Excessive cleavage of von Willebrand factor multimers by ADAMTS13 may predict the progression of transplant-associated thrombotic microangiopathy

Background Transplant-associated thrombotic microangiopathy (TA-TMA) is a fatal complication of hematopoietic stem cell transplantation and is characterized by severe thrombocytopenia, hemolytic anemia, and organ dysfunction. In response to several possible triggers, dynamic multimetric change in von Willebrand factor (VWF) may contribute to inducing microthrombi in circulation in TA-TMA. Objectives By performing VWF multimer analysis and measuring VWF-degradation product (DP), we unraveled the relationship between multimeric changes in circulating VWF and the pathogenesis of TA-TMA. Methods This study analyzed 135 plasma samples from 14 patients who underwent allogeneic hematopoietic stem cell transplantation at a single institute. VWF-associated markers, namely VWF:antigen (VWF:Ag), VWF-DP/VWF:Ag ratio, VWF:ristocetin cofactor activity, VWF:ristocetin cofactor activity/VWF:Ag ratio, and ADAMTS13 activity, were analyzed in these samples collected every 7 days. Results There were 2 patients with definite thrombotic microangiopathy (TMA) and 6 patients who presented with probable TMA that did not progress to definite TMA. Each plasma sample was classified into 3 groups: definite TMA, probable TMA, and non-TMA. VWF multimer analysis showed the absence of high-molecular-weight VWF multimers in probable TMA, whereas the appearance of unusually large VWF multimers was observed in definite TMA. The median value of the VWF-DP/VWF:Ag ratio in probable TMA was elevated to 4.17, suggesting that excessive cleavage of VWF multimers by VWF cleaving enzyme, ADAMTS13, resulted in the loss of high-molecular-weight VWF multimers. Conclusion During the transition from probable to definite TMA, drastic VWF multimer changes imply a switch from bleeding to thrombotic tendencies. Extensive VWF-DP and VWF multimer analyses provided novel insights.


| I N T R O D U C T I O N
Transplant-associated thrombotic microangiopathy (TA-TMA) occurs after hematopoietic stem cell transplantation (HSCT) and presents as microangiopathic hemolytic anemia, consumptive thrombocytopenia, and organ dysfunction due to microcirculatory insufficiency [1,2].
Various factors, such as conditioning therapy, immunosuppressive drugs, and allogeneic immune responses, cause vascular endothelial cell damage and platelet thrombus formation [3].With various diagnostic criteria for TA-TMA [4][5][6][7], the incidence and mortality rates of TA-TMA in allogeneic HSCT are 10% to 25% [8] and 60% to 75% at 3 months, respectively [4,9].von Willebrand factor (VWF) is a hemostatic glycoprotein produced in endothelial cells that is secreted as an unusually large (UL) multimer during vascular injury [10].UL-VWF multimers can be cleaved into smaller multimers by a disintegrin-like metalloproteinase with thrombospondin type 1 motif 13 (ADAMTS13) at the peptide bond between Tyr1605-Met1606 of the VWF A2 domain [11], thereby preventing excessive platelet aggregation.Since larger VWF multimers possess higher platelet aggregation capacity [12], insufficient cleavage of UL-VWF multimers due to a deficiency in ADAMTS13 activity results in lethal thrombosis in patients with thrombotic thrombocytopenic purpura (TTP).In contrast, there are conditions under which high-molecular-weight (HMW)-VWF multimers are reduced or absent.There are 2 types of VWF abnormalities: von Willebrand disease type 2A [13] and 2B [14], in which the production of HMW-VWF multimers is reduced or consumptively reduced, and acquired von Willebrand syndrome.Aortic stenosis [15], left ventricular assist devices [16], percutaneous cardiopulmonary support [17], and extracorporeal membrane oxygenation [18] are some of the conditions that can cause acquired von Willebrand syndrome and severe bleeding symptoms.
There have been several reports on the association of VWF and ADAMTS13 with complications of allogeneic HSCT; an elevated VWF:antigen (VWF:Ag; ≥325%) is a predictive marker for the development of TA-TMA, and an elevation of VWF:Ag before 37 days (IQR, 20-125 days) of the TA-TMA leads to a definitive diagnosis [19].
An increase in VWF:Ag levels on day 7 after HSCT was a predictive marker of the onset of TA-TMA [20].All these reports have investigated the relationship between VWF and pathologic conditions  • von Willebrand factor (VWF) multimer and its degradation product were studied in a small cohort.
• The loss of high-molecular-weight VWF multimer was observed just prior to the onset of TA-TMA.
• The excessive cleavage of VWF multimers by ADAMTS13 may trigger the development of TA-TMA.indirect Coombs test results.In this study, "definite thrombotic microangiopathy (TMA)" was defined when all 4 BMT-CTN criteria were met."Probable TMA" was defined as TMA that met all of the BMT-CTN criteria except for concurrent renal and/or neurologic dysfunction without other explanations.The test results at each blood sampling point were used to determine the probability of a patient having a TMA occurrence.Based on the test results at each blood collection point, we determined whether each sample could be classified as non-TMA, probable TMA, or definite TMA.

