9th Annual Meeting of the Austrian Society for Laboratory Medicine and Clinical Chemistry (ÖGLMKC)

Increased kynurenine indicates a fatal course of COVID-9

P Curcic, M Herrmann, A Meinitzer, S Pailer, M Holter, F Prüller, H Mangge

Medical University of Graz, Clinical Institute of Medical and Chemical Laboratory Diagnostics (CIMCL), Graz, Austria

Background: An inefficient immune response accompanied by an overwhelming inflammatory reaction involves in severe courses of COVID-19. Kynurenine (KYN) has important immune-modulatory functions and may contribute to a failure in controlling SARS-CoV-2. The present study aims to explore biomarkers that hint a fatal outcome of COVID-19 early on. Methods: We established a cohort of 148 hospitalized COVID-19 patients. Thirty-one patients died due to a severe COVID-19 course, and 117 recovered within 90 days. We built a biobank by collecting left-over material from these patients whenever blood arrived at the central laboratory of our University hospital for analysis of routine markers. The scientific laboratory analysis comprised KYN, Tryptophan (TRP), KYN/TRP ratio, ferritin, interleukin-6 (IL-6), C-reactive protein (CRP), creatinine, N-terminal pro-natriuretic peptide (NTproBNP), troponin T (TnT), fibrinogen, D-Dimer, prothrombin time (PT), activated partial thromboplastin time (aPTT), antithrombin (AT), protein C, protein S, factor XIII, lupus aPTT, angiotensin-2, vitamin D metabolites, and telomeres in all COVID-19 patients. Basic clinical characteristics and anteceding diseases including cardiovascular, oncologic, renal, hypertension, pulmonary, metabolic (diabetes, obesity) were recorded in a database together with the laboratory data. Results: At the time of diagnosis of SARS-CoV-2 infection those patients who deceased within 90 days afterwards due to COVID-19, had a significantly higher age, higher KYN, KYN/TRP ratio, ferritin, creatinine, and NTproBNP values than SARS-CoV-2 patients who survived COVID-19 along the same time span. In a Kaplan–Meier analysis the variables age, KYN, ferritin, D-Dimer, TnT, NTproBNP, and creatinine showed a significant influence on survival time. Gender, however, showed no influence. In a combined Cox regression analysis KYN had the highest hazard ratio (1.209, CI: 1.068–1.368) followed by age (1.036, CI: 1.036–1.101). In a ROC analysis, KYN values above the cut off limit of 4.82 nmol/L (as specified by Youden index) had a sensitivity of 82 % (95 % CI: 66–95 %) and a specificity of 72 % (95 % CI: 65–82 %) to predict COVID-19 related death within 90 days observation time.

Conclusions: Kynurenine predicts an increased risk of mortality in SARS-CoV-2 infected people already at the time of the first positive SARS-CoV-2 verification.

Reference

Mangge H, Herrmann M, Meinitzer A, Pailer S, Curcic P, Sloup Z, Holter M, Prüller F. Antioxidants 2021; 10:1960–67. PMID: 34943063

Dramatic decrease of vitamin K2 subtype menaquinone-7 in COVID-9 patients

P Curcic, M Herrmann, A Meinitzer, S Pailer, M Holter*, F Prüller, H Mangge

Medical University of Graz, Clinical Institute of Medical and Chemical Laboratory Diagnostics (CIMCL) and *Institute of Medical Computer Sciences, Statistics and Documentation, Graz, Austria

Background: Vitamin K (VK) is a fat-soluble compound with a common chemical structure, a 2-methyl-1,4-naphthoquinone ring, and a variable aliphatic side-chain. VK is involved in the synthesis of blood-clotting proteins, bone stability, anti-oxidative, and immune inflammatory-modulatory functions. Vitamin K also activates protein S, which owns anti-oxidant and anti-inflammatory functions. The fact that cytokine overproduction, oxidative stress, and disturbed microcirculation by thrombogenicity play a central role in severe COVID-19 prompted us to analyze this vitamin. Methods: We analysed by a validated liquid-chromatography tandem mass-spectrometry method serum vitamin K1, MK4, MK7, and VK epoxide levels in 104 healthy controls, 77 patients with non-COVID-19 pneumonia, and 135 hospitalized COVID-19 patients with potentially fatal outcomes admitted to our University Hospital between April and November 2020. We included the quotient between VK and triglyceride (TG, nmol/mmol/L) values in the analyses with respect to the TG transporter function for all VK subtypes. Additionally, we assessed anthropometric, routine laboratory, and clinical data from the laboratory and hospital information systems. Results: The COVID-19 patients had significantly lower MK7 levels than non-COVID-19 pneumonia patients and healthy controls. COVID-19 and non-COVID-19 pneumonia patients had significantly lower vitamin K1 and significantly, higher MK4 compared to healthy controls, but did not differ significantly from each other. Between COVID-19 non-survivors (n=30) and survivors (n=105) no significant differences were seen in all vitamin K subtypes, despite the fact that non-survivors had higher peak concentrations of IL-6, CRP, d-dimer, and higher oxygen needs, respectively. Conclusions: The present data identified significantly decreased vitamin K1, K2 (MK7), and increased MK4 levels in patients with COVID-19 compared to healthy controls. Vitamin K2 (MK7) was lowest in COVID-19 patients irrespective of potentially fatal courses, indicating consumption of this VK subtype by COVID-19 immanent effects, most probably inflammatory and oxidative stress factors.

Reference

Mangge H, Prueller F, Dawczynski C, Curcic P, Sloup Z, Holter M, Herrmann M, Meinitzer A. Antioxidants 2022; 11:1235–45. PMID: 35883726

Lipid profiles and outcome in diabetic and obese patients with COVID-19

S Mink ¹ , A Leiherer ^1,2 , H Drexel ² , P Fraunberger ¹

¹ Central Medical Laboratories, Feldkirch, Austria

² VIVIT Institute, Academic Teaching Hospital Feldkirch, Feldkirch, Austria

Background: Diabetic and obese patients are among the most severely affected subgroups in COVID-19 [1]. Several mechanisms have been suggested to account for worse outcomes in these patients including hypercoagulation, increased baseline inflammation, reduced respiratory function and a higher prevalence of comorbidities. In addition, COVID-19 has been associated with increased risks of diabetes and dyslipidaemia. At the same time, certain lipoproteins are thought to play a significant role in toxin clearance in severe infections [2]. However, there is of yet no data on lipid levels on hospital admission of diabetic and obese patients with COVID-19. Methods: This prospective multicenter cohort study included 1152 COVID-19 patients from five hospitals. Patients were stratified by diabetes, obesity and vaccination status. Anti-SARS-CoV2-spike antibodies and lipid levels were measured on hospital admission. Pre-specified endpoints were in-hospital mortality and oxygen administration. Findings: 395 patients were obese or had a history of diabetes. Mortality risk increased by 1.6 with decreasing LDL-C in diabetic and obese patients (aOR 1.57, 95 % CI 1.06–2.33, p=0.025). Risk of oxygen administration increased by 1.6 with decreasing LDL-C, by 1.5 with decreasing HDL-C and by 1.4 with decreasing total cholesterol levels (aOR 1.61, 95 % CI 1.25–2.08, p<0.001; aOR 1.51, 1.16–1.97, p=0.002; aOR 1.43, 1.13–1.80, p=0.003). Interpretation: LDL-C on hospital admission is inversely associated with in-hospital mortality in diabetic and obese patients with COVID-19. Patients with lower levels of LDL-C, HDL-C and total cholesterol have increased risks of requiring oxygen.

References

1. Steenblock C, Schwarz PEH, Ludwig B, et al. COVID-19 and metabolic disease: mechanisms and clinical management. Lancet Diabetes Endocrinol 2021; 9(11): 786–98 [PMID: 34619105]

2. Tanaka S, Tymowski C de, Assadi M, et al. Lipoprotein concentrations over time in the intensive care unit COVID-19 patients: Results from the ApoCOVID study. PloS one 2020; 15(9): e0239573 [PMID: 32970772]

Comparison between a laboratory and a point-of-care-method for measuring C-reactive protein

B Simon, H Rubey

LK Mistelbach-Gänserndorf, Department of medico-chemical laboratory diagnostics, Mistelbach, Austria

C-reactive protein (CRP) is a laboratory value used routinely to diagnose inflammation. High CRP levels indicate a possibly dangerous inflammation requiring swift treatment. Point-of-care tests (POCT) are tests that are performed near the patient, thus delivering results faster. This study aims to investigate whether CRP values derived from a POCT (Lumira DX) are comparable to those determined using classical lab tests (Roche COBAS). To this end, 103 plasma samples stored in Lithium-Heparin tubes were chosen randomly from routine samples, and the samples‘ CRP levels were tested with both test systems. The resulting values were compared using Passing–Bablok–Regression and Bland–Altman–Plots. The time to result of both test system was measured as well. 77 samples were determined to yield CRP levels in the linear range of both test systems and were included in the comparison. The linear regression analysis resulted in a correlation coefficient of r=0.96. The Passing–Bablok regression analysis also showed comparability between the two systems (slope 1,059 (CI 0.99–1.11); intercept: −0.11 (CI −0.40–0.23)). The Bland–Altman–Plot resulted in a minimal, clinically not relevant bias of −2.4 mg/L. Four data points were outside the calculated limits, which number is within the acceptable number of deviations (which was calculated to be six). Average time to result was about 20 min (44.5 %) shorter when using the POCT. In summary, our comparison showed that the values determined by the POCT were comparable to the traditional lab test while providing results faster, therefore potentially improving patient care.

Differences of fine speckled patterns on HEp-2 cells

W Klotz, M Herold

Medical University of Innsbruck, Department of Internal Medicine II, Rheumatology Laboratory, Innsbruck, Austria

The international consensus on ANA patterns (ICAP) describes 30 patterns categorized as AC (anticellular) numbers from 1 to 29 and AC-0 for samples without antibody binding to any cellular structure (www.anapatterns.org). The fine speckled nuclear pattern AC-4 is caused by different autoantibodies producing slightly variable patterns. The most common AC-4 pattern caused by antibodies against Ro60 is described as bright shining myriad nuclear speckles and can be differed from other AC-4 patterns. AC-4a was recommended for the anti-Ro60 pattern, AC-4b for all the other fine speckled patterns (1). Among others AC-4b is produced by myositis-specific or myositis-associated antibodies (MAA) (2). We looked for ANA results reported in 2022 in which specification into AC-4a and AC-4b was done in daily routine and certified by follow up testing.

Out of 7945 ANA tests on HEp-2 cells 296 samples had an isolated AC-4 pattern (176 AC-4, 82 AC-4a, 38 AC-4b). All other AC-4 positive samples were reported as mixed patterns.

In 82 samples showing an AC-4a pattern 36 were positive for SSA antibodies (2 anti-SSA, 17 anti-SSA and TRIM21, 17 anti-SSA and TRIM21 and anti-SSB), 2 samples were positive for anti TRIM21 and 8 samples were anti-ENA (extractable nuclear antigens) negative. In 36/82 samples antibodies against ENA were not tested.

38 samples were described as AC-4b. 5/38 were positive in MAA (1 anti-TRIM21, 1 anti-Ku, 3 anti-Mi-2). 4 samples were negative for anti-ENA as well as for MAA, 7 negative for anti-ENA, 1 negative for MAA. In 21 samples no follow up testing was performed.

Myositis is a very rare disease and MAA infrequently found in daily routine testing. But our data with a restricted number of myositis antibody positive samples show that separation of the AC-4 pattern into two categories enhances the probability to find myositis associated antibodies.

References

1. Röber N, Dellavance A, Ingénito F, Reimer ML, Carballo OG, Conrad K, Chan EKL, Andrade LEC. Strong Association of the Myriad Discrete Speckled Nuclear Pattern With Anti-SS-A/Ro60 Antibodies: Consensus Experience of Four International Expert Centers. Front Immunol. 2021 Oct 5;12:730102. doi: 10.3389/fimmu.2021.730102

2. Klotz W, Herold M. Commentary: Strong Association of the Myriad Discrete Speckled Nuclear Pattern With Anti-SS-A/Ro60 Antibodies: Consensus Experience of Four International Expert Centers. Front Immunol. 2022.18;13:840960. doi: 10.3389/fimmu.2022.840960

Antinuclear antibodies in patients with rheumatoid arthritis

W Klotz, M Herold

Medical University of Innsbruck, Department of Internal Medicine II, Rheumatology Laboratory, Innsbruck, Austria

Antinuclear antibodies (ANA) are frequently found in patients with Rheumatoid Arthritis (RA). This phenomenon is well known since many years (1). The number of ANA positive patients increases with disease duration (2).

Anti-citrullinated protein autoantibodies (ACPA) target different citrullinated proteins and peptides and are weighty findings in the 2010 defined classification criteria for RA (3) by the European League against Rheumatism (EULAR) and the American College of Rheumatology (ACR). Aim of our study was to estimate the frequency of ANA positivity in ACPA positive patients who are supposed to suffer from RA. We did a computer search in our data collected 2022 from January 1 until December 31 and selected samples in which ACPA and ANA were analysed. We found 2,411 patients with both ACPA and ANA data. In 2235 ACPA negative patients 1832 (81.97 %) were ANA negative, 403 (18.03 %) ANA positive. In 176 ACPA positive patients 99 (56.25 %) were ANA negative, 77 (43.75 %) ANA positive. Chi square test revealed statistical significance (p<0.001).

The difference in ANA positivity between patients positive or negative for ACPA cannot be explained by neither sex (66.7 % females in ACPA negative vs. 75.6 % in ACPA positive patients) nor age (mean 54 y in ACPA negative, 60 y in ACPA positive patients).

Our data confirm the knowledge that ANAs are more often found in patients with RA compared to healthy subjects. Though classification criteria have got more precise and specific autoantibodies like ACPA extend the diagnostic precision ANAs seem to be an accompanying phenomenon without any influence on prognosis or therapy. ANAs have never been included in any international recommendations for the management of patients with RA.

