Hereditary tyrosinaemia type 1 in the absence of succinylacetone: 4-oxo 6-hydroxyhepanoate (4OHHA), a putative diagnostic biomarker.

Hereditary tyrosinemia type 1 (HT1) is a rare metabolic disease resulting in acute liver failure in early infancy, hypophosphataemic rickets, neurological crises, liver cirrhosis and risk of hepatocellular carcinoma later on in life. It is caused by the deficiency of the enzyme fumarylacetoacetate hydrolase which is involved in the terminal step of the catabolic pathway of tyrosine. Diagnosis is made through clinical suspicion supported by biochemical abnormalities that result from accumulation of upstream metabolites. Detection of succinylacetone (SA) in dried blood spot or urine remains pathognomonic, however it is not always detectable. Here we describe three cases of HT1 presenting with atypical biochemistry, where SA was not always detectable, highlighting the importance of an additional disease biomarker, 4-oxo-6-hydroxyheptanoate.

Investigation of the relationship between phenylalanine in venous plasma and capillary blood using volumetric blood collection devices.

Measurement of plasma and dried blood spot (DBS) phenylalanine (Phe) is key to monitoring patients with phenylketonuria (PKU). The relationship between plasma and capillary DBS Phe concentrations has been investigated previously, however, differences in methodology, calibration approach and assumptions about the volume of blood in a DBS sub-punch has complicated this. Volumetric blood collection devices (VBCDs) provide an opportunity to re-evaluate this relationship. Paired venous and capillary samples were collected from patients with PKU (n = 51). Capillary blood was collected onto both conventional newborn screening (NBS) cards and VBCDs. Specimens were analysed by liquid-chromatography tandem mass-spectrometry (LC-MS/MS) using a common calibrator. Use of VBCDs was evaluated qualitatively by patients. Mean bias between plasma and volumetrically collected capillary DBS Phe was -13%. Mean recovery (SD) of Phe from DBS was 89.4% (4.6). VBCDs confirmed that the volume of blood typically assumed to be present in a 3.2 mm sub-punch is over-estimated by 9.7%. Determination of the relationship between plasma and capillary DBS Phe, using a single analytical method, common calibration and VBCDs, demonstrated that once the under-recovery of Phe from DBS has been taken into account, there is no significant difference in the concentration of Phe in plasma and capillary blood. Conversely, comparison of plasma Phe with capillary DBS Phe collected on a NBS card highlighted the limitations of this approach. Introducing VBCDs for the routine monitoring of patients with PKU would provide a simple, acceptable specimen collection technique that ensures consistent sample quality and produces accurate and precise blood Phe results which are interchangeable with plasma Phe.

Evaluation of volumetric blood collection devices for the measurement of phenylalanine and tyrosine to monitor patients with phenylketonuria.

BACKGROUND: Measurement of dried blood spot (DBS) phenylalanine (Phe) is central to the monitoring of patients with phenylketonuria. However, the volume and hematocrit (Hct) of the blood applied to conventional DBS cards significantly affects analytical results. Volumetric blood collection devices are reported to be more accurate, precise and less prone to Hct effects. METHODS: Accuracy, imprecision, effect of blood volume and Hct were evaluated for measurement of Phe and tyrosine using three volumetric devices and compared with the conventional PerkinElmer-226 filter-paper collection devices. i.e. conventional DBS cards. Applicability for use in a clinical laboratory was assessed qualitatively. RESULTS: Blood volume did not impact on the performance of the volumetric devices; however, significant biases were observed with the conventional DBS card. A higher Hct introduced unacceptable bias for Neoteryx-Mitra and conventional DBS card. All devices had a mean relative standard deviation (RSD) ≤ 4.1 %, except for the Neoteryx-Mitra (≤ 6.2 %). Relative to liquid blood, the mean biases of Phe for the various devices were -5.1 (HemaXis-DB10), -7.8 (Capitainer-qDBS), -12.0 (Neoteryx-Mitra) and -32.6 % (conventional DBS card). CONCLUSIONS: Introducing volumetric collection devices will overcome the significant pre-analytical issues associated with conventional DBS collection and improve the biochemical monitoring of patients with PKU.

