Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia

Hanya Mahmood; Adam Shephard; Paul Hankinson; Mike Bradburn; Anna Luiza Damaceno Araujo; Alan Roger Santos-Silva; Marcio Ajudarte Lopes; Pablo Agustin Vargas; Kris D. McCombe; Stephanie G. Craig; Jacqueline James; Jill Brooks; Paul Nankivell; Hisham Mehanna; Nasir Rajpoot; Syed Ali Khurram

doi:10.1038/s41416-023-02438-0

Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia

Hanya Mahmood^*, Adam Shephard, Paul Hankinson, Mike Bradburn, Anna Luiza Damaceno Araujo, Alan Roger Santos-Silva, Marcio Ajudarte Lopes, Pablo Agustin Vargas, Kris D. McCombe, Stephanie G. Craig, Jacqueline James, Jill Brooks, Paul Nankivell, Hisham Mehanna, Nasir Rajpoot, Syed Ali Khurram

^*Corresponding author for this work

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

45 Downloads (Pure)

Abstract

Background: Oral epithelial dysplasia (OED) is the precursor to oral squamous cell carcinoma which is amongst the top ten cancers worldwide. Prognostic significance of conventional histological features in OED is not well established. Many additional histological abnormalities are seen in OED, but are insufficiently investigated, and have not been correlated to clinical outcomes.

Methods: A digital quantitative analysis of epithelial cellularity, nuclear geometry, cytoplasm staining intensity and epithelial architecture/thickness is conducted on 75 OED whole-slide images (252 regions of interest) with feature-specific comparisons between grades and against non-dysplastic/control cases. Multivariable models were developed to evaluate prediction of OED recurrence and malignant transformation. The best performing models were externally validated on unseen cases pooled from four different centres (n = 121), of which 32% progressed to cancer, with an average transformation time of 45 months.

Results: Grade-based differences were seen for cytoplasmic eosin, nuclear eccentricity, and circularity in basal epithelial cells of OED (p < 0.05). Nucleus circularity was associated with OED recurrence (p = 0.018) and epithelial perimeter associated with malignant transformation (p = 0.03). The developed model demonstrated superior predictive potential for malignant transformation (AUROC 0.77) and OED recurrence (AUROC 0.74) as compared with conventional WHO grading (AUROC 0.68 and 0.71, respectively). External validation supported the prognostic strength of this model.

Conclusions: This study supports a novel prognostic model which outperforms existing grading systems. Further studies are warranted to evaluate its significance for OED prognostication.

Original language	English
Journal	British Journal of Cancer
Early online date	27 Sept 2023
DOIs	https://doi.org/10.1038/s41416-023-02438-0
Publication status	E-pub ahead of print - 27 Sept 2023

Bibliographical note

Funding Information:
HM received funding from the National Institute for Health Research for this study, which forms part of her Doctoral Research Fellowship (NIHR300904). AS, NR and SAK acknowledge funding from Cancer Research UK (C63489/A29674). Prof HM reports grants from UK National Institute of Health research, Cancer Research UK, the UK Medical Research Council, and AstraZeneca; advisory board fees from AstraZeneca, MSD, Merck, Nanobiotix, and Seagen; and is Director of Warwickshire Head Neck clinic and Docpsert Health. He is also an NIHR Senior Investigator. The views expressed in this paper are those of the author(s) and not necessarily those of the NIHR, or the Department of Health and Social Care.

Publisher Copyright:
© 2023, The Author(s).

ASJC Scopus subject areas

Oncology
Cancer Research

UN SDGs

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

Access to Document

10.1038/s41416-023-02438-0Licence: Creative Commons: Attribution (CC BY)

MahmoodH2023DevelopmentFinal published version, 1.33 MBLicence: Creative Commons: Attribution (CC BY)

AI Based Grading of Oral Epithelial Dysplasia and Prediction of Malignant Transformation
Nankivell, P. & Mehanna, H.
Cancer Research Uk
1/10/20 → 31/07/24
Project: Research

Cite this

Mahmood, H., Shephard, A., Hankinson, P., Bradburn, M., Araujo, A. L. D., Santos-Silva, A. R., Lopes, M. A., Vargas, P. A., McCombe, K. D., Craig, S. G., James, J., Brooks, J., Nankivell, P., Mehanna, H., Rajpoot, N., & Khurram, S. A. (2023). Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia. British Journal of Cancer. Advance online publication. https://doi.org/10.1038/s41416-023-02438-0

