Glowing in the dark: case report of a clue-poor melanoma unmasked by polarized dermatoscopy

We report a case of a melanoma arising in a congenital-type compound nevus, which was excised because it was observed by both the patient and the treating dermatologist to have changed. Because the lesion was routinely photo-documented with both polarized and non-polarized dermatoscopy images prior to excision, these images were available for subsequent examination. Matched images are presented in what appears to be unique in the published literature: polarizing-specific white lines are identified as a compelling clue to the diagnosis of melanoma in a lesion that contains no clues apparent in the non-polarized image. Dermatopathology images reveal that the melanoma is arising in conjunction with a congenital type nevus. As expected, dermatoscopic polarizing-specific white lines are evident on the melanoma but not the nevus, and while a possible explanation is discussed, this remains speculative.

The lesion was re-excised with one a centimeter peripheral clearance margin, deep to but not including muscle    here is the so-called "cobblestone" pattern (pattern of clods) described as the pattern of a congenital type nevus and this lesion had none of the specific criteria of melanoma according to that method. Likewise this lesion did not score as a melanoma according to the ABCD dermatoscopic algorithm [9], the 3-point checklist [10], the 7-point checklist [11], the Menzies method [12] or the CASH algorithm [13]. The "Chaos & Clues" algorithm [14,15] identifies suspicion for malignancy based on the presence of chaos (defined as asymmetry of structure and/or color) plus the presence of at least one of eight clues, including the clue of "white lines," and therefore that method would identify this lesion as suspicious but only if polarized dermatoscopy was employed.
PSWL, which were the critical diagnostic feature in this case, have been attributed to the presence of collagen in the context of dermal fibrosis, which has birefringent properties causing rapid randomization of polarized light thus making the collagen more conspicuous [5]. Dermatopathologic collagen bands are in fact evident in this case (Figures 4 and 5).
In attempting to explain the presence of PSWL in melanoma but not nevi, it has been proposed that basic fibroblast growth factor (bFGF), maximally expressed at the advancing front of the neoplasm adjacent to fibrotic changes in the dermis may be associated with tightly woven collagen bundles encircling microinvasive melanoma nests [5] and that these structures may correlate with PSWL. In melanomas, melanocyte-induced de novo collagen 1 type synthesis can be identified with Sirius red staining [5], which has not been demonstrated in nevi [16].
While these factors may play a role in the differential presence of PSWL in melanoma, but not in the associated nevus, the actual dermatopathological correlation remains speculative.
This case illustrates the critical role that polarized dermatoscopy can play and argues strongly for its inclusion in the clinical routine. We also suggest that the observation of polarizing-specific white lines should lead to consideration of excision biopsy in any lesion when either nevus or melanoma is in the differential diagnosis. If polarizing-specific white lines are subtle, they may be best appreciated by comparing the polarized and non-polarized view. For this reason we suggest that both polarizing and non-polarizing modes of dermatoscopy should be employed routinely, this practice being facilitated by dermatoscopes that can be conveniently switched between modes. The polarized image (Figures 2 and 3B) shows the same basic pattern as the non-polarized image, but in addition there is a pattern of white lines, whiter than surrounding skin color, and predominantly oriented in a perpendicular arrangement.
Because these white lines are not present in the non-polarized dermatoscopy image, they can be described with certainty as dermatoscopic polarizing-specific white lines (PSWL). In the polarized images hairs are visible peripherally but they are absent over the central area where the polarizing-specific white lines are present (Figures 2 and 3B).
PSWL are a published clue to melanoma, Spitz nevus, basal cell carcinoma, dermatofibroma, scar tissue and benign lichenoid keratosis [5]. Although there is a large list of differential diagnoses for lesions with this clue, it is notable that if this clue is seen in a melanocytic lesion that is not a Spitz nevus, then the diagnosis of melanoma can be expected. It is also known that polarizing specific white lines are more frequently seen in melanomas, which are invasive [5]. The likelihood of Spitz nevus decreases with increasing age [6], and some authors recommend the excision of all Spitz nevi at any age due the fact that the distinction from melanoma can be challenging dermatopathologically [7]. In the largest study evaluating PSWL, of the non-biopsied lesions, none of 9750 Clark nevi, 90 congenital melanocytic nevi (CMN), 168 intradermal nevi (IDN) and 15 blue nevi displayed this structure, while of the biopsied lesions one of 20 Clark nevi, two of 10 IDN and none of one CMN displayed PSWL [5].
According to the original dermatoscopic method of classic pattern analysis [8], the global pattern of the lesion presented