Annual meeting of the Royal Belgian Society of Laboratory Medicine: “Men’s health”

HEMOLYSIS EVALUATION USING ALINITY HEMOLYSIS INDEX: APPLICATION TO THE SURVEILLANCE OF MECHANICAL CIRCULATORY ASSIST DEVICES

V. Castiglione¹, L. Lutteri¹, R. Gadisseur¹, E. Cavalier¹, S. Robinet²

¹CHU de Liège / Clinical chemistry department, Liège, Belgium, ²CHU de Liège / Intensive Care Unit, Liège, Belgium

Objective

Impella (Abiomed) is a mechanical circulatory assist device used to improve the management of cardiogenic shock or cardiorespiratory failure. Hemolysis is a rare (<10%) complication, but if it is massive, the position of the Impella must be quickly checked, or even removed. Hence, it is recommended to measure plasma hemoglobin (PHb) within 8 hours after Impella installation. However, PHb is not measured 24/24h in our hospital because it is a time-consuming manual method. Instead, we wanted to assess the use of the Alinity hemolysis index (AHI)(Abbott), which is available 24 hours a day, 7 days a week.

Material and Methods

Seventy-nine samples were collected. PHb (mg/L) was measured by reference spectrophotometry method, and then compared within the hour to the AHI. The AHI results were multiplied 10 fold to be correlated with the level of PHb. Hence, the Impella hemolysis cut-off of 400mg/L of PHb correspond to an AHI of 40 (no unit).

Results

There was a bias between the methods, with AHI underestimating PHb (AHI×10 = 0.90 PHb -26.7). Yet, a linear correlation existed between both methods (p>0.05). Regarding the hemolysis cut-off for patients with Impella, the AHI had a false negative rate of 4% (n=3/79). As the AHI tends to under-estimate hemolysis, there was no false positive. Hence, in the absence of PHb measurement 24/24, we accepted the use of AHI in emergency to detect important hemolysis. In addition, this is in agreement with the only similar study published on the subject at this time (Burki C. et al, Clin Biochem, 2022). Nevertheless, AHI results should be considered positive above 30 instead of 40.

Conclusions

We showed that the AHI is able to identify patients with important hemolysis in emergency, provided that a lower cut-off is used. A more in-depth validation and comparison of the AHI need to be performed with more values around the cut-off.

DEVELOPMENT AND VALIDATION OF DIAGNOSTIC ALGORITHMS FOR THE LABORATORY DIAGNOSIS OF PORPHYRIAS

S. Lefever¹, N. Peersman¹, W. Meersseman¹, D. Cassiman¹, P. Vermeersch¹

¹University of Leuven, Leuven, Belgium

Objective: Porphyrias are rare metabolic disorders of the haem synthesis. They can present with acute neurovisceral attacks, cutaneous symptoms, or a combination of both. As they present with a wide variety of clinical symptoms, diagnosis is often delayed and correct interpretation of porphyria-related tests remains a challenge for many physicians. We developed and validated two algorithms for the laboratory diagnosis of porphyrias based on presenting symptoms.

Methods: Based on a literature search and clinical/laboratory expertise, we developed algorithms for acute and cutaneous porphyrias. We validated these algorithms using all porphyria related laboratory test requests between January 1st 2000 and September 30th 2020 in UZ Leuven. In addition, we also evaluated our algorithm using samples from the European porphyria network (EPNET) external quality assessment scheme (2010–2021).

Results: Sensitivity of the algorithm for acute porphyria (figure 1) was 100.0% [74.9%–100.0%] (13 acute intermittent porphyria (AIP) and 1 variegate porphyria [VP]) with a specificity of 98.5% [91.0%–100.0%] (65 patients). Sensitivity of the algorithm for cutaneous porphyria (fiugre 2) was 100% [95.1%–100.0%] (7 VP, 59 porphyria cutanea tarda (PCT), 23 erythropoietic protoporphyria (EPP), 2 X-linked erythropoietic protoporphyria [XLEPP]) with a specificity of 93.9% [82.9%–98.5%]. There were no diagnostic samples of other types of porphyria. The algorithms correctly identified 18 of the 19 EPNET porphyria cases. One of the two hereditary coproporphyria cases was missed.

Conclusion: The algorithms for acute and cutaneous porphyria showed high sensitivity and specificity and can be used to aid the clinician in correctly interpreting the laboratory findings of porphyria-related tests.

Figure 1: Diagnostic algorithm for acute porphyrias. *5% of AIP have a normal PBGD activity (including the non-erythroid form ofAIP). MDS: myelodysplastic syndrome.

Figure 2: Diagnostic algorithm for cutaneous porphyrias. *The sensitivity of plasma scan is 100% for symptomatic VP patients, but less than 85% in asymptomatic family members VP patients. **% zinc-chelated protoporphyrin depends on % clonal cells.

STUDY OF SERUM PARAOXONASE AND HIGH DENSITY LIPOPROTEIN FRACTIONS IN DIABETES WITH INFERTILITY

M. V. Rojekar ¹

¹Rajiv Gandhi Medical College, THANE, India

Objective: Significant alteration in lipid profile and antioxidant system occurs in response to diabetes mellitus (DM). The same is applicable to patients suffering from infertility. Paraoxonase (PON) enzyme is associated with high density lipoprotein (HDL). The HDL in human plasma consists of two main sub-fractions HDL2C and HDL3C. We studied the HDL subclasses and HDL associated enzyme paraoxonase with respect to diabetes with infertility.

Materials and Methods: The study was conducted in a tertiary care referral hospital in India. Total 80 subjects were included in the study. Lipid profile, PON1 arylesterase (ARE), PON1 lactonase (LACT) and HDL fractions were estimated. Regression analysis applied.

Results: PON1 ARE, LACT and HDL fractions are found to be decreased among cases than in controls. PON1 ARE & LACT showed negative correlation with blood glucose levels and HDL 3C while positive correlation with HDL 2C.

Conclusion: PON1 ARE and PON1 LACT activities reduction are due to increased oxidative stress. PON1 as well as HDL fraction levels are oxidative stress subjects. Among the HDL fractions, HDL2C is the more variable fraction and reflects changes in HDL. The study suggested that the protective role of total HDL against oxidative damage and complications is mainly mediated through HDL2C fraction.

EVALUATION OF THE SENTINEL DIAGNOSTICS REACTIVE FOR THE DETECTION OF SERUM LIPASE

N. Sanou¹, F. Yildirim¹, V. Farinella¹, A. Tamigniau¹

¹Charleroi University Hospital Center, Marie Curie Civil Hospital, Charleroi, Belgium, Charleroi, Belgium

Objectives: Lipase measurement is essential in the diagnosis and monitoring of acute or chronic pancreatitis, including therapy. The objective of this study was to compare results obtained with the Architect® (reference kit) and the Sentinel Diagnostics reagent kits (RKs). Furthermore we determined the clinical impact of switching to Sentinel Diagnostics RK. Both RKs use the principle of measuring the absorbance of a colored compound.

Materials and Methods: The evaluation of the RK was performed over 15 days and Biorad quality controls (QCs) were used. We analyzed repeatability, reproducibility, linearity interval, limits of detection and quantification (LoD, LoQ). We obtained bias and total acceptable error (TEa) from the results of a peer group available on the Biorad website. In order to compare the different RKs, 31 routine samples were chosen according to the thresholds for the diagnosis of acute pancreatitis and for normal values established for each RK: 8–78 U/L for Architect® and <60U/L for Sentinel Diagnostics. The Medcalc software was used for the statistical analysis; bias and TEa were compared with the Carmen Ricos table.

Results: Sentinel’s RK produced discrepancies in repeatability and reproducibility. The variation coefficients observed were higher than those advertized by the supplier for QC levels 1 and 2 (awaiting for company explanations). Calculated bias and TEa were lower than Ricos table’s thresholds (<11.31% &<37.88% respectively). LoQ and LoD were 4 and 0.8U/L, respectively, useful for the diagnosis of pancreatic insufficiency. Linear range was from 4 to 3500 U/L.

The bias was negative and a significant difference between RKs was observed (p<.001) especially for values >2000U/L. We designed a correspondence table in order to interpret the results obtained with the two RKs. The agreement between methods was 90.32% with major discrepancies in 3.24% of cases.

Conclusion: The Sentinel’s RK is suitable for the diagnosis of pancreatic insufficiency but could underestimate values of clinical significance. For a better comparison the analysis of a larger number of samples is required.

ASSOCIATIONS BETWEEN ENDOCRINE DISRUPTOR CONTAMINATION AND THYROID HORMONE HOMEOSTASIS IN BELGIAN TYPE 1 DIABETIC CHILDREN

P. Dufour¹, C. Pirard¹, M.-C. Lebrethon², C. Charlier¹

¹Laboratory of Clinical, Forensic and Environmental Toxicology, University of Liege (ULiège), CHU (B35), 4000 Liege, Belgium, Liege, Belgium, ²Department of Pediatrics, University of Liege (ULiège), CHU (B35), 4000 Liege, Belgium, Liege, Belgium

Objectives: Among the health concerns associated with the growing impact of human activity on environment, several chemicals contaminating our environment are suspected to interfere with the thyroid system homeostasis. According to the multiple hit hypothesis, individuals exposed to thyroid stressors (e.g.: anti-thyroid antibodies, pregnancy,…) are more susceptible to chemicals suspected to have thyroid-disrupting properties. Type 1 diabetic patients have been demonstrated to be at greater risk of developing thyroid problems. Therefore the aim of the present study was to explore associations between pollutants and thyroid and glucose homeostasis biomarkers in a Belgian population of type 1 diabetic children.

Material and Methods: Fifteen organochlorine pesticides, 4 polychlorinated biphenyls and 7 perfluoroalkyl substances were measured in the serum of 54 type 1 diabetic children (aged from 3 to 18 years old) while of 7 phthalate metabolites, 4 parabens, 7 bisphenols, benzophenone 3 and triclosan were measured in their urine. We explored the statistical associations of these pollutant concentrations and the blood levels of free thyroxine (fT4), thyroid stimulating hormone (TSH) and glycated hemoglobin (Hb1Ac).

Results: We observed that serum perfluorohexane sulfonate and urinary monoethylphthalate levels are positively correlated to TSH level in blood. PCB 138 was positively associated to fT4 while significant negative association between fT4 and bisphenol F was found. Finally, we highlighted positive associations between Hb1Ac levels and PCB 153 and two urinary phthalate metabolites: mono-2-ethyl-5-hydroxyhexyl phthalate and mono-2-ethyl-5-oxoxyhexyl phthalate.

Conclusions: Our results suggest that some environmental pollutants could interfere with thyroid and glucose homeostasis in type 1 diabetic children. Nevertheless, these findings should be confirmed by additional and larger scales studies.

SINGLEPLEX VERSUS MULTIPLEX KIT COMPARISON FOR TOTAL TAU AND NEUROFILAMENT LIGHT CHAINS USING SIMOA TECHNOLOGY ON SR-X DEVICE

A. Ladang¹, S. Kovacs¹, G. Sqalli¹, E. Cavalier¹

¹Clinical chemistry, CHU de Liège, Liège, Belgium

Introduction: The SiMoA technology is now taking the lead in the field of neurodegenerative biomarkers with an impressive amount of clinical studies published with this technology. Usually, it is commonly accepted that kits from the same company faces less bias than kits from different companies. However, before thinking to establish reference and use these kits in routine, the kit comparison needs to be done.

