Meta-Analysis Scientific Posters PUBLICATIONS & REPORTS on ... Posters on Sebia Solutions.pdf · MONDAY - June 22th Hemoglobinopathies Advanced technologies, new biomarker discovery,

Analytical Performances of HbA1c technique on CAPILLARYS 2 Flex Piercing and MINICAP Flex Piercing

Meta-Analysis PUBLICATIONS & REPORTS 2011-2015

Scientific Posters on Sebia Solutions

IFCC EuroMedLab JIB Paris 2015

Parc Téchnologique Léonard de Vinci CP 8010 Lisses - 91008 Evry CEDEX France Tel: +33 (0)1 69 89 80 80 e-mail: [email protected]

MONDAY - June 22th

Hemoglobinopathies

Advanced technologies, new biomarker discovery, microparticles, exosomes...

M017

FIVE HEMOGLOBIN VARIANTS DETECTED FOR THE FIRST TIME BY CAPILLARY ELECTROPHORESIS

S. Barella1, F.R. Demartis1, M.F. Desogus1, M.C. Sollaino1, R. Galanello1

1Ospedale Regionale per le Microcitemie-Laboratorio Ematologia-Cagliari, Italy

BACKGROUND-AIM

Hemoglobinopathies (Hbe) are inherited hemoglobin (Hb) disorders, leading to the synthesis of mutated globin chainsor a change in their expression level. Capillary electrophoresis (CE) is one of the methods used to test Hbe. Often use inroutine, CE permits fast and precise separation of the Hb fractions, as well as their quantification. Used in our laboratoryfor five months, CE allowed us to detect, among others, 5 Hb variants that we report here, and never described in theliterature by this technique.

METHODS

Whole blood samples were analyzed by CE (CAPILLARYS 2 Flex Piercing, Sebia) and "Hemoglobin(e)" kit. HPLC (Bio-RadVARIANT II) was used as comparison method.

RESULTS

During this study, we found 5 Hb variants never described by CE. (1) Hb Belfast migrates in “S” zone in CE but can bedifferentiated from HbS as migration position is 212 (214 for HbS), corresponding to the beginning of the zone. In HPLCmigrates after HbA2, in unknown zone. (2) Hb G-San José migrates in “F” zone. In HPLC, this variant elutes immediatelyafter HbS. (3) Hb J-Sardegna is located in the zone 12 whereas the minor peak is in the “D” zone. This variant co-eluteswith HbF in HPLC. (4) Hb Sassari migrates in “F” zone but cannot be confused with HbF: migration position is differentthan the HbF (173 against 178-191 for the HbF). The minor peak is located near the HbA2, indicated the presence ofan alpha variant. In HPLC elutes immediately after HbA2. (5) Hb Shelby on CE migrates in Z8 zone, between HbA andHbF. On HPLC elutes to late, after HbA2. During the study we also analyzed a sample with Hb Shelby without HbA(beta genotype: ß°39/ßShelby); we found a normal profile but without zone displaying, indicating a delay in the profilerecentering. This sample was mixed with normal control and re-analyzed. The new profile was now recentered: thepresumed HbA peak previously seen, is identified between HbA and HbF, in Z8 zone.

CONCLUSION

CE is a powerful technique allowing us to easily identify Hb variants. Separation is precise, especially if migrationposition related to the X-Axis is used. This technique can be implemented as a first-line screening test, keeping in mindthat a confirmatory testing is still required.*This work is dedicated to the memory and in honor of Prof. Renzo Galanello.

Poster Abstracts – EuroMedLab Paris 2015 – Paris, 21-25 June 2015 • DOI 10.1515/cclm-2015-5004 Clin Chem Lab Med 2015; 53, Special Suppl, pp S1 – S1450, June 2015 • Copyright © by Walter de Gruyter • Berlin • Boston S189

UnauthenticatedDownload Date | 5/27/15 10:16 AM

MONDAY - June 22th

Hemoglobinopathies


M053

IMPRECISION OF THE MIGRATION POSITIONS OF HEMOGLOBIN (HB) VARIANTS USING CAPILLARY ELECTROPHORESIS

j.h.m.j. Jean. Riou, Christian Godart, Henri Wacjman, Michel Goossens, Josiane Bardakdjian-michau.11Service de Biochimie Génétique Groupe Hospitalier Henri Mondor Créteil France

BACKGROUND-AIM

Capillary electrophoresis (CE) and ion-exchange High Performance Liquid Chromatography (HPLC) have becomemethods of choice for the screening of hemoglobin (Hb) disorders in routine laboratories. While retention time is usedin HPLC for the presumptive identification of the Hb variants, the migration time on CE is not commonly used andthe presumptive identification of the Hb variants is inferred from their electrophoretic mobility in migration zonesdefined by the manufacturer. These migration zones are usually thought to be less conclusive than retention timesince they can be quite wide and include several variant with close mobility. We assessed the imprecision of the CEmigration position on a selection of common Hb variants with the aim to value the use of the migration time for greaterdiscrimination between Hb variants of close mobility.

METHODS

Hb CE was performed on CAPILLARYS 2 instrument using the CAPILLARYS Hemoglobin(e) kit (Sebia, Lisses, France).Intra- and inter-assay precision of Hb variants migration position was assessed by analyzing 7 fresh samples (1 A/A, 1A/S, 1 A/C, 1 A/D, 1A/E, 1 A/Hb Hope and 1 Control), using 2 different lot numbers of buffer. Inter-individuals precisionwas assessed on 61 samples heterozygous for common Hb variant (20 A/S, 10 A/C, 10 A/E, 10 A/D and 11 A/Hb Hope).Mean values, standard deviations and coefficients of variation for the migration position of each Hb variant were thencalculated.

RESULTS

The intra- and inter-assay precision of the migration positions for the common Hb variants tested were shown to beexcellent, with CV <0.49% and <0.39%, respectively. The inter-individuals precision was also very good, with CV <0.62%.The variation of the migration position for all Hb variants tested was very low, comprised between +/- 1 unit on the x-axis. 200 samples from our collection stored at -80°C were then processed on the CAPILLARYS 2 in order to determinethe migration position of rare Hb variants.

CONCLUSION

The migration positions of Hb variants obtained by CE proved to be highly reproducible and it can be expected thatthese migration positions visualized on the X-axis be used rather than migration zones to refine the presumptiveidentification of Hb variants and increase their detection specificity.



MONDAY - June 22th

Myeloma

Biology of blood cancers

M198

BIOLOGICAL DIAGNOSIS OF MONOCLONAL IMMUNOGLOBULIN LIGHT CHAINS: BENEFIT OF URINE TESTING

V. Molinier-frenkel2, K. Belhadj3, T. Damy1

1APHP, Fédération de Cardiologie, Hôpital Henri Mondor, Créteil2APHP, Laboratoire d'Immunologie, Hôpital Henri Mondor, Créteil3APHP, Service Hémopathies Lymphoïdes, Hôpital Henri Mondor, Créteil

BACKGROUND-AIM

Two assays for serum free light chains (sFLC) quantification are commercially available. Although these assays arehighly sensitive, they measure total FLC (polyclonal and monoclonal) and give an indication of monoclonality when anabnormal κ/λ ratio is obtained. It has been proposed that sFLC assays can replace urine immunofixation (IF) for routinedetection. Several publications have now suggested that sFLC quantification is complementary to urine testing. Weshow here several clinical cases of our routine where sFLC alone could have been misleading and where urine IF showedthe Bence-Jones (BJ) protein.

METHODS

sFLC was performed by Freelite (The Binding Site), N-Latex FLC (Siemens) or both techniques. Urine IF was performedby Hydrasys Urine Profile (Sebia). Serum Protein Electrophoresis was run on capillary electrophoresis (Sebia) and serumIF on Hydrasis IF (Sebia).

RESULTS

We present herein 6 clinical cases from a retrospective analysis of our routine. The patients were either newly diagnosedor followed up for multiple myeloma or AL amyloidosis. In all the 6 cases, monoclonal FLC were clearly identified in urineby IF while the sFLC concentrations and κ/λ ratio from the sFLC assays were either normal or ambiguous (borderlineto the reference range).

CONCLUSION

These 6 cases illustrate the pitfalls of relying solely on the sFLC analysis in order to establish the presence ofmonoclonal FLC. They demonstrate the continued importance of urine IF for initial screening and follow up ofpatients for BJ protein. These results are consistent with the International Myeloma Working Group (IMWG) guidelinescautioning about the importance of urine electrophoresis (for peak quantification) and IF for response-to-treatmentdefinition and about its complementarity to the sFLC assay.



MONDAY - June 22th

Myeloma


M200

VALIDATION OF URINE MONOCLONAL PEAK QUANTIFICATION TOOL ON CAPILLARY ELECTROPHORESIS.

G. Proczek1, M. Baeza1, D. Simonin1, F. Robert1

1Sebia, Lisses

BACKGROUND-AIM

Urine monoclonal peak (MP) quantification is important for an appropriate disease management of patients withmonoclonal gammopathies. For Multiple Myeloma (MM), the International Myeloma Working Group (IMWG) hasdefined response-to-treatment criteria, based on the central tumoral biomarkers: the serum and urine MP. Thus,quantifying urine MP is crucial for an efficient disease management and response classification. With thousands ofinstalled capillary electrophoresis instruments worldwide, we decided to develop a “urine MP quantification tool” onboth Sebia Capillarys 2 (C2) and Capillarys 2 Flex Piercing (C2FP) instruments.

METHODS

Urine MP quantification was developed on the “Urine” program on both C2 and C2FP, available on Phoresis Core softwarerelease >8.51. Urines containing different concentrations of MP were processed. Obtained results were compared tothose obtained on HR3 program (Hydragel HR, Hydrasys 2 Scan). Precision studies were carried out on both urine MPand albumin quantifications: between assays, batches and preparations. We investigated linearity by serial dilutionsof urine with MP in a normal sample. Both limit of detection (LOD) and of quantification (LOQ) were determined.

RESULTS

The urine MP was quantified by both available software modes: “tangential” and “orthogonal”. Independently of thequantification method used, our results showed an excellent correlation of urine MP quantification between Capillarysand Hydrasys techniques (correlation coefficient of 0.999). Characterization of any MP can be done using the IT URINEtechnique on C2 or C2FP. The obtained inter-assay, inter-batch and inter-preparation precisions were excellent (meanCVs of 1.65%, 2.3% and 2.5% respectively; with a mean total CV of 2.5%). MP quantification on Capillarys was shownto be linear on the tested range (proteinuria ranging from 0 to 5.65 g/L) for both quantification methods. Finally, wedetermined urine peak’s LOD and LOQ: 20 mg/L and 150 mg/L respectively. Albumin quantification showed similarperformances.

CONCLUSION

We validated herein the new peak quantification tool on Urine program on both C2 and C2FP instruments. Clinicallaboratories can now use this new tool to accurately quantify urine peaks in order to comply with IMWG guidelines foran optimal MM disease management.



