Bioequivalence of liquid and freeze-dried recombinant human follicle-stimulating hormone

CURRENT MEDICAL RESEARCH AND OPINION® VOL. 21, NO. 1, 2005, 121–125 © 2005 LIBRAPHARM LIMITED 0300-7995 doi: 10.1185/030079904X18027 All rights reserved: reproduction in whole or part not permitted ORIGINAL ARTICLE Bioequivalence of liquid and freeze-dried recombinant human follicle-stimulating hormone I. Lugan a, S. Febbraro b, H. Lecuelle a, O. Papasouliotis a, Q. Ho-Nguyen a and M. Buraglio a a Serono International SA, R&Ph Development HPG, Geneva, Switzerland b SIMBEC Research Ltd, Merthyr Tydfil, Mid Glamorgan, UK Address for correspondence: Isabelle Lugan, Serono International SA, 15bis Chemin des Mines, case postale 54, 1211 Geneva 20, Switzerland. Tel. +41 22 739 3664; Fax +41 22 739 3635; email [email protected] Key words: Filled-by-mass – Freeze-dried – Follicle-stimulating hormone, recombinant human – Liquid – Pharmacokinetics – Therapeutic equivalency Objective: The bioequivalence and tolerability of freeze-dried and liquid formulations of recombinant human follicle-stimulating hormone (r-hFSH) filled- by-mass were assessed in a crossover, open-label, randomised, single-centre, phase I bioequivalence study. Methods: Following pituitary down-regulation with the gonadotrophin-releasing hormone agonist goserelin, healthy adult volunteers (18 years–45 years of age) received single subcutaneous injections of r-hFSH , 300 IU, from freeze-dried and liquid formulations in random order, separated by a 7-day washout period. Blood was obtained over 144 h for pharmacokinetic analysis. Main outcome measures: These were peak serum FSH concentrations (C max ), time to peak concentration (T max ) and area under the concentration–time curve from zero to the last measurable concentration (AUC last ), local and systemic tolerability. Results: Of 44 volunteers who underwent down-regulation, 39 (18 men, 21 women) completed the study. C max and AUC last were similar with the freeze-dried (mean 9.51 IU/L and 844 IU.h/L, respectively) and liquid (mean 8.99 IU/L and 841 IU.h/L, respectively) formulations, whereas T max was significantly higher with the liquid formulation (median 12 h vs 15 h, p = 0.0183). The 90% confidence intervals for the ratio of the treatment means for C max and AUC last were within the pre-defined bioequivalence range of 0.8–1.25. Conclusion: Both formulations were well tolerated with regard to both systemic and local adverse events. The freeze-dried and liquid formulations of r-hFSH are bioequivalent and show no significant differences in tolerability. Thus, the liquid formulation is expected to provide comparable efficacy and tolerability to the freeze- dried formulation in clinical use. Introduction Follicle-stimulating hormone (FSH) is a pituitary glycoprotein hormone that plays a key role in regulating reproductive function in both males and females1. Deficiency of this hormone leads to ovulatory failure or dysfunction in women and impaired spermatogenesis in men. Clinically, exogenous FSH is administered for the stimulation of follicular development in women under going assisted reproductive technologies and the induction of ovulation in women with anovulatory infertility2. A B S T R A C T Paper 2836 121 C ur r M ed R es O pi n D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y D ic le U ni v. o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. 122 Bioequivalence of liquid and freeze-dried r-hFSH © 2005 LIBRAPHARM LTD – Curr Med Res Opin 2005; 21(1) FSH for clinical use was previously derived from human urine; now a recombinant human FSH (r-hFSH) preparation, Gonal-F*, is available. This product is pro duced by transfecting the genes coding for the alpha and beta subunits of human FSH into Chinese hamster ovary cells, and possesses a higher specific activity (bio activity/unit mass) than can be achieved with urine-derived products. r-hFSH offers high batch-to-batch consistency, resulting in accurate and reliable dosing3. Recent improvements in manufacturing processes have allowed dosage vials of r-hFSH to be filled by mass (FbM, µg) rather than bioactivity (IU), resulting in a guaranteed dose from vial to vial8. r-hFSH FbM was originally developed as a sterile freeze-dried preparation, but a multidose liquid formulation has subsequently been developed to be used with a pre- filled pen. Reconstitution before administration is no longer required, thereby simplifying administration and providing accurate dosing by reducing the number of steps required for FSH administration. The present study was performed to assess the bioequivalence of liquid and freeze-dried r-hFSH FbM; a secondary objective was to compare the systemic and local tolerability of each preparation. Patients and methods The trial was an open, randomised, crossover bioequiv- alence4,5 study designed to compare the pharmacokinetics of the freeze-dried and liquid formulations of r-hFSH FbM. It was approved by the Simbec Independent Ethics Committee and conducted according to the principles of the Declaration of Helsinki as revised in Edinburgh (2000) and the International Conference on Harmon ization Tripartite Guidelines for Good Clinical Practice. Written informed consent was obtained from