Method: 60 left-over cerebrospinal fluid (CSF) from patient undergoing Alzheimer’s disease investigation were used to measure total Tau (tTau) and Neurofilament light chains (NfL) on respective singleplex kits as well as on Neurology 3-Plex A and Neurology 4-Plex A. The analysis was realized on SR-X device from Quanterix. tTau concentrations from SR-X were also compared to the one obtained with Lumipulse G1200 from Fujirebio.

Results: No bias was detected for NfL between the single and the multiplex kits. Yet, regarding tTau major proportional bias ranging from 22 to 50 % were observed between kits from Quanterix. When comparing to Lumipulse kit, the bias was even increased up to 75 %. Additionally, two samples showed clinically discordant results. For these samples, clinical data were in favor of Lumipulse data.

Conclusion: For tTau, the concentration range dramatically differs in-between kits. Reference ranges should not be transferred from one type of kit to another. All results coming from a single study should be all obtained with the same kit. Additionally, authors working with Quanterix should always mention the type of kits used for their study to prevent from a massive confusion in the literature. These data highlight the need for standardization of NfL and tTau kits.

BIOLOGICAL VARIATION AND INFLUENCE OF RENAL FUNCTION IN DIAGNOSTIC POTENTIAL OF NEUROFILAMENT LIGHT CHAINS

A. Ladang¹, L. Otte¹, S. Alaoui¹, J. Rulmont¹, E. Cavalier¹

¹Clinical chemistry, CHU de Liège, Liège, Belgium

Introduction: Neurofilament light chains (NfL), are neuron-specific proteins involved in axonal maintenance. Many studies have focused on NfL clinical potential for the diagnostic of Alzheimer’s disease (AD). Yet, some gray zone remains on the physiological variation of NfL. Indeed, serum NfL was recently shown to be increased in case of renal insufficiency but doubts remains on the impact of renal function on NfL cerebrospinal fluid (CSF) concentrations. In this study, we checked whether CSF diagnostic accuracy for AD depends on renal function. We also established serum biological variation of NfL.

Methods: For biological variation, serum samples were collected once a week between 7 am and 9 am, from 21 healthy volunteers aged between 19 and 25 years old, during a period of 5 weeks. For diagnostic accuracy, 33 left-over CSF from patients undergoing AD investigation aged between 40 and 86 were selected. NfL was measured for both serum and CSF samples with the SiMoA technology, using the SR-X instrument provided by Quanterix.

Results: Serum intra-individual coefficient of variation was 4.41 % and inter-individual coefficient of variation was 22.97 %. NfL was increased in AD patients compared to non-AD patients (p=0.007). No correlation was observed between NfL and Tau, pTau or ratio Amyloid β 42/40. Regarding patients suffering from chronic kidney disease (CKD), NfL was not correlated to creatinine levels, nor increased in CKD patients compared to non CKD (p=0.556). The ROC curve based on the entire cohort gave an area under the curve (AUC) of 0.774. AUC was not modified when the patient suffering from CKD were retrieved (AUC= 0.811).

Conclusion: NfL has a high inter-individual but a low intra-individual variability in serum. In AD, diagnostic accuracy is not modified by CKD patients. However, these results needs to be confirmed in a bigger cohort.

ANALYTICAL COEFFICIENT OF VARIATION AND BIOLOGICAL VARIATION OF MIRNAS INVOLVED IN BONE METABOLISM

J. Rulmont¹, L. Otte¹, E. Cavalier¹, A. Ladang¹

¹Clinical chemistry, CHU de Liège, Liège, Belgium

Objectives: Osteoporosis is a pathology characterized by an increased risk of bone fracture causing potentially disastrous consequences for the patient, including a loss of their autonomy or even a higher risk of mortality. The tools currently used to track down this pathology are efficient but have shown some limits, especially when it comes to diabetic osteoporosis or non-postmenopausal osteoporosis. In this context, microRNAs represent a potential hope for the future. Indeed, modification of miRNA plasma concentration can be observed years before the occurrence of a fracture. Although many miRNAs have been described in the context of bone metabolism, very few is known about their analytical properties. The aim of this work is to establish the analytical coefficient of variation and biological variation of some miRNAs involved in bone metabolism.

Materials and methods: Plasma samples were collected once a week between 7 am and 9 am, from 21 healthy volunteers aged between 19 and 25 years old, during a period of 5 weeks. The samples were directly centrifuged. MicroRNAs were extracted, transformed into their complementary DNA strand, and analyzed by a qRT-PCR using the osteomiR® kit from TAmiRNA.

Results: Analytical coefficient of variation (CVa) was comprised between 0.44 % and 2.09%. Intra-individual biological variation (CVi) fluctuating from 0,59% to 2,57% and inter-individual biological variation (CVg) fluctuating from 0% to 2,76% were also noticed for the 19 miRNAs. Moreover, no correlation between the number of cycles and the CVi or the CVg have been observed. This work has also led to the identification of a correlation between the analytical variation coefficient and the number of cycles.

Conclusion: Firstly, miRNAs are currently presented in the literature as stable markers, whose rates do not fluctuate over time. However, this work has shown that, like all biomarkers, miRNAs have a biological variation over time. Secondly, the absence of a correlation between the Cp and the CVI or CVG reinforces the hypothesis of the existence of physiological fluctuations of miRNAs levels in plasma. Thirdly, the existence of a correlation between the number of cycles and the CVa tends to confirm the idea, already present in the literature, that the results obtain by a PCR analysis decrease in reliability beyond 35 cycles.

A CASE OF VITAMIN B12 DEFICIENCY NEUROLOGICAL SYNDROME IN A YOUNG ADULT DUE TO LATE-ONSET COBALAMIN C (CBLC) DEFICIENCY: A DIAGNOSTIC CHALLENGE

S. O. Ailliet¹, R. Vandenberghe¹, T. Schiemsky¹, L. van Overbeke², P. Demaerel¹, W. Meersseman¹, D. Cassiman¹, P. Vermeersch¹

¹UZ Leuven, Leuven, Belgium, ²AZ Sint-Maarten, Mechelen, Belgium

Vitamin B12 deficiency can present with neurologic and psychiatric symptoms without macrocytic anaemia. We describe a case of late-onset cobalamin C deficiency which typically presents with normal serum vitamin B12 concentrations, posing an additional diagnostic challenge. A 23-year-old woman with decreased muscle strength and hallucinations was diagnosed with ‘catatonic depression’ and admitted to a residential mental health facility. She was referred to our hospital for further investigation 3 months later. Heteroanamnesis revealed that the symptoms had been evolving progressively over several months. Magnetic resonance imaging (MRI) of the brain showed diffuse symmetrical white matter lesions in both hemispheres. Routine laboratory tests including vitamin B12 and folic acid were normal except for a slight normocytic, normochromic anaemia. Over the next 6 weeks her symptoms deteriorated, and she became unresponsive to stimuli. A new MRI scan showed progression of the white matter lesions. The neurologist requested plasma homocysteine (Hcys) which was more than 8 times the upper limit of normal. Further testing revealed increased methylmalonic acid and the patient was diagnosed with adult-onset cobalamin C deficiency. This case illustrates that Hcys and/or methylmalonic acid should be determined in patients presenting with neuropsychiatric symptoms suggestive of vitamin B12 deficiency with a normal serum vitamin B12 to rule out a late-onset cobalamin C deficiency.

DEVELOPMENT OF A SENSITIVE LC/MS-MS METHOD FOR ESTRADIOL AND ESTRONE QUANTIFICATION

C. Le Goff¹, C. Defourny¹, J. Deleersnyder¹, S. Peeters¹, C. Nix¹, A.-S. Gendebien¹, N. Ferrante¹, M. Rechchad¹, E. Cavalier¹

¹Clinical Chemistry department, CHU of Liege and CIRM, Liege, Belgium, Liege, Belgium

Objectives: Estrogens are hormones responsible for the development and maintenance of the female reproductive organs and secondary women sexual characteristics. They are also involved in the regulation of the menstrual cycle and the maintenance of pregnancy. The main difficulty for the quantitative analysis of these molecules is their very low serum concentration (in the order of pg/mL). The aim of this work was the development of a LC/MS-MS assay method to concurrently determine the two major endogenous estrogens, estrone (E1) and 17β-estradiol (E2).

Material and methods: The sample preparation procedure is based on liquid-liquid extraction and derivatization with dansyl chloride. After chromatographic separation performed on a C18 column, the compounds were analyzed using a Waters® XEVO TQ-XS triple-quadrupole mass spectrometer operating in multiple reaction monitoring mode. The ionization source used is the electrospray in positive mode. The procedure was validated by testing 8 levels of concentration (from 1 to 800 pg/mL) in quadruplate during 3 different days. A calibration curve was prepared using 7 points and responses were determined by calculating the integrated peak area ratio between E1 and E2 to deuterated respective compound. According to these results, the precision, the trueness, and the linearity were calculated. A comparison was performed (n=27) between our method routinely used (QTrap 6500, Sciex) with our newly developed method (Xevo TQ-XS).

Results: The relative biases obtained were between 1.3% and 4.9% for estrone and between 0.1% and 11.5% for estradiol. The intra-assay CV did not exceed 4.6 and 6.9% and inter-assay CV, 6.9% and 7.6% for estrone and estradiol respectively. The measurement uncertainty values obtained were < 13.9% for estrone and < 16.2% for estradiol. Very good sensitivity was also demonstrated, with LLOQs calculated at 3.3 pg/mL and 3.7 pg/mL for estrone and estradiol respectively. No carry-over has been demonstrated. For the method correlation, regression curves were: Old E1 method = -0, 2655 + 1, 0157 New E1 method and Old E2 method = -0, 5051 + 1, 0379 New E2 method.

Conclusion: Our new method has been successfully validated according to all the criteria set by the European Medicines Agency (EMA) guidelines.

QUANTIFICATION AND POSSIBLE DEGRADATION KINETICS OF IOHEXOL IN HUMAN FLUIDS USING MASS SPECTROMETRY COUPLED TO LIQUID CHROMATOGRAPHY

P. Massonnet¹, S. Peeters¹, E. Grifnée¹, J. Demeuse¹, L. Huyghebaert¹, T. Dubrowski¹, P. Dufour¹, J. Farre Segura¹, P. Delanaye², C. Le Goff¹, E. Cavalier¹

¹Clinical Chemistry department, CHU of Liege and CIRM, Liege, Belgium, Liege, Belgium,

²Department of Nephrology-Dialysis-Transplantation, University of Liège, CHU , Liège, Belgium, Liege, Belgium

Objective

Iohexol clearance is a reference method to measure the glomerular filtration rate (GFR, an indicator of kidney function). After injection of a fixed volume of iohexol, plasma and/or urine iohexol clearance is evaluated by measuring iohexol in plasma and/or urine at different time points, which allows calculation of plasmatic/urine clearance of the product. However, some discrepancies have been observed between urine and plasma results from the same patient and no study has ever clearly explained this. Yet, this work aims at assessing molecule profile variations occurring over time with patients undergoing iohexol clearance using LC coupled with high resolution mass spectrometry.

Method

In this project, urine samples were taken at given timepoints from routine samples obtained in patients who underwent GFR measurement by iohexol clearance. The samples were first centrifuged, and the supernatant was diluted 100 times with water before injection in a NanoACQUITY UPLC system coupled with a SYNAPT XS instrument operating in positive ion mode. The mobile phases were composed of water and acetonitrile (+0.1% formic acid each). Standard curve points and quality controls were prepared by spiking a commercial preparation of iohexol (Omnipaque®) in serum free of the product at various concentrations.

