Validation of plasma amino acid profile using UHPLC-mass spectrometer (QDa) as a screening method in a metabolic disorders reference centre: Performance and accreditation concerns.

INTRODUCTION: Amino acid (AA) analysis in plasma is essential for diagnosis and monitoring of inborn errors of metabolism (IEM). The efficacy of patient management is governed by the rapidity of AA profile availability, along with the robustness of the method. French quality guidelines and progress made in analytical techniques have led biologists to develop AA profile exploration via mass spectrometry (MS). OBJECTIVES: The aim of this study was to validate an analytical method with a single quadrupole mass spectrometer (MS) and to suggest reference values in regard to French quality and IEM society recommendations. DESIGN AND METHODS: Plasma samples from patients were deproteinised and derivatised with AccqTag™ reagent. Analysis was performed by reverse-phase chromatography coupled to QDA detector. We evaluated accuracy, intra-days and inter-days precision and limit of quantification by the ß-expectation tolerance interval method for 27 AA. Method comparison was performed with the standard method (ion exchange chromatography, IEC) on Jeol Aminotac® and to tandem MS. Reference values were established on AA concentrations of the cohort of patients who had no IEM. RESULTS: Our method allowed the separations of almost all amino acids with a total run time of 12 min. Separation of isoleucine and alloisoleucine was incomplete (R = 0.55) but without impact on biological interpretation. Precision, accuracy and quantification were satisfactory (intra-days coefficient of variation (CV) was <5%, inter-days precision CV <10% and accuracy <15%). The limits of quantification were validated between 1 and 900 µmol/L. Results were comparable between the new method and IEC. CONCLUSION: Ultimately, we validated a rapid method on plasma for quantifying 27 amino acids that can be used in routine practice, according to French quality laboratories and SFEIM (French Society of Inborn Error of Metabolism) recommendations. Furthermore, estimated reference values were similar to those found in published studies focusing on other methods. Despite a lower specificity compared to tandem MS, the simplicity and rapidity of our method are the main advantage of this technique and place it as a major tool in IEM diagnosis and management.

Biological follow-up in amyotrophic lateral sclerosis: decrease in creatinine levels and increase in ferritin levels predict poor prognosis.

BACKGROUND AND PURPOSE: Amyotrophic lateral sclerosis (ALS) is a fatal disorder of the motor neuron system, with a median survival of 2 to 4 years and a wide variety of prognosis. Thus, there is a critical need for diagnosis and prognosis biomarkers to improve the care of patients in routine practice. In this study, we aimed to determine prognostic value of routine biochemical markers in sporadic ALS (SALS). METHODS: We retrospectively collected clinical and biological data obtained during the systematic routine monitoring of 216 sporadic ALS patients. The main outcomes were disease duration and annual decline of Revised ALS Functional Rating Scale (ALSFRS-R). Changes to these biological variables over time were assessed, in link with disease progression. RESULTS: We found that concentrations of creatinine (P=0.0166) and ferritin (P=0.0306) changed significantly during the progression of ALS. A reduction of creatinine levels and an increase of ferritin levels were associated with disease progression. Multivariate analysis showed that early variation of ferritin was an independent predictive factor of patient survival (P=0.0048). CONCLUSION: Changes to ferritin and creatinine levels with time are associated with ALS progression. This is the first study describing the changes to these biological variables during ALS progression.

New sampling strategy using a Bayesian approach to assess iohexol clearance in kidney transplant recipients.

BACKGROUND: Glomerular filtration rate (GFR) measurement is a major issue in kidney transplant recipients for clinicians. GFR can be determined by estimating the plasma clearance of iohexol, a nonradiolabeled compound. For practical and convenient application for patients and caregivers, it is important that a minimal number of samples are drawn. The aim of this study was to develop and validate a Bayesian model with fewer samples for reliable prediction of GFR in kidney transplant recipients. METHODS: Iohexol plasma concentration-time curves from 95 patients were divided into an index (n = 63) and a validation set (n = 32). Samples (n = 4-6 per patient) were obtained during the elimination phase, that is, between 120 and 270 minutes. Individual reference values of iohexol clearance (CL(iohexol)) were calculated from k (elimination slope) and V (volume of distribution from intercept). Individual CL(iohexol) values were then introduced into the Bröchner-Mortensen equation to obtain the GFR (reference value). A population pharmacokinetic model was developed from the index set and validated using standard methods. For the validation set, we tested various combinations of 1, 2, or 3 sampling time to estimate CL(iohexol). According to the different combinations tested, a maximum a posteriori Bayesian estimation of CL(iohexol) was obtained from population parameters. Individual estimates of GFR were compared with individual reference values through analysis of bias and precision. A capability analysis allowed us to determine the best sampling strategy for Bayesian estimation. RESULTS: A 1-compartment model best described our data. Covariate analysis showed that uremia, serum creatinine, and age were significantly associated with k(e), and weight with V. The strategy, including samples drawn at 120 and 270 minutes, allowed accurate prediction of GFR (mean bias: -3.71%, mean imprecision: 7.77%). With this strategy, about 20% of individual predictions were outside the bounds of acceptance set at ± 10%, and about 6% if the bounds of acceptance were set at ± 15%. CONCLUSIONS: This Bayesian approach can help to reduce the number of samples required to calculate GFR using Bröchner-Mortensen formula with good accuracy.