|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Ferring Pharmaceuticals A/S,5 International PharmaScience Center, Department of Non-Clinical Development, Copenhagen S, 2300 Denmark
Charles River Laboratories Preclinical Services,6 Sparks, Nevada 89431
ABSTRACT
Preterm labor (PTL) affects up to 25% of human pregnancies in developing countries, but there are few therapeutic options. Based on the key role of oxytocin (OXT) in labor and parturition, OXT antagonists are a potentially useful class of drugs for PTL. Barusiban is a new selective, potent, and long-acting OXT receptor antagonist. In this study barusiban was given by continuous i.v. infusion to monkeys during the last 3 wk of pregnancy; the monkeys were also given daily doses of OXT to induce uterine contractions and simulate PTL. Barusiban effectively suppressed OXT-induced PTL-like contractions and prevented early delivery. In contrast, fenoterol (a beta2-adrenoceptor [beta2-AR] agonist used as a comparative control) did not inhibit uterine contractions in this model. Barusiban was particularly effective in maintaining low intrauterine pressure (IUP) near the end of pregnancy, which is when IUP in both OXT controls and fenoterol-treated females increased substantially. Although barusiban delayed the onset of labor, it did not prevent normal delivery. These data demonstrate the safety and efficacy of barusiban in reducing uterine contractility in response to repeated OXT challenge, and suggest that barusiban may be therapeutically effective in long-term treatment of PTL.
neuropeptides, parturition, pregnancy, oxytocin, uterus
INTRODUCTION
Preterm labor (PTL) is a leading cause of premature delivery and associated neonatal morbidity and mortality, and represents a significant unmet medical need [1]. Estimates of preterm birth range from 5%12% in developed countries including the United States and up to 25% in developing countries [13]. The rate of preterm birth varies among ethnic groups, being highest in black women [1, 2]. Overall the rate of preterm births has increased 27% over the past 2 decades [4]. Most preterm infants survive, many after a long stay in a neonatal intensive care unit, but they are at risk for disabilities such as mental retardation, cerebral palsy, lung and gastrointestinal problems, and vision and hearing loss [4]. Treatment of complications associated with PTL has significant economic and social impact [57]. Therapeutic options to treat PTL are limited, with existing drug therapy (tocolytics) compromised by side effects and limited efficacy. Most drugs for PTL (including terbutaline, ritodrine, nifedipine, rofecoxib, indomethacin, and magnesium sulfate) are used off-label. To our knowledge, at the present time only two drugs have been approved specifically for PTL: fenoterol, a ß2-adrenoceptor (ß2-AR) agonist, and atosiban, a mixed antagonist of vasopressin (V1a) and oxytocin (OXT) receptors (OTR) [8].
Beta adrenoceptor agonists (betamimetics) in general and fenoterol specifically have been studied for treatment of preterm labor [9, 10]. The results indicate that beta adrenoceptor agonists can delay delivery by 2448 h but have side effects including pulmonary edema, cardiac arrhythmias, and hypokalemia. Also, prolonged treatment with beta adrenoceptor agonists is associated with loss of efficacy related to desensitization of the ß-adrenergic receptor system [11].
Endogenously generated OXT is thought to play an important role in the initiation of labor and in parturition [12, 13]. Both endocrine and paracrine sources of OXT are involved, emanating from the hypothalamic-pituitary axis and the uterine/fetal compartment, respectively [14]. Levels of OTR mRNA in the uterine myometrium increase substantially around the onset of labor [15]. In nonhuman primates and in pregnant women, a circadian rhythm of uterine activity in late gestation appears to be driven by maternal OXT [1618]. The enzyme oxytocinase (cystine aminopeptidase) degrades OXT in humans and some species of nonhuman primates [19]. Oxytocinase increases with normal gestation but decreases in patients with PTL, consistent with a role for OXT in exacerbating uterine contractions in PTL [20]. Collectively, these data support a role for OXT and its receptor in labor, both at term and preterm [21].