References

1. Ward DJ, Johnson GD, Holborow EJ. Antinuclear Factor in Rheumatoid Arthritis, its Incidence and Clinical Significance. Ann Rheum Dis. 1964;23:306–10. doi: 10.1136/ard.23.4.306

2. Condemi JJ, Barnett EV, Atwater EC, Jacox RF, Mongan ES, Vaughan JH. The significance of antinuclear factors in rheumatoid arthritis. Arthritis Rheum. 1965;8:1080–93. doi: 10.1002/art.1780080607

3. Rönnelid J, Turesson C, Kastbom A. Autoantibodies in Rheumatoid Arthritis - Laboratory and Clinical Perspectives. Front Immunol. 2021;14;12:685312. doi: 10.3389/fimmu.2021.685312

Antinuclear antibodies testing in autoimmune hepatitis, HEp-2 cells Vs. rodent tissues

W Klotz, M Herold

Medical University of Innsbruck, Department of Internal Medicine II, Rheumatology Laboratory, Innsbruck, Austria

Autoantibodies allow the classification of autoimmune hepatitis (AIH) into type 1 AIH, characterized by presence of antinuclear antibodies (ANA) or anti smooth muscle antibodies (SMA) and type 2 AIH, associated with anti LKM-1 antibodies. According to the simplified diagnostic criteria (1), detection of ANA should be performed on sections of rodent tissue to avoid false positive results. HEp-2 cells are widely used for detection of ANA. Recently the use of HEp-2 cells or ELISA instead of rodent tissue to detect ANA in AIH was validated using a restricted number of AIH patients and controls (2). The authors conclude that HEp-2 cells are comparable to tissue sections if the cut-off titer is adjusted by four-fold increase. In the present study, we verified these results using a large number (2,842) of consecutive patients’ samples sent for screening on autoantibodies associated with autoimmune liver diseases.

Figure 1 : ANA detected on HEp-2 cells and KSL slides.

Out of 2,842 samples 588 were tested positive for ANA using rodent kindey-stomach-liver sections (KSL), 1,538 samples were tested positive for ANA on HEp-2 cells using the same cutoff (1:40) as for KSL. Excluding isolated cytoplasmic or mitotic patterns 1,325 samples were tested positive on HEp-2 cells. If the cutoff for ANA on HEp-2 cells is adjusted to 1:160 as suggested in (2), 620 samples will be considered as ANA positive on HEp-2 cells. 181 samples were tested positive using a cutoff of 1:80 on KSL, 264 samples were tested positive on HEp-2 cells using the adapted cutoff of 1:320.

Using adjusted cutoffs on HEp-2 cells, the number of ANA positive patients still slightly exceeds the number of positive results on KSL both for 1:40 as well as for 1:80 dilution. As IIF on HEp-2 cells allows to differ ANA patterns and thus adds information on clinical relevance (e.g. indication of overlap to PBC in case of AC-6 or AC-12 patterns) HEp-2 assay is suitable for screening on ANA if autoimmune hepatitis is suspected.

References

1. Hennes EM, Zeniya M, Czaja AJ et al. Simplified criteria for the diagnosis of autoimmune hepatitis. Hepatology 2008;48:169–76.

2. Galaski J, Weiler-Normann C, Schakat M et al. Update of the simplified criteria for autoimmune hepatitis: evaluation of the methodology for immunoserological testing. J Hepatol 2021;74(2):312–20.

Longitudinal measurements of quinolinic acid in patients with SARS-CoV-2

S Michaelis ¹ , S Zelzer ² , C Schneider ¹ , A Meinitzer ² , I Stelzer ¹ , M Herrmann ² , D Enko ^1,2

¹ General Hospital Hochsteiermark Leoben, Institute of Clinical Chemistry and Laboratory Medicine, Leoben, Austria

² Medical University of Graz, Clinical Institute of Medical and Chemical Laboratory Diagnostics, Graz, Leoben, Austria

Quinolinic acid (QA) is a metabolite generated in the degradation of tryptophan (TRP) via kynurenine. This metabolic pathway plays a role in the immunological response against viral infections such as SARS-CoV-2. The aim of this study was to get insight into the behavior of QA in SARS-CoV-2 positives and its possible prognostic value. 104 patients with a SARS-CoV-2 infection hospitalized between August 2020 and April 2021 were included. 80 of these patients survived the disease (survival group), whereas 24 died (deceased group). The plasma concentrations of tryptophan and QA were measured on admission and seven days after admission. The difference between those two measurements (QA-DiFF) was calculated for each patient. On admission, there was no significant difference in the QA levels between the two groups. However, one week thereafter the deceased group showed significantly higher QA concentrations compared to the survival group (medians 1,075 vs. 1960 nmol/L, p<0.001). While QA levels significantly increased in deceased patients between admission and day seven (medians 1,410 vs. 1960 nmol/L, p<0.001), no statistically significant difference was observed in the patients who survived. Hence, QA-DIFF was significantly higher in the deceased group (medians 24.4 % vs. −19.0 %, p<0.001). In conclusion, QA is a promising marker for the survival prognosis of patients infected with SARS-CoV-2, particularly considering the development over the course of the disease.

Prevalence of vitamin D deficiency in a large cohort of interdisciplinary hospitalized patients before and during the coronavirus disease 2019 pandemic

Y Gao ¹ , F Caselotto ¹ , T Garmatiuk ¹ , C Graenitz-Trisko ¹ , W Huf ² , S Demyanets ^1,3

¹ Klinik Hietzing, Vienna Healthcare Group, Vienna, Austria

² Karl Landsteiner Institute for Clinical Risk Management, Vienna, Austria

³ Medical University of Vienna, Vienna, Austria

Vitamin D is not only implicated in bone metabolism through regulation of calcium homeostasis, but also known to have pleiotropic protective effects. Therefore, vitamin D deficiency was proposed as complimentary risk factor for aggravated clinical course including hospitalization in numerous disorders. We aimed to investigate the prevalence of vitamin D deficiency retrospectively in a large cohort of interdisciplinary hospitalized patients (n=33698, 60.5 % female; average age 62.6±20.4) in a municipal hospital of Vienna in relation to their age and gender. As changes in life style and socio-economic factors may influence vitamin D status, we additionally compared the prevalence of insufficient vitamin D level before (01/2017–12/2019) and after (01/2020–12/2022) onset of the COVID-19 pandemic in Austria. Serum 25-hydroxy vitamin D (25(OH)D) was measured by CLIA technology (LIAISON^® 25 OH Vitamin D Total Assay). In the entire sample, we found that 26.1 % had sufficiency (30–100 ng/mL), 30.2 % insufficiency (20–29 ng/mL), 25.7 % deficiency (10–19 ng/mL), and 18.0 % severe deficiency (<10 ng/mL) of vitamin D. Interestingly, higher prevalence of vitamin D deficiency (<30 ng/mL) was found in the younger subjects (age 18–30, p<0.001). In the entire cohort, women had higher levels of 25(OH)D (23.9±12.6 ng/mL) compared to males (21.0±12.3 ng/mL, p<0.001). Average level of vitamin D was slightly higher after (22.9±13.0 ng/mL) onset of the pandemic compared to before (22.6±12.3 ng/mL; p=0.005), mainly driven by females in the age group of 80–90 years (before 23.2±12.6, after 25.8±13.7 ng/mL, p<0.001, p-value corrected for daylight seasonal variation). We can conclude that the prevalence of vitamin D deficiency is high, slightly decreasing with age, particularly in females. Interestingly, vitamin D levels were slightly higher after onset of the pandemic, again primarily in females, reaching significance in the age group of 80–90.