MONDAY - June 22th

Myeloma


M203

MORNING URINE AS AN ALTERNATIVE TO 24-HOUR URINE ELECTROPHORESIS? RENAL LESIONS ASSESSMENT,MONOCLONAL PEAK DETECTION AND QUANTIFICATION COMPARISONS

P. Boulard1

1Evolab, Thionville

BACKGROUND-AIM

Urine protein electrophoresis (UPE) and immunofixation (uIF) are central in the diagnosis and follow up of patientswith monoclonal gammopathies. They allow: proteinuria screening, urine peak typing and quantification. Peakquantification is an important criterion used for response to treatment assessment for multiple myeloma patients.International Myeloma Working Group (IMWG) guidelines recommend 24 hour (24h) urine testing. In front ofdifficulties of 24h urine collection, and since proof of analyzes on morning urine is lacking, we decided to comparedata from both morning and 24h urines.

METHODS

24h urine collection was done for 230 patients with morning urine collected separately in a small flask. Both urinetypes from all patients were analyzed on Hydragel Urine Profile (UP) (Sebia, Lisses) on Hydrasys 2 Scan instrument.Urine peak quantification was performed after scanning the ELP track of the UP gel. The percentage of concordancebetween morning and 24h urine was calculated for renal lesion typing and for monoclonal component typing.

RESULTS

Hydragel UP kit allows, with a single analysis, to screen proteinuria content, type renal lesions, type and quantifymonoclonal components. Among the 230 patients we analyzed, 45 presented a monoclonal component. We obtainedan excellent concordance (99%) between morning and 24h urines for proteinuria typing (physiological, glomerular,tubular, mixed and overload proteinuria). For monoclonal component typing, an excellent concordance was observedas well (99%). Urine peaks at the limit of the technique detection from only 3 patients were picked up only by oneof the two urine collection types (2 picked up on morning urine only and 1 picked up on 24h urine only). As for peakquantification, we observed a very good correlation between the two urine collection types on the tested range.

CONCLUSION

To our knowledge, no study has compared morning to 24h urine analysis for such number of patients and for proteinuriatyping, monoclonal component detection, typing and quantification. Our results clearly show that morning urine is agood candidate to replace 24h urines for detection and quantification of urine peak, as well as for renal lesion typing.Before switching to morning urine, further studies are needed to confirm these findings. Comparisons in the contextof patient follow up are still lacking.



MONDAY - June 22th

Capillarys 3


M045

FIRST EVALUATION OF THE NEW AUTOMATED CAPILLARYS 3 TERA SYSTEM (SEBIA) FOR SERUM PROTEINELECTROPHORESIS

S. Baglioni1, A. Celli1, A. Menicacci1, L. Guidi1, P. Casprini11Central Laboratory Unit, S. Stefano Hospital, Prato.

BACKGROUND-AIM

We evaluated for the very first time the reliability and the analytical performances of the new fully automated capillaryelectrophoresis system CAPILLARYS 3 TERA (Sebia, France) for serum protein electrophoresis (SPE) testing. Resultscomparison was performed with our capillary electrophoresis instrument CAPILLARYS 2 (Sebia). We also valued theease-of-use and robustness of the CAPILLARYS 3 TERA within the framework of real routine work conditions.

METHODS

A total number of 2252 samples sent to our laboratory for routine SPE tests were analyzed the same day on CAPILLARYS3 TERA and CAPILLARYS 2. The correlation between the 2 systems has been assessed. For the sensitivity study, 3 patientsamples with different isotypes of monoclonal components (MCs) (ranged from 3.4g/L to 4g/L) were diluted in a normalserum (1:2, 1:4, 1:8, 1:16, 1:32) and analyzed on both systems; Sensitivity has been assessed by checking the highestdilution for which the residual monoclonal peak was still visible. The precision has been tested on 4 pools of freshsamples showing different electrophoretic patterns that have been analyzed 6 times a day during 6 consecutive days(n=36); the mean, SD and CV were calculated for each fraction.

RESULTS

The results of the method comparison for protein fractions quantification show excellent correlation withoutsignificant differences between methods for each fraction (Albumin: y=0.9722x+1.4363, r=0.99; Alpha-1: y=0.9946x+0.137, r=0.99; Alpha-2: y=0.9505x+0.8341, r=0.99; Beta-1: y=0.9767x+0.1103, r=0.95; Beta-2: y=0.9629x+0.0694,r=0.99; Gamma: y=0.9761x+0.2784, r=0.99). The detection limit for MCs on CAPILLARYS 3 TERA was around 0.25 g/L foreach monoclonal component, similar to the one of CAPILLARYS 2. Total CV was <1% for albumin and <5% for globulins.

CONCLUSION

The evaluation of the CAPILLARYS 3 TERA over a large number of samples proved to be a very reliable instrument. Thecorrelation was excellent when compared to CAPILLARYS 2, with electrophoretic patterns being strictly identical. Thesystem has proved to be able to correctly detect all monoclonal components. Furthermore, the enhanced automationand high throughput of the system permits to save a substantial amount of time, reducing significantly the resultturnaround time.



MONDAY - June 22th

Capillarys 3

Analytical technologies: flow cytometry, mass spectrometry, multiplexed analysis, metabolome analysis...

M092

CAPILLARYS 3 TERA: FIRST EVALUATION OF A HIGH THROUGHPUT CAPILLARY ELECTROPHORESIS INSTRUMENT FORHIGH RESOLUTION HBA1C SEPARATION

M. Gutsche1, M. Seydel1, B. Meissner1, M. Praus1

1DIAGNOSTICUM

BACKGROUND-AIM

The efficient measurement of HbA1c will become of increasing importance as the prevalence of type II diabetesmellitus increases. To this end the new CAPILLARYS 3 TERA instrument (Sebia, France) for high throughput capillaryelectrophoresis HbA1c testing has been developed. In this study we evaluated the instrument with respect to trueness,precision and correlation to CAPILLARYS 2.

METHODS

The trueness was assessed on 6 samples with target values assigned in one approved laboratory of the IFCC NetworkLaboratories for HbA1c. The precision was tested for 12 days in 4 pools of samples stored in aliquots at -80°C. Foreach pool, 2 determinations for each of the 12 capillaries were performed (n=24). For method comparison, 3129 HbA1croutine samples (min=4.2%; max=14.4%) were analyzed on both CAPILLARYS 3 TERA and CAPILLARYS 2 systems.

RESULTS

For the trueness study, the measured HbA1c values were slightly below the target values with a bias ranging from 0%to 0.2% (0.3 to 1.7 mmol/mol). At a mean HbA1c concentration of 4.4% (24.6 mmol/mol) the coefficient of variation(CV) = 1.3%, at 6.3% (45.4 mmol/mol) CV= 1%, at 7.5% (58.5 mmol/mol) CV= 0.6% and at 10.5% (91.3 mmol/mol) CV=0.9%. The regression line followed the equation y=0.996x+0.029 with a correlation coefficient r=0.994 and a meanbias= 0.00%. The implementation of capillary electrophoresis for routine testing of HbA1c has considerably improvedour workflow. Due to the high resolution of different hemoglobins it is possible to rely on the automated data analysisto reduce the time required by laboratory technicians to supervise the analytical process. In 2014, samples of 133140individuals were analyzed for HbA1c. A total of 280 abnormal hemoglobin profiles were detected indicating that theprevalence is in the order of 0.21%. For these patients further diagnostic testing is necessary to arrive at a correctinterpretation of the HbA1c result and improve clinical practice.

CONCLUSION

HbA1c testing by capillary electrophoresis is an accurate method improving workflow and quality of laboratory analysis.The newly developed CAPILLARYS 3 TERA is a high throughput multiparameter laboratory automate processing up to70 samples/hour with comparable performances to the well established CAPILLARYS 2 system.



MONDAY - June 22th

Capillarys 3


M113

SUITABILITY AND PERFORMANCES OF THE NEW CAPILLARYS 3 TERA FOR HIGH VOLUME TESTING ACTIVITY

N. Baidjibay1, M. Van De Loo1, D. Simonin2

1Laboratoire Bio7, 91090 Lisses, France2SEBIA, Parc Technologique Léonard de Vinci, CP 8010 Lisses, 91008 EVRY , France

BACKGROUND-AIM

Capillary electrophoresis (CE) is a high resolution separation method that was recently adapted for the measurementof HbA1c. After one year experience in our laboratory using this method for routine HbA1c testing, we evaluated theperformances of the new CAPILLARYS 3 TERA instrument (Sebia, France), a multiparameter CE instrument with 12capillaries in parallel.

METHODS

This evaluation was conducted during 8 weeks over 8.000 samples sent to the laboratory for routine HbA1c testing and1.500 samples for Serum Protein Electrophoresis (SPE), analyzed on the CAPILLARYS 3 TERA to assess its robustnessand ease-of-use. The comparison was based on the correlation between our current CE instruments (CAPILLARYS 2Flex Piercing, Sebia, France) and the CAPILLARYS 3 TERA on 863 whole blood samples covering a wide range of HbA1cvalues. The mean deviation between the 2 systems was calculated at 3 different HbA1c levels: 30, 60 and 90mmol/mol. The between run precision was evaluated on 2-levels daily HbA1c controls (Sebia, France) that were processed onthe 12 capillaries during 40 consecutive days (n=480).

RESULTS

The correlation between CAPILLARYS 2 Flex Piercing and CAPILLARYS 3 TERA proved to be excellent with a linearregression y= 1.0006x + 0.0093 and a mean bias= 0.01% for HbA1c results expressed in NGSP units (min=4.1%;max=13.4%), and y= 1.0006x + 0.1144 and a mean bias= 0.1mmol/mol for HbA1c results expressed in IFCC units(min=21mmol/mol; max=123mmol/mol;). The coefficient of correlation was r= 0.993. The mean deviations at 30, 60and 90mmol/mol were successively 0.15, 0.17 and 0.18mmol/mol. The between-run CVs were 1.94% and 1.13% for theHbA1c Control Level 1 (mean value= 33mmol/mol and 5.2%, respectively); the between-run CVs were 1.35% and 1.00%for the HbA1c Control Level 2 (mean value= 68mmol/mol and 8.4%, respectively).

CONCLUSION

This extensive evaluation of CAPILLARYS 3 TERA instrument over nearly 10.000 samples showed that this newmultiparameter instrument is reliable, easy to use and can absorb high volume testing activity thanks to its fullautomation and high throughput. We found excellent correlation and precision when compared to the CAPILLARYS 2Flex Piercing with resolution and separation for HbA1c and SPE profiles being similar.



MONDAY - June 22th

Gel


M118

ASSESSMENT OF LC-MRM MASS SPECTROMETRY AND SEMI-AUTOMATIC ISOFOCUSING APPROACHES FOR THEDETERMINATION OF APOLIPOPROTEIN E PHENOTYPING IN HUMAN SERA

C. Hirtz2, J. Vialaret2, G. Nouadje3, S. Schraen-maschke1, P. Benlian4, S. Mary4, L. Tiers2, P. Bros2, C. Delaby2, A. Gabelle2, S.

Lehmann2

1CHU de Lille, Centre de Biologie Pathologie; Université Lille-Nord de France; INSERM U837; Lille2CHU de Montpellier and University of Montpellier3SEBIA, Evry4U4M Lille

BACKGROUND-AIM

Apolipoprotein E (Apo E) is a 36 Kda glycoprotein involved in lipid transport. It exists in 3 major isoforms: ε2, ε3 andε4. Apo E status is known to be a major risk factor for late-onset Alzheimer’s and cardiovascular diseases. Genotypingis commonly used to obtain Apo E status but can show technical issues with ambiguous determinations. Phenotypingcan be an alternative, not requiring genetic material. We evaluate the ability to accurately type Apo E isoforms by 2phenotyping tests in comparison with genotyping.

METHODS

Two phenotyping techniques were used: (1) LC-MS/MS detection of 4 Apo E specific peptides (6490 Agilent triplequadripole): After its denaturation, serum was either reduced and alkylated, or only diluted, and then trypsin digested.Before analysis, desalting, evaporation and resuspension was performed. (2) Isofocusing electrophoresis: serum wasneuraminidase digested delipidated and electrophoresed on agarose gel using Hydrasys 2 Scan instrument (Sebia,Lisses, France). An immunofixation (Apo E primary antibody - HRP secondary antibody - TTF1/TTF2 staining) allowedthe visualization of Apo E bands. The results of the two techniques were compared to genotyping.

RESULTS

Sera from 35 patients previously genotyped were analyzed with the 2 phenotyping techniques. 100% concordancebetween both phenotyping assays was obtained for the tested phenotypes (ε2/ε2, ε2/ε3, ε2/ε4, ε3/ε3, ε3/ε4, ε4/ε4).When compared to genotyping, 3 samples were discordant. After reanalyzing them by both phenotyping tests andDNA sequencing, 2/ 3 discrepencies were confirmed. Those can be explained by variants or rare Apo E alleles or byunidentified technical issues. 102 additional samples were tested on LC-MS/MS only and compared to genotyping. Thedata showed 100% concordance.

CONCLUSION

Our 2 phenotyping methods represent a valuable alternative to genotyping. LC-MS/MS has the advantage of beingfully specific, with absolute quantification of the different isoforms and can be considered as a reference method.Sebia isofocusing technique was concordant with LC-MS/MS. Plus, it is a rapid, semi-automated assay that can be easilyimplemented in clinical laboratories.



MONDAY - June 22th

Gel

Inherited disorders, metabolic disorders, rare diseases

M425

SCREENING FOR ALPHA-1 ANTITRYPSIN DEFICIENCY USING DRIED BLOOD SPOTS

C.C. Colette1, Z. Farid2, O. Marie-françoise2, B. Malika2

1Laboratoire d'immunologie Centre Hospitalo Universitaire Lyon Sud2Laboratoire de biochimie et biologie moléculaire Centre Hospitalier Regional Universitaire Lille

BACKGROUND-AIM

Alpha- 1 Antitrypsin (A1AT) deficiency is a genetic disorder resulting in low levels of serum A1AT. It is associatedwith lung deteriorations and/or liver injuries and significantly underdiagnosed. Facilitating an earlier diagnosis of thisdeficiency might allow a better management of the lung disease. We have thus developed and standardized withinthree French laboratories a method for quantifying and phenotyping A1AT extracted from dried blood spots (DBS).

METHODS

DBS were prepared by spotting EDTA-anticoagulated whole blood on a filter paper which was air dried and storeduntil use. Paper disks were punched from the blood spots and eluted with water for the measurement of A1ATlevels and glycine 1M pH 7.4 buffer, for the phenotyping. Automated immunonephelemetric and immunoturbidimetrictechniques were set up for the quantitation of low levels of A1AT. Phenotyping was performed on ready-to-use agarosegels (Hydragel 18 A1AT Isofocusing; Sebia) run on a semi-automated system (Hydrasys System™; Sebia) with a specificprogramme devoted to diluted samples designed by the manufacturer (Sebia).All the results obtained with DBS within each laboratory were compared (1) to the results obtained with thecorresponding plasma and (2) to the results obtained in the other laboratories. The correlation between those resultswas studied with linear regression analysis using Statview™ and Excel™ (Microsoft) softwares.

RESULTS

90 DBS issued from 90 patients were studied. The correlation coefficients between the concentration of A1AT in DBSand in plasma were 0.965, 0.970 and 0.953 within the 3 laboratories. The regression lines issued from the comparisonbetween the laboratories appear to merge as one single line. So, for a target value of 0.500 g/L, the results obtainedwere between 0.50 and 0.54 g/L. A 100% of concordance was obtained for the interpretation of the phenotypes.

CONCLUSION

This study shows that the results obtained with DBS are highly correlated with those obtained with venous bloodsamples. It becomes then possible to undertake a large scale screening program of A1AT deficiency relying on a kitdesigned to perform a capillary blood sampling on filter paper.



MONDAY - June 22th

CDT

Evidence-based medicine, Lab medicine practice guidelines, decision making

M391

APPLICABILITY OF THE BIOMARKERS OF CHRONIC ALCOHOL ABUSE IN THE STRATEGIES TO IMPROVE TRAFFIC ANDWORKPLACE SAFETY

F. Tagliaro1, F. Bortolotti1, A. Bertaso1, L. Romeo2

1Unit of Forensic Medicine, Dept. of Public Health and Community Medicine, University of Verona, Verona, Italy2Unit of Occupational Medicine, Dept. of Public Health and Community Medicine, University of Verona, Verona, Italy

BACKGROUND-AIM

If the relationship between blood alcohol concentration (BAC) and road accident occurrence has extensively beenstudied, less attention has been devoted to the study of the “predictive value” of the biomarkers of chronic alcoholabuse. Aim of the present work was the investigation of the hypothesis of association of one or more of thesebiomarkers with the occurrence of traffic accidents, also among professional drivers.

METHODS

Subjects admitted to hospital for accident-related injuries (InjDr) (N= 468) were divided in two groups on the basis ofthe BAC (≤ 0.5 and > 0.5 mg/mL); a group of control subjects (drivers with no record of recent accidents, N=236) wasalso included. GGT and CDT in serum were determined in by using enzymatic analysis and HPLC, respectively.EtG in hair was studied by using GC/QQQ-MS in cases of fatal road accidents (N= 60).The association of the increase of these biomarkers with the occurrence of alcohol related traffic accidents (i.e. withBAC > 0.5 mg/mL) was verified by using statistical methods.CDT analysis was also applied to check the fitness-to-work in a group of professional bus drivers (n=503).

RESULTS

Using a cut-off of 1.9%, 36 of 100 InjDr with BAC > 0.5 g/L, showed elevated CDT. Comparing this subgroup with thecontrol group (CDT positives 0.4%), the Odds ratio was as high as 132, with a p value well below the 0.001 threshold(Fisher’s test).On the other hand, only 7 out of 368 InjDr with BAC ≤ 0.5 g/L (1.9%) showed elevated CDT concentrations, resulting notsignificantly different from the control group (Odds ratio).GGT proved also significantly elevated in the InjDr, but with a lower degree of statistical significance in comparisonwith CDT.Finally, EtG in hair was found increased in 44% of the alcohol related traffic fatalities (cut-off 30 pg/mg), but also in17% of the non-alcohol related accidents.The application of CDT analysis in the assessment of the fittness-to-work of professional drivers showed a low but notnegligible prevalence of alcohol abusers (about 2%), which were directed towards psychological counselling with rapidnormalization of the CDT values.

CONCLUSION

The use of CDT for the assessment of the fitness to drive is objectively justified. GGT and EtG show a promising potentialin this field.The use of biomarkers of chronic alcohol abuse, and particularly CDT, has proved also useful in the assessment of thefitness-to-work in case of safety sensitive jobs.



TUESDAY - June 23th

HbA1c

Diabetes

T009

IN VIVO EFFECT OF CARBAMYLATED HEMOGLOBIN ON HBA1C MEASUREMENT BY CAPILLARY ELECTROPHORESIS

J.M. Jou Turalles 1, M.E. García1

1Servei de Hemoterapia I Hemostasia, Laboratori CORE, Hospital Clínic de Barcelona

BACKGROUND-AIM

HbA1c is widely used as the gold standard marker to assess glycemic control. Carbamylated hemoglobin (cHb) is a welldescribed interference found in patients with chronic renal failure (CRF) interfering with HbA1c measurement due toits incomplete separation with HbA1c with some ion-exchange HPLC techniques. The aim of our work was to assesspotential interferences of in vivo cHb on HbA1c quantification by capillary electrophoresis (CAPILLARYS 2 Flex Piercing,Sebia) in CRF patients.

METHODS

The effect of in vitro cHb on HbA1c measurement by CAPILLARYS 2 Flex Piercing was assayed by incubating red bloodcells with Potassium Cyanate. The effect of in vivo cHb was assayed on samples from patients with CRF (diabetic andnon-diabetic) by correlation studies with an HPLC method (G8, Tosoh) that has been previously shown not to haveinterference with in vivo cHb.

RESULTS

cHb resulting from in vitro incubation with Potassium Cyanate showed no impact on HbA1c measurements usingCapillarys 2 Flex Piercing. Morevover, HbA1c quantification by Capillarys 2 Flex piercing was perfectly correlated withthe one obtained on HPLC G8 for both CRF and non-CRF patient groups.

CONCLUSION

Sebia Capillarys 2 Flex Piercing analyzer provides a clear HbA1c electrophoregram easy to interpret, with nointerference of in vivo cHb. It can be considered a suitable system for HbA1c measurement in laboratories. To ourknowledge, this is the first study demonstrating the absence of interference from in-vivo cHb on HbA1c measurementby Capillarys 2 Flex Piercing.



TUESDAY - June 23th

HbA1c

Diabetes

T014

INFLUENCE OF FETAL HEMOGLOBIN ON HBA1C MEASUREMENT USING 3 DIFFERENT CAPILLARY ELECTROPHORESISINSTRUMENTS

M. Baeza1, G. Deschamps1, F. Hologne1, D. Simonin1, F. Robert1

1Sebia, Lisses

BACKGROUND-AIM

Elevated Fetal Hemoglobin (HbF) has been reported to interfere with some assay methods for HbA1c. There are manyclinical conditions associated with elevated HbF (>1%), such as β-thalassemia, pregnancy, leukemia, and hereditarypersistence of fetal hemoglobin. Using capillary electrophoresis (CE) method for HbA1c measurement, HbF is clearlyseparated from HbA1c fraction. But as HbF migrates closely to HbA0 fraction, and because HbA0 fraction is included inthe calculation formula used to measure HbA1c value, an interference of HbF in the HbA1c measurement by CE methodmight be suspected. Here we evaluated the influence of HbF at different levels in the measurement of HbA1c by severalcapillary electrophoresis instruments.

METHODS

6 adult whole blood samples showing different HbA1c levels (from 32 to 138 mmol/mol) were serially diluted with acord blood sample with elevated HbF (>90%) to get different HbF levels (from 1.5% to 23.7 %). For all samples, theHbF level was determined on the CAPILLARYS 2 Flex Piercing Hemoglobin(e) technique (Sebia, France). Each nativeand diluted sample was then split in 4 aliquots. 3 aliquots were run on 3 routine CE instruments for HbA1c testing:MINICAP Flex Piercing (MCF), CAPILLARYS 2 Flex Piercing (C2FP) and CAPILLARYS 3 TERA (C3T) (Sebia, France). 1 aliquotwas analyzed on a NGSP secondary reference method (NGSP) that is known to be free of interference from HbF, usedas the comparative method.

