A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer
Background: Tumor-infiltrating lymphocytes (TILs) are clinically significant in triple-negative breast cancer (TNBC). Although a standardized methodology for visual TILs assessment (VTA) exists, it has several inherent limitations. We established a deep learning-based computational TIL assessment (C...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-08-01
|
| Series: | EBioMedicine |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352396421002851 |
| _version_ | 1818652139167154176 |
|---|---|
| author | Peng Sun Jiehua He Xue Chao Keming Chen Yuanyuan Xu Qitao Huang Jingping Yun Mei Li Rongzhen Luo Jinbo Kuang Huajia Wang Haosen Li Hui Hui Shuoyu Xu |
| author_facet | Peng Sun Jiehua He Xue Chao Keming Chen Yuanyuan Xu Qitao Huang Jingping Yun Mei Li Rongzhen Luo Jinbo Kuang Huajia Wang Haosen Li Hui Hui Shuoyu Xu |
| author_sort | Peng Sun |
| collection | DOAJ |
| description | Background: Tumor-infiltrating lymphocytes (TILs) are clinically significant in triple-negative breast cancer (TNBC). Although a standardized methodology for visual TILs assessment (VTA) exists, it has several inherent limitations. We established a deep learning-based computational TIL assessment (CTA) method broadly following VTA guideline and compared it with VTA for TNBC to determine the prognostic value of the CTA and a reasonable CTA workflow for clinical practice. Methods: We trained three deep neural networks for nuclei segmentation, nuclei classification and necrosis classification to establish a CTA workflow. The automatic TIL (aTIL) score generated was compared with manual TIL (mTIL) scores provided by three pathologists in an Asian (n = 184) and a Caucasian (n = 117) TNBC cohort to evaluate scoring concordance and prognostic value. Findings: The intraclass correlations (ICCs) between aTILs and mTILs varied from 0.40 to 0.70 in two cohorts. Multivariate Cox proportional hazards analysis revealed that the aTIL score was associated with disease free survival (DFS) in both cohorts, as either a continuous [hazard ratio (HR)=0.96, 95% CI 0.94–0.99] or dichotomous variable (HR=0.29, 95% CI 0.12–0.72). A higher C-index was observed in a composite mTIL/aTIL three-tier stratification model than in the dichotomous model, using either mTILs or aTILs alone. Interpretation: The current study provides a useful tool for stromal TIL assessment and prognosis evaluation for patients with TNBC. A workflow integrating both VTA and CTA may aid pathologists in performing risk management and decision-making tasks. |
| first_indexed | 2024-12-17T02:17:15Z |
| format | Article |
| id | doaj.art-52ea8a3555af41e19f6956cb71b42f2c |
| institution | Directory Open Access Journal |
| issn | 2352-3964 |
| language | English |
| last_indexed | 2024-12-17T02:17:15Z |
| publishDate | 2021-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | EBioMedicine |
| spelling | doaj.art-52ea8a3555af41e19f6956cb71b42f2c2022-12-21T22:07:22ZengElsevierEBioMedicine2352-39642021-08-0170103492A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast CancerPeng Sun0Jiehua He1Xue Chao2Keming Chen3Yuanyuan Xu4Qitao Huang5Jingping Yun6Mei Li7Rongzhen Luo8Jinbo Kuang9Huajia Wang10Haosen Li11Hui Hui12Shuoyu Xu13State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. China; Corresponding authors: Dr. Peng Sun, Department of Pathology, Sun Yat-sen University Cancer Center, 651 Dongfeng East Road, 510060, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaDepartment of Physiology, Zhongshan school of Medicine, Sun Yat-Sen University, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaState Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, P. R. China; Department of Pathology, Sun Yat-sen University Cancer Center, Guangzhou, P. R. ChinaBio-totem Pte Ltd, Foshan, P. R. ChinaBio-totem Pte Ltd, Foshan, P. R. ChinaBio-totem Pte Ltd, Foshan, P. R. ChinaBio-totem Pte Ltd, Foshan, P. R. ChinaBio-totem Pte Ltd, Foshan, P. R. China; Department of General Surgery, Nanfang Hospital, Southern Medical University, Guangzhou, P. R. China; Co-coresponding author: Dr. Shuoyu Xu: Bio-totem Pte Ltd, 321 State Road, 528231, Foshan, P.R.China; Department of General Surgery Nanfang Hospital, Southern Medical University, 1838 Guangzhou Avenue North, 510515, Guangzhou, P.R.China.; 1838 Guangzhou Avenue North, 510515, Guangzhou, P. R. ChinaBackground: Tumor-infiltrating lymphocytes (TILs) are clinically significant in triple-negative breast cancer (TNBC). Although a standardized methodology for visual TILs assessment (VTA) exists, it has several inherent