Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks
Abstract Background Ovarian cancer causes 151,900 deaths per year worldwide. Treatment and prognosis are primarily determined by the histopathologic interpretation in combination with molecular diagnosis. However, the relationship between histopathology patterns and molecular alterations is not full...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2020-08-01
|
| Series: | BMC Medicine |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12916-020-01684-w |
| _version_ | 1818450144925843456 |
|---|---|
| author | Kun-Hsing Yu Vincent Hu Feiran Wang Ursula A. Matulonis George L. Mutter Jeffrey A. Golden Isaac S. Kohane |
| author_facet | Kun-Hsing Yu Vincent Hu Feiran Wang Ursula A. Matulonis George L. Mutter Jeffrey A. Golden Isaac S. Kohane |
| author_sort | Kun-Hsing Yu |
| collection | DOAJ |
| description | Abstract Background Ovarian cancer causes 151,900 deaths per year worldwide. Treatment and prognosis are primarily determined by the histopathologic interpretation in combination with molecular diagnosis. However, the relationship between histopathology patterns and molecular alterations is not fully understood, and it is difficult to predict patients’ chemotherapy response using the known clinical and histological variables. Methods We analyzed the whole-slide histopathology images, RNA-Seq, and proteomics data from 587 primary serous ovarian adenocarcinoma patients and developed a systematic algorithm to integrate histopathology and functional omics findings and to predict patients’ response to platinum-based chemotherapy. Results Our convolutional neural networks identified the cancerous regions with areas under the receiver operating characteristic curve (AUCs) > 0.95 and classified tumor grade with AUCs > 0.80. Functional omics analysis revealed that expression levels of proteins participated in innate immune responses and catabolic pathways are associated with tumor grade. Quantitative histopathology analysis successfully stratified patients with different response to platinum-based chemotherapy (P = 0.003). Conclusions These results indicated the potential clinical utility of quantitative histopathology evaluation in tumor cell detection and chemotherapy response prediction. The developed algorithm is easily extensible to other tumor types and treatment modalities. |
| first_indexed | 2024-12-14T20:46:38Z |
| format | Article |
| id | doaj.art-870a4535a412474faaef410028f69084 |
| institution | Directory Open Access Journal |
| issn | 1741-7015 |
| language | English |
| last_indexed | 2024-12-14T20:46:38Z |
| publishDate | 2020-08-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medicine |
| spelling | doaj.art-870a4535a412474faaef410028f690842022-12-21T22:48:00ZengBMCBMC Medicine1741-70152020-08-0118111410.1186/s12916-020-01684-wDeciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networksKun-Hsing Yu0Vincent Hu1Feiran Wang2Ursula A. Matulonis3George L. Mutter4Jeffrey A. Golden5Isaac S. Kohane6Department of Biomedical Informatics, Harvard Medical SchoolDepartment of Bioengineering, University of California San DiegoDepartment of Electrical Engineering, Stanford UniversityDivision of Gynecologic Oncology, Dana-Farber Cancer InstituteDepartment of Pathology, Brigham and Women’s HospitalDepartment of Pathology, Brigham and Women’s HospitalDepartment of Biomedical Informatics, Harvard Medical SchoolAbstract Background Ovarian cancer causes 151,900 deaths per year worldwide. Treatment and prognosis are primarily determined by the histopathologic interpretation in combination with molecular diagnosis. However, the relationship between histopathology patterns and molecular alterations is not fully understood, and it is difficult to predict patients’ chemotherapy response using the known clinical and histological variables. Methods We analyzed the whole-slide histopathology images, RNA-Seq, and proteomics data from 587 primary serous ovarian adenocarcinoma patients and developed a systematic algorithm to integrate histopathology and functional omics findings and to predict patients’ response to platinum-based chemotherapy. Results Our convolutional neural networks identified the cancerous regions with areas under the receiver operating characteristic curve (AUCs) > 0.95 and classified tumor grade with AUCs > 0.80. Functional omics analysis revealed that expression levels of proteins participated in innate immune responses and catabolic pathways are associated with tumor grade. Quantitative histopathology analysis successfully stratified patients with different response to platinum-based chemotherapy (P = 0.003). Conclusions These results indicated the potential clinical utility of quantitative histopathology evaluation in tumor cell detection and chemotherapy response prediction. The developed algorithm is easily extensible to other tumor types and treatment modalities.http://link.springer.com/article/10.1186/s12916-020-01684-wDigital pathologyPlatinum responseGene expressionProteomicsMachine learningSerous ovarian carcinoma |
| spellingShingle | Kun-Hsing Yu Vincent Hu Feiran Wang Ursula A. Matulonis George L. Mutter Jeffrey A. Golden Isaac S. Kohane Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks BMC Medicine Digital pathology Platinum response Gene expression Proteomics Machine learning Serous ovarian carcinoma |
| title | Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks |
| title_full | Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks |
| title_fullStr | Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks |
| title_full_unstemmed | Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks |
| title_short | Deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks |
| title_sort | deciphering serous ovarian carcinoma histopathology and platinum response by convolutional neural networks |
| topic | Digital pathology Platinum response Gene expression Proteomics Machine learning Serous ovarian carcinoma |
| url | http://link.springer.com/article/10.1186/s12916-020-01684-w |
| work_keys_str_mv | AT kunhsingyu decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks AT vincenthu decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks AT feiranwang decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks AT ursulaamatulonis decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks AT georgelmutter decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks AT jeffreyagolden decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks AT isaacskohane decipheringserousovariancarcinomahistopathologyandplatinumresponsebyconvolutionalneuralnetworks |