Measuring factor IX activity of nonacog beta pegol with commercially available one-stage clotting and chromogenic assay kits: a two-center study

Transcription

1 Journal of Thrombosis and Haemostasis, 14: DOI: /jth ORIGINAL ARTICLE Measuring factor IX activity of nonacog beta pegol with commercially available one-stage clotting and chromogenic assay kits: a two-center study A. E. BOWYER,* A. HILLARP, M. EZBAN, P. PERSSON and S. KITCHEN* *Coagulation Department, Sheffield Teaching Hospitals, Sheffield, UK; Department of Clinical Chemistry, Skane University Hospital, Malm o, Sweden; Novo Nordisk A/S, Novo Nordisk Park, Maløv, Denmark; and Novo Nordisk A/S, Søborg, Denmark To cite this article: Bowyer AE, Hillarp A, Ezban M, Persson P, Kitchen S. Measuring factor IX activity of nonacog beta pegol with commercially available one-stage clotting and chromogenic assay kits: a two-center study. J Thromb Haemost 216; 14: Essentials Validated assays are required to precisely measure factor IX (FIX) activity in FIX products. N9-GP and two other FIX products were assessed in various coagulation assay systems at two sites. Large variations in FIX activity measurements were observed for N9-GP using some assays. One-stage and chromogenic assays accurately measuring FIX activity for N9-GP were identified. Summary. Background: Measurement of factor IX activity (FIX:C) with activated partial thromboplastin time-based one-stage clotting assays is associated with a large degree of interlaboratory variation in samples containing glyco- PEGylated recombinant FIX (rfix), i.e. nonacog beta pegol (N9-GP). Validation and qualification of specific assays and conditions are necessary for the accurate assessment of FIX:C in samples containing N9- GP. Objectives: To assess the accuracy of various onestage clotting and chromogenic assays for measuring FIX:C in samples containing N9-GP as compared with samples containing rfix or plasma-derived FIX (pdfix) across two laboratory sites. Methods: FIX:C, in severe hemophilia B plasma spiked with a range of concentrations (from very low, i.e..3 IU ml 1, to high, i.e..9 IU ml 1 ) of N9-GP, rfix (BeneFIX), and pdfix (Mononine), was determined at two laboratory sites with 1 commercially available one-stage clotting assays and two chromogenic FIX:C assays. Assays were performed Correspondence: Steve Kitchen, Department of Coagulation, Royal Hallamshire Hospital, Glossop Road, Sheffield S1 2JF, UK Tel.: ; fax: steve.kitchen@sth.nhs.uk Received 9 November 215 Manuscript handled by: D. DiMichele Final decision: F. R. Rosendaal, 2 April 216 with a plasma calibrator and different analyzers. Results: A high degree of variation in FIX:C measurement was observed for one-stage clotting assays for N9-GP as compared with rfix or pdfix. Acceptable N9-GP recovery was observed in the low-concentration to high-concentration samples tested with one-stage clotting assays using SynthAFax or DG Synth, or with chromogenic FIX:C assays. Similar patterns of FIX:C measurement were observed at both laboratory sites, with minor differences probably being attributable to the use of different analyzers. Conclusions: These results suggest that, of the reagents tested, FIX:C in N9-GP-containing plasma samples can be most accurately measured with one-stage clotting assays using SynthAFax or DG Synth, or with chromogenic FIX:C assays. Keywords: blood coagulation test; factor IX; hemophilia B; nonacog beta pegol; poly(ethylene glycol). Introduction Hemophilia B (HB) is a congenital X-linked bleeding disorder characterized by deficiency of functional coagulation factor IX. The current treatment of HB involves replacement therapy with either plasma-derived FIX (pdfix) or recombinant FIX (rfix) concentrate. FIX is administered either as required to treat bleeding episodes (on demand) or as scheduled regular injections (prophylaxis). The half-life of native FIX is h, and the most common protocol for prophylaxis comprises twiceweekly injections of FIX concentrate to maintain FIX levels at >.1 IU ml 1 [1]. Modification of the FIX molecule has been achieved by fusion to immunoglobulin Fc fragments or albumin, or by the covalent attachment of a poly(ethylene glycol) (PEG) residue (PEGylation) [2 4]. These modified molecules show prolonged half-lives as compared with nonmodified concentrates. Furthermore, the addition of a 216 International Society on Thrombosis and Haemostasis