Low levels of malondialdehyde IgM antibodies contribute to atherothrombotic risk

T Koller, T Afonyushkin, M Oszvar-Kozma, S Taqi, M Hoke, C J Binder

Medical University of Vienna, Vienna, Austria

The plaque milieu’s oxidative environment generates lipid peroxidation products such as malondialdehyde (MDA), an oxidation specific epitope (OSE) which modify plaque components. OSEs are recognized by natural IgM antibodies, titres of which inversely correlate with cardiovascular disease (CVD) risk, the mechanism of which is poorly understood. MDA IgMs inhibit extracellular vesicle (EV)-induced clotting and thrombosis. Besides EVs, other plaque components are prothrombotic, MDA modified and released during plaque rupture or erosion. We thus hypothesize that MDA IgMs exert their protection through atherothrombosis inhibition. Consequently, we assessed their anti-thrombotic potential in vitro and in vivo. In vitro, MDA plaque protein or plaque homogenates increased measures of platelet activation (Multiplate) and coagulation cascade activation (thrombin generation, thromboelastometry), which was inhibited by LR04, an MDA targeting natural IgM, but not its isotype. In vivo, LR04 protected from mouse pulmonary thrombosis triggered by injection of MDA plaque protein. Finally, ELISA measurements of IgM levels targeting MDA modified plaque protein in 704 cardiovascular patient samples showed significantly increased survival of statin treated patients with high antibody titres. Taken together, we describe a novel protective mechanism of MDA IgMs through atherothrombosis inhibition via reduction of platelet and coagulation cascade activation by MDA modified plaque components. Levels of these antibodies can be used to risk stratify patients in whom dyslipidaemia is addressed with statin therapy, representing a potential marker of residual CVD risk in patients receiving lipid-lowering therapy.

Measurement of NOAC plasma concentrations -- indicated or not indicated in clinical practice?

A Hausch, Y Herschel-Aydinli, T Mueller

Hospital Voecklabruck, Department of Laboratory Medicine, Voecklabruck, Austria

BACKGROUND: In contrast to vitamin K antagonists, which have a narrow therapeutic range and therefore require coagulation monitoring, non-vitamin K antagonist oral anticoagulants (NOACs) have a broad therapeutic window. Hence, they have the advantage of not requiring coagulation monitoring. However, there is consensus that in special situations, measuring NOAC plasma concentrations can be helpful. In this study, we sought to check whether NOAC determinations were indicated in daily clinical practice. METHODS: During a period of 7 months (i.e. from April to October 2022), we recorded all determinations of dabigatran, rivaroxaban, apixaban and edoxaban in the clinical setting of a tertiary care hospital. Two laboratory physicians recorded the medical indications of all NOAC determinations during this period from the medical records. This was done in accordance to the “2021 European Heart Rhythm Association Practical Guide on the Use of NOACs in Patients with Atrial Fibrillation” (ESC guideline). In case of discrepancies between the judgement of the two laboratory physicians or otherwise unclear indications, we consulted the requesting clinicians. In this way, each NOAC measurement was assigned a unique medical indication. RESULTS: During the study period, we had 351 NOAC measurements in 245 patients. A majority of patients was treated in the departments of trauma surgery (n=52), internal medicine (n=50), neurology (n=42) and intensive care (n=40). Of the 351 NOAC measurements, 297 (85 %) had a clear clinical indication according to the ESC guideline, whereas 54 (15 %) had no clinical indication. The guideline compliant indications were related to unplanned surgery/intervention (n=126; i.e. 4 acute, 118 urgent, and 4 expedite), severely impaired renal function (n=90), acute stroke (n=41), bleeding (n=28; i.e. 11 mild bleeding, 17 non-life-threatening major bleeding, and 0 life-threatening or bleeding into critical site), pharmacokinetics and drug-drug interaction (n=7), and extremes of body weight (n=5). In contrast, there were 31 non-indicated demands of NOAC determinations in elective surgery, and 23 demands without any indication for determining the NOAC plasma concentration. CONCLUSION: Our data show that 15 % of all NOAC measurements were not indicated. The overuse of NOAC measurement might go at the expense of patients’ health by misinterpreting laboratory data. Thus, our data might be a hint for other laboratories to reevaluate their demand management system. Furthermore, health expenditures are increasing. Reducing unnecessary analyses could contribute to saving resources.

Comparison of factor VIII activity as measured by one-stage clotting assay vs. chromogenic assay in a large cohort of patients with different diseases

K Giacomuzzi ¹ , J Sandforth ¹ , K Puechler ¹ , A Hausch ² , Y Herschel-Aydinli ² , T Mueller ^1,2

¹ Hospital Bolzano, Department of Clinical Pathology, Bolzano, Italy

² Hospital Voecklabruck, Department of Laboratory Medicine, Voecklabruck, Austria

BACKGROUND: Factor VIII (FVIII) is a non-enzymatic plasma protein that is essential for normal blood coagulation. Factor VIII coagulation activity (FVIII:C) can be measured by both, one-stage clotting assays and chromogenic assays. Here, we report real world data from simultaneous FVIII:C measurements with both methods.

METHODS: We retrospectively evaluated all orders for the determination of FVIII:C over a period of 2.5 years at our institution. During this time, we used two assays simultaneously for each FVIII:C plasma sample in clinical practice, namely a one-stage clotting assay (one-stage aPTT-based clotting assay with FVIII-deficient plasma, HemosIL SynthASil, Werfen) and a chromogenic assay (Coamatic Factor VIII, Chromogenix). For this study, we extracted the results of the hemostasis tests from the laboratory information system and reviewed the corresponding medical findings.

RESULTS: We analyzed 1,179 orders from 977 patients (223 pediatric and 754 adult individuals). In this cohort, there were no patients with emicizumab or extended half-life FVIII replacement drugs. We analyzed the results on 47 samples from 23 patients with congenital hemophilia A, 47 samples from 5 patients with acquired hemophilia A (AHA), 63 samples from 50 patients with other single factor deficiencies, 100 samples from 57 patients with inherited von Willebrand disease (vWD) or acquired von Willebrand syndrome (vWS), 43 samples from 41 patients with lupus anticoagulant, and 879 samples from 801 patients with other diseases. The comparison of the 1,179 parallel FVIII:C determinations with the one-stage clotting assay and the chromogenic assay by Spearman rank correlation yielded a correlation coefficient of 0.901 (95 % confidence interval, 0.889 to 0.911) and a significant deviation from linearity (p=0.01). In the Bland–Altman diagram we plotted FVIII:C as measured by the clotting assay (x-axis) against the differences between the two methods expressed as percentages of the results shown on the x-axis. In this analysis, we found that all samples with a difference of >200 % were from patients with lupus anticoagulant. In these measurements, the FVIII clotting assay gave falsely low results, while the chromogenic assay showed the true values.

CONCLUSION: In our cohort, we generally found good agreement between the two methods for determining FVIII:C. However, interference from lupus anticoagulants occurred with the one sage clotting test. This interference mimicked FVIII deficiency or a specific FVIII inhibitor. In contrast, the lupus anticoagulant did not interfere with the chromogenic assay. In some cases, chromogenic assays may therefore be preferable to one-stage clotting assays.

