http://informahealthcare.com/ddi ISSN: 0363-9045 (print), 1520-5762 (electronic) Drug Dev Ind Pharm, 2015; 41(3): 470–475 ! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/03639045.2013.879721
RESEARCH ARTICLE
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
Pharmacokinetics and pharmacodynamics of FSK0808 and Gran after single intravenous drip administration or single subcutaneous administration: comparative study in healthy Japanese adult male subjects Kyoko Matsuguma1, Shunji Matsuki1, Chung Eunhee1, Akimasa Watanabe1, Asuka Tanaka1, Kei Sakamoto1, Hiromi Takeshita1, Akiko Hitaka1, Kyoko Shigetome1, Miyuki Kimura1, Akiko Miyamoto1, Shin Irie1, Daiki Kaneko2, and Akihiro Ohnishi3 1
LTA Medical Corporation Kyushu Clinical Pharmacology Research Clinic, Jigyo Chuo-ku, Fukuoka, Japan, 2Fuji Pharma Co., Ltd., 5-7 Sanban-cho, Chiyoda-ku, Tokyo, Japan, and 3Department of Laboratory Medicine, The Jikei University School of Medicine, 4-11-1 Izumihoncho, Komae, Tokyo, Japan Abstract
Keywords
FSK0808 is a recombinant human granulocyte colony-stimulating factor developed by Fuji Pharma Co., Ltd and Mochida Pharmaceutical Co., Ltd. as a biosimilar product of GranÕ . We verified the pharmacokinetic/pharmacodynamic equivalence of FSK0808 and commercially available GranÕ by a randomized crossover study of single intravenous dose (200 mg/m2) and single subcutaneous dose (400 mg/m2) in healthy Japanese adult male subjects. According to the bioequivalence guidelines, the area under the blood concentration – time curve by 48 hours after administration (AUC0–48) in a single intravenous drip (IVD) study, and AUC0–48 and maximum blood concentration (Cmax) in a single subcutaneous (SC) dose study were used as primary endpoints, and the pharmacodynamic parameters including absolute neutrophil count (ANC) or number of CD34 positive cells (CD34+ cells) as secondary endpoints. The safety was evaluated based on the characteristics and incidence of adverse reactions. As a result, the 90% confidence interval (CI) of the difference in mean value for AUC0–48 among drugs ranged from log(0.8) to log(1.25), in the IVD study, and those for Cmax and AUC0–48 were within the range of log(0.8)–log(1.25) in the SC study. Those for secondary endpoints were all within the range of log(0.8)–log(1.25). Thus, the pharmacokinetics/pharmacodynamics of both drugs were considered equivalent for all routes of administration, and the profiles of adverse reactions were also very similar.
Biosimilar, clinical trial, filgrastim, G-CSF, pharmacodynamics, pharmacokinetics
Introduction Granulocyte colony-stimulating factor (G-CSF) is a glycoprotein composed of a single polypeptide chain of 174 amino acids with O-glycosylation at one threonine residue. It stimulates the proliferation of neutropenic progenitor cells and their differentiation to granulocytes, functionally activates mature neutrophils, and enhances neutrophilic function1. Recombinant human G-CSF (rhG-CSF) is in clinical use for bone marrow transplant, neutropenia accompanied by various hematological disorders, and chemotherapy-related neutropenia2. However, peripheral hematopoietic stem cells induced in cancer patients who received rhG-CSF3,4, causing concerns in clinical use5. The transplantation of allogeneic peripheral blood stem cells (PBSC) has now become available using PBSC
Address for correspondence: Daiki Kaneko, Fuji Pharma Co., Ltd., 5-7 Sanban-cho, Chiyoda-ku, Tokyo 102-0075, Japan. Tel: +81-3-3556-3344. E-mail: [email protected]
History Received 23 July 2013 Revised 21 December 2013 Accepted 27 December 2013 Published online 29 January 2014
induction, the optimal conditions for which are determined by the number of CD34 positive cells (CD34+ cells)2,6. Filgrastim (C845H1339N223O243S9, molecular mass of 18 798.61) is a non-glycosylated rhG-CSF (r-metHuG-CSF) produced by Escherichia coli (E. coli) bacteria with the addition of an N-terminal methionine. It received FDA approval under the trade name NeupogenÕ (Amgen Inc., Thousand Oaks, CA) in 1991, and in Japan under the brand name GranÕ (Kyowa Hakko Kirin Co., Ltd., Tokyo, Japan) in the same year. Guidance of biosimilar was issued and rule-ized in Japan in 2009. FSK0808 was developed as a Biosimilar of filgrastim. FSK0808 referred to the chemical synthesis of the cDNA, included the human bladder cancer cell G-CSF genes in E. coli, and produced the raw E. coli from human G-CSF like precedence medical supplies, FSK0808 cultivates E. coli, purification, and as is injectable formulation. The equally homogeneous thing is checked by physicochemical analysis. FSK0808 was developed by Fuji Pharmaceuticals and Mochida Pharmaceutical Co., Ltd; it is a biosimilar product of GranÕ and its active ingredient has a very similar structure. Since filgrastim has mobilization effect on
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
DOI: 10.3109/03639045.2013.879721
peripheral blood of hematopoietic stem cells and neutrophilia effect, it evaluated the equivalence of PD with Gran for the number of CD34+ cells, and absolute neutrophil count (ANC) by the study against the indication. In addition, in international biotechnology of filgrastim Biosimilar, it is assessed similarly. To verify the equivalence between FSK0808 and GranÕ with reference to the ‘‘Guidelines for Quality, Safety and Efficacy of Biosimilar Products’’ (Guidelines for Biosimilar Products)7, the pharmacokinetics (PK) and pharmacodynamics (PD) after single administration were evaluated in healthy adult male subjects. The Guidelines require the equivalence of PK/PD to be examined by route of administration of the preceding biological product (previously approved drug) for the target indication. In the case of GranÕ , since the routes of administration for the indication by neutrophilic effect are intravenous (IV) and subcutaneous (SC) administration and that for mobilization of hematopoietic stem cells is SC administration, we conducted one study by IV and another study by SC administration. In the IV study, however, although it was administered over one minute by IV infusion, anaphylactoid reaction occurred in one subject in the first administration (period I), so the study was discontinued and intravenous drip (IVD) was used instead.
Methods Study design This study was designed with reference to the Guidelines for Biosimilar Products7 and the ‘‘Guidelines of Bioequivalence Studies for Generic Drugs’’ (BE Guidelines)8. A two-group, two-period randomized crossover study was conducted to confirm the equivalence of pharmacokinetic parameters when the test drug (FSK0808) or the control drug (GranÕ : filgrastim) was intravenously or subcutaneously administered once. As the secondary endpoint, the equivalence of pharmacodynamic parameters was to be confirmed using ANC as an indicator, but since the control drug given by SC administration is also indicated for mobilization of hematopoietic stem cells in the peripheral blood, the PD parameters were also examined using the number of CD34+ cells as an indicator. The protocol of this study was approved by the Institutional Review Board of Kyushu Clinical Pharmacology Research Clinic (KPC) in Fukuoka, Japan. All the studies were conducted at KPC according to the Declaration of Helsinki, Ethical Principles for Medical Research Involving Human Subjects and The Japanese GCP (Good Clinical Practice) Guidelines. All subjects received a detailed explanation and gave informed consent before participating in the study. Healthy adult male subjects aged 20–40 years with a body weight of 50–80 kg and a Body Mass Index (BMI) of 18.5– 25.0 kg/m2, without any present illness, history of allergies to any medications, any medical histories that were disqualified by the physician, or previous experience with rhG-CSF formulations were recruited and screened. The subjects were then enrolled after medical, physical, and laboratory examinations (such as hepatitis B, hepatitis C, HIV, syphilis, and drugs of abuse urine tests). The subjects were randomly allocated to one of the two drugs. While the first SC study was conducted in an open trial, the IVD study was conducted in a double-blind trial: the allocation was known only to the drug assignor, who dispensed the drugs and confirmed the indistinguishability. For both the test drug FSK0808 and the control drug GranÕ , injections containing 150 mg/0.6 mL and 300 mg/0.7 mL were used in the IVD study and the SC study, respectively. In the IVD study, both drugs were diluted with 5% glucose to make 60 mL to adjust the dose to 200 mg/m2, and administered once a day by
Pharmacokinetics and pharmacodynamics of FSK0808
471
intravenous drip in 30 minutes. In the SC study, both drugs were subcutaneously administered to the upper arm once a day at a dose of 400 mg/m2. For both studies, the treatment-free interval was 21 days after the end of administration. For the pharmacokinetic evaluation using plasma filgrastim concentration as an indicator in the SC study, the blood was collected at 14 time points (before administration, and at 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24 and 48 hours after administration) in the IVD study, and at 13 time points (before administration, and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 36 and 48 hours after administration). For the pharmacodynamic evaluation using ANC as an indicator in the SC study, the blood was collected at 16 time points (before administration, and at 0.5, 1, 2, 4, 6, 8, 10, 12, 14, 24, 36, 48, 72, 96 and 168 hours after administration) in the IVD study, and at 14 time points (before administration, and at 0.5, 1, 2, 4, 6, 8, 12, 24, 36, 48, 72, 96 and 168 hours after administration). In the pharmacodynamic evaluation using CD34+ cells as an indicator in the SC study, the blood was collected at 7 time points (before administration, and at 24, 48, 72, 96, 168 and 338 hours after administration). The plasma filgrastim concentration was determined by the enzyme-linked immunosorbent assay (ELISA) test kit (Immuno-Biological Laboratories Co., Ltd., Gunma, Japan) according to the manufacture’s instructions, ANC was determined by flow cytometry by KPC using XT-2000i of Cysmex Corporation (Kobe, Japan), and CD34+ cells were determined by flow cytometry by Mitsubishi Chemical Medience Corporation (Tokyo, Japan) using FACSCalibur of Becton, Dickinson and Company (Franklin Lakes, NJ). The safety was evaluated based on the incidence and nature of adverse reactions. The symptoms, vital signs, the values for instrument analysis and the severity of abnormal laboratory test values were evaluated according to the Common Toxicity Criteria (NCI-CTCAE) v3.0 of the National Cancer Institute, and the severity of the symptoms with no standard was considered 1 for mild, 2 for moderate, 3 for severe, 4 for life-threatening or disabling, and 5 for fatal symptoms. Among the adverse events which occurred during the clinical study period, those that did not negate the causal relationship with the drug were counted as adverse reactions. Anti-G-CSF antibody was determined by the ELISA method by Mitsubishi Chemical Medience Corporation. Analytical method The required number of subjects was calculated based on the phase I study of biosimilar products9 approved overseas, and was 12 per group (24 in total) in the IVD study and 20 per group (40 in total) in the SC study, considering the number who discontinued or dropped out. In the pharmacokinetic evaluation, the descriptive statistics (arithmetic mean and standard deviation) of Cmax, AUC0–48, AUC by infinite time (AUC1), and the elimination rate constant (kel) were calculated according to the BE Guidelines8. Regarding Cmax, AUC0–48, AUC1 and kel, statistics were compared between both drugs by analysis of variance. As a primary endpoint, if the 90% CI of the difference in mean value between the drugs after logarithmic transformation of AUC0–48 in the IVD study and Cmax (observed peak value) and AUC0–48 in the SC study was within the range of log(0.8)–log(1.25), both drugs were considered equivalent. In addition, AUC was calculated by the linear trapezoidal method. In the pharmacodynamic evaluation using ANC as an indicator, the descriptive statistics of the maximum ANC (ANC Cmax), the area under the ANC – time curve by 168 hours after initial administration (ANC AUC0–168), and the time to reach the
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
472
K. Matsuguma et al.
Drug Dev Ind Pharm, 2015; 41(3): 470–475
maximum ANC (ANC tmax) were calculated, and statistics for ANC Cmax and ANC AUC0–410 were compared between both drugs by analysis of variance. The ratio of the median value for the control drug in the confidence interval of the difference in median value at ANC tmax between the drugs was calculated. Regarding ANC Cmax and ANC AUC0–410, when the 95% CI was within the range of log(0.8)–log(1.25), both drugs were considered equivalent. Regarding ANC tmax, when the ratio of the median value for the control drug in the 95% CI of the difference in median value between the drugs was within the range of 0.20 to +0.20, both drugs were considered equivalent. In the pharmacodynamic evaluation using the number of CD34+ cells as an indicator, the descriptive statistics of the maximum number of CD34+ cells (CD34+ Cmax), the area under the number of CD34+ cells – time curve by 338 hours after initial administration (CD34+ AUC0–338) and the time to reach the maximum number of CD34+ cells (CD34+ tmax) were calculated, and evaluated by the same analytical method as for ANC. All statistical analyses were conducted with SAS ver9.1 (SAS Institute Japan, Tokyo, Japan) and Win Nonlin Professional ver6.1 (PharsightÕ Corporation, Mountain View, CA).
Results The study was conducted on total of 64 healthy adult male subjects (24 subjects in IVD study and 40 subjects in SC study). After the end of administration in period I, two subjects discontinued the study, including one because of an adverse event and one because of personal reasons. The 62 subjects who completed the administration in period II (23 in the IVD study, 39 in the SC study) were included in the pharmacokinetic/ pharmacodynamic analysis set. In IVD study, the mean age (minimum–maximum), mean height (minimum–maximum), mean body weight (minimum–maximum) and mean BMI (minimum–maximum) were 26.3 years old (20–39), 172.2 cm (163.7– 181.7), 63.4 kg (50.9–73.3) and 21.3 kg/m2 (18.6–24.1), respectively (Table 1). In SC study, the mean age, mean height, mean body weight and mean BMI were 23.4 years old (20–39), 171.8 cm (156.1–185.1), 62.6 kg (50.4–75.8) and 21.4 kg/m2 (18.5–24.7), respectively (Table 1). For evaluation of safety, all of the 64 subjects who received the drug were included in the safety analysis set. The mean plasma filgrastim concentration after single IVD or SC administration of the test drug or control drug is shown in Figure 1, and the pharmacokinetic parameters are shown in Table 2. The mean plasma concentration profiles of filgrastim were similar in both studies. In IVD study, 90%CI for AUC0–48
Table 1. Summary of demographic characteristics. Study period
Study
n
Age (year)*
Height (cm)*
Weight (kg)*
BMI*
2011/3 to 2011/5 2009/7 to 2009/9 Total
IVD SC
23 39 62
26.3 (20–39) 23.4 (20–34) 24.5 (20–39)
172.19 (163.7–181.7) 171.83 (156.1–185.1) 171.96 (156.1–185.1)
63.43 (50.9–73.3) 62.60 (50.4–75.8) 62.90 (50.4–75.8)
21.32 (18.6–24.1) 21.14 (18.5–24.7) 21.20 (18.5–24.7)
*Mean (min–max).
Figure 1. Mean serum filgrastim profiles after administration of FSK0808 and GranÕ (Mean ± SD) (linear scale). The Gran data sets were shifted by 0.3 h to distinguish between different data sets.
Pharmacokinetics and pharmacodynamics of FSK0808
DOI: 10.3109/03639045.2013.879721
473
Table 2. Mean pharmacokinetic parameters and their 90% CI of difference in means.
Study
Parameter
FSK0808
Gran
Difference of means
Mean ± SD 90%CI of difference of means
IVD
AUC0–48 (ngh/mL) Cmax (ng/mL)
420.64 ± 61.77 101.9864 ± 13.2957
463.54 ± 55.08 112.1084 ± 12.4777
log(0.9051) log(0.9057)
log(0.8690)–log(0.9426) log(0.8655)–log(0.9476)
SC
AUC0–48 (ngh/mL) Cmax (ng/mL) AUC0–1 (ngh/mL) Kel (h1)
534.59 ± 120.91 35.4774 ± 9.0794 539.84 ± 120.41 0.10889 ± 0.01780
562.02 ± 116.33 37.4935 ± 8.6942 567.85 ± 116.44 0.10290 ± 0.01844
log(0.9470) log(0.9408) log(0.9467) log(1.0627)
log(0.9098)–log(0.9857) log(0.8900)–log(0.9944) log(0.9100)–log(0.9850) log(1.0061)–log(1.1226)
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
Figure 2. Changes of means of ANC and CD34+ numbers after administration of FSK0808 and GranÕ (Mean ± SD) (linear scale). The Gran data sets were shifted by 1 h to distinguish between different data sets.
which was the primary endpoint, is log(0.869)–log(0.9426) and for Cmax and AUC0–48, which were the primary endpoints in the SC study were log(0.8900)–log(0.9944) and log(0.9098)– log(0.9857), respectively. For primary endpoints of both studies, the 90% CI was within the range of log(0.8)–log(1.25) which is acceptable range of bioequivalence. AUC1 and kel, the secondary
endpoints, were the 90% CI was within the range of log(0.8)– log(1.25) which is acceptable range in both studies. The change in the mean ANC in the peripheral blood and that in the mean number of CD34+ cells after single IVD or SC administration are shown in Figure 2, and the pharmacodynamic parameters are shown in Table 3.
474
K. Matsuguma et al.
Drug Dev Ind Pharm, 2015; 41(3): 470–475
Table 3. Mean pharmacodynamic parameters and their 95% CI of difference in means.
Study
Parameter
FSK0808
Gran
Difference of means*
Mean ± SD 95%CI of difference of meansy
IVD
ANC Cmax (cells102/mL) ANC AUC0–168 (cells102 h/mL) ANC tmax (h)
206.77 ± 39.66 11 017.21 ± 1625.73 12.3 ± 1.1
203.60 ± 61.01 11 219.75 ± 2324.71 12.2 ± 0.8
log(1.0280) log(0.9867) 0.0000
log(0.9813)–log(1.0769) log(0.9581)–log(1.0161) 0.0000–0.0833
SC
ANC Cmax (cells102/mL) ANC AUC0–168 (cells102 h/mL) ANC tmax (h) CD34+ Cmax (cells102 h/mL) CD34+ AUC0–338 (cellsh/mL) CD34+ tmax (h)
252.06 ± 56.00 16 836.95 ± 3204.67 25.2 ± 4.6 9.323 ± 4.835 1498.91 ± 771.48 81.8 ± 14.3
252.68 ± 54.10 16 808.18 ± 3223.29 26.5 ± 4.9 9.580 ± 5.299 1475.29 ± 769.83 83.7 ± 13.3
log(0.9956) log(1.0011) 0.0000 log(0.9740) log(1.0169) 0.0000
log(0.9601)–log(1.0324) log(0.9738)–log(1.0291) 0.0000–0.0000 log(0.8981)–log(1.0564) log(0.9494)–log(1.0891) 0.0000–0.0000
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
*Difference of median at ANC tmax and CD34+ tmax. yRate against median of Gran in 95%CI of difference of median at ANC tmax and CD34+ tmax.
In SC study, the mean ANC profiles was also similar profile, and the 95% CI for both ANC Cmax and ANC AUC0–168, between the drugs were log(0.9601)–log(1.0324), log(0.9738)–log(1.0291), respectively, and they were within the range of log(0.8)–log(1.25). Regarding ANC tmax, the ratio of the median value for the control drug in the 95% CI was within the range of 0.20 to +0.20. In both studies, another secondary endpoints, between the drugs were within the range of log(0.8)–log(1.25). On the other hand, in the pharmacodynamic evaluation using the number of CD34+ cells after single SC administration, the mean CD34+ profile was similar, and the 95% CI for CD34+ Cmax and CD34+ AUC0–338, was log(0.8981)–log(1.0564), log(0.9494)–log(1.0891), respectively. They were within the range of log(0.8)–log(1.25). Regarding CD34+ tmax, the ratio of the median value for the control drug in the 95% CI was within the range of 0.20 to +0.20. In both study, another secondary endpoint, between the drugs was within the range of log(0.8)– log(1.25). On the other hand, in the pharmacodynamic evaluation using the number of CD34+ cells after single SC administration, the 95% CI for CD34+ Cmax and CD34+ AUC0–338, ANC tmax, another secondary endpoint, between the drugs was within the range of log(0.8)–log(1.25). Regarding CD34+ tmax, the ratio of the median value for the control drug in the 95% CI was within the range of 0.20 to +0.20. Regarding safety, death and serious adverse events were not observed in any study. The adverse reactions resulting in discontinuation were positive urinary blood and positive urinary protein in one subject (1.6%) upon administration of the control drug in the IVD study. These adverse reactions were mild and recovered with no treatment. The adverse reactions are shown in Table 4; the main ones were back pain, headache, increased laboratory test values (UA and reticulocytes) and malaise, all of which were mild to moderate. In this study, the anti-G-CSF antibody was not detected in both the drugs.
Discussion There are several biosimilar G-CSFs approved in Europe: Filgrastim HEXALÕ , RatiograstimÕ , BiograstimÕ , Õ Õ Õ 10,11 . The physicochemical Tevagrastim , Zartio and Nivestim comparative studies revealed that the physicochemical properties of these biosimilar filgrastims are similar to the originator product (NeupogenÕ )12,13. FSK0808, which is a biosimilar filgrastim, also confirmed the similarity of the physicochemical property with the originator product in Japan (GranÕ ). This study was carried out in accordance with the Guidelines for Biosimilar Products. In terms of PK/PD of G-CSF, the theoretical studies were performed by
Table 4. Summary of adverse reactions. FSK0808 (n ¼ 62) Adverse reactions* Back pain Headache Elevated UA level Increased reticulocyte count Fatigue Arthralgia Elevated ALP level Nausea Pyrexia Feeling of warmth Reduction of plasma cholesterol Increased CRP Positive uric blood Decreased neutrophil count Positive urine sediment Positive uric protein
No. of cases 18 14 10 9 4 3 1 1 0 0 0 0 0 0 0 0
GranÕ (n ¼ 64)
Incidence No. of Incidence (%) cases (%) 29.0 22.6 16.1 14.5 6.5 4.8 1.6 1.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
23 20 7 10 5 1 0 0 2 1 1 1 1 1 1 1
35.9 31.3 10.9 15.6 7.8 1.6 0.0 0.0 3.1 1.6 1.6 1.6 1.6 1.6 1.6 1.6
*MedDRA/J Ver.14.0 UA: uric acid, ALP: alkaline phosphatase, CRP: C-reactive protein.
