N8-GP is a recombinant factor VIII (FVIII) with a site-directed glycoPEGylation for the purpose of half-life prolongation.
N8-GP is a recombinant factor VIII (FVIII) with a site-directed glycoPEGylation for the purpose of half-life prolongation.
To evaluate the safety and pharmacokinetic profiles of N8-GP in comparison with those of the patients' previous FVIII products.
This dose-escalation trial included previously treated patients with severe hemophilia A who received one of three dose levels (25, 50 or 75 U kg−1) of N8-GP and FVIII product. Each dose escalation was preceded by safety and pharmacokinetic assessment. The trial was registered at www.clinicaltrials.gov (NCT01205724).
Twenty-six patients each received one dose of their previous FVIII product followed by the same, single dose of N8-GP. N8-GP, at any tested dose, was well tolerated, with a low frequency of adverse events. No new inhibitors against FVIII or N8-GP and no binding antibodies against N8-GP developed during the trial. The pharmacokinetics of N8-GP were dose-linear. The incremental recovery of N8-GP was 0.025 [(U mL−1)/(U kg−1)]. The clearance was 1.79 mL−1 h−1 kg−1. The estimated time from dosing of 50 U kg−1 N8-GP to a plasma activity of 1% was 6.5 days (range: 3.6–7.9 days). The mean terminal half-life of N8-GP was 19.0 h (range: 11.6–27.3 h), 1.6-fold longer than that of the patients' previous products.
A single dose of up to 75 U kg−1 N8-GP was well tolerated in patients with hemophilia A, with no safety concerns. N8-GP had a prolonged half-life, and FVIII:C activity remained at > 1% for longer than the patient's previous product. These results indicate that N8–GP has the potential to reduce dosing frequency during prophylaxis.
Replacement therapy with factor VIII (FVIII) concentrates is the standard of care for patients with hemophilia A. The therapeutic aim is to maintain FVIII activity at a level high enough to prevent bleeds, and to provide rapid and definitive treatment of bleeds when they occur.
In children and many adults with severe hemophilia A, prophylactic replacement is the preferred therapy, as it reduces the frequency of joint and other bleeds, and is superior at preventing chronic arthropathy than is on-demand replacement therapy . However, the relatively short plasma half-life of currently available FVIII products limits the convenience of prophylaxis. In order to maintain an FVIII activity level above a threshold level (minimum of 1%), intravenous administration of currently available FVIII products is required every other day or three times weekly [2-4]. This is often challenging, particularly in small children, who are at the highest risk of inhibitor formation and joint damage. Furthermore, even recommended dosing regimens cannot prevent trough levels of < 1% for some patients, and the time spent below this cut-off level has been demonstrated to be a major determinant for acute bleeds during prophylaxis therapy .
On-demand replacement therapy is the standard of care in patients who bleed less frequently, such as those with mild or moderate hemophilia A, and is also used for the treatment of acute bleeds in patients receiving prophylaxis. The aim of on-demand treatment is to reduce the extent of bleeding, thereby decreasing inflammation and joint damage, and to prevent rebleeds into injured tissue. Often, multiple injections are required for symptoms to resolve completely . Also, many patients with mild or moderate hemophilia are not familiar with injecting themselves at home, and may need to visit a hospital repeatedly to receive treatment, which increases the inconvenience.
N8-GP is a new recombinant FVIII (rFVIII) in which site-directed glycoPEGylation has been utilized to prolong the half-life of the molecule while leaving the hemostatic activity unchanged . The rFVIII component of N8-GP (turoctocog alfa; Novo Nordisk, Bagsværd, Denmark) is synthesized in Chinese hamster ovary cells in a serum-free production environment, with no human or animal proteins added during the purification process . Turoctocog alfa has a truncated B-domain of 21 amino acids , and has been demonstrated to have similar pharmacokinetic (PK) properties to another marketed rFVIII product (Advate) . N8-GP is produced through glycoPEGylation of the turoctocog alfa molecule, during which the terminal sialic acid on an O-glycan structure in the truncated B-domain is replaced by a conjugated sialic acid containing a branched 40-kDa PEG. This process results in a highly defined product with a single PEG attached to the B-domain of each rFVIII. Upon activation by thrombin, the truncated B-domain – with the attached PEG – is cleaved off, leaving the activated FVIII .
The aim of this phase 1, first-in-human clinical trial was to evaluate the safety and PK profiles of single, escalating doses of N8-GP in patients with severe hemophilia A, and to compare the FVIII activity of N8-GP with that of the patients' previous FVIII products.
The primary objective of this first-in-human trial was to determine the safety profile of N8-GP by evaluating adverse events and antibody formation (including inhibitors against FVIII and N8-GP, and binding antibodies against N8-GP). Safety was assessed by performing a physical examination and electrocardiography, recording vital signs, and undertaking standard laboratory safety assessments, including hematology, biochemistry, urinalysis, and troponin T. In addition, changes in the levels of coagulation-related parameters (fibrinogen, antithrombin, prothrombin fragment 1 + 2, D-dimers, activated partial thromboplastin time, and prothrombin time) were monitored. The trial was also designed to evaluate the PK profile of N8-GP, and to compare N8-GP FVIII activity with that of the patients' previous FVIII products. The time points for the assessment of the PK properties of the patients' previous FVIIII product were predose and at 30 min and 1, 4, 8, 12, 24, 30 and 48 h postdose. The PK samples for N8-GP were drawn at the same time points, and also at 72, 96, 120, 144 and 168 h postdose. The PK parameters assessed are listed in Table 1.
|N8-GP 25 U kg−1||N8-GP 50 U kg−1||N8-GP 75 U kg−1||Totala|
|Number of patients||7||9||10||26|
|AUC (U h mL−1)|
|Mean (SD)||14.74 (5.35)||38.85 (11.41)||46.76 (20.56)||NA|
|Minimum; maximum||8.63; 23.81||25.11; 52.61||15.32; 75.09||–|
|IR30 min ([U mL−1]/[U kg−1])|
|Mean (SD)||0.026 (0.005)||0.025 (0.006)||0.026 (0.008)||0.025 (0.006)|
|Minimum; maximum||0.020; 0.032||0.015; 0.033||0.016; 0.042||0.015; 0.042|
|Mean (SD)||15.81 (4.27)||23.08 (5.24)||17.96 (4.84)||19.04 (5.53)|
|Minimum; maximum||11.79; 21.35||11.57; 27.35||12.09; 26.58||11.57; 27.35|
|CL (mL−1 h−1 kg−1)|
|Mean (SD)||1.89 (0.64)||1.39 (0.42)||2.07 (1.31)||1.79 (0.92)|
|Minimum; maximum||1.05; 2.90||0.95; 1.99||1.00; 4.90||0.95; 4.90|
|C30 min (U mL−1)|
|Mean (SD)||0.65 (0.12)||1.24 (0.28)||1.93 (0.58)||NA|
|Minimum; maximum||0.49; 0.80||0.76; 1.64||1.23; 3.16||–|
|Vz (mL kg−1)|
|Mean (SD)||40.14 (7.02)||45.02 (14.19)||49.48 (25.55)||45.27 (17.78)|
|Minimum; maximum||29.56; 52.21||29.92; 72.98||25.69; 109.03||25.69; 109.03|
|Time to 1% (days)|
|Mean (SD)||3.91 (1.09)||6.53 (1.41)||5.49 (1.57)||5.73 (1.6)|
|Minimum; maximum||2.90; 5.20||3.56; 7.87||3.33; 7.71||3.42; 8.35|
|Time to 3% (days)|
|Mean (SD)||2.86 (0.82)||5.01 (1.08)||4.30 (1.27)||4.47 (1.29)|
|Minimum; maximum||2.07; 3.87||2.80; 6.08||2.54; 5.95||2.60; 6.60|
The trial was conducted at 13 sites in seven countries (Germany, Italy, Japan, Switzerland, Turkey, the UK, and the USA). Patients were selected according to the Committee for Medicinal Products of Human Use (CHMP) guidelines on the clinical investigation of plasma-derived FVIII and rFVIII products . Patients were previously treated males (aged ≥ 18 years) with severe hemophilia A, a residual FVIII activity of < 1%, and at least 150 days of exposure to an FVIII product. Patients with any history of FVIII inhibitors or with an increased risk of thromboembolic events were excluded, as were immunocompromised patients with a CD4+ lymphocyte count of < 200 μL−1. The trial was approved by independent ethics committees and institutional review boards in all participating countries, and was conducted in accordance with the Declaration of Helsinki  and Good Clinical Practice . All patients provided written informed consent prior to being included in the trial. The trial was registered at www.clinicaltrials.gov as NCT01205724.
