Dataset Information

Immunohistochemical Assays for Bladder Cancer Molecular Subtyping: Optimizing Parsimony and Performance of Lund Taxonomy Classifiers.

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 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.

SUBMITTER: Hardy CSC

PROVIDER: S-EPMC9058369 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Immunohistochemical Assays for Bladder Cancer Molecular Subtyping: Optimizing Parsimony and Performance of Lund Taxonomy Classifiers.

Hardy Céline S C CSC Ghaedi Hamid H Slotman Ava A Sjödahl Gottfrid G Gooding Robert J RJ Berman David M DM Jackson Chelsea L CL

The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 20220419 5

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 ...[more]

PMID: 35437049

Similar Datasets

Project description:ObjectivesRadiomics-based analyses encompass multiple steps, leading to ambiguity regarding the optimal approaches for enhancing model performance. This study compares the effect of several feature selection methods, machine learning (ML) classifiers, and sources of radiomic features, on models' performance for the diagnosis of clinically significant prostate cancer (csPCa) from bi-parametric MRI.MethodsTwo multi-centric datasets, with 465 and 204 patients each, were used to extract 1246 radiomic features per patient and MRI sequence. Ten feature selection methods, such as Boruta, mRMRe, ReliefF, recursive feature elimination (RFE), random forest (RF) variable importance, L1-lasso, etc., four ML classifiers, namely SVM, RF, LASSO, and boosted generalized linear model (GLM), and three sets of radiomics features, derived from T2w images, ADC maps, and their combination, were used to develop predictive models of csPCa. Their performance was evaluated in a nested cross-validation and externally, using seven performance metrics.ResultsIn total, 480 models were developed. In nested cross-validation, the best model combined Boruta with Boosted GLM (AUC = 0.71, F1 = 0.76). In external validation, the best model combined L1-lasso with boosted GLM (AUC = 0.71, F1 = 0.47). Overall, Boruta, RFE, L1-lasso, and RF variable importance were the top-performing feature selection methods, while the choice of ML classifier didn't significantly affect the results. The ADC-derived features showed the highest discriminatory power with T2w-derived features being less informative, while their combination did not lead to improved performance.ConclusionThe choice of feature selection method and the source of radiomic features have a profound effect on the models' performance for csPCa diagnosis.Critical relevance statementThis work may guide future radiomic research, paving the way for the development of more effective and reliable radiomic models; not only for advancing prostate cancer diagnostic strategies, but also for informing broader applications of radiomics in different medical contexts.Key pointsRadiomics is a growing field that can still be optimized. Feature selection method impacts radiomics models' performance more than ML algorithms. Best feature selection methods: RFE, LASSO, RF, and Boruta. ADC-derived radiomic features yield more robust models compared to T2w-derived radiomic features.

Project description:The global decline of coral reefs has driven considerable interest in active coral restoration. Despite their importance and dominance on mature reefs, relatively few coral restoration projects use slower growth forms like massive and encrusting coral species. Micro-fragmentation can increase coral cover by orders of magnitude faster than natural growth, which now allows cultivation of slow growing massive forms and shows promise and flexibility for active reef restoration. However, the major causes of variation in growth and survival of outplanted colonies remain poorly understood. Here, we report simple outplanting assays to aid in active reef restoration of slower growing species and increase the likelihood of restoration success. We used two different micro-fragmentation assays. Pyramid assays were used to examine variation associated with fragment size (ranging from ≈1-9 cm2), nursery residence time (for both in-situ and ex-situ nurseries), and 2D vs. 3D measurements of growth. Block assays were used to examine spatial variation among individual performance at outplanting sites in the field. We found 2D and 3D measurements correlated well, so measured survivorship and growth using top-down planar images for two of the main Hawaiian reef building corals, the plating Montipora capitata and the massive Porites compressa. Pyramid assays housed and outplanted from the in-situ nursery showed no effect of residence time or size on overall survivorship or growth for either species. Results from the ex-situ nursery, however, varied by species, with P. compressa again showing no effect of nursery residence time or size on survivorship or growth. In contrast, nursery culture resulted in improved survivorship of small M. capitata fragments, but net growth showed a weak positive effect of nursery time for medium fragments. Small fragments still suffered higher mortality than either medium or large fragments. Due to their lower mortality, medium fragments (≈3 cm2) appear to be the best compromise between growth and survivorship for outplanting. Likewise, given weak positive gains relative to the investment, our results suggest that it could be more cost-effective to simply outplant medium fragments as soon as possible, without intermediate culture in a nursery. Furthermore, the block assay revealed significant differences in survivorship and growth among sites for individuals of both species, emphasizing the importance of considering spatial variation in coral survival and growth following outplanting. These results highlight the value of using short-term micro-fragmentation assays prior to outplanting to assess size, and location specific performance, optimizing the efficiency of active reef restoration activities and maximizing the probability of success for active coral restoration projects.

Dataset Information

Immunohistochemical Assays for Bladder Cancer Molecular Subtyping: Optimizing Parsimony and Performance of Lund Taxonomy Classifiers.

Publications

Immunohistochemical Assays for Bladder Cancer Molecular Subtyping: Optimizing Parsimony and Performance of Lund Taxonomy Classifiers.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets