Collaborative Study On Monitoring Methods To Determine Direct Thrombin Inhibitors Lepirudin And Argatroban On behalf of the Control of Anticoagulation Subcommittee of the Scientific and Standardization Committee of the International Society of Thrombosis and Haemostasis E. Gray 1 and J Harenberg * SUMMARY Background: Direct thrombin inhibitors are now used clinically for the prophylaxis and treatment of thrombosis and related cardiovascular diseases. Although all these inhibitors bind to thrombin, their modes of action are different. Routine anticoagulant monitoring tests such as the APTT are used to estimate the activity of these inhibitors. However, no studies have been carried out to assess the suitability of these tests. Methods: An international collaborative study was carried out utilising a panel of plasmas spiked with lepirudin and argatroban to evaluate the robustness and the sensitivity of the different monitoring methods. Results: Thirteen laboratories took part in the trial. The point-of-care TAS-analyser, using cards with high or low concentrations of ecarin reagents gave the lowest intra- and inter- laboratory variation. APTT using local or common APTT reagents also showed acceptable within and between-laboratory variations. The chromogenic anti-IIa activity and the wet ecarin clotting time gave higher inter-laboratory variations. The ELISA for lepirudin had the highest inter-laboratory variation. Conclusion: APTT and the TAS-analyser with ECT-cards gave the most reproducible results compared to other methods. Further studies will evaluate the validity for patient samples. INTRODUCTION Direct and indirect serine protease inhibitors are used clinically for inhibition of blood coagulation (1). Direct inhibitors such as hirudin and fondaparinux inhibit individual serine proteases such as thrombin and factor Xa. Indirect coagulation inhibitors exemplified by the oral anticoagulant, warfarin, act through inhibition of the synthesis of vitamin K-dependent clotting factors. Unfractionated heparin, low molecular weight heparin and dermatan sulphate are examples of indirect inhibitors that act via potentiation of the endogenous plasma coagulation inhibitors, antithrombin and heparin co-factor II (2). Monitoring of the anticoagulant effect of most of these inhibitors is mandatory due to the necessity to obtain information on under- and overdosing. Underdosing of direct and indirect inhibitors are connected with an increased risk of thromboembolism or recurrent thromboembolism and overdose is associated with an enhanced risk of major haemorrhage (3). Recently, two direct thrombin inhibitors, recombinant hirudins and argatroban were introduced for clinical treatment of patients with unstable angina, during percutaneous coronary interventions, and heparin-induced thrombocytopenia (4). A variety of monitoring tests is available to monitor these direct thrombin inhibitors (DTI). The activated partial thromboplastin time (aPTT) is used most frequently to estimate the activity of these inhibitors (1). The ecarin clotting time (ECT) has been used to monitor DTI (5). Thrombin sensitive chromogenic assays are also able to detect the effect of the DTI (6). Bedside monitoring instruments, such as TAS analyser, use ECT reagent incorporated into cards for determination of a DTI effect (7). Lepirudin can be quantified also by an enzyme immunoassay (ELISA) technique (8, 9). However, no studies have been _____________________ *correspondence to: Prof. Dr. med. Job Harenberg, IV. Dept. of Medicine, University Hospital ,Theodor-Kutzer-Ufer 1-3, D-68167 Mannheim, Germany, Tel: +49/621-383 3378, Fax: +49/621-383 3808, E-mail: [email protected] 1Division of Haematology, National Institute for Biological Standards and Control, Hertfordshire, UK To cite: J Thromb Haemost. doi 10.1111/j.1538-7836.2005.01577.x 2 carried out to assess the suitability of these tests. The aim of this study is to utilise a panel of plasmas, spiked with lepirudin (irreversible inhibitor) or argatroban (reversible inhibitor), to evaluate the robustness and the sensitivity of different monitoring methods. MATERIALS AND METHODS Plasma samples and participants Pooled platelet poor plasma (PPP) was prepared from 40 donations of blood collected in CPD-adenine, at the North London Blood Transfusion Centre, Colindale, London, UK. It was buffered with HEPES to a final concentration of 0.04 mol/L. All units of plasma used were tested and found negative for Hepatitis B surface antigen, antibodies to Hepatitis C and antibodies to HIV-1 and HIV-2. Nine coded plasma samples (A – I) were prepared by spiking lepirudin (Refludan, Hoechst Marion Roussel, Germany) and argatroban (Mitsubishi-Tokyo Pharmaceut. Inc., Tokyo, Japan) at a final concentration of 0, 0.31, 0.63, 