OTR antagonists successfully inhibit OXT-induced uterine contractions in isolated human myometrium [8, 22] and in a primate model of PTL [23]. Thus, OTR antagonists are potentially useful therapeutic agents to delay or prevent PTL and delivery.
Atosiban, which is a mixed antagonist at V1a and OXT receptors [22], can be used short term to treat PTL and delay imminent preterm birth [24]. This therapy is typically of short duration (48 h), but this time window provides a chance to reduce respiratory distress syndrome by administration of antenatal glucocorticoids and allows in utero transfer to a neonatal intensive care unit. However, maintenance treatment of PTL over a longer time period still represents an unmet clinical need to reduce neonatal morbidity and mortality.
Barusiban is a new long-acting OTR antagonist that is being developed for PTL and is currently undergoing clinical trials. Barusiban's longer duration of action and higher potency compared with atosiban have been demonstrated in a nonhuman primate model of PTL, suggesting that is has pharmacokinetic (PK) and pharmacodynamic (PD) advantages for long-term treatment of PTL [23]. In contrast with atosiban, barusiban has a higher affinity for the human cloned OTRapproximately 300-fold greater than its affinity for the V1a receptor [22]. In studies of contractile patterns in isolated human myometrial tissue, barusiban also demonstrates more selective inhibitory effects than atosiban [8].
The goal of the present study was to investigate the feasibility of long-term treatment of PTL with barusiban, including in comparison with fenoterol (a ß2-AR agonist), by determining PK and PD after continuous infusion for up to 3 wk in a nonhuman primate model of PTL.
MATERIALS AND METHODS
Surgical Procedures
Eight pregnant female cynomolgus monkeys were assigned to the study, each of which was experimentally non-naive, aged 4.06.1 yr, and weighed 3.45.5 kg at study start. Surgical procedures for implantation of telemetry units (to measure intrauterine pressure [IUP]) and infusion catheters (for dosing and blood sampling) were as described previously [25]. Surgery was performed during the third trimester of each animal's pregnancy (Gestational Day [GD] 120 ± 3). Heparinized saline was infused continuously to maintain bidirectional catheter patency until the start of dosing. Remote techniques for dosing and blood sampling were used to avoid disturbance of the animals and to reduce iatrogenic influence on uterine contractions. The overall study plan was reviewed and approved by Charles River-Nevada Animal Care and Use Committee, and was conducted in accordance with published guidelines for laboratory animal care and use [26].
Treatments
Each treatment group consisted of two to three pregnant females. This number of animals per group was previously shown to be adequate for the detection of treatment-related effects [23]. It was also considered appropriate from both a technical and animal welfare perspective, considering the required surgical procedures. Dosing by i.v. infusion was accomplished from a room adjacent to the animal room, using clinical-grade Baxter Auto Syringe AS50 infusion pumps and a swivel-tethering system.
Pregnant females were dosed from approximately GD 141 to GD 163 (or delivery, whichever came first). Barusiban, fenoterol, or saline (control animals) were administered by continuous i.v. infusion for 24 h per day. The dose level for barusiban (150 µg kg1 h1) was determined based on maximal effects observed in a previous single-dose study [23]. The dose level for fenoterol (3 µg kg1 h1) was based on the maximally effective dose cited in the manufacturer's (Boehringer Ingelheim) package insert and adjusted for body weight of the monkeys. Fenoterol is a ß2-AR agonist approved for treatment of PTL, and was used in this study as a comparative control for inhibition of OXT-induced contractions.
Oxytocin was given daily (i.v., 36 h per day) to simulate PTL with daily periods of spontaneous contractions. Daily OXT infusions were customized to stimulate submaximal uterine activity in all animals (i.e., to achieve contractions at an amplitude of 3050 mmHg every 26 min). Rates of infusion were customized for each animal to produce the desired effect, and ranged from 10100 mU kg1 h1. Once the individualized dose level of OXT was determined for each animal, it was maintained for the duration of the treatment period. Oxytocin infusions were given at approximately the same time of day throughout the study (14001700 h for 3-h infusions and 10001600 h for 6-h infusions). The duration of OXT infusion started at 3 h and was then increased to 6 h for animals in all groups, based on controls that did not give birth after several days of 3-h infusions.