2 N9-GP activity measurement with commercial assay kits 1429 PEG moiety to the FIX molecule can reduce glomerular filtration and clearance by protein-specific receptors [5]. Longer-acting molecules have the potential to decrease the frequency of intravenous injections, improve compliance and contribute to an enhanced quality of life for individuals with HB. The development of a glycopegylated rfix, nonacog beta pegol (N9-GP) (Novo Nordisk A/S, Bagsværd, Denmark), has resulted in an extended half-life, increased FIX activity (FIX:C) levels and decreased FIX consumption as compared with pdfix and rfix [5,6]. N9-GP is glycopegylated via the covalent attachment of a 4-kDa PEG moiety to one of the N-glycans within the FIX activation peptide [2]. Following activation of N9-GP, the activation peptide, along with the PEG moiety, is cleaved off, and the molecule is converted to wild-type activated FIX (FIXa). This molecule retains the same level of activity as the non-pegylated form of rfix, and is subject to the same physiologic regulation [2]. The measurement of FIX:C is necessary for diagnosis, in some cases for clinical monitoring of FIX:C levels in HB patients, and for FIX concentrate potency assignments. The one-stage activated partial thromboplastin time (APTT)-based clotting assay is currently the most commonly used test worldwide for measuring FIX:C in the clinical setting [7]; chromogenic assays constitute an alternative. Chromogenic FIX:C assays have recently become commercially available, and are Conformite Europeenne-marked in the European Union, whereas they are currently only available for research use in the USA [7,8]. Recent recommendations on potency labeling of FVIII and FIX concentrates from the ISTH Scientific and Standardization Committee (SSC) suggest that it should be established whether valid potency estimates are possible with both one-stage clotting and chromogenic methods [9]. Inconsistencies associated with the FIX:C measurement of N9-GP with some one-stage FIX:C assays have previously been reported [8,1,11]; however, chromogenic FIX assays appear to be less variable [11 13]. Because of a functional interaction with the PEG moiety, the activity measured with a one-stage APTT-based clotting assay is highly dependent on the APTT reagent used [12,13]; therefore, one-stage APTT-based assays may be more sensitive to PEGylation than chromogenic assays [8,11]. The measurement of FIX:C with chromogenic assays has not been shown to be affected by PEGylation or other modifications [11]. Validation and qualification of specific assays and conditions are necessary for the assessment of FIX:C in samples containing N9-GP. The aim of this two-center study was to assess the accuracy of commercially available assays for measuring FIX:C in N9-GP-containing samples. A number of onestage clotting and chromogenic assays were used to determine FIX:C in FIX-deficient plasma spiked with N9-GP, pdfix, or rfix. Materials and methods Sample preparation Study kits were prepared by Quintiles (Edinburgh, UK), and distributed to two laboratory sites on dry ice. The two centers were Sheffield Teaching Hospitals, Sheffield, UK (Site 1), and Skane University Hospital, Malm o, Sweden (Site 2). Study kits contained blinded samples of SSC/ISTH secondary coagulation standard plasma lot no. 4 (SSC lot no. 4 plasma; National Institute for Biological Standards and Control [NIBSC], Potters Bar, UK) with an assigned FIX:C value of 1.5 IU ml 1, or plasma spiked with FIX concentrate products. Spiked plasma samples were prepared by diluting N9- GP, pdfix (Mononine; CSL Behring, Kankakee, IL, USA) or rfix nonacog alfa (BeneFIX; Pfizer, New York, NY, USA) in severe HB plasma (George King Bio-Medical, Overland Park, KS, USA) at nominal FIX:C levels of.3 IU ml 1 (very low),.2 IU ml 1 (low),.6 IU ml 1 (medium), and.9 IU ml 1 (high), based on label potency. The percentage target recovery of each sample was calculated according to the certified potency of each product. The certified potency of N9-GP was assigned by use of a one-stage clotting assay with SynthAFax (Instrumentation Laboratory [IL], A Werfen Company, Bedford, MA, USA) as the APTT reagent, in order to correlate biological activity with factor concentrations, and with an N9-GP reference standard that was calibrated to the World Health Organization 4th International Standard for FIX concentrate (NIBSC). FIX:C assays FIX:C recovery was determined with several one-stage clotting or chromogenic assays. Assays were performed with a CS51i analyzer (Sysmex, Kobe, Japan) or an ACL TOP (IL; used with SynthASil and SynthAFax assay reagents [both IL/Werfen] only) at Site 1, and a BCS XP analyzer (Siemens, Marburg, Germany) at Site 2. Samples were stored as 1-mL aliquots at 7 C until use, thawed in a water bath at 37 C for 5 min, and tested within 3 h. Calibration curves were obtained with SSC lot no. 4 plasma at eight different dilutions. FIX:C assays were performed with three doubling dilutions, and the results were extrapolated by the use of parallel lines. Over a 3-month period, samples were tested on three separate days, and the mean results were subsequently used to calculate the percentage of target recovery. One-stage clotting assays Six APTT reagents, i.e. SynthAFax (IL/Werfen), SynthA- Sil (IL/Werfen), Actin FS (Siemens), DG Synth (Grifols, Barcelona, Spain), Pathromtin Ò SL (Siemens), and APTT SP (IL/Werfen), were used at both sites, and a further 216 International Society on Thrombosis and Haemostasis

3 143 A. E. Bowyer et al Table 1 Summary of the characteristics of activated partial thromboplastin time (APTT) reagents included in the two-center study Reagent two reagents (Actin and Actin FSL; both Siemens) were tested at Site 1 only. A summary of the characteristics of the APTT reagents is provided in Table 1. Frozen FIXdeficient plasma (Precision Biologic, Dartmouth, Nova Scotia, Canada), CA-system buffer (Owren s Veronal buffer; Siemens) and calcium chloride (.25 M) were used in all of the aforementioned one-stage clotting assays, with the exception of SynthASil and SynthAFax assays performed on the ACL TOP (IL/Werfen), for which FIXdeficient plasma (IL/Werfen), calcium chloride (.2 M) and factor diluent (both IL/Werfen) were used. Chromogenic assays Chromogenic FIX:C assays were performed according to the manufacturers instructions at both sites with the Rossix Chromogenic FIX assay (Rossix, M olndal, Sweden), whereas Biophen FIX:C (Hyphen BioMed, Neuville-Sur- Oise, France) was tested at Site 1 only. Data analysis Activator Phospholipid source Manufacturer SynthAFax Ellagic acid Synthetic IL DG Synth Ellagic acid Rabbit brain Grifols cephalin SynthASil Colloidal silica Synthetic IL Actin Ellagic acid Rabbit brain Siemens cephalin Actin FS Ellagic acid Purified soy Siemens phosphatides Actin FSL Ellagic acid Purified soy Siemens phosphatides/ rabbit brain phosphatides Pathromtin Silicon dioxide Vegetable Siemens SL particles APTT SP Silica Synthetic IL IL, Instrumentation Laboratory. At both sites, the calibration curve was determined, and the FIX:C for a given sample was calculated in Microsoft Excel by use of the following algorithm: FIX:C IU ml 1 ¼ð½lnðraw dataþ interceptš=slopeþ dilution factor The slope and intercept were derived from the natural logarithm (ln) of the points of the calibration curve; raw data refer to the end time or final OD for a given sample. The mean of the data from three independent days was subsequently converted into percentage of the target recovery value, on the basis of the known FIX:C of N9- GP in the sample, which was determined from the certified potency of N9-GP and according to the following calculation: %target recovery ¼ðmean of 3 independent experiments= known concentration FIX:C in sampleþ 1% The results were rounded to two decimal places. The same formula was applied to calculate the target recovery values of pdfix (Mononine) and rfix (BeneFIX) samples. For the purposes of this study, a target recovery value range of 3% was considered to be acceptable. Results Measuring FIX:C in N9-GP samples FIX:C was measured with the different one-stage clotting