Drugs R D (2018) 18:149–159 https://doi.org/10.1007/s40268-018-0236-x ORIGINAL RESEARCH ARTICLE Pharmacokinetics of the Oral Selective CXCR2 Antagonist AZD5069: A Summary of Eight Phase I Studies in Healthy Volunteers 1 1 1 2 • • • • Marie Cullberg Cecilia Arfvidsson Bengt Larsson Anna Malmgren 3 1 1 • • Patrick Mitchell Ulrika Wa ¨ hlby Hamre ´n Heather Wray Published online: 31 May 2018 The Author(s) 2018 Abstract Intra- and inter-subject variability in AUC was 3–11 and Objective The aim of this study was to summarise the 29–64%, respectively. Less than 5% of the AZD5069 dose pharmacokinetic ﬁndings from eight phase I studies in was excreted as parent drug in the urine. Elderly subjects healthy volunteers given oral AZD5069, a selective small- had 39% higher AZD5069 AUC and 21% higher C than max molecule CXCR2 antagonist. younger adults. Japanese subjects had similar or slightly Methods 240 healthy volunteers across eight phase I higher exposure to AZD5069 than Caucasian subjects. Co- studies received single (0.1–200 mg) or multiple once- or administration with ketoconazole resulted in 2.1-fold twice-daily (10–120 mg) oral AZD5069 as solution, sus- higher AUC and 1.6-fold higher C . All formulations had max pension, capsules or tablets. Pharmacokinetics were eval- similar bioavailability. uated using non-compartmental analysis methods. Conclusions AZD5069 demonstrated predictive linear Results AZD5069 was rapidly absorbed (time to maximum pharmacokinetics with low intra- and moderate inter-sub- concentration * 2 h) under fasting conditions. A high-fat, ject variability and no major inﬂuences from ethnicity, age, high-calorie meal delayed and reduced the peak plasma food or formulation. Half-life data indicated suitability for AZD5069 concentration (C ) by 50%, but total exposure twice-daily dosing. max (AUC) was unchanged (fed:fasting geometric mean ratio Clinicaltrials.gov identiﬁers NCT00953888, NCT01051 90% conﬁdence interval within 0.80–1.25). The plasma 505, NCT01083238, NCT01100047, NCT01332903, NCT concentration of AZD5069 declined with an initial half-life 01480739, NCT01735240, NCT01989520. of 4 h and terminal half-life of 11 h. Steady-state plasma concentrations were achieved within 2–3 days and accu- mulation was * 1.1-fold with twice-daily dosing. Sys- Key Points temic exposure was approximately proportional to dose. AZD5069 has predictive, linear pharmacokinetics with no major inﬂuences of ethnicity, age, food or Electronic supplementary material The online version of this formulation. article (https://doi.org/10.1007/s40268-018-0236-x) contains supple- mentary material, which is available to authorized users. AZD5069 half-life data support twice-daily dosing. & Marie Cullberg email@example.com Early Clinical Development, IMED Biotech Unit, AstraZeneca, Pepparedsleden 1, Molndal, 431 83 Gothenburg, Sweden Respiratory, Inﬂammation and Autoimmunity, IMED Biotech Unit, AstraZeneca, Gothenburg, Sweden Early Clinical Development, IMED Biotech Unit, AstraZeneca, Boston, USA 150 M. Cullberg et al. 1 Introduction 2 Methods AZD5069 is a selective cysteine-x-cysteine receptor 2 2.1 Studies (CXCR2) antagonist developed as a potential treatment for inﬂammatory diseases, such as asthma, chronic obstructive Summaries of the study designs and clinicaltrials.gov pulmonary disease (COPD), and malignancies. CXCR2 is identiﬁers for all studies are included in Table 1. Study expressed on the surface of mainly neutrophils, but is also ﬂow charts are provided in Supplementary Figures S1–S8 present on other inﬂammatory effector cells. CXC (see electronic supplementary material [ESM]). All studies chemokines such as CXCL1 and CXCL8 interact with were conducted in healthy volunteer subjects. All partici- these receptors to induce neutrophil migration towards pants provided written informed consent prior to study inﬂamed tissues [1, 2]. AZD5069 is a small molecule that entry and study protocols were reviewed and approved by has demonstrated selective and reversible antagonism of local Independent Ethics Committees prior to commencing CXCR2 in human neutrophils in vitro . AZD5069 studies. All studies were performed in accordance with inhibited binding of radiolabelled CXCL8 to human ethical principles that have their origin in the Declaration CXCR2 in vitro with an IC (concentration that produces of Helsinki, and consistent with the International Confer- 50% inhibition) value of 0.79 nmol/L. ence on Harmonisation (ICH)/Good Clinical Practice Studies in human hepatocytes, liver microsomes and a (GCP) and applicable regulatory requirements. range of heterologously expressed human cytochrome P450 (CYP) enzymes have indicated that the oxidative 2.1.1 First-in-Human Studies: Single Ascending Dose metabolism of AZD5069 is primarily mediated by (SAD), Multiple Ascending Dose (MAD) and SAD/ CYP3A4 and CYP2C9 isozymes, but