Further delineation of the phenotype resulting from BRAF or MEK1 germline mutations helps differentiate cardio-facio-cutaneous syndrome from Costello syndrome

Because Cardio-facio-cutaneous (CFC) syndrome has significant phenotypic overlap with Costello syndrome, it may be difficult to establish the diagnosis on a clinical basis. The recent discoveries of germline HRAS mutations in patients with Costello syndrome and mutations in BRAF, MEK1, and MEK2 in CFC syndrome uncovered the biologic mechanism for the shared phenotypic findings based on the close interaction of the affected gene products within the MAP kinase pathway. We evaluated a series of patients who were either clinically diagnosed with Costello syndrome, or in whom the diagnoses of both Costello and CFC syndromes were considered. After excluding mutations in HRAS, we identified eight changes in BRAF and five in MEK1. Five mutations are novel, and all changes occurred de novo among those triads tested. A review of the clinical abnormalities showed important differences between patients with either a BRAF or MEK1 mutation, and those previously reported with an HRAS mutation. Statistical significance was achieved, despite the relatively small number of patients with BRAF and MEK1 mutations reported here, for polyhydramnios, growth hormone deficiency and the presence of more than one papilloma, which were less common in CFC compared to HRAS mutation positive patients. Although both CFC and Costello syndrome are characterized by cardiac abnormalities in about three-fourths of patients, the pattern of congenital heart defects (CHD), hypertrophic cardiomyopathy (HCM), and tachycardia differs somewhat. CHD, especially pulmonic stenosis associated with a secundum-type atrial septal defect, are more common in CFC than Costello syndrome (P = 0.02). Atrial tachycardia is less frequent in CFC patients with BRAF or MEK1 mutations, compared to Costello syndrome patients with HRAS mutation (P = 0.04). Chaotic atrial rhythm or multifocal atrial tachycardia was observed only in Costello syndrome. Malignant tumors have been viewed as characteristic for Costello syndrome due to HRAS mutations, however, we report here on a MEK1 mutation in a patient with a malignant tumor, a hepatoblastoma. Although this indicates that the presence of a tumor is not specific for Costello syndrome with HRAS mutation, it is noteworthy that the tumor histology differs from those commonly seen in Costello syndrome. Based on these clinical differences we suggest that patients with BRAF and MEK mutations should be diagnosed with CFC syndrome, and the diagnosis of Costello syndrome be reserved for patients with HRAS mutations. (c) 2007 Wiley-Liss, Inc.