Patients with Immune Thrombocytopenia Have Higher Proportion of Ultra Large VWF Multimers and Lower ADAMTS13 Activity Than Healthy Controls

Background Despite low platelet count and increased bleeding risk, patients with immune thrombocytopenia (ITP) have higher risk of thrombosis compared to healthy population. This risk seems to be further increased by some ITP therapies like Thrombopoietin receptor agonists (TPO-RA). The underlying mechanisms for the prothrombotic nature of ITP and TPO-RA treatment have been investigated; however, e roles of von Willebrand factor (VWF) multimers and VWF cleaving protease ADAMTS13 have not been fully explored. Aims To measure VWF multimer distribution and ADAMTS13 activity in ITP patients and to study the effect of TPO-RA treatment on these parameters. Methods Blood samples from ITP patients (n=15) before initiating TPO-RA treatment and after achieving response, defined as platelet count >50x109/L, and healthy controls (n=15) were obtained. Gel electrophoresis (the HYDRAGEL VW multimer assay) was performed followed by densitometric visualization of the VWF multimers. Conventionally, the multimers are divided into three fractions: low (LMWM), intermediate (IMWM) and high molecular weight multimers (HMWM). The area under the curve (AUC) is calculated for all three fractions. To better separate the ultra large molecular weight multimers (ULMWM) from the HMWM fraction, we constructed an alternative measure (AM) to describe the amount of ULMWM in patients compared to pooled normal plasma (PNP). AM is a horizontal line drawn from the start of the HMWM fraction to the end of the multimer pattern at a level where the amplitude of the PNP is 50% at the end of the multimer pattern. By using this AM to assess the lateral width instead of the conventional percentage HMWM, we are able to visualize the presence of ULMWM. The AM of the patient is compared with the AM of the PNP and expressed in relative width (%) to the migration distance of PNP. ADAMTS13 activity was measured with a chemiluminescent immunoassay on the Bioflash instrument. Statistical analysis was performed with MedCalc. Differences between two groups were tested with t-test for normally distributed data. Non-normally distributed data was log-transformed before statistical analysis was performed. Results Median age for patients was 58 years (interquartile range (IQR): 43-68), controls 55 years (IQR: 45-58); 7/15 were females in both groups. Visual inspection of VWF multimers on gel electrophoresis revealed that many of the ITP patients had a different VWF multimer pattern compared with healthy controls. The densitometric profiles showed that ITP patients displayed a wider lateral spread of multimers in the cathodal region, suggesting a higher degree of ULMWM. ITP patients before initiating TPO-RA had a higher amount of ULMWM (mean percentage AM: 16.6% (standard deviation (SD) 12.3)), compared with controls (mean percentage AM: 6.1% (SD 7.7)), p=0.009. There was no significant difference in ULMWM between ITP patients before initiation of TPO-RA (mean percentage AM: 16.6% (SD 12.3)), and after initiation in responders (mean percentage AM: 16.5% (SD 7.7)), p=0.956. ADAMTS13 activity (reference range: 61-131%) was significantly lower in ITP patients before initiation of TPO-RA treatment (mean: 97.4% (SD 28.1)) compared with controls (mean: 114.4 (SD 10)), p=0.019; but there was no difference for ITP patients before the initiation of TPO-RA (mean: 97.4% (SD 28.1)) compared to after response (mean: 97.4% (SD 24.5)), p=0.990. Conclusion ITP patients had significantly higher proportion of ULMWM than healthy controls. ULMWM are known to be procoagulant and can contribute to the hypercoagulable state in ITP. The origin of the higher proportion of ULMWM may be due to increased endothelial cell activation that previously has been shown in ITP, but platelets cannot be excluded as the source. Lower ADAMTS13 activity in ITP patients, that may be due to inflammation, may also contribute to higher levels of ULMWM. TPO-RA treatment was not associated with significant changes in the distribution of VWF multimers or ADAMTS13 activity suggesting that this treatment dose not contribute to hypercoagulability through a VWF-mediated mechanism.