Dermoscopy of pigmented basal cell carcinoma

Abstract

While the incidence varies with race, 7% of basal cell carcinomas (BCCs) are pigmented.1,2 When present, melanin is preferably seen in the superficial component of the tumor.1 Within the tumor mass, melanin can be confined to melanocytes or additionally taken up in surrounding tumor cells.3,4 As well, melanin can be found in dermal melanophages and in hyperplastic epidermal melanocytes in the overlying epidermis.1 By using dermoscopy, pigmented BCCs are always asymmetric in pattern and are relatively hypomelanotic lesions. Two thirds have 50% of their tumor area pigmented, and only 7% have 75% of their area pigmented.5 Because of their irregularity, the differential diagnosis of pigmented BCCs includes both invasive melanoma and benign pigmented lesions. With this in mind, a dermoscopy model has been previously reported to distinguish pigmented BCCs from both of these diagnostic groups.5 This model has been incorporated into the two-step procedure for the dermoscopic diagnosis of pigmented skin lesions (PSLs).6 For the model development, 142 pigmented BCCs were compared with 142 invasive melanomas and 142 benign PSLs (including nonmelanocytic lesions). This large lesion set was divided into an equal-sample-size training and test set. In the training set, each lesion was scored for 45 dermoscopy features from which a diagnostic method was created. Features were used in the model when they were found in a significantly different frequency in pigmented BCCs versus both invasive melanoma and benign PSLs. The dermoscopic method for diagnosing pigmented BCCs is shown in Table 1. Here, for a pigmented BCC to be diagnosed, it must not have the negative feature of pigment network and must have one or more of the following six positive features: (1) large, gray-blue ovoid nests; (2) multiple gray-blue globules; (3) mapleleaf-like areas; (4) spoke wheel areas; (5) ulceration; and (6) arborizing “treelike” vessels. This method is reproducible and gives an overall sensitivity of 93% for BCCs and specificity of 89% (using the invasive melanoma set) and 92% (using the benign PSL set). Within the model, the negative feature of pigment network excludes the diagnosis of pigmented BCCs. Such a feature has high specificity for melanocytic lesions and was found in only 2.8% of BCCs.5 The majority of positive features are variants of pigmentation confined to well-circumscribed morphologic areas. Large, gray-blue ovoid nests are found in 55% of pigmented BCCs. They are confluent or near-confluent, pigmented, ovoid or elongated areas, larger than globules, and not intimately connected to a pigmented tumor body. Multiple gray-blue globules are seen in 27% of lesions. They should be distinguished from multiple gray-blue dots (melanophages), which are smaller and “pepperlike” in morphology. Mapleleaf-like areas, while present in only 17% of lesions, are highly specific (100%). They are brown to gray-blue, discrete, bulbous extensions forming a leaflike pattern. They differ from pseudopods (found in melanoma) because mapleleaflike areas are discrete pigment nests (islands) never arising from a pigment network and usually not arising from an adjacent confluent, pigmented area. The least frequent (10%) but highly specific (100%) feature of BCCs in the diagnostic model is the spoke wheel area. These are seen as radial projections, usually tan but sometimes blue or gray, meeting at an often darker central axis. Arborizing treelike telangiectases are found in 52% of lesions and are also a positive feature. The final positive feature of ulceration is found in 27%