RESULTS

Methods comparison showed a good correlation between each CE method and the NGSP method when all native anddiluted samples were analyzed (linear regression y= 0.951x -0.382 and a coefficient of correlation R=0.998 for MCF; y=1.012x -2.904 and R=0.998 for C2FP; y= 0.980x -1.175 and R=0.997 for C3T). The mean deviations at 30, 60 and 90mmol/mol were successively 1.9, 3.3 and 4.8mmol/mol on MCF; 2.6, 2.2 and 1.9mmol/mol on C2FP; 1.8, 2.4 and 2.9mmol/molon C3T, showing no major deviation from the comparative method.

CONCLUSION

This evaluation showed that none of the CE methods tested is subject to interference with HbF up to 23% on themeasurement of HbA1c. MINICAP Flex Piercing, CAPILLARYS 2 Flex Piercing and CAPILLARYS 3 TERA can reliably reportaccurate HbA1c results in case of elevated HbF.



TUESDAY - June 23th

HbA1c

Diabetes

T025

APPLICATION OF CAPILLARYS 2 FLEX PIERCING HBA1C MEASUREMENT SYSTEM FOR ALPHA AND BETA-THALASSEMIASCREENING

L. Ji11DEPARTMENT OF LABORATORY MEDICINE, PEKING UNIVERSITY SHENZHEN HOSPITAL, SHENZHEN

BACKGROUND-AIM

HbA1c measurement can be affected by hemoglobin (Hb) disorders such as hemoglobinopathies and thalassemia.Recent studies showed that β-thalassemia could interfere with the measurement of HbA1c, similar to other knownanalytical interferences (e.g., carbamylatedHb, fetal Hb, or labile HbA1c). Beside the analytical interference, Hb variantsand β-thalassemia have been reported to be associated with a significant decrease in the lifespan of red blood cells.Thus, the HbA1c value in patients carrying such Hb abnormalities may be misinterpreted if compared to the usualreference cut-off value. Therefore, the assays which are incapable of identifying thalassemia might report misleadingHbA1c values for thalassemic patients.

METHODS

Whole blood samples from 258 healthy adult patients without Hb disorders, 80 α-thalassemia adult patients with --sea/αα genotype, and 225 adult patients with β-thalassemia trait from 10 common genotypes were measured usingCapillarys2 Flex Piercing HbA1c system and Capillarys2Hemoglobin(e)system (Sebia, Lisses, France). The samples withiron deficiency were excluded.Receiver operating characteristic curve (ROC) were performed to determine the HbA2cut-off values for screening α and β thalassemia.

RESULTS

For screening samples with α thalassemia, the optimal HbA2 cut-off value is 2.35% (area under curve (AUC) 0.969,sensitivity 88.1% and specificity 92.5%), and 2.55% (AUC 0.951, sensitivity 90.9% and specificity 88.6%) for theCapillarys 2 Flex Piercing HbA1c system and the Capillary2Hemoglobin(e) system, respectively. For screening sampleswith β thalassemia trait, the optimal HbA2 cut-off value is 3.38% (AUC 0.994, sensitivity 100% and specificity 98.2%),and 3.75% (AUC 0.993, sensitivity 98.2%% and specificity 100%) for theCapillarys 2 Flex Piercing HbA1c system andthe Capillary2 Hemoglobin(e) system, respectively.

CONCLUSION

The Capillarys 2 Flex Piercing HbA1c system can separate and accurately measure HbA2 values for screeningthalassemia besides reporting accurate HbA1c value, which provides valuable information to clinicians for theinterpretation of the HbA1c result in patients with thalassemia trait.



TUESDAY - June 23th

HbA1c

Diabetes

T037

HBG-COUSHATTA: AN UNEXPECTED DISCOVERY DURING HBA1C MEASUREMENT

X. Cheng1, M. Li1, J. Wu1, W. Su1

1Department of Clinical Lab, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking UnionMedical College, Beijing 100730, China

BACKGROUND-AIM

Hemoglobin (Hb) A1c is widely used as the gold standard for blood glucose level monitoring in diabetic patients.Various methods are being used in clinical laboratories for HbA1c measurement. However, the presence of Hb variantsmay interfere with HbA1c measurement.

METHODS

A 55-year old woman with type 2 diabetes and hyperlipidemia was prescribed with a long-term therapy of oral anti-diabetic medication, and monitoring of blood glucose and HbA1c was performed routinely. HbA1c measurment on theCE-HPLC showed that there was an Hb variant which interfered with the quantitation of HbA1c.To confirm the presence of an Hb variant in this patient, we profiled Hb through capillary electrophoresis. Moreover,DNA sequencing of the Hb β-chain gene (HBB) was performed to define its genotype.To evaluate the interference of the Hb variant on different methods of HbA1c measurements, we assessed the level ofHbA1c in this patient by capillary electrophoresis and tandem HPLC-capillary electrophoresis (LC/CE), one of the twoIFCC Reference Methods.

RESULTS

Hemoglobin electrophoresis results by capillary electrophoresis showed that proportions of HbA2 and HbA were 2.6%and 54.1%, respectively. As expected, an Hb variant was confirmed by capillary electrophoresis, with the presence ofan abnormal peak in the HbD zone, with a proportion of 43.3%.Further DNA sequencing of the Hb β-chain gene (HBB) revealed an alteration at codon 22 (GAA to GCA), resulting inan amino acid change from glutamic acid to alanine, which was previously named as HbG Coushatta. This patient washeterozygous Hb A/G, and the unknown peak detected in CE-HPLC was HbG0.The measurement of HbA1c through capillary electrophoresis yielded a result of 7.7% (60.6mmol/mol). Tandem HPLC-Capillary electrophoresis (LC/CE), reported an HbA1c value of 6.7% (49.7 mmol/mol). The calculation result of HbA1c,HbG1c, the sum of HbA1c and HbG1c was 7.7% (60.7mmol/mol), 6.7% (49.7mmol/mol) and 7.3% (56.3mmol/mol)respectively.

CONCLUSION

In conclusion, this case tells us that elimination of Hb variant interference is critical to achieve an accurate resultof HbA1c. High resolution methods as capillary electrophoresis and low interference methods as boronate affinitychromatography might be preferred in the regions with high prevalence of Hb variants. And staffs in clinical laboratoryshould pay more attention to analyze raw data of the test and find problems of the analysis.



TUESDAY - June 23th

HbA1c

Diabetes

T038

ACCURACY EVALUATION OF FIVE ROUTINE GLYCOSYLATED HEMOGLOBIN ASSAYS IN PATIENTS WITH HEMOGLOBINVARIANTS

Y.M. Yun2, S. Chun4, K.C. Kwon1, M. Ji2, S. Jeon3, J. Song3

1Department of Laboratory Medicine, Chungnam National University Hospital, Daejeon, South Korea2Department of Laboratory Medicine, Konkuk University School of Medicine, Seoul, South Korea3Department of Laboratory Medicine, Seoul National University College of Medicine, Seoul, South Korea4Department of Laboratory Medicine, University of Ulsan College of Medicine and Asan Medical Center, Seoul, South Korea

BACKGROUND-AIM

Hemoglobin A1c (HbA1c) is a widely used biomarker for the monitoring of glycemic status and also used for thediagnosis of diabetes. However, accurate quantification of HbA1c has been a challenge in patients with Hb variants orother interferences. We evaluated the impact of various Hb variants on five routine glycosylated hemoglobin assays(1 capillary electrophoresis assay, 1 immunoassay, and 3 high-performance liquid chromatography assays) comparedwith the reference measurement using LC-MS/MS.

METHODS

The whole blood samples showing the flags (suspected variant peak, duplex peak, high HbF, high labile HbA1c andlow HbA0) on routine HbA1c assays were collected between September 2013 and August 2014 in Korean population.Twenty normal samples without flag were collected as controls. Samples were assayed using the LC-MS/MS andfollowing glycohemoglobin analyzers: Sebia CAPILLARYS 2 Flex Piercing (CAP2 FP); Roche Cobas Integra (Roche); Bio-Rad Variant II turbo (Bio-Rad); Arkray ADAMS HA-8180 (Arkray); Tosoh G8 (Tosoh). Hb electrophoresis (HbEP) wasperformed for all samples.

RESULTS

Among 87 samples with flag and 20 controls, 44 samples showed normal Hb pattern in HbEP analysis and 63 samplesshowed Hb variants with the most frequent of HbD zone (73%), followed by HbF zone (19%). The mean absolute bias[(routine assay result (HbA1c unit, %)) – (LC-MS/MS result)] in Hb variants group (N=63) were -0.23 (CAP2 FP), -0.33(Roche), -0.55 (Bio-Rad), -1.09 (Arkray) and -1.37 (Tosoh) with the mean relative % bias ranged from -4.5% to -23.2%.In normal Hb pattern group (N=44), the mean absolute bias was ranged from -0.15 to -0.39 (HbA1c unit, %) with themean relative % bias of -1.6 to -4.6%.

CONCLUSION

The HbA1c results of 5 routine assays in patients with Hb variants showed negative bias compared with those of LC-MS/MS and the degree of the relative % bias (ranged from -4.5% to -23.2%) showed quite different among the assays.However, in patients with normal Hb pattern, the relative % bias of 5 routine assays was all within 5%.



TUESDAY - June 23th

HbA1c

Diabetes

T041

EVALUATION OF SEBIA CAPILLARYS 2 FLEX PIERCING FOR THE MEASUREMENT OF HBA1C AND STRUCTURAL VARIANTSOF HEMOGLOBIN DETECTION : ONE YEAR OF EXPERIENCE AT ROUEN UNIVERSITY HOSPITAL.

V. Brunel2, A. Lahary1, A. Chagraoui2, J. Wils2, G. Hue2

1Hematology Laboratory, Rouen University Hospital, Rouen, France2Laboratory of Medical Biochemistry, Rouen University Hospital, Rouen, France

BACKGROUND-AIM

Glycated hemoglobin (HbA1c) is considered the gold standard for assessing the compensation and treatment ofdiabetes. In addition, identification of variants hemoglobin during HbA1c measurement is not rare, and requiresscreening expertise by the clinical pathologist. Recently, a new HbA1c assay, using capillary electrophoresistechnology, has been introduced in clinical laboratories. We implemented this assay for HbA1c measurement in ourlaboratory in December 2013.

METHODS

We analysed whole blood sample with the Capillarys 2 Flex Piercing (C2FP, Sebia, Lisses, France) and Tosoh G7 (TosohCorporation, Tokyo, Japan) an high-performance liquid chromatography analyser previously used in our laboratory.We evaluated the analytical performances of the C2FP (intra-assay, between assays, linearity and bias). Having usedthe C2FP for one year in our laboratory, we sought to determine the extent to which results might be affected by thepresence of structural variant of hemoglobin.

