EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predicting Malignant Transformation in Oral Leukoplakia: A Multilayer Perceptron Approach Incorporating Clinicopathological Features and DNA Content

Guilherme Iani Pontes, Anna Luíza Damaceno Araújo, Andresa Borges Soares, Saman Warnakulasuriya iD, André Luis Santana de Freitas, Caroline Gennari Stevão iD, Marcelo Sperandio, Matheus Cardoso Moraes iD

Journal of Oral Pathology & Medicine · 2026

DOI: 10.1111/jop.70084 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Oral Pathology & Medicine, 2026. Verfasst von Guilherme Iani Pontes und 7 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Science and Technology, Federal University of São Paulo (ICT-UNIFESP), São José dos Campos, São Paulo, Brazil. Hauptthemen laut MEDLINE: Leukoplakia, Oral, Cell Transformation, Neoplastic, Machine Learning, Mouth Neoplasms. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação de Amparo à Pesquisa do Estado de São Paulo, Brazilian Ministry of Health (MoH), Brazilian National Program of Genomics and Precision Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Oral leukoplakia (OL) is a potentially malignant disorder of the oral mucosa. Accurate prediction of malignant transformation (MT) remains a clinical challenge. This study aimed to develop and evaluate a machine learning model that integrates histopathological, demographic, and DNA content features to predict MT risk in OL.

Methods

We conducted a retrospective cohort study of 97 OL cases—18 with confirmed MT and 79 non‐transformed controls—selected from a larger series. Each case included clinicopathological features, and DNA content data obtained by flow cytometry for cell cycle phases (G1, S‐phase, G2 and excess DNA beyond the tetraploid region [4cER]). All cases had a minimum 5‐year follow‐up or histologically confirmed transformation. A multilayer perceptron (MLP) model was trained on 27 features. Stratified five‐fold cross‐validation and minority class oversampling (positive filling) were used to improve learning and mitigate data imbalance. Performance was evaluated using accuracy, sensitivity, specificity, F1‐score, AUC, and Kaplan–Meier survival analysis.

Results

Significant predictors of MT included 4cER ( p = 0.005), G2 phase ( p = 0.04), dysplasia grading ( p = 0.003), and inflammatory infiltrate ( p = 0.01). The optimized model yielded 72% sensitivity, 96% specificity, and an AUC of 85.4%. Survival analysis showed significantly poorer outcomes in the high‐risk cases predicted by the model ( p < 0.0001).

Conclusion

Integrating DNA content analysis with machine learning provides an objective and clinically useful model to stratify malignant risk in OL, complementing conventional histopathology and supporting personalized patient management.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Leukoplakia, OralCell Transformation, NeoplasticMachine LearningMouth NeoplasmsHumansMaleFemaleRetrospective StudiesMiddle AgedAgedAdultAged, 80 and overMultilayer Perceptrons

Stichwörter der Autor:innen

DNA contentDNA content analysisOPMDepithelial dysplasiamachine learningmalignant transformationoral canceroral leukoplakiaploidyrisk prediction models

Autor:innen und Institutionen

  • Institute of Science and Technology, Federal University of São Paulo (ICT-UNIFESP), São José dos Campos, São Paulo, Brazil.
  • Head and Neck Surgery Department, University of São Paulo Medical School, São Paulo, São Paulo, Brazil.
  • Department of Surgery, Stomatology, Pathology, and Radiology, Area of Pathology, Bauru School of Dentistry, University of São Paulo (FOB USP), Bauru, São Paulo, Brazil.
  • Faculty of Dentistry, Oral and Craniofacial Sciences, King's College London, London, UK.
  • WHO Collaborating Centre for Oral Cancer, London, UK.
  • Faculdade São Leopoldo Mandic, Campinas, São Paulo, Brazil.

ORCID-Profile: Saman Warnakulasuriya, Caroline Gennari Stevão, Matheus Cardoso Moraes

Förderung

  • Fundação de Amparo à Pesquisa do Estado de São Paulo
  • Brazilian Ministry of Health (MoH)
  • Brazilian National Program of Genomics and Precision Health
  • National Council for Scientific and Technological Development (CNPq, Brazil)
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico

Zitieren

APA

Pontes, G. I., Araújo, A. L. D., Soares, A. B., Warnakulasuriya, S., de Freitas, A. L. S., Stevão, C. G., Sperandio, M., & Moraes, M. C. (2026). Predicting Malignant Transformation in Oral Leukoplakia: A Multilayer Perceptron Approach Incorporating Clinicopathological Features and DNA Content. Journal of Oral Pathology & Medicine. https://doi.org/10.1111/jop.70084

Vancouver

Pontes GI, Araújo ALD, Soares AB, Warnakulasuriya S, de Freitas ALS, Stevão CG, et al. Predicting Malignant Transformation in Oral Leukoplakia: A Multilayer Perceptron Approach Incorporating Clinicopathological Features and DNA Content. Journal of Oral Pathology & Medicine. 2026. doi:10.1111/jop.70084

BibTeX
@article{Pontes2026,
  author = {Pontes, Guilherme Iani and Araújo, Anna Luíza Damaceno and Soares, Andresa Borges and Warnakulasuriya, Saman and de Freitas, André Luis Santana and Stevão, Caroline Gennari and Sperandio, Marcelo and Moraes, Matheus Cardoso},
  title = {Predicting Malignant Transformation in Oral Leukoplakia: A Multilayer Perceptron Approach Incorporating Clinicopathological Features and DNA Content},
  journal = {Journal of Oral Pathology & Medicine},
  year = {2026},
  doi = {10.1111/jop.70084}
}
Kostenlos starten →