other metabolic MAD in Japanese Subjects (JSAD/JMAD) pathways (e.g. glucuronidation) may also play a signiﬁcant role in the elimination of AZD5069 . To date, AZD5069 The primary objective of these trials was to explore the has been studied in eight phase I studies in healthy vol- safety and tolerability of escalating single (0.1–200 mg) unteers [5–11], and in three phase II studies in inﬂamma- and/or twice daily doses (10–100 mg) of AZD5069. The tory respiratory diseases [11–14]. secondary objective was to investigate AZD5069 kinetics. Oral administration of AZD5069 markedly reduces The studies were randomised, double-blind, parallel-group, sputum neutrophilia in patients with stable bronchiectasis placebo-controlled studies and were each conducted at a and neutrophils in sputum and lung tissue in asthma single centre. A safety review committee assessed available patients at dose levels that do not affect neutrophil host pharmacokinetic, safety and tolerability data after each defence functions [9, 12, 15]. However, no objective cohort was dosed to determine the next dose level. measures of clinical improvement emerged in the bronchiectasis study , nor in the recently conducted 2.1.2 Absorption, Distribution, Metabolism and Excretion phase IIb study in severe asthma . This was despite an Mass Balance (ADME) Study estimated receptor occupancy of [ 90% over the dosing interval (45 mg twice daily [bid]) in more than 75% of the The primary objective of the ADME study was to char- patients , and signiﬁcant neutrophil reductions. acterise the pharmacokinetics, routes of excretion and AZD5069 is currently under investigation in patients with metabolism of a single oral dose of [ C]AZD5069 in cancer . healthy male subjects, who received a single dose of an Here, we report the pharmacokinetic (PK) proﬁle of oral solution containing 120 mg [ C]AZD5069 AZD5069 based on the data from the eight conducted (21.6 MBq). Blood and urine samples were collected pre- phase I studies in healthy volunteers. The objectives of dose and, along with faeces samples, at speciﬁed intervals these studies included characterisation of plasma and urine for up to 168 h after administration and were assessed for pharmacokinetic parameters of AZD5069, rates and routes radioactivity levels to determine the mass balance of of elimination, time to reach steady state, time and dose AZD5069. Plasma non-compartmental pharmacokinetic dependencies, diurnal variability, and the inﬂuence of parameters were assessed on unlabelled AZD5069. intrinsic (ethnicity, age) and extrinsic (food, CYP3A inhi- bition and formulation) factors on AZD5069 pharmacokinetics. Pharmacokinetics of AZD5069, a CXCR2 Antagonist 151 Table 1 Key information for phase I studies in healthy volunteers Study acronym Study description Dose/route/duration Demographics and clinicaltrials.gov ID SAD Single ascending Single oral doses of AZD5069 or placebo suspension: 64 males and 5 females (all dose Caucasian), median age NCT00953888 0.1, 0.5, 1.65, 5.45, 17.99, 60, 120, 200 mg 37 years (range 18–58); 45 DB, PC, PG, R on AZD5069, 24 on placebo MAD Multiple ascending Oral doses of AZD5069 or placebo suspension: 33 (32 Caucasian, 1 Black) dose male smokers, median age NCT01051505 10, 40, 100 mg 26 years (range 18–56); 25 DB, PC, PG, R Single doses on Days 1 and 8, bid administration on Days 2–7 on AZD5069, 8 on placebo JSAD/JMAD Single and Single and multiple oral doses of AZD5069 or placebo suspension: 63 Japanese males, median age multiple 10, 20, 40, 60, 80 mg; and a single oral dose of 120 mg 28 years (range 22–39); 42 NCT01100047 ascending dose on AZD5069, 21 on placebo Single doses on Days 1 and 11, bid administration on Days 4–10 in Japanese subjects DB, PC, PG, R ADME Mass balance Single oral 120-mg dose of [ C]-radiolabelled AZD5069 solution 6 Caucasian males, median study age 54 years (range 50–65); NCT01332903 all on AZD5069 ADME OL NF 24 h PK Neutrophil 100 mg AZD5069 bid oral capsule and placebo capsule 29 males and 1 female (all function 24-h Caucasian), median age NCT01480739 6 days in each of 2 treatment periods pharmacokinetic 28 years (range 19–45); 29 study received AZD5069 and 29 received placebo DB, PC, R, XO Food (Part A) Food effect Single oral doses of 120 mg AZD5069 suspension 15 Caucasian and 1 other non- speciﬁc males, median age NCT01083238 OL, R, XO 2 doses, 1 after fasting and 1 after a high fat meal 29.5 years (range 23–57); all on AZD5069 Age (Part B) Age effect Single oral dose of 120 mg AZD5069 suspension 3 males and 5 females (all Caucasian), median age 68 NCT01083238 OL (range 65–77); all on AZD5069 DDI Ketoconazole Single oral doses of AZD5069 15 mg (3 9 5-mg oral capsules) on 2 13 Caucasian, 1 Black, 1 CYP3A4 occasions, 2nd occasion on Day 3 of a 5-day treatment period with Asian males, median age NCT01735240 inhibition ketoconazole 400 mg daily 33 years (range 21–44); 15 received AZD5069 alone, 14 OL, ﬁxed sequence received AZD5069 plus ketoconazole Rel BA Formulation 2 9 20-mg ? 