Prognostic prediction models for rare skin cancer outcomes: a practical guide to handling nested-case control and case-cohort data (with care)

Background: Prognostic prediction models, whether based on classical statistics or artificial intelligence, are becoming increasingly popular in the medical community as an alternative or complement to more traditional tools, such as cancer staging systems. Population-based cohorts are the preferred study design for building and validating such models. However, they are generally expensive and frequently unfeasible, particularly when the outcome of interest is rare, and when the collected data include expensive or difficult-to-obtain biomarkers.