Urinary metabolomics using gas chromatography-mass spectrometry: potential biomarkers for autism spectrum disorder.

BACKGROUND: Diagnosis of autism spectrum disorder (ASD) is generally made phenotypically and the hunt for ASD-biomarkers continues. The purpose of this study was to compare urine organic acids profiles of ASD versus typically developing (TD) children to identify potential biomarkers for diagnosis and exploration of ASD etiology. METHODS: This case control study was performed in the Department of Pathology and Laboratory Medicine in collaboration with the Department of Pediatrics and Child Health, Aga Khan University, Pakistan. Midstream urine was collected in the first half of the day time before noon from the children with ASD diagnosed by a pediatric neurologist based on DSM-5 criteria and TD healthy controls from August 2019 to June 2021. The urine organic acids were analyzed by Gas Chromatography-Mass Spectrometry. To identify potential biomarkers for ASD canonical linear discriminant analysis was carried out for the organic acids, quantified in comparison to an internal standard. RESULTS: A total of 85 subjects were enrolled in the current study. The mean age of the ASD (n = 65) and TD groups (n = 20) was 4.5 ± 2.3 and 6.4 ± 2.2 years respectively with 72.3% males in the ASD group and 50% males in the TD group. Parental consanguinity was 47.7 and 30% in ASD and TD groups, respectively. The common clinical signs noted in children with ASD were developmental delay (70.8%), delayed language skills (66.2%), and inability to articulate sentences (56.9%). Discriminant analysis showed that 3-hydroxyisovalericc, homovanillic acid, adipic acid, suberic acid, and indole acetic were significantly different between ASD and TD groups. The biochemical classification results reveal that 88.2% of cases were classified correctly into ASD& TD groups based on the urine organic acid profiles. CONCLUSION: 3-hydroxy isovaleric acid, homovanillic acid, adipic acid, suberic acid, and indole acetic were good discriminators between the two groups. The discovered potential biomarkers could be valuable for future research in children with ASD.

Utilizing augmented artificial intelligence for aminoacidopathies using collaborative laboratory integrated reporting- A cross-sectional study.

Introduction: Plasma amino acids profiling can aid in the screening and diagnosis of aminoacidopathies. The goal of the current study was to analyze and report the metabolic profiles of plasma amino acid (PAA) and additionally to compare PAA-reference intervals (RI) from Pakistan with more countries utilizing Clinical Laboratory Integrated Reports (CLIR). Methods: This was a cross sectional prospective single center study. Twenty-two amino acids were analyzed in each sample received for one year at the clinical laboratory. Data was divided into reference and case data files after interpretation by a team of pathologists and technologists. All PAA samples were analyzed using ion-exchange high-performance chromatography. The CLIR application of Amino Acid in Plasma (AAQP) was used for statistical analysis for both data sets and post-analytical interpretive tools using a single condition tool was applied. Result: The majority of 92% (n = 1913) of PAA profiles out of the total 2081 tests run were non-diagnostic; the PAA values were within the age-specific RI. The PAA median was in close comparison close to the 50th percentile of reference data available in CLIR software. Out of the total 2081 tests run, one hundred and sixty-eight had abnormal PAA levels; 27.38% were labeled as non-fasting samples, and the main aminoacidopathies identified were Phenylketonuria and Maple Syrup Urine Disorder. Conclusion: An agreement of >95% was observed between the reporting done by the pathologists and technologists' team and then after the application of CLIR. Augmented artificial intelligence using CLIR can improve the accuracy of reporting rare aminoacidopathies in a developing country like ours.