We performed a meta-analysis of all the published studies with BAP1-mutated families. To avoid the overestimation of the cancer risks estimates usually due to publication bias in meta-analyses, we chose to compare BAP1 mutated vs non-mutated patients from the same families. We demonstrate that germline BAP1 mutations are associated with a significant increased overall risk of cancer and particularly of MM, UVM and CM (Table 2). In addition, two thirds of patients from the germline BAP1-mutated cohort presented MBAITs, while these tumors were not detected in the non-mutated individuals. Our results reveal that germline BAP1 mutations cause a novel autosomal dominant hereditary cancer syndrome, the BAP1 cancer syndrome, characterized predominantly by MM, UVM, CM and by MBAITs and possibly by other cancers. In fact, because of the relatively high incidence of carcinomas in the general population, a larger number of BAP1-mutant family members have to be studied before ruling out the possibility that additional cancers are linked to the BAP1 cancer syndrome.
Early diagnosis is crucial for curative resection of CM and UVM. For its anatomical localization, MM is a tumor in which early diagnosis is particularly difficult; it is indeed often diagnosed in advanced stages when patients have median survivals of 6–12 months. However, when MM patients are diagnosed at Stage 1a, survivals of five or more years are not uncommon . Indeed, a high degree of suspicion allowed us to detect four MM in the L and W families at an early stage and these patients experienced survivals of 5–10 and, hopefully, many more years.
MBAITs provide physicians with a marker to identify individuals who may carry germline BAP1 mutations and thus are at high risk of developing CM, UVM and MM. We identified MBAITs among the majority of germline BAP1 mutation carriers in the L and W families. Our meta-analysis demonstrates that the prevalence of MBAITs is significantly higher in germline BAP1 mutation carriers compared to controls. MBAITs have variable papular macroscopic appearance similar to dermal nevi; nevertheless they present histological (Figure 2), immunohistochemical (Figure 2) and molecular (BAP1 and BRAF mutations) features that allow their characterization in the broad spectrum of melanocytic lesions. We debated how to call these tumors, and we concluded that it was best to give them a new name to make a clear distinction between these tumors, and other melanocytic lesions, such as Spitz nevus and ASTs. Briefly, Spitz nevus consists of proliferation of large epithelioid or spindle shaped melanocytes, or a mixture of the two. At all ages, spindle cells are the most common cell type. Spitz nevi composed wholly of epithelioid cells occur mainly in early childhood. Spitz nevi go through the same junctional, compound and intradermal phases as common acquired nevi, but most are removed when they are compound lesions (when they have an epidermal and dermal component). Spitz nevi are roughly symmetrical and at any given level (epidermis, junction, upper dermis, lower dermis) the lesion shows similar architecture and cell type from side to side. The cellularity of the lesion and the size of the cells and their nuclei decrease toward the base of the Spitz nevus (so called maturation) and this is associated with loss of proliferative activity that is instead often present in the upper parts of the nevus. Also the architecture at the base (deep aspect) is infiltrative rather than expansile/pushing, the nevus cells lying dispersed between dermal collagen. In addition there is no nevus associate with them (unless it is part of a combined nevus). ATS are Spitzoid lesions that have some features that overlap with melanoma making the differential diagnosis challenging: for example, there is no maturation towards the deeper part of the dermis, and instead the AST cells show mitotic activity.
Instead, the tumors found in these BAP1 mutated patients show large epithelioid clonal cells (that resemble those found in Spitz nevus and in ASTs) but these cells are present only in the dermis (there is no epidermal component). In contrast to Spitz nevi there is no maturation towards the deeper part of the dermis, and in contrast to ASTs, Ki67 stain (a marker of cell proliferation) consistently showed absence of mitotic figures. In almost all of these BAP1 associated tumors, it was possible to detect remnants of a nearby conventional intradermal or compound nevus formed by smaller cells in close proximity to the large epithelioid cells, a very unlikely finding in Spitz and ASTs. In addition, at the cytophatological level, the nuclei of the large clonal cells in these BAP1 tumors are hyperchromatic, while in Spitz and ATS, the large cells have an open nuclear chromatin and a small distinct nucleolus. Also no Kamino bodies –common in Spitz- are detected in the BAP1 associated melanocytic tumors. At the molecular level, these lesions are characterized by BAP1 inactivation and almost always by concurrent BRAF mutation, features that are not found in Spitz and ASTs. Finally, in contrast to AST that at times mask a melanoma, these lesions are benign in appearance and behavior and had a history of being stable in morphology, per the patients, although it has been proposed that these tumors, that here we propose to name MBAITs, may rarely directly evolve into a CM, .
In summary, MBAITs have “Spitzoid features”, such as large epitheliod and spindle cells, pleomorphisms, etc., but are sufficiently distinct morphologically, cytologically, molecularly and clinically from ASTs that they should not be lumped together. The entity of AST has become a catch-all phrase to describe a very heterogenous group of lesions clinically and histologically. This has resulted in overly aggressive approaches to their management, such as the use of sentinel lymph node biopsy. Molecular genetic studies will ultimately help us divide the entity of ASTs into subgroups based on genetics and ultimately behavior. This progress will be valuable in guiding clinical management. Therefore, we believe that if we continue to lump different melanocytic lesions (such as MBAITs) under the umbrella term of AST we will miss an opportunity to help pathologists to identify and classify this subgroup and to guide clinicians in selecting patients who may need testing for BAP1 mutations.
The youngest individuals positively screened for MBAITs in the L and W families were in their third and fourth decades of life, but all of them described to us these lesions as having been present for several years. Accordingly, Wiesner et al. reported that in the BAP1-mutant families they studied, MBAITs appeared during the first two decades of life and increased in number with age [7, 8]. Therefore, MBAITs may exhibit anticipation of several years to the development of MM, CM and UVM, the characteristic malignancies of the BAP1 cancer syndrome. Along the guidelines used for FAMM-P families  we advise families with hereditary BAP1 mutation to have family members tested for mutant gene carrier status at the age of ten and, if positive, to begin routine screening with a total body dermatological examination as family members may develop melanoma at an early age. When these individuals become adults, annual eye examinations (indirect ophthalmoscopy, etc.) should be performed for early detection of UVM.