| Measurements of VWF-associated biomarkers
Protocols for the measurement of VWF-associated biomarkers have been described elsewhere.Briefly, VWF:Ag levels were measured by sandwich ELISA using rabbit antihuman VWF polyclonal antibodies (Dako Cytomation) [21].VWF:ristocetin cofactor (RCo) measurements were performed on a CS analyzer (Sysmex UK Ltd) using the BC von Willebrand Reagent (Siemens Healthineers).ADAMTS13 activity levels were measured via a chromogenic ADAMTS13-act-ELISA [22] (Kainos Laboratories).Pooled normal plasma was produced from 20 healthy volunteers, and then, the values of 100% VWF:Ag and ADAMTS13 levels were assigned to this pool.The World Health Organization Sixth International Standard for blood coagulation factor VIII and VWF was used as the standard plasma for VWF:RCo measurements.VWF-DP levels were measured using sandwich ELISA with N10, a monoclonal antibody recognizing a decapeptide (1596-DREQAPNLVY-1605) derived from the VWF-A2 domain, and an anti-VWF monoclonal antibody recognizing the N-terminal region of VWF, developed in our laboratory [23].The VWF-DP/VWF:Ag ratio was obtained by dividing the measured VWF-DP in each sample by VWF:Ag.

| VWF multimer analysis and VWF multimer index
VWF multimer analysis was performed as previously described [14] with some modifications [24,25].VWF:Ag level in each sample was not equal; however, to avoid technical errors due to the dilution process, the VWF:Ag was not matched to the same level by dilution.VWF multimer bands were classified into 4 groups and defined as follows: low molecular weight (LMW), including 5 bands at the bottom; intermediate molecular weight with the next 5 bands; HMW with upper bands above 10 bands; UL, which is not found in healthy individuals.
The VWF multimer index was measured according to the method described by Tamura et al. [26].Using ImageJ software (National Institutes of Health), the center of the lane was scanned, and the multimers were divided into 3 parts (LMW, intermediate molecular weight, and HMW + UL-VWF multimers).First, the VWF multimer ratio was determined as the ratio of the VWF multimer area of each part to the total VWF multimer area.Next, the VWF multimer ratio of each part in normal plasma, which was electrophoresed simultaneously, was determined.The VWF multimer index was defined as the percentage of the VWF multimer ratio in each part of the sample that corresponded to the VWF multimer ratio in each part of normal plasma (Supplementary Figure).

| Statistical analysis
Comparisons among the 3 groups (non-TMA, probable TMA, and definite TMA) were performed using the Kruskal-Wallis test, and comparisons between each group were performed using the Steel-Dwass test for those that showed significant differences among the 3 groups.Correlations were analyzed using Spearman's rank correlation coefficient test.Statistical significance was set at P < .05.All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University) [27], which is a graphical user interface for R (R Foundation for Statistical Computing).More precisely, it is a modified version of R Commander, designed to add statistical functions frequently used in biostatistics.

