Newborn screening follow-up for very long-chain acyl-CoA dehydrogenase deficiency in Colorado: Working towards a...

Very long chain acyl-CoA dehydrogenase deficiency (VLCADD) is an autosomal recessive fatty acid β-oxidation disorder that has been identified by newborn screening (NBS) in most states since the early 2000s. Despite over 20 years of experience, there are aspects of VLCADD NBS that remain challenging. We conducted a retrospective chart review of abnormal NBS for VLCADD in Colorado between 2017 and 2023. We analyzed confirmatory plasma acylcarnitine profiles (P-ACP), genetic sequencing of ACADVL, Collaborative Laboratory Integrated Reports (CLIR) scores, patient enzyme analysis of VLCAD, and cell-based variant expression analysis. A real-world "Clinical Designation" was then compared to a variety of algorithms trialed on the data. Of the 67 infants with abnormal screens during this timeframe, 5 (7 %) had a Clinical Designation of affected, 4 (6 %) remained unclassified, and 58 (87 %) were discharged based on a designation of unaffected. A Kruskal-Wallis rank sum test showed the biomarker with the best discrimination between affected and unaffected individuals was C14:1/C12:1 [chi-squared 10.4 (p = 0.001)]. The highest performing algorithm was (Molecular testing + cell-based expression) + (P-ACP C14:1 OR P-ACP C14:1/C12:1). Excluding the missing data, this algorithm showed 96 % (46 of 48) agreement with the Clinical Designation. We conclude that there is not a single biomarker that can specifically discern affected from unaffected individuals who screen positive on NBS for VLCADD. Thus, we developed a standardized diagnostic approach to more accurately classify patients that starts with the molecular findings and requires at least one of the P-ACP C14:1 or P-ACP C14:1/C12:1 to agree with molecular findings. The algorithm needs to be trialed with a different data set, and will advance the conversation around maximizing benefits and minimizing harms for infants who screen positive for VLCADD.