limitations. We established a deep learning-based computational TIL assessment (CTA) method broadly following VTA guideline and compared it with VTA for TNBC to determine the prognostic value of the CTA and a reasonable CTA workflow for clinical practice. Methods: We trained three deep neural networks for nuclei segmentation, nuclei classification and necrosis classification to establish a CTA workflow. The automatic TIL (aTIL) score generated was compared with manual TIL (mTIL) scores provided by three pathologists in an Asian (n = 184) and a Caucasian (n = 117) TNBC cohort to evaluate scoring concordance and prognostic value. Findings: The intraclass correlations (ICCs) between aTILs and mTILs varied from 0.40 to 0.70 in two cohorts. Multivariate Cox proportional hazards analysis revealed that the aTIL score was associated with disease free survival (DFS) in both cohorts, as either a continuous [hazard ratio (HR)=0.96, 95% CI 0.94–0.99] or dichotomous variable (HR=0.29, 95% CI 0.12–0.72). A higher C-index was observed in a composite mTIL/aTIL three-tier stratification model than in the dichotomous model, using either mTILs or aTILs alone. Interpretation: The current study provides a useful tool for stromal TIL assessment and prognosis evaluation for patients with TNBC. A workflow integrating both VTA and CTA may aid pathologists in performing risk management and decision-making tasks.http://www.sciencedirect.com/science/article/pii/S2352396421002851Triple-negative breast cancerTumor-infiltrating lymphocyteDeep learningPrognosis |
| spellingShingle | Peng Sun Jiehua He Xue Chao Keming Chen Yuanyuan Xu Qitao Huang Jingping Yun Mei Li Rongzhen Luo Jinbo Kuang Huajia Wang Haosen Li Hui Hui Shuoyu Xu A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer EBioMedicine Triple-negative breast cancer Tumor-infiltrating lymphocyte Deep learning Prognosis |
| title | A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer |
| title_full | A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer |
| title_fullStr | A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer |
| title_full_unstemmed | A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer |
| title_short | A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer |
| title_sort | computational tumor infiltrating lymphocyte assessment method comparable with visual reporting guidelines for triple negative breast cancer |
| topic | Triple-negative breast cancer Tumor-infiltrating lymphocyte Deep learning Prognosis |
| url | http://www.sciencedirect.com/science/article/pii/S2352396421002851 |
| work_keys_str_mv | AT pengsun acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT jiehuahe acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT xuechao acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT kemingchen acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT yuanyuanxu acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT qitaohuang acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT jingpingyun acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT meili acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT rongzhenluo acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT jinbokuang acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT huajiawang acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT haosenli acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT huihui acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT shuoyuxu acomputationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT pengsun computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT jiehuahe computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT xuechao computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT kemingchen computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT yuanyuanxu computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT qitaohuang computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT jingpingyun computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT meili computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT rongzhenluo computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT jinbokuang computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT huajiawang computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT haosenli computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT huihui computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer AT shuoyuxu computationaltumorinfiltratinglymphocyteassessmentmethodcomparablewithvisualreportingguidelinesfortriplenegativebreastcancer |