| VWF multimer analysis and diagnosis of TA- TMA
VWF multimer analysis and VWF-associated markers for the 2 cases that progressed from probable to definite TMA are shown in Figure 3.The VWF:antigen (VWF:Ag) level was not significantly different among the 3 groups.(B) ADAMTS13 activity was significantly lower in the definite TMA group than in the non-and probable TMA groups.(C) VWF:Ag/ADAMTS13 activity was significantly higher in the definite TMA group than in the non-and probable TMA groups.(D) The VWF-degradation product (DP) was significantly lower in the definite TMA group than in the non-and probable TMA groups.(E) The VWF-DP/VWF:Ag ratio was significantly lower in the definite TMA group than in the nonand probable TMA groups.(F) The VWF:ristocetin cofactor activity (RCo) was significantly higher in the definite TMA group than in the nonand probable TMA groups.(G) The VWF:RCo/VWF:Ag was significantly different among the 3 groups, with the highest to lowest being the definite, non-, and probable TMA groups.(H) The high-molecular-weight (HMW) + unusually large (UL)-VWF multimer index was significantly higher in the definite TMA group than in the non-and probable TMA groups.(I) The low-molecular-weight (LMW)-VWF multimer index was significantly lower in the definite TMA group than in the non-and probable TMA groups.*P < .05,**P < .01.
Figure 4 shows VWF multimer analysis and VWF-associated markers for the 2 cases.One patient presented with probable TMA but did not develop definite TMA (Figure 4A).The other patient presented with neither probable nor definite TMA (Figure 4B).UL-VWF multimers were present at only 5 points within the definite TMA group.

