Evaluation of a new semi-automated Hydragel 11 von Willebrand factor multimers assay kit for routine use

Background Accurate diagnosis and classification of von Willebrand disease (VWD) are essential for optimal management. The von Willebrand factor multimers analysis (VWF:MM) is an integral part of the diagnostic process in the phenotypic classification, especially in discrepant cases. The aim of this study was to evaluate the performance of a new Hydragel 11VWF multimer assay (H11VW). Methods Analytical performance characteristics such as repeatability (intra-assay variability, in gel between track variation), reproducibility (inter-assay variability, between gel variation), sensitivity, EQA performance and differences between two commercially available VWF:MM kits (H5VW and H11VW) were analysed in healthy volunteers' plasmas using in-house prepared reference plasma. Results Repeatability and reproducibility results of H11VW demonstrated acceptable and equivalent performance with previously verified H5VW. Participation in EQA was successful. No statistically significant difference was detected between H5VW and H11VW kits for different fractions of multimers: LMWM p=0.807; IMWM p=0.183; HMWM p=0.774. Conclusions H11VW demonstrated acceptable analytical performance characteristics. H11VW kit conveniently offers a more significant number of samples on a single gel. H5VW and H11VW kits can be used in daily practice interchangeably.

The aim of this study was to evaluate the performance characteristics of H11VW.

Study objects
Two types of normal citrated plasma samples were used in the H11VW performance evaluation: in-house reference plasma (IRP) and plasma from 10 healthy individuals recruited voluntarily. IRP has been used in North Estonia Medical Centre laboratory for a couple of years, and detailed procedure on the preparation of IRP was published previously elsewhere (11). For healthy individuals, a well-structured questionnaire was used to obtain information about age, gender, individual/family bleeding history, medication. Information provided enabled us to classify them preliminary as non VWD individuals. Venous blood samples were collected into 3.8% NC Buffered Citrate (Vacutest KIMA s.r.l., Arzergrande, Italy) tubes, which were centrifuged at 1500 g for 15 minutes at room temperature to generate platelet-free plasma (residual platelet count < 10×10 9 /L), aliquoted and stored frozen at -70 °C until further analysis. Aliquots were thawed in a water bath (+37 °C) for 5 minutes and mixed well before testing. All participating volunteers gave their informed consent. The study was performed according to the Declaration of Helsinki and was approved by the national ethics committee.

VWF:MM
VWF multimers evaluation was performed on Hydrasys 2 Scan instrument (Sebia, Lisses, France), using 2.0% SDS agarose gel, direct immunofixation, visualisation with peroxidase-labelled antibody and followed by densitometry, according to manufacturer recommendations. VWF multimers were classified as low, intermediate and high molecular weight multimers (LMWM, IMWM and HMWM respectively). Densitometry data was obtained using the Phoresis Klju~ne re~i: elektroforeza, multimeri, Fon Vilebrandov faktor, Fon Vilebrandova bolest software originating from Sebia. Principle of VWF:MM methodology for H5WV and separation of multimers fractions was previously described in detail (9)(10)(11)(12). Technical steps for both kits (H5WV and H11WV) are very similar. The important differences are the metal weight holding mechanism for the blotting steps (1.8 kg for H5WV and 2.3 kg for H11WV) and a number of sample positions (5 tracks gel for H5WV and 11 tracks gel for H11WV).

Analytical performance characteristics
We have chosen the following analytical performance characteristics of H11VW to analyse: repeatability (intra-assay variability, in gel between track variation), reproducibility (inter-assay variability, between gel variation), sensitivity, EQA performance and differences between two commercially available VWF:MM kits (H5VW and H11VW). For repeatability analysis, 11 measurements were done, each for a single non VWD volunteer individual's plasma, which was applied to 11 tracks of the gel. Consequently, intra-assay coefficient of variation (CV) % was calculated. For reproducibility analysis, VWF:MM results of the same IRP from 55 different gels runs were collected, and inter-assay CVa% was calculated. For sensitivity analysis, the single volunteer's plasma with respectively known VWF antigen value was diluted in series (1:2, 1:4, 1:6; 1:8; 1:16 and 1:32). For of the dilution series VWF:MM assay was performed on single H11VW gel together with IRP for comparison reasons. The external quality assessment (EQA) program for VWF:MM was issued by the ECAT (External quality Control of diagnostic Assays and Tests with a focus on Thrombosis and Haemostasis). In total, North Estonia Medical Centre laboratory participated in 6 cycles of EQA using H11VW. Finally, applying the results of previously performed H5VW kit performance verification (n=26) (11), corresponding results of IRP on H11VW (n=29) were collected, and statistical comparison was carried out.

Statistical analysis
Results were reported as the mean % of the respective molecular weight fraction of multimers ± standard deviation (SD) and the coefficient of variation (CV, %). The difference between the two commercial kits (H5WV versus H11VW) was evaluated by nonparametric Man-Whitney U test on the IBM SPSS software, version 21.0 (Chicago, IL, USA). Values were considered statistically significant at p<0.05.

