Urine Sorbitol and Xylitol for the Diagnosis of Sorbitol Dehydrogenase Deficiency-Related Neuropathy

Abstract

BACKGROUND AND OBJECTIVES: Sorbitol dehydrogenase (SORD) deficiency, due to biallelic loss-of-function variants in the SORD gene, is a recently recognized cause of autosomal recessive hereditary neuropathy. Specific diagnosis is difficult on clinical grounds alone, and molecular genetic testing of SORD is complicated by the presence of a pseudogene. Biochemical testing of serum sorbitol is suggested as a potential biomarker. We report a novel urine biochemical profile of elevated excretion of sorbitol and a second polyol, xylitol, to aid in the identification of individuals with SORD-related neuropathy. METHODS: Patients with confirmed or suspected SORD-related neuropathy were recruited by clinicians from 7 academic medical centers and 1 nonprofit specialty care center. Urine was analyzed by a clinically validated, gas chromatography-mass spectrometry assay to measure sorbitol and xylitol excretion. Over 700 reference samples were evaluated from residual clinical samples. Clinical and molecular findings were gathered using a standardized questionnaire. RESULTS: Nineteen individuals with a clinical and genetic diagnosis of SORD-related neuropathy (median age 31 years, 47% female) and 715 reference samples were used to determine the initial performance of this index urine test. The median sorbitol excretion in affected individuals was 638 mmol/mol creatinine (1st percentile of disease range = 456; reference median = 7, reference 99th percentile = 198). The median xylitol excretion was 1,577 mmol/mol creatinine (1st percentile of disease range = 1,242; reference median = 7, reference 99th percentile = 102). In this patient cohort, the combination of sorbitol and xylitol yielded 100% sensitivity and specificity for SORD-related neuropathy. Four additional individuals who had clinical phenotypes compatible with SORD-related neuropathy had abnormal urine sorbitol/xylitol profile, 2 with genotypes involving a variant of unknown significance in SORD and 2 with a single heterozygous pathogenic SORD variant detected, whereas normal urine sorbitol/xylitol levels were observed in 11 asymptomatic, heterozygous carriers. DISCUSSION: These data demonstrate the clinical utility of urine sorbitol and xylitol analysis in the screening and diagnosis of SORD deficiency-related neuropathy. In addition, this is the first description of elevated xylitol in SORD deficiency as a clinically informative biomarker, which may increase the specificity of biochemical testing.