1 9 5-mg capsules vs 3 9 45-mg tablet formulations 7 Caucasian, 3 Black and 6 relative Asian males, median age NCT01989520 4 single doses bioavailability 33 years (range 20–46); 16 study received AZD5069 OL, XO, R ADME absorption, distribution, metabolism, excretion, bid twice daily, DB double-blind, DDI drug–drug interaction, OL open-label, PC placebo- controlled, PG parallel group, R randomised, XO crossover 2.1.3 Steady-State 24-h Neutrophil Function circulating neutrophil numbers following a burst of stren- Pharmacokinetic Study (NF 24 h PK) uous exercise and following subcutaneous injection of granulocyte colony-stimulating factor. The secondary The primary objectives of the NF 24 h PK study were to objective was to evaluate the steady-state 24-h pharma- investigate the effect of AZD5069 100 mg bid on neu- cokinetic proﬁle of AZD5069 following twice-daily trophil number and function in the circulation, and on administration. For the pharmacokinetic analysis, 152 M. Cullberg et al. comparisons were made between evening and morning 2.3 Study Drugs administration by frequent blood sampling for 24 h. AZD5069 was administered in several formulations in 2.1.4 Inﬂuence of Food and Age Study (Food and Age) these phase I studies. In the SAD, MAD and JSAD/JMAD studies, oral suspensions containing either 1 mg/g or The primary objectives of this study were to compare the 50 mg/g AZD5069 were used. In the Food and Age study, only the 50-mg/g suspension was used. Capsules contain- pharmacokinetics of AZD5069 following administration of single 120-mg doses when fasting or after a high-fat meal, ing 5 mg AZD5069 were used in the DDI study to administer a 15-mg dose. For the ADME study, radiola- and to make an initial assessment of any pharmacokinetic differences in the fasted state between adult (18–65 years) belled [ C]AZD5069, providing 0.9 MBq/mL, was and elderly ([ 65 years) subjects. To assess the effect of administered as an oral 5-mg/mL solution. For the NF 24 h food, a single dose of AZD5069 was administered to adults PK study, AZD5069 was administered as capsules con- either 30 mins after commencing a standard meal or fol- taining 50 mg AZD5069. In the Rel BA study, a 45-mg lowing a fast of at least 10 h (Part A). The standard meal tablet formulation intended for use in phase III studies, and contained 970 kcal; 182, 260 and 529 kcal derived from two other 45-mg tablet formulations with slower dissolu- tion rates were compared with capsules used in a phase II protein, carbohydrate and fat, respectively, and followed FDA guidelines for a high-fat and high-calorie test meal dose-ﬁnding study containing 5 mg and 20 mg (2 9 20 mg ? 1 9 5 mg) of AZD5069. . In Part B of the study, a single dose of AZD5069 was The ketoconazole (Nizoral ) administered in the DDI administered to healthy elderly volunteers in the fasted study was a 200-mg tablet manufactured by Janssen-Cilag, state for comparison with the fasted adults from Part A of Belgium. the study. 2.4 Collection and Analysis of Pharmacokinetic 2.1.5 Ketoconazole Drug–Drug Interaction Study (DDI) Samples The primary objective of this DDI study was to evaluate Venous blood, urine and, when required, faeces samples for pharmacokinetic analysis were taken pre-dose and at the effect of ketoconazole on the pharmacokinetics of AZD5069. A single dose of 15 mg AZD5069 was admin- speciﬁed intervals throughout each study. All study sam- ples were analysed within the established stability using istered to healthy adults followed by once-daily adminis- tration of 400 mg ketoconazole with a single dose of fully validated bioanalytical methods in the laboratories of AstraZeneca or, if outsourced, at one of the preferred 15 mg AZD5069 administered concomitantly on the third day of ketoconazole administration. provider’s laboratories on behalf of AstraZeneca. To investigate the concentration of AZD5069 in human 2.1.6 Relative Bioavailability Study (Rel BA) plasma and urine samples, a method that employs protein precipitation (plasma) or dilution (urine) followed by liquid The primary objective of the Rel BA study was to assess chromatography with tandem mass spectrometric detection the relative bioavailability of AZD5069 from the putative (LC–MS/MS) in the positive ion mode was used. The AZD5069 methods (both plasma and urine) were initially phase III tablet formulation in comparison with the capsule formulation used in phase II studies. The secondary validated in the range 1.0–5000 nM with the lower limit of quantiﬁcation (LLOQ) set at 1.0 nM using 100 lL matrix. objective was to assess the bioavailability of AZD5069 from up to three phase III tablet formulation variants rel- As the plasma method was transferred to a new bioana- ative to the putative phase III tablet. Subjects received lytical provider, the range was slightly adapted to meet the single oral doses of 45 mg AZD5069 as 2 9 20-mg ? 