some groups. For example, Kryzanski et al. constructed the theoretical PK/PD model of a filgrastim14. Scoltz et al. also constructed the theoretical PK/PD model to understand the PK/PD differences between filgrastim and pegfilgrastim15. Their model was well fitted to clinical data, and could explain the PK/PD differences between filgrastim and pegfilgrastim. Thus, PK/PD studies of G-CSF have been reported because of their clinical importance. In this study, we applied FSK0808 and GranÕ to subjects, and analyzed PK/PD to evaluate the equivalence between these two drugs. In both the IVD study and SC study, the change in the plasma filgrastim concentration after administration of the drug was almost consistent. In the IVD study, the 90% CI for AUC0–48, the primary endpoint, between both drugs was within the range of log(0.8)–log(1.25) which is the acceptable range of bioequivalence, and the apparent Cmax showed a similar result. In the SC study, the 90% CIs for Cmax and AUC0–48, the primary endpoints, between the drugs were also within the acceptable range. Thus, the pharmacokinetics of FSK0808 and GranÕ were considered equivalent by both routes of administration. In the pharmacodynamic evaluation using ANC, the 95% CI for ANC Cmax and ANC AUC0–168, the endpoints in both studies, was within the acceptable range, and regarding ANC tmax, the ratio of the median value for the control drug in the 95% CI was also within the acceptable range. Based on this, the indices
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
DOI: 10.3109/03639045.2013.879721
showing the scale of pharmacological effects (ANC Cmax and ANC AUC0–168) and reactivity (ANC tmax) were equal to those for GranÕ by either route of administration, and when using ANC, these drugs were evaluated to be pharmacodynamically equivalent. On the other hand, in the pharmacodynamic evaluation using CD34+ cells after administration of the drug in the SC study, the 95% CI for CD34+ Cmax and CD34+ AUC0–338, the endpoints, between the drugs was within the acceptable range, and regarding CD34+ tmax, the ratio of the median value for the control drug in the 95% CI between the drugs was also within the acceptable range. Accordingly, even if using CD34+ cells, these drugs were evaluated to be pharmacodynamically equivalent. From the above results, the pharmacokinetics and pharmacodynamics of FSK0808 and GranÕ were considered to be equivalent. Regarding safety evaluation, the adverse reactions observed only with FSK0808 throughout both studies were nausea and increased ALP, both of which are known adverse reactions of GranÕ . When the IV study was conducted prior to the IVD study, one subject showed anaphylactoid response after the administration of FSK0808. In this case, since neither the presence of anti-G-CSF antibody nor the increase in nonspecific IgE antibody could be confirmed between before and after administration of FSK0808, this response was considered non-allergic. Similar events were reported with commercially available rhG-CSF or with antibody drugs and it is widely known that these events are resolved by adjusting the speed of administration of the drug16,17. We also confirmed that no subjects showed anaphylactoid response when we conducted the study again by intravenous drip. From this result, the anaphylactoid response with the IV study was considered to be an accidental event. The frequency of these adverse reactions was different from those of the report from Waller et al. This difference might be due to the difference of applied dose of filgrastim. In Japan, the equivalence evaluation between other biosimilar filgrastim (TKN732) and GranÕ has been confirmed by PD/PK study18,19. However, the efficacy and the safety of FSK0808 were evaluated by phase III trial in patients with breast cancer20 without reference product (GranÕ ). The results revealed that FSK0808 was safe and well tolerated in breast cancer patients undergoing chemotherapy and effectively stimulated the neutrophil recovery. This finding is consistent with our results. Our study also suggested that FSK0808 can be used for patients same as GranÕ .
Conclusions In these studies, the PK/PD study was conducted by all routes of administration of GranÕ used in clinical practice, and the type and incidence of adverse reactions were similar between both the drugs. In addition, safety was comparable between FSK0808 and GranÕ . In conclusion, FSK0808 is expected to show the same efficacy as GranÕ in clinical practice. We expect that the emergence of this biosimilar G-CSF will be reducing healthcare costs for the patients on recombinant biopharmaceutical treatment. Additionally, this may lead to even more advance in developing biosimilar G-CSF products in Japan.
Pharmacokinetics and pharmacodynamics of FSK0808
475
Declaration of interest Financial supports for these studies were provided by Fuji Pharma Co., Ltd and Mochida Pharmaceutical Co., Ltd. One author, Daiki Kaneko was an employee in Fuji Pharma Co., Ltd and the other authors have no conflicts of interest.
References 1. Nomura H, Imazeki I, Oheda M, et al. Purification and characterization of human granulocyte colony-stimulating factor (G-CSF). EMBO J 1986;5:871–6. 2. GranÕ Injection 75 150 M300 GranÕ Syringe 75 150 M300 interview forms (17th edition). Kyowa Hakko Kirin Co., Ltd. (revised June 2010). 3. Du¨hrsen U, Villeval JL, Boyd J, et al. Effects of recombinant human colony-stimulating factor on hematopoietic progenitor cells in cancer patients. Blood 1988;72:2074–81. 4. Sheridan WP, Begley CG, Juttner CA, et al. Effect of peripheral blood progenitor cells mobilised by filgrastim (G-CSF) on platelet recovery after high dose chemotherapy. Lancet 1992;339: 640–4. 5. Grigg AP, Roberts AW, Raunow H, et al. Optimizing dose and scheduling of filgrastim (granulocyte colony-stimulating factor) for mobilization and collection of peripheral blood progenitor cells in normal volunteers. Blood 1995;86:4437–45. 6. Guidelines on peripheral stem cell mobilization and collection from healthy donors for allogeneic peripheral blood stem cell transplantation. Revised 4th edition (June 30, 2010). 7. Guidelines for the quality, safety and efficacy assurance of follow-on biologics. Yakushoku shinsahatu 0304007, Ministry of Health, Labour and Welfare, Japan (March 4, 2009). 8. Guidelines on bioequivalence studies for generic drugs. Iyakushin 487, Ministry of Health, Labour and Welfare, Japan (December 22, 1997). 9. Assessment report for TEVAGRASTIM. (2008). European Medicines Agency, Procedure No. EMEA/H/C/827. 10. Gascon P. Presently available biosimilars in hematology-oncology: G-CSF. Targeted Oncol 2012;7:S29–34. 11. Waller CF, Bronchud M, Mair S, Challand R. Comparison of the pharmacodynamic profiles of a biosimilar filgrastim and Amgen filgrastim: results from a randomized, phase I trial. Ann Hematol 2010;89:971–8. 12. Skrlin A, Radic I, Vuletic M, et al. Comparison of the physicochemical properties of a biosimilar filgrastim with those of reference filgrastim. Biologicals 2010;38:557–66. 13. Soergel F, Lerch H, Lauber T. Physicochemical and biologic comparability of a biosimilar granulocyte colony-stimulating factor with its reference product. BioDrugs 2010;24:347–57. 14. Krzyzanski W, Wiczling P, Lowe P, et al. Population modeling of filgrastim PK-PD in healthy adults following intravenous and subcutaneous administrations. J Clin Pharmacol 2010;50:101S–12S. 15. Scholz M, Schirm S, Wetzler M, et al. Pharmacokinetic and -dynamic modelling of G-CSF derivatives in humans. Theor Biol Med Model 2012;9:32. 16. Ohta K, Masuzawa Y, Hirai M, et al. Two patients showing shocklike symptoms due to administration of rhG-CSF. Int J Hematol 1994;54:228. 17. Vogel WH. Infusion reactions: diagnosis, assessment, and management. Clin J Oncol Nurs 2010;14:E10–21. 18. Hasegawa S, Miyake J, Shioi Y, et al. Comparative studies of pharmacokinetic profiles of biosimilar filgrastim TKN732 and marketed filgrastim Gran. Jpn Phrmacol Ther 2013;41:251–60. 19. Hasegawa S, Miyake J, Shioi Y, et al. Comparative studies of pharmacodynamic profiles of biosimilar filgrastim TKN732 and marketed filgrastim gran. Jpn Phrmacol Ther 2013;41:261–74. 20. Sagara Y, Sato K, Fukuma E, et al. The efficacy and safety of FSK0808, Filgrastim Biosimilar: a multicenter, non-randomized study in Japanese patients with breast cancer. Jpn J Clin Oncol 2013; 43:865–73.
