H&E-based MSI/MMR testing with AI in colorectal cancer: a multi-centred blinded evaluation

Cher Bass; Foivos Ntelemis; Julian Schmidt; Steffen Wolf; André Geraldes; Debapriya Mehrotra; Shikha Singhal; Narender Kumar; Angelica Marcia; Nicholas Bennett; Oscar Maiques; Mitchell Hyde; Bejal Mistry; Grace Rogerson; Michele Cummings; Clare Freer; Elizabeth Walsh; Manuel Salto-Tellez; Maurice Loughrey; In Hwa Um; David J. Harrison; Richard Clarkson; James Blackwood; J. Carl Barrett; Jakob Nikolas Kather; Nicolas M. Orsi; Pahini Pandya; Salim Arslan

doi:10.1038/s41746-025-02218-5

H&E-based MSI/MMR testing with AI in colorectal cancer: a multi-centred blinded evaluation

Cher Bass^*, Foivos Ntelemis, Julian Schmidt, Steffen Wolf, André Geraldes, Debapriya Mehrotra, Shikha Singhal, Narender Kumar, Angelica Marcia, Nicholas Bennett, Oscar Maiques, Mitchell Hyde, Bejal Mistry, Grace Rogerson, Michele Cummings, Clare Freer, Elizabeth Walsh, Manuel Salto-Tellez, Maurice Loughrey, In Hwa UmDavid J. Harrison, Richard Clarkson, James Blackwood, J. Carl Barrett, Jakob Nikolas Kather, Nicolas M. Orsi, Pahini Pandya, Salim Arslan

^*Corresponding author for this work

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

Abstract

Mismatch repair (MMR) deficiency occurs in 10–20% of colorectal cancer (CRC) cases, leading to microsatellite instability (MSI). Although MSI/MMR testing is critical for CRC management, high costs and long turnaround times limit testing rates and clinical utility, highlighting the need for more accessible, cost-effective alternatives. PANProfiler Colorectal (PPC) is an artificial intelligence (AI)-based biomarker test that determines MSI/MMR status directly from haematoxylin and eosin (H&E)-stained slides. We conducted a blinded, multi-centred validation to assess PPC’s performance against standard testing. The study included 3,576 whole slide images from 1,243 CRC patients across three United Kingdom institutions. PPC produced definitive results for 86.55% of slides, achieving an overall agreement of 93.83%, positive agreement of 92.54%, and negative agreement of 94.02%. PPC accurately determined MSI/MMR status from routine H&E slides, offering a rapid, scalable alternative to conventional diagnostic methods.

Original language	English
Article number	44
Pages (from-to)	1-11
Number of pages	11
Journal	npj Digital Medicine
Volume	9
DOIs	https://doi.org/10.1038/s41746-025-02218-5
Publication status	Published - 15 Dec 2025

UN SDGs

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

SDG 3 Good Health and Well-being

Access to Document

10.1038/s41746-025-02218-5Licence: CC BY

Bass-et-al-2025-H&E-based-MSI-MMR-testing-npjDigMed-9-44-CCBY
Copyright © 2025, The Author(s). This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Final published version, 2.55 MBLicence: CC BY

ICAIRD: I-CAIRD: Industrial Centre for AI Research in Digital Diagnostics
Harrison, D. (PI)
Innovate UK
1/02/19 → 31/01/22
Project: Standard

Cite this

Bass, C., Ntelemis, F., Schmidt, J., Wolf, S., Geraldes, A., Mehrotra, D., Singhal, S., Kumar, N., Marcia, A., Bennett, N., Maiques, O., Hyde, M., Mistry, B., Rogerson, G., Cummings, M., Freer, C., Walsh, E., Salto-Tellez, M., Loughrey, M., ... Arslan, S. (2025). H&E-based MSI/MMR testing with AI in colorectal cancer: a multi-centred blinded evaluation. npj Digital Medicine, 9, 1-11. Article 44. https://doi.org/10.1038/s41746-025-02218-5