RESULTS

Intra-assay and between-assays CV were respectively lower than 3.0% and 2.6% (IFCC units) or lower than 1.4% and1.7% (NGSP units). The linearity was excellent. There was a good concordance between results of C2FP and TosohG7 : HbA1c[C2FP] = 1.000 x HbA1c[Tosoh G7] – 0.100 (n=44). In 2014, the measurement of HbA1c was performed inapproximately 11500 samples in our laboratory. During this one year period, we observed 139 electropherograms witha structural variant of hemoglobin which was a discovery in 26 patients. Twenty five were identified. Most commonhemoglobin variants observed were : HbS (64%) HbC (28%) HbD (3%). Other variants were observed (<1%) : HbE, HbTatras, Hb P-Nilotic, Hb Hope, Hb Abruzzo, Hb Athens-Georgia. One case is still under investigation. C2FP softwaredoes not allow HbA1c calculation because of an haemoglobin variant peak partially eluating in HbA0 zone in two cases.Manual off-line recalculation of HbA1c value was possible only for one case.

CONCLUSION

In conclusion, the present evaluation has shown that the analytical performance of C2FP for measurement of HbA1cfulfils the quality criteria requested for clinical use. HbA1c measurement by C2FP is not analytically altered by thepresence of the most common variants of hemoglobin which we encountered.



TUESDAY - June 23th

HbA1c

Diabetes

T042

EVALUATION OF THE HBA1C KIT ON SEBIA MINICAP FLEX PIERCING

S. Daniel2, R. Emilie1, U. Lydie2, C. Bernard1

1Hôpital Bel Air CHR Metz-Thionville, 57100 Thionville (F)2Laboratoire National de Santé, Génétique et Biologie Moléculaire, L3555 Dudelange (L)

BACKGROUND-AIM

HbA1c is a valuable tool to monitor and diagnose diabetes mellitus and can be measured via different analyticaltechniques including capillary electrophoresis. Here we report the results of the evaluation of MiniCap Flex Piercing, anew capillary electrophoresis analyzer for the separation and the quantification of HbA1c

METHODS

The aim of this study was to assess the performance of the MiniCap Flex Piercing for routine HbA1c quantification.A systematic evaluation of the HbA1c Kit on the Minicap Flex Piercing has been undertaken in an internationalmulticentric study. The analytical performances of the method as well as the influence of the most frequent analyticalinterferences (i.e., labile HbA1c and hemoglobin variants) have been studied. The broadly evaluated HbA1c Kit on theCapillarys 2 Flex Piercing has been used as a comparative reference method.

RESULTS

The Minicap Flex Piercing was compared to the reference method and a strong interassay correlation was achieved(R>0.99), with high resolution profiles being similar. Between-run CVs were lower than 2.02% (IFCC units). Linearity wasdetermined over a range of 21 to 127 mmol/mol of HbA1c, and a high correlation between theoretical and observedvalues (R>0.99) was achieved. The use of samples with target values assigned in approved laboratories of the IFCCNetwork Laboratories for HbA1c indicated a good accuracy of the method, with a low bias ranging from -0.2% to -0.05%.The correlation between Capillarys 2 Flex Piercing and MiniCap Flex Piercing for samples with structural hemoglobinvariants (e.g. S, C and E) or β-thalassemia trait was excellent (R>0.99, mean bias of -0.11%), suggesting that thesehemoglobin disorders do not affect the HbA1c measurement. Another typical interference, labile HbA1c did not alterthe accuracy of the determination of HbA1c values.

CONCLUSION

The study shows that the analytical performances of the MiniCap Flex Piercing analyzer for HbA1c assay arein accordance with the quality criteria required for clinical use, the instrument can thus be recommended forimplementation in clinical laboratories for routine practice.



TUESDAY - June 23th

HbA1c

Diabetes

T050

COMPARISON OF THE "HEMOGLOBIN(E)" AND "HBA1C" KITS FOR HBA2 AND HBS QUANTITATION’S ON THE CAPILLARYS2 FLEX PIERCING DEVICE

J. Philippe1, B. Joëlle1, A. Vincent1, L. Philippe1, F. Cécile1

1Laboratoire de Biochimie et de Biologie Moléculaire, Unité de Pathologie Moléculaire du Globule Rouge, Hôpital Edouard-Herriot (Hospices Civils de Lyon), CHU de Lyon, Lyon, France.

BACKGROUND-AIM

Capillary electrophoresis (CE) and cation-exchange HPLC are techniques of choice to detect hemoglobin (Hb) disordersbut, at least in France, the general practitioner rarely asks for this screening without any clear clinical context.Comparatively, the HbA1c dosage is much more prescribed and is performed by most of laboratories. We thus evaluatedthe possibility to use a CE HbA1c technique to quantify at the same time both the HbA2 and HbS fractions.

METHODS

HbA2 and HbS were measured both on the "Hemoglobin(e)" and the "HbA1c" kits of the Capillarys 2 Flex Piercing (Sebia)on 128 whole bloods with the following Hb genotypes : 9 heterozygous ß-thalassemia (thal), 1 homozygous ß-thal.with 24% HbF, 3 heterozygous δ-thal. or δ-globin variant, 8 A/A, 3 A/E, 29 A/S, 4 S/ß0-thal, 9 S/C, 1 S/D, 60 S/S beforeor after blood transfusion and 1 Hb H disease. For HbS, fidelity and intermediate repeatability were also comparedbetween the two kits.

RESULTS

The HbA2 correlation was excellent (r² = 0.99) with a systematic mean negative bias of -0.4% (absolute value) for the"HbA1c" kit (IC95 = [-0.8%; 0%]). Interestingly, this bias appeared to be slightly affected by the ß-globin genotype (-0.6to -0.8% for the heterozygous ß-thal patients versus -0.2 to -0.4% for the other genotypes). Consequently, with a cut-off point of 2.8% HbA2, all the heterozygous ß-thal could be detected with the "HbA1c" kit. The only important biasbetween the two kits concerned the patient with 24% HbF.The HbS correlation was also excellent (r² = 0.9949) but also revealed again a slight negative bias of -2% (in HbS unit)between the two kits (IC95 [-5.0%; 1.3%]) independently of the ß-globin genotype. The coefficients of variation (CVs)for repeatability (0.3 to 1.1% for the 2 kits) and intermediate fidelity (0.5 to 1.5% for the "HbA1c" kit and 0.6 to 1.6%for the "Hemoglobin(e)" kit) were very good.

CONCLUSION

The "HbA1c" kit of the Capillarys 2 Flex Piercing may be suitable for the first-line screening of ß-thal patients (with acut-off point of HbA2 of 2.8%) and for the follow-up of sickle-cell disease patients who sometimes require a preciseHbS determination (for example, before a surgery or to monitor their regular transfusion program).



TUESDAY - June 23th

HbA1c

Diabetes

T053

CAPILLARY ELECTROPHORESIS SIGNIFICANTLY IMPROVES CLINICAL UTILITY OF HEMOGLOBIN A1C IN GESTATIONALDIABETES

M. Vučić Lovrenčić1, S. Božičević1, M. Krhač1, V. Radišić Biljak1, M. Prašek2

1Department of Medical Biochemistry and Laboratory Medicine, Merkur University Hospital, Zagreb, CROATIA2Vuk Vrhovac University Clinic, Merkur University Hospital, Zagreb, CROATIA

BACKGROUND-AIM

Clinical utility of hemoglobin A1c (HbA1c) in diagnosis and monitoring of gestational diabetes (GDM) has been seriouslycompromised due to a lack of diagnostic accuracy originating from both biological variability and analytical limitationsof contemporary methodology. Recently introduced capillary electrophoresis system for HbA1c analysis (Capillarys 2Flex Piercing ®/ Minicap Flex Piercing®, Sebia, France) has been reported to have an excellent analytical performanceand was found to be free from the common analytical interferences and hemoglobin variants, as well as fetalhemoglobin, which might be particularly interesting when measuring HbA1c in pregnancy complicated with diabetes.In this study we aimed to validate the clinical utility of HbA1c, as measured by the capillary electrophoresis, in diagnosisof GDM.

METHODS

256 pregnant women (mean gestational age: 26±4,7 weeks) were screened for GDM with a standard 75g oral glucosetolerance procedure followed by the venous plasma glucose measurement (hexokinase;Beckman Coulter AU680, USA)at fasting, 1h and 2h after glucose load. Their glycaemic status was classified according to the WHO-2013 criteria. HbA1cwas sampled at fasting and assayed with capillary electrophoresis [HbA1c-CAP (Minicap Flex Piercing®, Sebia, France)]and an automated immunoturbidimetric procedure [HbA1c-IT (TinaQuant-Integra 400Plus, Roche Diagnostics, USA)].

RESULTS

GDM was diagnosed in 91 women (35,5%), who did not differ regarding age and gestational age from women withnormoglycaemia (NG; N=165). HbA1c-CAP was significantly lower than HbA1c-IT (4,7±0,30%/28±3,3 mmol/mol vs.5,2±0,24%/33±2,6 mmol/mol, P<0,0001), and both systematic and proportional differences were found between themethods (y=1,625+0,750x;intercept A/95%CI=1,625/1,233-2,033;slope B/95%CI=0,75/0,667-0,833; Passing Bablok).ROC-curve comparison showed a significantly better diagnostic accuracy of HbA1c-CAP vs. HbA1c-IT in discriminatingbetween GDM and NG:[AUC/sensitivity(%)/specificity(%)/criterion(%/mmol/mol)=0,727/70,5/69/>4,7/>28vs. 0,681/53,7/74,2/>5,2/>33, respectively; P=0,0176].

CONCLUSION

Our study indicates a significant improvement in clinical utility of hemoglobin A1c in GDM diagnosis with the use ofcapillary electrophoresis.



TUESDAY - June 23th

HbA1c

Diabetes

T056

CAPILLARY BLOOD SAMPLING KIT FOR HBA1C VERSUS VENOUS PUNCTURE ON CAPILLARYS 2 FLEX PERCING

C. Bendavid2, L. Peltier2, C. Letellier1

1Service de biochimie-toxicologie,CHU Rennes2Service de biochimie-toxicologie,CHU Rennes, INSERM U991

BACKGROUND-AIM

Capillary blood sample collected from finger prick presents many advantages over venous puncture: low volume,less invasive for the patient, better patient’s compliance with monitoring recommendations. The present study wasdesigned to compare the measurement of capillary blood hemoglobin A1c levels with venous blood hemoglobin A1clevels using the Capillarys 2 Flex Piercing system (C2FP) (Sebia, France) on a large range of HbA1c values and withdifferent storage conditions.

METHODS

Data was collected from samples of 60 volunteer patients and covering a wide range of HbA1c values (4.7% - 14%NGSP). Both venous and capillary blood samples obtained simultaneously from each subject were tested using theC2FP system. After an initial assessment of venous HbA1c at J0, capillary and venous samples were stored at roomtemperature (Room T°) and 4°C respectively, away from light, and re-analyzed together at J5 on the same C2FP systemin duplicates. To test stability, 4 different samples were simultaneously taken from venous puncture and finger prick,and stored at different T° (-20°C, 8 days; 2-8°C, 8 days; Room T°, 8 days; 30°C, 3 days). Respective duplicates values werecompared to capillary and venous (reference) result at J0.

