Reference intervals: past, present, and future.

Clinical laboratory test results alone are of little value in diagnosing, treating, and monitoring health conditions; as such, a clinically actionable cutoff or reference interval is required to provide context for result interpretation. Healthcare practitioners base their diagnoses, follow-up treatments, and subsequent testing on these reference points. However, they may not be aware of inherent limitations related to the definition and derivation of reference intervals. Laboratorians are responsible for providing the reference intervals they report with results. Yet, the establishment and verification of reference intervals using conventional direct methods are complicated by resource constraints or unique patient demographics. To facilitate standardized reference interval best practices, multiple global scientific societies are actively drafting guidelines and seeking funding to promote these initiatives. Numerous national and international multicenter collaborations demonstrate the ability to leverage combined resources to conduct large reference interval studies by direct methods. However, not all demographics are equally accessible. Novel indirect methods are attractive solutions that utilize computational methods to define reference distributions and reference intervals from mixed data sets of pathologic and non-pathologic patient test results. In an effort to make reference intervals more accurate and personalized, individual-based reference intervals are shown to be more useful than population-based reference intervals in detecting clinically significant analyte changes in a patient that might otherwise go unrecognized when using wider, population-based reference intervals. Additionally, continuous reference intervals can provide more accurate ranges as compared to age-based partitions for individuals that are near the ends of the age partition. The advantages and disadvantages of different reference interval approaches as well as the advancement of non-conventional reference interval studies are discussed in this review.

Preterm to term infant postmenstrual age reference intervals for thyroid-stimulating hormone and free thyroxine.

BACKGROUND: Infants born preterm are affected by a hypothalamic-pituitary-thyroid axis that is immature and still developing as they progress closer to corrected term gestation. Multiple risk factors place preterm infants at risk for a hypothyroid state. However, there is variability in thyroid-stimulating hormone cutoff values and limited data on free thyroxine reference intervals to guide clinicians. METHODS: 1584 thyroid-stimulating hormone and 1576 free thyroxine laboratory samples that were originally collected to screen hospitalized infants for delayed-onset of hypothyroidism were retrospectively evaluated from a group of 1087 infants who ranged in postmenstrual age from 25 to 43 weeks gestation at the time of laboratory sample collection. Median thyroid hormone values and reference intervals were established using R and the mixtools package. RESULTS: Thyroid-stimulating hormone reference intervals remained similar across gestational ages from 0.340-9.681 µIU/mL in 25-27 6/7-week infants to 1.090-7.627 µIU/mL in 40-43-weeks infants. For the same age groups, free thyroxine reference intervals increased from 0.42-0.91 ng/dL to 0.87-1.32 ng/dL. CONCLUSION: The reference intervals identified suggest that infants <31 weeks gestation have a higher thyroid-stimulating hormone and lower free thyroxine level at baseline than previously anticipated. IMPACT: The increasing free thyroxine values in preterm to term infants indicate a maturing hypothalamic-pituitary-thyroid axis. Clinicians need thyroid hormone reference intervals that also vary by postmenstrual age to aid the evaluation of sick preterm infants who are at risk of a delayed hypothyroidism diagnosis that can be missed on the initial newborn screen. This study provides one of the largest samples of thyroid-stimulating hormone and free thyroxine data to establish reference intervals in preterm infants. Clinicians may utilize the identified postmenstrual age-based reference intervals to inform follow-up thyroid testing in preterm infants at several weeks postnatal age.

Conversion of abiraterone to D4A drives anti-tumour activity in prostate cancer.

Prostate cancer resistance to castration occurs because tumours acquire the metabolic capability of converting precursor steroids to 5α-dihydrotestosterone (DHT), promoting signalling by the androgen receptor and the development of castration-resistant prostate cancer. Essential for resistance, DHT synthesis from adrenal precursor steroids or possibly from de novo synthesis from cholesterol commonly requires enzymatic reactions by 3ß-hydroxysteroid dehydrogenase (3ßHSD), steroid-5α-reductase (SRD5A) and 17ß-hydroxysteroid dehydrogenase (17ßHSD) isoenzymes. Abiraterone, a steroidal 17α-hydroxylase/17,20-lyase (CYP17A1) inhibitor, blocks this synthetic process and prolongs survival. We hypothesized that abiraterone is converted by an enzyme to the more active Δ(4)-abiraterone (D4A), which blocks multiple steroidogenic enzymes and antagonizes the androgen receptor, providing an additional explanation for abiraterone's clinical activity. Here we show that abiraterone is converted to D4A in mice and patients with prostate cancer. D4A inhibits CYP17A1, 3ßHSD and SRD5A, which are required for DHT synthesis. Furthermore, competitive androgen receptor antagonism by D4A is comparable to the potent antagonist enzalutamide. D4A also has more potent anti-tumour activity against xenograft tumours than abiraterone. Our findings suggest an additional explanation-conversion to a more active agent-for abiraterone's survival extension. We propose that direct treatment with D4A would be more clinically effective than abiraterone treatment.

Health Care and Precision Medicine Research: Analysis of a Scalable Data Science Platform.

BACKGROUND: Health care data are increasing in volume and complexity. Storing and analyzing these data to implement precision medicine initiatives and data-driven research has exceeded the capabilities of traditional computer systems. Modern big data platforms must be adapted to the specific demands of health care and designed for scalability and growth. OBJECTIVE: The objectives of our study were to (1) demonstrate the implementation of a data science platform built on open source technology within a large, academic health care system and (2) describe 2 computational health care applications built on such a platform. METHODS: We deployed a data science platform based on several open source technologies to support real-time, big data workloads. We developed data-acquisition workflows for Apache Storm and NiFi in Java and Python to capture patient monitoring and laboratory data for downstream analytics. RESULTS: Emerging data management approaches, along with open source technologies such as Hadoop, can be used to create integrated data lakes to store large, real-time datasets. This infrastructure also provides a robust analytics platform where health care and biomedical research data can be analyzed in near real time for precision medicine and computational health care use cases. CONCLUSIONS: The implementation and use of integrated data science platforms offer organizations the opportunity to combine traditional datasets, including data from the electronic health record, with emerging big data sources, such as continuous patient monitoring and real-time laboratory results. These platforms can enable cost-effective and scalable analytics for the information that will be key to the delivery of precision medicine initiatives. Organizations that can take advantage of the technical advances found in data science platforms will have the opportunity to provide comprehensive access to health care data for computational health care and precision medicine research.

Fast, simple, and sensitive high-performance liquid chromatography method for measuring vitamins A and E in human blood plasma.

Vitamins A and E are fat-soluble vitamins that play important roles in several physiological processes. Monitoring their concentrations is needed to detect deficiency and guide therapy. In this study, we developed a high-performance liquid chromatography method to measure the major forms of vitamin A (retinol) and vitamin E (α-tocopherol and γ-tocopherol) in human blood plasma. Vitamins A and E were extracted with hexane and separated on a reversed-phase column using methanol as the mobile phase. Retinol was detected by ultraviolet absorption, whereas tocopherols were detected by fluorescence emission. The chromatographic cycle time was 4.0 min per sample. The analytical measurement range was 0.03-5.14, 0.32-36.02, and 0.10-9.99 mg/L for retinol, α-tocopherol, and γ-tocopherol, respectively. Intr-aassay and total coefficient of variation were <6.0% for all compounds. This method was traceable to standard reference materials offered by the National Institute of Standards and Technology. Reference intervals were established using plasma samples collected from 51 healthy adult donors and were found to be 0.30-1.20, 6.0-23.0, and 0.3-3.2 mg/L for retinol, α-tocopherol, and γ-tocopherol, respectively. In conclusion, we developed and validated a fast, simple, and sensitive high-performance liquid chromatography method for measuring the major forms of vitamins A and E in human plasma.

Applications of monolithic solid-phase extraction in chromatography-based clinical chemistry assays.

Complex matrices, for example urine, serum, plasma, and whole blood, which are common in clinical chemistry testing, contain many non-analyte compounds that can interfere with either detection or in-source ionization in chromatography-based assays. To overcome this problem, analytes are extracted by protein precipitation, solid-phase extraction (SPE), and liquid-liquid extraction. With correct chemistry and well controlled material SPE may furnish clean specimens with consistent performance. Traditionally, SPE has been performed with particle-based adsorbents, but monolithic SPE is attracting increasing interest of clinical laboratories. Monoliths, solid pieces of stationary phase, have bimodal structures consisting of macropores, which enable passage of solvent, and mesopores, in which analytes are separated. This structure results in low back-pressure with separation capabilities similar to those of particle-based adsorbents. Monoliths also enable increased sample throughput, reduced solvent use, varied support formats, and/or automation. However, many of these monoliths are not commercially available. In this review, application of monoliths to purification of samples from humans before chromatography-based assays will be critically reviewed.

Lead poisoning: Clinical and laboratory considerations.

Lead has been a known source of toxicity for millennia due to widespread use until the 20th century. Consequently, there remains significant, though decreasing, exposure to lead throughout the world. Clinical signs and symptoms of lead toxicity are well-documented but is particularly concerning for children six years of age and under, as brain development is rapid and therefore, is likely to be affected by even low levels of lead. Therefore, in the United States, it is recommended that young children to be routinely screened for blood lead levels. Blood lead levels can be measured by various methods in laboratories with blood collection greatly impacting possible lead contamination of samples. The history, presentation, and laboratory testing methodologies will be discussed.

Artificial Intelligence Applications in Clinical Chemistry.

Artificial intelligence (AI) applications are an area of active investigation in clinical chemistry. Numerous publications have demonstrated the promise of AI across all phases of testing including preanalytic, analytic, and postanalytic phases; this includes novel methods for detecting common specimen collection errors, predicting laboratory results and diagnoses, and enhancing autoverification workflows. Although AI applications pose several ethical and operational challenges, these technologies are expected to transform the practice of the clinical chemistry laboratory in the near future.

Matrix-matched calibrators are necessary for robust and high-quality dried blood spots lead screening assays by inductively coupled plasma-mass spectrometry.

Background and aims: Reliable lead screening methods are necessary to support early identification of lead exposure in children. Sample collection using dried blood spots (DBS) offers advantages compared to traditional venipuncture and capillary collection. Here, we describe and compare three lead DBS inductively coupled plasma-mass spectrometry (ICP-MS) methods for lead screening. Materials and methods: Lead was extracted from Whatman 903 protein saver cards punches and analyzed by ICP-MS across three independent clinical laboratories. Each laboratory evaluated the performance of aqueous and matrix-matched DBS calibrators using external quality control samples (WI State of Laboratory of Hygiene Program). Leftover patient samples (n = 39) were used for an interlaboratory comparison of lead DBS. Lead DBS results were compared to whole blood methods. Results: The DBS ICP-MS methods using matrix-matched DBS calibrators had superior performance to the aqueous calibrations. There was a strong correlation between lead measured in DBS (matrix-matched) and whole blood for the three methods evaluated. Conclusion: Lead can be measured accurately by ICP-MS in DBS samples when matrix-matched calibrators are used. External quality control programs are valuable to assess the performance of DBS methods. DBS lead ICP-MS methods are a robust analytical option for lead screening even though the limitations of DBS are well recognized.

A simple and fast liquid chromatography-tandem mass spectrometry method for measurement of underivatized L-arginine, symmetric dimethylarginine, and asymmetric dimethylarginine and establishment of the reference ranges.

Asymmetric dimethylarginine (ADMA) is an endogenous inhibitor of nitric oxide synthase and an established biomarker for endothelial function, while symmetric dimethylarginine (SDMA), an emerging biomarker for renal function, has been shown to outperform creatinine-based equations for estimated glomerular filtration rate. In order to study these analytes for clinical research, a fast and simple method for measuring arginine (ARG), SDMA, and ADMA in plasma by liquid chromatography-tandem mass spectrometry (LC-MS/MS) has been developed. Plasma (50 µL) was mixed with 50 µL of internal standard of (13)C-arginine and d(7)-ADMA followed by protein precipitation with methanol containing 1% ammonium acetate (300 µL). After centrifugation, the supernatant (100 µL) was mixed with 300 µL of acetonitrile with 1% formic acid, and the mixture was injected onto a silica column monitored by a mass spectrometer. The analytical cycle time was 5.0 min. The method was linear from 5.7 to 489.7 µM for ARG, 0.06 to 5.15 µM for SDMA, and from 0.34 to 5.65 µM for ADMA, with an accuracy of 99.0-120.0%. Total coefficients of variation for all analytes ranged from 2.7% to 7.7% for three concentration levels. The effects of hemolysis, lipemia, uremia, icterus, specimen tube types, storage at different temperature, and freeze/thaw were thoroughly investigated. Reference ranges were established using 51 well-defined reference subjects (12 men and 39 women, age 19-64 years): 53.1-129.7 µM for ARG, 0.32-0.65 µM for SDMA, and 0.36-0.67 µM for ADMA. In conclusion, the validated LC-MS/MS method described here offers a fast and reliable ARG, SDMA, and ADMA quantitation in plasma with minimum sample preparation.

Indirect reference intervals using an R pipeline.

BACKGROUND: Indirect reference intervals require robust statistical approaches to separate the pathological and healthy values. This can be achieved with a data pipeline created in R, a freely available statistical programming language. METHODS: A data pipeline was created to ingest, partition, normalize, remove outliers, and identify reference intervals for testosterone (Testo; n  = 7,207) and aspartate aminotransferase (AST; n  = 5,882) using data sets from NHANES. RESULTS: The estimates for AST and Testo determined by this pipeline approximated current RIs. Care should be taken when using this pipeline as there are limitations that depend on the pathology of the analyte and the data set being used for RI estimation. CONCLUSIONS: R can be used to create a robust statistical reference interval pipeline.

Quantitation of non-derivatized free amino acids for detecting inborn errors of metabolism by incorporating mixed-mode chromatography with tandem mass spectrometry.

Introduction: Amino acids are critical biomarkers for many inborn errors of metabolism, but amino acid analysis is challenging due to the range of chemical properties inherent in these small molecules. Techniques are available for amino acid analysis, but they can suffer from long run times, laborious derivatization, and/or poor resolution of isobaric compounds. Objective: To develop and validate a method for the quantitation of a non-derivatized free amino acid profile in both plasma and urine samples using mixed-mode chromatography and tandem mass spectrometry. Methods: Chromatographic conditions were optimized to separate leucine, isoleucine, and allo-isoleucine and maintain analytical runtime at less than 15 min. Sample preparation included a quick protein precipitation followed by LC-MS/MS analysis. Matrix effects, interferences, linearity, carryover, acceptable dilution limits, precision, accuracy, and stability were evaluated in both plasma and urine specimen types. Results: A total of 38 amino acids and related compounds were successfully quantitated with this method. In addition, argininosuccinic acid was qualitatively analyzed. A full clinical validation was performed that included method comparison to a reference laboratory for plasma and urine with Deming regression slopes ranging from 0.38 to 1.26. Conclusion: This method represents an alternative to derivatization-based methods, especially in urine samples where interference from metabolites and medications is prevalent.

Applications of monolithic columns in liquid chromatography-based clinical chemistry assays.

Monolithic columns have slowly been applied to HPLC methods for clinical chemistry testing in the last 10 years. The application areas include therapeutic drug monitoring, drugs of abuse, vitamins, porphyrins, and steroids. In comparison with conventional particulate columns, the monolithic columns may offer shorter chromatography time, more robustness, and better resolution for certain analytes. The potential drawback of large mobile phase consumption may be improved with smaller id columns, which are currently on the market. Methods covered in this review are those searchable in PubMed up to December 2010. This review highlights the emergence of monolithic column technology in HPLC methods used for clinical chemistry testing. The goals of this review are threefold: (i) To identify the areas of clinical chemistry that analytical monolithic columns have been used in HPLC methods. (ii) To demonstrate the application of analytical monolithic columns in HPLC methods using different detection systems. (iii) To discuss the advantages and limitations of the monolithic columns compared with particulate columns in the clinical chemistry applications.

First- and Second-Trimester Reference Intervals for Thyroid Function Testing in a US Population.

OBJECTIVES: Thyroid dysfunction in pregnancy is associated with increased risk of adverse outcomes to mother and child. Trimester-specific reference intervals for thyroid function tests are not routinely provided by clinical laboratories. In this study, we present first- and second-trimester-specific reference intervals in a US population for thyroid-stimulating hormone (TSH), free thyroxine (FT4), total thyroxine (T4), and total triiodothyronine (T3) measured on Roche analyzers. METHODS: We used patient samples from first- and second-trimester prenatal screening. Samples were limited to singleton pregnancies and negative screening results for thyroid peroxidase and thyroglobulin antibodies. Analytes (TSH, FT4, T4, and T3) were measured on a Roche Modular e170 then verified on a Roche cobas e801. RESULTS: The reference intervals established on the e170 and verified on the e801 for the first trimester were 0.16 to 2.82 mIU/L for TSH, 12.0 to 18.5 pmol/L for FT4, 62.8 to 177.9 nmol/L for T4, and 1.5 to 3.4 nmol/L for T3. The reference intervals for the second trimester were 0.40 to 3.62 mIU/L for TSH, 10.2 to 16.6 pmol/L for FT4, 66.6 to 176.0 nmol/L for T4, and 1.56 to 3.6 nmol/L for T3. CONCLUSIONS: This is the first report of trimester-specific reference intervals for thyroid function tests on Roche analyzers in the United States, and it is consistent with worldwide reports.

A novel HPLC method for quantification of 10 antiepileptic drugs or metabolites in serum/plasma using a monolithic column.

Therapeutic drug monitoring of antiepileptic drugs (AEDs) is important in maximizing the therapeutic response while minimizing the adverse effects. High-performance liquid chromatography (HPLC) is the most commonly used technique for this purpose. Recently, commercial monolithic columns were introduced, which consist of a single rod of fused silica or polymer. The objective of this work was to develop a simple and fast method to quantify 10 commonly measured AEDs or metabolites [carbamazepine, carbamazepine-10,11-epoxide, felbamate, lamotrigine, 10,11-dihydro-10-hydroxy-carbamazepine (active metabolite of oxcarbazepine), pentobarbital, phenobarbital, phenytoin, primidone, and zonisamide] in serum/plasma by HPLC using a reverse-phase monolithic column. Serum/heparin plasma (100 microL) was mixed with an internal standard solution (5-ethyl-5-p-tolylbarbituric acid in methanol, 250 microL). After centrifugation at 15,500 g for 10 minutes, 15 microL of supernatant was injected into a monolithic column. The analytes were eluted with an isocratic solution of 0.1 M, pH 6.5, phosphate buffer:methanol:acetonitrile (77:20:3), monitored at 210 nm. The chromatography time was 16 minutes. The method was linear from 0.4-4.9 to 21.2-190.9 microg/mL depending on the analytes with analytical recovery of 80%-114%. The inter- and intra-assay coefficients of variation were <8% in 3 levels of serum-based controls for all the analytes. No significant carryover was observed. Commercial controls containing >100 therapeutic drugs and common endogenous substances were tested and showed no interference. Comparison studies for 6 AEDs or metabolites were performed against commercial HPLC methods. Three AEDs were compared with Food and Drug Administration-approved immunoassays. All comparisons had R > 0.96 with slope ranging from 0.86 to 1.20. This is a simple and fast HPLC method suitable for measuring the 10 AEDs or metabolites. The use of the monolithic column resulted in increased sensitivity, better resolution, and a shorter analytical time compared with a regular C18 column.

Urinary cannabinoid mass spectrometry profiles differentiate dronabinol from cannabis use.

BACKGROUND: Dronabinol is used to treat a variety of conditions, including loss of appetite in people with AIDS and severe nausea and vomiting caused by cancer chemotherapy. Its therapeutic potential for pain management is now being explored in specific populations. Monitoring dronabinol compliance is challenging because its active ingredient, Δ-9-tetrahydrocannabinol (THC), is also present in cannabis. We developed a rapid LC-MS/MS assay with minimal specimen preparation to quantitate 11 cannabinoids in urine. Using this assay coupled with urine samples from normal controls, cannabis, and dronabinol users, we show the ability to differentiate cannabis from dronabinol use. METHODS: Residual clinical urine samples from 55 cannabinoid positive subjects and 31 negative controls, as well as prospective samples from 5 patients receiving dronabinol therapy were obtained for analysis. RESULTS: In the dronabinol group, only the THC metabolites 11-nor-9-carboxy-tetrahydrocannabinol (THC-COOH) and 11-hydroxy-Δ-9-tetrahydrocannabinol (THC-OH) were detected. Minor cannabinoids were detected in 91% of cannabis group samples and their detection was more frequent in samples with increased THC metabolite concentrations. Of minor cannabinoids evaluated, cannabigerol (CBG) and cannabidiol (CBD) had the greatest sensitivity in detecting cannabis use. CONCLUSIONS: This method has a high sensitivity for the detection of cannabis use with implications for evaluating dronabinol compliance.

The 2018 AACC/SYCL PhD Clinical Chemist Compensation Survey.

BACKGROUND: Doctoral level board-certified clinical chemists play an invaluable role in many facets of laboratory medicine and healthcare. However, information concerning their total compensation is sparse. CONTENT: A confidential self-reported compensation survey was conducted by the American Association for Clinical Chemistry's Society for Young Clinical Laboratorians (AACC SYCL) Core Committee from April 1 to April 17, 2018. Respondents provided information on geographic location, employment sector, gender, and years of experience to account for the influence of these variables on compensation. There were 199 respondents in total from the United States and Canada, however, only respondents employed in the United States with an earned doctoral degree and certification by the American Board of Clinical Chemistry (n = 133), were included in the full analysis. In comparison to compensation reported in AACC SYCL salary surveys conducted in 2010 and 2013, early career median salaries are trending upwards after correction for inflation. SUMMARY: This survey is the first to collect the gender of respondents, and identify a pay gap for some geographic groups. However, this gap could be due in part to a difference in the years of experience, since males were highly represented in the group with >20 years of experience (25 out of 35, 71%). Future studies on compensation trends within clinical chemistry that do not rely on self-report are needed to ensure accuracy and completeness of the dataset.