Skin biopsies were taken from the EN on the abdomen and from adjacent normal skin after informed consent of the patient according to the guidelines of the local ethics committee and the Declaration of Helsinki. Separate fibroblast cultures were established from these biopsies. DNA was extracted directly from the skin biopsies as well as from cultured fibroblasts. In addition, formalin-fixed paraffin-embedded biopsy material, blood leukocytes, buccal brushings from lesional mucosa, scalp hair roots, and urothelial cells from urine sediment were available for analysis (Table ). DNA was extracted from these tissues and cells using standard protocols. FGFR3
mutations were analyzed using SNaPshot®
assays as described previously [6
]. We identified the FGFR3
hotspot mutation R248C in EN tissue, but not in the adjacent normal skin (Figure ). The R248C mutation was also detected in the EN tissue of the buccal mucosa harvested by buccal brushings. In contrast, the R248C mutation was not found in cultured fibroblasts from either affected or normal skin, nor in hair roots from affected skin of the scalp or in the urothelial cells. No mutations in the PIK3CA
gene were found in any of the tissue samples.
Genetic analysis of the FGFR3 gene by a SNaPshot® assay in various tissues revealed a mosaicism of the R248C hotspot mutation.
Immunohistochemical staining of the EN tissue with FGFR3 antibody (Santa Cruz Biotechnology, Santa Cruz, USA) revealed expression of FGFR3 protein in the complete epidermal layer, but the staining intensity was comparable with a normal skin control.
The results indicate that the systemic keratinocytic nevus syndrome in the present patient is caused by mosaicism of the R248C FGFR3
mutation. This mosaicism has extended to the oral mucosa, and caused intraoral lesions. The genetic mosaicism can be limited to the epidermis, but may also extend to other tissues, resulting in EN syndromes [1
]. The proportion of EN patients with additional abnormalities of other organ systems is not known, and may vary depending on the different subtypes of EN. One comprehensive study identified an EN syndrome in 33% of 119 EN patients [8
In keratinocytic nevi, activating point mutations in FGFR3
genes have been identified in about 40% of patients [5
]. Interestingly, the mutational spectrum is restricted to hotspots, including R248C for FGFR3
and E545G for PIK3CA
. The same mutations have been identified in sporadic malignant tumors, and confer oncogenic properties to cells in vitro
]. A non-mosaic constitutional R248C FGFR3
mutation causes thanatophoric dysplasia, a typically lethal skeletal dysplasia [11
]. Obviously, surviving patients with the R248C FGFR3
mutation must be mosaics [12
mutations have also been identified in seborrheic keratoses which are benign epidermal skin tumors arising in adult patients with an increasing age-related incidence [5
]. Although the identified FGFR3
mutations are oncogenic [10
] and can activate downstream signalling pathways [18
], neither EN nor seborrheic keratoses bear a significant risk of malignant transformation.
In the patient of this case report, the R248C FGFR3 hotspot mutation was identified in the EN tissue of the skin and buccal mucosa (ectoderm) as well as a substantial proportion of blood leukocytes (mesoderm). With the involvement of two embryologic tissues, one might conclude that the mutation had happened at a very early stage, in a cell whose descendant population included skin and hematogenous precursor cells. The early occurrence of the mutation might be the reason for the widespread skin involvement.
To the best of our knowledge, we could show for the first time that the mosaic FGFR3
mutation can be associated with intraoral EN lesions, an otherwise uncommon observation [20
]. The wildtype sequence of unaffected skin, hair roots and cultured fibroblasts in our patient confirms that the R248C mutation occurred in a mosaic form restricted to the epithelium and absent from the stromal component of skin. We propose that this case represents an EN syndrome, given the presence of the mutation in blood leukocytes indicating bone marrow involvement. Even her slight scoliosis may be part of the syndrome, although the finding could be coincidental and independent from the mosaicism. Hitherto the R248C FGFR3
mutation has been identified in two cases of an EN syndrome: - one with facial dysmorphism, and the other patient with cerebral involvement [21
]. A preliminary designation of "FGFR3 EN syndrome" has been proposed, while acknowledging that a genetic heterogeneity may underlie the clinical phenotype of this syndrome [1
]. For patients with extensive EN, a work-up for EN syndrome is indicated even without obvious extracutaneous features, as in our case.
Furthermore, it is important to be aware that these patients may have a predisposition towards malignancies of internal organs, especially low grade urothelial carcinomas [23
]. As yet, there has been no report identifying FGFR3
mutations in a patient with EN associated with urothelial cancer or other malignancies [26
]. However, FGFR3
mutations are involved in the pathogenesis of bladder tumors, cervix carcinoma, and multiple myeloma [9
]. The risk for the present patient is difficult to assess, although the absence of the FGFR3
mutation in urothelial cells suggests that at least for bladder cancer, it may be low. The risk for malignant transformation of an EN lesion (e.g., basal cell carcinoma or squamous cell carcinoma) appears likewise to be low, although occasional incidences have been reported [28
If the FGFR3
mosaicism involves the gonad, our patient is at risk of having offspring with thanatophoric dysplasia. There is precedent for this, in a female patient with extensive mosaicism of the R248C FGFR3
mutation, including widespread acanthosis nigricans of the skin and skeletal dysplasia [31