Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology
Abstract Determining tumor microsatellite status has significant clinical value because tumors that are microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) respond well to immune checkpoint inhibitors (ICIs) and oftentimes not to chemotherapeutics. We propose MSI-SEER, a deep...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-05-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-025-01580-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849730811701493760 |
|---|---|
| author | Sunho Park Morgan F. Pettigrew Yoon Jin Cha In-Ho Kim Minji Kim Imon Banerjee Isabel Barnfather Jean R. Clemenceau Inyeop Jang Hyunki Kim Younghoon Kim Rish K. Pai Jeong Hwan Park N. Jewel Samadder Kyo Young Song Ji-Youn Sung Jae-Ho Cheong Jeonghyun Kang Sung Hak Lee Sam C. Wang Tae Hyun Hwang |
| author_facet | Sunho Park Morgan F. Pettigrew Yoon Jin Cha In-Ho Kim Minji Kim Imon Banerjee Isabel Barnfather Jean R. Clemenceau Inyeop Jang Hyunki Kim Younghoon Kim Rish K. Pai Jeong Hwan Park N. Jewel Samadder Kyo Young Song Ji-Youn Sung Jae-Ho Cheong Jeonghyun Kang Sung Hak Lee Sam C. Wang Tae Hyun Hwang |
| author_sort | Sunho Park |
| collection | DOAJ |
| description | Abstract Determining tumor microsatellite status has significant clinical value because tumors that are microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) respond well to immune checkpoint inhibitors (ICIs) and oftentimes not to chemotherapeutics. We propose MSI-SEER, a deep Gaussian process-based Bayesian model that analyzes H&E whole-slide images in weakly-supervised-learning to predict microsatellite status in gastric and colorectal cancers. We performed extensive validation using multiple large datasets comprised of patients from diverse racial backgrounds. MSI-SEER achieved state-of-the-art performance with MSI prediction by integrating uncertainty prediction. We achieved high accuracy for predicting ICI responsiveness by combining tumor MSI status with stroma-to-tumor ratio. Finally, MSI-SEER’s tile-level predictions revealed novel insights into the role of spatial distribution of MSI-H regions in the tumor microenvironment and ICI response. |
| format | Article |
| id | doaj-art-3e9aca173f5f43b69259150733c4a1d2 |
| institution | DOAJ |
| issn | 2398-6352 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj-art-3e9aca173f5f43b69259150733c4a1d22025-08-20T03:08:44ZengNature Portfolionpj Digital Medicine2398-63522025-05-018111510.1038/s41746-025-01580-8Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histologySunho Park0Morgan F. Pettigrew1Yoon Jin Cha2In-Ho Kim3Minji Kim4Imon Banerjee5Isabel Barnfather6Jean R. Clemenceau7Inyeop Jang8Hyunki Kim9Younghoon Kim10Rish K. Pai11Jeong Hwan Park12N. Jewel Samadder13Kyo Young Song14Ji-Youn Sung15Jae-Ho Cheong16Jeonghyun Kang17Sung Hak Lee18Sam C. Wang19Tae Hyun Hwang20Vanderbilt University Medical CenterDivision of Surgical Oncology, Department of Surgery, University of Texas Southwestern Medical CenterDepartment of Pathology, Gangnam Severance Hospital, Yonsei University College of MedicineDepartment of Internal Medicine, Division of Medical Oncology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of KoreaVanderbilt University Medical CenterDepartment of Radiology, Mayo ClinicDepartment of Artificial Intelligence and Informatics, Mayo ClinicVanderbilt University Medical CenterVanderbilt University Medical CenterDepartment of Pathology, Yonsei University College of MedicineDepartment of Hospital Pathology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of KoreaDepartment of Laboratory Medicine and Pathology, Mayo ClinicDepartment of Pathology, SMG-SNU Boramae Medical Center, Seoul National University College of MedicineDepartment of Gastroenterology and Hepatology, Mayo ClinicDivision of Gastrointestinal Surgery, Department of Surgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of KoreaDepartment of Pathology, College of Medicine, Kyung Hee University hospital, Kyung Hee UniversityDepartment of Surgery, Department of Biochemistry and Molecular Biology, Department of Biomedical Systems Informatics, Yonsei University College of MedicineDepartment of Surgery, Gangnam Severance Hospital, Yonsei University College of MedicineDepartment of Hospital Pathology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of KoreaDivision of Surgical Oncology, Department of Surgery, University of Texas Southwestern Medical CenterVanderbilt University Medical CenterAbstract Determining tumor microsatellite status has significant clinical value because tumors that are microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) respond well to immune checkpoint inhibitors (ICIs) and oftentimes not to chemotherapeutics. We propose MSI-SEER, a deep Gaussian process-based Bayesian model that analyzes H&E whole-slide images in weakly-supervised-learning to predict microsatellite status in gastric and colorectal cancers. We performed extensive validation using multiple large datasets comprised of patients from diverse racial backgrounds. MSI-SEER achieved state-of-the-art performance with MSI prediction by integrating uncertainty prediction. We achieved high accuracy for predicting ICI responsiveness by combining tumor MSI status with stroma-to-tumor ratio. Finally, MSI-SEER’s tile-level predictions revealed novel insights into the role of spatial distribution of MSI-H regions in the tumor microenvironment and ICI response.https://doi.org/10.1038/s41746-025-01580-8 |
| spellingShingle | Sunho Park Morgan F. Pettigrew Yoon Jin Cha In-Ho Kim Minji Kim Imon Banerjee Isabel Barnfather Jean R. Clemenceau Inyeop Jang Hyunki Kim Younghoon Kim Rish K. Pai Jeong Hwan Park N. Jewel Samadder Kyo Young Song Ji-Youn Sung Jae-Ho Cheong Jeonghyun Kang Sung Hak Lee Sam C. Wang Tae Hyun Hwang Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology npj Digital Medicine |
| title | Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology |
| title_full | Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology |
| title_fullStr | Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology |
| title_full_unstemmed | Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology |
| title_short | Deep Gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology |
| title_sort | deep gaussian process with uncertainty estimation for microsatellite instability and immunotherapy response prediction from histology |
| url | https://doi.org/10.1038/s41746-025-01580-8 |
| work_keys_str_mv | AT sunhopark deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT morganfpettigrew deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT yoonjincha deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT inhokim deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT minjikim deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT imonbanerjee deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT isabelbarnfather deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT jeanrclemenceau deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT inyeopjang deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT hyunkikim deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT younghoonkim deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT rishkpai deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT jeonghwanpark deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT njewelsamadder deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT kyoyoungsong deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT jiyounsung deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT jaehocheong deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT jeonghyunkang deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT sunghaklee deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT samcwang deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology AT taehyunhwang deepgaussianprocesswithuncertaintyestimationformicrosatelliteinstabilityandimmunotherapyresponsepredictionfromhistology |