Hidradenitis Suppurativa Phenotypes: Validation of a Latent Class Model to Identify Specific Clusters of Patients

Introduction: Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease with heterogeneity in clinical presentation and associated comorbidities. Existing classification systems mainly assess severity and do not adequately capture phenotypic variation. We previously developed a 3-class phenotype model based on latent class analysis (LCA). The objective of this study was to externally validate the previously developed phenotypic LCA model in an independent group of patients. Methods: In this cross-sectional multicenter analysis, consecutive HS patients seen between January 2023 and December 2024 at three referral centers in Italy and Switzerland were included. Clinical, demographic, comorbidity, severity, treatment, and quality-of-life data were collected. Validation was performed by applying the original LCA model to the new dataset and comparing the classification with dermatologist judgement. In addition, new 3-class and 4-class LCA models were trained and compared. A freely accessible web application based on the original LCA model was also developed. Results: A total of 420 patients were included (55.7% females, mean age 32.4 years), with a similar distribution across Hurley disease stages. The mean disease duration was 10.1 years and the mean diagnostic delay 6.6 years. The original LCA assigned 8.6%, 32.6%, and 58.8% of patients to classes I, II and III, respectively. Dermatologists assigned 20.0%, 26.4%, and 33.3% to these classes, while 20.2% were non-classifiable. The agreement between the original LCA and dermatologist classification was fair (AC1=0.35; accuracy=55.2%). A new 3-class LCA again showed the most stable fit and partially reproduced the original structure (AC1=0.50). A 4-class model did not substantially improve agreement. Conclusion: The original LCA framework showed partial reproducibility and modest concordance with dermatologist-based classification. These findings support its use as a research-oriented framework to structure HS heterogeneity, although further refinement and validation are needed before routine clinical implementation.

Authors

Publication Details

Journal
Dermatology
Published
2026-11-19
DOI
https://doi.org/10.1159/drm/ackag034
Primary Topic
Hidradenitis Suppurativa and Treatments
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Hidradenitis Suppurativa Phenotypes: Validation of a Latent Class Model to Identify Specific Clusters of Patients

Vincenzo Bettoli, Simone Cazzaniga, S. Morteza Seyed Jafari, DAMIANO D ABENI et al.
Dermatology
Hidradenitis Suppurativa and Treatments
article

Hidradenitis Suppurativa Phenotypes: Validation of a Latent Class Model to Identify Specific Clusters of Patients

Vincenzo Bettoli, Simone Cazzaniga, S. Morteza Seyed Jafari, DAMIANO D ABENI, Alexandra Junge, Luigi Naldi, Robert Emil Hunger
article en

Abstract

Introduction: Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease with heterogeneity in clinical presentation and associated comorbidities. Existing classification systems mainly assess severity and do not adequately capture phenotypic variation. We previously developed a 3-class phenotype model based on latent class analysis (LCA). The objective of this study was to externally validate the previously developed phenotypic LCA model in an independent group of patients. Methods: In this cross-sectional multicenter analysis, consecutive HS patients seen between January 2023 and December 2024 at three referral centers in Italy and Switzerland were included. Clinical, demographic, comorbidity, severity, treatment, and quality-of-life data were collected. Validation was performed by applying the original LCA model to the new dataset and comparing the classification with dermatologist judgement. In addition, new 3-class and 4-class LCA models were trained and compared. A freely accessible web application based on the original LCA model was also developed. Results: A total of 420 patients were included (55.7% females, mean age 32.4 years), with a similar distribution across Hurley disease stages. The mean disease duration was 10.1 years and the mean diagnostic delay 6.6 years. The original LCA assigned 8.6%, 32.6%, and 58.8% of patients to classes I, II and III, respectively. Dermatologists assigned 20.0%, 26.4%, and 33.3% to these classes, while 20.2% were non-classifiable. The agreement between the original LCA and dermatologist classification was fair (AC1=0.35; accuracy=55.2%). A new 3-class LCA again showed the most stable fit and partially reproduced the original structure (AC1=0.50). A 4-class model did not substantially improve agreement. Conclusion: The original LCA framework showed partial reproducibility and modest concordance with dermatologist-based classification. These findings support its use as a research-oriented framework to structure HS heterogeneity, although further refinement and validation are needed before routine clinical implementation.

Dermatology
Hidradenitis Suppurativa and Treatments
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon