Department of Pathology, Division of Laboratories, Pharmacy and Biomedical Genetics, University Medical Center Utrecht, P.O. Box 85500, 3508, Utrecht, GA, Netherlands.
Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, Netherlands.
Virchows Arch. 2021 Oct;479(4):773-783. doi: 10.1007/s00428-021-03095-5. Epub 2021 Apr 13.
Cutaneous intermediate melanocytic neoplasms with ambiguous histopathological features are diagnostically challenging. Ancillary cytogenetic techniques to detect genome-wide copy number variations (CNVs) might provide a valuable tool to allow accurate classification as benign (nevus) or malignant (melanoma). However, the CNV cut-off value to distinguish intermediate lesions from melanoma is not well defined. We performed a systematic review and individual patient data meta-analysis to evaluate the use of CNVs to classify intermediate melanocytic lesions. A total of 31 studies and 431 individual lesions were included. The CNV number in intermediate lesions (median 1, interquartile range [IQR] 0-2) was significantly higher (p<0.001) compared to that in benign lesions (median 0, IQR 0-1) and lower (p<0.001) compared to that in malignant lesions (median 6, IQR 4-11). The CNV number displayed excellent ability to differentiate between intermediate and malignant lesions (0.90, 95% CI 0.86-0.94, p<0.001). Two CNV cut-off points demonstrated a sensitivity and specificity higher than 80%. A cut-off of ≥3 CNVs corresponded to 85% sensitivity and 84% specificity, and a cut-off of ≥4 CNVs corresponded to 81% sensitivity and 91% specificity, respectively. This individual patient data meta-analysis provides a comprehensive overview of CNVs in cutaneous intermediate melanocytic lesions, based on the largest pooled cohort of ambiguous melanocytic neoplasms to date. Our meta-analysis suggests that a cut-off of ≥3 CNVs might represent the optimal trade-off between sensitivity and specificity in clinical practice to differentiate intermediate lesions from melanoma.
具有模棱两可组织病理学特征的皮肤交界性黑色素细胞肿瘤的诊断具有挑战性。辅助细胞遗传学技术可检测全基因组拷贝数变异（CNV），这可能为准确分类良性（痣）或恶性（黑色素瘤）提供有价值的工具。然而，区分交界性病变与黑色素瘤的 CNV 截止值尚未明确。我们进行了系统评价和个体患者数据荟萃分析，以评估 CNV 在分类交界性黑色素细胞病变中的应用。共纳入 31 项研究和 431 个个体病变。交界性病变的 CNV 数量（中位数 1，四分位距 [IQR] 0-2）明显高于（p<0.001）良性病变（中位数 0，IQR 0-1），也明显低于（p<0.001）恶性病变（中位数 6，IQR 4-11）。CNV 数量可极好地区分交界性和恶性病变（0.90，95%置信区间 [CI] 0.86-0.94，p<0.001）。两个 CNV 截止值的敏感性和特异性均高于 80%。CNV 数量≥3 对应的敏感性和特异性分别为 85%和 84%，CNV 数量≥4 对应的敏感性和特异性分别为 81%和 91%。本个体患者数据荟萃分析基于迄今为止最大的混合交界性黑色素瘤队列，全面概述了皮肤交界性黑色素细胞病变中的 CNV。我们的荟萃分析表明，CNV 数量≥3 可能是区分交界性病变与黑色素瘤的最佳敏感性与特异性折衷点，适用于临床实践。