all participants before enrolment. Volunteers Healthy male and female volunteers aged between 18 years and 45 years were included in the study. They were required to be at least 50 kg in weight for females and 60 kg for males, to have a body mass index of no more than 28 kg/m2 and to have normal vital signs. Women were required to have had a negative pregnancy test and to be using combined oral contraceptives for at least two cycles prior to the study and for at least one cycle after the study. All volunteers were required to smoke fewer than 10 cigarettes per day and to agree to smoke fewer than five cigarettes per day during the study. Exclusion criteria included use of hormonal medication other than oral contraceptives, evidence of illness that might interfere with drug pharmacokinetics, a history of adverse drug reactions or hypersensitivity to drugs with a similar chemical structure to r-hFSH , infection with hepatitis B or C or HIV, a history of drug or alcohol abuse, admin istration of any investigational drug within 12 weeks prior to inclusion, and, in women, breast-feeding. Treatments All volunteers underwent pituitary down-regulation with goserelin (Zoladex, AstraZeneca, Macclesfield, UK), 3.6 mg, by a single subcutaneous injection. Female volunteers continued taking combined oral contraceptives until the seventh day after goserelin injection, and then discontinued oral contraception for the duration of the study. Following down-regulation, all volunteers received two doses of r-hFSH FbM, 300 IU (22 µg), by subcutaneous injection, separated by a washout period of at least 7 days; an additional 2 days washout was permitted if serum FSH concentrations were above 4 IU/L in women or 2 IU/L in men. One dose of r-hFSH consisted of 1 mL of the freeze-dried formulation, and the other of 0.48 mL of the liquid formulation. The order of the two doses was randomised according to a list generated at Serono SA, Geneva, Switzerland. The volunteers were instructed to abstain from alcohol for 24 h prior to the administration of r-hFSH , and to avoid caffeine or xanthine-containing products from 12 h before dosing to 24 h after dosing. A light breakfast was provided 2 h before dosing and volunteers were asked to avoid excessive eating or drinking throughout the study. Assessments Blood samples were taken for serum FSH measurement prior to dosing and at 1 h, 2 h, 4 h, 6 h, 8 h, 10 h, 12 h, 15 h, 24 h, 48 h, 72 h, 96 h, 120 h and 144 h after dosing. Samples of 5.5 mL of venous blood were withdrawn through a cannula inserted into the vein into poly propylene tubes containing barrier beads and allowed to stand for 15 min at room temperature to allow clotting. Thereafter, tubes were refrigerated for approximately 1 h, after which time samples were centrifuged at 3000 rpm for 10 min at 4oC. Two 1 mL aliquots of serum were transferred into fresh tubes and stored at or below –20oC prior to FSH analysis. Serum FSH was measured by means of a validated commercially available immunoradiometric assay (Daiichi Radioisotope Laboratories, Tokyo, Japan). The assay had a lower limit of quantification of 0.5 IU/L and intra- and inter-assay coefficients of variation between * Gonal-F is a registered trademark of Serono International SA, Geneva, Switzerland C ur r M ed R es O pi n D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y D ic le U ni v. o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. © 2005 LIBRAPHARM LTD – Curr Med Res Opin 2005; 21(1) Bioequivalence of liquid and freeze-dried r-hFSH Lugan et al. 123 3.0% and 5.3% over a sample concentration range of 0.52 IU/L–50 IU/L. The measured FSH concentration– time curves were assessed by means of non-compart mental analysis to derive the peak concentrations (C max ), the time to peak concentration (T max ) and the area under the concentration–time curve from zero to the last measurable concentration (AUC last ). All pharmacokinetic analyses were performed with WinNonLin Professional software (version 3.2, Pharsight, Mountain View, CA, USA). A safety evaluation, including medical examination, 12-lead electrocardiogram, blood sampling for haematology and clinical chemistry, and urinalysis, was performed within 2 weeks prior to down-regulation and within 7 days after the last pharmacokinetic sample was obtained following the second r-hFSH injection. Vital signs (heart rate, blood pressure and oral body temperature) and local tolerability were monitored prior to each dosing and at 2 h, 8 h, 12 h, 15 h and 24 h after each dosing. Inform ation on adverse events was recorded throughout the study. Statistics C max and AUC last were transformed to natural logarithms and analysed by an analysis of variance with the factors sequence, volunteer within sequence, period and treat- ment. Treatment by gender interaction was investigated but was found to have no statistically significant effect. The potential influence of baseline FSH concentrations was assessed by an analysis of covariance (ANCOVA). Based on the log-transformed data, a range of 0.8–1.25 for the 90% confidence interval for the ratios of the treatment means (liquid:freeze-dried) for C max and AUC last was used as the criterion for equivalence. T max values were analysed by the