Effects of automated tube sorting on specific clinical laboratory parameters – a pilot study

T Karl, LV Kipman, A Padinger, V Dellago-Dworak, C Hölzl, I Siorpaes, GJ Oostingh

Salzburg University of Applied Sciences, Biomedical Sciences, Department of Health Sciences, Puch/Salzburg, Austria

A bulk sorter optimizes the workflow in laboratories with large numbers of samples. The sample tubes are sorted on racks, in bulk depots or directly at the point of measurement. This sorting process agitates the tubes continuously. The effects of mechanical agitation on the sample tubes and hence on the laboratory parameters have hardly been investigated so far. We aimed to analyze possible deviations in laboratory parameters in the field of clinical chemistry, hematology and hemostaseology caused by the use of a bulk sorter. Therefore, a pilot project was performed at the Salzburg University of Applied Sciences in the division of Biomedical Sciences. Venous blood was drawn from 17 volunteers (2 EDTA, 2 serum and 2 citrate sample tubes from each volunteer). Thereafter, the transport of the samples to the lab was simulated. The sample tubes were then divided into 2 groups. One of the 2 identical samples was prepared manually and the other one was processed using the Automated Tube Registration and Sorting (ATRAS) system before being analyzed. In clinical chemistry, the hemolysis parameters potassium, lactate dehydrogenase and aspartate aminotransferase were analyzed on the Cobas c 311. In hematology the number of erythrocytes, lymphocytes and thrombocytes was measured on the Sysmex XN-530. The coagulation parameters partial thromboplastin time, the plasma thrombin time and the international normalized ratio were determined on the Siemens CS-2100i. To determine the validity of the results, 3 measurements per sample were performed for each parameter and used as such for the statistical evaluation of the mean values of the measurements. The deviations between the laboratory results of the 2 sample groups could not be classified as abnormal. By using the ATRAS bulk sorting system, no deviations in the results can be determined within the measured laboratory parameters. To confirm the results of the pilot study, a further study with more subjects is recommended.

Assessment of clinical chemistry parameters using a verification scheme based on CLSI EP15

C Schneider, S Michaelis, LI Jooda, D Enko

General Hospital Hochsteiermark, Institute of Clinical Chemistry and Laboratory Medicine, Leoben, Austria

Within the framework of accreditation by ISO 15189, the verification of certified tests is mandatory. We present a scheme based on CLSI EP15-A3, which was implemented in our laboratory for the verification of clinical chemistry analyses. Verification procedures were performed for selected liver parameters (i.e. direct bilirubin, total bilirubin, gamma-glutamyltransferase (GGT), aspartate transaminase (AST), alanine transaminase (ALT), cholinesterase (CHE)) on two cobas pro c503 (Roche, Rotkreuz, Switzerland) analyzers. Two analyte concentration levels (low and high) were included utilizing the manufacturer´s quality controls. The verification protocol consisted of five measurements per level per day for five days (5×5 scheme). Bias and imprecision were calculated for each analyte on each analyzer according to CLSI EP15 A3. Imprecision results deviating from the calculated verification limit were compared to biological variations and checked for clinical significance. The maximal allowable bias was obtained from RiliBäk or literature or was arbitrary set to 15 % if clinically acceptable. For each device, minor deviations from the limits given by the EP15 for some parameters were observed. This can be attributed to the narrow limits stated by the manufacturers due to fixed protocols under fixed conditions while our measurements represent more real life conditions. Nevertheless, all parameters were in the limits of biological variation and were suitable for routine use. In conclusion, the application of this verification scheme proved useful in terms of usability, statistical reliance, input of workforce and cost effectiveness.

Product problems in companion diagnostics – analysis of field safety notes published by BfArM until end 2022

R Siekmeier ¹ , J Hannig ¹ , W März ²

¹ Drug Regulatory Affairs, Pharmaceutical Institute, University Bonn, Germany

² Synlab Akademie, synlab Holding Deutschland AG, Augsburg, Germany

Introduction: Marketing/market surveillance of in vitro diagnostics (IVD) in Europe is subject of the Regulation (EU) 2017/746 of the European Parliament and of the Council replacing Directive 98/79/EC. Compared to Directive 98/79/EC the Regulation 2017/746 is more stringent and explicitly covers Companion diagnostics (CDx). In cases of incidents and Field Safety Corrective Actions (FSCA) manufacturers have to inform the responsible Competent Authority (D: Federal Institute for Drugs and Medical Devices (BfArM)) and the public by Field Safety Notices (FSN). Here we evaluated FSCA and FSN for CDx tests (not analyzers/common consumables and stand-alone software as a service (SaaS)) published by BfArM.

Materials and Methods: FSN published 2005–2022 on the BfArM homepage (https://www.bfarm.de/SiteGlobals/Forms/Suche/Expertensuche_Formular.html?nn=597716&cl2Categories_Format=kundeninfo) were analyzed. CDx were defined according definitions/lists published by FDA (https://www.fda.gov/medical-devices/in-vitro-diagnostics/list-cleared-or-approved-companion-diagnostic-devices-in-vitro-and-imaging-tools) and VfA (https://www.vfa.de/de/arzneimittel-forschung/datenbanken-zu-arzneimitteln/individualisierte-medizin.html) including each more than 100 compounds for which testing is mandatory/recommended prior to therapy.

Results: Out of 3417 FSCA for all IVD 39 FSCA were found for CDx (first FSCA 2009) of which 28 were related to molecular and 11 to immunological assays. All but 1 were for use in oncology, mostly KRAS (7), HER2 (6), EGFR (5), BRAF (3), BCR-ABL (3) and estrogen receptor (3). Typical product problems (multiple entries) were high/false-positive results (13), low/false-negative results (12), no/invalid results (10) or erroneous results (3). Typical corrective measures were (multiple entries) FSN (mandatory in cases of recall; 39, including information for problem handling), recall (26), retesting of samples (26), modification of the instructions for use (6) and software-upgrade (2).

Conclusions: The number of FSCA for CDx is small but increasing. As BfArM publishes only FSCA affecting the German market (the largest in Europe) and analyzers for CDx/common consumables and SaaS were excluded the number of FSCA affecting CDx in Europe tends to be higher. FSN play an important role for risk reduction in case of CDx product problems.

Erythropoetin rs67640 gene polymorphism affects erythropoiesis but not life expectancy of patients with peripheral arterial disease

W Renner, O Trummer, S Khuen, U Langsenlehner, H Mangge, T Langsenlehner

Medical University of Graz, Clinical Institute of Medical and Chemical Laboratory Diagnostics, Graz, Austria

A common functional polymorphism (rs1617640, A>C) in the promoter of the erythropoietin gene (EPO) has been associated with erythropoietin expression and markers of erythropoiesis. We aimed to analyze the potential role of this genetic marker for long term survival of patients with peripheral arterial disease (PAD).

EPO genotypes and laboratory markers for erythropoiesis were determined in a cohort of 946 patients with PAD. Survival follow-up was performed 20 years after recruitment of patients.

At the time of recruitment, the minor EPO rs1617640 C-allele was associated in an allele-dose-dependent manner with hemoglobin levels (p=0.006), hematocrit (p=0.029). Twenty years after recruitment, 752 (79.5 %) patients were dead, 103 (10.9 %) were still alive, and 91 (9.6 %) were lost-to-follow up. In a Cox regression analysis including sex, smoking habit, type-2 diabetes, arterial hypertension and hypercholesterolemia, EPO genotypes were not associated with overall survival (Hazard ratio 0.63; 95 % confidence interval 0.88–1.08, p=0.63). The functional EPO rs1617640 gene polymorphism, irrespective of its association with markers of erythropoiesis, does not affect survival of PAD patients.

