More than 90% of blue nevi in adult patients are stable in size after a median of 4.5 years, according to study findings published in Journal of the European Academy of Dermatology and Venereology.
Researchers conducted a retrospective cohort study to characterize the evolution of blue nevi and assess the need for monitoring among patients from a tertiary melanoma referral center in Spain. The participants underwent follow-up with periodic digital monitoring from December 1998 to November 2019.
The blue nevi were identified with use of an artificial intelligence (AI) algorithm, and images of the lesions were obtained from the databases of 2 digital epiluminescence microscopy devices. A segmentation neural network was trained to diagnose the lesions from the temporal sequences in the HAM100001 5 segmentation dataset.
Growing blue nevi were considered as lesions with a percentage change in their area of more than 50% when comparing the calculated area of the first image and last image of follow-up. Lesions with a change in area of over 100% were excised and characterized histologically and with molecular genomic profiling.
A total of 123 blue nevi from 103 participants were included in the analysis. The patients were a median age of 48.3 years (IQR, 35.5-67.3) at diagnosis, and 54.5% were men. The most frequently occurring morphologic subtype was common blue nevus (73.2%).
The median follow-up of the blue nevi was 4.52 years (IQR, 2.61-8.89), with a median of 7 images per lesion. Of the lesions, 108 (87.8%) were segmented, of which 99 (91.7%) were considered stable with a global percentage change of 0.04% (IQR, -0.11 to 0.17).
Among the 9 (8.3%) growing blue nevi observed, 2 (1.85%) had progressive growth with a percentage change of more than 100% and were excised for further analysis. Case 1 had a follow-up of 19.53 years, and case 2 had a follow-up of 3.72 years. Histopathologic examination of the 2 lesions revealed cellular blue nevus areas. No malignant blue nevus lesions were observed.
Somatic single-nucleotide variants were identified in the 2 growing blue nevi, ranging from 0.8 to 1.6 mutations/megabase, which suggested a low tumor mutation burden. The researchers detected the common GNAQ c.626A>T alteration leading to a p.Gln209Leu protein change in both lesions with variant allele fractions of 1.9% and 11.9%, respectively.
In case 1, 6 variants involving PTPRT, PIK3C2B, SUZ12, CDH2, and HNF1A genes were categorized as putative driver mutations. No gene fusions or copy number variants were found in either blue nevus.
Limitations of the study include the fact that the lesions had few digitized images or were followed up for a short period, which may have resulted in an underestimation of the percentage of growing blue nevi. Also, identifying blue nevi with an AI algorithm may not be able to correctly detect rare clinical variants such as the amelanotic, combined, or hypomelanotic variants.
“No atypical histopathological signs or additional pathogenic mutations related to melanoma development or involved in malignant progression were identified in this subset of lesions,” the researchers conclude.
Alamon-Reig F, Combalia M, Albero-González R, et al. Analysis of dermoscopic changes of blue nevi on digital follow-up: a 21-year retrospective cohort study. J Eur Acad Dermatol Venereol. Published online January 25, 2023. doi:10.1111/jdv.18915