Correspondence to: Elizabeth D. Drugge, PhD, MPH, Department of Epidemiology and Community Health, New York Medical College School of Health Sciences and Practice, 40 Sunshine Cottage Rd, Valhalla, NY 10595
Two photographic approaches provide critical and complementary information in the differential diagnosis of malignant melanoma. Total body photography (TBP) traditionally provides baseline images from which macroscopic lesion changes can be detected, whereas digital epiluminescence (dermoscopic) microscopy reveals subtle changes in preexisting nevi.
Time and cost barriers restrict the use of both modalities to a select group of high-risk patients in pigmented skin lesion clinics.
Automation of TBP using a 25-camera array enables the routine capture of clinical images as an adjunct to a total body skin examination. Computer-assisted comparison of serial images exposes new and changed lesions, which are then photographed dermoscopically.
We describe the clinical, dermoscopic, and histopathologic features of melanomas ≤3 mm in size (micromelanomas) identified using this process.
We performed a retrospective study of 268 consecutive melanocytic lesions from which biopsy specimens were obtained from 218 patients between January 2015 and June 2016 in a single practice dermatology clinic. Lesion diameter was obtained from dermoscopic images taken before the biopsy specimens were obtained, and depth was obtained from pathology reports.
Eighty-one of 268 melanocytic lesions (30.22%) were melanoma in situ (MIS) or invasive melanoma (range 0.12-3.5; median 0.34), and 28 (34.57%) were ≤3 mm in diameter; 27 (33.33%) were >3 and ≤6 mm; and 26 (32.10%) were >6 mm. Of the lesions ≤3 mm, 21 (75.00%) were MIS and 7 (25.00%) were invasive (range 0.22-0.42; median 0.3 mm). Of lesions >3 and ≤6 mm, 13 (48.15%) were MIS and 14 (51.85%) were invasive (range 0.2-2.3; median 0.395 mm). Of lesions >6 mm, 16 (61.54%) were MIS and 10 (38.46%) were invasive (range 0.12-3.5; median 0.38 mm).
Nineteen of the 28 micromelanomas (68%) had diameters ≤2 mm and were sent for 2 additional blinded histopathologic assessments. Eleven of these (58%) were diagnosed as melanoma by all 3 dermatopathologists. The remaining 8 lesions were called melanoma by 1 pathologist and severely atypical by another with a comment that they could represent evolving melanoma and should be excised.
Time-lapse clinical images and the corresponding dermoscopic image of a melanoma can be seen in Fig 1. Dermoscopic features of chaos, clods, and amorphous areas were identified in all malignant lesions, but our sample size was not large enough to determine the relative value of each feature.
There are other reports of micromelanomas identified using various TBP and dermoscopy platforms with yields in the range of 23 (11%) of 206 lesions biopsied,
we found a significantly lower number needed to biopsy (3.1 vs 12.01), with a similar MIS: INV ratio (1.56:1). These studies suggest that routine comparison of complete sets of TBP images combined with dermoscopy can identify very small lesions of melanoma, some of which are already invasive.
Cutaneous melanoma in situ: translational evidence from a large population-based study.
Rhett Drugge, MD, is the inventor and holder of the intellectual property rights (US patent 7,359,748) of the Melanoscan system. E. Drugge is a first-degree relative of Dr Drugge. The other authors have no conflicts of interest to declare.
We thank George F. Murphy, MD, Director of Dermatopathology at BWH Consulting.