@article{cd887fffcdc240a1915076103e180a86,

title = "Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia",

abstract = "Background: Oral epithelial dysplasia (OED) is the precursor to oral squamous cell carcinoma which is amongst the top ten cancers worldwide. Prognostic significance of conventional histological features in OED is not well established. Many additional histological abnormalities are seen in OED, but are insufficiently investigated, and have not been correlated to clinical outcomes. Methods: A digital quantitative analysis of epithelial cellularity, nuclear geometry, cytoplasm staining intensity and epithelial architecture/thickness is conducted on 75 OED whole-slide images (252 regions of interest) with feature-specific comparisons between grades and against non-dysplastic/control cases. Multivariable models were developed to evaluate prediction of OED recurrence and malignant transformation. The best performing models were externally validated on unseen cases pooled from four different centres (n = 121), of which 32% progressed to cancer, with an average transformation time of 45 months. Results: Grade-based differences were seen for cytoplasmic eosin, nuclear eccentricity, and circularity in basal epithelial cells of OED (p < 0.05). Nucleus circularity was associated with OED recurrence (p = 0.018) and epithelial perimeter associated with malignant transformation (p = 0.03). The developed model demonstrated superior predictive potential for malignant transformation (AUROC 0.77) and OED recurrence (AUROC 0.74) as compared with conventional WHO grading (AUROC 0.68 and 0.71, respectively). External validation supported the prognostic strength of this model. Conclusions: This study supports a novel prognostic model which outperforms existing grading systems. Further studies are warranted to evaluate its significance for OED prognostication.",

author = "Hanya Mahmood and Adam Shephard and Paul Hankinson and Mike Bradburn and Araujo, {Anna Luiza Damaceno} and Santos-Silva, {Alan Roger} and Lopes, {Marcio Ajudarte} and Vargas, {Pablo Agustin} and McCombe, {Kris D.} and Craig, {Stephanie G.} and Jacqueline James and Jill Brooks and Paul Nankivell and Hisham Mehanna and Nasir Rajpoot and Khurram, {Syed Ali}",

note = "Funding Information: HM received funding from the National Institute for Health Research for this study, which forms part of her Doctoral Research Fellowship (NIHR300904). AS, NR and SAK acknowledge funding from Cancer Research UK (C63489/A29674). Prof HM reports grants from UK National Institute of Health research, Cancer Research UK, the UK Medical Research Council, and AstraZeneca; advisory board fees from AstraZeneca, MSD, Merck, Nanobiotix, and Seagen; and is Director of Warwickshire Head Neck clinic and Docpsert Health. He is also an NIHR Senior Investigator. The views expressed in this paper are those of the author(s) and not necessarily those of the NIHR, or the Department of Health and Social Care. Publisher Copyright: {\textcopyright} 2023, The Author(s).",

year = "2023",

month = sep,

day = "27",

doi = "10.1038/s41416-023-02438-0",

language = "English",

journal = "British Journal of Cancer",

issn = "0007-0920",

publisher = "Cancer Research UK",

}

Mahmood, H, Shephard, A, Hankinson, P, Bradburn, M, Araujo, ALD, Santos-Silva, AR, Lopes, MA, Vargas, PA, McCombe, KD, Craig, SG, James, J, Brooks, J , Nankivell, P , Mehanna, H, Rajpoot, N & Khurram, SA 2023, 'Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia', British Journal of Cancer. https://doi.org/10.1038/s41416-023-02438-0

TY - JOUR

T1 - Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia

AU - Mahmood, Hanya

AU - Shephard, Adam

AU - Hankinson, Paul

AU - Bradburn, Mike

AU - Araujo, Anna Luiza Damaceno

AU - Santos-Silva, Alan Roger

AU - Lopes, Marcio Ajudarte

AU - Vargas, Pablo Agustin

AU - McCombe, Kris D.

AU - Craig, Stephanie G.

