Unmet Need for Reliable Immunological Detection Method for Anti-von Willebrand Factor Autoantibodies

Autoimmune coagulation factor deficiency (AiCFD) is a disease in which the formation of hemostatic thrombi is inhibited because of the acquisition of autoantibodies to coagulation factors, resulting in bleeding symptoms.[1] [2] Autoantibodies develop against any coagulation factors, mainly in the elderly.[3] [4] [5] The number of patients with AiCFD is gradually increasing in Japan,[2] which is considered the world's first super-aging society. Aging per se might be the greatest risk factor for this disorder,[6] along with genetic and environmental factors.[7] [8] The frequency of AiCFD is 1.85, 0.044, and 0.038 per million people each year for autoimmune factor VIII, XIII, and V deficiency (AiF8D, AiF13D, and AiF5D), respectively, in the Gunma prefecture or all Japan.[9] [10] [11] The number of autoimmune von Willebrand factor deficiency (AiVWFD) or autoimmune factor X deficiency (AiF10D) cases is too low to calculate the frequency.

In particular, there are currently no standardized or generalized methods for detecting anti-VWF autoantibodies (a-VWF-Ab). As a result, the a-VWF-Ab-positive rate (a-VWF-Ab positive cases/a-VWF-Ab tested cases, %) reported in Ichinose et al,[12] Tiede et al,[13] Mohri et al,[14] Siaka et al,[15] Franchi et al,[16] Dicke et al,[17] Dicke et al,[18] varied greatly from 0 to 80% depending on the investigators, i.e., the number of patients with a positive result for a-VWF-Ab/the number of patients who underwent a-VWF-Ab testing (percentage) was 3/11 (17%),[12] 4/27 (15%),[13] 8/25 (32%),[14] 8/10 (80%),[15] 9/23 (39%),[16] 0/14 (0%),[17] or 0/6 (0%)[18].

Therefore, it is very probable that the diagnosis, especially the differential diagnosis between AiVWFD and other types of acquired von Willebrand syndrome, might be insufficient or incorrect. AiVWFD diagnostic criteria were established by the Japanese Ministry of Health, Labor and Welfare (JMHLW), enacted as “Designated Intractable Disease 288-3,”[12] which are modeled based on the established diagnostic criteria of AiF13D, AiF5D, and AiF10D.[2] [6] [11] [19] Among governmental criteria, a-VWF-Ab positivity is a requirement for “definite diagnosis.”

Since 2012, the Japanese Collaborative Research Group (JCRG) has been conducting detailed examinations based on case consultations, referred by the attending physicians, of 13 patients suspected of having AiVWFD (P1–P13). VWF activity, measured by a ristocetin cofactor assay (VWF:Rco), was <6–37% in 12 of 13 patients and the remaining case P12 had VWF:Ag 49% ([Table 1]), which met the requirement of a “possible diagnosis” according to JMHLW diagnostic criteria.

Table 1 Summary of AiVWFD-suspected patients' information and VWF-related parameters

	Age	Sex	Underlying disease	Bleeding site/symptom	Hb (g/dL)	VWF:Rco (%)	VWF:Ag (%)	Spe. Act.[a]	F8:C (%)
					–	60–170	50–155	–	60–150
1	42	F	SLE	Nasal, SC, HM	9.9	<6	6	1.00	21.1
2	76	F	ITP	GI, SC, post-ope	10.4	<6	276	0.03	22
3	72	M	Waldenstrom MG	Urinary, nasal	7.9 [b]	<6	9	0.67	n.d.
4	70	F	Plasma cell dyscrasia	Oral	12.3	<6	11	0.55	6
5	14	F	None	HM, nasal	4.2	25	45	0.56	39
6	74	M	None	Oral, GI	7.6	<6	21	0.29	22
7	67	M	Polycythemia vera	IM	5.0	37	89	0.42	43
8	80	F	Common cold	Gingival	12.2	<6	7	0.86	17.8
9	67	F	ITP	Orbital, gingival	11.0	7	56	0.13	57
10	77	F	AL-amyloidosis	Oral, post-ope	9.6	<6	13	0.46	5
11	38	F	MALT-lymphoma, SS, hypothyroidism	Nasal, post-ope	10	16	35	0.46	20
12	59	F	None	IM, SC, post-ope	n.d.	53	49	1.08	41
13	90	M	Colon Ca.	IM, urinary	6.0	22	22	1,00	31

Abbreviations: Ag, antigen; AiVWFD, autoimmune von Willebrand factor deficiency; Ca., cancer; F, female; F8:C, factor VIII coagulant activity; GI, gastrointestinal; Hb, hemoglobin; HM, hypermenorrhea; IM, intramuscular; ITP, immune thrombocytopenic purpura; MG, macroglobulinemia; M, male; MALT, mucosa-associated lymphoid tissue; ope, operation; Rco, ristocetin cofactor activity; SC, subcutaneous; SLE, systemic lupus erythematosus; Spe. Act., specific activity (Rco/Ag); SS, Sjögren syndrome; VWF:Ag, von Willebrand factor antigen; VWF:Rco, von Willebrand factor ristocetin cofactor.

