ECE2020 Audio ePoster Presentations Adrenal and Cardiovascular Endocrinology (121 abstracts)
1Cochin Institute, Paris, France; 2Cochin Hospital, Paris, France; 3Lille Hospital, Lille, France; 4Grenoble Hospital, Grenoble, France; 5Saint-Antoine Hospital, Paris, France; 6Rouen Hospital, Rouen, France; 7Ambroise Paré Hospital, Boulogne-Billancourt, France; 8Birmingham Women’s and Children’s NSH Foundation Trust, Birmingham, United Kingdom; 9UniversitätsSpital Zürich, Zürich, Switzerland; 10Endocrinology in Charlottenburg, Berlin, Germany; 11National and Kapodistrian University of Athens, Athens, Greece; 12Conception Hospital, Marseille, France; 13Nice Hospital, Nice, France; 14Curry Cabral Hospital, Lisboa, Portugal; 15Bordeaux Hospital, Bordeaux, France; 16Lyon Hospital, Lyon, France; 17Universitätsklinikum Würzburg, Würzburg, Germany; 18University of Birmingham, Birmingham, United Kingdom; 19Bicêtre Hospital, Le Kremlin-Bicêtre, France; 20University of Munich, Munich, Germany; 21Cochin Hospital, Division of Genetics, Paris, France
Introduction: PBMAH is a rare but heterogeneous disease, characterized by multiple benign adrenal macronodules with variable levels of cortisol excess. In 2013, our team discovered germline heterozygous inactivating mutations of ARMC5, acting as a tumor suppressor gene. ARMC5 mutation rate is 50% in patients with PBMAH treated by adrenalectomy for severe hypercortisolism, 80% in familial cases and 20% in sporadic cases according to the current literature. The aim of this study was to identify predictive criteria for ARMC5 mutation.
Methods: Since 2014, 454 French and European index PBMAH cases have been genotyped for ARMC5 in Cochin Institute or in the Genetic Department of Cochin Hospital. To date, among these 454 patients, we have reviewed the 264 consecutive index cases referred from five endocrinology French departments belonging to the COMETE network. A retrospective analysis of the clinical, radiological and biological data was done with a central review of the adrenal CT-scans.
Results: Sixty-five out of the 454 index cases (14.3%) present a germline inactivating ARMC5 mutation. Similarly, 40 (15.2 %) out of the subgroup of the 264 reviewed patients had a germline ARMC5 mutation. Patients with ARMC5 mutations have a more severe disease than wild-type patients, in terms of cortisol excess (UFC 1.92 [0.2 −12.1] vs 0.84 [0.08 − 10.1] fold ULN, respectively, P = 0.017), adrenal morphology (10.8 [2−27] vs 3.4 [1−13] nodules, respectively, P < 0.001) and complications (87.5 vs 68.3% of patients treated for hypertension, respectively, P = 0.013). 100% of the mutated patients have bilateral adrenal nodules AND plasma cortisol after dexamethasone 1 mg suppression test above 50 nmol/l, only 61.6% of the non-mutated patients have these two associated criteria. Therefore, the 100% negative predictive value of the association of bilateral nodules and autonomous cortisol secretion allows to exclude ARMC5 mutation when both are absent. Using more stringent criteria improved the specificity for ARMC5 mutations : among the mutated patients, 75.9% had at least 4 and 65.5% had at least 6 adrenal nodules, vs 18.0 and 6.8% of wild-type patients respectively, improving the mutation rate to 37.3% and 57.6%, respectively. However by using these criteria some mutated patients would be missed (24.1 and 34.5%, respectively).
Conclusion: Restricting the indication of ARMC5 sequencing to patients with bilateral adrenal nodules and a cortisol above 50 nmol/l after 1 mg dexamethasone suppression test could avoid useless genotyping in a third of patients, and would allow a cost and time economy and earlier results, without missing any mutated patient.