Pharmacokinetics, adverse effects and tolerability of a novel analogue of human pancreatic polypeptide, PP 1420

Authors


  • T.T and B.C.T.F. contributed equally to this study.

Professor Stephen R. Bloom, Division of Diabetes, Endocrinology and Metabolism, Imperial College London, 6th Floor, Commonwealth Building, London W12 0HS, UK. Tel.: +44 20 8383 8242, Fax: +44 20 8383 8142. E-mail: s.bloom@imperial.ac.uk

Abstract

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT

• Human pancreatic polypeptide (hPP) is a natural pancreatic hormone that suppresses appetite after meals.

• When exogenous hPP is given to healthy human volunteers, it causes a reduction in food intake.

• PP 1420 is a longer acting analogue of hPP that has been developed as a novel treatment for obesity.

WHAT THIS STUDY ADDS

• PP 1420 has been shown to be well tolerated in healthy human volunteers.

• PP 1420 has been shown to have an extended terminal elimination half-life compared with hPP.

AIMS The objectives of this phase 1 study were to confirm the tolerability of single ascending subcutaneous doses of PP 1420 in healthy subjects, to assess its adverse effects and to investigate the drug's pharmacokinetics and dose proportionality.

METHODS This was a double-blind, placebo-controlled, randomized study. There were three dosing periods. Each subject (n= 12) was randomized to receive one dose of placebo and two ascending doses of PP 1420, given as a subcutaneous injection. Blood samples were taken over 24 h to assess pharmacokinetics. Standard safety and laboratory data were collected. The primary endpoint was the tolerability of PP 1420. The secondary endpoint was exposure to PP 1420 as assessed by Cmax and AUC(0,∞).

RESULTS PP 1420 was well tolerated by all subjects with no serious adverse effects. Following single subcutaneous doses of PP 1420 at 2, 4 and 8 mg to male subjects, Cmax was reached at a median tmax of approximately 1 h post dose (range 0.32–2.00 h). Thereafter, plasma concentrations of PP 1420 declined with geometric mean apparent terminal elimination t1/2 ranging from 2.42–2.61 h (range 1.64–3.95 h) across all dose levels.

CONCLUSIONS Subcutaneous PP 1420 was well tolerated in healthy human subjects at single doses between 2–8 mg, with no tolerability issues arising. Where observed, adverse events were not serious, and there was no evidence of a dose-relationship to frequency of adverse events. The results therefore support the conduct of clinical trials to investigate efficacy, tolerability and pharmacokinetics during repeated dosing.

Introduction

Obesity is a growing global epidemic. The Health Survey for England and Wales reports that 24% of adults are obese and a further 41% of men and 32% of women are classified as overweight [1]. The WHO forecasts that by 2015, 4 billion adults will be overweight and over 700 million will be obese [2]. The serious health consequences of obesity are well documented and include type 2 diabetes, hyperlipidaemia and hypertension, all of which are independent risk factors for cardiovascular disease [3–6]. Furthermore, obesity confers an increased risk of arthritis, sleep apnoea, infertility and cancer [7].

Surgical treatments for obesity are currently the most effective means of achieving significant and sustained weight loss, and improving morbidity and mortality [8]. These procedures, however, are reported to confer a peri-operative risk of mortality of 0.5% [9]. Non-surgical treatments for obesity have included the medications orlistat, sibutramine and rimonabant. Sibutramine has recently been withdrawn from the European market, as a recent study has suggested an increased incidence of cardiovascular events [10]. Similarly, rimonabant was withdrawn due to an associated increase in anxiety and depression [11]. Orlistat is, in fact, the only drug still on the market that is specifically indicated for obesity. It causes modest weight loss and its use is limited due to poor tolerance [12]. Therefore it is clear that new pharmacological strategies are urgently needed to tackle obesity.