RESEARCH ARTICLE
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
Pharmacokinetics and pharmacodynamics of FSK0808 and Gran after single intravenous drip administration or single subcutaneous administration: comparative study in healthy Japanese adult male subjects Kyoko Matsuguma1, Shunji Matsuki1, Chung Eunhee1, Akimasa Watanabe1, Asuka Tanaka1, Kei Sakamoto1, Hiromi Takeshita1, Akiko Hitaka1, Kyoko Shigetome1, Miyuki Kimura1, Akiko Miyamoto1, Shin Irie1, Daiki Kaneko2, and Akihiro Ohnishi3 1
LTA Medical Corporation Kyushu Clinical Pharmacology Research Clinic, Jigyo Chuo-ku, Fukuoka, Japan, 2Fuji Pharma Co., Ltd., 5-7 Sanban-cho, Chiyoda-ku, Tokyo, Japan, and 3Department of Laboratory Medicine, The Jikei University School of Medicine, 4-11-1 Izumihoncho, Komae, Tokyo, Japan Abstract
Keywords
FSK0808 is a recombinant human granulocyte colony-stimulating factor developed by Fuji Pharma Co., Ltd and Mochida Pharmaceutical Co., Ltd. as a biosimilar product of GranÕ . We verified the pharmacokinetic/pharmacodynamic equivalence of FSK0808 and commercially available GranÕ by a randomized crossover study of single intravenous dose (200 mg/m2) and single subcutaneous dose (400 mg/m2) in healthy Japanese adult male subjects. According to the bioequivalence guidelines, the area under the blood concentration – time curve by 48 hours after administration (AUC0–48) in a single intravenous drip (IVD) study, and AUC0–48 and maximum blood concentration (Cmax) in a single subcutaneous (SC) dose study were used as primary endpoints, and the pharmacodynamic parameters including absolute neutrophil count (ANC) or number of CD34 positive cells (CD34+ cells) as secondary endpoints. The safety was evaluated based on the characteristics and incidence of adverse reactions. As a result, the 90% confidence interval (CI) of the difference in mean value for AUC0–48 among drugs ranged from log(0.8) to log(1.25), in the IVD study, and those for Cmax and AUC0–48 were within the range of log(0.8)–log(1.25) in the SC study. Those for secondary endpoints were all within the range of log(0.8)–log(1.25). Thus, the pharmacokinetics/pharmacodynamics of both drugs were considered equivalent for all routes of administration, and the profiles of adverse reactions were also very similar.
Biosimilar, clinical trial, filgrastim, G-CSF, pharmacodynamics, pharmacokinetics
Introduction Granulocyte colony-stimulating factor (G-CSF) is a glycoprotein composed of a single polypeptide chain of 174 amino acids with O-glycosylation at one threonine residue. It stimulates the proliferation of neutropenic progenitor cells and their differentiation to granulocytes, functionally activates mature neutrophils, and enhances neutrophilic function1. Recombinant human G-CSF (rhG-CSF) is in clinical use for bone marrow transplant, neutropenia accompanied by various hematological disorders, and chemotherapy-related neutropenia2. However, peripheral hematopoietic stem cells induced in cancer patients who received rhG-CSF3,4, causing concerns in clinical use5. The transplantation of allogeneic peripheral blood stem cells (PBSC) has now become available using PBSC
Address for correspondence: Daiki Kaneko, Fuji Pharma Co., Ltd., 5-7 Sanban-cho, Chiyoda-ku, Tokyo 102-0075, Japan. Tel: +81-3-3556-3344. E-mail: [email protected]
History Received 23 July 2013 Revised 21 December 2013 Accepted 27 December 2013 Published online 29 January 2014
induction, the optimal conditions for which are determined by the number of CD34 positive cells (CD34+ cells)2,6. Filgrastim (C845H1339N223O243S9, molecular mass of 18 798.61) is a non-glycosylated rhG-CSF (r-metHuG-CSF) produced by Escherichia coli (E. coli) bacteria with the addition of an N-terminal methionine. It received FDA approval under the trade name NeupogenÕ (Amgen Inc., Thousand Oaks, CA) in 1991, and in Japan under the brand name GranÕ (Kyowa Hakko Kirin Co., Ltd., Tokyo, Japan) in the same year. Guidance of biosimilar was issued and rule-ized in Japan in 2009. FSK0808 was developed as a Biosimilar of filgrastim. FSK0808 referred to the chemical synthesis of the cDNA, included the human bladder cancer cell G-CSF genes in E. coli, and produced the raw E. coli from human G-CSF like precedence medical supplies, FSK0808 cultivates E. coli, purification, and as is injectable formulation. The equally homogeneous thing is checked by physicochemical analysis. FSK0808 was developed by Fuji Pharmaceuticals and Mochida Pharmaceutical Co., Ltd; it is a biosimilar product of GranÕ and its active ingredient has a very similar structure. Since filgrastim has mobilization effect on
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
DOI: 10.3109/03639045.2013.879721
peripheral blood of hematopoietic stem cells and neutrophilia effect, it evaluated the equivalence of PD with Gran for the number of CD34+ cells, and absolute neutrophil count (ANC) by the study against the indication. In addition, in international biotechnology of filgrastim Biosimilar, it is assessed similarly. To verify the equivalence between FSK0808 and GranÕ with reference to the ‘‘Guidelines for Quality, Safety and Efficacy of Biosimilar Products’’ (Guidelines for Biosimilar Products)7, the pharmacokinetics (PK) and pharmacodynamics (PD) after single administration were evaluated in healthy adult male subjects. The Guidelines require the equivalence of PK/PD to be examined by route of administration of the preceding biological product (previously approved drug) for the target indication. In the case of GranÕ , since the routes of administration for the indication by neutrophilic effect are intravenous (IV) and subcutaneous (SC) administration and that for mobilization of hematopoietic stem cells is SC administration, we conducted one study by IV and another study by SC administration. In the IV study, however, although it was administered over one minute by IV infusion, anaphylactoid reaction occurred in one subject in the first administration (period I), so the study was discontinued and intravenous drip (IVD) was used instead.
Methods Study design This study was designed with reference to the Guidelines for Biosimilar Products7 and the ‘‘Guidelines of Bioequivalence Studies for Generic Drugs’’ (BE Guidelines)8. A two-group, two-period randomized crossover study was conducted to confirm the equivalence of pharmacokinetic parameters when the test drug (FSK0808) or the control drug (GranÕ : filgrastim) was intravenously or subcutaneously administered once. As the secondary endpoint, the equivalence of pharmacodynamic parameters was to be confirmed using ANC as an indicator, but since the control drug given by SC administration is also indicated for mobilization of hematopoietic stem cells in the peripheral blood, the PD parameters were also examined using the number of CD34+ cells as an indicator. The protocol of this study was approved by the Institutional Review Board of Kyushu Clinical Pharmacology Research Clinic (KPC) in Fukuoka, Japan. All the studies were conducted at KPC according to the Declaration of Helsinki, Ethical Principles for Medical Research Involving Human Subjects and The Japanese GCP (Good Clinical Practice) Guidelines. All subjects received a detailed explanation and gave informed consent before participating in the study. Healthy adult male subjects aged 20–40 years with a body weight of 50–80 kg and a Body Mass Index (BMI) of 18.5– 25.0 kg/m2, without any present illness, history of allergies to any medications, any medical histories that were disqualified by the physician, or previous experience with rhG-CSF formulations were recruited and screened. The subjects were then enrolled after medical, physical, and laboratory examinations (such as hepatitis B, hepatitis C, HIV, syphilis, and drugs of abuse urine tests). The subjects were randomly allocated to one of the two drugs. While the first SC study was conducted in an open trial, the IVD study was conducted in a double-blind trial: the allocation was known only to the drug assignor, who dispensed the drugs and confirmed the indistinguishability. For both the test drug FSK0808 and the control drug GranÕ , injections containing 150 mg/0.6 mL and 300 mg/0.7 mL were used in the IVD study and the SC study, respectively. In the IVD study, both drugs were diluted with 5% glucose to make 60 mL to adjust the dose to 200 mg/m2, and administered once a day by
Pharmacokinetics and pharmacodynamics of FSK0808
471
intravenous drip in 30 minutes. In the SC study, both drugs were subcutaneously administered to the upper arm once a day at a dose of 400 mg/m2. For both studies, the treatment-free interval was 21 days after the end of administration. For the pharmacokinetic evaluation using plasma filgrastim concentration as an indicator in the SC study, the blood was collected at 14 time points (before administration, and at 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24 and 48 hours after administration) in the IVD study, and at 13 time points (before administration, and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 36 and 48 hours after administration). For the pharmacodynamic evaluation using ANC as an indicator in the SC study, the blood was collected at 16 time points (before administration, and at 0.5, 1, 2, 4, 6, 8, 10, 12, 14, 24, 36, 48, 72, 96 and 168 hours after administration) in the IVD study, and at 14 time points (before administration, and at 0.5, 1, 2, 4, 6, 8, 12, 24, 36, 48, 72, 96 and 168 hours after administration). In the pharmacodynamic evaluation using CD34+ cells as an indicator in the SC study, the blood was collected at 7 time points (before administration, and at 24, 48, 72, 96, 168 and 338 hours after administration). The plasma filgrastim concentration was determined by the enzyme-linked immunosorbent assay (ELISA) test kit (Immuno-Biological Laboratories Co., Ltd., Gunma, Japan) according to the manufacture’s instructions, ANC was determined by flow cytometry by KPC using XT-2000i of Cysmex Corporation (Kobe, Japan), and CD34+ cells were determined by flow cytometry by Mitsubishi Chemical Medience Corporation (Tokyo, Japan) using FACSCalibur of Becton, Dickinson and Company (Franklin Lakes, NJ). The safety was evaluated based on the incidence and nature of adverse reactions. The symptoms, vital signs, the values for instrument analysis and the severity of abnormal laboratory test values were evaluated according to the Common Toxicity Criteria (NCI-CTCAE) v3.0 of the National Cancer Institute, and the severity of the symptoms with no standard was considered 1 for mild, 2 for moderate, 3 for severe, 4 for life-threatening or disabling, and 5 for fatal symptoms. Among the adverse events which occurred during the clinical study period, those that did not negate the causal relationship with the drug were counted as adverse reactions. Anti-G-CSF antibody was determined by the ELISA method by Mitsubishi Chemical Medience Corporation. Analytical method The required number of subjects was calculated based on the phase I study of biosimilar products9 approved overseas, and was 12 per group (24 in total) in the IVD study and 20 per group (40 in total) in the SC study, considering the number who discontinued or dropped out. In the pharmacokinetic evaluation, the descriptive statistics (arithmetic mean and standard deviation) of Cmax, AUC0–48, AUC by infinite time (AUC1), and the elimination rate constant (kel) were calculated according to the BE Guidelines8. Regarding Cmax, AUC0–48, AUC1 and kel, statistics were compared between both drugs by analysis of variance. As a primary endpoint, if the 90% CI of the difference in mean value between the drugs after logarithmic transformation of AUC0–48 in the IVD study and Cmax (observed peak value) and AUC0–48 in the SC study was within the range of log(0.8)–log(1.25), both drugs were considered equivalent. In addition, AUC was calculated by the linear trapezoidal method. In the pharmacodynamic evaluation using ANC as an indicator, the descriptive statistics of the maximum ANC (ANC Cmax), the area under the ANC – time curve by 168 hours after initial administration (ANC AUC0–168), and the time to reach the
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
472
K. Matsuguma et al.
Drug Dev Ind Pharm, 2015; 41(3): 470–475
maximum ANC (ANC tmax) were calculated, and statistics for ANC Cmax and ANC AUC0–410 were compared between both drugs by analysis of variance. The ratio of the median value for the control drug in the confidence interval of the difference in median value at ANC tmax between the drugs was calculated. Regarding ANC Cmax and ANC AUC0–410, when the 95% CI was within the range of log(0.8)–log(1.25), both drugs were considered equivalent. Regarding ANC tmax, when the ratio of the median value for the control drug in the 95% CI of the difference in median value between the drugs was within the range of 0.20 to +0.20, both drugs were considered equivalent. In the pharmacodynamic evaluation using the number of CD34+ cells as an indicator, the descriptive statistics of the maximum number of CD34+ cells (CD34+ Cmax), the area under the number of CD34+ cells – time curve by 338 hours after initial administration (CD34+ AUC0–338) and the time to reach the maximum number of CD34+ cells (CD34+ tmax) were calculated, and evaluated by the same analytical method as for ANC. All statistical analyses were conducted with SAS ver9.1 (SAS Institute Japan, Tokyo, Japan) and Win Nonlin Professional ver6.1 (PharsightÕ Corporation, Mountain View, CA).
Results The study was conducted on total of 64 healthy adult male subjects (24 subjects in IVD study and 40 subjects in SC study). After the end of administration in period I, two subjects discontinued the study, including one because of an adverse event and one because of personal reasons. The 62 subjects who completed the administration in period II (23 in the IVD study, 39 in the SC study) were included in the pharmacokinetic/ pharmacodynamic analysis set. In IVD study, the mean age (minimum–maximum), mean height (minimum–maximum), mean body weight (minimum–maximum) and mean BMI (minimum–maximum) were 26.3 years old (20–39), 172.2 cm (163.7– 181.7), 63.4 kg (50.9–73.3) and 21.3 kg/m2 (18.6–24.1), respectively (Table 1). In SC study, the mean age, mean height, mean body weight and mean BMI were 23.4 years old (20–39), 171.8 cm (156.1–185.1), 62.6 kg (50.4–75.8) and 21.4 kg/m2 (18.5–24.7), respectively (Table 1). For evaluation of safety, all of the 64 subjects who received the drug were included in the safety analysis set. The mean plasma filgrastim concentration after single IVD or SC administration of the test drug or control drug is shown in Figure 1, and the pharmacokinetic parameters are shown in Table 2. The mean plasma concentration profiles of filgrastim were similar in both studies. In IVD study, 90%CI for AUC0–48
Table 1. Summary of demographic characteristics. Study period
Study
n
Age (year)*
Height (cm)*
Weight (kg)*
BMI*
2011/3 to 2011/5 2009/7 to 2009/9 Total
IVD SC
23 39 62
26.3 (20–39) 23.4 (20–34) 24.5 (20–39)
172.19 (163.7–181.7) 171.83 (156.1–185.1) 171.96 (156.1–185.1)
63.43 (50.9–73.3) 62.60 (50.4–75.8) 62.90 (50.4–75.8)
21.32 (18.6–24.1) 21.14 (18.5–24.7) 21.20 (18.5–24.7)
*Mean (min–max).
Figure 1. Mean serum filgrastim profiles after administration of FSK0808 and GranÕ (Mean ± SD) (linear scale). The Gran data sets were shifted by 0.3 h to distinguish between different data sets.
Pharmacokinetics and pharmacodynamics of FSK0808
DOI: 10.3109/03639045.2013.879721
473
Table 2. Mean pharmacokinetic parameters and their 90% CI of difference in means.
