A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing
Abstract Microsatellite instability (MSI) is a critical phenotype of cancer genomes and an FDA-recognized biomarker that can guide treatment with immune checkpoint inhibitors. Previous work has demonstrated that next-generation sequencing data can be used to identify samples with MSI-high phenotype....
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-54970-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559241002844160 |
|---|---|
| author | John Ziegler Jaclyn F. Hechtman Satshil Rana Ryan N. Ptashkin Gowtham Jayakumaran Sumit Middha Shweta S. Chavan Chad Vanderbilt Deborah DeLair Jacklyn Casanova Jinru Shia Nicole DeGroat Ryma Benayed Marc Ladanyi Michael F. Berger Thomas J. Fuchs A. Rose Brannon Ahmet Zehir |
| author_facet | John Ziegler Jaclyn F. Hechtman Satshil Rana Ryan N. Ptashkin Gowtham Jayakumaran Sumit Middha Shweta S. Chavan Chad Vanderbilt Deborah DeLair Jacklyn Casanova Jinru Shia Nicole DeGroat Ryma Benayed Marc Ladanyi Michael F. Berger Thomas J. Fuchs A. Rose Brannon Ahmet Zehir |
| author_sort | John Ziegler |
| collection | DOAJ |
| description | Abstract Microsatellite instability (MSI) is a critical phenotype of cancer genomes and an FDA-recognized biomarker that can guide treatment with immune checkpoint inhibitors. Previous work has demonstrated that next-generation sequencing data can be used to identify samples with MSI-high phenotype. However, low tumor purity, as frequently observed in routine clinical samples, poses a challenge to the sensitivity of existing algorithms. To overcome this critical issue, we developed MiMSI, an MSI classifier based on deep neural networks and trained using a dataset that included low tumor purity MSI cases in a multiple instance learning framework. On a challenging yet representative set of cases, MiMSI showed higher sensitivity (0.895) and auROC (0.971) than MSISensor (sensitivity: 0.67; auROC: 0.907), an open-source software previously validated for clinical use at our institution using MSK-IMPACT large panel targeted NGS data. In a separate, prospective cohort, MiMSI confirmed that it outperforms MSISensor in low purity cases (P = 8.244e-07). |
| format | Article |
| id | doaj-art-c7c189e896b046a68c488f67361962c2 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-c7c189e896b046a68c488f67361962c22025-01-05T12:40:13ZengNature PortfolioNature Communications2041-17232025-01-0116111110.1038/s41467-024-54970-zA deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencingJohn Ziegler0Jaclyn F. Hechtman1Satshil Rana2Ryan N. Ptashkin3Gowtham Jayakumaran4Sumit Middha5Shweta S. Chavan6Chad Vanderbilt7Deborah DeLair8Jacklyn Casanova9Jinru Shia10Nicole DeGroat11Ryma Benayed12Marc Ladanyi13Michael F. Berger14Thomas J. Fuchs15A. Rose Brannon16Ahmet Zehir17Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterMarie-Josée and Henry R. Kravis Center for Molecular Oncology, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterDepartment of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer CenterAbstract Microsatellite instability (MSI) is a critical phenotype of cancer genomes and an FDA-recognized biomarker that can guide treatment with immune checkpoint inhibitors. Previous work has demonstrated that next-generation sequencing data can be used to identify samples with MSI-high phenotype. However, low tumor purity, as frequently observed in routine clinical samples, poses a challenge to the sensitivity of existing algorithms. To overcome this critical issue, we developed MiMSI, an MSI classifier based on deep neural networks and trained using a dataset that included low tumor purity MSI cases in a multiple instance learning framework. On a challenging yet representative set of cases, MiMSI showed higher sensitivity (0.895) and auROC (0.971) than MSISensor (sensitivity: 0.67; auROC: 0.907), an open-source software previously validated for clinical use at our institution using MSK-IMPACT large panel targeted NGS data. In a separate, prospective cohort, MiMSI confirmed that it outperforms MSISensor in low purity cases (P = 8.244e-07).https://doi.org/10.1038/s41467-024-54970-z |
| spellingShingle | John Ziegler Jaclyn F. Hechtman Satshil Rana Ryan N. Ptashkin Gowtham Jayakumaran Sumit Middha Shweta S. Chavan Chad Vanderbilt Deborah DeLair Jacklyn Casanova Jinru Shia Nicole DeGroat Ryma Benayed Marc Ladanyi Michael F. Berger Thomas J. Fuchs A. Rose Brannon Ahmet Zehir A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing Nature Communications |
| title | A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing |
| title_full | A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing |
| title_fullStr | A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing |
| title_full_unstemmed | A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing |
| title_short | A deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing |
| title_sort | deep multiple instance learning framework improves microsatellite instability detection from tumor next generation sequencing |
| url | https://doi.org/10.1038/s41467-024-54970-z |
| work_keys_str_mv | AT johnziegler adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT jaclynfhechtman adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT satshilrana adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT ryannptashkin adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT gowthamjayakumaran adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT sumitmiddha adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT shwetaschavan adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT chadvanderbilt adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT deborahdelair adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT jacklyncasanova adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT jinrushia adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT nicoledegroat adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT rymabenayed adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT marcladanyi adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT michaelfberger adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT thomasjfuchs adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT arosebrannon adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT ahmetzehir adeepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT johnziegler deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT jaclynfhechtman deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT satshilrana deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT ryannptashkin deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT gowthamjayakumaran deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT sumitmiddha deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT shwetaschavan deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT chadvanderbilt deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT deborahdelair deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT jacklyncasanova deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT jinrushia deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT nicoledegroat deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT rymabenayed deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT marcladanyi deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT michaelfberger deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT thomasjfuchs deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT arosebrannon deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing AT ahmetzehir deepmultipleinstancelearningframeworkimprovesmicrosatelliteinstabilitydetectionfromtumornextgenerationsequencing |