1.25, and 2.5 µg/ml into the PPP. All plasmas were filled into glass sealed ampoules and then freeze-dried according to the conditions used for International Biological Standards (10). Four sets of plasma samples for each method were sent to the investigators. They were requested to assay one set of samples for four days in quadruplicate by the various assay methods in their laboratories. Fourteen laboratories agreed to take part in the study, with 13 laboratories (Appendix I) returning data for analysis. The participants included manufacturers of thrombin inhibitors, diagnostic equipment and reagent manufacturers, clinical laboratories, and academic institutes. Assays APTT – using local reagents and methods (L-APTT): Ten laboratories carried out the APTT, using their own in-house reagents. Seven different reagents were used (Actin FS, Thrombosil I, Pathromtin SL, Synthasil APTT, Automated APTT, STA-PTT, STA – CK Prest 5). APTT – using a common reagent (C-APTT): The common APTT reagent used was Actin FS (Aventis Pharma, Marburg, Germany). Ten laboratories returned results for this test. Anti-IIa Chromogenic assay: 13 laboratories performed an anti-IIa assay with the S2238 chromogenic substrate from Instrumentation Laboratory/Haemochrom Diagnostica (Essen, Germany). Ecarin clotting time–wet chemistry (Wet ECT): The ecarin reagents were donated by Dr. Götz Nowak (University Jena, Germany) (5). Ten laboratories carried out this test. ECT–dry chemistry: ECT reagent was available on test cards, which were analysed with the TAS analyser (Cardiovascular Diagnostics Inc., Raleigh, USA). Two test cards were used; the low ecarin reagent card is referred as the dry ECT, while the higher concentration ecarin card is referred as TIM. Enzyme immunoassay (ELISA): The ELISA kit was only available for the measurement of lepirudin samples (Immuno Bind, Hirudin Elisa kit, American Diagnostica Inc., Greenwich, USA, kindly supplied by its president Richard Hart). Statistical Analysis The data from the lepirudin-spiked plasma were analysed separately from the results from the argatroban- spiked plasma. The ability of each assay to detect the lowest and the highest concentrations of the inhibitors permitted an assessment of the sensitivity of the different methods. 3 Intra-laboratory (day-to-day) and inter-laboratory variations, as represented by the % geometric coefficient of variation (% GCV), were determined to assess the reproducibility and robustness of the different methods. The intra-laboratory variability was calculated by comparison of dose response curves from each day, using results from the first set of samples as the standard and the other 3 sets as test samples. Relative potency was assessed from each laboratory by random parallel line bioassay (11). For the clot based assays, the APTTs and the ECTs, and the ELISA, the log of the observed responses was found to give best the linearity with log dose, while the untransformed responses were found to be more satisfactory for the anti- IIa chromogenic assays. The statistical validity of linearity and parallelism of the assays was assessed by analysis of variance. The geometric means of the relative potencies were then used to calculate the % GCV (12). With the exception of the ELISA method, the inter-laboratory variations were calculated by taking the ratios of the inhibitor plasma response relative to the negative plasma responses for each day (this took into account each laboratory’s own day-to-day variation). The geometric means of the ratios from each lab for each concentration of inhibitors were then used for calculation of inter-laboratory % GCV. For the ELISA method, the negative plasma gave essentially no readouts, therefore ratios were calculated from the responses for the lowest concentration (0.31 µg/ml) relative to the responses of the other concentrations of lepirudin-spiked plasma. RESULTS Sensitivity of the methods The sensitivity of the methods was assessed by their detection limits. Ratios of the responses from the spiked plasmas to the negative plasma responses were used as indicators. A greater than 1 ratio indicate the inhibitor-spiked sample was giving longer clotting times than the negative plasma. Individual ratios of the inhibitors to negative plasma were calculated for all methods. The APTT using local reagents (L-APTT) showed normal ranges between 32 and 44 sec in the participating laboratories. The geometric mean ratios and their respective ranges for L-APTT are shown in Tables 1a and 1b. Except for one laboratory, all concentrations of lepirudin and argatroban were measured with coagulation times ranging from 88 to 177 sec for lepirudin and 98 to 179 sec with argatroban. Of the 7 local reagents used in the study, one reagent, which may be related