Digital pathology does not improve the prognostic performance of Ki‐67 in a large cohort of penile cancer
The Journal of Pathology: Clinical Research · 2026-11 · 12 (6)
DOI: 10.1002/2056-4538.70122 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Pathology: Clinical Research, online erschienen am 10.11.2026. Verfasst von Jan Hrudka und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pathology, 3rd Faculty of Medicine, Charles University, University Hospital Kralovske Vinohrady, Prague, Czech Republic. Hauptthemen laut MEDLINE: Ki-67 Antigen, Penile Neoplasms, Carcinoma, Squamous Cell, Biomarkers, Tumor. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Univerzita Karlova v Praze, Cooperatio Medical Diagnostics, Ministerstvo Zdravotnictví Ceské Republiky, Ministerstvo Obrany České Republiky.
Abstract (Originalsprache)
Penile squamous cell carcinoma (pSCC) is a rare but prognostically unfavourable genitourinary malignancy in which the search for reliable prognostic biomarkers has intensified in recent years. The prognostic role of Ki‐67 in pSCC remains uncertain. This study aimed to assess the prognostic value of Ki‐67 in pSCC using digital image analysis (HALO) and assessment by two pathologists, to compare their prognostic performance and interobserver agreement, and to explore associations with clinicopathological, immune and molecular features. In a large retrospective cohort of 224 invasive pSCCs, Ki‐67 was quantified by digital image analysis and independently estimated by two senior pathologists. Optimal cut‐points (67%, 25%, and 75%) were derived using maximally selected rank statistics. Ki‐67‐high status was significantly associated with 5‐year overall survival with HALO in univariate analysis [hazard ratio (HR) = 1.62, 95% CI 1.03–2.55, p = 0.037], and for HALO (HR 1.63, 95% CI 1.03–2.56, p = 0.035) and Pathologist 2 (HR 1.63, 95% CI 1.02–2.58, p = 0.04) after adjustment for age. When assessed by Pathologist 1, high Ki‐67 expression (≥25%) was significantly associated with shorter cancer‐specific survival. Further covariate‐adjusted analyses showed variable results across assessment methods, with the most consistent prognostic associations observed for Pathologist 2. Penalized spline analysis demonstrated a gradual, approximately linear increase in mortality risk with increasing Ki‐67 values across all three raters, without identifying a clear prognostic threshold. Interobserver agreement was moderate to substantial (Cohen's kappa 0.42–0.61). High Ki‐67 was associated with high grade, p16, and HPV positivity by quantitative PCR (qPCR) in all three raters. Exploratory analyses suggested an inverse association between Ki‐67 and B‐cell/plasma‐cell infiltration. The prognostic value of Ki‐67 in pSCC was non‐robust and depended on the assessment method and chosen cut‐off. Digital image analysis did not significantly outperform manual scoring or improve interobserver agreement. Finally, the inverse association between Ki‐67 and B‐cell/plasma‐cell infiltration represents a potentially novel observation warranting further validation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Pathology, 3rd Faculty of Medicine, Charles University, University Hospital Kralovske Vinohrady, Prague, Czech Republic.
- Department of Pathology, 1st Faculty of Medicine, Charles University, General University Hospital, Prague, Czech Republic.
- Department of Pathology and Molecular Medicine, 3rd Faculty of Medicine, Charles University, Thomayer University Hospital, Prague, Czech Republic.
- Department of Urology, 3rd Faculty of Medicine, Charles University, University Hospital Kralovske Vinohrady, Prague, Czech Republic.
- Department of Urology, 3rd Faculty of Medicine of Charles University, Thomayer University Hospital, Prague, Czech Republic.
- Šikl´s Institute of Pathology, Faculty of Medicine in Pilsen, Charles University, University Hospital Pilsen, Pilsen, Czech Republic.
ORCID-Profile: Jan Hrudka, Radoslav Matěj, Jan Hojný, Nicolette Zavillová, Květoslava Michalová, Jiří Soukup, Petr Waldauf
Förderung
- Univerzita Karlova v Praze, Cooperatio Medical Diagnostics
- Ministerstvo Zdravotnictví Ceské Republiky
- Ministerstvo Obrany České Republiky
- Univerzita Karlova v Praze
Zitieren
APA
Hrudka, J., Matěj, R., Grus, M., Marková, A., Prouzová, Z., Hojný, J., Kendall Bártů, M., Čapka, D., Zavillová, N., Michalová, K., Soukup, J., Záleský, M., & Waldauf, P. (2026). Digital pathology does not improve the prognostic performance of Ki‐67 in a large cohort of penile cancer. The Journal of Pathology: Clinical Research, 12(6). https://doi.org/10.1002/2056-4538.70122
Vancouver
Hrudka J, Matěj R, Grus M, Marková A, Prouzová Z, Hojný J, et al. Digital pathology does not improve the prognostic performance of Ki‐67 in a large cohort of penile cancer. The Journal of Pathology: Clinical Research. 2026;12(6). doi:10.1002/2056-4538.70122
BibTeX
@article{Hrudka2026,
author = {Hrudka, Jan and Matěj, Radoslav and Grus, Michal and Marková, Anežka and Prouzová, Zuzana and Hojný, Jan and Kendall Bártů, Michaela and Čapka, David and Zavillová, Nicolette and Michalová, Květoslava and Soukup, Jiří and Záleský, Miroslav and Waldauf, Petr},
title = {Digital pathology does not improve the prognostic performance of Ki‐67 in a large cohort of penile cancer},
journal = {The Journal of Pathology: Clinical Research},
year = {2026},
volume = {12},
number = {6},
doi = {10.1002/2056-4538.70122}
}
EUVIMED