<i>PDE11A</i> Is a Phenotype Modulator of Primary Bilateral Macronodular Adrenal Hyperplasia: Results of a 334-Patient Series
Patricia Vaduva, Lucas Bouys, Anne Jouinot, Stéphanie Espiard, Albain Chansavang, Annabel Berthon, Mario Néou, Anna Vaczlavik, Florian Violon, Helaine Laiz Silva Charchar, Matthias Kroiß, Gérald Raverot, Hélène Lasolle, Laurence Guignat, Rossella Libé, Guillaume Assié, Antoine Tabarin, Peter Kamenický, Éric Pasmant, Maria Candida Barisson Villares Fragoso, Constantine A. Stratakis, Bruno Ragazzon, Jérôme Bertherat
Abstract
CONTEXT: Primary bilateral macronodular adrenal hyperplasia (PBMAH), the most common cause of Cushing syndrome due to bilateral nodules, is a heterogeneous disease at the clinical, hormonal, and morphological levels. ARMC5-inactivating pathogenic variants are causative of PBMAH, and rare variants of PDE11A have been associated with PBMAH. OBJECTIVE: The aim of this study, on a large cohort of individuals with PBMAH from Europe and America, was to study the ARMC5 and PDE11A genotype to determine the genotype/phenotype correlation and to investigate the hypothesis that PDE11A could be a modifying gene of the adrenal phenotype. METHODS: Leukocyte DNA of 354 PBMAH index cases was sequenced for ARMC5 and PDE11A genes by next-generation sequencing. Phenotypic characteristics of 334 of these patients were analyzed to study the genotype/phenotype correlations. RESULTS: Seven out of 16 PDE11A variants were considered damaging according to in silico predictions: 6 missense variants (p.Tyr727Cys, p.Met623Arg, p.Tyr658Cys, p.Ag867Trp, p.Asn298Ser, p.Glu840Lys) and 1 stop-gain variant (p.Arg307Ter). In the cohort, 11.4% of patients had one of these variants and 19.2% had ARMC5-pathogenic variants. There was no statistically significant difference in the distribution of PDE11A-damaging variants according to ARMC5 status (P = .83; OR = 0.79; 95% CI, 0.26-2.03) nor in the distribution of ARMC5 pathogenic variants according to PDE11A status (P = .83; OR = 0.81; 95% CI, 0.27-2.04). Patients with PDE11A-damaging variants had lower urinary free cortisol (0.7 vs 1.25 upper limit of normal; P = .0002), midnight plasma cortisol (157.81 vs 222.19 nmol/L, P = .016), and number of adrenal nodules (3.46 vs 4.74; P = .048) compared to PDE11A wild-type patients. Patients with ARMC5-pathogenic variants had a more severe phenotype with more frequent comorbidities and were more often treated by adrenalectomy (60%). CONCLUSION: PDE11A appears to be a modulator of PBMAH phenotype, damaging variants being associated with an attenuated form. This may contribute to the heterogeneity of PBMAH and could affect patient management.