Study
Parameter
FSK0808
Gran
Difference of means
Mean ± SD 90%CI of difference of means
IVD
AUC0–48 (ngh/mL) Cmax (ng/mL)
420.64 ± 61.77 101.9864 ± 13.2957
463.54 ± 55.08 112.1084 ± 12.4777
log(0.9051) log(0.9057)
log(0.8690)–log(0.9426) log(0.8655)–log(0.9476)
SC
AUC0–48 (ngh/mL) Cmax (ng/mL) AUC0–1 (ngh/mL) Kel (h1)
534.59 ± 120.91 35.4774 ± 9.0794 539.84 ± 120.41 0.10889 ± 0.01780
562.02 ± 116.33 37.4935 ± 8.6942 567.85 ± 116.44 0.10290 ± 0.01844
log(0.9470) log(0.9408) log(0.9467) log(1.0627)
log(0.9098)–log(0.9857) log(0.8900)–log(0.9944) log(0.9100)–log(0.9850) log(1.0061)–log(1.1226)
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
Figure 2. Changes of means of ANC and CD34+ numbers after administration of FSK0808 and GranÕ (Mean ± SD) (linear scale). The Gran data sets were shifted by 1 h to distinguish between different data sets.
which was the primary endpoint, is log(0.869)–log(0.9426) and for Cmax and AUC0–48, which were the primary endpoints in the SC study were log(0.8900)–log(0.9944) and log(0.9098)– log(0.9857), respectively. For primary endpoints of both studies, the 90% CI was within the range of log(0.8)–log(1.25) which is acceptable range of bioequivalence. AUC1 and kel, the secondary
endpoints, were the 90% CI was within the range of log(0.8)– log(1.25) which is acceptable range in both studies. The change in the mean ANC in the peripheral blood and that in the mean number of CD34+ cells after single IVD or SC administration are shown in Figure 2, and the pharmacodynamic parameters are shown in Table 3.
474
K. Matsuguma et al.
Drug Dev Ind Pharm, 2015; 41(3): 470–475
Table 3. Mean pharmacodynamic parameters and their 95% CI of difference in means.
Study
Parameter
FSK0808
Gran
Difference of means*
Mean ± SD 95%CI of difference of meansy
IVD
ANC Cmax (cells102/mL) ANC AUC0–168 (cells102 h/mL) ANC tmax (h)
206.77 ± 39.66 11 017.21 ± 1625.73 12.3 ± 1.1
203.60 ± 61.01 11 219.75 ± 2324.71 12.2 ± 0.8
log(1.0280) log(0.9867) 0.0000
log(0.9813)–log(1.0769) log(0.9581)–log(1.0161) 0.0000–0.0833
SC
ANC Cmax (cells102/mL) ANC AUC0–168 (cells102 h/mL) ANC tmax (h) CD34+ Cmax (cells102 h/mL) CD34+ AUC0–338 (cellsh/mL) CD34+ tmax (h)
252.06 ± 56.00 16 836.95 ± 3204.67 25.2 ± 4.6 9.323 ± 4.835 1498.91 ± 771.48 81.8 ± 14.3
252.68 ± 54.10 16 808.18 ± 3223.29 26.5 ± 4.9 9.580 ± 5.299 1475.29 ± 769.83 83.7 ± 13.3
log(0.9956) log(1.0011) 0.0000 log(0.9740) log(1.0169) 0.0000
log(0.9601)–log(1.0324) log(0.9738)–log(1.0291) 0.0000–0.0000 log(0.8981)–log(1.0564) log(0.9494)–log(1.0891) 0.0000–0.0000
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
*Difference of median at ANC tmax and CD34+ tmax. yRate against median of Gran in 95%CI of difference of median at ANC tmax and CD34+ tmax.
In SC study, the mean ANC profiles was also similar profile, and the 95% CI for both ANC Cmax and ANC AUC0–168, between the drugs were log(0.9601)–log(1.0324), log(0.9738)–log(1.0291), respectively, and they were within the range of log(0.8)–log(1.25). Regarding ANC tmax, the ratio of the median value for the control drug in the 95% CI was within the range of 0.20 to +0.20. In both studies, another secondary endpoints, between the drugs were within the range of log(0.8)–log(1.25). On the other hand, in the pharmacodynamic evaluation using the number of CD34+ cells after single SC administration, the mean CD34+ profile was similar, and the 95% CI for CD34+ Cmax and CD34+ AUC0–338, was log(0.8981)–log(1.0564), log(0.9494)–log(1.0891), respectively. They were within the range of log(0.8)–log(1.25). Regarding CD34+ tmax, the ratio of the median value for the control drug in the 95% CI was within the range of 0.20 to +0.20. In both study, another secondary endpoint, between the drugs was within the range of log(0.8)– log(1.25). On the other hand, in the pharmacodynamic evaluation using the number of CD34+ cells after single SC administration, the 95% CI for CD34+ Cmax and CD34+ AUC0–338, ANC tmax, another secondary endpoint, between the drugs was within the range of log(0.8)–log(1.25). Regarding CD34+ tmax, the ratio of the median value for the control drug in the 95% CI was within the range of 0.20 to +0.20. Regarding safety, death and serious adverse events were not observed in any study. The adverse reactions resulting in discontinuation were positive urinary blood and positive urinary protein in one subject (1.6%) upon administration of the control drug in the IVD study. These adverse reactions were mild and recovered with no treatment. The adverse reactions are shown in Table 4; the main ones were back pain, headache, increased laboratory test values (UA and reticulocytes) and malaise, all of which were mild to moderate. In this study, the anti-G-CSF antibody was not detected in both the drugs.
Discussion There are several biosimilar G-CSFs approved in Europe: Filgrastim HEXALÕ , RatiograstimÕ , BiograstimÕ , Õ Õ Õ 10,11 . The physicochemical Tevagrastim , Zartio and Nivestim comparative studies revealed that the physicochemical properties of these biosimilar filgrastims are similar to the originator product (NeupogenÕ )12,13. FSK0808, which is a biosimilar filgrastim, also confirmed the similarity of the physicochemical property with the originator product in Japan (GranÕ ). This study was carried out in accordance with the Guidelines for Biosimilar Products. In terms of PK/PD of G-CSF, the theoretical studies were performed by
Table 4. Summary of adverse reactions. FSK0808 (n ¼ 62) Adverse reactions* Back pain Headache Elevated UA level Increased reticulocyte count Fatigue Arthralgia Elevated ALP level Nausea Pyrexia Feeling of warmth Reduction of plasma cholesterol Increased CRP Positive uric blood Decreased neutrophil count Positive urine sediment Positive uric protein
No. of cases 18 14 10 9 4 3 1 1 0 0 0 0 0 0 0 0
GranÕ (n ¼ 64)
Incidence No. of Incidence (%) cases (%) 29.0 22.6 16.1 14.5 6.5 4.8 1.6 1.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
23 20 7 10 5 1 0 0 2 1 1 1 1 1 1 1
35.9 31.3 10.9 15.6 7.8 1.6 0.0 0.0 3.1 1.6 1.6 1.6 1.6 1.6 1.6 1.6
*MedDRA/J Ver.14.0 UA: uric acid, ALP: alkaline phosphatase, CRP: C-reactive protein.
some groups. For example, Kryzanski et al. constructed the theoretical PK/PD model of a filgrastim14. Scoltz et al. also constructed the theoretical PK/PD model to understand the PK/PD differences between filgrastim and pegfilgrastim15. Their model was well fitted to clinical data, and could explain the PK/PD differences between filgrastim and pegfilgrastim. Thus, PK/PD studies of G-CSF have been reported because of their clinical importance. In this study, we applied FSK0808 and GranÕ to subjects, and analyzed PK/PD to evaluate the equivalence between these two drugs. In both the IVD study and SC study, the change in the plasma filgrastim concentration after administration of the drug was almost consistent. In the IVD study, the 90% CI for AUC0–48, the primary endpoint, between both drugs was within the range of log(0.8)–log(1.25) which is the acceptable range of bioequivalence, and the apparent Cmax showed a similar result. In the SC study, the 90% CIs for Cmax and AUC0–48, the primary endpoints, between the drugs were also within the acceptable range. Thus, the pharmacokinetics of FSK0808 and GranÕ were considered equivalent by both routes of administration. In the pharmacodynamic evaluation using ANC, the 95% CI for ANC Cmax and ANC AUC0–168, the endpoints in both studies, was within the acceptable range, and regarding ANC tmax, the ratio of the median value for the control drug in the 95% CI was also within the acceptable range. Based on this, the indices
Drug Dev Ind Pharm Downloaded from informahealthcare.com by Nyu Medical Center on 06/16/15 For personal use only.