non-parametric Wilcoxon-signed rank test. Based on the assumptions that the true mean ratio of AUC (liquid:freeze-dried) would be between 0.90 and 1.10, a range of 0.8–1.25 would indicate bioequivalence and that intra-subject variability would be approximately 20%, it was calculated that a sample size of 37 evaluable volunteers would provide 80% power to detect equivalence using Schuirmann’s method. Hence, it was planned to recruit 44 volunteers in order to accommodate an expected dropout rate of 15%. Results A total of 44 volunteers (22 men, 22 women) entered the study and underwent down-regulation. Their mean age (± SD) was 28.0 ± 6.8 years (27.4 ± 7.0 years for men and 28.6 ± 6.7 years for women) and their mean weight was 70.1 ± 11.6 kg (72.2 ± 11.9 kg for men and 63.0 ± 5.4 kg for women). Of these, two were withdrawn from the study before receiving study treatment and three were withdrawn after receiving the first dose of r-hFSH FbM because they did not fulfil the down- regulation criteria. Pharmacokinetic data were based on the 39 volunteers (18 men, 21 women) who received both doses of r-hFSH and the three who received only one treatment; hence, this analysis included a total of 81 pharmacokinetic profiles from 42 volunteers. The safety analysis was based on the 42 volunteers who received at least one dose of r-hFSH. Pharmacokinetics of r-hFSH The median serum concentration–time curves after administration of freeze-dried and liquid r-hFSH were similar (Figure 1), as were values of C max , T max and AUC last Figure 1. Median concentration–time profiles after administration of freeze-dried and liquid r-hFSH FbM, 300 IU C ur r M ed R es O pi n D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y D ic le U ni v. o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. 124 Bioequivalence of liquid and freeze-dried r-hFSH © 2005 LIBRAPHARM LTD – Curr Med Res Opin 2005; 21(1) (Table 1). T max was significantly ( p = 0.018) higher with the liquid formulation than with the freeze-dried formulation. ANCOVA showed a significant relationship between baseline FSH concentrations and pharmaco- kinetic parameters, and hence a correction for baseline FSH was applied. Following this correction, there was a significant period effect for AUC last , in that values were higher during the second treatment period than during the first ( p = 0.010). Analysis of the ratios of the treatment means for C max and AUC last showed that the 90% confidence limits for these ratios fell within the range 0.8–1.25 (Table 2). Thus, the two formulations could be considered to be bioequivalent. This finding was obtained irrespective of whether an adjustment for baseline FSH concentrations was used. The residual (intra-volunteer) variability was approximately 10% or less (Table 2). Since the study power calculation was based on the assumption that this variability would be about 20%, this allowed precise estimates of the treatment mean ratios. Safety Both formulations were well tolerated. In total, 39 adverse events were reported during the study. Of these, eight were reported after administration of goserelin and 31 developed after administration of r-hFSH. The incidence of adverse events was similar with both formu- lations [freeze-dried: 32.5% (15 events in 13 volunteers); liquid: 26.8% (16 events in 11 volunteers)]. All adverse events were transient and mild or moderate in severity. The only adverse events occurring in 5% or more of volunteers after either treatment were headache, which occurred in seven volunteers (17.5%) after administration of the freeze-dried formulation and in four volunteers (9.8%) after administration of the liquid formulation, and upper respiratory tract infection in two volunteers (5%) after administration of the freeze-dried formulation. There were no clinically significant changes in vital signs or laboratory tests during the study. Local tolerability was good with both formulations. Overall, less than 2% of injections were associated with itching, swelling or bruising, and there were no significant differences between the two formulations. Redness at the injection site occurred after 60% of injections with the freeze-dried formulation and after 46% of injections with the liquid formulation; this was generally mild and transient, occurring within the first 2 h after dosing. Only three cases were reported as moderate, none was painful. Discussion This study has shown bioequivalence of the freeze-dried and liquid formulations of r-hFSH FbM, with the 90% confidence intervals for the ratios of the treatment means of C max and AUC last lying within the pre-defined limits for equivalence (0.8–1.25). Furthermore, both formulations showed good systemic and local tolerability, C max (IU/L) T max (h) AUC last (IU.h/L) Freeze-dried N 40 40 40 Mean 9.5 15.8 844 SD 2.3 8.2 181 Median 9.4 12.0 840 Minimum 4.0 4.0 462 Maximum 15.1 48.0 1170 Geometric mean 9.2 14.1 824 Geometric mean coefficient of variation (CV) 26.1% 51.3% 22.8% Liquid N 41 41 41 Mean 9.0 18.8 841 SD 2.4 10.6 279 Median 8.9 15.0 808 Minimum 4.7 6.0 557 Maximum 16.7 48.0 2280 Geometric mean 8.7 16.4 811 Geometric mean CV 26.5% 56.2% 26.0% C max , peak serum follicle-stimulating hormone concentration T max , time to peak concentration AUC last , area under the concentration–time curve from zero to the last measurable concentration Table 1. Pharmacokinetic parameters after administration of freeze-dried and liquid recombinant human follicle- stimulating hormone FbM, 300 IU C ur r M ed R es O pi n D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y D ic le U ni v. o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. © 2005 LIBRAPHARM LTD – Curr Med Res Opin 2005; 21(1) Bioequivalence of liquid and freeze-dried r-hFSH Lugan et al. 125 with no significant differences in tolerability between the two. The overall safety data may show a trend in favour of the liquid formulation for some measures; for example, occurrence of redness at the injection site and incidence of adverse events. The variability in C max and AUC last was low in this study. As shown in Table 1, the coefficients of variation for these parameters were approximately 25%, while the residual, intra-individual variability was approximately 9% (Table 2). Since the calculation of study power was based on the assumption that this residual variability would be approximately 20%, this resulted in very precise measurements of treatment ratios and the detection of small differences in the rate and extent of absorption (approximately 8% and 5%, respectively). In contrast, T max showed greater variability and values differed significantly between the two formulations. This may reflect the diverse mechanisms by which r-hFSH enters the circulation after subcutaneous injection. In some individuals, diffusion from subcutaneous tissue to adjacent small blood vessels is the predominant route of absorption, resulting in a short T max . In contrast, in other individuals, r-hFSH enters the circulation primarily via the lymphatics, a process with a longer T max . Alternatively, it could reflect differences in the two formulations that do not affect the bioequivalence. This study showed a significant period effect in that AUC last was higher during the second treatment period than during the first. This probably reflects minor fluctuations resulting from incomplete suppression of endogenous FSH secretion as can be expected after a single goserelin injection. However, the finding of bioequivalence between the two formulations was seen irrespective of whether the pharmacokinetic parameters were corrected for baseline FSH concentrations. Meta-analysis of clinical trials has shown that r-hFSH provides a greater likelihood of achieving pregnancy in patients undergoing in vitro fertilisation techniques, compared with preparations derived from human urine6, and that the dose of r-hFSH required for ovarian stimulation in such patients is lower than with urine- derived FSH6,7. Moreover, the recent development and introduction of r-hFSH FbM reflects the achievement of a high level of product consistency. The bioequivalence between the freeze-dried and liquid r-hFSH FbM observed in the present study suggests that the liquid formulation may be considered pharmaceutically equiv- alent to the freeze-dried product and is expected to be equivalent in terms of efficacy, consistency and tolerability. References 1. Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Annu Rev Biochem 1981;50:465-95 2. Follitropin. In: Dollery C, editor. Therapeutic drugs, 2nd ed. Edinburgh: Churchill Livingstone; 1999: F146-9 3. Lunenfeld B. Development of gonadotrophins for clinical use. Reprod Biomed Online 2002;4(Suppl 1):11-7 4. Marzo A, Balant LP. Bioequivalence. An updated reappraisal addressed to applications of interchangeable multi-source pharmaceutical products. Arzneimittelforschung 1995;45:109-15 5. Pabst G, Jaeger H. Review of methods and criteria for the evaluation of bioequivalence studies. Eur J Clin Pharmacol 1990;38:5-10 6. Daya S. Updated meta-analysis of recombinant follicle-stimulating hormone (FSH) versus urinary FSH for ovarian stimulation in assisted reproduction. Fertil Steril 2002;77:711-4 7. Ravhon A, Lavery S, Aurell R, Trew G, Margara R, Winston R. Clinical experience with recombinant follicle-stimulating hormone (FSH) and urinary FSH: a retrospective case-controlled analysis. Fertil Steril 2001;75:920-5 8. Driebergen R, Baer G. Quantification of follicle stimulating hormone (follitropin alfa): is in vivo bioassay still relevant in the recombinant age? Curr Med Res Opin 2003;19:41-6 Estimated ratio (liquid: freeze-dried) 90% confidence interval Intra-volunteer coefficient of variation C max 0.918 (0.886, 0.951) 9.21% AUC last 0.951 (0.922, 0.981) 8.03% C max , peak serum follicle-stimulating hormone concentrations AUC last , area under the concentration–time curve from zero to the last measurable concentration Table 2. Treatment mean ratios for C max and AUC last , corrected for baseline serum follicle-stimulating hormone concentration, with associated confidence intervals CrossRef links are available in the online published version of this paper: http://www.cmrojournal.com Paper CMRO-2836_3, Accepted for publication: 02 December 2004 Published Online: 06 January 2005 doi: 10.1185/030079904X18027 C ur r M ed R es O pi n D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y D ic le U ni v. o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y.