Results

The iohexol LC-MS/MS method on a triple quadripole instrument was successfully implemented on the SYNAPT XS – NanoACQUITY UPLC system. The first results on different patient urines showed the presence of other peaks different from iohexol in patient samples. Interestingly, those peaks were not present in standards and in urines of patients who did not receive the drug.

Conclusion

Investigation of all mass spectra is in progress and these results open the possibility for a large screening over time. Comparison with data obtained using LC-MS/MS is also in progress and the next step is the comparison with data obtained on plasma samples.

LIPOPRINT: A NEW TOOL FOR THE CARDIOVASCULAR RISK ASSESMENT?

C. Le Goff¹, A. Gbahouo Guelasso¹, S. Cugnata¹, E. Brevers¹, J. Pincemail¹, S. Peeters¹, E. Cavalier¹

¹Clinical Chemistry department, CHU of Liege and CIRM, Liege, Belgium, Liege, Belgium

Objectives: Lipoproteins are spherical particles responsible for the transport of cholesterol, triglycerides and phospholipids. Increasing LDL cholesterol has been identified as a major risk factor for coronary artery disease. It is notified that the lipoprotein classes are heterogeneous as they consist of a set of subfractions diversified in particle size, density and chemical composition. This heterogeneity has been demonstrated by various analytical procedures. The Lipoprint system is able to break down LDL into up to seven subfractions. These are numbered from LDL-1 for the largest particles to LDL-7 for the smallest. The aim of our study was the evaluation of the Lipoprint system to improve the management of cardiovascular diseases.

Material and methods: The LDL subfractions kit (Quantimetrix) was evaluated in plasma. The intra- and inter-assay precision was determined for the different LDL subfractions at 4 levels in triplicate during 3 days (3 patient samples and 1 quality control). Fifteen patients for which one the lipid profil was requested were selected according to their cholesterol levels. Age, gender and treatment were recorded to study the relationship between these and the LDL subfractions results.

Results: The intra-and inter-assay CV for the different LDL subfractions evaluable did not exceed 11.9 % (for LDL-1 for concentration ranged from 19 to 30 mg/dL) and 12.2% (for LDL-2 for concentration ranged from 13 to 27 mg/dL). The maximum relative bias was 2.1%. In the clinical study, we observed that the patients with a high cholesterol level have systematically too many LDL 3–4 subfractions which are atherogenic fraction. Something interesting to note is that some patients with a normal total cholesterol level showed also increased small LDL subfractions. Indeed, differences in LDL heterogeneity among subjects are related to both genetic and environmental factors.

Conclusions: The method shows a good agreement for the high concentration of LDL subfractions that we were able to test. In the patient study, we observed clearly the effect of the different treatment on the LDL 3 to 7 which are the most atherogenic. The monitoring of LDL subfractions is therefore a new tool useful before the start of a treatment (diet, drugs) as well as for the follow-up of their effectiveness.

VALIDATION OF A LC-MS/MS METHOD FOR THE QUANTITATION OF PLASMA RENINE ACTIVITY (PRA) IN HUMAN

L. Huyghebaert¹, J. Demeuse¹, J. Farre Segura¹, S. Peeters¹, E. Grifnée¹, N. Vanden Broeck¹, G. Collard¹, P. Dufour¹, P. Massonnet¹, T. Dubrowski¹, E. Cavalier¹, C. Le Goff¹

¹Clinical Chemistry department, CHU of Liege and CIRM, Liege, Belgium, Liege, Belgium

Objective

The renin-angiotensin-aldosterone-system is important to control blood pressure. In order to establish a difference between primary/secondary hyperaldosteronism and also reninoma diagnosis the quantitation of PRA as a part of diagnosis has been widely use. The aim of this work was to validate an LC-MS/MS method to measure angiotensin I (Ang I) in plasma.

Materials and Method

Quantification was achieved on a Nexera X2 UPLC (Shimadzu Corporation, Kyoto, Japan) coupled to a QT5500 mass spectrometer (Sciex, CA, USA). Chromatographic separation was performed on a Luna Omega® C18 100Å core-shell column (100 × 2,1 mm, 1.6 µm) from Phenomenex (Torrance, CA, USA). The mobile phases composition was H₂O and acetonitrile both containing 0,2% formic acid (FA). After an incubation for generation of Ang I, internal standard (Ang I¹³C,¹⁵N) in H₂O 10% FA were added. The samples were then extracted on an OASIS HLB 96-well plate (Waters, Massachusetts, USA), evaporated, reconstituted and then filtered. Four levels of concentration (0.075, 0.3, 18, 36ng/mL) in quintuplicate during 5 different days were tested according to the European Medicines Agency (EMA) guidelines. Integrated peak area ratio between Ang I to Ang I¹³C,¹⁵N was used for quantification. According to these results, the assay performance characteristics were assessed. Statistical analysis were performed using EP Evaluator software.

Results

Within-run and between-run CVs didn’t exceed 7.8 % for the concentration between 0.075 and 36 ng/mL. LOQ was 0.075 ng/mL. Mean matrix effects (plasma) ranged from 77 to 92%, so no matrix effect was observed. The linearity was fit for the interval 0.075–45 ng/mL. The linearity of dilution is acceptable up to 1/10. Mean recoveries ranged from 38 to 49%. The measurement uncertainty was < 5.4% and the bias was < 3.7%.

Conclusion

Our LC-MS/MS method to determine plasma Ang I concentration has been validated according to the EMA guidelines and shows a good specificity and accuracy.

URINARY TEST STRIPS ADD VALUE IN THE IDENTIFICATION OF PATIENTS WITH CHRONIC KIDNEY DISEASE

T. Y. Chin¹, K. De Waele¹, S. Lambrecht¹, M. Speeckaert², M. Oyaert¹

¹Department of Laboratory Medicine, Ghent University Hospital, Ghent, Belgium, ²Department of Nephrology, Ghent University Hospital, Ghent, Belgium

Objective

Identifying patients at risk for chronic kidney disease (CKD) using urinary test strips may be a viable alternative. We aimed to evaluate the performance of urinary protein-creatinine ratio (PCR) and albumin-creatinine ratio (ACR) determined by urine test strips, as compared to quantitative lab tests.

Material and methods

Three hundred randomly selected urine samples were analysed. Urinary PCR and ACR were determined using both quantitative lab tests (Abbott Architect c16000 and Roche Cobas 8000, respectively) and semi-quantitative urinary test strips (Sysmex UC-3500). Patients were classified in CKD-risk categories according to the KDIGO guidelines.

Results

Proteinuria (PCR > 0.15 g/g creatinine) and albuminuria (ACR > 30 mg/g creatinine) were found in 46.1% and 44.9% of patients. Urinary test strip PCR showed a concordance of 76.5% with the quantitative methods. Sensitivity, specificity, negative and positive predictive value at a cut-off of 0.15 g/g were 85.1% (95%CI: 76.9–91.2%), 77.4% (95%CI: 69.0–84.4%), 85.7% (95%CI: 79.1–90.5%), and 76.5% (95%CI: 69.9–82.0%). Urinary test strip ACR showed a concordance of 77.7% with the quantitative methods. Sensitivity, specificity, negative and positive predictive value at a cut-off of 30 mg/g were 90.7% (95%CI: 84.1–95.3%), 76.7% (95%CI: 69.0–83.3%), 91.1% (95%CI: 85.2–94.5%), and 76.1% (95%CI: 70.2–81.1%). When classified into CKD-risk categories, there was a class discrepancy in 39 (18.9%) of the samples between the quantitative and semi-quantitative test strip results. 30 of these 39 samples were classified one risk class worse and 9 were classified one class better than the reference, showing a trend of overestimating the risk score by test strip ACR.

Conclusions

Because of the high sensitivity and negative predictive values, both semiquantitative ACR and PCR test strip results may add value to identify CKD patients. The poor specificity can be overcome by performing a quantitative urinary albumin and protein reflex test on samples positive for ACR and PCR, respectively.

AN OVERVIEW ON THE BIOCHEMICAL ASSESSMENT OF MALE ANDROLOGICAL STATUS USING SEX HORMONE-BINDING GLOBULIN AND TOTAL AND FREE TESTOSTERONE IN EUROPEAN CLINICAL LABORATORIES

N. Narinx^1,2, K. David^1,3, J. Walravens⁴, G. Snaterse⁴, P. Vermeersch², T. Fiers⁵, B. Lapauw^4,6, D. Vanderschueren^1,3, L. Antonio^1,3

¹ Laboratory of Clinical and Experimental Endocrinology, Department of Chronic Diseases and Metabolism, KU Leuven, Belgium

² Department of Laboratory Medicine, University Hospitals Leuven, Belgium

³ Department of Endocrinology, University Hospitals Leuven, Belgium

⁴ Department of Internal Medicine and Pediatrics, Ghent University, Belgium

⁵ Department of Laboratory Medicine, Ghent University Hospital, Belgium

⁶ Department of Endocrinology, Ghent University Hospital, Belgium

Background: Male hypogonadism (HG) manifests by one or more signs of androgen deficiency. However, prior to starting testosterone replacement therapy (TRT) in patients with HG, biochemical confirmation of deficient gonadal production of testosterone (T) is necessary. Despite the recommendation in clinical practice guidelines of assessing circulating total T levels, to date there is no agreement on differentiating low from normal circulating total T. Several studies have demonstrated the potential added value of free T in context of HG. The Endocrine Society (ES) suggests adding free T as an extra biochemical marker when total T is near the lower limit of normal or in men that have a condition altering sex hormone-binding globulin (SHBG) concentrations.

Surveys conducted in clinical laboratories in the United States of America, the United Kingdom and the Republic of Ireland uncovered that inconsistencies exist between guidelines on HG and clinical practice. These studies highlighted a high variability in the methodology on biochemical assessment of male andrological status. Different assay types for total and free T are used, lacking standardization and harmonization between laboratories, resulting in divergent reference values for total and free T. As diagnosis of HG and prescription of TRT is highly dependent on biochemical assessment, and TRT is not without risk of adverse events, harmonization of laboratory practice concerning total and free T is strongly recommended.

Objective: The purpose of our study is to map the diagnostic landscape concerning methodology on biochemical assessment of male andrological status (total T, free T and SHBG) throughout Europe.

Methods: A web-based survey has been issued to European clinical laboratories via several platforms (including the RBSLM). The survey contains questions on patient sampling, analytical methodology, reference ranges and post-analytical phase of SHBG and total, free and bioavailable T. Distribution of the survey has started on the 1^st of September 2022.

Results: As the survey has only just been launched, results are not available at the time of abstract submission. Responses are expected to come in through the course of September and October. By November we will have collected and analyzed preliminary data to present at the RBSLM annual meeting.

VERIFICATION OF THE CLINICAL CUTOFF FOR SERUM CORTISOL AFTER ACTH STIMULATION MEASURED BY THE SIEMENS ATELLICA IM CORTISOL IMMUNOASSAY

K. De Letter¹, S. Vandewalle², A. Vanden Bruel², M. Langlois¹

¹Dept. Laboratory Medicine AZ Sint-Jan Brugge, Brugge, Belgium, ²Dept. Endocrinology AZ Sint-Jan Brugge, Brugge, Belgium

Objective: The synacthen test – used for the assessment of adrenal insufficiency – evaluates the serum cortisol concentration after IV administration of synthetic ACTH. Since 2021, the proposed clinical cutoff for normal serum cortisol response 30 minutes after ACTH stimulation has been reduced from 18 µg/dL to 14 µg/dL. This applies to the newer generation of cortisol immunoassays, which employ specific monoclonal antibodies, and LC-MS/MS (1). To our knowledge, this is the first study aiming to verify the 30-minutes cutoff value with the Siemens Atellica IM cortisol immunoassay.

