---
title: "Deep learning on H&E slides predicts HPV/p16 status and molecular subtypes in vulvar SCC"
id: "medrxiv-5-predicting-hpv-p16-status-and-molecular-subtypes-from-histopathology-slides"
canonical_url: "https://medichelpline.com/clinical-feed/medrxiv-5-predicting-hpv-p16-status-and-molecular-subtypes-from-histopathology-slides"
content_type: "clinical_feed_article"
specialty: "Oncology"
source_name: "medRxiv (Clinical Preprints)"
source_url: "https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1?rss=1"
published_at: "2026-10-24T12:00:00.000Z"
evidence_level: "Verified Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Deep learning on H&E slides predicts HPV/p16 status and molecular subtypes in vulvar SCC
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/medrxiv-5-predicting-hpv-p16-status-and-molecular-subtypes-from-histopathology-slides
- **Specialty:** [Oncology](https://medichelpline.com/clinical-feed/oncology.md)
- **Primary Source:** medRxiv (Clinical Preprints)
- **Source URL:** [Original Journal Publication](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1?rss=1)
- **Published At:** 2026-10-24T12:00:00.000Z
- **Evidence Rating:** Verified Feed
## Executive GIST (TL;DR)
- The authors developed two attention-based multiple-instance learning models, **HERA** (binary HPV/p16 prediction) and **HERMES** (three-class molecular subtype prediction), using features extracted with H-optimus-1 from routine **hematoxylin and eosin** whole-slide images. - Model development used 644 patients drawn from three Dutch VSCC cohorts; independent testing used 286 patients from two external cohorts, as reported by the study team. - In external testing, **HERA** achieved an AUC of 0.84 (95% CI 0.75–0.91), specificity 0.93, negative predictive value 0.94, and sensitivity 0.68; predicted groups reproduced HPV-stratified prognostic separation (log-rank p<0.001). - **HERMES** produced a macro-average AUC of 0.70 (0.66–0.75); class AUCs were 0.80 for HPV-associated, 0.68 for HPV-independent/p53-abnormal, and 0.63 for HPV-independent/p53-wildtype, per the authors. - HERMES-predicted classes reproduced a prognostic hierarchy with reported 5-year recurrence rates of 14.0% (HPV-associated), 31.1% (HPV-independent/p53-wildtype) and 43.3% (HPV-independent/p53-abnormal) (p=0.0051). - Visual inspection of highly attended image tiles showed basaloid and koilocytotic-like atypia for **HPV**-associated tumors; keratinizing nests with desmoplastic stroma for HPV-independent/p53-abnormal; and well-differentiated keratinocytes with spongiosis for HPV-independent/p53-wildtype. - The authors note that **HERA** could assign HPV/p16 status without consuming tissue and enable HPV-stratified analyses in retrospective cohorts lacking ancillary testing; they explicitly call for external validation of discrimination and calibration across scanners and populations. - The study team observed that while molecular subtype information appears encoded in routine H&E slides, improved discrimination between HPV-independent subtypes is required before clinical application. - The source did not report specific model hyperparameters, training epochs, slide-level pre/post-processing pipelines, or per-cohort demographic breakdowns in the abstract; such methodological details may be in the full text or supplementary material but are not stated in the abstract itself.