Measurements
Overall health of the animals and any adverse effects of the test articles or implanted devices were monitored by recording clinical signs and food consumption (daily) and body weight (weekly). Intrauterine pressure was monitored continuously by telemetry. Telemetry data output was monitored remotely from a room adjacent to the animal room. Data generated during the recording periods were computed by data acquisition software (Data Sciences International), recorded, and plotted graphically. Telemetric recordings of IUP data were further evaluated by integration as area under the curve (AUC) for half-hour intervals, using ChromPerfect Spirit software (Justice Laboratory Software, Denville, NJ).
Blood samples were collected every other day from barusiban-treated animals, approximately 2.75 h after the start of OXT infusion. Plasma levels of barusiban were determined by liquid chromatography with tandem mass spectrometry (LC-MS/MS).
At the end of the study, animals were returned to the laboratory's animal colony.
Analysis of Data and Statistics
Group means and SEM were calculated as follows: IUP values were averaged for 0.5 h intervals (AUC30) for all animals in each treatment group in order to present a representative day of the study (Fig. 2). The number of animals available at each interval varied depending on the duration of OXT infusion (3 h or 6 h) and the number of post-OXT evaluations. These AUC30 IUP values were then combined into mean daily IUP values for each group, to demonstrate a mean IUP progression over time (Fig. 3). Daily IUP values were then combined to calculate the mean overall IUP for each group over the entire duration of the study (Fig. 4).
|
|
|
Graphical presentation of the data was conducted using GraphPad Prism version 4.03 for Windows (GraphPad Software, San Diego, CA). Curve fitting for PK data was conducted in Microsoft Excel 2003 using a logarithmic fit. Statistical analyses of IUP data were performed with the SAS System, Version 8.1. Significant intergroup differences were evaluated by single-factor ANOVA, with animal grouping as the factor, utilizing a P
0.05 level of significance. If the parametric ANOVA was significant at P
0.05, Tukey test was used to identify statistically significant differences among the three groups (animals treated with barusiban, fenoterol, or OXT control). Statistical analysis of gestational data in Table 1 was performed with a Student t-test, utilizing P
0.05.
|
RESULTS
Both barusiban and fenoterol were well tolerated and no signs of toxicity were present in the study. Plasma levels of barusiban were relatively stable over the duration of the study, and generally ranged between 4 and 6 µg/ml (Fig. 1).
|
Representative IUP profiles on one study day (Day 7) for the OXT control, barusiban, and fenoterol groups are presented in Figure 2. In the controls, OXT infusion resulted in an increase in mean IUP values from 1.6 ± 0.5 to 7.1 ± 0.6 kmmHg·h (mean ± SEM). After OXT infusion, IUP levels declined toward pre-OXT levels, generally within 1 h. Barusiban was highly effective in inhibiting OXT-induced increases in IUP. In the barusiban group, mean IUP values before and after OXT were 5.0 ± 1.2 kmmHg·h and 4.5 ± 0.4 kmmHg·h, respectively. In the fenoterol group, mean IUP values increased from 2.3 ± 0.8 to 8.0 ± 1.0 kmmHg·h during OXT infusion.
Prior to daily challenge with OXT to simulate PTL, spontaneous uterine activity in barusiban- and fenoterol-treated females was indistinguishable from saline-infused controls (Fig. 3A). These data show that neither barusiban nor fenoterol were inhibitory to the point of uterine atony. Taken together with previous data that showed that there is no desensitization to OXT over time [20], these results confirm the usefulness of the nonhuman primate model for evaluating long-term inhibition of OXT-induced contractions.