AU - James, Jacqueline

AU - Brooks, Jill

AU - Nankivell, Paul

AU - Mehanna, Hisham

AU - Rajpoot, Nasir

AU - Khurram, Syed Ali

N1 - Funding Information: HM received funding from the National Institute for Health Research for this study, which forms part of her Doctoral Research Fellowship (NIHR300904). AS, NR and SAK acknowledge funding from Cancer Research UK (C63489/A29674). Prof HM reports grants from UK National Institute of Health research, Cancer Research UK, the UK Medical Research Council, and AstraZeneca; advisory board fees from AstraZeneca, MSD, Merck, Nanobiotix, and Seagen; and is Director of Warwickshire Head Neck clinic and Docpsert Health. He is also an NIHR Senior Investigator. The views expressed in this paper are those of the author(s) and not necessarily those of the NIHR, or the Department of Health and Social Care. Publisher Copyright: © 2023, The Author(s).

PY - 2023/9/27

Y1 - 2023/9/27

N2 - Background: Oral epithelial dysplasia (OED) is the precursor to oral squamous cell carcinoma which is amongst the top ten cancers worldwide. Prognostic significance of conventional histological features in OED is not well established. Many additional histological abnormalities are seen in OED, but are insufficiently investigated, and have not been correlated to clinical outcomes. Methods: A digital quantitative analysis of epithelial cellularity, nuclear geometry, cytoplasm staining intensity and epithelial architecture/thickness is conducted on 75 OED whole-slide images (252 regions of interest) with feature-specific comparisons between grades and against non-dysplastic/control cases. Multivariable models were developed to evaluate prediction of OED recurrence and malignant transformation. The best performing models were externally validated on unseen cases pooled from four different centres (n = 121), of which 32% progressed to cancer, with an average transformation time of 45 months. Results: Grade-based differences were seen for cytoplasmic eosin, nuclear eccentricity, and circularity in basal epithelial cells of OED (p < 0.05). Nucleus circularity was associated with OED recurrence (p = 0.018) and epithelial perimeter associated with malignant transformation (p = 0.03). The developed model demonstrated superior predictive potential for malignant transformation (AUROC 0.77) and OED recurrence (AUROC 0.74) as compared with conventional WHO grading (AUROC 0.68 and 0.71, respectively). External validation supported the prognostic strength of this model. Conclusions: This study supports a novel prognostic model which outperforms existing grading systems. Further studies are warranted to evaluate its significance for OED prognostication.

AB - Background: Oral epithelial dysplasia (OED) is the precursor to oral squamous cell carcinoma which is amongst the top ten cancers worldwide. Prognostic significance of conventional histological features in OED is not well established. Many additional histological abnormalities are seen in OED, but are insufficiently investigated, and have not been correlated to clinical outcomes. Methods: A digital quantitative analysis of epithelial cellularity, nuclear geometry, cytoplasm staining intensity and epithelial architecture/thickness is conducted on 75 OED whole-slide images (252 regions of interest) with feature-specific comparisons between grades and against non-dysplastic/control cases. Multivariable models were developed to evaluate prediction of OED recurrence and malignant transformation. The best performing models were externally validated on unseen cases pooled from four different centres (n = 121), of which 32% progressed to cancer, with an average transformation time of 45 months. Results: Grade-based differences were seen for cytoplasmic eosin, nuclear eccentricity, and circularity in basal epithelial cells of OED (p < 0.05). Nucleus circularity was associated with OED recurrence (p = 0.018) and epithelial perimeter associated with malignant transformation (p = 0.03). The developed model demonstrated superior predictive potential for malignant transformation (AUROC 0.77) and OED recurrence (AUROC 0.74) as compared with conventional WHO grading (AUROC 0.68 and 0.71, respectively). External validation supported the prognostic strength of this model. Conclusions: This study supports a novel prognostic model which outperforms existing grading systems. Further studies are warranted to evaluate its significance for OED prognostication.

UR - http://www.scopus.com/inward/record.url?scp=85172774999&partnerID=8YFLogxK

U2 - 10.1038/s41416-023-02438-0

DO - 10.1038/s41416-023-02438-0

M3 - Article

C2 - 37758836

AN - SCOPUS:85172774999

SN - 0007-0920

JO - British Journal of Cancer

JF - British Journal of Cancer

ER -

Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia

Abstract

Bibliographical note

ASJC Scopus subject areas

UN SDGs

Access to Document

Fingerprint

Projects

AI Based Grading of Oral Epithelial Dysplasia and Prediction of Malignant Transformation

Cite this