Risk Prediction Models for Melanoma: A Systematic Review on the Heterogeneity in Model Development and Validation

Abstract

The rising incidence of cutaneous melanoma over the past few decades has prompted substantial efforts to develop risk prediction models identifying people at high risk of developing melanoma to facilitate targeted screening programs. We review these models, regarding study characteristics, differences in risk factor selection and assessment, evaluation, and validation methods. Our systematic literature search revealed 40 studies comprising 46 different risk prediction models eligible for the review. Altogether, 35 different risk factors were part of the models with nevi being the most common one (n = 35, 78%); little consistency in other risk factors was observed. Results of an internal validation were reported for less than half of the studies (n = 18, 45%), and only 6 performed external validation. In terms of model performance, 29 studies assessed the discriminative ability of their models; other performance measures, e.g., regarding calibration or clinical usefulness, were rarely reported. Due to the substantial heterogeneity in risk factor selection and assessment as well as methodologic aspects of model development, direct comparisons between models are hardly possible. Uniform methodologic standards for the development and validation of risk prediction models for melanoma and reporting standards for the accompanying publications are necessary and need to be obligatory for that reason.

Keywords: melanoma; risk prediction; statistical models; validation.

Figure 1

Flow diagram for the identification of studies developing risk prediction models for melanoma.

Figure 2

Geographical location of studies. (a) Distribution of studies according to the continents of their origin. (b) Distribution of studies according to their country of origin. (n = 40 studies). * Four studies used data sets from multiple countries with high melanoma incidences for the development of their risk prediction model.

Figure 3

Temporal distribution of the reviewed studies showing the number of publications in eight time intervals of four years each. (n = 40 studies).

Figure 4

Heatmap indicating joint occurrences of risk factor pairs in risk prediction models for melanoma. Only risk factors occurring in more than two risk prediction models are included. Each number represents the absolute frequency of the corresponding risk factor combination. The darker the field the more frequent is the corresponding risk factor combination (n = 45 models). MC1R = melanocortin 1 receptor.