Material and Methods: Serum samples of 36 patients undergoing synacthen test were evaluated. Basal and 30-minutes values of cortisol were measured in parallel by the Siemens Atellica IM cortisol immunoassay (using polyclonal cortisol antibodies) and the Roche Cobas Elecsys cortisol II immunoassay (using monoclonal cortisol antibodies).

Results: Passing-Bablok regression of the 30-minutes cortisol concentration is y(Siemens) = –1.125754 + 1.300794x(Roche) with significant correlation (r=0.955, P<0.0001). Median difference between the 30-minutes values is 4.55 µg/dL (IQR 3.15–5.90). Bland-Altman plot of the 30-minutes values shows a mean bias of +5 µg/dL (95%CI 0.3–9.7) of Siemens relative to Roche immunoassay. Using a 14 µg/dL cutoff with Siemens immunoassay would result in discordant interpretation in 8% of patients compared to the historical 18 µg/dL cutoff (which yields only 1 discordant classification compared to Roche immunoassay with 14 µg/dL cutoff).

Conclusion: The original clinical cutoff of 18 µg/dL for the 30-minutes cortisol value should be maintained to allow correct interpretation of adrenal function using the less specific Siemens Atellica IM cortisol immunoassay.

Reference: (1) Javorsky BR et al. New cutoffs for the biochemical diagnosis of adrenal insufficiency after ACTH stimulation using specific cortisol assays. J. Endocrine Society 2021;5:1–11. https://doi.org/10.1210/jendso/bvab022.

THE ADDED VALUE OF METHYLMALONIC ACID AS INDICATOR OF VITAMIN B12 DEFICIENCY

L. Dedeene¹, S. Theunynck¹, O. Heylen¹, N. Callewaert¹, K. Croes¹

¹Laboratory Medicine, AZ Groeninge Hospital, Kortrijk, Belgium, Kortrijk, Belgium

Objectives

To improve the diagnostics for vitamin B12 deficiency in the clinical lab of AZ Groeninge, we aimed to assess the added value of MMA in a hospital-specific patient population and to evaluate the usefulness of eGFR-corrected MMA concentrations (Van Loon et al., 2018).

Material and methods

The study was approved by the local Ethics Committee. The MMA MassChrom reagent kit (Chromsystems) was implemented on LC-MS/MS and the reference values of Erdogan et al., 2010 were verified (60–360 nmol/L) using 20 healthy volunteers (CLSI EP28-A3C). As recommended in literature, a grey zone for MMA was applied (360–800 nmol/L) to account for mildly increased MMA levels associated with renal impairment or increasing age. Afterwards, 124 randomly selected leftover patient samples with B12 concentrations <400 ng/L (Cobas e602) were analyzed for MMA and active B12 (Cobas e602). Homocysteine (LC-MS/MS) was determined in 43 of the 124 patients.

Results

Only 21% (9/42) of the patients with B12 lower than the cut-off of 200 ng/L had an MMA concentration indicative of a functional B12 deficiency (>800 nmol/L). In 11% (9/82) of the patients with apparently normal B12 concentrations (200–400 ng/L), MMA analysis revealed a functional deficiency. When using the combined B12 score based on three or four biochemical markers (Fedosov et al., 2015) to distinguish (sub)clinical B12 deficiency from B12 adequacy, the MMA cut-off of 360 nmol/L had a sensitivity of 93% and a specificity of 78%, whereas the total B12 cut-off of 200 ng/L had a lower sensitivity (63%), but a comparable specificity (81%). Finally, in patients with an eGFR<90 mL/min/1.73m², 16% (7/43) of the samples with an observed MMA >360 nmol/L are reclassified as B12 non-deficient after correction for eGFR (corrected MMA <360 nmol/L).

Conclusions

Our data demonstrate the added value of MMA to determine a patient’s B12 status. Furthermore, eGFR-corrected MMA concentrations are useful to prevent overdiagnosis of B12 deficiency.

IMPACT OF IMPLEMENTATION OF FETAL MEDICINE FOUNDATION ALGORITHM FOR FIRST TRIMESTER PREECLAMPSIA SCREENING IN PERIPHERAL HOSPITAL

M. Briers¹, I. Brandt¹, E. Scheire²

¹Department of Laboratory Medicine, OLV Hospital, Aalst, Belgium, ²Department of Gynaecology, OLV Hospital, Aalst, Belgium

Objective: To evaluate the impact of introducing the Fetal Medicine Foundation (FMF) algorithm for the first trimester screening for preeclampsia (PE) into clinical practice in OLV hospital campus Aalst and Asse.

Materials and methods: This study included 301 pregnant women between 11 and 14 weeks of gestation. For each patient a serum sample was collected for measuring the placental growth factor (PlGF) using the Elecsys® immunoassay (Roche Diagnostics). According the FMF algorithm, the screening for preterm PE was performed based on a combination of maternal risk factors, mean arterial pressure, mean uterine artery pulsatility index and the serum biomarker PlGF. Women with a calculated preterm PE risk of ≥1/150 were classified as high-risk. The risk assessment, generated by the FMF algorithm, was reported to the attending gynaecologist, so in case of high-risk for PE, treatment with low-dose aspirin (160 mg/day) could be initiated. The goal of this study was to evaluate the added value and the challenges of implementing this screening tool.

Results: For 204 out of 301 women all data were available to complete the FMF algorithm. In this subpopulation 28 women were additionally classified as high-risk for PE by the FMF algorithm compared to the current screening (i.e. the assessment by the gynaecologist, as noted in the medical file). Five women were classified as low-risk based on the FMF algorithm unlike the high-risk assessment of the gynaecologist. McNemar test showed a significant difference in the risk assessment (p=0.0001) of 11.27%. Some gynaecologists experienced more time pressure during consultation, because of the four blood pressure and two Doppler measurements. In addition, the collection and import of the data in the algorithm were time consuming and sensitive to errors.

Conclusion: The implementation of the FMF algorithm caused a significant different risk assessment compared to the current screening, but it requires effort from both the department of gynaecology and from the laboratory.

STATISTICAL APPROACHES TO COMPARE LEVELS OF 9 ENDOCRINE DISRUPTING CHEMICALS IN HAIR, SERUM AND URINE

J. Claessens¹, C. Charlier¹, C. Pirard¹

¹Laboratory of Clinical, Forensic, Industrial and Environmental Toxicology, University Hospital of Liege, Liege, Belgium

Objectives

Human biomonitoring studies (HBM) are generally performed using blood or urine, although hair has gained attention as an alternative matrix. Correlations have been studied between the levels of pollutants measured in traditional and alternative matrices to evaluate the reliability of using hair in HBM. However, the results use to be inconsistent and frequently meaningless, questioning the initial hypothesis or the study design. Therefore, the aims of this work were to investigate different statistical methods to assess the comparability of the results provided by different matrices, and to study the legitimacy of hair analyses in HBM.

Material and method

Thirty participants provided spot urine, blood and hair samples. Bisphenols A and S (BPA, BPS), methyl, ethyl and propylparaben (MeP, EtP, PrP) were measured in urine, while PFOA, PCB-153 and -180 were determined in serum. All of them were also analyzed in hair. Spearman’s coefficient (r_s), intraclass correlation coefficient (ICC) and Cohen’s Kappa (k) were calculated to measure correlations or agreements of exposure classification between matrices.

Results

ICC wasn’t significant for any of the compounds. R_s were significant for EtP, PrP, and PCB-180, while k showed good agreement for PCB-180, moderate for PrP and EtP, weak for MeP, very weak for PCB-153 and PFOA, and no agreement for BPA and BPS.

Conclusion

The lack of correlation for BPA and BPS between hair and urine was expected since their urinary levels are known to show high variability. Significant correlations between matrices for parabens and PCB-180 could be attributed respectively to the more constant exposure and to persistent character. The lack of correlation and the very weak agreement for PFOA and PCB-153 were not expected. R_s and k provided comparable and useful information while ICC seemed less relevant for comparison between matrices. This work contributes to provide data on the reliability of hair as an alternative matrix for HBM.

INTEGRATION OF DIGITAL MICROSCOPY AND FLOW CYTOMETRIC ANALYSIS OF SOLID ELEMENTS IN URINE: THE BEST OF BOTH WORLDS AND THE GATE TO TOTAL AUTOMATION

T. Vanhove¹, J. Billen¹

¹University Hospitals Leuven (UZ Leuven), Department of Laboratory Medicine, Leuven, Belgium, Leuven, Belgium

Background

Urinalysis, one of the oldest medical techniques, has arrived in the 21st century. [1] With the introduction of automated analytical systems, solutions have been developed to automate the counting of particles in urine.

Objectives

Multiple goals are pursued in this development. To reduce the timeframe in which results can be expected and thus avoid treatment in cases in which otherwise unnecessary. The time-saving effect frees resources for additional samples, increasing the daily throughput of the laboratory. Further, to free the qualified personnel for cases in which their expertise is actually needed instead of occupying them with readily distinguishable negatives. And, to allow generation of reproducible results with standardized procedures [2].

Material and Methods

A literature review was carried out in light of a technical, economic and medical content replacement. It addresses the performance of traditional, routine chemical and manual microscopic analysis, as well as automated techniques.

Results

While test strip analyzers can apparently offer (similarly) quick results, they detect only a certain subset of diseases and infections. To this end, technical progress can be based on flowcytometry, along with automated microscopic particle analysis [3].

The use of fluorescence flowcytometry allow the rapid identification and differentiation of particles such as leukocytes, bacteria and even iso- or dysmorphic red blood cells. [2] Besides, it allows a reproducible assessment of particle concentration in urine and consequentially a reliable screening procedure for urinary tract infections and hematuria cases. [4] Along with other more arcane parameters like epithelial cells, spermatozoids, casts, crystals and even atypical cells. [5]

The decision to perform microscopy, independently from the physician’s request, should be made by the laboratory. In case of samples showing abnormalities, either laboratory-defined or as indicated by the analyzer, visual microscopic review remains necessary. [6] Definitely in consideration of parameters that can’t be identified unambiguously using flow cytometry; for instance differentiation of casts, non-squamous epithelial cells, dysmorphic erythrocytes, fungi, parasites and clinically significant crystals.

Conclusion

The added value and complementarity of digital microscopy for qualitative reviews on indication becomes clear. The sample throughput of “next-generation” analytical systems combined with the experience in fluorescence flowcytometry and digital microscopy can contribute greatly to the laboratory workflow.

References

1. Rifai N, Horvath A, Wittwer C. Tietz Textbook of Clinical Chemistry and Molecular Diagnostics. Sixth Edit. Rifai N, editor. Elsevier; 2018.

2. Oyaert M, Delanghe J. Progress in automated urinalysis. Ann Lab Med. 2018;39(1):15–22.

3. Delanghe JR, Kouri TT, Huber AR, Hannemann-Pohl K, Guder WG, Lun A, et al. The role of automated urine particle flow cytometry in clinical practice. Clinica Chimica Acta. 2000;301(1–2):1–18.

4. Oyaert M, van Meensel B, Cartuyvels R, Frans J, Laffut W, Vandecandelaere P, et al. Laboratory diagnosis of urinary tract infections: Towards a BILULU consensus guideline. J Microbiol Methods. 2018;146(November 2017):92–9.

5. Yuno T. Fully Automated Urine Particle Analyzer. UF-series Clinical Case Study. Sysmex Corporation; 2016.

6. Gai M, Piccoli G, Segoloni G, Lanfranco G. Microscopic urinalysis and automated flow cytometry in a nephrology laboratory. Clin Chem. 2003;49(9):1559–60.

LABORATORY EXPLORATION OF A VERY HIGH LEVEL OF INSULIN MEASURED IN A LONG-STANDING TYPE 1 DIABETIC PATIENT

P. Dufour¹, E. Grifnée¹, E. Brevers¹, N. Paquot², C. Le Goff¹, E. Cavalier¹, A. Ladang¹

¹Department of Clinical Chemistry, University Hospital of Liège, CHU (B35), 4000 Liège, Belgium, Liege, Belgium,

²Division of Diabetes, Nutrition and Metabolic Disorders, Department of Medicine, University Hospital of Liège, 4000 Liège, Belgium, Liege, Belgium

Case description: During a routine biological control prescribed by her general practitioner, an insulin concentration of 11850 pmol/L was measured with Abbott Alinity in the serum of a type 1 diabetes 67-year old woman. The disease has been present for 50 years, currently treated with insulin aspart and insulin detemir, and characterized by the occurrence of sudden hypoglycaemia and sometimes significant hyperglycaemia. Despite the very high level of insulin, blood glucose concentration was above the normal range (159 mg/dL) and the patient did not present any signs of hypoglycaemia which, a priori, excluded therapeutic insulin overdose.

Laboratory investigations: Treatment of the sample with VeraPrep^TM excluded the presence of heterophilic antibodies. Yet, the presence of macroinsulin was suspected. After polyethylene glycol precipitation (25%), a recovery of 8% was observed, this was in accordance with the macroinsulin hypothesis. The presence of anti-insulin antibodies (16.09 U/mL measured with Medipan Medizym IAA ELISA) was then discovered. Finally, insulin serum concentrations were also measured with the Siemens Atellica and Roche cobas platforms and were found at 1590 and 6.74 pmol, respectively.

Discussion: The important differences observed between apparatus are consistent with the different cross-reactions between immunoassays and commercial insulins (Parfitt et al., 2015). The presence of macroinsulin in patient could have clinical consequences. Indeed, on one hand, anti-insulin antibodies could reduce the efficiency of insulin treatment by inactivating the injected hormone and, on the other hand, insulin could be released from the insulin-antibody complex between 2 injections, potentially inducing hypoglycemia. From an analytical perspective, the exploration of macroinsulin in suspicious samples should better be standardized and from a clinical perspective, this exploration could be interesting in patients presenting poorly controlled diabetes.

Reference: Parfitt, C. et al., 2015. Clin Biochem 48, 1354–1357.

SNPS POTENTIALLY AFFECTING SHBG OR TOTAL TESTOSTERONE DO NOT SUBSTANTIALLY AFFECT CALCULATED ESTIMATES OF FREE TESTOSTERONE

J. Walravens¹, G. Snaterse¹, T. van den Eynde¹, T. Reyns², N. Narinx³, L. Antonio³, T. Fiers², J.-M. Kaufman¹, B. Lapauw¹

¹Ghent University Hospital, Department of Endocrinology, Ghent, Belgium,

²Ghent University Hospital, Laboratory for Clinical Biology, Ghent, Belgium,

³Leuven University Hospital, Department of Endocrinology, Leuven, Belgium

Objective

Investigating prevalence of single-nucleotide polymorphisms (SNPs) and their effects on calculated (cFT) and measured FT (mFT).

Material and Methods

Population-based study (999 healthy men, 24–46 years); genotyping was performed using microarray (Illumina®) for SNPs suggested to affect binding affinity and/or concentration of SHBG or total T (rs6258, rs6259, rs1799941, rs12150660, rs727428, rs5934505). All SNPs are in or near the SHBG gene except rs5934505 which is located on the X chromosome. SHBG and total T concentrations were measured by immunoassay and LC-MS/MS, respectively. FT levels were calculated and in a subset (232 individuals) measured directly using LC-MS/MS after equilibrium dialysis. Linkage disequilibrium was determined using the LDlinkR package for RStudio. Linear mixed modelling with a correction for age and BMI was used for comparison of hormonal levels.

Results

Allelic frequencies for the analyzed SNPs ranged from 0.5% to 41.2%. Rs12150660 and rs1799941 were in strong linkage disequilibrium (R² = 0.90) and analyzed together as rs1799941. Compared to wild-type, SHBG levels were significantly lower in heterozygote carriers of rs6258 (-9.71 nmol/L) and higher in carriers of all other SNPs (+3.76 to 9.42 nmol/L) except rs5934505 and rs727428. Total T was higher in rs1799941 homozygotes and rs5934505 carriers (+59.00 and 67.43 ng/dl, respectively). No effects on mFT were found whereas rs6259 homozygotes or rs5934505 carriers (+1.80 and 0.46 ng/dl, respectively) had increased cFT.

Conclusion

We show that some of the analyzed SNPs are relatively prevalent and affected serum levels of total T, SHBG and cFT but not mFT. Effects on SHBG can be attributed to effects on SHBG structure, expression or metabolism. The results did not disprove the use of cFT. However, it can be influenced by SNPs affecting only total T or SHBG levels (rs6259 and rs5934505).

Acknowledgments

This research was funded by the Flemish FWO-TBM program (grant number: T004321N).

PRE-ANALYTICAL EFFECT OF DIFFERENT URINE TRANSFER TUBES ON URINE SEDIMENT RESULTS

N. Debunne¹, J. Delanghe¹, M. Oyaert¹

¹Department of Laboratory Medicine, Ghent University Hospital, Ghent, Belgium, Ghent, Belgium

Introduction

Due to recent technological progress, the reproducibility of urinalysis has been improved. This makes that the reliability of results are not solely dependent on the analytical performance and increases the importance of the pre-analytical requirements. Here, we investigated the effect of different urine collection techniques on urine test strip and sediment results.

Materials and Methods

In total, 150 randomly selected urine samples were subdivided into 3 different collection containers and subsequently transferred into its accompanying collection tube (Sarstedt, MLS and BD). As reference, the original urine sample was manually poured into a blank collection tube. Both flow cytometry (Sysmex UF-5000) and routine chemical dipstick analysis (Sysmex UC-3500) were performed on all samples. To compare the individual testing groups, a repeated measure one-way ANOVA was used for the quantitative data and kappa agreements between the test and the control group were used for the ordinal and nominal data.

Results

We found no statistical differences between the different collection tubes for any semi-quantitative chemical analysis performed on a dipstick. On contrary, the use of vacuum tubes was found to be less suitable for particle analysis on a flow cytometer as its results in a statistically significant decrease of 35–45% for pathological cast (MLS, BD and Sarstedt), 20–26% for hyaline cast and 18–22% for renal epithelial cells (BD and MLS) and a significant increase in erythrocytes of approximately 10–15% (MLS and BD). No statistical differences were found in crystal and leukocyte counts. However, as vacuum tubes, aspiration tubes can also affect the integrity of particles with a significant increase of bacterial cells.

Conclusion

The results of this study indicate that manual urine aspiration collection tubes are preferred over vacuum aspiration systems when fragile particles need to be detected/quantified in human urine.

WELCOME TO THE REAL WORLD IN POCT CRP

J. van Elslande¹, A.-M. van den Abeele¹, C. Verfaillie¹, S. Degandt¹, H. Louagie¹, T. Ghys¹, J. van Acker¹

¹AZ Sint-Lucas, Clinical Laboratory, Groenebriel 1, 9000 Ghent, Belgium., Ghent, Belgium

AZ Sint-Lucas, Clinical Laboratory, Groenebriel 1, 9000 Ghent, Belgium

Background: In children with febrile illness, point-of-care testing (POCT) for C-reactive protein (CRP) can be used to decide on the prescribing of antibiotics, hospital admission (triage) and/or the need for further diagnostic investigations.

Methods: We performed a retrospective analysis of POCT CRP measurements (Afinion AS100, Abbott, IL, USA) in the emergency department and pediatric consultation of the AZ Sint-Lucas Hospital Ghent over a period of 566 days (n=4314). We compared results with central laboratory CRP (Cobas c702, Roche Diagnostics, Rotkreuz, Switzerland), if a serum sample was available within 72 hours after POCT CRP.

Results: 18.3% (n=791) of POCT CRP measurements were performed in adult patients (>=18 years old). There was a clear relationship between the height of the POCT CRP level, and whether a sample was sent to the central laboratory for confirmation. For values <5 mg/L, 6–50 mg/L, 51–100 mg/L and >100 mg/L, confirmation followed within 72 hours, respectively, in 9.4%, 13.4%, 29.1% and 41.6% of cases. Confirmatory serum samples were usually taken within 3 hours (n=776, 59.9%). In this time frame, correlation with the Cobas c702 was good (y = 2.3 + 0.97x, r² = 0.98). Variance increased if POCT CRP values exceeded 100 mg/L (Figure 1).

Conclusions: Correlation between POCT and central laboratory CRP results was excellent in low to medium concentration ranges, which are the most relevant for triage purposes. Despite being intended for pediatric patients, a significant amount of POCT CRP analyses was nonetheless performed in adults (18.3%). Notably, almost 10% of low POCT CRP (<5 mg/L) cases were further investigated with central laboratory testing while confirmation was lacking in almost 60% of high POCT CRP (>100 mg/L) cases.

Figure 1: Residual plot of POCT CRP and Cobas c702 CRP results, measured within 3 hours of each other (n=776). The red dotted line represents no difference.

References

1. Lemiengre MB, Verbakel JY, Colman R, Van Roy K, De Burghgraeve T, Buntinx F, et al. Point-of-care CRP matters: normal CRP levels reduce immediate antibiotic prescribing for acutely ill children in primary care: a cluster randomized controlled trial. Scand J Prim Health Care. 2018 Oct 2;36(4):423–36.

2. Van Hecke O, Raymond M, Lee JJ, Turner P, Goyder CR, Verbakel JY, et al. In-vitro diagnostic point-of-care tests in paediatric ambulatory care: A systematic review and meta-analysis [Internet]. Vol. 15, PLoS ONE. PLoS One; 2020.

3. Goyder C, Tan PS, Verbakel J, Ananthakumar T, Lee JJ, Hayward G, et al. Impact of point-of-care panel tests in ambulatory care: A systematic review and meta-analysis. BMJ Open. 2020 Feb 27;10(2).

4. Verbakel JY, Lee JJ, Goyder C, Tan PS, Ananthakumar T, Turner PJ, et al. Impact of point-of-care C reactive protein in ambulatory care: A systematic review and meta-analysis. Vol. 9, BMJ Open. BMJ Publishing Group; 2019.

5. Brouwer N, van Pelt J. Validation and evaluation of eight commercially available point of care CRP methods. Clin Chim Acta. 2015 Jan 15;439:195–201.

