Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2026 Jan;47(Suppl 1):9-17.
doi: 10.1007/s00292-025-01518-7. Epub 2025 Dec 3.

Evaluation of single-nucleotide variants in bladder cancer using prediction algorithms

June Möller  1   2 , Lancelot Seillier  1   2 , Axel Fürstberger  3 , Michael Rose  1   2   3 , Danny D Jonigk  1   2   4 , Nadina Ortiz-Brüchle  1   2 , Nadine T Gaisa  5   6   7
Affiliations

Evaluation of single-nucleotide variants in bladder cancer using prediction algorithms

June Möller et al. Pathologie (Heidelb). 2026 Jan.

Abstract
in English, German

Background: Bladder and urinary tract cancer show poor survival rates and demand novel therapeutic strategies. Advances in the omics domain have expanded genetic analysis, with prediction tools offering potential support. However, their performance may differ by tumor entity.

Objective: This study aimed to evaluate prediction tool performance using genetic data from bladder and urinary tract cancer.

Methods: Variant data were obtained from ClinVar and cBioPortal for bladder cancer (n = 441), PanCancer (n = 361), and benign variants (n = 357). Sixteen prediction algorithms were assessed individually and in combinations of two or three; oncogenes and tumor suppressors were compared. A PanCancer dataset of Suybeng et al. was also analyzed.

Results: Prediction performance varied across datasets. Combinations of three tools achieved the highest sensitivity (100%: MutationTaster/MetaSVM/List-S2) and specificity (97.45%: MutationTaster/DEOGEN2/FATHMM_XF). Entity-specific and gene-type differences were observed.

Conclusion: Combining prediction tools enhances genetic analysis. Tool selection should depend on cancer entity, gene function, and study objective.

Zusammenfassung: HINTERGRUND: Harnblasen- und Harnwegskarzinome weisen eine schlechte Überlebensrate auf und erfordern neue therapeutische Ansätze. Fortschritte im Omics-Bereich haben die genetische Analyse erweitert, wobei Prädiktionstools eine potenzielle Unterstützung darstellen. Ihre Leistung kann jedoch zwischen Tumorentitäten variieren.

Fragestellung: Ziel der Studie war die Bewertung der Leistungsfähigkeit von Prädiktionstools bei der Analyse genetischer Daten von Harnblasen- und Harnwegskarzinomen.

Material und methoden: Variantendaten wurden aus den Datenbanken ClinVar und cBioPortal für Blasenkarzinome (n = 441), PanCancer (n = 361) und aus benignen Varianten (n = 357) extrahiert. Einzeln sowie in Kombinationen von 2 und 3 Tools wurden 16 Algorithmen getestet; Onkogene und Tumorsuppressorgene wurden verglichen. Zusätzlich wurde ein PanCancer-Datensatz von Suybeng et al. einbezogen.

Ergebnisse: Die Prädiktionsleistung variiert zwischen den Datensätzen. Kombinationen aus 3 Tools erzielten die höchste Sensitivität (100 %: MutationTaster/MetaSVM/LIST-S2) und Spezifität (97,45 %: MutationTaster/DEOGEN2/FATHMM.XF). Unterschiede zwischen Entitäten sowie zwischen Onkogenen und Tumorsupressoren wurden beobachtet.

Schlussfolgerung: Kombinationen von Algorithmen können genetische Analysen verbessern. Die Auswahl der Tools sollte im Hinblick auf Entität, Gen und Ziel der Analyse erfolgen.

Keywords: Cancer of the upper urinary tract; Next-generation sequencing; Pathogenicity prediction; Urothelial Carcinoma; Variant interpretation.

PubMed Disclaimer

Conflict of interest statement

Declarations. Conflict of interest: J. Möller, L. Seillier, A. Fürstberger, M. Rose, D.D. Jonigk, N. Ortiz-Brüchle, and N.T. Gaisa declare that they have no competing interests. Ethics approval and consent to participate: Not applicable, as the study involved use of public data only. Consent for publication: All authors have read and revised the final version of the manuscript. For this article no studies with human participants or animals were performed by any of the authors. All studies mentioned were in accordance with the ethical standards indicated in each case. The supplement containing this article is not sponsored by industry.

References

    1. Am J Hum Genet. (2016). https://doi.org/10.1016/j.ajhg.2016.08.016
    1. Bioinformatics. (2015). https://doi.org/10.1093/bioinformatics/btv195
    1. Bioinformatics. (2018). https://doi.org/10.1093/bioinformatics/btx536
    1. Bioinformatics. (2021). https://doi.org/10.1093/bioinformatics/btab529
    1. BMC Genomics. (2013). https://doi.org/10.1186/1471-2164-14-S3-S7

MeSH terms

LinkOut - more resources