Intra-assay and Inter-assay variability
Example of repeatability analysis is depicted in Figure 1A. Visually, 11 tracks of one single plasma on the same H11VW gel look pretty much the same, but a visual inspection is too subjective. Densitometric analysis of the gels, and consequently calculated % of different fractions of multimers (LMWM mean value 14.4%, SD 1.0; IMWM mean value 27.5%, SD 2.8; and HMWM mean value 58.1%, SD 2.8) demonstrate intra-assay variability performance equivalent to previously published H5VW kit repeatability values (12): CV were 6.9% for LMWM, 10.3% for IMWM, and 4.8% for HMWM.
Statistical data in reproducibility analysis for different multimers fractions were as follows: LMWM mean value 18.1%, SD 3.0; IMWM mean value 33.1%, SD 2.0; HMWM mean value 48.9%, SD 3.9. These results yielded higher but more or less acceptable coefficients of variation for LMWH and HMWM when compared to repeatability data, but the variability of IMWM was lower. Inter-assay CV values were 16.6% for LMWM, 6.2% for IMWM, and 8.1% for HMWM, respectively.

Sensitivity analysis
Sensitivity analysis with serial dilutions revealed a cut-off (VWF antigen values 9 % and below), which aggravates visual inspection of gels, worsens densitometric analysis by Phoresis software (Sebia, France). An example of sensitivity analysis is given in Figure 1C.
Plasma of a healthy individual (VWF antigen of 96% and VWF activity of 105%) was diluted according to protocol and resulted in final VWF antigen values of 52%, 19%, 13%, 9%, 5%, and 2%. As shown in Figure  1C, at the level of 9% multimer, bands are still clearly recognisable, densitometric distribution of different fractions is substantially lower when compared to IRP graph, but proportions of LMWM, IMWM and HMWM values resembled the normal values.

EQA survey
The performance of the North Estonia Medical Centre laboratory in the ECAT Foundation EQA programs was considered successful because results of VWF:MM of all 6 cycles were in agreement with the corresponding goals of the ECAT Foundation. Summary of EQA results is provided in Table I.
Following the success of the EQA and above provided performance data, in May 2020 VWF:MM assay with H11VW kit was accredited in the North Estonia Medical Centre laboratory according to ISO15189:2012.

Discussion
The measurement of VWF multimers has become a part of the laboratory workflow for the identification and classification of VWD (1,4). Several home-made methods have been developed in the past decades for evaluating the VWF multimeric structure in expert-level laboratories, characterised by varying analytical performances (5-7), occasional differences in interpreting the results (12).
Both visual and densitometry-based investigation makes interpretation easier, allows the overlay patients curves with normal control and enables estimating the relative quantification of each multimer subset, providing useful information for the diagnosis of VWD. Implementation of VWF multimers assay for routine use is important for the classification of VWD, leading to the improvement of VWD diagnosis and monitoring of treatment response in Baltic countries.
The role of quality assurance in a hemostasis laboratory is very important (14). Because high precision is needed in VWF multimers quantification, standardisation of its measurements is crucial for an accu-rate diagnosis. One important aspect is the type of plasma sample used for internal quality control (IQC) (14). It is known that present type 2A VWD-like controls are not provided by VWF test manufacturers; thus, laboratories may be able to use previously diagnosed 2A VWD patients' samples. Using normal commercial plasma for IQC multimeric evaluation might end up with a relative loss of the highest HMWM, probably due to the lyophilisation process while preparing commercial plasma (6,11). To our knowledge, at least one group of researchers have verified commercial normal reference plasma (Standard Human Plasma, Siemens) as acceptable quality control for VWF multimers evaluation (15). This study was the first to report the H11VW kit validation results, including analysis of the VWD patients' samples, and has also noted the benefits and limitations of semi-automated VWF:MM assay, including the smaller sample size H5VW kit. In the present study, the results do not only support the previously published (15) but also provide additional analytical performance characteristics evaluation, especially for the larger sample size H11VW kit.
We concluded that the analytical performance of HYDRAGEL 11 VON WILLEBRAND MULTIMERS assay is acceptable and gives a perspective to standardisation of the VWF:MM assay by Sebia (France). NewH11VW kit conveniently offers a larger number of samples on a single gel, thus saves precious time. The choice of kit (H5VWvs H11VW) can be generally based on the volume of laboratory workload (the number of collected patient samples). Considering the performance data, H5VW and H11VW kits can be used in daily practice for the visual investigation of gel and quantitative estimation of VWF multimer fractions interchangeably.
Acknowledgements. Research funding: Sebia (Lisses, France) donated the Hydragel 5von Willebrand multimers kits. The authors would like to express gratitude to dr. Lenna Örd for language editing and laboratory technicians Galina Trofimova and Tatjana Tverskaja for excellent technical assistance related to sample testing.

Conflict of interest statement
All the authors declare that they have no conflict of interest in this work. VWF, von Willebrand factor; LMWM, low molecular weight multimers; IMWM, intermediate molecular weight multimers; HMWM, high molecular weight multimers; H5VW, Hydragel 5VWF multimer assay kit; H11VW, Hydragel 11VWF multimer assay kit.