This was an open-label, first-in-human, dose escalation trial of three ascending doses of N8-GP in patients with hemophilia A in a non-bleeding state. N8-GP was supplied as a freeze-dried powder in single-use vials, and was reconstituted in 4.3 mL of sterile 0.9% NaCl for intravenous injection. N8-GP potency was assigned by the chromogenic assay against the 7th World Health Organization (WHO) standard, and expressed in units (U). The doses evaluated in this study were 25 U kg−1, 50 U kg−1, and 75 U kg−1. At least six patients completed each of the three dose cohorts. The trial consisted of three visits in total. At the first visit (visit 1), the patient received a single dose of their previous FVIII product (plasma-derived FVIII or rFVIII) at the same dose level as the N8-GP dose to be administered at the following visit (visit 2). A wash-out period of 4 days was required before dosing at visits 1 and 2. If patients received FVIII treatment within these 4 days, the visit was postponed accordingly. There was also a 3-day wash-out period prior to the antibody assessment at visit 3 in order to avoid the interference of FVIII activity with the antibody assays.
Dosing of patients with N8-GP was staggered by at least 24 h in the first cohort to allow for reporting of any serious or severe adverse events before another patient was dosed with N8-GP. Dosing with N8-GP in the subsequent cohorts was not performed before at least six patients in the previous cohort had been treated, and data were reviewed by a safety group in accordance with the protocol.
The PK assessment was based on activity assays, as reported previously for coagulation factors [13, 14]. The PK data reported here were assessed with a two-stage chromogenic assay (Coamatic FVIII; Chromogenix, Milan, Italy) performed on a Coasys Plus C coagulation analyzer (Roche, Basel, Switzerland). In this assay, FVIII supports the activation of factor X (FX) to activated FX (FXa) by activated FIX (FIXa), in the presence of calcium ions and phospholipids. The rate of activation of FX is positively correlated with the amount of FVIII in the sample. FXa hydrolyzes the chromogenic substrate S-2765, whereby a chromophoric group (pNA) is released. The change in the absorbance per minute (ΔA min−1) is then measured at 405 nm, and is proportional to the FVIII activity in the sample. N8-GP FVIII activity was assessed against an N8-GP product-specific standard, in which the biological activity was determined relative to the 7th WHO International Standard FVIII concentrate with the chromogenic assay. For all other products (Advate, Kogenate, Refacto, and Haemate), a commercial plasma standard (Siemens Healthcare Diagnostics, Eschborn, Germany), calibrated against the 5th WHO International Standard for FVIII and von Willebrand factor (vWF) in plasma, was used as the reference standard.
A Nijmegen-modified Bethesda assay was used for the determination of inhibitors against FVIII [15, 16]. Inhibitors were defined as positive at ≥ 0.6 Bethesda units (BU), according to the CHMP guidelines . Patients were classified as having a confirmed inhibitor if they had tested positive in two consecutive assays.
A radioimmunoassay, in which study samples were incubated with radioactively labeled N8-GP, was used to test for the potential presence of N8-GP-binding antibodies.
