Digital dermoscopic changes during follow-up of de-novo and nevus-associated melanoma: a cohort study.

2020 
BACKGROUND: Nevus-associated melanoma (NAM) has been regarded as a distinct biological entity from de-novo melanoma (DNM); however, static dermoscopy often fails in differentiating these entities. Digital dermoscopic monitoring allows to identify dynamic changes occurring during follow-up; this may improve diagnostic accuracy and potentially our knowledge on NAM biology. We aimed to define main independent factors associated with NAM diagnosis and those influencing follow-up time in a population of melanomas excised at follow-up. METHODS: A cohort of melanomas excised at follow-up was retrospectively and consecutively selected. NAMs and DNMs were compared according to baseline features and main dermoscopic changes occurring during follow-up. Univariate and multivariable logistic and Cox's regression analysis were performed to respectively define factors associated with NAM diagnosis and those influencing the risk for excision. RESULTS: Eighty-six melanomas were enrolled, of which 21 (24.4%) were nevus-associated. During follow-up NAMs mainly underwent atypical network modifications (47.6%), followed by inverse network (28.6%) and dermoscopic island (23.8%) worsening or appearance. DNMs were also mainly characterized by atypical network modifications (47.7%), however, a significant proportion of cases underwent irregular pigmentation/dots/globules or regression changes (29.2%), which were rarely seen among NAMs. Furthermore, both multivariable logistic and Cox's regression analysis demonstrated a significant association between NAM and a longer follow-up. CONCLUSIONS: We demonstrated that among melanomas excised at follow-up, different patterns of dermoscopic changes may be found between NAMs and DNMs. This finding, together with the association of NAM with a longer follow-up time, supports the hypothesis of different biological behavior of these two entities.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    20
    References
    3
    Citations
    NaN
    KQI
    []