Note: F8:C values were measured in a JCRG-contracted commercial laboratory (SRL Ltd., Hachioji, Japan) and their reference ranges are shown in italics. Cases P11 and P12 had less than 50% VWF activity and/or antigen and are therefore suspected of having acquired VWF deficiency and also of AiVWFD. Cases P5 and P7 were excluded from further study because they had low VWF levels only once.

^a For calculation, Rco levels lower than the indicated values were considered as the indicated values per se (e.g., <6% as 6%).

^b Hb concentrations below 8.0 g/dL are included in the criteria for severe bleeding (ref.19) and are shown in bold.

The a-VWF-Ab was first detected by an immunoblot assay in two AiVWFD patients (P1[20] and P2[21]; data not shown) using a therapeutic commercial VWF/factor VIII concentrate (Confact-F, Japan Blood Products Organization, Tokyo, Japan). However, we have determined that it is difficult to quantify the amount of a-VWF-Ab with this antibody detection method. Therefore, we next tried to detect a-VWF-Ab using the immunochromatography test (ICT) developed as part of our AMED (Japan Agency for Medical Research and Development) research project. In strips coated with purified human VWF (PhVWF; Purified native Human VWF protein [Factor VIII Free], Fitzgerald Industries International, North Acton, Massachusetts, United States), samples from P1 and P2 were positive, whereas samples of P3, P4, P8, and P9 had reduced test line intensities similar to those of healthy control samples and were thus judged as negative ([Fig. 1A] and [Supplementary Fig. S1A] [online only]).

Fig. 1 Immunological detection of anti-von Willebrand factor antibodies (a-VWF-Abs). (A) Immunochromatography test (ICT) using purified human VWF (PhVWF). Samples except for P2 were diluted fivefold with a buffer (P2, 10-fold). Samples of P1 were collected from January 24, 2014 to April 22, 2016. (B) Sandwich-ICT using a homemade mouse anti-human VWF monoclonal antibody (mAb; 6–3E8E or 8–12H12G). Samples of P9 were collected on July 26, 2019, December 20, 2019, and February 4, 2020. (C) Conventional ELISA using PhVWF and 2% skim milk. Samples were diluted with 2% skim milk/phosphate buffered saline (PBS), as indicated. The cutoff value is 0.22 (broken line). (D) Conventional ELISA using PhVWF. Samples except for P2 were diluted twofold with 2% bovine serum albumin (BSA)/PBS (P2, 300- or 1,000-fold). The cutoff value is 0.03 (broken line). (E) Sandwich-ELISA using a homemade rat anti-human VWF mAb (7A12B9). Samples were diluted 100-fold with Sample Diluent (Zymutest; Hyphen BioMed, Neuville sur Oise, France). The cutoff value is 0.89 (broken line). (F) Conventional ELISA using PhVWF. Samples were diluted 100-fold with Sample Diluent (Zymutest). The cutoff value is 0.61 (broken line). (G) Conventional ELISA using rhVWF (recombinant hVWF). Samples were diluted 25-fold with 2% BSA/tris buffered saline (TBS). HC, healthy control; NC, negative control.

Later attempts to detect VWF/a-VWF-Ab complexes with sandwich-ICT using strips coated with a homemade mouse anti-human VWF monoclonal antibody (mAb; 6–3E8E or 8–12H12G) resulted in the P2 sample positive, but P9 and P10 were negative ([Fig. 1B]).

Since our AMED research project was completed in 3 years, we reverted to the conventional enzyme-linked immunosorbent assay (ELISA) to detect a-VWF-Ab using a 96-well microtiter plate coated with PhVWF at 200 ng/well. In line with the ICT results, samples of P1 and P2 were positive, while those of P3, P4, and P9–11 were negative ([Fig. 1C], [D] and [Supplementary Fig. S1B] [online only]). In contrast, whereas a sample of P8 was negative when tested by ICT, it was positive when tested by ELISA (slightly higher than the cutoff value).