Performance of the automated DNA extraction with MagNA Pure 24^® (Roche^®) for further genetic testing for hemoglobinopathies with Globin StripAssays^® (ViennaLab^®)

G Leixner ¹ , J Gaugeler-Kurzweil ¹ , A Voill-Glaninger ¹ , L Kager ² , W Novak ² , T Kusstatscher ¹ , A Viveiros ¹

¹ Klinik Landstrasse, Vienna Healthcare Group, Institute for Laboratory Medicine, Vienna, Austria

² St. Anna Children’s Hospital, Hematology and Oncology Outpatient Clinic, Vienna, Austria

Introduction: Genetic testing for hemoglobinopathies through reverse hybridization with the α- and β-Globin StripAssays^® (ViennaLab^®) is widely performed. There are currently only two authorized DNA extraction methods for these assays: Spin Micro DNA Extraction Kit (ViennaLab^®) for the α-Globin StripAssay^® and GenXTractTM Blood DNA Extraction System (ViennaLab^®) for the β-Globin StripAssay^®, both of which are time consuming due to their hands-on character. Further downsides are the different extraction protocols, analog lot documentation and the absence of a CE certificate for the Spin Micro kit. In this study, we set out to evaluate the performance of an automated DNA extraction using the MagNA Pure 24^® (Roche^®) platform for further testing with the Globin StripAssays.

Materials and Methods: We analyzed blood samples of 33 patients with clinical suspicion of hemoglobinopathies who had given their written consent. As the gold standard, we stipulated the DNA extraction methods according to the Globin StripAssay instruction manual and compared it to the MagNA Pure 24 Total Isolation Kit performed on the MagNA Pure 24. The results of the different strips were qualitatively classified into three categories: strip read out weak, good and/or divergent. PCR was performed according to the Globin StripAssay instruction manual and reverse hybridization with the Autoblot 3,000.

Results: From a total of 990 α-Globin StripAssay analyzed lines, identical strip read outs were found in 98.1 % (971/990). After extraction with Spin Micro, faint lines were detected in 1.2 % (12/990) and false negative lines in 0.7 % (7/990). Out of 1,323 β-Globin StripAssay lines, identical strip read outs were found in 98.3 % (1,300/1,323). Samples extracted with GenXTract showed faint lines in 0.2 % (3/1,323) and (weakly) false positive lines in 1.1 % (15/1,323). After extraction with MagNA Pure, a weak read out was observed in 0.4 % (5/1,323) of total lines.

Conclusions: The automated DNA extraction with MagNa Pure 24 showed non-inferior results when compared with the Spin Micro and Gen^XTract DNA extraction kits. Furthermore, and in contrast to the latter, no false negatives nor positives were detected using the automated system. The results of this study allow the usage of the MagNa Pure 24 for the purpose of DNA extraction for further genetic testing with Globin StripAssays.

Comparison of two fully automated high-pressure liquid chromatography methods (Hb NEXT vs. HA-8180T) for detection of variant hemoglobins

I Herbring, A Black, K Wiesinger, B Dieplinger, M Egger-Salmhofer

Konventhospital Barmherzige Brueder Linz and Ordensklinikum Linz, Department of Laboratory Medicine, Linz, Austria

Background: Hemoglobin variants are a result of genetic accent="true" changes that lead to abnormal production of hemoglobin chains (thalassemia) or the generation of hemoglobin chain variants. The aim of this study was to compare the performance of the thalassemia mode of two fully automated high-pressure liquid chromatography (HPLC) methods in detecting common hemoglobin variants.

Methods: For the present study EDTA whole blood samples of 61 patients with suspected hemoglobinopathies were parallel tested with the thalassemia mode of the Hb NEXT (A.Menarini diagnostics) and HA-8180T (Arkray) HPLC methods.

Results: Of the 61 patients we found 12 patients with HbS, 6 patients with HbE, 6 patients with HbD, 18 patients with beta-thalassemia minor, 1 patient with beta-thalassemia major, 1 patient with elevated HbF and 2 patients with Hb-Lepore. No evidence of beta-thalassemia or any other hemoglobinopathy was found in the remaining 15 patients. For all 61 patients we found concordant results between the Hb NEXT and HA-8180T HPLC methods. Both analyzers are very user-friendly, whereby switching to thalassemia mode on the Hb NEXT provides the option of only one machine for both variant determination and HbA1c measurement, with a time saving for HbA1c measurement only (60 s without variant determination). In contrast the thalassemia mode runs faster on HA-8180T (210 vs. 360 s). The chromatogram display is currently somewhat clearer on the HA-8180T when compared to the Hb NEXT due to a scaling of the x-axis (seconds) and a labeling of the most important peaks (HbF, HbA1c, HbA2).

Conclusions: In the present comparison study of the Hb NEXT and the HA8180T we found a perfect agreement between the two methods. Therefore, the Hb Next HPLC is well suitable for detecting common hemoglobin variants in patients with suspected hemoglobinopathies. However, due to our rather small sample size, only covering common hemoglobin variants, further investigations in larger cohorts of patients including also other hemoglobin variants are necessary.

Analytical and clincal evaluation of a novel 11-color 1-tube flow cytometry panel for measurable residual disease detection in multiple myeloma patients

K Deutsch-Biedermann ¹ , J Gruber ¹ , S Machherndl-Spandl ² , I Strassl ² , P Bettelheim ² , I Herbring ¹ , C Kimbacher ¹ , K Hefler-Frischmuth ¹ , B Dieplinger ¹

¹ Department of Laboratory Medicine, Konventhospital Barmherzige Brueder Linz and Ordensklinikum Linz, Linz Austria

² Department of Internal Medicine I: Hematology with Stem Cell Transplantation, Hemostaseology and Medical Oncology, Ordensklinikum Linz Elisabethinen, Linz, Austria

Background: Multiparametric flow cytometry is a highly valuable method for assessment of measurable residual disease (MRD) in multiple myeloma patients. Euroflow Next Generation Flow (NGF) utilizes a contemporary 8-color 2-tube panel which measures a total of 10 specific antigens (cyto kappa, cyto lambda, CD19, CD27, CD38, CD45, CD56, CD81, CD117, CD138). The aim of the present study was an evaluation of a novel 11-color 1-tube panel, consisting of all recommended Euroflow NGF markers plus CD200.

Methods: A DxFLEX flow cytometer equipped with 3 lasers (405/488/638nm), 13 fluorescence detectors, standard filter configuration, reagents, and antibodies (cyto kappa, cyto lambda, CD19, CD27, CD38, CD45, CD56, CD81, CD117, CD138, and CD200) from Beckman Coulter were used for our novel 11-color 1-tube panel (novel MRD-panel). The bone marrow aspirates underwent an ammonium chloride-based bulk lysis, followed by staining of the surface antibodies and, after permeabilization, light-chain intracellular staining. We acquired 1x107 cells per sample and determined limit of detection (LOD) and lower limit of quantification (LLOQ) as recommended by the International Myeloma working group. For the clinical evaluation samples from 20 patients with multiple myeloma under or after treatment were analyzed with our novel MRD-panel and results were compared with our routine plasma cell panel (RPC-panel) for which we collected 500 plasma cells using a 10-color 1-tube method on the same flow cytometer. For 6 patients a second bone marrow aspirate was drawn in parallel, sent to another laboratory and analyzed with the contemporary Euroflow NGF 8-color 2-tube panel (contemporary MRD-panel) using the CYT-MM-MRD8 kit (Cytognos) on a BD FACSCanto (Becton Dickinson) flow cytometer.

