Abstract
Background and Aims
Bartter (BS) and Gitelman syndrome (GS) are autosomal recessive rare inherited disorders characterized by hypokalemic metabolic alkalosis and secondary hyperaldosteronism. The primary defect is a genetically determined impairment of sodium chloride reabsorption in the renal tubule, thus resulting in salt loss, dehydration and acid-base homeostasis perturbations.
Although the diagnosis can be suspected based on presenting features, the clinical diagnosis of BS and GS can be challenging, as they are rare and phenotypically overlapping. As a consequence, the current clinical classification lacks of specificity and genetic testing represents the gold standard for the diagnosis. Driven by the rapidly decreasing costs and turn-around time, next-generation sequencing technologies are increasingly utilized in diagnostics and research of inherited tubulopathies, including BS and GS. Recently, sequencing of selected gene panels provided the advantage of achieving high coverage of genes of interest at lower costs, providing high diagnostic yield and new insights into the phenotypic spectrum of these rare disorders. However, whole-exome (WES) is not routinely performed for the molecular diagnosis of BS and GS. The aim of our study was to assess the diagnostic performance of WES in BS and GS and to establish genotype-phenotype correlations.
Method
We performed WES in all consecutive patients referred for genetic testing with a clinical suspect of BS or GS. Variant prioritization was carried out according to the American College of Medical Genetics and Genomics guidelines (ACMG). Parents and first-degree relatives were included, whenever available. Demographic, clinical and laboratory data were collected retrospectively, in order to establish genotype-phenotype correlations.
Results
We enrolled 50 patients (22 males, 46 Caucasians) with a clinical diagnosis of BS (19), GS (24) or BS/GS (7). All the patients showed hypokalemic metabolic alkalosis at onset (serum bicarbonate=29.5 mEq/l ± 4.4, potassium= 2.7 mEq/l ± 0.6). The median age at clinical diagnosis was 7 years (range 0-67 years). Three patients had familial history of tubulopathies. WES showed pathogenic variants in 42/50 patients (84%), thus establishing a conclusive diagnosis. Interestingly, a dedicated analytic pipeline allowed us to identify copy number variations (CNVs) in 7/42 patients with a confirmed genetic diagnosis. In detail, WES allowed us to confirm the clinical diagnosis in 33/50 patients, with an improvement in classification in at least 14 cases (i.e. subtype I-V of BS). In 9 additional patients, genetic testing changed the clinical diagnosis: 6 patients with a clinical of BS turned out to have pathogenic variants in SLC12A3, resulting in GS; in 3 patients, genetic testing revised the clinical diagnosis indicating inherited disorders outside the BS/GS spectrum (HELIX syndrome, Primary familial hypoparatiroidism, Type 2 renal hypomagnesemia). Only 38% of patients with a genetic diagnosis of BS showed nephrocalcinosis. Strikingly, this was present in 8% of patients with GS. On the other hand, hypomagnesemia, a distinctive feature of GS, was similarly distributed among BS and GS patients (45% vs. 68%, respectively). Finally, although patients with GS showed a median age at onset higher than patients with BS, some overlap did exist, making differential diagnosis challenging at single-patient level.
Conclusion
The results of our study demonstrate that WES ensures a high diagnostic yield (84%) in patients with a clinical diagnosis of BS or GS, especially if coupled with analysis of CNVs. This approach showed to be useful in dealing with the phenotypic heterogeneity typical of these rare disorders, improving differential diagnosis by detecting phenocopies also outside the BS/GS spectrum, enabling additional specific work-up, genetic counseling, and screening of at-risk relatives.
Comprehensive screening shows that mutations in the known syndromic genes are rare in individuals presenting with hyperinsulinaemic hypoglycaemia