@article{678f307bc5064def90b622430d1aee58,

title = "H\&E-based MSI/MMR testing with AI in colorectal cancer: a multi-centred blinded evaluation",

abstract = "Mismatch repair (MMR) deficiency occurs in 10–20\% of colorectal cancer (CRC) cases, leading to microsatellite instability (MSI). Although MSI/MMR testing is critical for CRC management, high costs and long turnaround times limit testing rates and clinical utility, highlighting the need for more accessible, cost-effective alternatives. PANProfiler Colorectal (PPC) is an artificial intelligence (AI)-based biomarker test that determines MSI/MMR status directly from haematoxylin and eosin (H\&E)-stained slides. We conducted a blinded, multi-centred validation to assess PPC{\textquoteright}s performance against standard testing. The study included 3,576 whole slide images from 1,243 CRC patients across three United Kingdom institutions. PPC produced definitive results for 86.55\% of slides, achieving an overall agreement of 93.83\%, positive agreement of 92.54\%, and negative agreement of 94.02\%. PPC accurately determined MSI/MMR status from routine H\&E slides, offering a rapid, scalable alternative to conventional diagnostic methods.",

author = "Cher Bass and Foivos Ntelemis and Julian Schmidt and Steffen Wolf and Andr{\'e} Geraldes and Debapriya Mehrotra and Shikha Singhal and Narender Kumar and Angelica Marcia and Nicholas Bennett and Oscar Maiques and Mitchell Hyde and Bejal Mistry and Grace Rogerson and Michele Cummings and Clare Freer and Elizabeth Walsh and Manuel Salto-Tellez and Maurice Loughrey and Um, \{In Hwa\} and Harrison, \{David J.\} and Richard Clarkson and James Blackwood and Barrett, \{J. Carl\} and Kather, \{Jakob Nikolas\} and Orsi, \{Nicolas M.\} and Pahini Pandya and Salim Arslan",

note = "Funding: This project was (partially) funded by Innovate UK Smart Grant 10054824. The results shown in this study are partially based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga. We acknowledge the support of iCAIRD (Industrial Centre for Artificial Intelligence Research in Digital Diagnostics), funded by Innovate UK, for the provision of the I1-UK-CRC-CZI-1-DEV cohort. Biosamples in the W1-UK-CRC-CZI-1-BLIND cohort were obtained from the Wales Cancer Biobank (https://openbioresources.metajnl.com/articles/10.5334/ojb.46) which is funded by Health and Care Research Wales. Other investigators may have received specimens from the same subjects. Biosamples in the N1-UK-CRC-SVS-1-BLIND cohort were obtained from the Northern Ireland Biobank (https://openbioresources.metajnl.com/articles/10.5334/ojb.47 which received core funding from the Health and Social Care Research and Development Division of the Public Health Agency in Northern Ireland. O.M is funded by Barts Charity (MGU0504).",

year = "2025",

month = dec,

day = "15",

doi = "10.1038/s41746-025-02218-5",

language = "English",

volume = "9",

pages = "1--11",

journal = "npj Digital Medicine",

issn = "2398-6352",

publisher = "Springer Nature",

}

Bass, C, Ntelemis, F, Schmidt, J, Wolf, S, Geraldes, A, Mehrotra, D, Singhal, S, Kumar, N, Marcia, A, Bennett, N, Maiques, O, Hyde, M, Mistry, B, Rogerson, G, Cummings, M, Freer, C, Walsh, E, Salto-Tellez, M, Loughrey, M, Um, IH , Harrison, DJ, Clarkson, R, Blackwood, J, Barrett, JC, Kather, JN, Orsi, NM, Pandya, P & Arslan, S 2025, 'H&E-based MSI/MMR testing with AI in colorectal cancer: a multi-centred blinded evaluation', npj Digital Medicine, vol. 9, 44, pp. 1-11. https://doi.org/10.1038/s41746-025-02218-5

TY - JOUR

T1 - H&E-based MSI/MMR testing with AI in colorectal cancer

T2 - a multi-centred blinded evaluation

AU - Bass, Cher

AU - Ntelemis, Foivos

AU - Schmidt, Julian

AU - Wolf, Steffen

AU - Geraldes, André

AU - Mehrotra, Debapriya

AU - Singhal, Shikha

AU - Kumar, Narender

AU - Marcia, Angelica

AU - Bennett, Nicholas

AU - Maiques, Oscar

AU - Hyde, Mitchell

AU - Mistry, Bejal

AU - Rogerson, Grace

AU - Cummings, Michele

AU - Freer, Clare

AU - Walsh, Elizabeth

AU - Salto-Tellez, Manuel

AU - Loughrey, Maurice

AU - Um, In Hwa

AU - Harrison, David J.