## Clinical Analysis & Structured Key Points
[Skip to main content](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1?rss=1#main-content) [![medRxiv](https://www.medrxiv.org/sites/default/files/medrxiv_inner_logo.png)](https://www.medrxiv.org/) * [Home](https://www.medrxiv.org/) * [Submit](https://www.medrxiv.org/submit-a-manuscript) * [FAQ](https://www.medrxiv.org/about/FAQ) * [Blog](https://connect.medrxiv.org/news/) * [ALERTS / RSS](https://www.medrxiv.org/content/alertsrss) * [Resources](https://connect.medrxiv.org/resources/) * [About](https://www.medrxiv.org/content/about-medrxiv) Search for this keyword [Advanced Search](https://www.medrxiv.org/search) Follow this preprint # Predicting HPV/p16 Status and Molecular Subtypes from Histopathology Slides using Deep Learning in Vulvar Squamous Cell Carcinoma Nikki van den Berg, Willemijn van der Kolk, Kim Landsbergen, Kim Kortekaas, Linda Nooij, Anne Fleur van Velzen, Nienke te Grootenhuis, the AIRMEC Consortium, Nanda Horeweg, Lena van Doorn, Maaike Oonk, Jos Bart, Tjalling Bosse doi: https://doi.org/10.64898/2026.10.22.26366203 [This article is a preprint and has not been certified by peer review [](https://www.medrxiv.org/about/FAQ#unrefereed)[what does this mean?](https://www.medrxiv.org/about/FAQ#unrefereed)]. It reports new medical research that has yet to be evaluated and so should _not_ be used to guide clinical practice. Nikki van den Berg 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Nikki%2Bvan%2Bden%2BBerg%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=van%20den%20Berg%20N&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3ANikki%2Bvan%2Bden%2BBerg%2B) Willemijn van der Kolk 2 University Medical Center Groningen; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Willemijn%2Bvan%2Bder%2BKolk%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=van%20der%20Kolk%20W&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3AWillemijn%2Bvan%2Bder%2BKolk%2B) Kim Landsbergen 2 University Medical Center Groningen; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Kim%2BLandsbergen%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Landsbergen%20K&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3AKim%2BLandsbergen%2B) Kim Kortekaas 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Kim%2BKortekaas%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Kortekaas%20K&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3AKim%2BKortekaas%2B) Linda Nooij 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Linda%2BNooij%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Nooij%20L&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3ALinda%2BNooij%2B) Anne Fleur van Velzen 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Anne%2BFleur%2Bvan%2BVelzen%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=van%20Velzen%20AF&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3AAnne%2BFleur%2Bvan%2BVelzen%2B) Nienke te Grootenhuis 2 University Medical Center Groningen; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Nienke%2Bte%2BGrootenhuis%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=te%20Grootenhuis%20N&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3ANienke%2Bte%2BGrootenhuis%2B) 3 -; Nanda Horeweg 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Nanda%2BHoreweg%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Horeweg%20N&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3ANanda%2BHoreweg%2B) Lena van Doorn 4 University Medical Center Rotterdam * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Lena%2Bvan%2BDoorn%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=van%20Doorn%20L&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3ALena%2Bvan%2BDoorn%2B) Maaike Oonk 2 University Medical Center