1 expected sample concentrations to 2.0–4000 nM. In addi- 9 5-mg capsules and three different 45-mg tablet formu- tion, a new combined AZD5069 and AZ13587715 (a lations, in a randomised cross-over design. metabolite) method in the range 1–1000 nM was validated on the same LC/MS/MS platform, using 100 lL matrix. Where appropriate because of concentrations above the 2.2 Study Populations upper quantiﬁcation limit, samples could also be diluted up to 10 times (100 for the combined AZD5069 and All subjects who participated in the studies were judged to be healthy, based on standard inclusion and exclusion AZ13587715 method) with blank human matrix and then quantiﬁed within the validated calibration range. criteria. An overview of the demographic characteristics of the participants from each study are shown in Table 1. Quality controls (QCs) prepared at three different con- centration levels were analysed in duplicate with each Pharmacokinetics of AZD5069, a CXCR2 Antagonist 153 batch of samples against separately prepared calibration analysis was conducted using data from the SAD and MAD standards to assess the day-to-day performance of the studies. Since there are single-dose data included in the assay. In addition, a dilution QC at one concentration level MAD study, these data were pooled with the single-dose was analysed with the study samples in all runs where data collected in the SAD study. This analysis was repeated dilution was applied to any of the study samples. The intra- for each of single-dose (Day 1) AUC and C and the max batch and inter-batch precision, reported as coefﬁcient of corresponding AUC and C following multiple dosing max variation (CV), had to be B 15% at all levels (ex- (Day 8). Dose proportionality was assessed using a power cept B 20% at LLOQ) to be acceptable and intra-batch and model that, while presented on a linear scale, was based on inter-batch bias had to be within 15% of the nominal a linear regression model of log-transformed AUC (and concentration at all levels (except ± 20% at LLOQ). In all C separately) as a function of log-transformed dose. The max of the studies included in this publication, the overall model was then presented on a linear scale as: intercept slope precision for the QC samples (undiluted) at all three con- AUC = exp dose . The 90% CI for the slope centration levels was B 12.9% and the overall bias ranged estimate is also presented along with the adjusted correla- between – 6 and 7%. The overall precision for the dilution tion coefﬁcient (R ). QC sample was \ 12.5% and the overall bias ranged Pharmacokinetic differences between Japanese and between – 6.8 and 3.5%. In addition, up to 10% of the Caucasian subjects were assessed by comparing geometric study samples in each study were reanalysed to conﬁrm mean ratios (GMRs) of dose-normalised AUC and C max acceptable incurred sample reproducibility of the methods values and by graphs of exposure across all dose levels. used. AUC and C inter- and intra-subject components of max variability were estimated with a mixed-effect analysis 2.5 Pharmacokinetic Evaluation (ANOVA) model for the values on Days 6 and 8, with subject as a random effect, for each dose group (MAD Pharmacokinetic parameters were estimated by non-com- study). Similarly, the variability components were esti- partmental analysis using WinNonLin Enterprise Version mated with a ﬁxed-effect analysis (ANOVA) model for the 5.2 or later (Pharsight Corporation, Mountain View, CA, values on Day 3 (morning and evening), with subject as a USA) or SAS Version 9.2 (SAS Institute Inc., Cary, NC, ﬁxed effect in the NF 24 h PK study. USA). The plasma parameters assessed for AZD5069 were For most studies (SAD, MAD, JSAD/MAD, Food and maximum (peak) drug concentration (C ), time to reach Age, ADME, NF 24 h PK), the sample sizes were based on max maximum concentration (t ), area under plasma con- previous experience from similar studies with other com- max centration–time curve from zero to inﬁnity after a single pounds and were determined without formal statistical dose (AUC), AUC from zero to time 12 h (AUC ), AUC considerations, hypotheses or formal power calculations. 