Immunohistochemical Assays for Bladder Cancer Molecular Subtyping : Optimizing Parsimony and Performance of Lund Taxonomy Classifiers
(2022) In Journal of Histochemistry and Cytochemistry 70(5). p.357-375- Abstract
Transcriptomic and proteomic profiling classify bladder cancers into luminal and basal molecular subtypes, with controversial prognostic and predictive associations. The complexity of published subtyping algorithms is a major impediment to understanding their biology and validating or refuting their clinical use. Here, we optimize and validate compact algorithms based on the Lund taxonomy, which separates luminal subtypes into urothelial-like (Uro) and genomically unstable (GU). We characterized immunohistochemical expression data from two muscle-invasive bladder cancer cohorts (n=193, n=76) and developed efficient decision tree subtyping models using 4-fold cross-validation. We demonstrated that a published algorithm using routine... (More)
Transcriptomic and proteomic profiling classify bladder cancers into luminal and basal molecular subtypes, with controversial prognostic and predictive associations. The complexity of published subtyping algorithms is a major impediment to understanding their biology and validating or refuting their clinical use. Here, we optimize and validate compact algorithms based on the Lund taxonomy, which separates luminal subtypes into urothelial-like (Uro) and genomically unstable (GU). We characterized immunohistochemical expression data from two muscle-invasive bladder cancer cohorts (n=193, n=76) and developed efficient decision tree subtyping models using 4-fold cross-validation. We demonstrated that a published algorithm using routine assays (GATA3, KRT5, p16) classified basal/luminal subtypes and basal/Uro/GU subtypes with 86%–95% and 67%–86% accuracies, respectively. KRT14 and RB1 are less frequently used in pathology practice but achieved the simplest, most accurate models for basal/luminal and basal/Uro/GU discrimination, with 93%–96% and 85%–86% accuracies, respectively. More complex models with up to eight antibodies performed no better than simpler two- or three-antibody models. We conclude that simple immunohistochemistry classifiers can accurately identify luminal (Uro, GU) and basal subtypes and are appealing options for clinical implementation.
(Less)
- author
- Hardy, Céline S.C. ; Ghaedi, Hamid ; Slotman, Ava ; Sjödahl, Gottfrid LU ; Gooding, Robert J. ; Berman, David M. and Jackson, Chelsea L.
- organization
- publishing date
- 2022
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- CDKN2A, histopathology, LundTax, urothelial carcinoma
- in
- Journal of Histochemistry and Cytochemistry
- volume
- 70
- issue
- 5
- pages
- 19 pages
- publisher
- Histochemical Society
- external identifiers
-
- scopus:85129206215
- pmid:35437049
- ISSN
- 0022-1554
- DOI
- 10.1369/00221554221095530
- language
- English
- LU publication?
- yes
- id
- e945f051-9794-4ea6-b69b-9a83805fce3b
- date added to LUP
- 2022-07-07 12:36:51
- date last changed
- 2024-06-13 14:07:18
@article{e945f051-9794-4ea6-b69b-9a83805fce3b,
abstract = {{<p>Transcriptomic and proteomic profiling classify bladder cancers into luminal and basal molecular subtypes, with controversial prognostic and predictive associations. The complexity of published subtyping algorithms is a major impediment to understanding their biology and validating or refuting their clinical use. Here, we optimize and validate compact algorithms based on the Lund taxonomy, which separates luminal subtypes into urothelial-like (Uro) and genomically unstable (GU). We characterized immunohistochemical expression data from two muscle-invasive bladder cancer cohorts (n=193, n=76) and developed efficient decision tree subtyping models using 4-fold cross-validation. We demonstrated that a published algorithm using routine assays (GATA3, KRT5, p16) classified basal/luminal subtypes and basal/Uro/GU subtypes with 86%–95% and 67%–86% accuracies, respectively. KRT14 and RB1 are less frequently used in pathology practice but achieved the simplest, most accurate models for basal/luminal and basal/Uro/GU discrimination, with 93%–96% and 85%–86% accuracies, respectively. More complex models with up to eight antibodies performed no better than simpler two- or three-antibody models. We conclude that simple immunohistochemistry classifiers can accurately identify luminal (Uro, GU) and basal subtypes and are appealing options for clinical implementation.</p>}},
author = {{Hardy, Céline S.C. and Ghaedi, Hamid and Slotman, Ava and Sjödahl, Gottfrid and Gooding, Robert J. and Berman, David M. and Jackson, Chelsea L.}},
issn = {{0022-1554}},
keywords = {{CDKN2A; histopathology; LundTax; urothelial carcinoma}},
language = {{eng}},
number = {{5}},
pages = {{357--375}},
publisher = {{Histochemical Society}},
series = {{Journal of Histochemistry and Cytochemistry}},
title = {{Immunohistochemical Assays for Bladder Cancer Molecular Subtyping : Optimizing Parsimony and Performance of Lund Taxonomy Classifiers}},
url = {{http://dx.doi.org/10.1369/00221554221095530}},
doi = {{10.1369/00221554221095530}},
volume = {{70}},
year = {{2022}},
}