Pattern analysis of multi component melanoma on dermoscopy
The 2017 South Coast Dermoscopy skin lesion recognition and dermoscopy course has just started with a group of about 75 GPs, dermatology trainees, dermatologists, plastic surgeons, nurses and podiatrists assembling at The Holiday Inn Fareham last Friday.
This is the first case I sent out to our on line case discussion group. I like to put up some easy cases to start with. No plain view to hand. In fact I posted this here a few years back, but never mind, its a nice case.
This is an obvious melanoma, at one glance. Professor Peter Soyer calls this the heuristic method, after ‘Eureka!’ which means ‘I have it!’. OK, but why is it a melanoma?
Any of the accepted diagnostic methods will get us there. Let’s start with the 2 step algorithm.
step 1: is it melanocytic? There are brown globules (clods in the new terminology) particularly from 6 to 9 o’clock. Brown globules are a positive feature for melanocytic lesions. Can we see features of any non-melanocytic lesion (i.e. haemangioma, seborrhoeic keratosis, dermatofibroma etc)? No. So it is melanocytic by default and also on positive grounds (brown globules).
step 2; could it be a melanoma? Yes, because of gross asymmetry and multiple patterns and colours. Therefore excision.
The 3 step pathway. In this, we look for asymmetry, blue-white structures, and atypical network. Our lesion here shows the grossest of asymmetry, not so much of shape but of colour and patterns (which are both more important than geometry). We can see blue-white structures (also called blue white veil) and we can also see atypical network, in this case inverse network (the lacy white net-like structure at right and right of centre.) So again positive.
The 7 point check list of Argenziano. Please click on the link for a discussion of this 7 point check list. Again, positive. I can see gross asymmetry, irregular dots and globules, blue-white veil, pseudopods (basically a pseudopod is a blob of the end of a streak) an a few other structures.
Chaos and clues. This is chaotic as it has many structures and colours all mixed up. Step 2 of ‘chaos and clues’ is, having said ‘yes, chaos’ we look for clues, initially to ask the question ‘Is it a seborrheoic keratosis?’ Plainly not, the lesion lacks the fissured keratin, comedo like openings, milia like cysts, cerebriform appearance or looped vessels we expect to see in a seb k. When we look for melanoma clues, they are abundant (see above link to 7 point check list).
using SCOPE, our own mnemonic, we see no Symmetry, many COlours, many Patterns and odd things happening at the Edge.
The lesson being that whichever algorithm or scoring technique we use, this is a melanoma (as proved on histology, 3.5mm Breslow thickness). Professor Harald Kittler has said that experts learn all the algorithms and then invent their own. Sometimes one is better, sometimes another, sometimes a mixture.
Pattern analysis is defined as the simultaneous evaluation of all dermoscopic features of a lesion. The link takes you to a useful discussion on dermnetnz.