0–12 from zero to time 24 h (AUC ), AUC during a dose For the DDI study, the variability observed in the MAD 0–24 interval at steady state (AUC ), terminal elimination rate study was used: the log scale standard deviation (SD) for ss constant (k ), terminal half-life (t ), apparent oral plasma C and AUC were 0.298 and 0.134, respectively, for the z max clearance (CL/F), renal clearance (CL ) and apparent oral pooled doses. Based on a 90% CI (two one-sided testing volume of distribution during terminal phase (V /F). procedure at 5% type I error level for each test), 12 patients for AUC and 18 patients for C would yield [ 90% max 2.6 Statistical Evaluation power to declare no effect based on default 0.8–1.25 no- effect boundaries. For the Rel BA study, the variabilities Statistical analyses were performed using SAS Version observed in the MAD, DDI and NF 24 h PK studies were 8.1 or later. Pharmacokinetic variables were summarised used for sample size estimation: within-subject log-scale using descriptive statistics. The following factors were SD ranged from 0.03 to 0.14 for AUC and 0.16 to 0.22 for assessed by ANOVA as a repeated-ﬁxed effect using point C . Assuming an observed ratio of 1 in the AUC and max estimates of the geometric least-squares (LS) mean ratios C between two treatments and a within-subject SD of max and in some cases associated 90% conﬁdence intervals 0.1 (AUC) and 0.2 (C ) on the log scale, corresponding max (CIs) for AUC (and C where applicable): accumulation 90% CIs for the treatment ratios are 0.93–1.08 and max (AUC [or AUC ] after multiple dosing to AUC 0.86–1.16, respectively. 0–12 0–24 0–12 [or AUC ] after the ﬁrst dose), time dependency (AUC 0–24 ss to AUC [or AUC ] after the ﬁrst dose), diurnal 0–12 0–24 variability (evening to morning), fed to fasted, elderly to adults and tablet to capsule. Time to steady state was assessed by visual inspection of graphs of individual trough concentrations (C ) over time. Dose-proportionality 12h 154 M. Cullberg et al. 3.2 Absorption 3 Results A summary of the pharmacokinetic parameters from all of AZD5069 was well absorbed as 65% of the radioactivity was found in urine following a single 100-mg oral dose of the phase I studies is presented in Supplementary Table S1 [ C]AZD5069 solution (ADME study). Absorption from (see ESM). Dose-normalised AUC values are shown in the oral solution was rapid with C being reached at a Fig. 1. max median time of 0.5 h after administration (ADME study). AZD5069 was also relatively quickly absorbed from an 3.1 Plasma Concentration Proﬁle oral suspension during fasting conditions in healthy sub- jects (median t 2 h (range 0.5–4 h), and this was inde- Following an oral single dose of an AZD5069 suspension max pendent of the dose administered (SAD study). The rate to fasting healthy subjects, the peak plasma concentration and extent of absorption of AZD5069 when administered (C ) was observed at approximately 2 h. The concen- max as capsules was similar to that with the oral suspension tration declined in a multi-exponential manner with approximate initial and terminal half-lives of 4 and 11 h, (Fig. 1a, Table S1 [ESM]). Geometric mean AUC, C max and median t were 8390 nmol h/L, 2210 nmol/L and respectively (Fig. 2). max 2.0 h after 40 mg AZD5069 in suspension (MAD study, Fig. 1 Boxplots of dose-normalised AUC (area under plasma concentration–time curve) by dose and study (see Table 1 for acronyms) following: a single-dose; and b repeated-dose (bid) administration of AZD5069 Pharmacokinetics of AZD5069, a CXCR2 Antagonist 155 Fig. 2 Arithmetic mean plasma AZD5069 concentration–time proﬁles following a single oral administration of 0.1–200 mg (SAD study). Inserted graph shows the same data on a logarithmic y-axis scale Day 1) compared with 11,500 nmol h/L, 2610 nmol/L and elimination accounting for the remaining 35% of the dose 3.0 h after 45 mg in capsules (Rel BA study). (ADME study). AZD5069 is likely to be extensively metabolised in humans as the fraction of an oral dose 3.3 Distribution excreted as unchanged drug in urine was \ 10% (SAD study, MAD study, ADME study). Extensive metabolism The geometric mean apparent oral volume of distribution was also indicated by mean plasma radioactivity equivalent (V /F) ranged from 56 to 134 L at relevant doses across concentrations being greater than mean AZD5069 plasma studies in adult Caucasian subjects. concentrations at all sampling times following ingestion of [ C]AZD5069 solution (ADME study). 3.4 Elimination 3.5 Steady-State Pharmacokinetics The total apparent plasma clearance (CL/F) was estimated to be approximately 8 L/h (mean across all studies in adult Steady-state conditions were attained within 2–3 days of Western subjects). CL/F appeared to be independent of bid dosing (MAD study). The mean accumulation ratio was dose with mean values of 10.8, 10.4 and 8.85 L/h in the 1.07–1.19 on Day 6 and 1.01–1.26 on Day 8 following bid 10-, 40- and 100-mg dose groups on Day 1, respectively, dosing (MAD study).Total AUC after the ﬁrst dose was and 11.0, 10.4 and 8.23 L/h on Day 8, respectively (MAD similar in magnitude to AUC during a dose interval at study). Renal clearance (CL ) also appeared to be inde- steady state (AUC ), indicating time-independent phar- R 0–12 pendent of dose with mean values of 0.270, 0.302 and macokinetics. The variability in AUC was moderate 0.245 