Potential for melanoma diagnosis “over-calling” and “under-calling” from variation in histopathology interpretation

ABSTRACT Background: Pathologist disagreement over the diagnosis of melanoma may harm individuals and healthcare systems.Objective: To estimate the extent of “over-calling” and “under-calling” of melanoma for a population undergoing one excision per person and to model the impact of potential solutions.Methods: We performed an epidemiological modelling study, using data from population-based and diagnostic accuracy studies. We simulated biopsy results for 10,000 patients and modelled the impact of an intervention to improve pathologist agreement on “over-calling” and “under-calling” of melanoma. We then estimated the impact of four potential second review strategies for borderline lesions on the number requiring expert dermatopathologist review.Results: For 10,000 individuals undergoing excision of one melanocytic lesion, interpreted by a community pathologist, an intervention to improve histopathology agreement reduced the number of benign lesions “over-called” from 308 to 164 and the number of melanomas “under-called” from 289 to 240. If all borderline lesions were sent for expert review, the number of referrals would decrease from 1,500 to 737 cases if formal second community pathologist was used, and to 701 cases if a hypothetical intervention was additionally used. Limitations: Hypothetical intervention and post-test probabilities used to rule in or rule out melanoma are undefined.Conclusions: Interventions to improve agreement in the classification of melanocytic lesions and formal second community review may reduce the rates of “over-calling” and “under-calling” of melanoma.Supplementary TableseTable 1: Summary of the MPATH-Dx reporting schema for classifying melanocytic lesionseTable 2: Definitions and formulae used to calculate interval likelihood ratios and post-test probabilitieseTable 3: Intervention Effect Models for each Hypothetical ScenarioeTable 4: Likelihood ratios of melanoma for a hypothetical intervention to increase community pathologist diagnostic agreement with increasing effect sizes (Models 1 to 5)eTable 5: Pre and post-test probabilities of melanoma for a hypothetical intervention to increase community pathologist diagnostic agreement with increasing effect sizes (Models 1 to 5)Supplementary FigureseFigure 1: Histological appearance of MPATH-Dx classes II, III, IV and V. a) MPATH-Dx class II b) MPATH-Dx class III MPATH-Dx class IV MPATH-Dx class VeFigure 2: Pathologist diagnosis of versus expert reference diagnosis with and without an intervention to improve diagnostic agreement (Model 1 and Model 4)eFigure 3A – F: Pre and Post Test Probabilities for each MPATH-Dx category to predict melanoma, using single community pathologist review, with an intervention to improve diagnostic agreement. A. No intervention, B. Model 1, C. Model 2, D. Model 3, E. Model 4, F. Model 5THIS DATASET IS ARCHIVED AT DANS/EASY, BUT NOT ACCESSIBLE HERE. TO VIEW A LIST OF FILES AND ACCESS THE FILES IN THIS DATASET CLICK ON THE DOI-LINK ABOV