Next, we attempted to detect the VWF/a-VWF-Ab complex by sandwich-ELISA using a homemade rat anti-human VWF mAb (7A12B9) as a capture antibody.[22] Only the P6 sample was clearly positive, whereas P1, P2, P4, P8, P12, P13, and P9–11 samples were judged as negative ([Fig. 1E] and [Supplementary Fig. S1C] [online only]). However, when coating with 50 ng/well of PhVWF, both P2 and P6 samples became a-VWF-Ab-positive ([Fig. 1F]). Nevertheless, it has been argued that PhVWF is unsuitable for a-VWF-Ab detection because it contains ABO blood group A and B antigens and that recombinant human VWF antigen (rhVWF) produced in animal cells is preferable.[23] [24] In 2014, Franchi et al prototyped a new ELISA system using rhVWF and reported that 9 of 23 (39%) patients were positive,[16] whereas Dicke et al did not detect positive cases among 14 cases in 2014 and 6 cases in 2016.[17] [18] Most recently, we obtained this product, specifically von Willebrand factor/vWF Protein, Human, Recombinant (His Tag) expressed in CHO stable cells (Sino Biological Inc., Beijing, China), and applied it to ELISA. As a result, among the five patient samples available at this time, only P8 was positive, whereas P1, which was previously judged as a-VWF-Ab-positive, was negative ([Fig. 1G]).

Unfortunately, we cannot explain these conflicting results at this time given that the measurement method is identical except that rhVWF was used as an antigen instead of PhVWF. Since the previously reported positive rate of ELISA using rhVWF as an antigen varies widely,[16] [17] [18] this might suggest diversity in the target sites and properties of autoantibodies against VWF at the individual level.

Finally, judging from the vicissitude of clinical bleeding symptoms and VWF parameters, including VWF activity and antigen levels, in the two previously reported a-VWF-Ab-positive patients (P1 and P2) [20, 21; a-VWF-Abs identified in other laboratories] and two unpublished a-VWF-Ab-positive patients (P6 and P8), the authors believe that these four cases definitely retained a-VWF-Ab. Nonneutralizing a-VWF-Ab is the cause of this disease for at least one Japanese AiVWFD patient[20] and another patient reported in Canada.[25] It is impossible to diagnose all AiVWFD patients by the functional test method to detect the VWF inhibitor. Therefore, to avoid overlooking AiVWFD cases, carrying out a-VWF-Ab detection for all suspected AiVWFD cases is indispensable. The authors hope that more reliable antibody detection methods will be developed and marketed in the near future.

In conclusion, the authors propose that the patient sample is “a-VWF-Ab-positive” when two or more different (1) methods, such as ICT and ELISA (and immunoblotting), (2) antigens (plasma-derived or recombinant VWF) for free a-VWF-Ab, (3) antibodies (mouse or rat mAb) to capture VWF/a-VWF-Ab complexes, (4) sample diluents, (5) laboratories, etc. give positive results, because so far there has been no single reliable a-VWF-Ab detection method.

Authors' Contributions

A.I. initiated and designed the study, wrote, edited, and proofread the manuscript. T.O., M.S., and C.Y. performed experimental examinations and proofread the manuscript. Y.M. developed and supplied ICT test kits, and proofread the manuscript.

Publication History

Received: 13 July 2022

Accepted: 17 December 2022

Accepted Manuscript online:
20 December 2022

Article published online:
25 January 2023

• References

• 1 Cugno M, Gualtierotti R, Tedeschi A, Meroni PL. Autoantibodies to coagulation factors: from pathophysiology to diagnosis and therapy. Autoimmun Rev 2014; 13 (01) 40-48 Erratum in: Autoimmun Rev. 2015 Jul;14(7):650
• 2 Ichinose A. Immune-mediated acquired coagulation factor deficiencies: state-of-the-art in diagnosis and management [in Japanese]. Rinsho Ketsueki 2019; 60 (06) 667-679
• 3 Ahmed AE. Autoantibodies to coagulation factors and bleeding disorders. Clin Rev Allergy Immunol 1998; 16 (03) 313-319
• 4 Macik BG, Crow P. Acquired autoantibodies to coagulation factors. Curr Opin Hematol 1999; 6 (05) 323-328
• 5 Boggio LN, Green D. Acquired hemophilia. Rev Clin Exp Hematol 2001; 5 (04) 389-404 , quiz 431
• 6 Ichinose A. Japanese Collaborative Research Group on AH13. Autoimmune acquired factor XIII deficiency due to anti-factor XIII/13 antibodies: a summary of 93 patients. Blood Rev 2017; 31 (01) 37-45
• 7 Osaki T, Souri M, Ichinose A. Plasma proteomics associated with autoimmune coagulation factor deficiencies reveals the link between inflammation and autoantibody development. Int J Hematol 2022; 115 (05) 672-685
• 8 Osaki T, Souri M, Ichinose A. Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII deficiency via whole-exome sequencing analysis. PLoS One 2021; 16 (09) e0257322
• 9 Ogawa Y, Yanagisawa K, Uchiumi H. et al. Clinical characteristics and outcomes of acquired hemophilia A: experience at a single center in Japan. Int J Hematol 2017; 106 (01) 82-89
• 10 Ichinose A, Osaki T, Souri M. Autoimmune acquired factor XIII deficiency in Japan 2021 update: focused on annual incidence and clinical features. Haemophilia 2022; 28 (05) e121-e124
• 11 Ichinose A, Osaki T, Souri M. A review of coagulation abnormalities of autoimmune acquired factor V deficiency with a focus on Japan. Semin Thromb Hemost 2022; 48 (02) 206-218
• 12 Ichinose A, Osaki T, Souri M, Favaloro EJ. A review of autoimmune acquired von Willebrand factor deficiency in Japan. Semin Thromb Hemost 2022; 48 (08) 911-925
• 13 Tiede A, Priesack J, Werwitzke S. et al. Diagnostic workup of patients with acquired von Willebrand syndrome: a retrospective single-centre cohort study. J Thromb Haemost 2008; 6 (04) 569-576
• 14 Mohri H, Motomura S, Kanamori H. et al. Clinical significance of inhibitors in acquired von Willebrand syndrome. Blood 1998; 91 (10) 3623-3629 Erratum in: Blood 1999;93(1):413
• 15 Siaka C, Rugeri L, Caron C, Goudemand J. A new ELISA assay for diagnosis of acquired von Willebrand syndrome. Haemophilia 2003; 9 (03) 303-308
• 16 Franchi F, Biguzzi E, Stufano F, Siboni SM, Baronciani L, Peyvandi F. A two-step approach (enzyme-linked immunosorbent assay and confirmation assay) to detect antibodies against von Willebrand factor in patients with acquired von Willebrand syndrome. Thromb Res 2014; 134 (06) 1316-1322
• 17 Dicke C, Holstein K, Schneppenheim S. et al. Acquired hemophilia A and von Willebrand syndrome in a patient with late-onset systemic lupus erythematosus. Exp Hematol Oncol 2014; 3: 21
• 18 Dicke C, Schneppenheim S, Holstein K. et al. Distinct mechanisms account for acquired von Willebrand syndrome in plasma cell dyscrasias. Ann Hematol 2016; 95 (06) 945-957
• 19 Ichinose A, Kohler HP, Philippou H. Factor XIII and Fibrinogen SSC Subcommittee of the ISTH. Recommendation for ISTH/SSC Criterion 2015 for autoimmune acquired factor XIII/13 deficiency. Thromb Haemost 2016; 116 (04) 772-774
• 20 Ihara A, Suzuki N, Matsushita T, Ichinose A. Acquired von Willebrand syndrome in a patient with immune thrombocytopenic purpura [in Japanese]. Rinsho Ketsueki 2015; 56 (07) 901-904
• 21 Kobayashi N, Ogawa Y, Yanagisawa K. et al. Acquired immune-mediated von Willebrand syndrome accompanied by antiphospholipid syndrome [in Japanese]. Rinsho Ketsueki 2017; 58 (06) 613-618
• 22 Yokoyama C, Ikeda S, Osaki T, Souri M, Ichinose A. Generation and application of rat monoclonal antibodies specific for a human blood coagulation protein: von Willebrand factor. Monoclon Antib Immunodiagn Immunother 2019; 38 (03) 133-136
• 23 Tiede A, Rand JH, Budde U, Ganser A, Federici AB. How I treat the acquired von Willebrand syndrome. Blood 2011; 117 (25) 6777-6785
• 24 James AH, Eikenboom J, Federici AB. State of the art: von Willebrand disease. Haemophilia 2016; 22 (Suppl 5): 54-59
• 25 Lee A, Sinclair G, Valentine K, James P, Poon MC. Acquired von Willebrand syndrome: von Willebrand factor propeptide to von Willebrand factor antigen ratio predicts remission status. Blood 2014; 124 (05) e1-e3