and chromogenic assays to determine recovery as compared with the certified potency in the spiked samples. Test curves were visually determined to be parallel to standard curves. The individual results and calculated percentage of target recovery for one-stage APTT and chromogenic assays are shown in Tables S1 and S2, respectively. Assays measuring acceptable N9-GP recovery Acceptable N9-GP recovery was observed for the medium and high samples tested with a one-stage FIX:C assay with SynthAFax and DG Synth reagents, and for samples at all activity levels with both chromogenic FIX:C assays (Figs 1A,B and 2). The percentage of target recovery for the samples with acceptable N9-GP recovery levels ranged from 83% to 115% for one-stage FIX:C assays with SynthAFax or DG Synth, and from 74% to 126% with Fig. 1. Two-center comparison of FIX products with different one-stage clotting assays. The FIX activity in samples spiked with either nonacog beta pegol (N9-GP), plasma-derived FIX (Mononine) or recombinant FIX (BeneFIX) in comparison with a Scientific and Standardization Committee (SSC) standard (SSC/ISTH) secondary coagulation standard plasma lot no. 4 (SSC lot no. 4 plasma; National Institute for Biological Standards and Control, Potters Bar, UK) was determined with one-stage clotting assays performed with activated partial thromboplastin time (APTT) reagents that included SynthAFax (A), DG Synth (B), SynthAFax (ACL TOP) (C), SynthASil (D), SynthASil (ACL TOP) (E), Actin (F), Actin FS (G), Actin FSL (H), Pathromtin SL (I), and APTT SP (J). Assays were performed with a CS51i analyzer at Site 1, and with a BCS XP analyzer at Site 2. In addition, Site 1 used an ACL TOP with SynthASil and SynthAFax assay reagents only. Samples were spiked with FIX diluted to a range of nominal activity levels, including very low (.3 IU ml 1 ), low (.2 IU ml 1 ), medium (.6 IU ml 1 ), and high (.9 IU ml 1 ). Assays were performed in three independent replicates, and data are presented as the percentage target recovery based on certified potency and range () for each sample, measured at one or both laboratory sites. A target recovery of 1% is indicated by a dashed line. ND, not done. 216 International Society on Thrombosis and Haemostasis

4 N9-GP activity measurement with commercial assay kits 1431 (a) SynthAFax (c) (e) (g) (i) N9-GP Mononine BeneFIX SSC (b) DG Synth Site 1 Site SynthAFax (ACL TOP) (d) SynthASil SynthASil (ACL TOP) (f) Actin 5 Actin FS (h) Actin FSL Pathromtin SL (j) APTT SP International Society on Thrombosis and Haemostasis

5 1432 A. E. Bowyer et al (a) Rossix Chromogenic FIX Site 1 Site 2 (b) Biophen FIX:C Fig. 2. Two-center comparison of FIX products with two chromogenic assays. The FIX activity (FIX:C) in samples spiked with either nonacog beta pegol (N9-GP), plasma-derived FIX (Mononine) or recombinant FIX (BeneFIX) in comparison with a Scientific and Standardization Committee (SSC) standard (SSC/ISTH) secondary coagulation standard plasma lot no. 4 (SSC lot no. 4 plasma; National Institute for Biological Standards and Control, Potters Bar, UK) was determined with chromogenic assays performed with the two commercially available kits Rossix Chromogenic FIX assay (A) (Rossix, M olndal, Sweden) and Biophen FIX:C (B) (Hyphen BioMed, Neuville-Sur-Oise, France). Samples were spiked with FIX diluted to a range of nominal activity levels, including very low (.3 IU ml 1 ), low (.2 IU ml 1 ), medium (.6 IU ml 1 ), and high (.9 IU ml 1 ). Assays were performed in three independent replicates, and data are presented as the percentage target recovery based on certified potency and range () for each sample, measured at one or two laboratory sites. A target recovery of 1% is indicated by a dashed line. ND, not done. the chromogenic FIX:C assays. The percentage of target recovery for the very low-concentration samples ranged from 111% to 185% with the one-stage assay. SynthAFax tested with the ACL TOP instrument at Site 1 produced results that were close to target for the medium and high samples (113% and 18%, respectively; Fig. 1C). Assays that underestimated N9-GP recovery The recovery of N9-GP was considerably underestimated in low, medium