American Association for Clinical Chemistry
Better health through laboratory medicine
Education Rounds

Sihe Wang, PhD

1. Measurement of vitamin D metabolites by LC-MS/MS.

Vitamin D nutritional status is important to health and prevention of disease including bone health, muscle strength, cardiovascular disease, autoimmune disease, and cancer.

25-hydroxyvitamin D is the biomarker for evaluating vitamin D nutritional status. Monitoring the active metabolite (1, 25-dihydoxyvitamin D) is useful in evaluating patients with chronic kidney disease and vitamin D dependent rickets. Among all the methods, liquid chromatography-tandem mass spectrometry (LC-MS/MS) represents the most accurate and sensitive technology for measuring vitamin D metabolites.

LC-MSMS has unique advantages over immunoassays in clinical laboratory services due to its high specificity and sensitivity. It can also simultaneously measure multiple analytes in a single run. However, LC-MS/MS also presents pitfalls which include matrix effect, interference, and in-source transformation. Therefore, proper development and rigorous validation are necessary before introducing such an assay into patient care.

The presentation discussed the development and validation of LC-MS/MS methods for measuring 25-hydroxyvitamin D and 1, 25-dihydroxyvitamin D following a strict validation protocol. Also discussed were the challenges of measuring vitamin D metabolites by LC-MS/MS and means to improve the performance.

2. Measurement of 19 drugs and metabolites important for pain management by liquid chromatography-tandem mass spectrometry.

Monitoring pain management and illicit drugs/metabolites in urine is an important tool to pain management physicians in order to ensure compliance and detect drug abuse. Mass spectrometry-based methods have the advantage of high specificity and sensitivity. Preparation of the sample involved the use of glucoronidase and the supernatant was loaded into a Cyclone Max TurboFlow column and a Cyclone-P TurboFlow column followed by a Hypersil Gold PFP analytical column and monitored by a Quantum Ultra mass spectrometer in selective reaction monitoring mode. Identification was achieved by calculating the ratio of another transition to the base peak and quantification was based on peak area ratios of analytes to internal standards.

This method had a wide linear range for each analyte with an analytical recovery of 83.6-119.8%. For spiked urine samples, the within-run coefficients of variation (CV) were ≤9.8% and the total CV were ≤12.5% for all analytes at 3 levels. No significant carryover was observed. Commercial controls containing >100 therapeutic drugs and common endogenous substances were tested and showed no interference with this method. Comparison using 152 de-identified patient samples and spiked urine samples whose values spanned over linear ranges were concordant with commercially mass spectrometry-based methods.

David Carpentiari, PhD

1. Therapeutic Drug Target Interactions by Mass Spectrometry Proteomic Analysis of Formalin Fixed Paraffin Embedded Pediatric Brain Tumors.

Proteomic platforms are emerging as an invaluable tool for personalized medicine particularly in the field of oncology. The recent availability of protocols to extract high quality proteins from tissue samples fixed in formalin and stored in paraffin blocks for mass spectrometry analysis opens an array of diagnostic and therapeutic possibilities. This pilot study was designed to test the potential utility of this approach to four pediatric brain tumors: Atypical teratoid rhabodid tumor (ATRT), anaplastic astrocytoma (AA), ependymoma (EP) and medulloblastoma (MED). Twelve formalin fixed paraffin embedded (FFPE) tissue samples representing four pediatric brain tumors (ATRT, AA, EP and MED) were retrieved from the pathology files. The diagnoses were confirmed by a pediatric pathologist and a neuropathologist. A representative block was selected from each case and areas of interest were laser captured for protein extraction per protocol (Expression Pathology). Proteins were analyzed with a ThermoFischer LTQ-Orbitrap XL and peptides were matched to protein libraries (X!Tandem). Proteins with at least 95% confidence were selected. Unique proteins for each tumor category and associated therapeutic drug-target interactions (TDTI) were determined through software analysis (GeneGO). A total of 1,070 proteins with 95% confidence were identified. Among these, 397 were unique proteins: ATRT (161), AA (64), EP (61) and MED (111). Nineteen TDTI were associated with the unique proteins: ATRT (5), AA (8), MED (6). No TDTI were associated with EP unique proteins. While a larger set of samples must be analyzed before diagnostic tests and therapeutic pathways are clearly defined, this technique for mass spectrometry now provides an opportunity for a greater array of tissue analysis than previous. FFPE spectrometry proteomic analysis offers a significant potential for improved personalized medicine in pediatric oncology and should be strongly considered as an adjunct in future clinical trials involving pediatric brain tumors.

2. Procalcitonin Reference Range and Blood Culture Correlation in Hospitalized Children.

Procalcitonin (PCT) is a blood biomarker associated with bacterial sepsis. Unfortunately, there is significant interpretative uncertainty about the optimal reference range. Values less than 0.05 ng/mL are considered normal and values above 2.0 ng/mL are highly suspicious for clinical sepsis. PCT results between these two values are difficult to interpret. We attempted to optimize the PCT reference range by measuring PCT levels in stored specimens that were obtained from children hospitalized for suspected sepsis.

We calculated the PCT predictive values for bacteremia at the 2.0 ng/mL cut off value. Children hospitalized at our institution in 2009 were included based on a positive or negative bacterial blood culture result. PCT was measured on stored specimens drawn within a six hour period from the culture collection time. Contaminants were identified by medical record review. The percentage of true positive blood cultures associated with a PCT value between 0.05 ng/mL and 2.0 ng/mL, and the predictive values at the 2.0 ng/mL cut off point were calculated.

The study included 98 patients aged 3 weeks to 18 years. In the final analysis, 38% (37/98) blood cultures were true (non-contaminant) positives and 41% (15/37) of these cultures were associated with PCT <2.0 ng/mL. The positive predictive value (PPV) and the negative predictive value (NPV) for bacteremia at the 2.0 ng/mL PCT cut off value were 60% and 75%, respectively. No true positive culture was seen with a PCT value <0.09 ng/mL.

There is a high percentage (41%) of true positive cultures associated with PCT values between 0.05 ng/mL and 2.0 ng/mL. The PPV (60%) and the NPV (75%) of PCT at the cut-off value of 2.0 ng/mL appear to be inadequate for clinical utility. The lowest PCT value with a true positive culture (NPV = 100%) was 0.09 ng/ml.