| D I S C U S S I O N
To our knowledge, this is the first report to investigate the association between TA-TMA onset, VWF-associated markers (VWF:Ag, VWF-DP/VWF:Ag ratio, VWF:RCo, VWF:RCo/VWF:Ag, and ADAMTS13 activity), and VWF multimers over time and in detail.Unlike previous reports [19,20,28,29], our report found that VWF:Ag elevation alone was not a predictive marker for the development of TA-TMA.In addition, VWF:Ag was consistently elevated regardless of the TA-TMA status; VWF:Ag may be more influenced by endothelial damage derived from conditioning chemotherapy, total body irradiation, or calcineurin inhibitors [8] than by TA-TMA.Intriguingly, VWF multimer analysis revealed drastic changes in VWF-related markers in patients with TA-TMA.Upon transition from non-TMA to probable TMA, the HMW-VWF multimer was absent, and VWF:RCo/VWF:Ag was elevated (UPN-7 days 21 and 28; UPN-10 day 14; UPN-14 days 28, 35, and 42).Furthermore, upon transition from probable TMA to definite TMA, the appearance of the UL-VWF multimer and an increase in VWF:RCo/VWF:Ag were observed (UPN-7 days 35, 42, and 49; UPN-10 days 21 and 28).Interestingly, this movement of VWF-associated markers was also observed during recovery from a definite TMA.In other words, during recovery from definite TMA to probable TMA, there was a loss of UL-VWF multimers, an absence of HMW-VWF multimers, and a decrease in VWF:RCo/VWF:Ag; during recovery from probable TMA to non-TMA, there was a reappearance of HMW-VWF multimers and normalization of VWF:RCo/VWF:Ag (UPN-10 days 35 and 42).Similar VWF multimer behavior has been reported in SOS/VOD [30] caused by damage to sinusoidal endothelial cells [31].Although TMA and SOS/VOD are not the same disease, in that TMA is caused by vascular endothelial cell injury [32], it is interesting to observe similar VWF multimer behavior in the 2 thrombotic diseases after HSCT.The reason for the loss of the HMW-VWF multimer was not revealed in the SOS/VOD report [30].VWF-DP, developed in our laboratory, is a measuring system used to quantify VWF cleavage by ADAMTS13, and a high VWF-DP/VWF:Ag ratio indicates increased VWF cleavage by ADAMTS13 [23].Another disease with dynamic VWF multimer behavior is TTP.In TTP, HMW-VWF multimer is deficient in the acute phase, and HMW-VWF multimer increases immediately before recurrence.HMW-VWF multimer is usually consumed by massive thrombus formation in the acute phase [33,34], which is different from the mechanism of HMW-VWF multimer loss in TA-TMA proposed in our study.
VWF-DP/VWF:Ag was elevated in probable TMA but significantly decreased in definite TMA.HMW-VWF multimer defects after HSCT were derived from the excessive cleavage of the VWF multimer by ADAMTS13.
Our results led to 2 hypotheses regarding the role of VWF multimers in TA-TMA progression.In the first mechanism, microthrombus formation due to vascular endothelial damage causes stenosis of the vascular lumen.This stenosis generates shear stress [35], which stretches the VWF multimeric structure, leading to cleavage by ADAMTS13 [36].Even in patients without TMA who do not meet the diagnostic criteria for probable or definite TMA, stenosis of the vascular lumen can cause this pathology when microthrombi form due to endothelial damage.This can be explained by the increased VWF-DP/VWF:Ag ratios in the non-TMA and probable TMA groups.and 28 met the diagnostic criteria for definite TMA, and recovery from definite to probable TMA was seen on day 28.HMW-VWF multimer was deficient during probable TMA, and the unusually large VWF multimer appeared during definite TMA.The loss of HMW-VWF multimer was thought to be caused by ADAMTS13 cleavage of the VWF multimer based on the increased VWF-DP/VWF:Ag ratio.Transplant-associated TMA improved in this patient, but she died due to the recurrence of the primary disease.In both cases, there was a point of HMW-VWF multimer loss that did not meet the diagnostic criteria for probable TMA.NP, normal plasma; VWF:RCo, von Willebrand factor ristocetin cofactor activity.
considered.A multicenter prospective validation study, including records of thrombotic/hemorrhagic events in a larger number of patients, should be performed in the future.In addition, complement is important in TA-TMA.In our analysis of VWF-associated markers, citrate plasma was collected and stored; however, for complement analysis, EDTA plasma must be stored at −80 • C or measured immediately after blood collection.Although complement measurement was not possible in this study, it is an important issue to be considered in the future.
Our findings suggest a potential treatment strategy for TA-TMA.
There have been reports suggesting that treatment options for TA-TMA include reduction or discontinuation of calcineurin inhibitors that exacerbate vascular endothelial damage, control of infection and graft-vs-host disease [4], and recombinant thrombomodulin [38] and fresh frozen plasma infusion [30].However, the efficacy of plasma exchange remains controversial [39].Completion of TA-TMA leads to organ ischemia, multiorgan failure, and death.To prevent probable TMA from turning into definite TMA and to prevent the transition from definite TMA to organ ischemia and multiorgan failure, treatment that does not induce UL-VWF multimers or eliminate or degrade UL-VWF multimers may be an effective option for definite TMA.
Recombinant ADAMTS13 [40] may improve the pathophysiology of definite TMA by degrading the UL-VWF multimer but may not improve the pathophysiology of probable TMA and may induce bleeding due to HMW-VWF multimer deficiency in probable TMA.
Repeated fresh frozen plasma infusions supplemented with ADAMTS13 and appropriately sized multimers may be the best TA-TMA therapy currently available.In addition, as a measurement of the clinical response, a decrease in VWF:RCo/VWF:Ag may be used as an indicator of improvement from definite TMA to probable TMA, and an increase in VWF:RCo/VWF:Ag may be used as an indicator of improvement from probable TMA to non-TMA.

2 . 1 |
involving endothelial damage and microcirculatory disturbance.A VWF-degradation product (DP) enzyme-linked immunosorbent assay (ELISA) was recently developed in our laboratory to quantify the amount of cleavage of VWF multimers by ADAMTS13.The measurement of VWF-DP explains the mechanism of VWF multimer changes.Little is known about the behavior of VWF-DP and VWF multimer patterns in patients with TA-TMA.Thus, this study investigated the association between temporal changes in VWF-associated biomarkers, including VWF-DP, in patients with and without TA-TMA.2 | M E T H O D S Patients and blood sampling Japanese patients aged 20 years or older who underwent allogeneic HSCT for hematopoietic malignancies between June 2020 and March 2021 at the Department of Hematology, Toyama Prefectural Central Hospital, were included.Peripheral whole blood was collected with sodium citrate as an anticoagulant every 7 days, starting on day −7 until a maximum of 63 days or until discharge from the hospital, with the day of allogeneic HSCT as day 0.However, deviations of up to ±2 days were allowed at each blood collection point.All samples were stored at −80 • C, thawed at 37 • C immediately before testing, and were not refrozen.The study protocol complied with the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committees of Nara Medical University (#2442) and Toyama Prefectural Central Hospital (#58-62).Written informed consent was obtained from all the participants.