DOI: 10.3109/03639045.2013.879721
showing the scale of pharmacological effects (ANC Cmax and ANC AUC0–168) and reactivity (ANC tmax) were equal to those for GranÕ by either route of administration, and when using ANC, these drugs were evaluated to be pharmacodynamically equivalent. On the other hand, in the pharmacodynamic evaluation using CD34+ cells after administration of the drug in the SC study, the 95% CI for CD34+ Cmax and CD34+ AUC0–338, the endpoints, between the drugs was within the acceptable range, and regarding CD34+ tmax, the ratio of the median value for the control drug in the 95% CI between the drugs was also within the acceptable range. Accordingly, even if using CD34+ cells, these drugs were evaluated to be pharmacodynamically equivalent. From the above results, the pharmacokinetics and pharmacodynamics of FSK0808 and GranÕ were considered to be equivalent. Regarding safety evaluation, the adverse reactions observed only with FSK0808 throughout both studies were nausea and increased ALP, both of which are known adverse reactions of GranÕ . When the IV study was conducted prior to the IVD study, one subject showed anaphylactoid response after the administration of FSK0808. In this case, since neither the presence of anti-G-CSF antibody nor the increase in nonspecific IgE antibody could be confirmed between before and after administration of FSK0808, this response was considered non-allergic. Similar events were reported with commercially available rhG-CSF or with antibody drugs and it is widely known that these events are resolved by adjusting the speed of administration of the drug16,17. We also confirmed that no subjects showed anaphylactoid response when we conducted the study again by intravenous drip. From this result, the anaphylactoid response with the IV study was considered to be an accidental event. The frequency of these adverse reactions was different from those of the report from Waller et al. This difference might be due to the difference of applied dose of filgrastim. In Japan, the equivalence evaluation between other biosimilar filgrastim (TKN732) and GranÕ has been confirmed by PD/PK study18,19. However, the efficacy and the safety of FSK0808 were evaluated by phase III trial in patients with breast cancer20 without reference product (GranÕ ). The results revealed that FSK0808 was safe and well tolerated in breast cancer patients undergoing chemotherapy and effectively stimulated the neutrophil recovery. This finding is consistent with our results. Our study also suggested that FSK0808 can be used for patients same as GranÕ .
Conclusions In these studies, the PK/PD study was conducted by all routes of administration of GranÕ used in clinical practice, and the type and incidence of adverse reactions were similar between both the drugs. In addition, safety was comparable between FSK0808 and GranÕ . In conclusion, FSK0808 is expected to show the same efficacy as GranÕ in clinical practice. We expect that the emergence of this biosimilar G-CSF will be reducing healthcare costs for the patients on recombinant biopharmaceutical treatment. Additionally, this may lead to even more advance in developing biosimilar G-CSF products in Japan.
Pharmacokinetics and pharmacodynamics of FSK0808
475
Declaration of interest Financial supports for these studies were provided by Fuji Pharma Co., Ltd and Mochida Pharmaceutical Co., Ltd. One author, Daiki Kaneko was an employee in Fuji Pharma Co., Ltd and the other authors have no conflicts of interest.
References 1. Nomura H, Imazeki I, Oheda M, et al. Purification and characterization of human granulocyte colony-stimulating factor (G-CSF). EMBO J 1986;5:871–6. 2. GranÕ Injection 75 150 M300 GranÕ Syringe 75 150 M300 interview forms (17th edition). Kyowa Hakko Kirin Co., Ltd. (revised June 2010). 3. Du¨hrsen U, Villeval JL, Boyd J, et al. Effects of recombinant human colony-stimulating factor on hematopoietic progenitor cells in cancer patients. Blood 1988;72:2074–81. 4. Sheridan WP, Begley CG, Juttner CA, et al. Effect of peripheral blood progenitor cells mobilised by filgrastim (G-CSF) on platelet recovery after high dose chemotherapy. Lancet 1992;339: 640–4. 5. Grigg AP, Roberts AW, Raunow H, et al. Optimizing dose and scheduling of filgrastim (granulocyte colony-stimulating factor) for mobilization and collection of peripheral blood progenitor cells in normal volunteers. Blood 1995;86:4437–45. 6. Guidelines on peripheral stem cell mobilization and collection from healthy donors for allogeneic peripheral blood stem cell transplantation. Revised 4th edition (June 30, 2010). 7. Guidelines for the quality, safety and efficacy assurance of follow-on biologics. Yakushoku shinsahatu 0304007, Ministry of Health, Labour and Welfare, Japan (March 4, 2009). 8. Guidelines on bioequivalence studies for generic drugs. Iyakushin 487, Ministry of Health, Labour and Welfare, Japan (December 22, 1997). 9. Assessment report for TEVAGRASTIM. (2008). European Medicines Agency, Procedure No. EMEA/H/C/827. 10. Gascon P. Presently available biosimilars in hematology-oncology: G-CSF. Targeted Oncol 2012;7:S29–34. 11. Waller CF, Bronchud M, Mair S, Challand R. Comparison of the pharmacodynamic profiles of a biosimilar filgrastim and Amgen filgrastim: results from a randomized, phase I trial. Ann Hematol 2010;89:971–8. 12. Skrlin A, Radic I, Vuletic M, et al. Comparison of the physicochemical properties of a biosimilar filgrastim with those of reference filgrastim. Biologicals 2010;38:557–66. 13. Soergel F, Lerch H, Lauber T. Physicochemical and biologic comparability of a biosimilar granulocyte colony-stimulating factor with its reference product. BioDrugs 2010;24:347–57. 14. Krzyzanski W, Wiczling P, Lowe P, et al. Population modeling of filgrastim PK-PD in healthy adults following intravenous and subcutaneous administrations. J Clin Pharmacol 2010;50:101S–12S. 15. Scholz M, Schirm S, Wetzler M, et al. Pharmacokinetic and -dynamic modelling of G-CSF derivatives in humans. Theor Biol Med Model 2012;9:32. 16. Ohta K, Masuzawa Y, Hirai M, et al. Two patients showing shocklike symptoms due to administration of rhG-CSF. Int J Hematol 1994;54:228. 17. Vogel WH. Infusion reactions: diagnosis, assessment, and management. Clin J Oncol Nurs 2010;14:E10–21. 18. Hasegawa S, Miyake J, Shioi Y, et al. Comparative studies of pharmacokinetic profiles of biosimilar filgrastim TKN732 and marketed filgrastim Gran. Jpn Phrmacol Ther 2013;41:251–60. 19. Hasegawa S, Miyake J, Shioi Y, et al. Comparative studies of pharmacodynamic profiles of biosimilar filgrastim TKN732 and marketed filgrastim gran. Jpn Phrmacol Ther 2013;41:261–74. 20. Sagara Y, Sato K, Fukuma E, et al. The efficacy and safety of FSK0808, Filgrastim Biosimilar: a multicenter, non-randomized study in Japanese patients with breast cancer. Jpn J Clin Oncol 2013; 43:865–73.