and high samples tested with the onestage FIX:C assay with SynthASil, Actin, Actin FS or Actin FSL reagents (Fig. 1D H). The percentage of target recovery for these samples ranged from 31% (high sample with Actin FS at Site 1) to 72% (low samples with SynthASil [using ACL TOP] and Actin, both at Site 1), which was just within the acceptable target range. Only the very low N9-GP sample was within the acceptable target range with Actin FS and Actin FSL at Site 1 and SynthASil at Site 2 (all 74%; Fig. 1D,G,H), but the recovery was considerably underestimated with SynthASil tested on the CS51i at Site 1 (37%; Fig. 1D). Assays that overestimated N9-GP recovery The use of Pathromtin SL and APTT SP resulted in a dramatic overestimation of N9-GP recovery in all samples (Fig. 1I,J). With Pathromtin SL, the recovery ranged from 531% (very low sample at Site 1) to 2811% (high sample at Site 2), and with APTT SP it ranged from 333% (very low sample at Site 1) to 1247% (high sample at Site 2). Measuring FIX:C in pdfix (Mononine) samples The target FIX:C recovery in the low, medium and high samples spiked with pdfix (Mononine) was close to target with SynthAFax, DG Synth or Actin FS reagents and the Rossix Chromogenic FIX assay at both sites (Figs 1A,G and 2A), and with SynthASil, Actin, Actin FSL, Pathromtin SL, APTT SP and Biophen FIX:C reagents at Site 1 (Figs 1B,D,F,H J and 2B). Some overestimation of recovery was observed at Site 2 with SynthASil (139% for the high sample), Pathromtin SL (134% for the low sample and 142% for the high sample) and APTT SP (ranging from 141% to 15% of target) (Fig. 1B,D,I,J). With the exception of Biophen FIX:C, which was only tested at Site 1, the very low samples generally showed an overestimation of recovery, with values ranging from 126% (Pathromtin SL at Site 2) to 33% (SynthASil on ACL TOP at Site 1). Measuring FIX:C in rfix (BeneFIX) samples The one-stage FIX:C measurement of the low, medium and high samples spiked with different concentrations of rfix (BeneFIX) was acceptable with one-stage DG Synth, Actin FS and Pathromtin SL reagents at both sites (Fig. 1B,G,I), and with Actin and Actin FSL, which were tested at Site 1 only (Fig. 1F,H). SynthAFax results were acceptable at Site 1, but underestimated the recovery in the medium and high samples at Site 2 (66% and 64% of target, respectively) (Fig. 1A). The recovery with SynthASil was acceptable at Site 1, but slightly elevated in the high samples at Site 2 (132% of target) (Fig. 1D). The results obtained with the Rossix Chromogenic FIX:C kit were close to target at both sites (Fig. 2A). For reasons unknown, the chromogenic Biophen FIX:C kit underestimated the recovery in the very low and low samples (52% and 62%), but was closer to target for the medium and high samples (73% and 77%) (Fig. 2B). 216 International Society on Thrombosis and Haemostasis

6 N9-GP activity measurement with commercial assay kits 1433 Measuring FIX:C in SSC lot no. 4 samples The measurement of SSC lot no. 4 plasma was within acceptable limits at both sites with all assays tested (Figs 1 and 2). Discussion The clinical laboratory assessment of FIX:C worldwide is currently almost exclusively performed with one-stage APTT-based clotting assays. Clinical laboratories perform one-stage assays with a large variety of commercially available APTT reagents, calibration plasmas, diluents, factordeficient plasmas, and instrumentation [7]. It is therefore not surprising that differences in FIX:C have been reported, e.g. in patients with mild HB [14], for recovery of rfix products [15], and in estimations of FIX:C by the use of modified FIX concentrates [3,11 13]. In addition, differences of ~ 3% between one-stage and chromogenic assays have been reported in the measurement of FIX:C [16]. The chromogenic FIX:C assay is infrequently utilized in routine clinical laboratory practice; however, the US National Hemophilia Foundation Medical and Scientific Advisory Council has recently recommended the addition of chromogenic assays for clinical monitoring of FIX clotting factors once they are approved [17]. New guidance from the ISTH SSC has indicated that, although FIX potency labeling