2. 2 |
Diagnosis of TA-TMA The diagnosis of TA-TMA is based on the Blood and Marrow Transplant Clinical Trial Network (BMT-CTN) criteria [4]: (1) red blood cell fragmentation and ≥2 schistocytes per high-power field on peripheral smear, (2) concurrent increased serum lactate dehydrogenase level above institutional baseline, (3) concurrent renal and/or neurologic dysfunction without other explanation, and (4) negative direct and Essentials • Transplant-associated thrombotic microangiopathy (TA-TMA) is a fatal thrombotic complication.

3 |
R E S U L T S 3.1 | Patient background Blood samples were drawn from all the patients at 135 points.Patient backgrounds are shown in the Table.The median age was 57 years (37-69 years), and the male-to-female ratio was 1:1.The primary diseases were acute myeloid leukemia in 5 patients, acute lymphoid leukemia in 2 patients, myelofibrosis in 2 patients, myelodysplastic neoplasms in 2 patients, adult T-cell lymphoma/leukemia in 1 patient, mixed-phenotype acute leukemia in 1 patient, and blast phase of chronic myeloid leukemia in 1 patient.Of the 14 patients, 3 underwent transplantation in complete remission (CR), and 11 underwent transplantation in non-CR.The ABO blood types were major mismatches in 3 cases, minor mismatches in 5 cases, major minor mismatches in 1 case, and matches in 5 cases.The transplant source was related peripheral blood in 1 case, unrelated peripheral blood in 6 cases, unrelated bone marrow in 2 cases, and cord blood in 5 cases.Conditioning included fludarabine, melphalan, and busulfan in 13 patients and total body irradiation/cyclophosphamide in 1 patient.Graft-vs-host disease prophylaxis included tacrolimus + short-term methotrexate in 3 patients, tacrolimus + mycophenolate mofetil in 4 patients, and tacrolimus + mycophenolate mofetil + antithymocyte globulin in 7 patients.No grade III to IV graft-vs-host disease, sepsis, hemophagocytic syndrome, or sinusoidal obstruction syndrome/venoocclusive disease (SOS/VOD) occurred during the observation period.

F
I G U R E 1 Comparison of von Willebrand factor (VWF)-associated markers in nonthrombotic microangiopathy (TMA), probable TMA, and definite TMA groups.In all 135 blood sampling points, 5 samples were classified as definite TMA, 27 as probable TMA, and 103 as non-TMA.(A)
U R E 3 von Willebrand factor (VWF) multimer and VWF-associated markers in 2 patients who developed definite thrombotic microangiopathy (TMA).(A) Data from unique patient number (UPN)-7 are shown.On days 14-to 28, patients met the diagnostic criteria for probable TMA and had high-molecular-weight (HMW)-VWF multimer defects, while on days 35 to 49, an unusually large VWF multimer appeared during definite TMA.The loss of HMW-VWF multimer was thought to be caused by ADAMTS13-induced cleavage of VWF multimer, based on the increased VWF-degradation product (DP)/VWF:antigen (VWF:Ag) ratio.The patient died of multiorgan dysfunction due to transplant-associated TMA on day 49.(B) Data from UPN-10 are shown.Days 7 and 14 met the diagnostic criteria for probable TMA, days 21

5 |
C O N C L U S I O N SDespite a small sample size in a single institute, we found a dynamic change in VWF multimer analysis from the loss of HMW-VWF multimers to the appearance of UL-VWF multimers during the transition from probable to definite TMA, suggesting that excessive cleavage of VWF multimers by ADAMTS13 may predict the progression of TA-TMA.Based on the VWF multimer analysis, it is desirable to develop safe TA-TMA therapy from the viewpoint of both hemorrhage and thrombosis.