Analyses were descriptive rather than being based on a statistical approach. Thus, no formal power calculation was performed. To ensure adequate evaluation of the safety and PK properties of N8-GP, a total of 18 patients (a minimum of six in each of the three cohorts) were needed to complete the trial.
The safety evaluation was based on summary and individual data, and included all of the patients exposed to N8-GP.
In the PK assessment, data were adjusted for the actual dose administered. This was performed by multiplying the measured individual plasma activities by (planned dose)/(actual dose), and then calculating the PK parameters with these concentrations. No subtraction of baseline values was performed, as all patients had severe hemophilia A with an FVIII activity level of ≤ 1% and had a 96-h wash-out period since their previous FVIII administration. The actual time at which the blood samples were taken was used in the calculation of PK parameters. If consecutive samples were measured below the lower limit of quantification (LLoQ), without any following value above the LLoQ, the first value was set to LLoQ/2 and the following values were set to 0.
The PK parameters were calculated with standard non-compartmental methods. All PK endpoints for N8-GP were modeled and analyzed by analysis of variance (anova) on the log-transformed parameter values, with dose as an independent variable. Mean residence time was analyzed in the same way, but not log-transformed. Estimates of 95% confidence intervals (CIs) were provided, back-transformed to the original scale.
Dose linearity for N8-GP was evaluated on the basis of the area under the plasma activity curve from administration to infinity (AUC) and FVIII plasma activity 30 min after administration (C30 min). This was performed by analysis of covariance (ancova) on log-transformed parameter values, with the log of the dose as a covariate. The slope estimate was provided with a 95% CI.
On the assumption of adequate dose linearity, the incremental recovery (determined as the peak level recorded 30 min after administration [IR30 min]), the volume of distribution (Vz), plasma clearance (CL), AUC and t½ were also compared between the patient's previous FVIII treatment and N8-GP, normalized to a dose of 50 U kg−1. The comparison of N8-GP and the previous FVIII treatment was performed with a mixed model, with patient as a random factor, and treatment (N8-GP, previous FVIII) as an independent variable.
A total of 27 male patients with hemophilia A were screened, one of whom did not satisfy the screening criteria, owing to a history of inhibitors; therefore, 26 patients were exposed to N8-GP, and all patients then completed the trial. All 26 patients exposed to N8-GP were included in the safety analysis set. The median age was 36.5 years, ranging from 20 to 60 years. Of the 26 patients who were exposed to N8-GP, 22 were Caucasian, one was African-American, and three were Asian/Japanese. The median body mass index was 24.7 kg m−2 (range: 19.4–34.7 kg m−2). Five patients were HIV-positive and 18 patients were hepatitis C-positive. There was one protocol deviation, consisting of a patient at dose tier 2 (50 U kg−1) with a previously unrecognized inhibitor at visit 1, screening (0.7 BU), visit 2, baseline (0.9 BU), and visit 3, final (1.7 BU). The patient completed the trial as per protocol. As the inhibitor was expected to influence the FVIII activity assay and the PK evaluation, the data obtained in this patient were excluded from the primary PK analysis.
N8-GP was well tolerated in the dose range 25–75 U kg−1. No adverse events leading to withdrawal or deaths were reported. Seventeen adverse events were reported in 11 (42%) patients following a dose of N8-GP; all were rated as mild or moderate.
The number of adverse events reported after exposure to N8-GP did not appear to be dose-related. Furthermore, no treatment-dependent patterns in the type or frequency of adverse events reported following dosing with N8-GP, either within or among system organ classes, were apparent. No type of event was reported more than once, except for headache, which was reported in one patient following a 50 U kg−1 dose of N8-GP, and in one patient following a dose of 75 U kg−1 N8-GP. The headache event reported in dose tier 2 (50 U kg−1) was regarded as possibly being related to N8-GP administration by the investigator.
One serious adverse event (tuberculosis) was reported 5 days following dosing with 50 U kg−1 N8-GP in one patient, which was evaluated as unlikely to be related to N8-GP as judged by the investigator and sponsor.
