Comparison of hemosil and innovance antithrombin assays using Hemophilia A plasma and antithrombin lowering (to mimic fitusiran)

Abstract Introduction: There are several assays for measuring antithrombin (AT) activity that each use different reagents, targeting human or bovine factor Xa or human factor IIa. As these assays were developed to detect physiologic (not pharmacologic) antithrombin deficiency, their lower limit of detection were set fairly high (~20-35%). Fitusiran is an AT lowering therapeutic that increases thrombin generation to restore hemostasis in people with hemophilia. It was recently approved for prophylaxis in persons with hemophilia A or B, with or without inhibitors, and requires AT monitoring including accurate measurement of AT levels between 15-35%, the target therapeutic range. Previously, a field study by Chhabra et al. demonstrated that only the Siemens INNOVANCE assay (INN) measured AT levels in plasma from patients on fitusiran accurately at all clinical decision points. Using different methodology than Chhabra, we compare the differences between the INN assay and the HemoSIL liquid AT assay (SIL) at various AT levels. Methods: Plasma from patients with moderate or severe hemophilia A was collected and baseline AT activity was measured using INN and SIL. Samples were spiked with different concentrations of anti-AT antibody (determined through a previously established standard curve that was created using SIL reagents) to achieve targeted AT levels (75%, 50%, 35%, 25%, 15%, 10%). Samples were tested locally using the INN and SIL assays. Summary statistics were computed for each assay at the target AT levels. Paired, one-sample t-tests were used to assess statistical evidence of the difference at the 0.05 level without adjustment for multiple testing. Results: Plasma samples from 21 males with severe hemophilia A [median age 17 years (range 5-44 years)] were tested using both SIL and INN AT assays. The mean baseline levels were higher in SIL (112.6%) compared to INN (109%), p = 0.02. This difference was also seen at AT levels of 75% (SIL: 74.9%, INN: 71.1%, p= 0.01). There was no difference at 50% AT (SIL: 49.3%, INN: 49.9%, p= 0.63) and at 35% AT (SIL: 37.2%, INN: 38.6%, p= 0.29). There were significant discrepancies at target AT levels of 25%, 15%, and 10% AT with SIL consistently reporting lower mean compared to INN: 28.5% vs. 34.1% (p<0.001), 21.3 % vs. 28.6% (p<0.001), and 19.6 vs. 26.3% (p<0.001), respectively. True AT levels of 15% and 10% were not achieved with our AT depletion method. Conclusions: The SIL and INN AT assays are both licensed for the detection of physiologic AT deficiency. Both function well and demonstrate good concordance at higher AT levels (≥50%), but significant discrepancies arise at lower AT levels, particularly below 25%. Specifically, the INN assay, the only one approved in the US for fitusiran monitoring, consistently reports higher AT activity compared to SIL at lower target levels. This data confirms data from the field study by Chhabra, though the methods in this study are different. Inferring from this study, we believe that our data demonstrating lower levels of AT with SIL are in fact the inaccurate levels and therefore laboratories using the SIL assay will likely underestimate AT activity in patients on fitusiran, which could result in potential inaccurate dosing of fitusiran. Clinicians should be aware of assay specific differences when interpreting AT results, especially at low AT concentrations and should follow the prescribing information and only use the INN assay when measuring AT levels in patients on fitusiran prophylaxis.