Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer

Byungsoo Ahn; Damin Moon; Hyun Soo Kim; Chung Lee; Nam Hoon Cho; Heung Kook Choi; Dongmin Kim; Jung Yun Lee; Eun Ji Nam; Dongju Won; Hee Jung An; Sun Young Kwon; Su Jin Shin; Hye Ra Jung; Dohee Kwon; Heejung Park; Milim Kim; Yoon Jin Cha; Hyunjin Park; Yangkyu Lee; Songmi Noh; Yong Moon Lee; Sung Eun Choi; Ji Min Kim; Sun Hee Sung; Eunhyang Park

doi:10.1038/s41467-024-48667-6

Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer

Byungsoo Ahn, Damin Moon, Hyun Soo Kim, Chung Lee, Nam Hoon Cho, Heung Kook Choi, Dongmin Kim, Jung Yun Lee, Eun Ji Nam, Dongju Won, Hee Jung An, Sun Young Kwon, Su Jin Shin, Hye Ra Jung, Dohee Kwon, Heejung Park, Milim Kim, Yoon Jin Cha, Hyunjin Park, Yangkyu LeeSongmi Noh, Yong Moon Lee, Sung Eun Choi, Ji Min Kim, Sun Hee Sung, Eunhyang Park

Department of Pathology

Research output: Contribution to journal › Article › peer-review

4 Scopus citations

Abstract

Platinum-based chemotherapy is the cornerstone treatment for female high-grade serous ovarian carcinoma (HGSOC), but choosing an appropriate treatment for patients hinges on their responsiveness to it. Currently, no available biomarkers can promptly predict responses to platinum-based treatment. Therefore, we developed the Pathologic Risk Classifier for HGSOC (PathoRiCH), a histopathologic image–based classifier. PathoRiCH was trained on an in-house cohort (n = 394) and validated on two independent external cohorts (n = 284 and n = 136). The PathoRiCH-predicted favorable and poor response groups show significantly different platinum-free intervals in all three cohorts. Combining PathoRiCH with molecular biomarkers provides an even more powerful tool for the risk stratification of patients. The decisions of PathoRiCH are explained through visualization and a transcriptomic analysis, which bolster the reliability of our model’s decisions. PathoRiCH exhibits better predictive performance than current molecular biomarkers. PathoRiCH will provide a solid foundation for developing an innovative tool to transform the current diagnostic pipeline for HGSOC.

Original language	English
Article number	4253
Journal	Nature Communications
Volume	15
Issue number	1
DOIs	https://doi.org/10.1038/s41467-024-48667-6
State	Published - 18 May 2024

Bibliographical note

Publisher Copyright:
© The Author(s) 2024.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1038/s41467-024-48667-6

Cite this

Ahn, B., Moon, D., Kim, H. S., Lee, C., Cho, N. H., Choi, H. K., Kim, D., Lee, J. Y., Nam, E. J., Won, D., An, H. J., Kwon, S. Y., Shin, S. J., Jung, H. R., Kwon, D., Park, H., Kim, M., Cha, Y. J., Park, H., ... Park, E. (2024). Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer. Nature Communications, 15(1), Article 4253. https://doi.org/10.1038/s41467-024-48667-6

@article{7bceb1c404354e02af9b30557b4e142b,

title = "Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer",

abstract = "Platinum-based chemotherapy is the cornerstone treatment for female high-grade serous ovarian carcinoma (HGSOC), but choosing an appropriate treatment for patients hinges on their responsiveness to it. Currently, no available biomarkers can promptly predict responses to platinum-based treatment. Therefore, we developed the Pathologic Risk Classifier for HGSOC (PathoRiCH), a histopathologic image–based classifier. PathoRiCH was trained on an in-house cohort (n = 394) and validated on two independent external cohorts (n = 284 and n = 136). The PathoRiCH-predicted favorable and poor response groups show significantly different platinum-free intervals in all three cohorts. Combining PathoRiCH with molecular biomarkers provides an even more powerful tool for the risk stratification of patients. The decisions of PathoRiCH are explained through visualization and a transcriptomic analysis, which bolster the reliability of our model{\textquoteright}s decisions. PathoRiCH exhibits better predictive performance than current molecular biomarkers. PathoRiCH will provide a solid foundation for developing an innovative tool to transform the current diagnostic pipeline for HGSOC.",

author = "Byungsoo Ahn and Damin Moon and Kim, {Hyun Soo} and Chung Lee and Cho, {Nam Hoon} and Choi, {Heung Kook} and Dongmin Kim and Lee, {Jung Yun} and Nam, {Eun Ji} and Dongju Won and An, {Hee Jung} and Kwon, {Sun Young} and Shin, {Su Jin} and Jung, {Hye Ra} and Dohee Kwon and Heejung Park and Milim Kim and Cha, {Yoon Jin} and Hyunjin Park and Yangkyu Lee and Songmi Noh and Lee, {Yong Moon} and Choi, {Sung Eun} and Kim, {Ji Min} and Sung, {Sun Hee} and Eunhyang Park",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2024.",

year = "2024",

month = may,

day = "18",

doi = "10.1038/s41467-024-48667-6",

language = "English",

volume = "15",

journal = "Nature Communications",

issn = "2041-1723",

publisher = "Nature Research",

number = "1",

}

Ahn, B, Moon, D, Kim, HS, Lee, C, Cho, NH, Choi, HK, Kim, D, Lee, JY, Nam, EJ, Won, D, An, HJ, Kwon, SY, Shin, SJ, Jung, HR, Kwon, D, Park, H, Kim, M, Cha, YJ, Park, H, Lee, Y, Noh, S, Lee, YM, Choi, SE, Kim, JM, Sung, SH & Park, E 2024, 'Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer', Nature Communications, vol. 15, no. 1, 4253. https://doi.org/10.1038/s41467-024-48667-6

TY - JOUR

T1 - Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer

AU - Ahn, Byungsoo

AU - Moon, Damin

AU - Kim, Hyun Soo

AU - Lee, Chung

AU - Cho, Nam Hoon

AU - Choi, Heung Kook

AU - Kim, Dongmin

AU - Lee, Jung Yun

AU - Nam, Eun Ji

AU - Won, Dongju

AU - An, Hee Jung

AU - Kwon, Sun Young

AU - Shin, Su Jin

AU - Jung, Hye Ra

AU - Kwon, Dohee

AU - Park, Heejung

AU - Kim, Milim

AU - Cha, Yoon Jin

AU - Park, Hyunjin

AU - Lee, Yangkyu

AU - Noh, Songmi

AU - Lee, Yong Moon

AU - Choi, Sung Eun

AU - Kim, Ji Min

AU - Sung, Sun Hee

AU - Park, Eunhyang

PY - 2024/5/18

Y1 - 2024/5/18

N2 - Platinum-based chemotherapy is the cornerstone treatment for female high-grade serous ovarian carcinoma (HGSOC), but choosing an appropriate treatment for patients hinges on their responsiveness to it. Currently, no available biomarkers can promptly predict responses to platinum-based treatment. Therefore, we developed the Pathologic Risk Classifier for HGSOC (PathoRiCH), a histopathologic image–based classifier. PathoRiCH was trained on an in-house cohort (n = 394) and validated on two independent external cohorts (n = 284 and n = 136). The PathoRiCH-predicted favorable and poor response groups show significantly different platinum-free intervals in all three cohorts. Combining PathoRiCH with molecular biomarkers provides an even more powerful tool for the risk stratification of patients. The decisions of PathoRiCH are explained through visualization and a transcriptomic analysis, which bolster the reliability of our model’s decisions. PathoRiCH exhibits better predictive performance than current molecular biomarkers. PathoRiCH will provide a solid foundation for developing an innovative tool to transform the current diagnostic pipeline for HGSOC.

AB - Platinum-based chemotherapy is the cornerstone treatment for female high-grade serous ovarian carcinoma (HGSOC), but choosing an appropriate treatment for patients hinges on their responsiveness to it. Currently, no available biomarkers can promptly predict responses to platinum-based treatment. Therefore, we developed the Pathologic Risk Classifier for HGSOC (PathoRiCH), a histopathologic image–based classifier. PathoRiCH was trained on an in-house cohort (n = 394) and validated on two independent external cohorts (n = 284 and n = 136). The PathoRiCH-predicted favorable and poor response groups show significantly different platinum-free intervals in all three cohorts. Combining PathoRiCH with molecular biomarkers provides an even more powerful tool for the risk stratification of patients. The decisions of PathoRiCH are explained through visualization and a transcriptomic analysis, which bolster the reliability of our model’s decisions. PathoRiCH exhibits better predictive performance than current molecular biomarkers. PathoRiCH will provide a solid foundation for developing an innovative tool to transform the current diagnostic pipeline for HGSOC.

UR - http://www.scopus.com/inward/record.url?scp=85193468053&partnerID=8YFLogxK

U2 - 10.1038/s41467-024-48667-6

DO - 10.1038/s41467-024-48667-6

M3 - Article

C2 - 38762636

AN - SCOPUS:85193468053

SN - 2041-1723

VL - 15

JO - Nature Communications

JF - Nature Communications

IS - 1

M1 - 4253

ER -

Histopathologic image–based deep learning classifier for predicting platinum-based treatment responses in high-grade serous ovarian cancer

Abstract

Bibliographical note

UN SDGs

Access to Document

Fingerprint

Cite this