currently relies on one-stage clotting assays, the use of newly available chromogenic FIX:C assays should be evaluated individually for each FIX product [9]. The results of the present study show a high degree of variation with the use of different one-stage clotting assays to measure FIX:C in samples containing N9-GP as compared with non-pegylated FIX molecules. The results demonstrate similar patterns between the two test sites, and any differences may be caused by the use of different analyzers. These findings are consistent with previous observations that the level of clotting factor activity measured with one-stage clotting assays in samples containing PEGylated factor is dependent on the APTT reagent used [12,13,15]. In the present study, two APTT reagents, DG Synth and SynthAFax, measured the expected FIX:C recovery for N9-GP. This would be expected for SynthAFax, as it was also used to determine N9-GP potency. Pathromtin SL and APTT SP grossly overestimated the recovery of N9-GP by several-fold. Actin, Actin FS, Actin FSL and SynthASil underestimated N9-GP recovery by approximately 3 7%. Previous studies have suggested that the phospholipid composition and concentration of APTT reagents used in the one-stage clotting assay may affect the potency measurement for recombinant FVIII proteins [18 2], as can the type of activator present [21]. The APTT reagents used in this study encompassed a variety of activators and phospholipid sources (Table 1). The constituents of each individual APTT reagent may partly explain some of the discrepancies observed between one-stage clotting assays for N9-GP. Recently, Rosen et al. showed that most silica-containing APTT reagents caused overestimation of N9-GP activity, owing to adsorption of N9-GP on the surface of the silica particles, with the exception of SynthASil, which contains only small amounts of silica, and underestimated N9-GP [22]. This colocalization of N9-GP with the contact activators FXIa and plasma kallikrein leads to premature N9-GP activation during the contact phase, thereby resulting in an overestimation of recovery. This result was confirmed in the present study; the two reagents associated with severe overestimation of N9-GP recovery, Pathromtin SL and APTT-SP, are both silicabased. SynthASil, with a low silica content, underestimated N9-GP activity. In contrast, some reagents containing ellagic acid, such as SynthAFax and DG Synth, produced acceptable recoveries of N9-GP; however, Actin FS underestimated N9-GP recovery. A previous study also observed that batch-to-batch variation of up to 2% in the concentration of phospholipid components of APTT reagents may contribute to discrepancies in the outcomes of one-stage clotting assays [23], and this may have contributed to the differences in the results between sites in the present study. The source of phospholipid and the concentrations of all reagent constituents are likely to contribute to assay sensitivity to N9-GP activity. The recovered potencies of SSC plasma lot no. 4, rfix and pdfix were within acceptable limits for all APTT reagents tested, with the exception of a slight, but probably clinically insignificant, underestimation of rfix (BeneFIX) activity with one-stage assays performed with SynthAFax at Site 2. In the majority of assays, the very low samples (.3 IU ml 1 ) showed a high percentage recovery. The two chromogenic FIX:C assays tested in this study recovered the certified potency for N9-GP and pdfix (Mononine); however, some underestimation of recovery was observed with the Hyphen BioMed kit with rfix (BeneFIX). On the basis of our results, chromogenic assays appear to be more accurate for the assessment of samples containing low FIX:C levels. Currently, there is a strong focus on the measurement of trough factor activity levels in hemophilic patients; however, the available assay products and reagents are not optimal for measuring such low factor activity levels, and improvements are required. A plasma standard was used to generate calibration curves for all assays. As it has been well established that the measurement of recombinant products by use of a plasma-derived standard is not ideal or recommended [24], it is possible that the differences for rfix (BeneFIX) observed in our study may be partly attributable to this issue. The observations from this study, together with previous studies, make it clear that the choice of APTT reagent can affect the accurate measurement of FIX:C for 216 International Society on Thrombosis and Haemostasis

7 1434 A. E. Bowyer et al recombinant and modified proteins. As a result, the establishment of a panel of specific assay reagents and conditions required for accurate measurement of factor activity for the new longer-acting clotting factor proteins that are currently emerging onto the market will be necessary [8,9,25]. The results of this study suggest that, of the reagents tested, FIX:C can be most accurately measured in plasma samples containing N9-GP with one-stage clotting assays with SynthAFax or DG Synth and with the Rossix or Hyphen BioMed chromogenic assays. These findings have been recently confirmed with clinical samples from HB patients treated with N9-GP [26]. Furthermore, the type of activator and phospholipid source of APTT reagents cannot be used as a predictor of efficiency for measuring N9-GP. This study could be expanded to investigate additional commercially available APTT reagents. Addendum A. Bowyer, S. Kitchen, and A. Hillarp performed the research and analyzed the data. M. Ezban, S. Kitchen, A. Hillarp, and A. Bowyer designed the research study. P. Persson organized logistics and coordinated the study. All authors contributed to the writing and review of the manuscript, and approved the final version. Acknowledgements The authors would like to thank M. H. Sørensen (Novo Nordisk) for writing the instructions for preparation of spiked samples, Quintiles (Edinburgh, UK) for preparing the samples, and M. Persson (Skane University Hospital, Malm o) for technical assistance. Medical writing support was provided by Physicians World Europe GmbH (Mannheim, Germany), and was financially supported by Novo Nordisk A/S (Bagsværd, Denmark). Disclosure of Conflict of Interests A. Bowyer has received grants and travel support from Novo Nordisk. S. Kitchen has received consultancy fees and travel support from Novo Nordisk. A. Hillarp has received consultancy fees from Novo Nordisk and Octapharma AG, and has received consultancy and speaker fees from Bayer Pharma AG. M. Ezban and P. Persson are employees of Novo Nordisk A/S, Maløv, Denmark. Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1 Individual results of the two-center study of FIX:C of PEGylated and non-pegylated FIX molecules measured with one-stage clotting assays. Table S2 Individual results of the two-center study of FIX:C of PEGylated and non-pegylated FIX molecules measured with chromogenic assays. References 1 Srivastava A, Brewer AK, Mauser-Bunschoten EP, Key NS, Kitchen S, Llinas A, Ludlam CA, Mahlangu JN, Mulder K, Poon MC, Street A; Treatment Guidelines Working Group on Behalf of The World Federation of Hemophilia. Guidelines for the management of hemophilia. Haemophilia 213; 19: e Ostergaard H, Bjelke JR, Hansen L, Petersen LC, Pedersen AA, Elm T, Moller F, Hermit MB, Holm PK, Krogh TN, Petersen JM, Ezban M, Sorensen BB, Andersen MD, Agerso H, Ahmadian H, Balling KW, Christiansen ML, Knobe K, Nichols TC, et al. Prolonged half-life and preserved enzymatic properties of factor IX selectively PEGylated on native N-glycans in the activation peptide. Blood 211; 118: Sommer JM, Buyue Y, Bardan S, Peters RT, Jiang H, Kamphaus GD, Gray E, Pierce GF. Comparative field study: impact of laboratory assay variability on the assessment of recombinant factor IX Fc fusion protein (rfixfc) activity. Thromb Haemost 214; 112: Stennicke HR, Ostergaard H, Bayer RJ, Kalo MS, Kinealy K, Holm PK, Sorensen BB, Zopf D, Bjorn SE. Generation and biochemical characterization of glycopegylated factor VIIa derivatives. Thromb Haemost 28; 1: Collins PW, Young G, Knobe K, Karim FA, Angchaisuksiri P, Banner C, Gursel T, Mahlangu J, Matsushita T, Mauser- Bunschoten EP, Oldenburg J, Walsh CE, Negrier C; Paradigm 2 Investigators. Recombinant long-acting glycopegylated factor IX in hemophilia B: a multinational randomized phase 3 trial. Blood 214; 124: Collins PW, Moss J, Knobe K, Groth A, Colberg T, Watson E. Population pharmacokinetic modeling for dose setting of nonacog beta pegol (N9-GP), a glycopegylated recombinant factor IX. J Thromb Haemost 212; 1: Kitchen S, Signer-Romero K, Key NS. Current laboratory practices in the diagnosis and management of haemophilia: a global assessment. Haemophilia 215; 21: Kitchen S, Gray E, Mertens K. Monitoring of modified factor VIII and IX products. Haemophilia 214; 2(Suppl. 4): Hubbard AR, Dodt J, Lee T, Mertens K, Seitz R, Srivastava A, Weinstein M. Factor VII, Factor IXSoTS, Standardisation Committee of The International Society on Thrombosis and Haemostasis. Recommendations on the potency labelling of factor VIII and factor IX concentrates. J Thromb Haemost 213; 11: Bowyer A, Hillarp A, Persson M, Ezban M, Persson P, Kitchen S. The laboratory measurement of a glycopegylated recombinant FIX, N9-GP: a two centre study. J Thromb Haemost 214; 12 (Suppl. 1): (abstract EHF11). 11 Holm PK, Sørensen MH, Hermit BM, Ezban M. The activity of GlycoPEGylated recombinant FIX (N9-GP) can be measured in two-stage chromogenic and one-stage clotting assays. J Thromb Haemost 213; 11(Suppl. 2): 828 (abstract PB ). 12 Hermit MB, Sorensen MH, Ezban M, Lochu A, Magdelaine A. Pegylation of FIX influences the clotting time in APTT assays. Use of product specific standard eliminates PEG interference. Haemophilia 213; 19: 45 (abstract PO59). 13 Lochu A, Romari N, Beltran J, Magdelaine A, Ezban M, Hermit MB. Influence of FIX and FVIII PEGYLATION on FIX and FVIII activity based on APTT assays. J Thromb Haemost 214; 11(Suppl. 2): 97 (abstract PB ). 216 International Society on Thrombosis and Haemostasis

8 N9-GP activity measurement with commercial assay kits Pouplard C, Trossaert M, Le Querrec A, Delahousse B, Giraudeau B, Gruel Y. Influence of source of phospholipids for APTT-based factor IX assays and potential consequences for the diagnosis of mild haemophilia B. Haemophilia 29; 15: Gritsch H, Romeda-Finger S, Scheiflinger F, Turecek PL. Potency assignment and measurement of recombinant FIX activity in human plasma impact of aptt reagents on the 1-stage clotting assay. Haemophilia 214; 2(Suppl. 3): 37 (abstract). 16 Wilmot HV, Hogwood J, Gray E. Recombinant factor IX: discrepancies between one-stage clotting and chromogenic assays. Haemophilia 214; 2: MASAC statement regarding use of various clotting factor assays to monitor factor replacement therapy. National Hemophilia Foundation MASAC Document # pdf. Accessed 16 March Mikaelsson M, Oswaldsson U, Sandberg H. Influence of phospholipids on the assessment of factor VIII activity. Haemophilia 1998; 4: Caron C, Dautzenberg MD, Delahousse B, Droulle C, Pouzol P, Dubanchet A, Rothschild C. A blinded in vitro study with Refacto mock plasma samples: similar FVIII results between the chromogenic assay and a one-stage assay when using a higher cephalin dilution. Haemophilia 22; 8: Mikaelsson M, Oswaldsson U. Assaying the circulating factor VIII activity in hemophilia A patients treated with recombinant factor VIII products. Semin Thromb Hemost 22; 28: Lusher JM, Hillman-Wiseman C, Hurst D. In vivo recovery with products of very high purity assay discrepancies. Haemophilia 1998; 4: Rosen P, Rosen S, Ezban M, Persson E. N9-GP overestimation in one-stage clot method due to silica-mediated conversion to fixa during contact activation. J Thromb Haemost 215; 13: Kitchen S, Cartwright I, Woods TA, Jennings I, Preston FE. Lipid composition of seven APTT reagents in relation to heparin sensitivity. Br J Haematol 1999; 16: Barrowcliffe TW, Hubbard AR, Kitchen S. Standards and monitoring treatment. Haemophilia 212; 18(Suppl. 4): Dodt J, Hubbard AR, Wicks SJ, Gray E, Neugebauer B, Charton E, Silvester G. Potency determination of factor VIII and factor IX for new product labelling and postinfusion testing: challenges for caregivers and regulators. Haemophilia 215; 21: Sørensen MH, Andersen S, Ezban M. Factor IX-deficient plasma spiked with N9-GP behaves similarly to N9-GP postadministration clinical samples in N9-GP ELISA and FIX activity assays. Haemophilia 215; 21: International Society on Thrombosis and Haemostasis