L/h in the 10-, 40- and 100-mg dose groups on Day between subjects: 29–62% CV in the MAD study, and 64% 1, respectively, and 0.367, 0.352 and 0.386 L/h on Day 8, CV in the NF 24 h PK study, and very low between dosing respectively (MAD study). The fraction of the AZD5069 occasions within subjects (3–11% CV in the MAD study dose excreted unchanged thus ranged from 2.5 to 4.7%. and 8% CV in the NF 24 h PK study). Corresponding C max Following administration of a single 120-mg dose of values were 13–35 and 44% for the inter-subject variability [ C]AZD5069 as an oral solution, excretion via urine was and 19–22 and 16% for the intra-subject variability. the primary route of elimination (65% of dose) with faecal 156 M. Cullberg et al. 3.6 Diurnal Variability 3.8 Pharmacokinetics in Populations of Special Interest The geometric mean plasma concentration of AZD5069 observed at 12 h post-dose (C ) was greater for the eve- Elderly (median 68 [range 65–77] years) healthy Caucasian ning dose (564 nmol/L) compared with the morning dose subjects had, on average, 39% higher AUC (90% CI 4–85) (459 nmol/L) and median t was 1 h later after the and 21% higher C (90% CI -4 to 54) values for ss,max max evening dose (NF 24 h PK study). The 90% CIs of the AZD5069 than younger subjects (median 30 [range 23–57] evening:morning ratios for the exposure parameters were, years) (Food and Age study). however, all within the default bioequivalence boundaries Healthy Japanese subjects appeared to have similar, or of 0.80–1.25; the AUC ratio was 1.17 (90% CI higher, exposure to AZD5069 compared with Caucasian 0–12,ss 1.13–1.21), C ratio 1.15 (90% CI 1.07–1.24), and subjects, depending on the study and dose level being ss,max C , ratio 1.01 (90% CI 0.99–1.03). investigated (Fig. 1). Geometric mean AUC ratios (Ja- ss min panese:Caucasians) for AZD5069 varied between 0.80 and 3.7 Dose Linearity 1.06 when Japanese SAD data were compared with Wes- tern SAD data at equal doses (JSAD/JMAD study, SAD AUC and C were approximately proportional to the study). When comparing Japanese SAD/MAD AZD5069 max AZD5069 dose following both single (0.1–200 mg) and data with Western MAD data at equal doses, the exposure repeated (10–100 mg bid) administration (Fig. 3). was higher in Japanese subjects, after both the ﬁrst dose and repeated dosing (JSAD/JMAD study, MAD study). Fig. 3 AUC and C versus dose following single-dose (a) and study. Lines and equations are based on regression analysis (see max repeated-dose (bid) (b) administration of AZD5069. Red symbols Sect. 2.6 for details). AUC area under plasma concentration–time represent data from the SAD study and blue symbols from the MAD curve, C peak plasma concentration max Pharmacokinetics of AZD5069, a CXCR2 Antagonist 157 Geometric mean ratios varied between 1.20 and 1.97. When pooling all AZD5069 doses, the dose-normalised AUC at steady state was 40% higher in the Japanese subjects. Smokers, who were only included in the MAD study, had generally somewhat lower dose-normalised AUC val- ues than non-smokers from the SAD study, but similar to those from the NF 24 h PK study and the fasted adults in the Food and Age study (Fig. 1). 3.9 Drug–Food and Drug–Drug interactions Administration of AZD5069 oral suspension immediately after a high-fat meal resulted in a 50% lower C , which max was delayed by 3 h in comparison with administration during the fasting state; however, the fed:fasted AUC ratio was within the standard bioequivalence boundaries 80–125% (Food and Age study). The fed:fasted geometric mean ratio for AZD5069 AUC was 0.89 (90% CI 0.83–0.94), while for C it was 0.50 (90% CI 0.40–0.61). max The mean concentration–time proﬁle following admin- istration of AZD5069 with the CYP3A4 inhibitor keto- conazole was higher than when AZD5069 was administered alone, from the ﬁrst quantiﬁable sample at 30 mins post-dose. In most subjects, AZD5069 plasma con- centrations dropped below the LLOQ by 48 and 72 h post- dose in the AZD5069 and AZD5069 ? ketoconazole treatment groups, respectively (Fig. 4). Co-administration of AZD5069 and ketoconazole resulted in a 2.1-fold increase in AUC (90% CI 1.9–2.3) and a 1.6-fold (90% CI 1.5–1.9) increase in C for AZD5069. max 3.10 Formulation Factors The relative bioavailability study showed similar AUC for all formulations tested, but the AZD5069 absorption was slightly faster from the intended phase III tablet formula- tion than from the capsule formulation used in the phase II Fig. 4 Geometric mean (± SD) plasma concentration versus time proﬁles of AZD5069 when administered alone on Day 1 and in studies. Geometric LS mean for AZD5069 AUC (nmol*h/ combination with ketoconazole on Day 3 of a 5-day treatment period L) was 11,570 (95% CI 9425–14,200) for the phase II with ketoconazole 400 mg daily (DDI study) on a linear scale (a) and capsules compared with 12,080 (95% CI 9846–14,820) for a log scale (b) the intended phase III tablet formulation. AUC geometric LS mean values for the two slower dissolution tablets were 4 Discussion 10,710 (95% CI 8725–13,140) and 10,870 (95% CI 8861–13,350), respectively. The geometric mean ratio of These studies provide a comprehensive portrait of the tablet:capsule formulations was 1.04 (90% CI 0.99–1.10) pharmacokinetic proﬁle of AZD5069 in healthy human for AUC and 1.20 (90% CI 1.05–1.38) for C . The tablet max volunteers. AZD5069 is rapidly absorbed during fasting and capsule formulations of AZD5069 were considered conditions; administration with food delays and reduces the pharmacokinetically comparable. Overall, there were no peak plasma concentration, but the total exposure (AUC) consistent differences in AUCs between the studies that remains unchanged. could be related to formulation (Fig. 1a), suggesting that The plasma concentration proﬁle of AZD5069 has a the solid formulations of AZD5069 had similar bioavail- terminal half-life of 11 h, which supports twice-daily ability to the liquid formulations. 158 M. Cullberg et al. administration. Steady state is achieved within 2–3 days 5 Conclusion and no major time dependency is observed in the phar- macokinetic proﬁle. Systemic exposure to AZD5069 is AZD5069 demonstrated a predictive, linear pharmacoki- approximately proportional to the dose following both netic proﬁle with low intra- and moderate inter-subject single and repeated dosing. In a 6-month study in patients variability and no major inﬂuences from ethnicity, age, with severe asthma, plasma exposure of AZD5069 was food or formulation. The estimated half-life indicated approximately proportional to the dose across the range suitability for twice-daily administration in future clinical evaluated (5–45 mg bid) and the morning pre-dose plasma development to assess the therapeutic potential of levels were consistent over the time period investigated AZD5069. (1 week to 6 months) . Acknowledgements We thank all the participants, investigators and AZD5069 is extensively metabolised in humans, trial-site staff who were involved in the conduct of these trials. We with \ 10% of the oral dose excreted unchanged. In vitro also thank David Candlish, InScience Communications, for medical data have indicated that the oxidative metabolism of writing and editorial assistance. Our thanks to the Bioanalytical lab- AZD5069 is primarily mediated by CYP3A4 and CYP2C9 oratories for analysing the samples from the clinical trials included in the manuscript. Many thanks to Xiao Tong for programming isozymes, but other metabolic pathways (for example, contributions. glucuronidation) may also play a signiﬁcant role . Based on ex vivo studies, AZD5069 is extensively bound (99%) Compliance with Ethical Standards to plasma proteins and does not enter into red blood cells Funding All studies were sponsored and funded by AstraZeneca. (blood:plasma ratio = 0.56) (AstraZeneca, data on ﬁle). David Candlish, of InScience Communications, Tattenhall, UK, CL/F data thus suggest that AZD5069 is a low-clearance provided editorial and medical writing assistance, which was funded drug (extraction ratio \ 0.18). Assuming that the fraction by AstraZeneca. absorbed is at least 65% and the absolute bioavailability is Conﬂicts of Interest All authors are full-time permanent employees more than 50%, the renal clearance can be estimated to of AstraZeneca. be \ 20% of the total clearance. This suggests that the elimination of AZD5069 is not likely to be sensitive to Ethical Approval All procedures performed in studies involving differences in renal function, but that hepatic function human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 might be relevant. Japanese subjects appeared to have Helsinki declaration and its later amendments or comparable ethical similar, or somewhat higher, exposure to AZD5069 than standards. Caucasian subjects, depending on the study and dose level. The pharmacokinetic variability within subjects was very Informed Consent Informed consent was obtained from all indi- vidual participants included in the studies. low, while between-subject variability was moderate. There was no obvious inﬂuence of smoking on the Open Access This article is distributed under the terms of the AZD5069 exposure, but the results should be interpreted Creative Commons Attribution-NonCommercial 4.0 International with caution as smokers and non-smokers were not inclu- License (http://creativecommons.org/licenses/by-nc/4.0/), which per- ded in the same study. Smoking was, however, not mits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit to the original expected to be an inﬂuential factor, as AZD5069 is not author(s) and the source, provide a link to the Creative Commons metabolised by any of the iso-enzymes known to be license, and indicate if changes were made. induced by smoking, i.e. CYP1A2 or CYP2B6 . Co- administration with the potent CYP3A4 inhibitor keto- conazole resulted in an increase in AZD5069 AUC and C of 2.1- and 1.6-fold, respectively. The \ 5-fold max References increase indicates that AZD5069 is not a sensitive substrate of CYP3A4 . The clinical relevance of this remains, 1. Gernez Y, Tirouvanziam R, Chanez P. Neutrophils in chronic inﬂammatory airway diseases: can we target them and how? Eur however, to be assessed and will depend on the selected Respir J. 2010;35(3):467–9. clinical dose and the safety margins at this dose. Until the 2. Nair P, Aziz-Ur-Rehman A, Radford K. Therapeutic implications therapeutic window has been characterised in the target of ‘neutrophilic asthma’. Curr Opin Pulm Med. 2015;21(1):33–8. patient population, we have taken a cautious approach, 3. Nicholls D, Wiley K, Dainty I, MacIntosh F, Phillips C, Gaw A, Ka¨rrman Mardh C. Pharmacological characterization of recommending that AZD5069 should not be combined with AZD5069, a slowly reversible CXC chemokine receptor 2 potent CYP3A4 inhibitors in clinical studies. antagonist. J Pharmacol Exp Ther. 2015;353(2):340–50. 4. Gardiner P, Ekdahl A, Palmgren A-P, Cullberg M, Larsson B. The mechanisms and routes of clearance of the CXCR2 antago- nist AZD5069 in human. American Association of Pharmaceu- tical Scientists (AAPS); San Antonio, Texas, USA. 2013. Pharmacokinetics of AZD5069, a CXCR2 Antagonist 159 5. Wray H, Sparrow A. Safety and tolerability of single doses of patients with moderate-to-severe COPD. Pulm Pharmacol Ther. AZD5069 in healthy volunteers. Eur Respir J. 2011;38(Supple- 2015;31:36–41. ment 55):3984. 14. O’Byrne P, Metev H, Puu M, Richter K, Keen C, Uddin MK, 6. Wray H, Wilbraham D. Safety and tolerability of AZD5069 in Larsson B, Cullberg M, Nair P. Efﬁcacy and safety of a CXCR2 healthy smokers following multiple ascending doses. Int J antagonist, AZD5069, in patients with uncontrolled persistent Immunorehab. 2012;14(1):78. asthma receiving ICS/LABA therapy: a phase 2b randomised trial 7. Lorch U, Negi H, Tabata H, Wray H, Cullberg M, Larsson B. The Lancet. Respir Med. 2016;4(10):797–806. safety, tolerability and pharmacokinetics of AZD5069, a novel 15. Watz H, Uddin M, Pedersen F, Kirsten A, Goldmann T, Stell- CXCR2 antagonist, in healthy Japanese volunteers. Eur Respir J. macher F, Groth E, Larsson B, Bottcher G, Malmgren A, Kraan 2012;40(Supplement 56):P4842. M, Rabe KF. Effects of the CXCR2 antagonist AZD5069 on lung 8. Mant T, Wray H, Cullberg M, Larsson B. The absorption, dis- neutrophil recruitment in asthma. Pulm Pharmacol Ther. tribution, metabolism and excretion (ADME) of single oral doses 2017;45:121–3. of AZD5069, a novel CXCR2 antagonist, in healthy male vol- 16. Hong D, Falchook G, Cook CE, Harb W, Lyne P, McCoon P, unteers. Eur Respir J. 2012;40(Supplement 56):P2149. Mehta M, Mitchell P, Mugundum GM, Scott M, Wang JS. A 9. Jurcevic S, Humfrey C, Uddin M, Warrington S, Larsson B, Keen phase 1b study (SCORES) assessing safety, tolerability, phar- C. The effect of a selective CXCR2 antagonist (AZD5069) on macokinetics, and preliminary anti-tumor activity of durvalumab human blood neutrophil count and innate immune functions. Br J combined with AZD9150 or AZD5069 in patients with advanced Clin Pharmacol. 2015;80(6):1324–36. solid malignancies and SCCHN. Ann Oncol. 2016;27(supplement 10. Wray H, Mant T. Effects of food and age on pharmacokinetics, 6):1049PD. safety and tolerability of single doses of AZD5069. Int J 17. US Dept of Health and Human Services, Food and Drug Immunorehab 2012;14(1):78. Administration, Center for Drug Evaluation and Research 11. Cullberg M. Effect of the CYP3A4 inhibitor ketoconazole on the (CDER). Guidance for Industry. Food-Effect Bioavailability and pharmacokinetics and pharmacodynamics of AZD5069, an oral Fed Bioequivalence Studies. 2002. CXCR2 antagonist. Eur Respir J. 2014;44(Supplement 58):P961. 18. Lucas C, Martin J. Smoking and drug interactions. Aust Prescrib. 12. De Soyza A, Pavord I, Elborn JS, Smith D, Wray H, Puu M, 2013;36(3):102–4. Larsson B, Stockley R. A randomised, placebo-controlled study 19. U.S. Department of Health and Human Services Food and Drug of the CXCR2 antagonist AZD5069 in bronchiectasis. Eur Respir Administration (FDA) Center for Drug Evaluation and Research J. 2015;46(4):1021–32. (CDER) Center for Biologics Evaluation and Research (CBER). 13. Kirsten A, Forster K, Radeczky E, Linnhoff A, Balint B, Watz H, Guidance for industry: drug interaction studies—study design, Wray H, Salkeld L, Cullberg M, Larsson B. The safety and tol- data analysis, and implications for dosing and labeling recom- erability of oral AZD5069, a selective CXCR2 antagonist, in mendations. 2012.
Drugs in R&D – Springer Journals
Published: May 31, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera