DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data
Abstract Background Identifying breast cancer patients with DNA repair pathway-related germline pathogenic variants (GPVs) is important for effectively employing systemic treatment strategies and risk-reducing interventions. However, current criteria and risk prediction models for prioritizing genet...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2022-02-01
|
| Series: | Genome Medicine |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13073-022-01027-9 |
| _version_ | 1818501349310988288 |
|---|---|
| author | Jiaqi Liu Hengqiang Zhao Yu Zheng Lin Dong Sen Zhao Yukuan Huang Shengkai Huang Tianyi Qian Jiali Zou Shu Liu Jun Li Zihui Yan Yalun Li Shuo Zhang Xin Huang Wenyan Wang Yiqun Li Jie Wang Yue Ming Xiaoxin Li Zeyu Xing Ling Qin Zhengye Zhao Ziqi Jia Jiaxin Li Gang Liu Menglu Zhang Kexin Feng Jiang Wu Jianguo Zhang Yongxin Yang Zhihong Wu Zhihua Liu Jianming Ying Xin Wang Jianzhong Su Xiang Wang Nan Wu |
| author_facet | Jiaqi Liu Hengqiang Zhao Yu Zheng Lin Dong Sen Zhao Yukuan Huang Shengkai Huang Tianyi Qian Jiali Zou Shu Liu Jun Li Zihui Yan Yalun Li Shuo Zhang Xin Huang Wenyan Wang Yiqun Li Jie Wang Yue Ming Xiaoxin Li Zeyu Xing Ling Qin Zhengye Zhao Ziqi Jia Jiaxin Li Gang Liu Menglu Zhang Kexin Feng Jiang Wu Jianguo Zhang Yongxin Yang Zhihong Wu Zhihua Liu Jianming Ying Xin Wang Jianzhong Su Xiang Wang Nan Wu |
| author_sort | Jiaqi Liu |
| collection | DOAJ |
| description | Abstract Background Identifying breast cancer patients with DNA repair pathway-related germline pathogenic variants (GPVs) is important for effectively employing systemic treatment strategies and risk-reducing interventions. However, current criteria and risk prediction models for prioritizing genetic testing among breast cancer patients do not meet the demands of clinical practice due to insufficient accuracy. Methods The study population comprised 3041 breast cancer patients enrolled from seven hospitals between October 2017 and 11 August 2019, who underwent germline genetic testing of 50 cancer predisposition genes (CPGs). Associations among GPVs in different CPGs and endophenotypes were evaluated using a case-control analysis. A phenotype-based GPV risk prediction model named DNA-repair Associated Breast Cancer (DrABC) was developed based on hierarchical neural network architecture and validated in an independent multicenter cohort. The predictive performance of DrABC was compared with currently used models including BRCAPRO, BOADICEA, Myriad, PENN II, and the NCCN criteria. Results In total, 332 (11.3%) patients harbored GPVs in CPGs, including 134 (4.6%) in BRCA2, 131 (4.5%) in BRCA1, 33 (1.1%) in PALB2, and 37 (1.3%) in other CPGs. GPVs in CPGs were associated with distinct endophenotypes including the age at diagnosis, cancer history, family cancer history, and pathological characteristics. We developed a DrABC model to predict the risk of GPV carrier status in BRCA1/2 and other important CPGs. In predicting GPVs in BRCA1/2, the performance of DrABC (AUC = 0.79 [95% CI, 0.74–0.85], sensitivity = 82.1%, specificity = 63.1% in the independent validation cohort) was better than that of previous models (AUC range = 0.57–0.70). In predicting GPVs in any CPG, DrABC (AUC = 0.74 [95% CI, 0.69–0.79], sensitivity = 83.8%, specificity = 51.3% in the independent validation cohort) was also superior to previous models in their current versions (AUC range = 0.55–0.65). After training these previous models with the Chinese-specific dataset, DrABC still outperformed all other methods except for BOADICEA, which was the only previous model with the inclusion of pathological features. The DrABC model also showed higher sensitivity and specificity than the NCCN criteria in the multi-center validation cohort (83.8% and 51.3% vs. 78.8% and 31.2%, respectively, in predicting GPVs in any CPG). The DrABC model implementation is available online at http://gifts.bio-data.cn/ . Conclusions By considering the distinct endophenotypes associated with different CPGs in breast cancer patients, a phenotype-driven prediction model based on hierarchical neural network architecture was created for identification of hereditary breast cancer. The model achieved superior performance in identifying GPV carriers among Chinese breast cancer patients. |
| first_indexed | 2024-12-10T20:54:55Z |
| format | Article |
| id | doaj.art-60de1be1db0e43cdb084be6bde5c2274 |
| institution | Directory Open Access Journal |
| issn | 1756-994X |
| language | English |
| last_indexed | 2024-12-10T20:54:55Z |
| publishDate | 2022-02-01 |
| publisher | BMC |
| record_format | Article |
| series | Genome Medicine |
| spelling | doaj.art-60de1be1db0e43cdb084be6bde5c22742022-12-22T01:34:00ZengBMCGenome Medicine1756-994X2022-02-0114111510.1186/s13073-022-01027-9DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype dataJiaqi Liu0Hengqiang Zhao1Yu Zheng2Lin Dong3Sen Zhao4Yukuan Huang5Shengkai Huang6Tianyi Qian7Jiali Zou8Shu Liu9Jun Li10Zihui Yan11Yalun Li12Shuo Zhang13Xin Huang14Wenyan Wang15Yiqun Li16Jie Wang17Yue Ming18Xiaoxin Li19Zeyu Xing20Ling Qin21Zhengye Zhao22Ziqi Jia23Jiaxin Li24Gang Liu25Menglu Zhang26Kexin Feng27Jiang Wu28Jianguo Zhang29Yongxin Yang30Zhihong Wu31Zhihua Liu32Jianming Ying33Xin Wang34Jianzhong Su35Xiang Wang36Nan Wu37Department of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesFintech Innovation Center, Southwestern University of Finance and EconomicsDepartment of Pathology, National Cancer Center /National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesInstitute of Biomedical Big Data, Wenzhou Medical UniversityDepartment of Laboratory Medicine, National Cancer Center /National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgery, Guiyang Maternal and Child Healthcare HospitalDepartment of Breast Surgery, the Affiliated Hospital of Guizhou Medical UniversityDepartment of Molecular Pathology, the Affiliated Cancer Hospital of Zhengzhou UniversityDepartment of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesDepartment of Breast Surgery, the Affiliated Yantai Yuhuangding Hospital of Qingdao UniversityDepartment of Breast Surgery, the Fourth Hospital of Hebei Medical UniversityDepartment of Breast Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesDepartment of Breast Surgery, Beijing Tiantan Hospital, Capital Medical UniversityDepartment of Oncology, National Cancer Center /National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Ultrasound, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegePET-CT Center, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeMedical Research Center, Beijing Key Laboratory for Genetic Research of Skeletal Deformity & Key Laboratory of Big Data for Spinal Deformities, All at Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, Cancer Hospital of HuanXingDepartment of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesMachine Intelligence Group, University of EdinburghBeijing Key Laboratory for Genetic Research of Skeletal Deformity, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesState Key Laboratory of Molecular Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Pathology, National Cancer Center /National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeInstitute of Biomedical Big Data, Wenzhou Medical UniversityDepartment of Breast Surgical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical CollegeDepartment of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical SciencesAbstract Background Identifying breast cancer patients with DNA repair pathway-related germline pathogenic variants (GPVs) is important for effectively employing systemic treatment strategies and risk-reducing interventions. However, current criteria and risk prediction models for prioritizing genetic testing among breast cancer patients do not meet the demands of clinical practice due to insufficient accuracy. Methods The study population comprised 3041 breast cancer patients enrolled from seven hospitals between October 2017 and 11 August 2019, who underwent germline genetic testing of 50 cancer predisposition genes (CPGs). Associations among GPVs in different CPGs and endophenotypes were evaluated using a case-control analysis. A phenotype-based GPV risk prediction model named DNA-repair Associated Breast Cancer (DrABC) was developed based on hierarchical neural network architecture and validated in an independent multicenter cohort. The predictive performance of DrABC was compared with currently used models including BRCAPRO, BOADICEA, Myriad, PENN II, and the NCCN criteria. Results In total, 332 (11.3%) patients harbored GPVs in CPGs, including 134 (4.6%) in BRCA2, 131 (4.5%) in BRCA1, 33 (1.1%) in PALB2, and 37 (1.3%) in other CPGs. GPVs in CPGs were associated with distinct endophenotypes including the age at diagnosis, cancer history, family cancer history, and pathological characteristics. We developed a DrABC model to predict the risk of GPV carrier status in BRCA1/2 and other important CPGs. In predicting GPVs in BRCA1/2, the performance of DrABC (AUC = 0.79 [95% CI, 0.74–0.85], sensitivity = 82.1%, specificity = 63.1% in the independent validation cohort) was better than that of previous models (AUC range = 0.57–0.70). In predicting GPVs in any CPG, DrABC (AUC = 0.74 [95% CI, 0.69–0.79], sensitivity = 83.8%, specificity = 51.3% in the independent validation cohort) was also superior to previous models in their current versions (AUC range = 0.55–0.65). After training these previous models with the Chinese-specific dataset, DrABC still outperformed all other methods except for BOADICEA, which was the only previous model with the inclusion of pathological features. The DrABC model also showed higher sensitivity and specificity than the NCCN criteria in the multi-center validation cohort (83.8% and 51.3% vs. 78.8% and 31.2%, respectively, in predicting GPVs in any CPG). The DrABC model implementation is available online at http://gifts.bio-data.cn/ . Conclusions By considering the distinct endophenotypes associated with different CPGs in breast cancer patients, a phenotype-driven prediction model based on hierarchical neural network architecture was created for identification of hereditary breast cancer. The model achieved superior performance in identifying GPV carriers among Chinese breast cancer patients.https://doi.org/10.1186/s13073-022-01027-9Hereditary breast cancerDeep learningBRCA1/2Genetic testGenotype-phenotype correlation |
| spellingShingle | Jiaqi Liu Hengqiang Zhao Yu Zheng Lin Dong Sen Zhao Yukuan Huang Shengkai Huang Tianyi Qian Jiali Zou Shu Liu Jun Li Zihui Yan Yalun Li Shuo Zhang Xin Huang Wenyan Wang Yiqun Li Jie Wang Yue Ming Xiaoxin Li Zeyu Xing Ling Qin Zhengye Zhao Ziqi Jia Jiaxin Li Gang Liu Menglu Zhang Kexin Feng Jiang Wu Jianguo Zhang Yongxin Yang Zhihong Wu Zhihua Liu Jianming Ying Xin Wang Jianzhong Su Xiang Wang Nan Wu DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data Genome Medicine Hereditary breast cancer Deep learning BRCA1/2 Genetic test Genotype-phenotype correlation |
| title | DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data |
| title_full | DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data |
| title_fullStr | DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data |
| title_full_unstemmed | DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data |
| title_short | DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data |
| title_sort | drabc deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data |
| topic | Hereditary breast cancer Deep learning BRCA1/2 Genetic test Genotype-phenotype correlation |
| url | https://doi.org/10.1186/s13073-022-01027-9 |
| work_keys_str_mv | AT jiaqiliu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT hengqiangzhao drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT yuzheng drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT lindong drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT senzhao drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT yukuanhuang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT shengkaihuang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT tianyiqian drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jializou drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT shuliu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT junli drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT zihuiyan drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT yalunli drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT shuozhang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT xinhuang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT wenyanwang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT yiqunli drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jiewang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT yueming drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT xiaoxinli drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT zeyuxing drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT lingqin drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT zhengyezhao drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT ziqijia drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jiaxinli drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT gangliu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT mengluzhang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT kexinfeng drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jiangwu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jianguozhang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT yongxinyang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT zhihongwu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT zhihualiu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jianmingying drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT xinwang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT jianzhongsu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT xiangwang drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata AT nanwu drabcdeeplearningaccuratelypredictsgermlinepathogenicmutationstatusinbreastcancerpatientsbasedonphenotypedata |