Results: For our novel MRD-panel we determined a LOD of 0.00006 % (6×10⁻⁷) and a LLOQ of 0.0006 % (6×10⁻⁶). In the 20 multiple myeloma patients we found 11 patients with abnormal plasma cells (MRD positive) and 9 patients with only normal plasma cells (MRD negative), when using our novel MRD-panel. Comparing these results to our RPC-panel we detected 16 concordant results. Four MRD positive patients with the novel MRD-panel were not detected with the RPC-panel. All 6 patients with parallel measurements using the novel MRD-panel and the contemporary MRD-panel showed concordant results (2 MRD positive patients vs. 4 MRD negative patients for both MRD-panels).

Conclusions: The novel MRD-panel meets the quality specifications of the International Myeloma working group. Our clinical evaluation found higher sensitivity for the novel MRD-panel when compared to our RPC-panel and concordant results with the contemporary MRD-panel. Therefore, our novel MRD-panel is well suitable for detecting MRD in multiple myeloma patients.

Utility of CD148 for the differential diagnosis of mantle cell lymphoma (MCL)

A Klemenjak ¹ , G Slavka ¹ , I Fang ² , W Hübl ¹

¹ Klinik Ottakring, Zentrallabor, Vienna, Austria

² KH Barmherzige Brüder Wien, Zentrallabor und Blutdepot, Vienna, Austria

Background: Expression of CD148 on clonal B-cells has been reported to be of value in the differential diagnosis of mature B-cell neoplasms, especially CLL and MCL, either as single parameter or in relation to the expression of CD200 or CD180. We aimed to verify this data by retrospective analysis of hospital cases. Methods: CD148, CD200 and CD180 expression of 42 cases of CLL and 16 cases of MCL was retrospectively analyzed. Diagnoses were based on smear-, bone marrow- and lymph node morphology, immunohistochemistry, immunophenotype (excluding CD148, CD200, and CD180), and genetic analysis. Additionally, CLL-cases were categorized according to the Matutes-Score into cases with maximum score (5) and cases lacking at least one typical characteristic of CLL. Results: Median fluorescence intensity (MFI) of CD148, CD148/CD200 ratio and CD148/CD180 ratio of CLL and MCL were 1888 vs. 8,702 (p<0.0001), 0.78 vs. 86.3 (p<0.0001), and 10.8 vs. 34.6 (p=0.0054), respectively. ROC analyses yielded respective AUC values of 0.969, 0.999, and 0.738. Cut-off points for optimal diagnostic efficiency were>4,205, >2.9, and >30.4, respectively. CLL subgroups with weak/absent CD5 expression (n=9), weak/absent CD23 expression (n=9), strong CD20 expression (n=7), and strong light-chain expression/CD79b positivity (n=12) had median CD148 expressions of 1,476, 2,781, 2,811, and 3,622, respectively. CD148/CD200 ratio was 0.81, 0.81, 0.91, and 1.08, while CD148/CD180 ratio of these subgroups was 7.5, 12.2, 8.1, and 16.6, respectively. The low number of cases within the subgroups did not allow a valid statistical evaluation. Discussion: Our data showed a highly significant difference in median CD148 expression and CD148/CD200 ratio between CLL and MCL. Median CD148 expression alone provided far superior results to the CD148/180 ratio. However, compared to CD200 as a single marker we found no significant benefit of adding CD148 for the differential diagnoses of CLL and MCL. Notably, two of our cases displayed an aberrant CD200 expression (one CD200 negative CLL case and one CD200 positive MCL case; diagnosed by histopathology/immunohistochemistry). Using our determined MFI cut-off point, these two cases were classified correctly by CD148 expression, but their values were close to the cut-off point. Further studies with a higher number of cases with aberrant expression of CD200 are needed to determine the value of CD148 in this subgroup. Conclusion: Median CD148 expression, CD148/CD200 ratio and CD148/CD180 ratio seem to be of little additional value to immunophenotyping panels evaluating CD200 expression for the differential diagnoses of CLL and MCL.

CUL3^LZTR1 E3 ubiquitin ligase-based regulation of RAS proteostasis in hematopoietic malignancies

JW Bigenzahn ^1,2 , F Kartnig ² , M Vollert ² , G Hoermann ³ , G Superti-Furga ^2,4

¹ Department of laboratory medicine, Medical University Vienna, Vienna, Austria

² CeMM, Research Center for Molecular Medicine of the Austrian Academy of Sciences, Vienna, Austria

³ MLL Munich Leukemia Laboratory, Munich, Germany

⁴ Center for Physiology and Pharmacology, Medical University of Vienna, Vienna, Austria

The proteostatic regulation of signaling proteins, modulating their abundance and activity, is an important regulatory mechanism to coordinate cellular proliferation, survival, and differentiation. By studying the effect of single gene loss-of-function effects on the response of chronic myeloid leukemia (CML) cells to BCR-ABL targeting tyrosine kinase inhibitors (TKI), we have previously identified the CUL3^LZTR1 E3 ubiquitin ligase complex as a conserved proteostatic regulator of RAS GTPase proteins. Missense mutations of LZTR1 associated with Noonan syndrome are loss-of-function mutations and are unable to rescue the observed resistance phenotype in CML cells as well as RAS abundance. However, the molecular mechanistic details of how CUL3^LZTR1 specifically regulates RAS proteins as well as the broader implication of LZTR1 mutations in hematopoietic malignancies is still unclear.

To address these questions, we have (1) performed FACS-based genome-wide CRISPR screens on KRAS abundance to identify additional genes important for CUL3^LZTR1 function and (2) surveyed a collection of 5,000 whole genome sequencing profiles of patients with different hematological malignancies to identify potential novel LZTR1 alterations.

Findings of both approaches will be presented portraying a more refined understanding of the mechanistic details of the CUL3^LZTR1 complex as well as its significance in drug response and proliferation of hematopoietic cells.

