Truace
Subscribe

Machine-learning-based Phenomapping of Patients with Keratinocyte Carcinoma: Data-driven Subgrouping by Disease Burden, Comorbidities and Socioeconomic Status

Keratinocyte carcinoma (KC) places a considerable and growing burden on healthcare systems. Given the KC population's heterogeneity, tailored clinical pathways are needed to accommodate diverse management needs. This study applied machine learning (ML)-based phenomapping to identify distinct real-world subgroups within a national KC population using demographic and medical history variables. The study included KC patients treated in publicly-funded, office-based dermatology practices and registered in the Danish…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-0768· cite & verify
Machine-learning-based Phenomapping of Patients with Keratinocyte Carcinoma: Data-driven Subgrouping by Disease Burden, Comorbidities and Socioeconomic Status
Military Dermatology by U.S. Army. Office of The Surgeon General. Borden Institute. Public domain

In brief

On 2026-08-13, a peer-reviewed study reported machine-learning-based phenomapping of 106,490 keratinocyte carcinoma patients from the Danish Skin Cancer Registry (2014-2022). The model derived seven clusters ranging from young, well-educated, high-income, medically noncomplex females with low-risk BCCs to highly comorbid patients with more SCCs and immunosuppressive drug exposure.

The finding matters because keratinocyte carcinoma places a considerable and growing burden on healthcare systems and the population is heterogeneous. By stratifying patients by disease burden, comorbidities and socioeconomic status, the approach provides a basis for tailored clinical pathways with differentiated resource needs, though the source does not report prospective implementation or outcomes of such pathways.

Main points

  1. Study included 106,490 KC patients treated in publicly-funded, office-based dermatology practices and registered in the Danish Skin Cancer Registry from 2014-2022.
  2. One cluster (22.2%; n=23,590) was young, well educated high-income medically noncomplex females with low-risk basal cell carcinomas.
  3. Four clusters (55.2%; n=58,736) had greater dermatological disease burden including facial BCCs, multiple BCCs, more squamous cell carcinomas, and history of skin cancer or actinic keratosis treatment.
  4. Two clusters (22.7%; n=24,164) were highly comorbid patients with more SCCs, high number of KCs on head and neck and high immunosuppressive drug exposure.

The gain

ML-based phenomapping of 106,490 Danish KC patients identified 7 distinct subgroups differentiated by disease burden, comorbidities and socioeconomic status, providing a basis for tailored clinical pathways.

The rundown

The analysis used demographic and medical history variables from the Danish Skin Cancer Registry covering 2014-2022 to perform data-driven subgrouping. The population was drawn from publicly-funded, office-based dermatology practices, reflecting real-world care.

The seven clusters separate along age, sex, education, income, comorbidity, immunosuppressive exposure, tumor type (BCC vs SCC), location, multiplicity, and prior actinic keratosis-related treatment, illustrating heterogeneity beyond single risk factors.

Sources

  1. Peer-reviewedActa Dermato-Venereologica2026-08-13

The debate