VALIDATION OF AN AUTOMATED METHOD FOR DIRECT QUANTIFICATION OF URINARY AMMONIUM ON ROCHE COBAS C702

A. Ockerman¹, A. Coussee², G. Frans¹

¹Department of Laboratory Medicine, University Hospitals Leuven, Leuven, Belgium,

²Department of Laboratory Medicine, AZ Damiaan, Oostende, Belgium

Objectives

Urinary ammonium (uNH₄⁺) is a valuable tool to determine the cause of metabolic acidosis, especially in renal tubular acidosis. While it is mostly evaluated indirectly by calculating urinary osmolal/anion gaps, several studies have shown that these surrogate measures are not accurate enough and should be replaced by direct uNH₄⁺ measurements. The goal of this study was to validate the plasma NH₄⁺ method on Roche Cobas c702 for direct quantification of uNH₄⁺.

Materials and Methods

uNH₄⁺ was measured using the plasma NH₄⁺ assay on Roche Cobas c702 (NH3L2). Urine was collected in a non-additive tube (Greiner Bio-One, Vacuette® Urine Tube), centrifuged at 3000g for 10min and the supernatant was separated. Samples were aliquoted and stored at −20°C until analysis.

The presence of a matrix effect was evaluated by spiking 3 urine samples of different uNH₄⁺ concentrations with 15 000, 30 000 and 45 000 μmol/L NH₄I (Sigma-Aldrich). Imprecision was determined according to CLSI EP15 with a 15 000 μmol/L NH₄I solution and two urine pools (±25 000 and 50 000 μmol/L uNH₄⁺). Urine samples were diluted 1/100 in CLR water before analysis. Linearity (10–1000 μmol/L for plasma NH₄⁺) was verified for uNH₄⁺ according to CLSI EP06 with serial 1/2 dilutions starting from an undiluted urine sample with ± 7500 µmol/L uNH₄⁺. Desirable quality specifications based on biological variation were used as acceptance criteria (imprecision=12.4%, bias=9.2% and TEA=29.6%).

Results

Urine samples spiked with NH ₄I showed recoveries ranging from 93.2–107.1% (mean±SD = 101.1±4.4%), thereby complying with the bias specification and excluded a matrix effect. Within-lab CVs between 4.3–5.5% were found for the 3 tested solutions/pools and complied with the imprecision goal. The linearity study also confirmed that uNH₄⁺ analysis was linear from 10–1000 μmol/L.

Conclusions

The Roche plasma NH₄⁺ method on Roche Cobas c702 fulfilled all validation criteria and can be used for direct quantification of uNH₄⁺.

FALSELY LOW BETA-HCG RESULTS IN PREGNANT WOMAN: DON’T FORGET THE “HOOK EFFECT”

A. Nizet¹, P. Jeanmart¹, L. Dewalque¹, Q. Bodson¹

¹Department of Laboratory Medicine, Clinique CHC MontLégia, Liège, Belgium

Objectives

Human chorionic gonadotrophin (hCG) is a well-known marker for pregnancy or gestational trophoblastic diseases, daily assayed in most of the laboratory medicine (1). However, the serum βhCG concentration may be sometimes misreported. Indeed, falsely low results can occur and may directly influence patient care management.

Here, we reported the admission in the emergency department of a 27-year-old pregnant woman for uncontrollable vomiting unresponsive to antiemetic for three weeks associated to a five-kilogram weight loss.

Material and methods

Physical examination, blood test and ultrasound scan were performed at admission. The serum βhCG assays were performed using the Siemens hCG test Atellica IM Total hCG (ThCG) (Siemens Healthineers, Erlangen, Germany) on Atellica IM1600 analyser (Siemens Healthineers, Erlangen, Germany). Following the discrepancy between the clinic and the βhCG level, analysis of diluted samples was also performed on Atellica IM 1600 analyser. Then, comparison of native serum βhCG assays was made using Cobas e801 (Roche Diagnostics, Mannheim, Germany) and Alinity I (Abbott Laboratories, IL, USA) analysers.

Results

The physical examination showed an absence of metrorrhagia and no abdominal pain. Ultrasound scan showed a uterus suggestive of molar pregnancy. Laboratory tests indicated a mild inflammatory syndrome associated with hepatic and hyperthyroidism. The βhCG assay was positive, but low for gestational age (948 UI/L). Analysis of diluted sample (dilutions performed to 1:1600) on Atellica IM1600 analyser provided the βhCG concentration at 1,437,304 IU/L corresponding to a βhCG concentration 1400 times greater than the limit of linearity (LoL) of the method. Analysis of undiluted sample on Cobas e801 and Alinity I analysers provided results above the LoL of the assay, requiring dilutions to obtain a quantitative βhCG value.

Conclusion

HCG assay uses a sandwich method involving two antibodies to the beta subunit of hCG molecules. In case of extremely high concentrations of βhCG, which can occur in complete hydatidiform mole, the detection antibodies can be oversaturated, decreasing the detected signal. This phenomenon, while rare, is known as “hook effect” and may delay diagnosis and lead to mismanagement of patients. To improve patient outcomes and prevent misdiagnosis, a good knowledge of the analytical limitations of laboratory tests associated with a complete physical examination remains essential.

References

1. Nodler JL, Kim KH, Alvarez RD. Abnormally low hCG in a complete hydatidiform molar pregnancy: The hook effect. Gynecol Oncol Case Reports. 2011 Dec;1(1):6–7. Available from: https://doi.org/10.1016/j.gynor.2011.10.003.

ALTERED EXTENDED INFLAMMATORY PARAMETERS ON SYSMEX XN-SERIES HEMATOLOGY ANALYZERS IN MALES WITH HUMAN MONKEYPOX

L. Nevejan¹, N. Boeckx², C. van Laer³

¹Department of Laboratory Medicine, University Hospitals Leuven, Leuven, Belgium,

²Department of Laboratory Medicine, University Hospitals Leuven & Department of Oncology, Laboratory of Experimental Hematology, University of Leuven, Leuven, Belgium,

³Department of Laboratory Medicine, University Hospitals Leuven & Department of Cardiovascular Sciences, Centre for Molecular and Vascular Biology, University of Leuven, Leuven, Belgium

Introduction

Starting early May 2022, a new nontravel-related outbreak of human Monkeypox (MPX) was declared as public health emergency by the World Health Organization. Considering the risk of MPX virus transmission in community and health care settings, early identification of potentially infected patients is warranted. Sysmex XN-series hematology analyzers (Sysmex, Kobe, Japan) measuring extended inflammatory parameters in peripheral blood cells via fluorescence signal intensity, size and complexity of their internal structure may be a comprehensive diagnostic tool for this purpose.

Materials and Methods

Adult male patients with PCR confirmed MPX presenting at University Hospitals Leuven between May 1, 2022 and September 15, 2022 with concomitantly determined complete blood count (CBC) were retrospectively included (n=17). Patient control groups included 1) adult males infected with Herpes simplex virus 1 (HSV-1) or Varicella-zoster virus (VZV) (n=19) and 2) adult males in whom a negative PCR for detecting MPXV/HSV-1/VZV was obtained (n=27). Healthy controls included adult males with normal CBC (n=70). Next to conventional CBC parameters, investigated extended inflammatory parameters included neutrophil reactivity index (NEUT-RI), neutrophil granularity index (NEUT-GI), reactive lymphocytes (RE-LYMPH) and antibody secreting B-lymphocytes (AS-LYMPH).

Results

MPX patients (median age [interquartile range]: 33 [27–41] years old) showed significantly increased relative/absolute AS-LYMPH and RE-LYMPH compared to HSV-1/VZV patients (p<0.01), PCR negative patient controls (p<0.001) and healthy controls (p<0.001). No significant differences were noticed in AS-LYMPH or RE-LYMPH between HSV-1/VZV patients and PCR negative patient controls. Neither NEUT-RI nor NEUT-GI were significantly altered between MPX patients and patient controls groups. The diagnostic accuracy in distinguishing MPX from non-MPX patients was excellent for AS-LYMPH and RE-LYMPH, with area under the curve (AUC) 0.922–0.951 (p<0.001).

Conclusion

Among extended inflammatory parameters obtained via CBC using Sysmex XN-series, AS-LYMPH and RE-LYMPH could be valuable tools in distinguishing MPX from non-MPX patients.

WHEN ORIGINALLY EPITHELIAL CELLS END UP SYNTHETIZING IMMUNOGLOBULINS

M. Lauwers¹, V. Daubie¹, J. Farhat¹, L. Vanoverschelde¹, L. Rocq², A.-L. Trépant², N. Yin³, D. Martiny³, L. Dewispelaere⁴, P. Heimann⁴, H. Farhat¹

¹Department of Hematology, Laboratoire Hospitalier Universitaire de Bruxelles – Universitair Laboratorium Brussel (LHUB-ULB), Brussels, Belgium,

²Department of Pathology, Erasme University Hospital, Brussels, Belgium,

³Department of Microbiology, Laboratoire Hospitalier Universitaire de Bruxelles – Universitair Laboratorium Brussel (LHUB-ULB), Brussels, Belgium,

⁴Department of Molecular Hemato-Oncology, Laboratoire Hospitalier Universitaire de Bruxelles – Universitair Laboratorium Brussel (LHUB-ULB), Brussels, Belgium

A 72-year-old woman presented to the emergency department for severe and worsening dyspnea. The patient had a 10-year history of immunosuppressive therapy following a liver transplant for post-hepatitis C cirrhosis complicated by hepatocarcinoma. She underwent surgical resection for urothelial and lung carcinoma in 2007 and 2021. Clinical and radiological investigations revealed right pleural effusion. A relapsed lung cancer was clinically suspected given the exudative nature of the liquid. Cytological analysis revealed a pleiocytosis consisting of numerous medium to large non cohesive cells showing varying nucleocytoplasmic ratios and basophilic cytoplasm. Some cells displayed plasmablastic features. Flow cytometric analyses showed an absence of B lymphoid cells but revealed a mixture of T lymphoid cells (normal phenotype) and a predominant population of cells expressing CD38, CD45 (dim), CD56, CD117, CD138 and cytoplasmic kappa light chains. Immunophenotypic and morphological features were highly suggestive of malignant plasma cell proliferation. Polymerase chain reaction (PCR) testing performed on the pleural fluid was negative for human herpesvirus 8 (HHV8) while it was positive for Epstein-Barr virus (EBV). Concurrently, immunohistochemical staining of the pleural fluid cell block highlighted a carcinomatous and neuroendocrine profile of the cells with expression of cytokeratin, chromogranin, INSM1 and synaptophysin markers. Hematopoietic B cell lymphoma-2 (BCL2), CD56, CD138 and kappa light chains markers were also positive. Both flow cytometric and immunohistochemical features were in favor of a single tumor component. Finally, next-generation sequencing analysis revealed that the tumor cells carried the same tumor protein 53 (TP53) NM_000546:c.880G>T mutation (p.(E294*)) as the original lung carcinoma, reflecting the relationship between the two tumors.

Based on these findings, the final diagnosis was narrowed down to a proliferation of tumor cells co-expressing both neuroendocrine and plasma markers. As suggested by Balanis et al.¹, this case illustrates a process of transdifferentiation of an initially epithelial carcinoma as reflected by the acquired neuroendocrine and plasma cell characters of tumor cells leading to their unexpected immunoglobulin production. This process is reported for the first time in the absence of any targeted therapy but within a context of a post liver transplant immunosuppressive therapy associated with a latent EBV infection.

¹Balanis NG, Sheu KM, Esedebe FN, et al. Pan-cancer Convergence to a Small-Cell Neuroendocrine Phenotype that Shares Susceptibilities with Hematological Malignancies. Cancer Cell. 2019;36(1):17–34.

COMPARISON OF IMPEDANCE AND FLUORESCENCE IN THE DETERMINATION OF PLATELET-POOR PLASMA

L. Proost¹, P. De Kesel¹, M. Hofmans¹, S. Lambrecht¹

¹Ghent University Hospital, Gent, Belgium

Introduction: Coagulation tests are influenced by pre-analytical errors. As platelets contain phospholipids which can interfere with coagulation tests, preparation of platelet-poor plasma is crucial for correct interpretation of results. In whole blood, different methods for platelet enumeration on automated haematology systems can be calibrated by the RBC/platelet ratio method recommended by ISLH. In plasma, no reference method is available for assigning platelet counts. This makes it difficult to determine which method is preferable to decide if plasma is platelet-poor.

Methods: We developed a novel method for platelet counting in plasma based on spectral flow cytometry (Cytek Northern Lights), using antigen/fluoro CD61/41 combinations and the volumetric unit of the flow cytometer. 20 EDTA whole blood samples with a broad range of platelet count were analysed both with the RBC/platelet method and the new developed method. Agreement of values were determined with Bland Altman analysis and Pearson correlation. In a second stage, 35 plasma samples were analysed after double centrifugation at 1999g for 8 minutes and 2230g for 15 minutes to obtain platelet-poor plasma. For comparison, samples were analysed with the spectral flow cytometry method and PLT-I (impedance)/ PLT-F (fluorescence) on Sysmex XN10/XN20.

Results: Correlation between the novel spectral flow cytometry based method and the reference method yielded an r value of 0.99. Bland-Altman-analysis showed that the limits of agreement did not exceed the maximum allowed difference between methods of 15% (Quality assurance in Haematology, WHO/LAB/98.4). The two methods are therefore considered to be in agreement and can be used interchangeably. Correlation between PLT-F and the newly developed method yielded an r-value of 0.99, while for PLT-I a correlation of only 0.96 was obtained. Additionally, Bland-Altman resulted in a significant difference for both PLT-F/PLT-I and the volumetric flow cytometric method. However, this difference was more pronounced for PLT-I (mean difference of 40%) compared to PLT-F (mean difference of 8.0%). 97% of the platelet count (PLT-I) resulted in a positive bias of > 15%, while this was only 3% for PLT-F.

Conclusions: The discordance between PLT-F/PLT-I illustrated the necessity of a reliable method, as developed in this study, to evaluate different methods for platelet enumeration in plasma. Platelet counts performed with PLT-I can result in underestimation of the true count which could lead to misinterpretation of coagulation tests.

HEMATOFLOW REVISITED; INTEGRATION OF MULTIPARAMETRIC CYTOMETRY AND HEMATOLOGY INTO A SINGLE TEAM IN A NON-UNIVERSITY GENERAL HOSPITAL LABORATORY

O. P. Pradier¹, C. T. Tilmant¹, N. D. Daoud¹

¹Department of Laboratories; CHIREC, Brussels, Belgium

Introduction: Four years’ experience, replacing digital microscopy by multiparametric cytometry (FCM) into routine hematology (Hematoflow) with a team of 9 multidisciplinary technologists (7.6 FTEs) performing hematology, coagulation, immuno-hematology / blood bank and cytometry.

Method: Two sites, 810 beds, general hospitals. 2 DxH900/10C-Navios (Beckman Coulter). Body fluids, platelets and RBCs are checked by conventional microscopy. WBC flags and counts anomalies are controlled by cytometry using 8 colors routine panel named CellDiffEvo (CDE), also the back bone of the screening cytometry (CDE, HIV, SuperLymphoDiff (SLD); inflammation/immunology panel and MyeloDiff (MD), a one tube bone marrow analysis). CDE = (CD64-FITC; HLA-DR-ECD; C16-PC5; CD19-PC7; CD34-AA700; CD38-AA750; CD3-PB;CD45-KO), SLD = CDE + (CD45RA-FITC; CD27-PE; CD138-PE; CD4-PC7; TCRgd-PC7; CD14-APC; CD8-AA700); MDF = CDE + (CD71-FITC; CD27-PE; CD138-PE; CD117PC5.5; CD4-PC7, CD13-APC, CD8-AA700). in addition to screening there are two B lymphocyte panels (kl and LNH).

Results: Hematoflow makes it possible to have 7d/w, a differential count, counting 20,000 cells/sample (120 cells for digital microscopy) and allowing 11 populations diff (5 diff + T,B,NK, blastes (circulating hematoipietic or leukemic cells; CD45low/CD34±/CD38±/HLA-DR+, as low as 1 cell/µl), Immature gran (SS/CD16dim), variant lymphocytes (CD3+/CD38+/HLA-DR+). It also makes it possible to detect sepsis (CD64 on neutrophils), monocytes immunodeficiency (loss of HLA-DR), unconventional monocytes (CD64dim/CD16+), plasmablastes (CD19,CD38hi).

Only 6 panels allow the formation of a highly versatile team performing the control, preparation and acquisition of data and positioning of the windows in Kaluza©. Validation is the responsibility of biologists.

SLD panel allows analyzing the CD4, CD8, gamma/delta, DP (CD4hiCD8low) maturation (from naïve to TemRA cells) and the B maturation, suggesting the presence of a chronic activation of immunity. Increased circulating plasmablastes and HLA-DR expression en CD4 T cells evoke acute immune activation. T cell clonality is suggested as a maturation pathway abnormality. SLD reports actually 21 populations.

Conclusion: Hematoflow is more efficient than microscopy for WBC analysis and allows integrating the diagnostic cytometry of a non-academic hospital, directly into a team of versatile technologists.

IGG ANTI-SPIKE ANTIBODIES AND SURROGATE NEUTRALIZING ANTIBODY LEVELS DECLINE FASTER 3 TO 10 MONTHS AFTER BNT162B2 VACCINATION THAN AFTER SARS-COV-2 INFECTION IN HEALTHCARE WORKERS

B. Decru¹, J. van Elslande², S. Steels², G. van Pottelbergh³, L. Godderis⁴, B. van Holm⁵, X. Bossuyt¹, J. van Weyenbergh⁵, P. Maes⁵, P. Vermeersch⁶

¹University Hospitals Leuven, Clinical Department of Laboratory Medicine and National Reference Center for Respiratory Pathogens, Leuven, Belgium, ²University Hospitals Leuven, Clinical Department of Laboratory Medicine and National Reference Center for Respiratory Pathogens, Leuven, Belgium, Leuven, Belgium, ³Academic Centre of General Practice, Department of Public Health and Primary Care, KU Leuven, Leuven, Belgium, ⁴Academic Centre of General Practice Environment and Health, Department of Public Health and Primary Care, KU Leuven. Group IDEWE, External Service for Prevention and Protection at Work, Leuven, Belgium, ⁵Laboratory of Clinical and Epidemiological Virology, Department of Microbiology, Immunology and Transplantation, Rega Institute, KU Leuven, Leuven, Belgium, ⁶University Hospitals Leuven, Clinical Department of Laboratory Medicine and National Reference Center for Respiratory Pathogens Department of Cardiovascular Sciences, KU Leuven, Leuven, Belgium

Background: IgG anti-spike (S) antibodies arise after SARS-CoV-2 infection as well as vaccination. Levels of IgG anti-S are linked to neutralizing antibody titers and protection against (re)infection.

Methods: We measured IgG anti-S and surrogate neutralizing antibody kinetics against Wild Type (WT) and 4 Variants of Concern (VOC) in health care workers (HCW) 3 and 10 months after natural infection (“infection”, n=83) or vaccination (2 doses of BNT162b2) with (“hybrid immunity”, n=17) or without prior SARS-CoV-2 infection (“vaccination”, n=97).

Results: The humoral immune response in the “vaccination” cohort was higher at 3 months, but lower at 10 months, compared to the “infection” cohort due to a faster decline. The “hybrid immunity” cohort had the highest antibody levels at 3 and 10 months with a slower decline compared to the “vaccination” cohort (figure 1). Surrogate neutralizing antibody levels (expressed as %inhibition of ACE-2 binding) showed a linear relation with log10 of IgG anti-S against WT and four VOC. IgG anti-S corresponding to 90% inhibition ranged from 489 BAU/mL for WT to 1756 BAU/mL for Beta variant. Broad pseudoneutralization predicted live virus neutralization of Omicron BA.1 in 20 randomly selected high titer samples.

Conclusions: Hybrid immunity resulted in the strongest humoral immune response. Antibodies induced by natural infection decreased more slowly than after vaccination, resulting in higher antibody levels at 10 months compared to vaccinated HCW without prior infection. There was a linear relationship between surrogate neutralizing activity and log10 IgG anti-S for WT and 4 VOC, although some VOC showed reduced sensitivity to pseudoneutralization.

ANALYTICAL PERFORMANCE OF THE VIASURE MONKEYPOX VIRUS REAL-TIME PCR DETECTION ASSAY

J. Maes¹, L. Vanstraelen¹, B. van Meensel¹, R. Cartuyvels¹, L. Waumans¹, K. Magerman¹

¹Department of Clinical Microbiology, Jessa Hospital, Hasselt, Belgium

Objective

To evaluate the analytical performance of the Viasure (CerTest Biotec, Spain) Monkeypox virus real-time PCR assay for the qualitative detection of Monkeypox virus DNA in skin, nasopharyngeal and rectal swabs. Accuracy, reproducibility and the possibility for sample collection in lysis buffer were evaluated.

Material and methods

The assay targets the F3L and G2R Monkeypox virus genes, as recommended by the WHO (1). DNA extraction was performed on Maxwell (Promega, USA). Amplification and detection were performed on Quantstudio7 (ThermoFisher, USA). Patient samples (collected in UTM medium) were initially tested by an in-house real-time PCR in the clinical laboratory of University Hospitals Leuven (UZ Leuven). This method was used as reference method to evaluate the results. Six positive skin, 9 low level positive nasopharyngeal, 5 positive rectal and 5 negative swabs were tested. A positive and low level positive sample were tested 3 times per day on 3 different days to evaluate the reproducibility. A dilution series in UTM medium and lysis buffer was tested, and the lowest concentration was tested again on the next three days.

Results

23/25 (92%) of patient samples were concordant with the analysis by UZ Leuven (Ct-values were similar). Two low level positive samples (a skin and nasopharyngeal swab, Ct-value with reference method around 37 and 36 respectively) tested negative. Reanalysis of these samples yielded a positive result. Reproducibility was 100%. Analysis in UTM medium and in lysis buffer yielded the same results. Also, storage conditions up to 10 days in the fridge could be guaranteed without affecting the sample result.

Conclusion

The analytical performance of the Viasure Monkeypox virus real-time PCR assay for the qualitative detection of Monkeypox virus DNA in skin, nasopharyngeal and rectal swabs was excellent, both in UTM medium and in lysis buffer.

References

1. Jiang Z, Sun J, Zhang L, Yan S, Li D, Zhang C, et al. Laboratory diagnostics for monkeypox: An overview of sensitivities from various published tests. Travel Med Infect Dis [Internet]. 2022 Sep 1 [cited 2022 Sep 13];49:102425.

PHYSIOPATHOLOGY OF MALE PUBERTY

A.-S. Parent ¹

¹Department of Pediatrics, CHU Liège, Liège, Belgium

Puberty results from the re-awakening of a complex neuroendocrine machinery. A subset of neurons in the anterior hypothalamus secrete gonadotropin-releasing hormone (GnRH) which, through its pattern of release, controls all aspects of reproductive function throughout life. The secretory activity of GnRH neurons depends on trans-synaptic and glial inputs provided by different neurotransmitters and cell-cell signaling molecules, derived from either neuronal or glial cells functionally connected to GnRH neurons. The recent years have brought fascinating data regarding the mechanisms leading to the reactivation of the GnRH network.

Whilst the timing of pubertal onset is a highly heritable trait with up to 80% of individual variation being under genetic regulation, it is also particularly sensitive to environmental conditions. Reports about the advancing onset of puberty in several countries have led to the hypothesis that the increasing burden of endocrine disrupting chemicals could be an explanation. We have shown that, in fact, pubertal timing currently shows complex changes since advancement of some manifestations of puberty (e.g. breast development or testicular enlargement) and no change or delay of others (e.g. menarche, adult testicular volume) can be observed.

Male pubertal onset takes place between age 9 and 14 in boys and is marked by testicular enlargement. Subsequent hallmarks of puberty in boys include deepening of the voice and a growth spurt. While precocious puberty is rare in boys, male delayed puberty is common, affecting up to 3% of the population. Management of patients with pubertal delay depends on the underlying cause. The main differential diagnoses include constitutional delay of growth and puberty, idiopathic hypogonadotropic hypogonadism and hypergonadotropic hypogonadism. The presentation will review the diagnosis, challenges and treatment of delayed puberty in boys.

INFERTILITY: SPERM ANALYSIS IN MEN

I. Goovaerts ^1,2

¹Centre for Reproductive Medicine, Antwerp University Hospital, Antwerp, Belgium, ²Faculty of Medicine and Health Sciences, University of Antwerp, Antwerp, Belgium

The first step in the diagnosis of male subfertility or infertility is routine semen testing. In 2021, WHO launched a new edition of the Manual for the Laboratory Examination and Processing of Human Semen, the reference document for laboratories. At the same time, ISO Standard 23162:2021 was published, which generally describes the same procedures for the basic analysis of semen samples. The 2021 manual ensures quality and global standardization in the basic examination of ejaculates, is easier to follow and reduces unnecessary workload, compared with the previous manual published in 2010. One preferred method is chosen for morphology and vitality. Old tests, such as sperm mucus interaction, have been abolished. Computer-assisted sperm analysis (CASA) moved from ‘Optional’ in 2010 to the ‘Advanced testing’ section in 2021, and thus is not recommended in the routine. Since conventional parameters do not fully reflect fertilization potential, functional tests such as DNA fragmentation, aneuploidy and assessment of interleukins are added to the ’Advanced Examination’ section. The most intriguing change, however, is the section ‘Interpretation of Semen Examination Results’, which has become a topic in the ‘Appendices’. Updated global distribution data of conventional semen parameters of fertile men are published, but interpretation is beyond the scope of the 2021 Manual. Since fertility is a couple’s problem, the female factor should not be underestimated, but with the introduction of intracytoplasmic sperm injection (ICSI), almost no attention has been paid to restoring male sperm production, since a few spermatozoa are sufficient to obtain fertilization. In recent years, treatment of men has been receiving more attention again. Not only can less invasive reproductive techniques be used if spermatogenesis is stimulated, but the man’s overall health could also improve, as there is growing evidence that male infertility may be a harbinger of his future health and a predictor of hospitalization and mortality.

HYPOGONADISM

D. Vanderschueren ¹

¹Department of Endocrinology, UZ Leuven, Leuven, Belgium

During the presentation the most important take home messages of the recent guidelines of the European academy of andrology on the investigation, treatment and monitoring will be discussed (reference).The most controversial issue with respect to hypogonadism is the diagnosis, treatment and monitoring on functional hypogonadism. Therefore, clinicians should be able to differentiate organic from functional hypogonadism; The diagnosis of functional hypogonadism should be based on both the presence of clinical symptoms supported by repeatedly low morning fasting serum total testosterone (T) measured with a well-validated assay, after exclusion of organic causes of hypogonadism. Lifestyle changes and weight reduction should be the first approach in all overweight and obese men. Whenever possible, withdrawal/modification of drugs potentially interfering with T production should be advised. Testosterone replacement therapy (TRT) is contraindicated in men with untreated prostate or breast cancer, as well as severe heart failure. Severe low urinary tract symptoms and high haematocrit represent relative contraindications for TRT. Prostate-specific antigen and digital rectal examination of the prostate should be undertaken in men >40 years of age before initiating TRT to exclude occult prostate cancer. Transdermal T should be preferred for initiation of TRT, whereas gonadotrophin therapy is only recommended when fertility is desired in men with secondary hypogonadism. TRT is able to improve sexual function in hypogonadal men. Other potential positive outcomes of TRT remain uncertain and controversial. TRT can reliably improve global sexual function in men with hypogonadism in the short term. Long-term clinical benefits, and safety of TRT in functional hypogonadism, remain to be fully documented. Clinicians should therefore explicitly discuss the uncertainties and benefits of TRT and engage them in shared management decision-making.

Reference: European Academy of Andrology (EAA) guidelines on investigation, treatment and monitoring of functional hypogonadism in males: Endorsing organization: European Society of Endocrinology. Giovanni Corona, Dimitrios G Goulis, Ilpo Huhtaniemi, Michael Zitzmann, Jorma Toppari, Gianni Forti, Dirk Vanderschueren and Frederick C Wu Andrology. 2020 Sep;8(5):970–987.

HORMONE TREATMENT IN TRANSGENDER PERSONS, INTERPRETING LABORATORY TESTS

G. T’Sjoen ¹

¹Department of Endocrinology, UZ Gent, Gent, Belgium

As the number of transgender and gender diverse people seeking gender-affirming hormone therapy rises; endocrinologists are increasingly asked to assist with interpretation of laboratory tests as many results may come back flagged. There are many common laboratory tests such as creatinine, hemoglobin, iron studies, troponin, PSA (Prostate-Specific Antigen) and hCG (Human chorionic gonadotropin) that are affected by sex steroids or body size. The influence of exogenous gender-affirming hormone therapy on body composition also determines interpretation of laboratory tests with sex-specific differences.

The exponential rise in transgender people attending so called the gender clinics represent an opportunity for laboratories to improve quality of care and experiences in healthcare for trans people. Costs are often cited as a barrier, to implement this, but we can also not be blind to negative attitudes towards transgender people among (some) healthcare staff as a reason. We should be aware that this bias implicit or explicit can affect healthcare delivery.

Finding a solution to this problem is not that difficult. As trans specific reference ranges do not exist, dual reporting – reporting both male and female reference ranges for all tests with sex specific ranges – is a pragmatic approach. This leaves greater flexibility to clinicians and patients when interpreting test results, and this is particularly valuable for people in early stages of hormone treatment (3 months). In fact, dual reporting of reference ranges should be the norm for all tests with sex specific ranges regardless of people’s gender experience or body. This approach removes the need for clinicians to state the sex, or gender identity, of patients on requests. Including two reference ranges on all test reports will not disadvantage cisgender people, but there is major benefit for trans people.

TESTOSTERONE AND OSTEOPOROSIS

J.-M. Kaufman ¹

¹Department of Endocrinology, UZ Gent, Gent, Belgium

Skeletal effects of testosterone are exerted in part via the androgen receptor and for a substantial part via the estrogen receptors following aromatization of testosterone to estradiol. Androgen receptor-mediated effects of testosterone, and its 5α-reduced active metabolite dihydrotestosterone, can be exerted both directly on the bone cells and indirectly through effects on striated muscles and mechanical loading on the bones. Sex steroids, both testosterone and estradiol, play an important role in bone maturation and acquisition of peak bone mass during growth as well as in the regulation of bone homeostasis and preservation of skeletal integrity in adult life. Hypogonadism is a well-recognized secondary cause of osteoporosis in men. Acquired profound hypogonadism such as in men receiving androgen deprivation therapy for prostate cancer, induces high bone turnover, accelerated bone loss and increased fracture risk. The much more progressive and only modest age-related decline of serum testosterone and its biologically active serum free fraction has been found to be associated with some skeletal changes. However, the main effects of sex steroids to help preserve skeletal health in ageing men by limiting bone turnover, bone loss, deterioration of trabecular microarchitecture, and cortical porosity appears to be mediated by estradiol acting on the estrogen receptor-α. Low serum estradiol is associated with increased fracture risk with indications for existence of a critical threshold level. Notwithstanding these found associations, in a large US-based prospective study, measurement of serum testosterone and estradiol did not contribute significantly to prediction of fracture risk above the predictive power of bone densitometry by DXA and clinical risk factors. Presently, the role of testosterone measurements in the work-up of osteoporosis in men is limited to the exclusion of hypogonadism. In postmenopausal women aromatization of androgens from adrenal origin, and in the first years after menopause also originating in part from the ovary, plays an important role as the only source of circulating estrogen. It has indeed been shown that lower postmenopausal estrogen levels adversely affect bone health and high endogenous testosterone levels have also been associated with decreased fracture risk in postmenopausal women. Potential meaningful clinical implementation of testosterone and estradiol measurements in the context of bone health is presently still hampered by analytical issues and by ill-defined reference ranges for postmenopausal women.

CLINICAL AND LABORATORY WORK-UP FOR TESTICULAR TUMORS

B. Tombal ¹

¹Service d’Urologie, Cliniques universitaires Saint Luc, Brussels, Belgium

Testicular cancer is considered a rare cancer, with 440 new cases in 2020 in Belgium. It mostly affects young males between 14 and 40 years, the age-standardized incidence being as 14.6 new cases per 100,000 person years in 2020. It has remained mostly stable over the last 10 years. The 5-year relative survival proportion is excellent 98.5 % for the period 2016–2020. At diagnosis, 1–2% of cases are bilateral, and the predominant histology is germ cell tumours (GCT) (90–95% of cases). The peak incidence is in the third decade of life for non-seminoma testis and mixed GCT patients, and in the fourth decade for seminoma testis patients.

Most GCT present as a palpable testicular mass and the primary treatment is orchidectomy, i.e. a surgical removal of the testis and its annexes via inguinal routes. Upon examination by a pathologist, GCTs are clinically and histologically subdivided into seminomas (SGCT) and non-seminomas (NSGCT), the later encompassing somatic and extra-embryonal elements of embryonal carcinoma, yolk sac, choriocarcinoma and teratoma. This subclassification will drive treatment recommendations.

Serum tumour markers alphafetoprotein (aFP), beta subunit of human Chorionic Gonadotropin (β-hCG) and LDH should be determined before and after orchidectomy as they support the diagnosis, may be indicative of GCT histology and are used for disease staging and risk stratification), as well as to monitor treatment response and detect disease relapse. Serum levels of aFP, β-hCG and LDH following orchidectomy provide staging and prognostic information [61]. As the serum half-life of aFP and β-hCG are five to seven days and one to three days respectively, it may take several weeks until normalization occurs. The persistence of, or increase in, serum tumour marker elevation following orchidectomy indicate the likely presence of metastatic disease. Whilst normalization of marker levels after orchidectomy is a favorable indicator, it does not exclude the possibility of metastatic disease. With metastatic TC, risk stratification is based on serum tumour marker levels immediately before initiation of systemic treatment. Before chemotherapy aFP and LDH levels may act as prognostic factors for OS in non-seminoma intermediate risk groups.