AU - Clarkson, Richard

AU - Blackwood, James

AU - Barrett, J. Carl

AU - Kather, Jakob Nikolas

AU - Orsi, Nicolas M.

AU - Pandya, Pahini

AU - Arslan, Salim

N1 - Funding: This project was (partially) funded by Innovate UK Smart Grant 10054824. The results shown in this study are partially based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga. We acknowledge the support of iCAIRD (Industrial Centre for Artificial Intelligence Research in Digital Diagnostics), funded by Innovate UK, for the provision of the I1-UK-CRC-CZI-1-DEV cohort. Biosamples in the W1-UK-CRC-CZI-1-BLIND cohort were obtained from the Wales Cancer Biobank (https://openbioresources.metajnl.com/articles/10.5334/ojb.46) which is funded by Health and Care Research Wales. Other investigators may have received specimens from the same subjects. Biosamples in the N1-UK-CRC-SVS-1-BLIND cohort were obtained from the Northern Ireland Biobank (https://openbioresources.metajnl.com/articles/10.5334/ojb.47 which received core funding from the Health and Social Care Research and Development Division of the Public Health Agency in Northern Ireland. O.M is funded by Barts Charity (MGU0504).

PY - 2025/12/15

Y1 - 2025/12/15

N2 - Mismatch repair (MMR) deficiency occurs in 10–20% of colorectal cancer (CRC) cases, leading to microsatellite instability (MSI). Although MSI/MMR testing is critical for CRC management, high costs and long turnaround times limit testing rates and clinical utility, highlighting the need for more accessible, cost-effective alternatives. PANProfiler Colorectal (PPC) is an artificial intelligence (AI)-based biomarker test that determines MSI/MMR status directly from haematoxylin and eosin (H&E)-stained slides. We conducted a blinded, multi-centred validation to assess PPC’s performance against standard testing. The study included 3,576 whole slide images from 1,243 CRC patients across three United Kingdom institutions. PPC produced definitive results for 86.55% of slides, achieving an overall agreement of 93.83%, positive agreement of 92.54%, and negative agreement of 94.02%. PPC accurately determined MSI/MMR status from routine H&E slides, offering a rapid, scalable alternative to conventional diagnostic methods.

AB - Mismatch repair (MMR) deficiency occurs in 10–20% of colorectal cancer (CRC) cases, leading to microsatellite instability (MSI). Although MSI/MMR testing is critical for CRC management, high costs and long turnaround times limit testing rates and clinical utility, highlighting the need for more accessible, cost-effective alternatives. PANProfiler Colorectal (PPC) is an artificial intelligence (AI)-based biomarker test that determines MSI/MMR status directly from haematoxylin and eosin (H&E)-stained slides. We conducted a blinded, multi-centred validation to assess PPC’s performance against standard testing. The study included 3,576 whole slide images from 1,243 CRC patients across three United Kingdom institutions. PPC produced definitive results for 86.55% of slides, achieving an overall agreement of 93.83%, positive agreement of 92.54%, and negative agreement of 94.02%. PPC accurately determined MSI/MMR status from routine H&E slides, offering a rapid, scalable alternative to conventional diagnostic methods.

U2 - 10.1038/s41746-025-02218-5

DO - 10.1038/s41746-025-02218-5

M3 - Article

SN - 2398-6352

VL - 9

SP - 1

EP - 11

JO - npj Digital Medicine

JF - npj Digital Medicine

M1 - 44

ER -

H&E-based MSI/MMR testing with AI in colorectal cancer: a multi-centred blinded evaluation

Abstract

UN SDGs

Access to Document

Fingerprint

Projects

ICAIRD: I-CAIRD: Industrial Centre for AI Research in Digital Diagnostics

Cite this