > 自动化研究所 > 中国科学院自动化研究所 > 中国科学院分子影像重点实验室

	Non-invasive measurement of PD-L1 status and prediction of immunotherapy response using deep learning of PET/CT images
	Mu, Wei 1; Jiang, Lei 2; Shi, Yu 3; Tunali, Ilke 1; Gray, Jhanelle E.4; Katsoulakis, Evangelia 5; Tian, Jie 6,7; Gillies, Robert J.1; Schabath, Matthew B.4,8
刊名	JOURNAL FOR IMMUNOTHERAPY OF CANCER
	2021
卷号	9 期号:6 页码:15
关键词	tumor biomarkers immunotherapy
DOI	10.1136/jitc-2020-002118
通讯作者	Tian, Jie(tian@ieee.org) ; Gillies, Robert J.(Robert.Gillies@moffitt.org) ; Schabath, Matthew B.(matthew.schabath@moffitt.org)
英文摘要	Background Currently, only a fraction of patients with non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitors (ICIs) experience a durable clinical benefit (DCB). According to NCCN guidelines, Programmed death-ligand 1 (PD-L1) expression status determined by immunohistochemistry (IHC) of biopsies is the only clinically approved companion biomarker to trigger the use of ICI therapy. Based on prior work showing a relationship between quantitative imaging and gene expression, we hypothesize that quantitative imaging (radiomics) can provide an alternative surrogate for PD-L1 expression status in clinical decision support. Methods F-18-FDG-PET/CT images and clinical data were curated from 697 patients with NSCLC from three institutions and these were analyzed using a small-residual-convolutional-network (SResCNN) to develop a deeply learned score (DLS) to predict the PD-L1 expression status. This developed model was further used to predict DCB, progression-free survival (PFS), and overall survival (OS) in two retrospective and one prospective test cohorts of ICI-treated patients with advanced stage NSCLC. Results The PD-L1 DLS significantly discriminated between PD-L1 positive and negative patients (area under receiver operating characteristics curve >= 0.82 in the training, validation, and two external test cohorts). Importantly, the DLS was indistinguishable from IHC-derived PD-L1 status in predicting PFS and OS, suggesting the utility of DLS as a surrogate for IHC. A score generated by combining the DLS with clinical characteristics was able to accurately (C-indexes of 0.70-0.87) predict DCB, PFS, and OS in retrospective training, prospective testing and external validation cohorts. Conclusion Hence, we propose DLS as a surrogate or substitute for IHC-determined PD-L1 measurement to guide individual pretherapy decisions pending in larger prospective trials.
资助项目	US Public Health Service[U01 CA143062] ; US Public Health Service[R01 CA190105]
WOS关键词	CELL LUNG-CANCER ; EXPRESSION ; BLOCKADE ; MUTATIONS ; ANTIBODY ; DRIVER ; NSCLC ; EGFR ; ALK
WOS研究方向	Oncology ; Immunology
语种	英语
出版者	BMJ PUBLISHING GROUP
WOS记录号	WOS:000662983800002
资助机构	US Public Health Service
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/45346]
专题	自动化研究所_中国科学院分子影像重点实验室
通讯作者	Tian, Jie; Gillies, Robert J.; Schabath, Matthew B.
作者单位	1.H Lee Moffitt Canc Ctr & Res Inst, Dept Canc Physiol, Tampa, FL 33612 USA 2.Tongji Univ, Shanghai Pulm Hosp, Dept Nucl Med, Sch Med, Shanghai, Peoples R China 3.China Med Univ, Dept Radiol, Shengjing Hosp, Shenyang, Peoples R China 4.H Lee Moffitt Canc Ctr & Res Inst, Dept Thorac Oncol, Tampa, FL 33612 USA 5.James A Haley Vet Affairs Med Ctr, Dept Radiat Oncol, Tampa, FL USA 6.Beihang Univ, Beijing Adv Innovat Ctr Big Data Based Precis Med, Sch Engn Med, Beijing, Peoples R China 7.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China 8.H Lee Moffitt Canc Ctr & Res Inst, Dept Canc Epidemiol, Tampa, FL 33612 USA
推荐引用方式 GB/T 7714	Mu, Wei,Jiang, Lei,Shi, Yu,et al. Non-invasive measurement of PD-L1 status and prediction of immunotherapy response using deep learning of PET/CT images[J]. JOURNAL FOR IMMUNOTHERAPY OF CANCER,2021,9(6):15.
APA	Mu, Wei.,Jiang, Lei.,Shi, Yu.,Tunali, Ilke.,Gray, Jhanelle E..,...&Schabath, Matthew B..(2021).Non-invasive measurement of PD-L1 status and prediction of immunotherapy response using deep learning of PET/CT images.JOURNAL FOR IMMUNOTHERAPY OF CANCER,9(6),15.
MLA	Mu, Wei,et al."Non-invasive measurement of PD-L1 status and prediction of immunotherapy response using deep learning of PET/CT images".JOURNAL FOR IMMUNOTHERAPY OF CANCER 9.6(2021):15.