Groningen; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Maaike%2BOonk%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Oonk%20M&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3AMaaike%2BOonk%2B) Jos Bart 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Jos%2BBart%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Bart%20J&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3AJos%2BBart%2B) Tjalling Bosse 1 Leiden University Medical Center; * [Find this author on Google Scholar](https://www.medrxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Tjalling%2BBosse%2B "Open in new tab") * [Find this author on PubMed](https://www.medrxiv.org/lookup/external-ref?access_num=Bosse%20T&link_type=AUTHORSEARCH "Open in new tab") * [Search for this author on this site](https://www.medrxiv.org/search/author1%3ATjalling%2BBosse%2B) * For correspondence: t.bosse@lumc.nl * [Abstract](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1)[](https://www.medrxiv.org/panels_ajax_tab/biorxiv_tab_art/node:1330107/1) * [Info/History](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.article-info)[](https://www.medrxiv.org/panels_ajax_tab/biorxiv_tab_info/node:1330107/1) * [Metrics](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.article-metrics)[](https://www.medrxiv.org/panels_ajax_tab/article_tab_metrics/node:1330107/1) * [Supplementary material](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.supplementary-material)[](https://www.medrxiv.org/panels_ajax_tab/biorxiv_tab_data/node:1330107/1) * [Data/Code](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.external-links)[](https://www.medrxiv.org/panels_ajax_tab/article_tab_data_code/node:1330107/1) * [ Preview PDF](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.full.pdf+html)[](https://www.medrxiv.org/panels_ajax_tab/biorxiv_tab_pdf/node:1330107/1) ![Loading](https://www.medrxiv.org/sites/all/modules/contrib/panels_ajax_tab/images/loading.gif) ## Abstract Vulvar squamous cell carcinoma (VSCC) comprises three prognostically distinct molecular subtypes: human papillomavirus (HPV)-associated, HPV-independent/p53-wildtype, and HPV-independent/p53-abnormal. Classification requires p16/p53 immunohistochemistry, which consumes tissue and is often incomplete in retrospective cohorts. We investigated whether deep learning could predict HPV/p16 status and molecular subtypes directly from routine hematoxylin and eosin-stained whole-slide images. Using features extracted with H-optimus-1, we developed two attention-based multiple-instance learning models: HERA, predicting HPV/p16 status, and HERMES, predicting the three molecular subtypes. Models were developed on 644 patients from three Dutch VSCC cohorts and tested on 286 patients from two independent cohorts. In independent testing, HERA achieved an area under the curve (AUC) of 0.84 (95% CI 0.75-0.91), with specificity 0.93, negative predictive value 0.94, and sensitivity 0.68. HERA-predicted groups reproduced the prognostic separation of HPV testing (log-rank p ## Funder Information Declared The Hanarth Foundation, No specific number Dutch Cancer Society, https://ror.org/0368jnd28, 2016–10168 Copyright The copyright holder for this preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a [CC-BY-NC-ND 4.0 International license](http://creativecommons.org/licenses/by-nc-nd/4.0/). bioRxiv and medRxiv thank the following for their generous financial support: > The Chan Zuckerberg Initiative, Cold Spring Harbor Laboratory, the Sergey Brin Family Foundation, California Institute of Technology, Centre National de la Recherche Scientifique, Fred Hutchinson Cancer Center, Imperial College London, Massachusetts Institute of Technology, Stanford University, The University of Edinburgh, University of Washington, and Vrije Universiteit Amsterdam. [Donate to openRxiv ](https://www.zeffy.com/en-US/donation-form/donate-to-make-a-difference-10981) [ Back to top](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1?rss=1#page) [ Previous](https://www.medrxiv.org/content/10.64898/2026.10.21.26366153v1 "Glucoregulatory and Inflammatory Trajectories in Pregnancy with Obesity: Secondary Analysis of a Randomized Exercise Trial") Posted October 24, 2026. [ Download PDF](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.full.pdf) Print/Save Options [Download PDF](https://www.medrxiv.org/content/medrxiv/early/2026/10/24/2026.10.22.26366203.full.pdf)Full Text & In-line FiguresXML [More Info](https://www.medrxiv.org/about/FAQ#PrintOptions "More Information on Print/Save Options") [ Author Declarations](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1#p-5) [Supplementary Material ](https://www.medrxiv.org/content/10.64898/2026.10.22.26366203v1.supplementary-material) [ Data/Code](https://www.medrxiv.org/content/early/2026/10/24/2026.10.22.26366203.external-links) [ Email](https://www.medrxiv.org/ "Email this Article") [ Share](https://www.medrxiv.org/) Predicting HPV/p16 Status and Molecular Subtypes from Histopathology Slides using Deep Learning in Vulvar Squamous Cell Carcinoma Nikki van den Berg, Willemijn van der Kolk, Kim Landsbergen, Kim Kortekaas, Linda Nooij, Anne Fleur van Velzen, Nienke te Grootenhuis, the AIRMEC Consortium, Nanda Horeweg, Lena van Doorn, Maaike Oonk, Jos Bart, Tjalling Bosse medRxiv 2026.10.22.26366203; doi: https://doi.org/10.64898/2026.10.22.26366203 [This article is a preprint and has not been certified by peer review [](https://www.medrxiv.org/about/FAQ#unrefereed)[what does this mean?](https://www.medrxiv.org/about/FAQ#unrefereed)]. It reports new medical research that has yet to be evaluated and so should _not_ be used to guide clinical practice. Share This Article: Copy [![Twitter logo](https://www.medrxiv.org/sites/all/modules/highwire/highwire/images/twitter.png)](https://www.medrxiv.org/highwire_log/share/twitter?link=http%3A%2F%2Ftwitter.com%2Fshare%3Furl%3Dhttps%253A%2F%2Fwww.medrxiv.org%2Fcontent%2F10.64898%2F2026.10.22.26366203v1%26text%3DPredicting%2520HPV%2Fp16%2520Status%2520and%2520Molecular%2520Subtypes%2520from%2520Histopathology%2520Slides%2520using%2520Deep%2520Learning%2520in%2520Vulvar%2520Squamous%2520Cell%2520Carcinoma "Share this on Twitter") [![Facebook logo](https://www.medrxiv.org/sites/all/modules/highwire/highwire/images/fb-blue.png)](https://www.medrxiv.org/highwire_log/share/facebook?link=http%3A%2F%2Fwww.facebook.com%2Fsharer.php%3Fu%3Dhttps%253A%2F%2Fwww.medrxiv.org%2Fcontent%2F10.64898%2F2026.10.22.26366203v1%26t%3DPredicting%2520HPV%2Fp16%2520Status%2520and%2520Molecular%2520Subtypes%2520from%2520Histopathology%2520Slides%2520using%2520Deep%2520Learning%2520in%2520Vulvar%2520Squamous%2520Cell%2520Carcinoma "Share on Facebook") [![LinkedIn logo](https://www.medrxiv.org/sites/all/modules/highwire/highwire/images/linkedin-32px.png)](https://www.medrxiv.org/highwire_log/share/linkedin?link=http%3A%2F%2Fwww.linkedin.com%2FshareArticle%3Fmini%3Dtrue%26url%3Dhttps%253A%2F%2Fwww.medrxiv.org%2Fcontent%2F10.64898%2F2026.10.22.26366203v1%26title%3DPredicting%2520HPV%2Fp16%2520Status%2520and%2520Molecular%2520Subtypes%2520from%2520Histopathology%2520Slides%2520using%2520Deep%2520Learning%2520in%2520Vulvar%2520Squamous%2520Cell%2520Carcinoma%26summary%3D%26source%3DmedRxiv "Publish this post to LinkedIn") [![Mendeley logo](https://www.medrxiv.org/sites/all/modules/highwire/highwire/images/mendeley.png)](https://www.medrxiv.org/highwire_log/share/mendeley?link=http%3A%2F%2Fwww.mendeley.com%2Fimport%2F%3Furl%3Dhttps%253A%2F%2Fwww.medrxiv.org%2Fcontent%2F10.64898%2F2026.10.22.26366203v1%26title%3DPredicting%2520HPV%2Fp16%2520Status%2520and%2520Molecular%2520Subtypes%2520from%2520Histopathology%2520Slides%2520using%2520Deep%2520Learning%2520in%2520Vulvar%2520Squamous%2520Cell%2520Carcinoma "Share on Mendeley") [ Citation Tools](https://www.medrxiv.org/ "Citation Tools") [ Get QR code](https://connect.medrxiv.org/qr/2026.10.22.26366203) * [Tweet Widget](http://twitter.com/share?url=https%3A//www.medrxiv.org/content/10.64898/2026.10.22.26366203v1&count=horizontal&via=&text=Predicting%20HPV/p16%20Status%20and%20Molecular%20Subtypes%20from%20Histopathology%20Slides%20using%20Deep%20Learning%20in%20Vulvar%20Squamous%20Cell%20Carcinoma&counturl=www.medrxiv.org/content/10.64898/2026.10.22.26366203v1 "Tweet This") Reviews and Context 0 Comment 0 TRIP Peer Reviews 0 Community Reviews 0 Automated Services 0 Blogs/Media 0 Author Videos **Subject Areas** [**All Articles**](https://www.medrxiv.org/content/early/recent) * [Addiction Medicine](https://www.medrxiv.org/collection/addiction-medicine) (634) * [Allergy and Immunology](https://www.medrxiv.org/collection/allergy-and-immunology) (914) * [Anesthesia](https://www.medrxiv.org/collection/anesthesia) (348) * [Cardiovascular Medicine](https://www.medrxiv.org/collection/cardiovascular-medicine) (4947) * [Dentistry and Oral Medicine](https://www.medrxiv.org/collection/dentistry-and-oral-medicine) (490) * [Dermatology](https://www.medrxiv.org/collection/dermatology) (425) * [Emergency Medicine](https://www.medrxiv.org/collection/emergency-medicine) (674) * [Endocrinology (including Diabetes Mellitus and Metabolic Disease)](https://www.medrxiv.org/collection/endocrinology-including-diabetes-mellitus-and-metabolic-disease) (1670) * [Epidemiology](https://www.medrxiv.org/collection/epidemiology) (16171) * [Forensic Medicine](https://www.medrxiv.org/collection/forensic-medicine) (36) * [Gastroenterology](https://www.medrxiv.org/collection/gastroenterology) (1236) * [Genetic and Genomic Medicine](https://www.medrxiv.org/collection/genetic-and-genomic-medicine) (7216) * [Geriatric Medicine](https://www.medrxiv.org/collection/geriatric-medicine) (747) * [Health Economics](https://www.medrxiv.org/collection/health-economics) (1092) * [Health Informatics](https://www.medrxiv.org/collection/health-informatics) (5229) * [Health Policy](https://www.medrxiv.org/collection/health-policy) (1457) * [Health Systems and Quality Improvement](https://www.medrxiv.org/collection/health-systems-and-quality-improvement) (1815) * [Hematology](https://www.medrxiv.org/collection/hematology) (597) * [HIV/AIDS](https://www.medrxiv.org/collection/hiv-aids) (1373) * [Infectious Diseases (except HIV/AIDS)](https://www.medrxiv.org/collection/infectious_diseases) (16431) * [Intensive Care and Critical Care Medicine](https://www.medr
## Related Clinical Research

- [Significant Findings from ESMO 2026 on Cancer Vaccines and Small Cell Lung Cancer Treatments](https://medichelpline.com/clinical-feed/stat-news-0-the-esmo-presentation-everyone-s-been-waiting-for.md)
- [STAT readers weigh in: Theranos, Galleri screening, integrative oncology, and medication adherence](https://medichelpline.com/clinical-feed/stat-news-1-opinion-stat-readers-on-elizabeth-holmes-cancer-screening-and-more.md)
- [Glucoregulatory and Inflammatory Trajectories in Pregnancy with Obesity: Exercise Trial Secondary](https://medichelpline.com/clinical-feed/medrxiv-8-glucoregulatory-and-inflammatory-trajectories-in-pregnancy-with-obesity.md)
- [Lipid Panels Predict First Heart Failure: LDL-C, ApoB and Sex-Specific Signals](https://medichelpline.com/clinical-feed/medrxiv-10-cholesterol-paradox-multi-point-heart-failure-prediction-from-lipid-panels-in.md)
- [Safety and Immunogenicity of the GBS-AlpN Vaccine in Adults 55–75: Phase 1 Randomised Trial](https://medichelpline.com/clinical-feed/pubmed-42759245.md) (DOI: 10.1016/j.vaccine.2026.129133)

## Navigation
- [← Back to Oncology Feed](https://medichelpline.com/clinical-feed/oncology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.