The gut and pancreatic hormones involved in appetite regulation represent an emergent drug development target for treatments that aim to cause effective weight loss with minimal side effects. Human pancreatic polypeptide (hPP) is a 36 amino acid, C-terminally amidated peptide, secreted by the PP cells of the islets of Langerhans [13]. Food intake stimulates hPP secretion [14], proportional to the size of the meal eaten [15]. hPP binds with highest affinity to the neuropeptide Y4 receptor [16] in the CNS, particularly in the hypothalamus and area postrema [17]. hPP acts as a feedback satiety hormone, suppressing appetite post-prandially. Exogenous intravenous administration of hPP to healthy human volunteers for 90 min acutely inhibits food intake and reduces overall food intake over 24 h by 25% [18, 19]. Therefore hPP is an attractive candidate treatment for obesity.

The half-life in circulation of hPP is short at 7 min [20] and previous studies on its physiological effects have overcome this problem by using prolonged, continuous intravenous infusions. As with other peptide-based hormones such as insulin, the simplest and most practical method of hPP administration would be subcutaneous injection. However, subcutaneous hPP would not be expected to deliver sustained concentrations for long. To overcome the short half-life problem, we have developed peptidase-resistant analogues of hPP to possess the following properties: longer half-lives, selective Y4 receptor binding activity and powerful appetite-suppressive properties in validated pre-clinical models of obesity (unpublished data). One of these, PP 1420, has an amino acid sequence that is similar to that of hPP, with one additional amino acid compared with the hPP sequence (a glycine residue located at position 0) and substitutions of five other residues of PP within the 37-residue peptide. The peptide is C-terminally amidated and contains only standard L-amino acids. The chemical name for PP 1420 is L-glycyl-L-alanyl-L-prolyl-L-leucyl-L-glutamyl-L-prolyl-L-valyl-L-tyrosyl-L-prolyl-L-glycyl-L-aspargyl-L-asparginyl-L-alanyl-L-threonyl-L-prolyl-L-glutamyl-L-glutaminyl-L-lysyl-L-alanyl-L-lysyl-L-tyrosyl-L-alanyl-L-alanyl-L-glutamyl-L-leucyl-L-arginyl-L-arginyl-L-tyrosyl-L-isoleucyl-L-aspargyl-L-arginyl-L-leucyl-L-threonyl-L-arginyl-L-prolyl-L-arginyl-L-tyrosinamide, hydrochloride salt.

The present study aimed to investigate the tolerability and pharmacokinetics of PP 1420 for the first time in healthy human volunteers in a phase 1 single ascending dose study.

Methods

Design of study

This was a single centre study performed in the UK at the Sir John McMichael Clinical Investigation Unit, Hammersmith Hospital, London, UK. It was a randomized, placebo-controlled, double-blind study in healthy male subjects.

Subjects

It was planned that 12 subjects would be recruited with two reserves during each dosing period. Inclusion criteria were healthy adult males aged between 18 and 50 years with body mass index 18 to 35 kg m−2 (inclusive) and body weight ≥70 kg. Exclusion criteria included a positive pre-study drug/alcohol screen, positive hepatitis B surface antigen or positive hepatitis C antibody result within 3 months of screening, a positive test for human immunodeficiency virus (HIV) antibody, history of migraine, history of or evidence of abnormal eating behaviour, as observed through the Dutch Eating Behaviour [21] and SCOFF questionnaires [22], history of excessive alcohol consumption within 6 months of the study, urinary cotinine concentrations indicative of smoking or history or regular use of tobacco- or nicotine-containing products within 6 months prior to screening, an electrocardiographic corrected QT interval at screening >450 ms, systolic blood pressure outside the range 85–160 mmHg, diastolic blood pressure outside the range 45–100 mmHg and/or heart rate outside the range 40–110 beats min–1. The use of any medicine that the investigator considered could interfere with the trial results was not allowed.