Automation of tacrolimus measurement on volumetric absorptive microsampling devices by tandem mass spectrometry.

Aims: An automated method for the measurement of blood tacrolimus on volumetric absorptive microsampling (VAMS) devices was developed. Materials & methods: VAMS devices prepared by the automated method were compared with those prepared by the existing manual method (n = 284; mean concentration: 8.0 µg/l; range: 0.6-18.1). Results: The performance of both methods was comparable. Passing-Bablok regression demonstrated an acceptable correlation (y = -0.449 + 1.06x). Bland-Altman analysis demonstrated acceptable agreement (mean bias: -0.007 µg/l; standard deviation: 1.536). Automation reduced operator touch time by 40 min (48-sample batch). Conclusion: Automated preparation of VAMS devices reduced touch time and improved process consistency, facilitating high-throughput testing and transformation of existing laboratory workflows. Automation did not improve precision for VAMS devices but did so for liquid blood samples.

After a kidney transplant, many patients take a drug called tacrolimus to help prevent their new kidney from being rejected. Blood levels of tacrolimus are checked regularly to ensure each patient is receiving the right dose. This means regular visits to the hospital for blood tests, which can be inconvenient and time-consuming for the patient. Microsampling devices are now available that would enable patients to collect blood from a finger prick sample, at home, and post it back to the lab for testing. However, to date, access to home sampling is limited because measuring tacrolimus from blood collected on a microsampling device relies on a manual laboratory process that is difficult to do and takes a long time. Measurement of tacrolimus from blood collected on a microsampling device can be successfully automated with a Gerstel MPS robot. The robot extracts the tacrolimus from the blood on the microsampling device and injects the resulting sample into a mass spectrometer for measurement. Two sets of microsamples were prepared. One set of samples was extracted by the robot and one set of VAMS samples was extracted manually. Tacrolimus was measured by mass spectrometry for both sets of samples and the results compared well. The automated method requires less operator input than the manual method, which will make it easier to measure large numbers of microsamples quickly and safely, increasing the number of patients who can benefit from the advantages of remote sampling.

A Machine Learning Approach for the Automated Interpretation of Plasma Amino Acid Profiles.

BACKGROUND: Plasma amino acid (PAA) profiles are used in routine clinical practice for the diagnosis and monitoring of inherited disorders of amino acid metabolism, organic acidemias, and urea cycle defects. Interpretation of PAA profiles is complex and requires substantial training and expertise to perform. Given previous demonstrations of the ability of machine learning (ML) algorithms to interpret complex clinical biochemistry data, we sought to determine if ML-derived classifiers could interpret PAA profiles with high predictive performance. METHODS: We collected PAA profiling data routinely performed within a clinical biochemistry laboratory (2084 profiles) and developed decision support classifiers with several ML algorithms. We tested the generalization performance of each classifier using a nested cross-validation (CV) procedure and examined the effect of various subsampling, feature selection, and ensemble learning strategies. RESULTS: The classifiers demonstrated excellent predictive performance, with the 3 ML algorithms tested producing comparable results. The best-performing ensemble binary classifier achieved a mean precision-recall (PR) AUC of 0.957 (95% CI 0.952, 0.962) and the best-performing ensemble multiclass classifier achieved a mean F4 score of 0.788 (0.773, 0.803). CONCLUSIONS: This work builds upon previous demonstrations of the utility of ML-derived decision support tools in clinical biochemistry laboratories. Our findings suggest that, pending additional validation studies, such tools could potentially be used in routine clinical practice to streamline and aid the interpretation of PAA profiles. This would be particularly useful in laboratories with limited resources and large workloads. We provide the necessary code for other laboratories to develop their own decision support tools.