Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides
Abstract Background Molecular profiling of estrogen receptor (ER), progesterone receptor (PR), and ERBB2 (also known as Her2) is essential for breast cancer diagnosis and treatment planning. Nevertheless, current methods rely on the qualitative interpretation of immunohistochemistry and fluorescence...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-12-01
|
| Series: | Communications Medicine |
| Online Access: | https://doi.org/10.1038/s43856-024-00695-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571397091622912 |
|---|---|
| author | Gil Shamai Ran Schley Alexandra Cretu Tal Neoran Edmond Sabo Yoav Binenbaum Shachar Cohen Tal Goldman António Polónia Keren Drumea Karin Stoliar Ron Kimmel |
| author_facet | Gil Shamai Ran Schley Alexandra Cretu Tal Neoran Edmond Sabo Yoav Binenbaum Shachar Cohen Tal Goldman António Polónia Keren Drumea Karin Stoliar Ron Kimmel |
| author_sort | Gil Shamai |
| collection | DOAJ |
| description | Abstract Background Molecular profiling of estrogen receptor (ER), progesterone receptor (PR), and ERBB2 (also known as Her2) is essential for breast cancer diagnosis and treatment planning. Nevertheless, current methods rely on the qualitative interpretation of immunohistochemistry and fluorescence in situ hybridization (FISH), which can be costly, time-consuming, and inconsistent. Here we explore the clinical utility of predicting receptor status from digitized hematoxylin and eosin-stained (H&E) slides using machine learning trained and evaluated on a multi-institutional dataset. Methods We developed a deep learning system to predict ER, PR, and ERBB2 statuses from digitized H&E slides and evaluated its utility in three clinical applications: identifying hormone receptor-positive patients, serving as a second-read tool for quality assurance, and addressing intratumor heterogeneity. For development and validation, we collected 19,845 slides from 7,950 patients across six independent cohorts representative of diverse clinical settings. Results Here we show that the system identifies 30.5% of patients as hormone receptor-positive, achieving a specificity of 0.9982 and a positive predictive value of 0.9992, demonstrating its ability to determine eligibility for hormone therapy without immunohistochemistry. By restaining and reassessing samples flagged as potential false negatives, we discover 31 cases of misdiagnosed ER, PR, and ERBB2 statuses. Conclusions These findings demonstrate the utility of the system in diverse clinical settings and its potential to improve breast cancer diagnosis. Given the substantial focus of current guidelines on reducing false negative diagnoses, this study supports the integration of H&E-based machine learning tools into workflows for quality assurance. |
| format | Article |
| id | doaj-art-abfbf11a53974b8bb2007b8e0e2d5fc5 |
| institution | Kabale University |
| issn | 2730-664X |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Medicine |
| spelling | doaj-art-abfbf11a53974b8bb2007b8e0e2d5fc52025-02-02T12:40:06ZengNature PortfolioCommunications Medicine2730-664X2024-12-014111610.1038/s43856-024-00695-5Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slidesGil Shamai0Ran Schley1Alexandra Cretu2Tal Neoran3Edmond Sabo4Yoav Binenbaum5Shachar Cohen6Tal Goldman7António Polónia8Keren Drumea9Karin Stoliar10Ron Kimmel11Department of Computer Science, Technion—Israel Institue of TechnologyDepartment of Computer Science, Technion—Israel Institue of TechnologyDepartment of Pathology, Carmel Medical CenterDepartment of Computer Science, Technion—Israel Institue of TechnologyDepartment of Pathology, Carmel Medical CenterDivision of Pediatric Hematology-Oncology, Boston Children’s HospitalDepartment of Computer Science, Technion—Israel Institue of TechnologyDepartment of Pathology, Haemek Medical CenterInstitute of Molecular Pathology and Immunology of the University of PortoBreast Cancer Unit, Carmel Medical CenterFaculty of Medicine, Technion—Israel Institue of TechnologyDepartment of Computer Science, Technion—Israel Institue of TechnologyAbstract Background Molecular profiling of estrogen receptor (ER), progesterone receptor (PR), and ERBB2 (also known as Her2) is essential for breast cancer diagnosis and treatment planning. Nevertheless, current methods rely on the qualitative interpretation of immunohistochemistry and fluorescence in situ hybridization (FISH), which can be costly, time-consuming, and inconsistent. Here we explore the clinical utility of predicting receptor status from digitized hematoxylin and eosin-stained (H&E) slides using machine learning trained and evaluated on a multi-institutional dataset. Methods We developed a deep learning system to predict ER, PR, and ERBB2 statuses from digitized H&E slides and evaluated its utility in three clinical applications: identifying hormone receptor-positive patients, serving as a second-read tool for quality assurance, and addressing intratumor heterogeneity. For development and validation, we collected 19,845 slides from 7,950 patients across six independent cohorts representative of diverse clinical settings. Results Here we show that the system identifies 30.5% of patients as hormone receptor-positive, achieving a specificity of 0.9982 and a positive predictive value of 0.9992, demonstrating its ability to determine eligibility for hormone therapy without immunohistochemistry. By restaining and reassessing samples flagged as potential false negatives, we discover 31 cases of misdiagnosed ER, PR, and ERBB2 statuses. Conclusions These findings demonstrate the utility of the system in diverse clinical settings and its potential to improve breast cancer diagnosis. Given the substantial focus of current guidelines on reducing false negative diagnoses, this study supports the integration of H&E-based machine learning tools into workflows for quality assurance.https://doi.org/10.1038/s43856-024-00695-5 |
| spellingShingle | Gil Shamai Ran Schley Alexandra Cretu Tal Neoran Edmond Sabo Yoav Binenbaum Shachar Cohen Tal Goldman António Polónia Keren Drumea Karin Stoliar Ron Kimmel Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides Communications Medicine |
| title | Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides |
| title_full | Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides |
| title_fullStr | Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides |
| title_full_unstemmed | Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides |
| title_short | Clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin-stained slides |
| title_sort | clinical utility of receptor status prediction in breast cancer and misdiagnosis identification using deep learning on hematoxylin and eosin stained slides |
| url | https://doi.org/10.1038/s43856-024-00695-5 |
| work_keys_str_mv | AT gilshamai clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT ranschley clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT alexandracretu clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT talneoran clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT edmondsabo clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT yoavbinenbaum clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT shacharcohen clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT talgoldman clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT antoniopolonia clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT kerendrumea clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT karinstoliar clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides AT ronkimmel clinicalutilityofreceptorstatuspredictioninbreastcancerandmisdiagnosisidentificationusingdeeplearningonhematoxylinandeosinstainedslides |