All subjects gave their written consent prior to any trial-related procedures. The study was conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice. The study protocol and informed consent information were approved by the Outer West London Research Ethics Committee (reference number 10/H0709/10). Clinical Trial Authorization was obtained from the Medicines and Healthcare products Regulatory Agency, UK (EudraCT number 2009–017522-39). The trial was registered with clinicaltrials.gov, number NCT01052493.

PP 1420 administration and sample collection

PP 1420 active pharmaceutical ingredient was synthesized to current International Committee on Harmonization Q7 Good Manufacturing Practice standards by PolyPeptide Laboratories (Hillerød, Denmark) and the Clinical Trial Material was manufactured by Nextpharma (Braine l'Alleud, Belgium). Coded syringes containing PP 1420 or placebo were prepared by Imperial College Healthcare NHS Trust Pharmacy Department. The doses of PP 1420 were 2, 4 or 8 mg given as subcutaneous injection into the anterior abdominal wall. Placebo was 0.9% saline. The starting dose, exposure escalation strategy and stopping exposures in this study were based on predicted safety margins utilizing no observed adverse effect limits observed from non-clinical toxicology studies. The dose range investigated was informed by previous studies in healthy volunteers of the physiological role of hPP [18, 19] and by the knowledge that at daily doses above 8 mg, production costs might render PP 1420 commercially unviable.

Subjects were admitted to the clinical research unit the evening prior to dosing (day −1). They received their first single dose of study medication the next morning (day 1) and were discharged from the unit on day 2 after all 24 h assessments had been completed, following satisfactory review by the investigators. Subjects were required to fast from 22.00 h on day −1 until 2 h after study drug administration. Blood samples were obtained for pharmacokinetics at the following times: pre dose, 0, 15, 30, 45, 60, 90, 120, 150, 180, 210, 240, 360, 720 min and 24 h. Heart rate, blood pressure, temperature, 12-lead ECG, physical examination and blood and urine sampling for standard laboratory assessments were performed during all three dosing periods. Subjects returned for a follow-up visit 7 to 10 days after each dosing period. The duration of each subject's participation in the study from screening to the follow-up visit was approximately 7–12 weeks.

There were three dosing periods. For each dosing period, four volunteers received placebo and eight PP 1420. Each subject was randomized to receive one dose of placebo and two ascending doses of PP 1420, each dose being spaced at least 2 weeks apart. A sentinel dosing strategy was used at the start of each dosing period. On the first day of each dosing period, two subjects were dosed, one receiving active PP 1420 and the other receiving placebo. The remaining subjects were dosed as soon as possible thereafter, depending on scheduling in the clinical unit.

Dose adjustment and stopping criteria

The Investigators and Independent Data Monitoring Committee (IDMC) separately reviewed safety and tolerability data from all subjects, and preliminary pharmacokinetic data from at least four subjects, prior to dose escalation. The Investigators assessed adverse events (AEs) in the blinded state. Standard pharmacokinetic parameters were derived for all available samples and the pharmacokinetic profile was analyzed for data up to and including the 24 h post dose time point. Whilst it was expected that a linear relationship would exist between dose and drug exposure, if it had become apparent that this relationship was non-linear, the dose escalation would have been modified according to the advice of the study pharmacokineticist. The doses could also have been adjusted on the basis of safety and tolerability data from previous doses. The maximum dose that was given in this study was 8 mg.

Dose escalation in this study would have been stopped if three or more subjects experienced dose limiting, drug-related AEs or if the pattern of AEs observed in a group had been consistent across subjects, poorly tolerable and clinically significant. Dose escalation in this study would also have been stopped if three or more subjects developed significantly high titres of antibodies to human sequence PP or PP 1420 analogue.

Analysis of PP 1420

Concentrations of PP 1420 were determined in plasma samples using a validated analytical methodology by Quotient Bioresearch (Fordham, UK). This utilizes a liquid chromatography/ tandem mass spectrometric method, with an internal PP 1420 standard incorporating 13C and 15N atoms to distinguish the standard by mass. Chromatographic separation is by uHPLC system using a reverse phase analytical column (100 mm × 2.1 mm, 1.7 µm i.d., C18) with a gradient elution using 0.2% (v/v) formic acid in acetonitrile and water. Samples are analyzed using a Sciex API5000 triple quadrupole mass spectrometer using electrospray ionization. Single specific MRM transitions are monitored for both PP 1420 and internal PP 1420 standard. The validated analytical range of this assay is 0.5–250 ng ml−1 with 0.5 ng ml−1 being the lowest limit of quantification. The precision, as measured by coefficient of variation (%CV), at 0.5 ng ml−1, 25 ng ml−1, 100 ng ml−1 and 250 ng ml−1 is 5.8%, 3.6%, 5.7% and 3.5%, respectively. The intra-assay accuracy, as measured by relative error (%RE), is ≤±12.0% and the inter-assay accuracy is ≤± 5.0%.

Endpoints and statistical methods

The pharmacokinetic endpoints, as listed below, were estimated for each subject using a fully validated version of WinNonlin Pro (Version 5.2.1 – Pharsight Products, Mountain View, CA, USA). The following parameters were derived, where appropriate, from the individual plasma concentration vs. time profiles after a single dose: Cmax, tmax, AUC(0,tlast) (the area under the concentration vs. time curve from time zero to the last sampling time, calculated by the linear trapezoidal rule), λz (the apparent terminal rate constant), terminal elimination t1/2 (calculated from loge 2/λz), AUC(0,∞) (the area under the concentration vs. time curve estimated from time zero to infinity). Actual sampling times were used for the pharmacokinetic analysis. Plasma concentrations below the limit of quantification of the assay (BLQ) were taken as zero for calculation of concentration summary statistics and all pharmacokinetic parameters.

A non-linear power model was used to assess dose-proportionality [23]. The proportional relationship between each parameter and dose is written as a power function:

image(1)

where a is a constant, b is the proportionality constant and y is the parameter of interest (AUC(0,∞) or Cmax). The exponent, b, was estimated by performing a linear regression of the logged parameter on log dose. The exponent, b, is the estimated slope of the resulting regression line since taking logs of equation (1) gives the linear relationship, log y = log a + b × log dose. The relationship is dose-proportional when b = 1. The exponents and 95% confidence intervals (CIs), blower (bl) and bupper (bu), were estimated. There would have been evidence of non dose-proportionality if this CI excluded 1. The estimate of the fold increase in exposure for a doubling in dose (with 95% CI) was also calculated. The increase in exposure expected for a doubling in dose was calculated as 2b (95% CI 2bl, 2bu).

The assumption of a linear relationship between loge transformed parameter and loge dose was tested by fitting an analysis of variance (anova) model and portioning the sum of squares for number of treatments (number of treatments −1 degree of freedom [d.f.]) into those for linearity (1d.f.) and departures from linearity or lack of fit (number of treatments −2 d.f.).

The P values from this test would indicate significant lack of fit of the power model if <0.05, and would only be reported if there was evidence of lack of fit.

Since each subject received two of the three dose levels, the assessment of dose proportionality required the covariate ‘subject’ to be included as a random effect in the power model. Subject was included as a random effect using the Linear Mixed Effects Wizard in WinNonlin Pro Version 5.2.1.

Safety endpoints were AEs, laboratory parameters (haematology, biochemistry, urinalysis), physical examination, vital signs and ECG. Subjective feelings of hunger and nausea were measured using 100 mm visual analogue scales (VAS) [24].

Results

Study groups

A total of 33 subjects were screened for the study of which 13 were recruited. These 13 subjects were randomized and exposed to PP 1420. One subject was replaced after the first dosing period for reasons unconnected to the study. This subject's evaluable data were included in the safety and PK analyses. The baseline demographic characteristics of the study group were as follows (mean ± SD): BMI 28.1 ± 3.3 kg m−2, weight 87.8 ± 9.2 kg, height 1.80 ± 0.1 m, age 34.0 ± 8.9 years.

PP 1420 pharmacokinetics

The mean concentration–time profiles of PP 1420 are presented in Figure 1. The pharmacokinetic variables calculated from the concentration–time profiles are presented in Table 1. Following single subcutaneous doses of PP 1420 at 2, 4 and 8 mg to male subjects, the median tmax was at approximately 1 h post dose (range of individual values 0.32–2.00 h). Thereafter, plasma concentrations of PP 1420 declined mono-exponentially with geometric mean apparent terminal elimination t1/2 ranging from 2.42–2.61 h (range of individual values 1.64–3.95 h) across all dose levels.

Figure 1.

Mean plasma concentrations of PP 1420 following single, subcutaneous injections of PP 1420 at 2 mg (filled circles), 4 mg (filled squares) and 8 mg (filled triangles). Plotted on a log scale (y axis). Error bars indicate SD

Table 1.  Pharmacokinetic parameters following single subcutaneous doses of PP 1420 to healthy male subjects
Dose (mg)ParameterCmax (ng ml−1)tmax (h)AUC(0,tlast) (ng ml−1 h)AUC(0,∞) (ng ml−1 h)t1/2 (h)
  1. Geometric mean, range and percentage coefficient of variation (%CV) are presented, with the exception of tmax which is presented as median (*). NC = not calculated.

2Geometric mean26.30.875*93.61012.42
Range15.9–39.10.32–2.0049.7–12667.6–1311.64–3.38
%CV28.6NC31.322.722.0
4Geometric mean55.11.00*2292412.49
Range37.5–74.20.75–1.50107–316109–3561.69–3.60
%CV21.5NC35.637.929.0
8Geometric mean95.71.00*4034182.61
Range74.2–1260.50–1.50205–589210–6002.13–3.95
%CV15.1NC37.538.520.8

Following single dosing of 2, 4 and 8 mg PP 1420, systemic exposure (Cmax and AUC(0,∞)) to PP 1420 increased with increasing doses and the increase was dose proportional. The relationship of AUC(0,∞) to increasing doses is shown in Figure 2 and for Cmax is shown in Figure 3. For a doubling in dose, systemic exposure to PP 1420 was predicted to increase, on average, 1.91 to 1.93-fold.

Figure 2.

Relationship between AUC(0,∞) values and dose of PP 1420 following single subcutaneous doses of 2, 4 and 8 mg PP 1420 to male subjects. The dotted line shows the dose-proportional relationship passing through the parameter at the 2 mg dose level. The solid line shows the power function obtained from the linear regression from the log parameter against log dose

Figure 3.

Relationship between Cmax values and dose of PP 1420 following single subcutaneous doses of 2, 4 and 8 mg PP 1420 to male subjects. The dotted line shows the dose-proportional relationship passing through the parameter at the 2 mg dose level. The solid line shows the power function obtained from the linear regression from the log parameter against log dose

Following single dosing of 2, 4 and 8 mg PP 1420, between-subject variability in systemic exposure (AUC(0,∞) and Cmax) to PP 1420 was low; CVs were 15.1–38.5%.

Tolerability of PP 1420

PP 1420 appeared to be well tolerated with no evidence of a dose-relationship for adverse events (AEs) and no serious AEs that were attributed to PP 1420. Five subjects receiving 4 mg of PP 1420 experienced an AE compared with three subjects receiving 8 mg and two subjects each receiving 2 mg and placebo. Four subjects did not experience any adverse event during treatment. The most common AE experienced was headache. Inspection of the AEs showed no unexpected pattern in their nature (Table 2). Because nausea was anticipated to be a likely AE, this was monitored using a 100 mm VAS scale. There was no statistically significant alteration in the placebo-subtracted VAS from baseline for the sensation of nausea (‘how sick do you feel right now?’). There was no statistically significant alteration in the placebo-subtracted VAS from baseline for the sensation of hunger (Figure 4).

Table 2.  Summary of adverse events in PP 1420 study
EventPlaceboPP-1420 2 mgPP-1420 4 mgPP-1420 8 mg
  1. Empty cells indicate that there were no AEs.

Number of subjects with AE2253
Total number of AEs4256
Headache2121
Injection site reaction/bruising  12
Nausea11  
Vomiting1  1
Diarrhoea  1 
Abdominal pain/bloating   1
Infected finger   1
Cold sore  1 
Figure 4.

Visual analogue scores (VAS) for gastrointestinal tolerability parameters. VAS depicted as placebo-corrected mean (±SEM) change from baseline after subcutaneous injection of PP1420 at 2 mg (circles), 4 mg (triangles) and 8 mg (inverted triangles)

There were no effects of any concern in any vital signs, ECG or laboratory parameters. There was no evidence of an antibody response in any subject receiving PP 1420.

Discussion

This study constitutes the first clinical investigation of PP 1420, a Y4 receptor agonist being developed as a new treatment for obesity. As expected, the peptidase resistance evident in pre-clinical studies was reflected in a prolonged terminal elimination t1/2 compared with native hPP.

The current study was not powered to determine the duration of satiety that may be induced by exposure to PP 1420. Nevertheless, the pharmacokinetic profile of PP 1420 may suggest that a twice or three times daily dosing regimen would be required to sustain plasma PP 1420 concentrations over 24 h if this proves necessary for efficacy. Whilst once daily or less frequent dosing would be desirable, the precedent set by exenatide, a marketed peptide hormone analogue of glucagon-like peptide-1 that is delivered by subcutaneous injection, indicates that twice daily dosing can be acceptable to patients.

A potential advantage of targeting the Y4 receptor is that the satiety induced by a 90 min i.v. infusion of hPP may persist for up to 24 h even after plasma concentrations have returned to baseline [18]. Thus, a relatively brief exposure to a typical post-prandial concentration of hPP can reduce spontaneous food consumption several hours later. Several dosing strategies will thus have to be investigated in future trials to determine PP 1420's pharmacodynamics and hence the optimal dosing regimen.

There were no adverse events in the current study of any clinical concern, nor were there any differences between placebo and PP 1420 in any vital sign, ECG or laboratory parameters measurements. Notably, given that nausea was anticipated to be a likely adverse event, PP 1420 did not cause any significant alteration in the VAS scores for nausea.

Overall, the study demonstrated that administration of PP 1420 in healthy human subjects was well tolerated at each dose level, with no tolerability issues arising from either measured clinical endpoints or observed adverse events. The results thus support the conduct of further trials of PP 1420 in humans to explore its efficacy, tolerability and pharmacokinetics in multiple doses.

Competing Interests

The authors accept direct responsibility for this paper. CB and IW were remunerated for their consultancy on this project.

The Division is funded by grants from the MRC, BBSRC, NIHR, an Integrative Mammalian Biology (IMB) Capacity Building Award, an FP7-HEALTH-2009–241592 EuroCHIP grant and is supported by the NIHR Imperial Biomedical Research Centre Funding Scheme. This study was supported by the Wellcome Trust Seeding Drug Discovery Initiative (081627). The following authors are also funded as follows: BCTF by a NIHR Clinical Lectureship, JSM by a MRC DPFS grant, JCS by a EuroCHIP grant, ESC by the BBSRC DRINC programme and SZV by a Wellcome Trust Clinical Research Training Fellowship and NIHR Clinical Lectureship. We would like to acknowledge the following for their assistance and support in completing the trial: (i) the staff of the Sir John McMichael Centre, Imperial College Healthcare NHS Trust, especially Ms Lili Lheureux and Ms Gaia Mahalingam, (ii) Prof Martin Wilkins, Prof Karim Meeran and Prof Waljit Dhillo as members of the IDMC, (iii) Ms Sheila Schwartz for project management and (iv) the clinical trial volunteers.

Ancillary