RESULTS

The trendline of J5 values using mmol/mol IFFC units (slope: y=0.9904x + 0.1387; R2=0.997) or %NGSP units (slope:y=0.9896x + 0.046; R2=0.997) showed a good correlation. Bland Altman plots showed a 0.4mmol/mol IFCC and 0%NGSP mean differences. All values were included in the recommended +/-6% bias on the bias plot.Room T° storage during 5 days resulted in a small additional peak of degradation but HbA1c value was still accurate.Reproducibility was assessed using the mean biases between the NGSP duplicates and showed the same 0% for venousand capillary results. Stability study on low, medium and high HbA1c levels showed that ideal conservation was 4°C.Room T° and -20°C give rise to degradation without alteration of HbA1c result. After 3 days at 30°C, only one sampleresult was slightly out of uncertainty of measurement.

CONCLUSION

The Sebia capillary sampling kit offers full automation and full positive ID. We have demonstrated a good correlationwith venous sample results. Storage study showed a sufficient robustness for usual sample delivery to centrallaboratory.



TUESDAY - June 23th

HbA1c

Diabetes

T063

COMPARISON OF 4 METHODS FOR HEMOGLOBIN A1C

L. Rogers1, P. Woolaver1, F. Kiechle1

1Memorial Healthcare System, Hollywood, Florida

BACKGROUND-AIM

We compared 4 methods for hemoglobin A1c (HbA1c) analysis at a large healthcare system with a diverse population.

METHODS

We analyzed 226 samples from patients with routine screening and 21 preserved samples with knownhemoglobinopathies or elevated HbA1c. Four methods were used including 2 immunoassays, Roche Cobas® Tina-quant HbA1cDx Gen. 2 assay and Siemens Dimension Vista®, HPLC using the BioRad Variant™ Turbo II, and capillaryelectrophoresis using the Sebia Capillarys™2 Minicap Flex Piercing system. Correlations were performed Demingregression analysis using EP Evaluator™. 13 specimens (P. Galveston, 2 J-Baltimore, N-Baltimore and 3 samples eachof HbC, HbD, HbE) from the National Glycohemoglobin Standardization Program (NGHP) were analyzed by BioRad andSebia methods.

RESULTS

Correlation data for all patient samples:Roche vs. Siemens: y=1.030x-0.44 R=0.983Roche vs. BioRad: y=1.029x-0.44 R=0.985Roche vs. Sebia: y=1.112x-0.95 R=0.983Siemens vs. BioRad: y=0.995x-0.14 R=0.980Siemens vs. Sebia: y=1.075x-0.61 R=0.982Sebia vs BioRad: y=0.927x+0.42 R=0.993Correlation data for 41 specimens with heterozygote hemoglobinopathies:Roche vs. Siemens: y=1.108x-0.86 R=0.975Roche vs. BioRad: y=0.968x+0.09 R=0.971Roche vs. Sebia: y=1.122x-0.88 R=0.969Siemens vs. BioRad: y=1.149x-0.99 R=0.948Siemens vs. Sebia: y=0.988x+0 R=0.955Sebia vs BioRad: y=0.880x+0.70 R=0.991BioRad and/or Sebia methods identified 22 patients (9.7%) of 226 routine screening specimens with aheterozygote hemoglobinopathy undetected by the immunoassay methods. Of the 13 NGHP samples with knownhemoglobinopathies analyzed by the BioRad and Sebia methods, all were identified as “atypical” by Sebia. BioRadidentified HbC as C, Hb D and E as “variant” and the 4 abnormal hemoglobins were unidentified. J-Baltimore was foundto co-migrate with HbA1c by capillary electrophoresis.

CONCLUSION

The 4 methods correlated well for samples with normal hemoglobin. Results from samples with heterozygoushemoglobinopathies showed more variability and less robust correlation, with the exception of Sebia and BioRadmethods. Sebia appears to be the most robust system for screening for Hb variants in HbA1c specimens.



TUESDAY - June 23th

HbA1c

Diabetes

T066

EFFECTS OF HEMOGLOBIN S, E, D OR C ON MEASUREMENT OF HBA1C – A COMPARISON BETWEEN SEBIA CAPILLARYS 2FLEX PIERCING AND BIO-RAD VARIANT TURBO 2.0 (RESIN D)

B. Landin1, L. Fredriksson1

1Dept Clinical Chemistry, Karolinska University Hospital, Stockholm

BACKGROUND-AIM

Each new HbA1c method must be evaluated for hemoglobin (Hb) variant interference. In our laboratory we routinelyuse the Bio-Rad Variant Turbo 2.0 method (VT2.0). In 2013 a new resin (resin D) introduced a slight change in the VT2.0integration, e.g. the HbA2 peak is now integrated separately. In 2014 we evaluated the Sebia Capillarys 2 Flex Piercingmethod (CAP) and studied groups of samples from patients without Hb variants (AA) or heterozygous for HbS (AS),HbE (AE), HbD-Punjab (AD) or HbC (AC).

METHODS

EDTA samples were analyzed fresh or after freezing at -80°C. Two VT2.0 instruments were randomly used. A CAPinstrument was provided for evaluation by Sebia (Lisses, France). Control samples (ERL, target value 35, 59 and 80mmol/mol, n=22 at each level) were regularly analyzed on both methods throughout the evaluation period. Meanresults were 35, 58 and 81 mmol/mol for both methods, and no significant difference between the two methods wasfound (p 0.578).

RESULTS

Sample groups AA (n=40), AS (n=45; HbS content 22-42 % according to CAP HbA1c), AE (n=24; 14-25 %), AD (n=13;35-38 %) and AC (n=10; 29-37 %) were also compared. The correlations between the two methods were excellent (R2>0.99) for all groups. The CAP results were significantly higher than VT2.0 results for all groups (mean difference AA0.9, AS 1.9, AE 1.8, AD 5.1 and AC 2.3 mmol/mol). The effect of HbS was significant when all AS samples were included(p 0.0351), but insignificant when 6 samples outside the range 27-93 mmol/mol were omitted (p 0.1531). The effect ofHbD was clearly significant (p 0.002), the relative difference between the VT2.0 and CAP results being most pronouncedat higher HbA1c levels (>76 mmol/mol).

CONCLUSION

The CAP method correlates well with the VT2.0 using resin D, both when testing normal samples (AA) and samplescontaining common Hb variants (AS, AE, AD, AC). However, in spite of identical results for ERL controls, CAP results forhuman samples were constantly somewhat higher than VT2.0 results. The mean difference between the two methodswas 1.8 % for the AA group and <5.3 % for both AS, AE and AC groups. A clear effect of a common Hb variant was seenonly in the AD group, in which the mean difference between the two methods was 10%.



TUESDAY - June 23th

Hemoglobinopathies

Diabetes

T018

PREVALENCE OF HEMOGLOBIN DISORDER BASED ON HB A1C CAPILLARY ELECTROPHORESIS ASSAY: EXPERIENCE FROMA LABORATORY IN FRANCE

N. Baidjibay 1, M. Van De Loo1

1Laboratoire Bio7, 91090 Lisses, France

BACKGROUND-AIM

Hemoglobinopathies (HbP) are now common worldwide due to migration. At least 5.2% of the world population carrya hemoglobin (Hb) variant and around 1.1% of couples are at risk for having children with an Hb disorder. Thus,genetic counseling for these couples is essential. In France, there is no systematic screening and all laboratories donot routinely realize Hb studies. We evaluated the possibility to use Hb A1c measurement by capillary electrophoresis(CE), much used in routine, to incidentally discover HbP and calculated their prevalence in our recruitment area.

METHODS

A retrospective study was carried out over an 8-days period. A total of 3,233 patient samples were received for diabetesscreening and follow-up. All samples were analyzed by CE on the CAPILLAYS 2 Flex Piercing (Sebia), with the "HbA1c"kit. Profiles showing Hb variants, Beta-thalassemia (Hb A2 > 3%) or elevated Hb F (> 2%) were counted up.

RESULTS

Our 47 collecting centers are located in 5 departments around Paris and cover more than 430 km2 and 866.000inhabitants. We use CE for Hb and Hb A1c, both assays allowing to highlight Hb abnormalities in clear-cut and preciseprofiles. Regarding the amount of tests, Hb A1c CE assay represents 98% whereas Hb CE assay only 2%. As there isno systematic screening for HbP in France, Hb A1c CE assay appears to be a good option to screen the population. Atotal of 121 HbP were found: prevalence can be considered equal to 3.74%, which is compatible with ones previouslycalculated. Among them, 23 Beta-thalassemias were found, 87 Beta, 7 Delta and 3 Alpha variants and one HbP leadingto an Hb F increase. These Hb disorders have a prevalence of 0.71%, 2.69%, 0.22%, 0.09% and 0.03%, respectively. HbS-like variants are the most common of the discovered Beta variants (64.4%, with 2 homozygous cases). Interestingly,no Hb D-like were found.

CONCLUSION

The detection of undiagnosed HbP during Hb A1c CE assay is not rare. Beyond the Hb A1c measurement, around 10-20cases of HbP are found per day. This technique can be used as a massive screening test. The incidental observationshould be reported to the clinicians and must lead to further investigations (e.g. Hb testing by CE) and geneticcounseling for the individuals.



TUESDAY - June 23th

Hemoglobinopathies

Haematology

T290

PHYSIOLOGICAL VARIABILITY OF FOETAL HEMOGLOBIN’S RATE IN THE INFANT AND THE CHILD

A.T. Dalila1, H. Amina2, T. Chafika2, B. Fatiha1, G. Z'hor2

1Nafissa Hamoud Hospital CHU HUSSEIN DEY2Nafissa Hamoud Hospital CHU HUSSEIN DEY

BACKGROUND-AIM

Hemoglobin is the respiratory pigment in erythrocytes; In adults, the major hemoglobin is Hb A(α2β2), with 2,3-3,5%HbA2(α2δ2) and the rest being H F(α2γ2).The Hb F appears with the fœtal life of the individual; it is in a majority at thenew-born baby and its rate gradually decreases until the 6 months to 1 year of life.Expression of residual Hb F amongadult individuals is highly variable and the diversity of ontogenetic evolution is a study model. Its rate is a major factorfor modifying the severity of hemoglobinopathies.The study of the variation of the rate of Hb F in infants and childrenin the physiological state shows a great interest. Our goal is to establish normal values for different age groups and touse them as thresholds for the early detection of hereditary and acquired hemoglobinopathies especially as we haveno estimate of that rate in Algeria. More over, the border is limited between the physiological state and the HPFH(hereditary persistence of Hb F).

