Diagnostic discordance among histopathological reviewers of melanocytic lesions

Gregory A. Hosler,Matthew S. Goldberg,Sarah I. Estrada,Brendan O'Neil,Sapna M. Amin,Jose A. Plaza
DOI: https://doi.org/10.1111/cup.14635
2024-05-10
Journal of Cutaneous Pathology
Abstract:Background Histopathological examination is adequate for the diagnosis of most cutaneous melanocytic neoplasms. However, there is a subset that is either difficult to definitively diagnose or would have diagnostic disagreement upon review by multiple dermatopathologists if a more exhaustive review was performed. Methods Melanocytic lesions underwent an independent, blinded diagnostic histopathological review of hematoxylin and eosin‐stained sections. Each lesion was reviewed by three to six dermatopathologists and categorized as benign, malignant, or unknown malignant potential (UMP). Diagnoses were grouped as concordant (all the same designation); opposing (received benign and malignant designations); majority (single designation with the highest number of diagnoses, no benign/malignant opposing designations); and non‐definitive (equal number of non‐opposing designations [i.e., benign/UMP or malignant/UMP]). Lesions with equivocal designations (concordant or majority UMP, opposing, majority, and non‐definitive) were utilized in a patient treatment model of projected surgical treatment discrepancies. Results In total, 3317 cases were reviewed, and 23.8% of lesions received equivocal diagnoses. Of these, 7.3% were majority benign, 4.8% were majority malignant, 2.7% were majority UMP, 0.5% were concordant UMP, 6.9% were opposing, and 1.6% were non‐definitive. Patient treatment models of those with equivocal lesions (n = 788) revealed a potential of overall surgical treatment variations ranging from 18% to 72%, with the highest variation amongst lesions with opposing, non‐definitive, or majority UMP (40%–72%) diagnoses. Conclusion Histopathologic review in this large cohort demonstrated substantial diagnostic variation, with 23.8% of cases receiving equivocal diagnoses. We identified diagnostic ambiguity even in lesions where a definitive diagnosis was previously rendered by a single real‐world dermatopathologist. The combined clinical impact of diagnostic discordance or a final diagnosis of UMP is highlighted by high diagnosis‐dependent treatment variation in the patient treatment model, which could be underreported in a real‐world setting, where review by more than one to two dermatopathologists is relatively rare.
dermatology,pathology
What problem does this paper attempt to address?