Discrepancies in the classification of Myelodysplastic Neoplasia (MDS) regarding ring sideroblasts between two classification systems

I Stelzer ¹ , S Michaelis ¹ , C Schneider ¹ , L Schoeffmann ² , F R Vagena ³ , G Hoefler ³ , D Enko ^1,4

¹ Institute of Clinical Chemistry and Laboratory Medicine, General Hospital Hochsteiermark, Leoben, Austria

² Department of Hemato Oncology , General Hospital Hochsteiermark, Leoben, Austria

³ Diagnostic and Research Institute of Pathology , Medical University of Graz, Graz, Austria

⁴ Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Graz, Austria

According to the revised 4th edition of the WHO Classification of Tumors of Hematopoietic and Lymphoid Tissues (2017), myelodysplastic syndrome with ring sideroblasts (MDS-RS) was defined as myelodysplasia with ≥15 % ring sideroblasts or ≥5 % ring sideroblasts with a concomitant SF3B1 mutation. In the current WHO classification (5th edition, 2022), a new entity, namely myelodysplastic neoplasia (MDS) with SF3B1, has been introduced. However, a detection of ≥15 % ring sideroblasts may substitute for an SF3B1mutation (MDS with low blasts and ring sideroblasts). In contrast, the International Consensus Classification of Myeloid Neoplasms and Acute Leukemias (ICC, 2022) solely includes MDS with an SF3B1 mutation. Cases of ≥15 % ring sideroblasts without SF3B1 mutation are assigned to the group of myelodysplastic syndrome (MDS) not otherwise specified (NOS). This may have therapeutic consequences, since Luspatercept is only approved for this diagnosis. To address the issue of possible discrepant classifications (WHO vs. ICC), we present a case series of MDS with detected ring sideroblasts. Twelve patients with a diagnosed MDS-RS according to WHO (2017) were included. Ring sideroblasts were identified by Prussian blue staining. Classification was repeated using the diagnostic criteria published by the WHO (2022) classification and the ICC. According to WHO (2022), eight cases were classified as MDS with SF3B1 mutation and four cases as MDS with low blasts and ring sideroblasts. Using the ICC, eight cases were classified as MDS-SF3B1 but four cases with ≥15 % ring sideroblasts and wild-type SF3B1 were assigned to MDS-NOS. Differences in the classification of MDS with the detection of ring sideroblasts between WHO (2022) and the ICC may have an impact on therapeutic options. Particularly, without Prussian blue staining one third would not have been diagnosed as MDS with low blasts and ring sideroblasts because no SF3B1 mutation was found.

UBA1 gene mutation establishes the diagnosis of VEXAS-syndrome in a patient with MDS

B Strasser ¹ , G Hoermann ² , A Haushofer ¹

¹ Klinikum Wels-Grieskirchen , Institut für Klinische und Medizinische Labordiagnostik, Wels, Austria

² MLL Munich Leukemia Laboratory, Munich, Germany

A novel syndrome:

VEXAS is newly discovered hematological/rheumatological overlapping syndrome, which subsumes the terms Vacuoles, E1 enzyme, X-linked, Autoinflammatory, and Somatic. It was first described by Beck et al. in December, 2020. (1)

Case presentation:

A male patient was diagnosed with a myelodysplastic neoplasm (MDS) at the age of 67. In addition to this hematological neoplasm, the patient suffered from intermittent fever episodes and painful redness of the ears (aural perichondritis). Microscopic bone marrow investigation detected conspicuous vacuolized proerythroblasts and promyelocytes. The diagnosis of VEXAS syndrome was established after consecutive molecular-genetic analyses obtained the pathognomonic UBA1 p.(Met41Leu) mutation.

Cytomorphological hallmark: Precursor vacuolization

Cytoplasmatic vacuoles are detectable in precursor cells. Confounding etiologies such as alcohol abuse or copper deficiency should be excluded. Almost all patients present with significant myelodysplasia. In most cases diagnostic criteria of MDS are met, typically with a low or absent blast increase (MDS-LB or MDS-IB1). In addition, a few cases of multiple myeloma have been reported in VEXAS patients. (2)

Molecular-genetic hallmark: UBA1 mutation

To date the diagnosis of VEXAS syndrome has been restricted to UBA1 mutations at codon 41. The clinical significance of alternative UBA1 mutations has yet to be clarified. The UBA1 gene is located on the X-chromosome. Therefore, strong gender-specific differences are derivable with a male dominance. Exceptional cases of female patients with VEXAS syndrome have been reported in combination with monosomy X.

Phenotype of VEXAS:

Primary symptoms and manifestations of auto-inflammatory diseases observed in VEXAS syndrome include fever, chondritis, skin involvement, alveolitis, and arthritis. Thromboembolic complications have a high frequency in VEXAS patients. (1, 2)

Conclusion:

Myelodysplasia and precursor vacuolization should evoke the suspicion of VEXAS syndrome. The detection of UBA1 p.(Met41Leu) mutation can confirm the diagnosis.

References

1. Beck DB, Ferrada MA, Sikora KA, Ombrello AK, Collins JC, Pei W, et al. Somatic Mutations in UBA1 and Severe Adult-Onset Autoinflammatory Disease. N Engl J Med. 2020;383(27):2,628–38.

2. Patel N, Dulau-Florea A, Calvo KR. Characteristic bone marrow findings in patients with UBA1 somatic mutations and VEXAS syndrome. Semin Hematol. 2021;58(4):204–11.

A multi-omics based anti-inflammatory immune signature characterizes long COVID-9 syndrome

JJ Kovarik, A Bileck, G Hagn, SM Meier-Menches, T Frey, A Kaempf, M Hollenstein, T Shoumariyeh, L Skos, B Reiter, MC Gerner, A Spannbauer, E Hasimbegovic, D Schmidl, G Garhöfer, M Gyöngyösi,*KG Schmetterer,* and C Gerner

Medical University of Vienna, Department of Laboratory Medicine, Vienna, Austria

To investigate long COVID-19 syndrome (LCS) pathophysiology, we performed an exploratory study with blood plasma derived from three groups: (1) healthy vaccinated individuals without SARS-CoV-2 exposure; (2) asymptomatic recovered patients at least three months after SARS-CoV-2 infection and; 3) symptomatic patients at least 3 months after SARS-CoV-2 infection with chronic fatigue syndrome or similar symptoms, here designated as patients with long COVID-19 syndrome (LCS). Multiplex cytokine profiling indicated slightly elevated proinflammatory cytokine levels in recovered individuals in contrast to patients with LCS. Plasma proteomics demonstrated low levels of acute phase proteins and macrophage-derived secreted proteins in LCS. High levels of anti-inflammatory oxylipins including omega-3 fatty acids in LCS were detected by eicosadomics, whereas targeted metabolic profiling indicated high levels of anti-inflammatory osmolytes taurine and hypaphorine, but low amino acid and triglyceride levels and deregulated acylcarnitines. Thus, this study provides first insight into the pathophysiology of LCS and points at potential biomarkers for diagnosis.

Analysis of the parodental microbiome: Detection of periodontopathogenic microorganisms and evaluation of risk factors

S. Sirovina

Labors.at, Molecular Biology, Vienna, Austria

Periodontitis is a common inflammatory disease that leads to destructive changes of the periodontium. Both, the subgingival colonization of pathogenic bacteria and influence of risk factors, lead to this dysbiotic state. The detection of A. actinomycetemcomitans (A.a complex), P. intermedia (orange complex), P. gingivalis (red complex), T. forsythia (red complex) and T. denticola (red complex) are particularly associated with periodontitis and are targeted by the micro-IDent test system. By extracting, using the previously validated chemagic 360-D, and analyzing subgingival plaque samples (n=50), using the micro-IDent test system, the presence of pathogenic microorganisms could be confirmed in a healthy population, too. The evaluation of two questionnaires showed a highly significant correlation between the detection of microbial complexes and the determined risk score of periodontitis. Microorganisms of the red complex were most frequently isolated, while A. actinomycetemcomitans was the least common. The type of toothbrush and a professional oral hygiene session were the most important factors influencing the complexity of the subgingival microbiota.