METHODS

Our study included 116 healthy subjects with no anomalies at hemoglobin, consisted of infants and children whose ageis between 1 day to 5 years, 67males (57,76%)and49 females(42,24%) with a 1.37sex ratio.Peripheral blood samplesanalysed by automated cell counters (Medonic 620), capillary electrophoresis hemoglobin by Capillarys SEBIA

RESULTS

According to our results,the switching is delayed at1 year, to consider the 5,12% threshold that was found and the1,12for the age groups from1 to5 years with a peak for 4%.This residual Hb F continues to be synthesized during the postnatal and adult life at variable rates.These quantitative variations are related to polymorphisms,the Hb γ > β switch is astudy model,the rate of HbF modulates the severity of hémoglobinopathies.This study showed that the normal ratesin our population is different from those set by the literature,but no statistical difference was found in Hb F values(P0,203)A mutation C > T at the position - 158 of the gene Hb γ2 (Xmn1-Hb γ 2) has already been found in thalassemiain Algeria,associated with a very high expression of HbF

CONCLUSION

The variability of HbF is a quantitative genetic feature controlled by a set of polymorphisms, which one of mostimportant would be located on the long arm of chromosome 6 in 6q23, and its presence seems necessary, with themarkers of chromosomes 11 and 2 (site 2p15).it might explain the normal HbF→HbA switch and its pathophysiologicalderegulation.



TUESDAY - June 23th

Gel

Haematology

T330

INTEREST OF THE NEW RAPID TEST “HYDRAGEL 5 VON WILLEBRAND MULTIMERS” FOR THE ANALYSIS OF VONWILLEBRAND MULTIMERS

V. Marc1, K. Jean-emmanuel2, F. Dominique1, B. Roseline1, N. Georges3

1Biology department, Hospital Foch, Suresnes2Internal Medicine? Hospital Foch, Suresnes3Sebia, Lisses

BACKGROUND-AIM

Diagnosis of von Willebrand Disease (vWD) requires measurement of von Willebrand factor (vWF) by bothimmunological (vWF:Ag) and functional tests (vWR:Co), based on ristocetin cofactor activity. A visible decrease ofvWR:Co is observed in case of qualitative defects of vWF [inherited/type 2 or acquired/avWD], with a decreased ratiovWR:Co/vWF:Ag (< 0.5). In some cases, this ratio is in a grey zone (0.5- 0.75). The gold standard to identify qualitativevWD is the analysis of the multimers’distribution. It is a time-consuming technique, available in specialized labs. Wepresent the interest in 3 cases of a new assay allowing a within day (6 hours) multimers’ analysis.

METHODS

Samples were analysed on the Hydrasys 2 instrument (Sebia, Lisses France) with a ready to use SDS agarose gel.Multimers were visualized directly on the gel (w/o protein transfer), using an immunofixation with antibodiesconjugated to horseradish peroxidase/ TTF1/TTF2.

RESULTS

Case 1: woman (38th week/1st gestation) who reported unusual bruises from childhood. Bleeding time (BT) was veryprolonged, platelet count slightly decreased (128 G/L), ratio vWR:Co/vWF:Ag = 0.38. Gel electrophoresis showed theabsence of high-molecular-weight multimers (HMWM) and a large excess of vWF dimers, suggesting a type 2:C.Case 2: woman who reported a familial history of vWD (24th week/1st pregnancy), first investigation. BT wasvery prolonged, platelet count normal, ratio vWR:Co/vWF:Ag = 0.58. Electrophoresis indicated the absence HMWM,suggesting a type 2:A.Case 3: 60 year-old woman who needed gastric biopsies (suspicion of cancer). She reported recently important bruises.Previously she went through different surgeries and 3 deliveries without bleeding. BT was very prolonged, ratiovWR:Co/vWF:Ag = 0.38. Electrophoresis indicated the absence HMWM, in basal conditions and 2 hours after vWFtherapy, suggesting an avWD.

CONCLUSION

This new assay is a valuable tool for vWD diagnosis, useful when a molecular abnormality of vWD is suspected. Itprovides clear pattern of wWF multimer distribution and has the major advantage to be performed within day, withthe instrument used in laboratory.



TUESDAY - June 23th

Gel

Quality assessment, laboratory errors, patient safety, ethics

T399

EVALUATION PERFORMANCE OF PROTEIN ELECTROPHORESIS AND IMMUNOFIXATION USING SEBIA HYDRASYS 2SCAN, SEBIA CAPILLARYS 2 FLEX PIERCING AND INTERLAB G26

B. M. Hussain1, S.S. Siti Suhana Abdullah Soheimi11, T. Tiara Hassan2, A.K. Kamarudin1, N. Abdul Hamid1

1Ampang Hospital, Ministry of Health, Malaysia2Institute of Biological Sciences, University of Malaya, Kuala Lumpur

BACKGROUND-AIM

To evaluate the analytical performance of the serum protein electrophoresis (SPE) assay on CAPILLARYS 2 Flex Piercing(SEBIA, USA), HYDRASYS 2 SCAN (SEBIA, USA) and G26 (INTERLAB, Italy) by precision and correlation .To evaluate the analytical performances of the Immunofixation (IF) assay on HYDRASYS 2 SCAN (SEBIA, USA) and G26(INTERLAB, Italy) by reproducibility and specificity.

METHODS

Specimens from Pooled healthy blood donors (normal serum) and patients samples were subjected to SPE for precisionand correlation. For IFE reproducibility, normal serum was spiked with IgG/A/m/Kappa/Lambda paraprotein andfor specificity, normal serum was diluted with quality control material in various ratios. Haemolysed samples wereexcluded. Each sample was analysed on all instruments on the same day to prevent degradation effects. SPE and IFEassays were performed using manufacturer’s recommendation.

RESULTS

1. Precision study demonstrated on Sebia Hydrasys 2 Scan, Sebia Capillarys 2 Flex Piercing and Interlab G 26. Allanalyzers showed less than targeted 10% CV.2. Overall, the R2 value for protein fractions between all the analyzers correlates well (R2> 0.9) except for Beta fractionHydrasys 2 vs IG26. Hydrays 2 correlates well with Capillarys 2 Flex despites using different method. Capillarys 2 doesnot correlate well with Alpha 2 and Beta of IG26.3. Interlab G26 did not detect IgA/ IgM and λ at low levels.4. Interlab G26 did not detect IgA/ IgM and λ at low levels.

CONCLUSION

The performance of the three analyzers for SPE assays demonstrated good precision. Sebia Hydrasys 2 Scan and SebiaCapillarys 2 Flex Piercing showed good correlation with each other in all parameters. IF assays cannot be done on SebiaCapillarys 2 Flex Piercing. Specificity study demonstrated that Sebia Hydrasys 2 is able to detect low level of IgA, IgMand λ. The Pre-set setting of Interlab G 26 were not able to detect low levels of IgA, IgM and λ. We were able to detectafter optimization. In conclusion, Sebia Hydrasys 2 shows that the analyzer is optimized to be used where else, InterlabG26 requires optimization prior to use.



WEDNESDAY - June 24th

HbA1c

Reference ranges, standardization and decision levels

W153

GLYCATED HEMOGLOBIN: REFERENCE VALUES ACCORDING TO GENDER AND AGE.

M. Pieri1, F.G. Martino2, R. Zenobi1, S. Pignalosa1, F. Duranti1, F. De Gregorio1, S. Bernardini1, M. Dessi11Department of Experimental Medicine and Surgery, "Tor Vergata" University Hospital, Rome (Italy).2Radiological Sciences and Laboratory Medicine Department, S. Filippo Neri Hospital, Rome, (Italy).

BACKGROUND-AIM

Diabetes has become “a global epidemic” and affects more young people, showing gender differences both forcomplications that for used therapies. Glycated Hemoglobin (HbA1c) reflects glycemia up to 6-8 weeks and representsthe "gold standard" for the diagnosis and monitoring of diabetes.According to the World Health Organization (WHO), HbA1c levels > 48 mMol/Mol (6.5%) are diagnostic for the disease;however, values between 42-47 mMol/Mol are considered high risk for the onset of the pathology. HbA1c values areinfluenced not only by the concentration of blood glucose, but also by genetic factors, race, smoking and obesity. Ouraim was to determine HbA1c normal range by gender and age in a sample of healthy population aged 18-64 years inorder to identify those values corresponding to pre-diabetic forms to implement effective prevention.

METHODS

We evaluated 558 samples of healthy donors attending the University Hospital of "Tor Vergata" and the "San FilippoNeri" Hospital. Whole blood samples, collected in K2-EDTA tubes and stored at -80°C were analyzed in a single sessionby Capillarys 2-FP (Sebia, Lisses, France) in the University Hospital of "Tor Vergata".

RESULTS

The sample consisted of 273 men (48.9%) and 285 women (51.1%). Our data showed a significant difference in meanHbA1c values between genders (men: 32.4 ± 4.6 mMol/Mol; women: 30.1 ± 3.8 mMol/Mol; mean ± SD; p<0.001).Samples were divided in six groups (A:18-34 years; B:35-49 years; C:50-64 years for men and A1, B1, C1 for women). Inparticular, C group (34.1±4.6 mMol/Mol) had mean values of HbA1c statistically significant (p<0.001) versus A (30.5±3.2mMol/Mol), B (31.7±7.7 mMol/Mol), A1 (29.6±3.4 mMol/Mol), B1 (30.2±3.7 mMol/Mol) and C1 (30.9±4.1 mMol/Mol). Nostatistical differences were found between women groups.

CONCLUSION

Data observed showed significant differences in HbA1c mean values measured for both gender and age. Setting morestringent values of HbA1c, based on the above parameters, could promote early diagnosis of diabetes and help theclinician to improve therapy, customizing it, and to reduce the risk of disease serious long-term complications.




CDT

Liver, pancreas, gastrointestinal diseases, microbiome

W218

THE PROFILE OF TRANSFERRIN ISOFORMS IN PANCREATIC CANCERS

E. Gruszewska1, L. Matus3, M. Gudowska1, B. Cylwik2, M. Szmitkowski1, B. Kedra3, L. Chrostek1

1Department of Biochemical Diagnostics, Medical University of Bialystok, Poland2Department of Pediatric Laboratory Diagnostics, Medical University of Bialystok, Poland3II Department of General and Gastroenterological Surgery, Medical University of Bialystok, Poland

BACKGROUND-AIM

Glycan structures on serum transferrin are terminated by sialic acid, which plays various important roles. The changesin transferrin sialylation are observed in many diseases, especially in cancers. The aim of this study was to assess theeffect of pancreatic cancers on serum profile of transferrin isoforms.

METHODS

Serum samples were obtained from 44 patients suffering from pancreatic cancers and 25 healthy volunteers (controls)recruited from hospital workers. The samples were analyzed by capillary electrophoresis (CE) technology on a MINICAPelectrophoresis system (Sebia, France). The normal serum transferrin isoforms were separated into five major fractionsaccording to their sialylation level: asialo-, disialo-, trisialo-, tetrasialo- and pentasialotransferrin.

RESULTS

There were no differences in the relative concentrations of disialo- (mean±SD; 0.63±0.30%) and trisialotranferrin(3.0±1.39%) in patients with pancreatic cancers when compared to the control group (0.70±0.42%, 3.54±1.08%;respectively) (Mann-Whitney U test: P>0.05 for both). But the mean concentrations of tetrasialo- (81.06±2.22%) weresignificantly higher and pentasialotransferrin (15.33±2.72%) were significantly lower in cancer patients than that inthe controls (77.98±4.21%, 17.82±4.38%; respectively) (P<0.001, P=0.009; respectively). There were no differences inthe concentrations of transferrin isoforms according to the size of tumor, presence of regional lymph nodes and distantmetastasis (P>0.05 for all comparisons). The exception was the concentrations of trisialotransferrin isoforms whichwas significantly higher in patients without (M0) (3.25±1.09%) than in those with presence (M1) (2.04±0.85%) of thedistant metastasis (P=0.008). The ratio of tetrasialo- to pentasialotransferrin in pancreatic cancers (5.45±0.97) wassignificantly higher than in the controls (4.66±1.27) (Mann-Whitney U test: P=0.007).

CONCLUSION

The electrophoretic visualization of the total transferrin isoforms shows the occurrence of alterations in transferrinsialylation in pancreatic cancers. It is clearly visible that tetrasialylated isoforms were frequently present than high-sialylated (pentasialotransferrin).




CDT

Toxicology, therapeutic drug monitoring, drug addiction

W411

CARBOHYDRATE-DEFICIENT TRANSFERRIN: BIOMARKER OF CHRONIC ALCOHOLISM. THE EARLY DETECTION OF ASOCIAL AND HEALTH PROBLEM

A. García Narbón1, M. Martínez Villanueva1, M.C. Ramírez Ruiz1, A.I. Sánchez Bermúdez1, C. Martínez Ros2, F. Espí

Martínez2, J.A. Noguera Velasco1

1Clinical Laboratory Sevices /Clinical Universitary Hospital Virgen de la Arrixaca, Murcia2Emergency Sevices Hospital/Clinical Universitary Hospital Virgen de la Arrixaca, Murcia

BACKGROUND-AIM

Carbohydrate-deficient transferrin (CDT) is a biomarker of chronic alcohol ingestion. Intake of 60-80 grams of alcoholdaily for two weeks, increases in blood the hyposialylated forms, disialotransferrin and asialotransferrin, whichconstitute CDT. The aim of our study is to determine the prevalence of chronic alcoholism in patients who come to theemergency department based on the values of CDT and assess the health impact related to this consumption.

METHODS

Cross-sectional study of 261 patients (194 men, 67 women) who come to the emergency department, makingthe determination of blood alcohol (quantitative colorimetric test,Siemens®) when there is a suspicion of alcoholconsumption. When alcohol is positive (>15mg/dl) CDT is determined (capillary electrophoresis,Capillarys Sebia®).CDT>1.6% is associated with chronic alcohol ingestion and CDT≤1.6 with acute intake.

RESULTS

Based on the value of CDT, 73.2% (191) patients presented acute alcohol consumption and 26.8% (70) chronicuse. In the group with pathological CDT, 85.7% are men and 14.3% women(p=0.011). The average age of thisgroup differs significantly (p=0.000) compared to acute intake group (47.13±12.99 vs 40.15 ± 15.27). In patientswith alcohol dependence: 94.3%(66) presents repeat hospital admissions due to alcohol consumption (p=0.000),37.1% (26) consume other psychoactive substances (cocaine, cannabis, opiates)(p = 0.010), 45.7% (32) requireshospitalization(p=0.021) and 62.9%(44) develops a pathology related to this addiction (hypertension, heart disease,hepatitis, cirrhosis, psychiatric disorders) requiring other treatments and interventions (p = 0.024), observingsignificant differences respect to the other group.

CONCLUSION

CDT is an specific biomarker of alcoholism that has allowed us to identify 27% of chronic alcohol consumers in ourstudy population. As an added social and health problem, we found an alarming rate of chronic alcohol consumers whoare also users of other drugs and which develops pathologies associated with their addiction. Prevention strategiesshould establish appropriate actions due to the health, social and economic benefits that can be obtained with earlyintervention. A simple way to start is the detection of these patients in the emergency department.




CDT

Toxicology, therapeutic drug monitoring, drug addiction

W425

HARMONISATION OF SEBIA CAPILLARYS CDT RESULTS ACCORDING TO WG-CDT RECOMMENDATIONS

A. Helander1, H. Dahl21Karolinska Institute, Department of Laboratory Medicine, and Karolinska University Laboratory, Clinical Chemistry,Stockholm, Sweden2Karolinska University Laboratory, Clinical Chemistry, Stockholm, Sweden

BACKGROUND-AIM

Carbohydrate-deficient transferrin (CDT) is a specific biomarker for long-term alcohol overconsumption. Drawbacksin CDT testing are that routine methods differ in their definition of the measurand and in reference intervals, whichprompted initiation of an IFCC Working Group on CDT Standardisation (WG-CDT). This study examined the possibilityto harmonise CDT results of the SEBIA Capillarys CDT method according to WG-CDT recommendations, using high-performance liquid chromatography (HPLC) as reference method.

METHODS

525 serum samples, including 60 with genetic transferrin variants, chromatographic interferences or congenitaldisorder of glycosylation, were obtained from the routine HPLC analysis of CDT (i.e. relative amount of thedisialotransferrin glycoform, %DST). The samples were analysed in parallel by the SEBIA Capillarys capillaryelectrophoresis (CE) assay.

RESULTS

There was good overall correlation between the %DST results obtained by SEBIA CE and HPLC (r2=0.91, p<0.0001), butthe CE values were on average ~0.5% lower over the entire measuring range. Using the regression equation betweenthe methods, the CE %DST results were adjusted (harmonised) to corresponding HPLC %DST equivalents. At a cut-off of 1.9% DST, which is routinely used in Sweden, the harmonised CE values showed good agreement (sensitivity90%, specificity 99%) with the measured HPLC values. A closer evaluation of the deviating results (“false positive/falsenegative”) revealed that almost all of those were very close to the cut-off, suggesting CE and HPLC method imprecisionas a main reason. The method imprecision of the SEBIA Capillarys CE assay was ≤6% near or above the cut-off. For afew samples, quantification of %DST was not possible by CE due to analytical interferences but usually by HPLC.

CONCLUSION

These data demonstrated good linearity and reproducibility of CDT results by SEBIA Capillarys CDT when analysed inparallel with HPLC. The results further demonstrated that routine harmonisation of SEBIA Capillarys CDT (%DST) resultsby way of the IFCC WG-CDT recommendation is possible. The results also pointed at the value of having access to HPLCas confirmative CDT method, in the few cases showing analytical interferences in the CE assay.




Myeloma

Reference ranges, standardization and decision levels

W160

ESTABLISHMENT OF REFERENCE RANGES FOR SERUM PROTEINS CAPILLARY ELECTROPHORESIS IN THE PEDIATRICPOPULATION

B. Valériane2, P. Bruno1, E. Bertrand3, P. Loic2, S. Vincent2, F. Anne2

1Délégation à la Recherche Clinique et à l’Innovation (DRCI), CHU de Clermont-Ferrand, Clermont-Ferrand, France2Service de Biochimie et Biologie Moléculaire, CHU de Clermont-Ferrand, Clermont-Ferrand, France3Service d’Immunologie, CHU de Clermont-Ferrand, Clermont-Ferrand, France

BACKGROUND-AIM

Serum protein electrophoresis is a routine exam performed in the majority of French medical analysis laboratories.This exam allows the separation and the quantification of total serum proteins into usually six fractions: albumin,alpha-1 globulins, alpha-2 globulins, beta-1 globulins, beta-2 globulins and gamma-globulins. It is useful in numerouspathological situations to make diagnoses, to follow evolution of a disease or to evaluate the efficiency of a treatment.Its interpretation relies on qualitative and quantitative analyses of each proteins fractions. For this, reference rangeshave been established for each fraction using serums from healthy adult patients. However, theses classical referenceranges used to interpret serum protein electrophoresis, and particularly the gamma-globulins fraction, are not suitablefor the pediatric population. Then we proposed to establish reference intervals for each electrophoresis fraction in thepediatric population.

METHODS

We collected 1053 serums from healthy patients from 15 days to 18 years old, and performed serum proteinelectrophoreses using the automated capillary electrophoresis (CE) system Capillarys Flex Piercing® from Sebia.

RESULTS

The gamma-globulins appeared as the most discriminant fraction with age in our pediatric population in comparisonto adult population. We found a statistical correlation between the percentage of gamma-globulins and age (r=0.52)with an increase of percentage gamma-globulins from 15 days to 10 years old. Then we determined seven statisticallydifferent classes of age and established the reference range in percentage and gamma-globulins concentrationsfor these seven classes using the intervals method (Q5-Q95). We deducted the relative (%) and concentration (g/L)reference ranges for the five other electrophoretic fractions.

CONCLUSION

Our results will allow the use of suitable reference ranges to interpret correctly the serum protein electrophoresis, suchas hyper- or hypo-gammaglobulinemia appreciation, from babies and children.




Gel

Vascular biology, endothelium, haemostasis, cardiovascular diseases

W187

INDUSTRIAL DEVELOPMENT OF A WITHIN-DAY VON WILLEBRAND FACTOR MULTIMERS ASSAY USING AGAROSE GELELECTROPHORESIS

L. Thierry1, B. Ghislaine 1, N. Georges 11Sebia, Research and development department, Lisses, France

BACKGROUND-AIM

The analysis of von Willebrand factor (vWF) multimers is necessary for the classification of hereditary and acquiredforms of von Willebrand disease (vWD). Only a few specialized laboratories are skilled enough to perform this analysisdue to the complexity of the method itself and to its very slow turnaround time (2 to 3 days). vWF multimers are usuallyseparated by “Home- made” discontinuous SDS agarose gel (difficult to produce) followed by a western blotting step.Multimers are then identified by immunofluorescence or other staining techniques.We developed a new method for multimer analysis on a commercially available instrument (Hydrasys 2 Scan, Sebia,France), which is rapid (within 6 hours), easy to perform (no western blot), reproducible (ready to use SDS agarose gel).This method assesses the overall size distribution of vWF multimers (low, intermediate and high molecular weight).

METHODS

Plasma vWF were loaded and separated on Hydrasys 2 Scan, in continuous SDS agarose gel system (no stacking andrunning gel) within 110 minutes. Multimers were probed in gel by immunofixation using horse-radish peroxide (HRP)conjugated rabbit anti-vWF (90 minutes). Visualisation of multimers was achieved by colorimetry using commerciallyavailable TTF1/TTF2 Sebia reagents. Curves were produced using GelScan and Sebia Phoresis software.

RESULTS

The results obtained with this new method show at least 9 to 12 main multimers bands in normal plasma. The absenceof intermediate and high molecular weight multimers often seen in type II variants was easily discernible. No multimerbands were visible in type III vWD plasma sample. Increasing or decreasing of each size multimer was easily noticeable.Densitometric analysis performed on Gelscan for comparison of patients samples with normal plasma run on the samegel were an added tool for interpretation.

CONCLUSION

The method for vWF multimer analysis describe here is simple to carry out, produces results within 6 hours, performedon a commercially available instrument. This technique which is the first line multimer analysis could also help formore extensive use.



Meta-Analysis Scientific Posters PUBLICATIONS & REPORTS on ... Posters on Sebia Solutions.pdf · MONDAY - June 22th Hemoglobinopathies Advanced technologies, new biomarker discovery,

Documents