Role of TGFBR1 and TGFBR2 genetic variants in Marfan syndrome.

OBJECTIVE: Genetic variants in transforming growth factor beta (TGF-β) receptors type 1 (TGFBR1) and type 2 (TGFBR2) genes have been associated with different hereditary connective tissue disorders sharing thoracic aortic aneurysm and dissection (TAA/D). Mutations in both TGFBR1/2 genes have been described in patients with TAA/D and Marfan syndrome (MFS), and they are associated consistently with Loeys-Dietz syndrome. The existing literature shows discordant data resulting from mutational screening of TGFBR1/2 genes in patients with MFS. The aim of the study was to investigate the role of TGFBR1/2 genetic variants in determining and/or modulating MFS clinical phenotype.

METHODS: We investigated 75 unrelated patients with MFS referred to the Center for Marfan Syndrome and Related Disorders (Careggi University Hospital, Florence) who were subjected to FBN1 and TGFBR1/2 Sanger mutational screening.

RESULTS: Forty-seven patients with MFS (63%) carried a pathogenetic FBN1 mutation. No pathogenetic mutations were detected in TGFBR1/2 genes. Ten common polymorphisms were identified in TGFBR2 and 6 in TGFBR1. Their association with cardiovascular manifestations was evaluated. Carriers of the A allele of rs11466512, delA allele of c.383delA or delT allele of c.1256-15del1T polymorphisms had a trend toward or significantly reduced z-scores (median [interquartile range (IQR)], 2.2 [1.13-4.77]; 2.1 [1.72-3.48]; 2.5 [1.85-3.86]) with respect to homozygous patients with wild-type MFS (median [IQR], 4.20 [2.39-7.25]; 3.9 [2.19-7.00]; 3.9 [2.14-6.93]). Carriers of the A allele of the rs2276767 polymorphism showed a trend toward increased z-score (median [IQR], 4.9 [2.14-7.16]) with respect to patients with wild-type MFS (median [IQR], 3.3 [1.75-5.45]). The protective effect of TGFBR1/2 genetic score including all the 4 variants was also evaluated. Patients with MFS with two or more protective alleles included in the score had statistically significant reduced aortic z-scores (median [IQR], 2.20 [1.48-3.37]) with respect to patients with 1 or no protective alleles (median [IQR], 4.20 [2.48-7.12]; P = .007). Patients with severe aortic manifestations (aortic z-score ≥ 2 or aortic surgery) showed a significantly lower prevalence of subjects with two or more protective alleles included in the genetic score (29.7%) than patients with no or milder cardiovascular involvement (63.6%; P = .029). The genetic score protective effect on global aortic manifestations severity (aortic z-score ≥ 2 or aortic surgery) was also observed at the logistic regression analysis adjusted for the presence of FBN1 gene mutations (odds ratio, 0.21; 95% CI, 0.05-0.84; P = .028).

CONCLUSIONS: In conclusion, our data reappraise the role of TGFBR1 and TGFBR2 as major genes in patients with MFS, and suggest that TGFBR1/2 genetic variants (in particular when evaluated as a burden by score) might play a role in modulating the severity of cardiovascular manifestation in MFS.