The results of physical examinations were as expected, reflecting the underlying disease. Other safety-related parameters and blood samples analyzed prior to, and after, N8-GP exposure did not reveal any unexpected findings.
Blood samples for antibody assessment were taken from all patients before N8-GP exposure and 4–5 weeks after exposure. None of the patients developed inhibitors or binding antibodies after exposure to any single dose of N8-GP.
Plasma concentration data (FVIII activity) obtained after dosing with N8-GP and the patients' previous FVIII products showed prolonged plasma activity after administration of N8-GP as compared with the previous product (Fig. 1). The PK results for the three dose levels of N8-GP are outlined in Table 1.
The estimated means for IR30 min, Vz, CL, t½ and AUC (adjusted to a dose of 50 U kg−1) were calculated for N8-GP and the previous FVIII product, and compared under the assumption of dose linearity. The results are presented in Table 2.
|N8-GP||Previous FVIII||Treatment difference, N8-GP/previous FVIII ratio (95% CI), P-value|
|IR30 min ([U mL−1]/[U kg−1])||25||0.025||25||0.032||0.79 (0.70–0.89), < 0.001|
|t½ (h)||24||18.35||25||11.73||1.56 (1.42–1.72), < 0.001|
|CL (mL−1 h−1 kg−1)||24||1.61||25||2.41||0.67 (0.55–0.81), < 0.001|
|Vz (mL kg−1)||24||42.61||25||40.73||1.05 (0.91–1.21), 0.522|
|AUC (U* h mL−1)a||24||30.04||25||20.12||1.49 (1.23–1.81), < 0.001|
The mean terminal half-life of N8-GP was 19.0 h (range: 11.6–27.3 h), representing an approximately 1.6-fold prolongation as compared with patients' previous FVIII products (mean half-life of 11.7 h). Clearance of N8-GP was reduced by ~ 30% as compared with the previous product. Volumes of distribution were comparable.
The terminal half-life of N8-GP was higher than that of the patients' previous products in all patients except one (Fig. 2), and the difference between treatments was statistically significant (P < 0.001; Table 2). The half-life obtained with N8-GP correlated with that of the patient's previous products; that is, patients with a long half-life for their previous FVIII product had a long half-life for N8-GP as well (Fig. 2). The mean time from dosing of N8-GP to a plasma activity of 1% was also substantially longer after treatment with N8-GP than after treatment with the patients' previous products: 3.9 days (25 U kg−1), 6.5 days (50 U kg−1) and 5.5 days (75 U kg−1; Fig. 3), as compared with values for patients' previous products of 2.9 days (25 IU kg−1), 3.7 days (50 IU kg−1), and 3.7 days (75 IU kg−1).
The single-dose PK profile of N8-GP appeared to be dose-linear in the dose range 25–75 U kg−1. Dose linearity for N8-GP was evaluated on the basis of AUC and C30 min. The slope estimate for AUC was 1.024 U h mL−1, and the 95% CI was 0.612–1.436 U h mL−1. The slope estimate for C30 min was 0.975 U mL−1, and the 95% CI was 0.754–1.196 U mL−1. Both slopes were close to 1 and, on the basis of these results, no deviation from dose linearity was detected; thus, dose linearity was assumed within the dose range investigated (25–75 U kg−1).
The results of this first-in-human clinical trial demonstrate that a single dose of up to 75 U kg−1 N8-GP was well tolerated in patients with hemophilia A, with no safety concerns. N8-GP had a prolonged half-life, and FVIII:C activity remained at > 1% for a longer period of time than with the patients' previous product.
PEG conjugation is widely used as a method for enhancing the PK properties of therapeutic proteins [17, 18], and reportedly reduces the clearance of drugs by interfering with glomerular filtration, enzymatic digestion, and interaction with clearance receptors . The 40-kDa PEG used for N8-GP is the same as that used for N9-GP, a recombinant FIX product with enhanced PK properties that is currently under clinical investigation . It is also used for several marketed drugs, including peginterferon alfa 2a (Pegasys), peginterferon alfa 2b (Pegintron), Macugen, Somavert, and certolizumab pegol (Cimzia). Clinical trials and postmarketing surveillance for these drugs have not revealed PEG-related safety issues .
No clinically relevant PEG-related or other safety findings have been observed in a toxicology program of N8-GP (Novo Nordisk, data on file). However, as there are no specific biomarkers related to the potential retention of PEG, comprehensive general safety surveillance was applied in the present trial in order to capture potential PEG-related safety issues (and will also be applied in any future trials with N8-GP).
In the present trial, a single intravenous dose of N8-GP was well tolerated. There were no reports of allergic reactions or intolerance, no clinically relevant changes in clinical chemistry or hematology parameters, and no formation of inhibitors against FVIII or N8-GP. Three of the patients in the trial were Asian/Japanese, and there were no apparent differences regarding safety or PK data between these and the patients of other, primarily Caucasian, ethnicities.
Our PK analyses demonstrate the dose linearity of N8-GP over a range of 25–75 U kg−1. As compared with the patients' previous products, a significant reduction in clearance and an associated increase in terminal half-life (11.7 vs. 18.4 h) were observed. The time for which residual FVIII activity was maintained at > 1% was prolonged as compared with the patients' previous products.
The estimated mean IR30 min was 0.025 (U mL−1)/(U kg−1) for N8-GP. This compares favorably with the recovery of currently marketed FVIII products, and is also similar to the IR30 min of turoctocog alfa (0.019 [IU mL−1]/[IU kg−1]) and Advate (0.019 [IU mL−1]/[IU kg−1]) determined in a recent study . Unexpectedly, the IR30 min for the patients' previous products in this trial was higher than for N8-GP (0.032 vs. 0.025 [U mL−1]/[U kg−1], respectively). For some of the patients' previous products, the strength stated on the label was based on potency assigned according to the one-stage clot assay; as the PK reported here was measured with the chromogenic assay, this may have contributed to this finding. Moreover, dosing was based on labeled activity for patients' previous products, and overfilling of vials cannot be excluded. The volume of distribution of N8-GP was the same as for the other marketed FVIII products used by these patients, indicating that N8-GP is distributed in a similar way as currently available FVIII products.
A close correlation was seen between the half-life of N8-GP and those of the patients' previous FVIII products. This observation suggests that N8-GP is cleared by the same pathways as currently marketed FVIII products, and that major determinants of half-life, such as the concentration of VWF and its half-life, are also relevant for N8-GP.
PK data for new FVIII products are generally accepted as surrogate endpoints for efficacy [10, 13]. Therefore, the prolonged half-life of N8-GP is of great clinical interest. A regimen involving frequent injections is known to be one of the main reasons for the failure of compliance with or adherence to prophylaxis . The prolonged half-life of N8-GP may permit prophylaxis with fewer injections, thereby improving adherence to treatment and reducing the discomfort of the treatment. A less frequent dosing regimen could also enable effective prophylaxis in small children with poor venous access. Such children may not be eligible for standard prophylaxis regimens without the placement of indwelling central venous catheters, which are associated with frequent complications, such as thrombosis and infection [23, 24].
Furthermore, the enhanced PK properties of N8-GP may also help to improve the efficacy of treatment. With current prophylactic dosing regimens, major determinants of acute bleeds in individuals aged 10–65 years are dosing frequency, adherence to dosing frequency, and time spent below an FVIII trough level of 1% . The annual bleeding rate increases, in general, by 2.2% per hour per week spent with an FVIII level of < 1%. Therefore, extending the half-life of an FVIII product, and thereby, the time to a plasma activity of 1%, could potentially reduce the frequency of bleeds during prophylaxis. With current treatment options, patients spend a median of 18 h (10th and 90th percentiles, 2.1 h and 46 h) per week with an FVIII level of < 1%, based on a mean half-life of approximately 11 h  (as confirmed for patients' previous products [11.7 h] in the current trial). Because the time to 1% and the half-life of N8-GP observed in the current trial were significantly prolonged as compared with patients' previous FVIII products, N8-GP may have the potential to further reduce the annual bleeding rate during prophylaxis, by significantly reducing the time spent with a level of < 1% with the same dosing schedule. It could also be speculated that a prophylactic regimen with a product with a prolonged half-life may facilitate trough levels of FVIII activity of > 3%, which theoretically may further reduce the risk of bleeds during prophylaxis. Therefore, the time to an FVIII activity of 3% was also estimated post hoc to be 4.5 days (based on dose normalization to 50 U kg−1).
A longer half-life may also translate into clinical benefit for on-demand regimens, or for the treatment of bleeds during prophylaxis. Acute bleeds occurring in patients receiving prophylaxis are treated with more than one infusion in ~ 20% of cases . The WFH guideline for treatment of hemophilia recommends prolonged treatment of joint bleeds for 2–3 days . A similar approach has been used in a randomized clinical trial of prophylactic vs. prolonged episodic treatment . A longer-lasting FVIII product could also advance management in this scenario.
In conclusion, N8-GP is a novel rFVIII product with enhanced PK properties that has the potential to improve both prophylactic and on-demand treatment of hemophilia A. The first-in-human safety and PK trial reported here did not identify any safety issues after a single dose of N8-GP (25–75 U kg−1), and demonstrated a 1.6-fold prolonged half-life as compared with the patients' previous FVIII products. The prolonged half-life of N8-GP results in a longer period with FVIII activity of > 1% (~ 6 days after a dose of 50 U kg−1). The anticipated benefits of N8-GP remain to be verified in clinical trials, but N8-GP has the potential to improve current treatment regimens for hemophilia A.
D. Viuff, K. Knobe, and A. Tiede: designed and performed the research, collected and interpreted data, and wrote and critically revised the manuscript; B. Brand, R. Fischer, K. Kavakli, S. R. Lentz, T. Matsushita, and C. Rea: collected and interpreted data, and critically revised the manuscript.
The other principal investigators who participated in the NN7088-3776 trial are thanked for their contributions. These were: W. Miesbach (Frankfurt, Germany); G. Castaman (Vicenza, Italy); T. Suzuki (Tokyo, Japan); M. Shima (Nara, Japan); B. Sørensen and E. Tuddenham (London, UK); A. Cuker (Philadelphia, USA); M. Escobar (Houston, USA); V. Radulescu (Lexington, USA); and C. Takemoto (Baltimore, USA). L. K. Amby, medical writer at Novo Nordisk A/S, provided editorial support for this manuscript, as did S. Eastwood and C. Price of PAREXEL (supported by Novo Nordisk A/S). T. Saugstrup, statistician at Novo Nordisk, provided the statistical analyses and programming.
A. Tiede has received research support and honoraria for lectures and consultancy from Novo Nordisk. B. Brand is an advisory board member for Novo Nordisk and Baxter, and has received honoraria for presentations from Novo Nordisk and Bayer. R. Fischer is an advisory board member for Novo Nordisk, has received travel support from Novo Nordisk, Bayer, CSL Behring, Pfizer, Baxter, and Biotest, and has received honoraria for lectures from Novo Nordisk, CSL Behring, and Pfizer. K. Kavakli is an advisory committee member for Novo Nordisk, and has received research funding, honoraria and travel costs for scientific events from Novo Nordisk. S. Lentz is a consultant for Novo Nordisk. C. Rea has received research funding from Baxter and SOBI, and travel and education funding from Novo Nordisk. K. Knobe is a Novo Nordisk employee. D. Viuff is a Novo Nordisk shareholder and former Novo Nordisk employee. T. Matsushita has no conflict of interest to declare. The trial was sponsored by Novo Nordisk A/S.