Continuous i.v. infusion of the OTR antagonist barusiban (150 µg kg1 h1) in the last 3 wk of pregnancy was highly effective in suppressing OXT-induced increases in IUP (Fig. 3B). Compared with OXT controls and the comparative control fenoterol (ß2-AR agonist), IUP in barusiban-treated animals was generally reduced throughout the study. In the first 17 days of the study, during which all three groups were dosed, IUP in the barusiban group was statistically significantly lower than controls on 8 days, and lower than the fenoterol group on 11 days (analysis not presented on graph). Barusiban was particularly effective in maintaining low IUP near the end of gestation, when IUP in both the OXT controls and the fenoterol group increased markedly for 45 days prior to delivery. Only in those animals treated with barusiban did births occur after the planned end of OXT challenge and tocolytic dosing on GD 163; one barusiban-treated mother delivered on GD 165, another on GD 169.
When compared with pre-OXT values across the entire duration of the study, group-average IUP values during OXT infusion were significantly higher for each group (Fig. 4). However, group-average IUP during OXT infusion was only 1.3-fold higher than pre-OXT values in the barusiban group, compared with 4.1-fold in OXT controls and 2.9-fold in the fenoterol group. Intergroup comparison of data collected during OXT infusion revealed that group-average IUP was significantly lower in the barusiban group compared with the OXT control and fenoterol groups (Fig. 4). In contrast, IUP during OXT infusion in the fenoterol group was not significantly different than OXT controls. These data demonstrate the overall effectiveness of barusiban in suppressing OXT-induced increases in IUP.
Although not statistically significant, the time to delivery (i.e., number of days from the start of dosing until birth) and length of gestation appeared to be increased in barusiban-treated females compared with the OXT control and fenoterol groups (Table 1). Additionally, for all barusiban-treated females, delivery did not occur until 16 days after dosing with the antagonist was stopped (Table 1). Gestation length in females given barusiban (163 ± 4 days [mean ± SEM]) was within the normal range for cynomolgus monkeys in the testing facility (165 ± 10 days). The length of gestation in controls and fenoterol-treated females was shortened to 154 ± 3 days and 148160 days, respectively.
All infants born by barusiban-treated females were healthy and morphologically normal. One control infant and both fenoterol infants died at the time of delivery; two of these were breech births; these results did not have a definitive relationship to treatment, as background stillbirth rates of up to 39% have been reported in macaques [27].
DISCUSSION
The data from this study extend the results of previous single-dose studies with barusiban [23] and reveal the efficacy and safety of long-term treatment with barusiban in reducing uterine contractions in a nonhuman primate model of PTL. Barusiban was effective in suppressing OXT-stimulated contractions that mimicked PTL and in extending the duration of pregnancy, thereby resulting in the birth of healthy infants at the expected time of parturition. In contrast, the ß2-AR agonist fenoterol (Partusisten), which is approved for PTL, was not efficacious in this model. These data provide the first proof of principle that barusiban is a potentially useful new agent for long-term treatment of PTL.
Continuous infusion of barusiban in the nonhuman primate model used for this study resulted in barusiban plasma levels of approximately 46 µg/ml. These exposure levels were from 2.5 to 10 times higher than the Cmax (maximum plasma concentration) values associated with efficacy in previous single-dose studies [23]. Therefore, it is likely that continuously-infused barusiban will be efficacious at lower doses than those administered in the present study. In spite of the relatively high plasma levels of barusiban achieved by infusion, there were no adverse effects on clinical condition or body weight of the monkeys.
Another dimension to the safety of barusiban is that its OXT antagonism can be reversed by administration of high doses of OXT [23]. This is important because in the event that emergency procedures (Caesarean section) need to be used clinically to deliver a baby, effective control of postpartum uterine bleeding requires the ability to alleviate any existing OXT antagonism so that the uterus can be constricted through administration of OXT or other vasoconstrictors.
Fenoterol (Partusisten) was used as a comparative control agent in this study, chosen because beta adrenoceptor agonists are the most commonly used class of drug for clinical treatment of PTL. However, in the nonhuman primate model of the present study, fenoterol was not effective in suppressing OXT-induced increases in IUP. Time to delivery (days to birth after start of dosing) and length of gestation in fenoterol-treated female monkeys were unchanged compared with OXT controls, and were indicative of birth before the expected time of parturition for this species. The dose of fenoterol (3 µg kg1 h1) given to the monkeys was considered adequate since it was based on the maximally effective dose cited in the package insert for clinical dosing, adjusted for the body weight of the monkeys. Desensitization of ß2-AR-mediated responses due to long-term infusion of a relatively high fenoterol dose could have been responsible for its lack of efficacy [28, 29].
Compared with fenoterol, barusiban led to a significantly greater reduction in IUP over time and a longer time to delivery (i.e., greater number of days from the start of dosing until birth). Although not statistically significant, overall, the time to delivery (i.e., number of days from the start of dosing until birth) and length of gestation appeared to be increased in the barusiban group compared with the OXT control and fenoterol groups. Barusiban was particularly effective in maintaining low IUP near the end of gestation, when IUP in both the OXT controls and the fenoterol group increased markedly for 45 days prior to delivery.
The efficacy of barusiban described in this paper is consistent with its known mode of action as an OTR antagonist and with a variety of biological data on the role of OXT in the initiation and maintenance of preterm and term labor [12, 13]. OXT antagonists successfully inhibit oxytocin-induced uterine contractions in isolated human myometrium [8, 22]. Collectively these data support a role for OXT and its receptor in preterm and term labor [21]. Also, the inhibitory effects of barusiban on OXT-induced contractions and increases in IUP have been previously described in studies that used the same nonhuman primate model of PTL as employed in the present study [23, 25].
In the current study, barusiban was given by continuous intravenous infusion for the last 3 wk of gestation, and was effective in offsetting daily challenges with OXT (36 h per day). This suggests the exciting possibility that barusiban may be effective in a maintenance or preventative modality, in addition to being effective when administered after PTL has started [23]. Given the long half-life and duration of action of barusiban [23], long-lasting efficacy may well be achieved using periodic single doses, helping to ensure greater patient compliance with intended clinical use. Ultimately, development of a non-parenteral formulation would also be attractive to enable outpatient dosing in a clinical setting. Short-term treatment with barusiban is currently the subject of double-blind, placebo-controlled Phase II clinical trials.
In conclusion, data from this study indicate that barusiban is both safe and effective in reducing uterine activity in response to daily OXT challenge in a nonhuman primate model that mimics PTL. When administered by continuous infusion for the last 3 wk of pregnancy, barusiban prevented early birth induced by OXT and extended the length of pregnancy to normal duration. This is the first proof of principle that barusiban is effective for long-term management of PTL-like contractions. Since neonatal survival rates improve by 3% with each day that pregnancy can be extended [30], any new therapeutic for PTL that is effective in a long-term treatment paradigm will be a significant contribution to this unmet clinical need and will convey substantial socioeconomic benefit.
ACKNOWLEDGMENTS
We recognize William Baughman and the late Michael Cook (Oregon National Primate Research Center, Beaverton, Oregon), and Jan Bernal (Charles River Preclinical Services, Sparks, Nevada) for their guidance and expert surgical assistance with the animal model. Independent consultants Andrew Hendrickx, Pamela Peterson, and William Hobson assisted with analysis and interpretation of the IUP data, and Lotte Seiding Larsen (Ferring Pharmaceuticals A/S) provided the PK analyses.
FOOTNOTES
1 Presented in part at the 46th spring meeting of the German Society for Experimental and Clinical Pharmacology and Toxicology, 1517 March 2005, Mainz, Germany; and at the 52nd annual meeting of the Society for Gynecologic Investigation, 2326 March 2005, Los Angeles, CA. ![]()
2 Correspondence: G.J. Chellman, Developmental and Reproductive Toxicology Program, Charles River Laboratories Preclinical Services, 587 Dunn Circle, Sparks, NV 89431. FAX: 775 331 2289; gary.chellman{at}us.crl.com ![]()
3 Current address: MPI Research, Inc., Mattawan, MI 49071. ![]()
4 Current address: Scios, Inc., Fremont, CA 94555. ![]()
Received: 5 May 2006.
First decision: 26 May 2006.
Accepted: 10 August 2006.
REFERENCES
Related articles in Biol Reprod:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |