Pharmacokinetics/Pharmacodynamics
The Pharmacokinetics, Pharmacodynamics, and Safety of Baricitinib, an Oral JAK 1/2 Inhibitor, in Healthy Volunteers
The Journal of Clinical Pharmacology 54(12) 1354–1361 © 2014, The American College of Clinical Pharmacology DOI: 10.1002/jcph.354
Jack G. Shi, PhD, Xuejun Chen, PhD, Fiona Lee, MS, Thomas Emm, BS, Peggy A. Scherle, PhD, Yvonne Lo, MS, Naresh Punwani, PhD, William V. Williams, MD, and Swamy Yeleswaram, PhD
Abstract Baricitinib (also known as LY3009104 or INCB028050), a novel and potent small molecule inhibitor of Janus kinase family of enzymes (JAKs) with selectivity for JAK1 and JAK2, is currently in clinical development for the treatment of rheumatoid arthritis (RA) and other inflammatory disorders. Two double-blind, randomized, and placebo-controlled studies were conducted to evaluate single ascending doses of 1–20 mg and multiple ascending doses of 2–20 mg QD and 5 mg BID for 10 or 28 days in healthy volunteers. Following oral administration, baricitinib plasma concentration typically attains its peak value within 1.5 hours postdose and subsequently declines in a bi-exponential fashion. Baricitinib demonstrates dose-linear and timeinvariant pharmacokinetics, with low oral-dose clearance (17 L/h) and minimal systemic accumulation observed following repeat dosing. The mean renal clearance of baricitinib was determined to be 12 L/h. [Correction added after publication 12 November 2014: in the preceding sentence, “2 L/h” was changed to “12 L/h.”] The effect of a high-fat meal on baricitinib pharmacokinetics was insignificant. The pharmacodynamics of baricitinib, evaluated by the inhibition of STAT3 phosphorylation following cytokine stimulation in the whole blood ex vivo, was well correlated with baricitinib plasma concentrations. Baricitinib was generally safe and well tolerated, with no serious treatment-related adverse events (AEs) reported from either of the studies. An expected rapidly reversible, dose-related decline in absolute neutrophil count was seen with baricitinib.
Keywords baricitinib, JAK, rheumatoid arthritis, pharmacokinetics, pharmacodynamics
Rheumatoid arthritis (RA) is a systemic, autoimmune disease that affects approximately one percent of the population.1 Active RA is characterized by chronic joint inflammation that can result in progressive destruction of cartilage and juxtarticular bone leading to functional impairment and disability.2 Conventional RA therapies consist of treatments with non-steroidal anti-inflammatory drugs (NSAIDs), corticosteroids, and disease-modifying anti-rheumatic drugs (DMARDs). The emergence of biologic therapies for RA, such as tumor necrosis factor-a (TNFa) inhibitors, represents treatment options that mitigate the clinical manifestations of RA by selectively targeting a key inflammatory molecule. However, approximately 30% of patients with RA, usually those with the more severe disease, will not experience satisfactory clinical improvement with use of currently available therapies.3 Therefore, a significant unmet medical need remains for more effective and better tolerated treatments for RA. Chronic inflammatory conditions, including RA, are associated with aberrant production of cytokines and growth factors, and the Janus kinase family of protein tyrosine kinases (JAKs) plays an important role in the signaling of a number of cytokines and hematopoietic growth factors.4 Many of the pro-inflammatory cytokines
implicated in the pathogenesis of RA, including IL-6 and IFN-g, utilize cell signaling that involves the JAK/signal transducers and activators of transcription (STAT) pathways. Inhibition of JAK-STAT signaling can target multiple RA-associated cytokine pathways and thereby reduce inflammation, cellular activation, and proliferation of key immune cells as demonstrated in studies of tofacitinib, a JAK inhibitor that has shown clinical efficacy in patients with RA.5 Baricitinib (LY3009104 also previously known as INCB028050) is a potent and selective small molecule inhibitor of JAK1/2 enzymes. In vitro, baricitinib inhibits JAK1 and JAK2 with IC50 values in the single-digit nM range, yet it does not significantly inhibit a diverse panel
From Incyte Corporation, Wilmington, DE, USA, Submitted for publication 12 May 2014; accepted 23 June 2014. [Correction added on 17 November 2014 after publication 12 November 2014; a typographical error was corrected in the Abstract section.] Corresponding Author: Jack G. Shi, PhD, Experimental Station, Building E400, Rt. 141 & Henry Clay Road, Wilmington, DE 19880, USA Email: [email protected]
1355
Shi et al
of 28 kinases when tested at 100-fold of the IC50 against JAK1 and JAK2.6 In multiple murine disease models that are relevant to arthritis, baricitinib showed excellent potential of antirheumatic efficacy without evidence of suppression of humoral immunity or adverse hematologic effects.6 In a Phase IIa proof-of-concept study conducted in RA patients, baricitinib demonstrated rapid and sustained efficacy in patients with inadequate responses to DMARDs across the three doses (4, 7, and 10 mg QD) evaluated over a 24-week treatment period.7 All doses were generally safe and well tolerated, and the adverse events (AEs) were predominantly mild to moderate with frequencies similar to placebo. Subsequently, in a Phase IIb dose-ranging study conducted in RA patients on stable background of methotrexate, baricitinib achieved the primary endpoint by demonstrating a statistically significant difference (P < 0.001) in the American College of Rheumatology 20 (ACR20) response for the 4 mg QD and 8 mg QD dose groups compared to placebo after 12-weeks of treatment.8 This report summarizes the findings of the first-inhuman (FIH) and single ascending dose (SAD) and multiple ascending dose (MAD) studies of baricitinib, which were conducted to evaluate the safety, tolerability, pharmacokinetic, and pharmacodynamic profiles of baricitinib in healthy adult volunteers.
Methods Two randomized, double-blind (to the subjects and investigator), and placebo-controlled clinical studies were conducted in full accordance with the Declaration of Helsinki, principals of good clinical practices (GCP), and local laws regarding the protection of the rights and welfare of human participants in biomedical research. The protocols were approved by an independent institutional review board (IRB), and informed consent for all participants was obtained prior to screening. An immediate release capsule formulation of baricitinib phosphate was used in both studies, and all doses
reported herein equivalents.
were
the
baricitinib
free
base
Study Population Males and females, 18–55 years of age, with a body mass index (BMI) of between 18 and 32 kg/m2 were eligible for participation in the studies if they were judged to be in good health based on their medical history and physical examinations including vital signs, electrocardiogram, and clinical laboratory test results. Women of childbearing potential enrolled in these studies were determined to be in a nongravid state and had agreed to take appropriate precautions (with at least 99% certainty) to avoid pregnancy from screening through follow-up. Subjects with hemoglobin levels and white blood and platelet counts below the lower reference limit for the parameters were excluded. Non-study medications were not allowed for 7 and 14 days, for over-the-counter and prescription drugs, respectively, prior to the first dose of study medication until the completion of the studies, unless deemed necessary and acceptable by the investigator. Use of investigational medications and any medications known to affect cytochrome P450 enzymes or Pglycoprotein activity was prohibited within 30 days or five half-lives (whichever was longer) prior to the study start and throughout the study. Study Design Study 1 was composed of 2 parts, and the detailed dosing paradigm is described in Table 1. Part 1 was a 3-period, ascending single-dose tolerability and PK/PD study with a planned enrollment of 24 subjects. Participants in Cohort 1 (n ¼ 12) received single oral doses of 1, 5, and 10 mg baricitinib and matching placebo in the fasted state, while participants in Cohort 2 (n ¼ 12) received single oral doses of 2 and 10 mg baricitinib in the fasted state, or 5 mg baricitinib following a high fat meal, and matching placebo. For each treatment/period, participants were randomized to receive the active treatment or placebo at a 2:1 ratio. A washout period of at least 10 days was
Table 1. Design of the Dosing Paradigm in Study 1
Cohort 1 1 1 2 2 2 3 3 N/A ¼ not applicable.
Sequence
Period 1
Period 2
Period 3
No. of subjects planned for enrollment
1 2 3 4 5 6 7 8
placebo 1 mg 1 mg placebo 2 mg 2 mg 5 mg fed 5 mg fasted
5 mg placebo 5 mg 10 mg placebo 10 mg 5 mg fasted 5 mg fed
10 mg 10 mg placebo 5 mg fed 5 mg fed placebo N/A N/A
4 4 4 4 4 4 12 12
1356 instituted between the treatment periods. Part 2 of the study with a planned enrollment of 24 subjects commenced upon completion of Part 1 to evaluate the effect of food on baricitinib single-dose PK. Using a 2-sequence, 2-period, cross-over design, the subjects were randomized with a 1:1 ratio to receive 5 mg baricitinib either fasted or after a standardized high-fat, high calorie meal. Study 2 was a sequential, 1-period, ascending multiple-dose tolerability and PK/PD study. Sequentially, higher multiple-doses of 2 mg QD, 5 mg QD, 10 mg QD, 5 mg BID, and 20 mg QD of baricitinib (after testing the safety of a single dose of 20 mg) were administered in the fasted state for 10 days; in addition, 10 mg QD and 5 mg BID were studied for 28 days in two additional cohorts. There were altogether 7 cohorts (5 cohorts for 10-day dosing, n ¼ 12 per cohort and 2 cohorts for 28-day dosing, n ¼ 16 per cohort) as indicated above. In each cohort, subjects were randomized to receive either baricitinib or placebo at a 3:1 ratio. In both studies, the AM dosing in fasted state was administered after at least a 10 hours overnight fast, and all PM dosing (Study 2 only) was preceded by at least a 3 hours fast from food. PK blood samples were collected following the single-dose administrations, the first dose and the last dose of multiple-dose administrations at 0, 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 24, 36, and 48 hours post-dose (in the multiple-dose study, the last PK samples collected following the first doses were at 12 and 24 hours, respectively, for the q12h and q24h administrations, immediately prior to the next dosing). PD blood samples were collected at 0, 1, 2, 4, 6, 12, and 24 hours post-dose in Study 1, and at 0, 2, 6 hours on Day 10 in Study 2. Additional trough (pre-dose) blood samples were collected in Study 2 before the AM dose on Days 2, 3, 5, 7, and 9 for the 10-day cohorts, and Days 2, 8, 15, and 22 for the 28-day cohort. A pre-dose urine sample and complete urine output from 0–24 hours post-dose were collected on Day 10 (for the 10-day cohorts only) for determination of urine concentrations of baricitinib. In both studies, safety and tolerability were assessed by monitoring AEs, measuring vital signs and ECGs, and clinical laboratory blood and urine sample assessments. In addition, serial measurements of white blood cells (WBCs) with differential counts were carried out predose and 4, 8, 12, 16, and 24 hours post-dose for each dose in Study 1, on Days 1 and 10 for the 10-day dosing cohorts in Study 2, and on Days 1 and 28 for the 28-day dosing cohorts in Study 2. Analytical Methods The PK samples were assayed by a validated, GLP, LC/ MS/MS method at Incyte Corporation (Wilmington, DE, USA) to determine baricitinib concentrations in plasma and urine. The plasma and urine standard curves ranged from 0.5 to 500 nM and 0.025 to 25 mM, respectively. The
The Journal of Clinical Pharmacology / Vol 54 No 12 (2014)
plasma samples were prepared using liquid/liquid extraction with methyl-t-butyl ether, and the urine samples directly diluted with acetonitrile, both with the 13 C4-labeled baricitinib (M þ 4) as an internal standard. Chromatographic separations were performed over a Waters Atlantis HILIC Silica column (50 2.1 mm, 3 mm) (Waters corporation, Milford, MA, USA) under isocratic condition using a mobile phase consisting of 97% acetonitrile and 3% 2 mM ammonium formate aqueous solution (pH ¼ 3). MS/MS analysis was performed using a positive Turbo IonSpray interface on a Sciex API-3000 or API-4000 mass spectrometer (Applied Biosystems, Foster City, CA, USA) operating in multiplereaction monitoring mode, monitoring the transitions of m/z 372.1 ! 251.2 for baricitinib, and m/z 376.1 ! 255.2 for the internal standard. For the plasma assay, intra-assay precision (CV %) and accuracy (Bias %) for quality control samples ranged from 2.6 to 10.5% and 13.5 to 5.0% respectively, while inter-assay precision and accuracy ranged from 4.5 to 12.5% and 5.9 to 1.7%, respectively. For the urine assay, intra-assay precision and accuracy for quality control samples ranged from 2.1 to 12.2% and 11.5 to 9.0% respectively, while inter-assay precision and accuracy ranged from 7.5 to 12.1% and 3.9 to 0.4%, respectively. The PD samples (ex vivo whole blood) were stimulated with interleukin-6 (IL-6) or thrombopoietin (TPO) to activate the JAK/STAT pathway, the blood cells lysed, and the total cell extracts were analyzed for levels of phosphorylated STAT3 (pSTAT3) using a specific enzyme-linked immunosorbent assay (ELISA). This assay was performed under a specific standard operating procedure (SOP). The details for the assay procedure and validation results have been described elsewhere.6,9 For each sample analyzed, duplicate PD analyses were performed and the average value was reported. For each subject, the percent inhibition of pSTAT3 was calculated by comparing pre-dose values obtained before the first dose with values obtained at different times after dosing. Pharmacokinetic and Pharmacodynamic Analysis Per study protocol, scheduled times were used for PK analysis, except when the difference between the actual time and nominal time was to be greater than 5 minutes for samples collected up to 4 hours after dose and greater than 15 minutes for samples collected more than 4 hours after dose; in those cases, actual time were used for the PK analysis. The vast majority of samples were collected on the scheduled times with only a few samples slightly outside of the pre-defined windows (corrected to real collection times). Standard non-compartmental PK analysis (NCA) methods were used to analyze the baricitinib plasma concentration data using WinNonlin version 5.0.1 (Pharsight Corporation, Mountain View, CA, USA).
1357
Shi et al
The relationship of baricitinib PD and concentration was evaluated using a sigmoid Imax/IC50 model with the following equation: I ¼ Imax Cg =ðCg þ IC50 g Þ where I is the pSTAT3 inhibition by baricitinib expressed as percent decrease of pSTAT3 level compared to the baseline value observed at pre-dose (t ¼ 0), C is the observed plasma concentration of baricitinib, Imax is the maximum inhibition (100%) theoretically archived at C ¼ 1, IC50 is the baricitinib concentration at which I ¼ 50%, and g is the Hill Coefficient. The PK–PD curve fitting was performed using Prism 5.0 (GraphPad Software, La Jolla, CA, USA), with the boundary for I set between 0 (at C ¼ 0) and 100% (at C ¼ 1) and all data points were included as observed. To estimate baricitinib cumulative PD effect, the area under the mean pSTAT3 inhibition–time curve from 0 to 24 hours (AUCE0–24h) was calculated at the PK steady-state using NCA with the linear trapezoidal rule. PD data (at later time points) showing negative pSTAT3 inhibition values were assigned a value of zero for AUCE0–24h calculation. The 24-hour average PD effect (Iavg) was defined as AUCE0–24h/24. The NCA portion of PD analysis was performed using WinNonlin version 5.0.1. Statistical Evaluation For the single and multiple dose escalation studies, the dose-proportionality of baricitinib Cmax (maximum concentration) and AUC (area under the concentration– time curve) were evaluated using a power-function regression (e.g., AUC0–t ¼ a Doseb), and the 90% Confidence Interval (CI) of b is calculated. For the food effect study, the log-transformed PK parameters were compared between the treatments (reference ¼
fasted dosing). The adjusted least square geometric mean ratios (GMRs) and the corresponding 90% CI for Cmax and AUC was calculated using SAS Proc Mixed (REML) for a crossover study design with the fixed factor for treatment, sequence, and period and a random factor for subject nested within sequence. All statistical analyses were performed using SAS version 9.1 (SAS Institute, Inc., Cary, NC). PD variables were summarized using descriptive statistics and graphics.
Results Study Population and Safety Analysis The baseline demographics of the study participants are summarized in Table 2. Out of the 27 subjects who participated in Part 1 of Study 1 (the dose escalation phase), 24 (88.9%) subjects completed all planned study procedures. Among the 3 subjects who prematurely discontinued the study, 2 (7.4%) subjects withdrew consent, and 1 (3.7%) subject was withdrawn at the Sponsor’s discretion due to abnormal hemoglobin values although the degree of this abnormality was not considered to be clinically significant by the investigator; this subject received a placebo and a single dose of 5 mg baricitinib in Periods 1 and 2, respectively (see Table 1). Twenty-six treatmentemergent adverse effects (TEAEs) were reported for 17 (63.0%) subjects during the dose escalation phase. There was no increase in the overall incidence of TEAE with increasing dose and the frequency of TEAEs was similar following dosing with placebo compared to the active doses. The most frequently reported TEAEs for all subjects (placebo or baricitinib treatment), which occurred in 2 (7.4%) subjects each during the dose escalation phase, were catheter site hemorrhage at the venipuncture site, back pain, somnolence, and contact
Table 2. The Baseline Demographic Characteristics of the Study Populations Study 1 Part 1 (n ¼ 27) Sex, n (%) Male Female Race, n (%) White Black Asian American Indian or Alaska Native Other Mean age, y (range) Mean weight, kg (range) Mean BMI, kg/m2 (range) BMI ¼ body mass index.
17 (63.0) 10 (37.0) 20 (74.1) 6 (22.2) 1 (3.7) – – 26.8 (19–54) 72.5 (50.9–98.9) 24.5 (19.2–30.5)
Study 2 Part 2 (n ¼ 26)
24 (92.3) 2 (7.7) 17 (65.4) 8 (30.8) 1 (3.8) – – 28.2 (19–46) 77.8 (59.1–102.1) 24.8 (19.7–29.6)
(n ¼ 93)
61 (65.6) 32 (34.4) 72 (77.4) 15 (16.1) – 1 (1.1) 5 (5.4) 30.6 (18–55) 76.8 (51.1–106.6) 25.5 (18.2–32.0)
1358 dermatitis. Two AEs (both mild in severity) considered possibly related to treatment by the Investigator were reported for 2 (7.4%) subjects, including 1 event of somnolence for one subject following placebo treatment and one subject following the 5 mg fed treatment. Out of the 26 subjects who participated in Part 2 of Study 1 (the food effect phase), 24 (92.3%) subjects completed all planned study procedures. Two (7.7%) subjects prematurely discontinued the study: 1 (3.8%) subject did not complete the study due to loss of contact, and 1 (3.8%) subject was withdrawn due to a low grade elevation of ALT which was not considered an AE by the investigator. Six TEAEs were reported for 6 subjects (23.2%) during the food-effect phase. AEs of constipation, upper respiratory infection, chlamydia urethritis, decreased appetite, headache, and contact dermatitis were reported for 1 subject each. Three AEs assessed by the Investigator as possibly related to treatment (all mild in severity) were reported for 3 (11.5%) subjects, including 1 AE each of constipation, headache, and decreased appetite. Out of the 93 subjects who participated in Study 2, 79 (84.9%) subjects completed all study procedures. Among the 14 (15.1%) subjects who prematurely discontinued the study, 9 (9.7%) subjects withdrew consent, 1 (1.1%) subject was discontinued by the Investigator for cause unrelated to the study, and 4 (4.3%) subjects were withdrawn due to AEs (1 AE each subject). Of the 4 AEs, 2 AEs were assessed to be unlikely related to the study drug and 2 AEs (elevated bilirubin and flu-like symptom, both mild in severity) were assessed to be possibly related to the study drug. One subject with a history of migraine headaches discontinued the study due to a serious AE of complicated migraine (severe and judged unlikely related to study medication) approximately 2 days after receiving a single dose of 20 mg INCB028050. No trends or clinically relevant changes were noted in mean clinical laboratory, vital sign, or ECG data following dosing across all cohorts, with the exception of an expected decrease in mean absolute neutrophil count (ANC) after dosing that reached a nadir approximately 8 hours after dosing then returned to predose levels by approximately 16–24 hours postdose, consistent with a neutrophil margination effect.9 Similar but less marked changes were seen in white blood cell count (WBC). An expected dose-dependent decrease in absolute reticulocyte count (ARC) was apparent with some evidence of compensation in the 10 mg QD dose group dosed for 28 days (maximal mean decrease of 22% in ARC on Day 8, with a median decrease of 4% on Day 28 for the 10 mg QD group). Thirteen AEs of neutropenia were reported for 12 subjects (10 events assessed as mild in severity and 3 events assessed as moderate in severity); all were resolved at study conclusion with several resolving during ongoing dosing and resulting in few dose holds.
The Journal of Clinical Pharmacology / Vol 54 No 12 (2014)
To summarize the safety analysis for the two clinical studies, the study drug was generally safe and well tolerated with no deaths or no treatment related serious adverse event (SAE) reported. Pharmacokinetic Profiles The PK profiles of baricitinib were comparable between the single and multiple dose administrations, and shown in Figure 1 are the steady-state PK profiles following the last dose of the 10-day QD dosing regimens (the profile from 28-day dosing cohort was similar and not shown). Following fasting, oral dose administration, baricitinib was rapidly absorbed, typically attaining the peak concentrations (Cmax) in plasma within 1.5 hours postdose, and subsequently, baricitinib plasma concentrations declined in an apparent bi-exponential fashion, with a geometric mean terminal-phase disposition t1/2 of about 8 hours in the healthy subjects. Based on baricitinib trough plasma concentration time course (data not shown), the PK steady-state was attained with multiple dosing within 48 hours (or 6x half-life) after the first dose. There were no significant differences in observed PK parameters on Day 10 (or Day 28) versus Day 1 for all regimens. The average accumulation index (GMR) for Cmax and AUC0–24h was 1.08 and 1.13, respectively, for the QD dosing regimen (steady-state vs. first dose). There was no evidence to suggest that baricitinib pharmacokinetics was time-dependent throughout the repeat dosing. The observed minimal systemic drug accumulation was consistent with the estimated disposition t1/2. Following the oral administrations, the variability observed in baricitinib plasma exposures was moderate. For Cmax, the overall intersubject percent coefficient of variation (%CV) was 35.3% and 34.6%, respectively, in Study 1/Part 1 (ascending single dose study) and Study 2 (the QD cohorts), and the overall intrasubject %CV was 14.4% in Study 1/Part 1. For AUC0–1, the overall
Figure 1. Baricitinib steady-state mean plasma concentrations in healthy subjects following the last dose in the 10-Day dosing regimen. Error bars represent standard deviations.
1359
Shi et al
decreased the geometric mean Cmax by approximately 29%, the changes were thought unlikely to have any clinical significance.
intersubject and intrasubject %CV was 29.6% and 4.4%, respectively, in Study 1/Part 1. For steady-state AUC0–24h following repeat QD dosing, the overall intersubject %CV was 20.2% in Study 2.
Pharmacodynamics Baricitinib demonstrated dose and time-dependent inhibition of cytokine-induced pSTAT3, with maximal inhibition of pSTAT3 occurring 1–2 hours after administration for all doses, coincident with the observed Cmax. Levels of pSTAT3 generally returned to predose control baseline by 16–24 hours in all single dose and QD dose cohorts examined. Similar magnitudes of pSTAT3 inhibition were observed using either IL-6 or TPO as the cytokine stimulus. The extent of pSTAT3 inhibition observed in fed versus fasted individuals was similar although the maximal inhibition occurred at a slightly later time point (3–4 hours post dose) following administration with the meal, consistent with delayed absorption. The PD profile of baricitinib was similar following the single and multiple dose administrations with no evidence of hysteresis, and therefore the data were pooled from both the studies for analysis of relationship between baricitinib PD and plasma concentration. A sigmoid Imax/IC50 model fitted to the combined PK–PD data (Figure 2) indicates that cytokine-induced pSTAT3 can be inhibited by baricitinib with an ex vivo IC50 value of 90 nM (95% CI: 84–97 nM) and a Hill Coefficient (g) of 0.87 (95% CI: 0.81–0.94). WBC and differential, including absolute ANC and absolute lymphocyte count (ALC), was also serially determined following dosing in both the single dose and repeat dose studies with similar results. ANC decreased in a dose-related manner reaching a nadir at 8 hours following dosing and returning to baseline 12–24 hours post-dose. Similar changes were seen in the multiple dose study with a return to baseline ANC within 24 hours following the final dose of study medication on Day 10 or Day 28 of repeat dosing (data not shown). In contrast ALC tended to increase peaking 6 hours after dosing and returning toward normal by 24 hours after dosing (Figure 3).
Dose Proportionality In both the single and multiple dose escalation studies, baricitinib exhibited linear PK characteristics with the plasma Cmax and AUC values rising proportionally to dose. Over the 1–10 mg single-dose range, the powerfunction regression analysis produced dose-proportionality equations of Cmax ¼ 26.8 Dose0.991 (90% CI of b: 0.907–1.08) and AUC0–1¼175 Dose0.998 (90% CI of b: 0.928–1.07), and over the 2–20 mg QD dose range, the dose-proportionality equations were Cmax ¼ 25.8 Dose0.968 (90% CI of b: 0.861–1.08) and AUC0– 0.926 (90% CI of b: 0.852–1.00). In all 24h ¼ 178 Dose the above equations, the 90% CI’s for the exponent, b, of the power function (or equivalently the slope of the logtransformed equation) included the value of 1, suggesting that the baricitinib Cmax and AUC were approximately proportionally related to the dose. Renal Clearance At the steady-state following 10-day QD dosing, the average oral-dose clearance (CL/F) of baricitinib was 17.8 L/h for the 5 cohorts dosed for 10 days and 17.0 L/h for all cohorts. For the QD dosing cohorts, the average percent dose excreted over a 24 hours period (on Day 10) as unchanged parent compound was 64.1% (range: 60.3– 65.6%). The average renal clearance (CLR) of 11.8 L/h (range: 10.5–13.0 L/h) represented approximately 2/3 of the total oral-dose clearance (Table 3). Food Effect Co-administration of a high-fat, high-calorie meal did not alter baricitinib AUC (Table 4). The GMR of baricitinib AUC0–1 comparing the fed and fasted administrations was approximately 99.7%, with the 90% CI of 96–103%, meeting the bioequivalence criterion. Although the highfat meal delayed the median Tmax by 3 hours and
Table 3. Baricitinib Steady-State PK Parameters Following QD or BID Dosing for 10 Days Dose
n
Cmax (nM)
2 mg QD 5 mg QD 10 mg QD 20 mg QD 5 mg BID Average
8 8 8 8 8
45.7 136 206 415 146
(16.7) (27.1) (30.4) (22.1) (16.3) –
Tmax (h) 1.5 1.2 1.0 1.2 1.5
(1.0–2.0) (0.5–2.0) (0.5–4.0) (0.5–2.0) (0.5–3.0) –
t1/2 (h) 8.4 7.4 9.1 6.8 11
(21.1) (17.5) (25.9) (18.3) (26.9) 8.5
AUC0–t (nM*h) 312 831 1460 2490 867
(19.7) (17.4) (17.5) (23.2) (17.6) –
CL/F (L/h)
CLR (L/h)
17.3 (21.5) 16.2 (18.3) 18.4 (14.2) 21.6 (22.0) 15.5 (18.1) 17.8
11.3 (21.9) 10.5 (27.2) 12.1 (24.4) 13.0 (19.8) 12.1 (42.1) 11.8
PK parameters are reported as geometric mean (%CV) value, expect that Tmax is reported as median (range) value; Average PK value is the arithmetic mean of all cohorts. Cmax, peak plasma concentration; Tmax, time to Cmax; t1/2, elimination half-time in the terminal disposition phase; AUC0–t, area under concentrationtime curve within one dosing interval; CL/F, oral-dose plasma clearance; CLR, renal clearance.
1360
The Journal of Clinical Pharmacology / Vol 54 No 12 (2014)
Table 4. The Food Effect on Baricitinib PK Parameters Treatment
n
Fasted Fed
12 11
Cmax (nM) 127 (24.0) 90.3 (51.3)
Tmax (h) 1.0 (0.50–1.5) 4.0 (0.50–6.0)
AUC0–t (nM*h)
AUC0–1 (nM*h)
718 (24.3) 713 (24.8)
730 (24.5) 722 (24.9)
Adjusted Geometric Mean Ratios and 90% Confidence Intervals (Reference = Fasted)* 0.716 (0.572, 0.897)
1.00 (0.971, 1.03)
0.997 (0.964, 1.03)
PK parameters values are geometric means (%CV). Cmax, peak plasma concentration; Tmax, time to Cmax; AUC0–t, area under concentration–time curve estimated up to the last observed concentration time point; AUC0–1, area under concentration–time curve extrapolated to infinity. * Statistical analysis was based on data from the 11 subjects who completed the study through both periods.
Similar changes were seen in the 10-day and 28-day repeat dose studies with transient decreases in ANC and increases in ALC that reversed by 24 hours after each dose, with the exception of the BID regimens where there was not sufficient time after each dose for these parameters to return to baseline. There was no evidence to indicate the changes in ANC or ALC were a consequence of myelosuppression.
Discussion Two FIH studies of baricitinib were conducted in healthy volunteers to evaluate the PK, PD, and safety profiles following single and multiple oral dose administrations. The safety analysis showed that following the single dose administrations of 1–20 mg, the overall incidence of TEAE did not increase with increasing doses and was similarly compared to the placebo groups. Following multiple dose administrations of 2–20 mg up to 10 or 28 days, baricitinib was generally well tolerated with no SAE reported in the active dose groups (the only one SAE of migraine reported was assessed by the Investigator to be
Figure 2. The PK–PD relationship established for baricitinib using ex vivo data (n ¼ 782, R2 ¼ 0.66). The PD marker is inhibition of phosphorylated STAT3 (pSTAT3) in whole blood. The solid line and dashed lines are the best-fit curve and 95% predictive bands, respectively, using a sigmoid Imax/IC50 model, which estimated Hill Coefficient and IC50 values to be 0.87 and 90 nM, respectively.
Figure 3. Percent change in mean absolute neutrophil count (ANC, Graph A) and absolute lymphocyte count (ALC, Graph B) following single doses of baricitinib in healthy subjects. Error bars represent standard error of mean.
1361
Shi et al
unlikely related to baricitinib). As dose limiting toxicity was not observed, the maximum tolerated dose (MTD) was not determined in the two studies conducted. Baricitinib appeared to be safe and well tolerated with an overall safety profile supporting its further clinical development. The transient decrease in ANC seen with other JAK inhibitors was confirmed with baricitinib. This appeared to be linearly related to dose, and is unlikely to represent bone marrow suppression given the rapid return to baseline following dosing for up to 28 days. There was also a transient increase in ALC, which has been seen for other JAK inhibitors not as well documented (Incyte data on file). It is also of note that the decreases in reticulocyte counts seen appeared to partially reverse in the 10 mg QD dose group which was dosed for 28 days. This suggests that with prolonged daily dosing of baricitinib, tachyphylaxis may occur for the effects seen on hematopoiesis. Baricitinib PK profile following multiple-dose administration was similar to that following single-dose administration, and the modest increase in plasma exposures observed was consistent with linear systemic drug accumulation. There was no evidence in these studies to suggest baricitinib induced its own clearance or saturated its elimination pathway (mainly renal excretion). Of note, the mean terminal phase t1/2 of approximate 8 hours was estimated in healthy subjects using a noncompartmental PK method. It is anticipated that the value of t1/2 may increase in disease states associated with reduced renal function. The value may also change when a compartmental PK analysis method is used and the terminal phase is better characterized. Consistent with the PK data, the PD profiles (using pSTAT3 as the PD biomarker) showed the pharmacological activity of baricitinib was maintained upon repeat dosing. The observed PK and PD response appeared to be directly linked and there was no evidence to suggest any PD hysteresis. The PD IC50 value from analyzing the ex vivo whole blood samples was comparable to that determined in vitro.
Conclusion An immediate release formulation of baricitinib, a new and potent JAK1/2 inhibitor, was studied in two Phase 1 studies. The investigational drug demonstrated a favorable pharmacokinetic and pharmacodynamics profile suitable for once-daily administration. Baricitinib exhibited linear
and time-independent PK, and its pharmacological activity was maintained with repeat dosing. All doses of baricitinib were safe and well tolerated, and the overall safety profile of this potent JAK1/2 inhibitor supports its further clinical development.
Conflict of Interest Baricitinib is being co-developed by Eli Lilly and Company and Incyte Corporation, and all authors of this manuscript are employees of Incyte Corporation.
Author Contributions JGS conducted PK/PD data analysis. JGS, WVW, and SY prepared the manuscript. WVW, NGP, EGM, XC, and SY designed study protocol and/or conducted safety data analysis. FL and TE performed GLP bioanalysis for baricitinib concentrations in plasma and urine samples. YL and PAS performed ex vivo pSTAT3 analysis. References 1. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov. 2003;2:473–488. 2. Klareskog L, Catrina AI, Paget S. Rheumatoid arthritis. Lancet. 2009;373:659–672. 3. Rindfleisch JA, Muller D. Diagnosis and management of rheumatoid arthritis. Am Fam Physician. 2005;72:1037–1047. 4. Ghoreschi K, Laurence A, O’Shea JJ. Janus kinases in immune cell signaling. Immunol Rev. 2009;228:273–287. 5. Kremer JM, Cohen S, Wilkinson BE, et al. A Phase IIb dose-ranging study of the oral JAK inhibitor tofacitinib (CP-690,550) versus placebo in combination with background methotrexate in patients with active rheumatoid arthritis and an inadequate response to methotrexate alone. Arthritis Rheum. 2012;64:970–981. 6. Fridman JS, Scherle PA, Collins R, et al. Selective inhibition of JAK1 and JAK2 is efficacious in rodent models of arthritis: preclinical characterization of INCB028050. J Immunol. 2010;184:5298–5307. 7. Greenwald M, Fidelus-Gort RK, Levy R, et al. A randomized, doseranging, placebo-controlled study of INCB028050, a selective JAK1 and JAK2 inhibitor, in subjects with active rheumatoid arthritis. Presentation at the annual scientific meeting of American College of Rheumatology (ACR), Atlanta, GA, USA, Nov. 6–11, 2010. 8. Keystone E, Taylor P, Genovese M, et al. 12-Week Results of a Phase IIb Dose-Ranging Study of LY3009104 (INCB028050), an Oral JAK1/JAK2 Inhibitor, in Combination with Traditional DMARDs in Patients with Rheumatoid Arthritis. Late-breaking oral presentation (Abstract LB0005). European League against Rheumatism's (EULAR) Annual European Congress of Rheumatology, Berlin, Germany, June 8, 2012. 9. Shi JG, Chen X, McGee RF, et al. The pharmacokinetics, pharmacodynamics, and safety of orally dosed INCB018424 phosphate in healthy volunteers. J Clin Pharmacol. 2011;51:1644–1164.
The Pharmacokinetics, Pharmacodynamics, and Safety of Baricitinib, an Oral JAK 1/2 Inhibitor, in Healthy Volunteers
The Journal of Clinical Pharmacology 54(12) 1354–1361 © 2014, The American College of Clinical Pharmacology DOI: 10.1002/jcph.354
Jack G. Shi, PhD, Xuejun Chen, PhD, Fiona Lee, MS, Thomas Emm, BS, Peggy A. Scherle, PhD, Yvonne Lo, MS, Naresh Punwani, PhD, William V. Williams, MD, and Swamy Yeleswaram, PhD
Abstract Baricitinib (also known as LY3009104 or INCB028050), a novel and potent small molecule inhibitor of Janus kinase family of enzymes (JAKs) with selectivity for JAK1 and JAK2, is currently in clinical development for the treatment of rheumatoid arthritis (RA) and other inflammatory disorders. Two double-blind, randomized, and placebo-controlled studies were conducted to evaluate single ascending doses of 1–20 mg and multiple ascending doses of 2–20 mg QD and 5 mg BID for 10 or 28 days in healthy volunteers. Following oral administration, baricitinib plasma concentration typically attains its peak value within 1.5 hours postdose and subsequently declines in a bi-exponential fashion. Baricitinib demonstrates dose-linear and timeinvariant pharmacokinetics, with low oral-dose clearance (17 L/h) and minimal systemic accumulation observed following repeat dosing. The mean renal clearance of baricitinib was determined to be 12 L/h. [Correction added after publication 12 November 2014: in the preceding sentence, “2 L/h” was changed to “12 L/h.”] The effect of a high-fat meal on baricitinib pharmacokinetics was insignificant. The pharmacodynamics of baricitinib, evaluated by the inhibition of STAT3 phosphorylation following cytokine stimulation in the whole blood ex vivo, was well correlated with baricitinib plasma concentrations. Baricitinib was generally safe and well tolerated, with no serious treatment-related adverse events (AEs) reported from either of the studies. An expected rapidly reversible, dose-related decline in absolute neutrophil count was seen with baricitinib.
Keywords baricitinib, JAK, rheumatoid arthritis, pharmacokinetics, pharmacodynamics
Rheumatoid arthritis (RA) is a systemic, autoimmune disease that affects approximately one percent of the population.1 Active RA is characterized by chronic joint inflammation that can result in progressive destruction of cartilage and juxtarticular bone leading to functional impairment and disability.2 Conventional RA therapies consist of treatments with non-steroidal anti-inflammatory drugs (NSAIDs), corticosteroids, and disease-modifying anti-rheumatic drugs (DMARDs). The emergence of biologic therapies for RA, such as tumor necrosis factor-a (TNFa) inhibitors, represents treatment options that mitigate the clinical manifestations of RA by selectively targeting a key inflammatory molecule. However, approximately 30% of patients with RA, usually those with the more severe disease, will not experience satisfactory clinical improvement with use of currently available therapies.3 Therefore, a significant unmet medical need remains for more effective and better tolerated treatments for RA. Chronic inflammatory conditions, including RA, are associated with aberrant production of cytokines and growth factors, and the Janus kinase family of protein tyrosine kinases (JAKs) plays an important role in the signaling of a number of cytokines and hematopoietic growth factors.4 Many of the pro-inflammatory cytokines
implicated in the pathogenesis of RA, including IL-6 and IFN-g, utilize cell signaling that involves the JAK/signal transducers and activators of transcription (STAT) pathways. Inhibition of JAK-STAT signaling can target multiple RA-associated cytokine pathways and thereby reduce inflammation, cellular activation, and proliferation of key immune cells as demonstrated in studies of tofacitinib, a JAK inhibitor that has shown clinical efficacy in patients with RA.5 Baricitinib (LY3009104 also previously known as INCB028050) is a potent and selective small molecule inhibitor of JAK1/2 enzymes. In vitro, baricitinib inhibits JAK1 and JAK2 with IC50 values in the single-digit nM range, yet it does not significantly inhibit a diverse panel
From Incyte Corporation, Wilmington, DE, USA, Submitted for publication 12 May 2014; accepted 23 June 2014. [Correction added on 17 November 2014 after publication 12 November 2014; a typographical error was corrected in the Abstract section.] Corresponding Author: Jack G. Shi, PhD, Experimental Station, Building E400, Rt. 141 & Henry Clay Road, Wilmington, DE 19880, USA Email: [email protected]
1355
Shi et al
of 28 kinases when tested at 100-fold of the IC50 against JAK1 and JAK2.6 In multiple murine disease models that are relevant to arthritis, baricitinib showed excellent potential of antirheumatic efficacy without evidence of suppression of humoral immunity or adverse hematologic effects.6 In a Phase IIa proof-of-concept study conducted in RA patients, baricitinib demonstrated rapid and sustained efficacy in patients with inadequate responses to DMARDs across the three doses (4, 7, and 10 mg QD) evaluated over a 24-week treatment period.7 All doses were generally safe and well tolerated, and the adverse events (AEs) were predominantly mild to moderate with frequencies similar to placebo. Subsequently, in a Phase IIb dose-ranging study conducted in RA patients on stable background of methotrexate, baricitinib achieved the primary endpoint by demonstrating a statistically significant difference (P < 0.001) in the American College of Rheumatology 20 (ACR20) response for the 4 mg QD and 8 mg QD dose groups compared to placebo after 12-weeks of treatment.8 This report summarizes the findings of the first-inhuman (FIH) and single ascending dose (SAD) and multiple ascending dose (MAD) studies of baricitinib, which were conducted to evaluate the safety, tolerability, pharmacokinetic, and pharmacodynamic profiles of baricitinib in healthy adult volunteers.
Methods Two randomized, double-blind (to the subjects and investigator), and placebo-controlled clinical studies were conducted in full accordance with the Declaration of Helsinki, principals of good clinical practices (GCP), and local laws regarding the protection of the rights and welfare of human participants in biomedical research. The protocols were approved by an independent institutional review board (IRB), and informed consent for all participants was obtained prior to screening. An immediate release capsule formulation of baricitinib phosphate was used in both studies, and all doses
reported herein equivalents.
were
the
baricitinib
free
base
Study Population Males and females, 18–55 years of age, with a body mass index (BMI) of between 18 and 32 kg/m2 were eligible for participation in the studies if they were judged to be in good health based on their medical history and physical examinations including vital signs, electrocardiogram, and clinical laboratory test results. Women of childbearing potential enrolled in these studies were determined to be in a nongravid state and had agreed to take appropriate precautions (with at least 99% certainty) to avoid pregnancy from screening through follow-up. Subjects with hemoglobin levels and white blood and platelet counts below the lower reference limit for the parameters were excluded. Non-study medications were not allowed for 7 and 14 days, for over-the-counter and prescription drugs, respectively, prior to the first dose of study medication until the completion of the studies, unless deemed necessary and acceptable by the investigator. Use of investigational medications and any medications known to affect cytochrome P450 enzymes or Pglycoprotein activity was prohibited within 30 days or five half-lives (whichever was longer) prior to the study start and throughout the study. Study Design Study 1 was composed of 2 parts, and the detailed dosing paradigm is described in Table 1. Part 1 was a 3-period, ascending single-dose tolerability and PK/PD study with a planned enrollment of 24 subjects. Participants in Cohort 1 (n ¼ 12) received single oral doses of 1, 5, and 10 mg baricitinib and matching placebo in the fasted state, while participants in Cohort 2 (n ¼ 12) received single oral doses of 2 and 10 mg baricitinib in the fasted state, or 5 mg baricitinib following a high fat meal, and matching placebo. For each treatment/period, participants were randomized to receive the active treatment or placebo at a 2:1 ratio. A washout period of at least 10 days was
Table 1. Design of the Dosing Paradigm in Study 1
Cohort 1 1 1 2 2 2 3 3 N/A ¼ not applicable.
Sequence
Period 1
Period 2
Period 3
No. of subjects planned for enrollment
1 2 3 4 5 6 7 8
placebo 1 mg 1 mg placebo 2 mg 2 mg 5 mg fed 5 mg fasted
5 mg placebo 5 mg 10 mg placebo 10 mg 5 mg fasted 5 mg fed
10 mg 10 mg placebo 5 mg fed 5 mg fed placebo N/A N/A
4 4 4 4 4 4 12 12
1356 instituted between the treatment periods. Part 2 of the study with a planned enrollment of 24 subjects commenced upon completion of Part 1 to evaluate the effect of food on baricitinib single-dose PK. Using a 2-sequence, 2-period, cross-over design, the subjects were randomized with a 1:1 ratio to receive 5 mg baricitinib either fasted or after a standardized high-fat, high calorie meal. Study 2 was a sequential, 1-period, ascending multiple-dose tolerability and PK/PD study. Sequentially, higher multiple-doses of 2 mg QD, 5 mg QD, 10 mg QD, 5 mg BID, and 20 mg QD of baricitinib (after testing the safety of a single dose of 20 mg) were administered in the fasted state for 10 days; in addition, 10 mg QD and 5 mg BID were studied for 28 days in two additional cohorts. There were altogether 7 cohorts (5 cohorts for 10-day dosing, n ¼ 12 per cohort and 2 cohorts for 28-day dosing, n ¼ 16 per cohort) as indicated above. In each cohort, subjects were randomized to receive either baricitinib or placebo at a 3:1 ratio. In both studies, the AM dosing in fasted state was administered after at least a 10 hours overnight fast, and all PM dosing (Study 2 only) was preceded by at least a 3 hours fast from food. PK blood samples were collected following the single-dose administrations, the first dose and the last dose of multiple-dose administrations at 0, 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 24, 36, and 48 hours post-dose (in the multiple-dose study, the last PK samples collected following the first doses were at 12 and 24 hours, respectively, for the q12h and q24h administrations, immediately prior to the next dosing). PD blood samples were collected at 0, 1, 2, 4, 6, 12, and 24 hours post-dose in Study 1, and at 0, 2, 6 hours on Day 10 in Study 2. Additional trough (pre-dose) blood samples were collected in Study 2 before the AM dose on Days 2, 3, 5, 7, and 9 for the 10-day cohorts, and Days 2, 8, 15, and 22 for the 28-day cohort. A pre-dose urine sample and complete urine output from 0–24 hours post-dose were collected on Day 10 (for the 10-day cohorts only) for determination of urine concentrations of baricitinib. In both studies, safety and tolerability were assessed by monitoring AEs, measuring vital signs and ECGs, and clinical laboratory blood and urine sample assessments. In addition, serial measurements of white blood cells (WBCs) with differential counts were carried out predose and 4, 8, 12, 16, and 24 hours post-dose for each dose in Study 1, on Days 1 and 10 for the 10-day dosing cohorts in Study 2, and on Days 1 and 28 for the 28-day dosing cohorts in Study 2. Analytical Methods The PK samples were assayed by a validated, GLP, LC/ MS/MS method at Incyte Corporation (Wilmington, DE, USA) to determine baricitinib concentrations in plasma and urine. The plasma and urine standard curves ranged from 0.5 to 500 nM and 0.025 to 25 mM, respectively. The
The Journal of Clinical Pharmacology / Vol 54 No 12 (2014)
plasma samples were prepared using liquid/liquid extraction with methyl-t-butyl ether, and the urine samples directly diluted with acetonitrile, both with the 13 C4-labeled baricitinib (M þ 4) as an internal standard. Chromatographic separations were performed over a Waters Atlantis HILIC Silica column (50 2.1 mm, 3 mm) (Waters corporation, Milford, MA, USA) under isocratic condition using a mobile phase consisting of 97% acetonitrile and 3% 2 mM ammonium formate aqueous solution (pH ¼ 3). MS/MS analysis was performed using a positive Turbo IonSpray interface on a Sciex API-3000 or API-4000 mass spectrometer (Applied Biosystems, Foster City, CA, USA) operating in multiplereaction monitoring mode, monitoring the transitions of m/z 372.1 ! 251.2 for baricitinib, and m/z 376.1 ! 255.2 for the internal standard. For the plasma assay, intra-assay precision (CV %) and accuracy (Bias %) for quality control samples ranged from 2.6 to 10.5% and 13.5 to 5.0% respectively, while inter-assay precision and accuracy ranged from 4.5 to 12.5% and 5.9 to 1.7%, respectively. For the urine assay, intra-assay precision and accuracy for quality control samples ranged from 2.1 to 12.2% and 11.5 to 9.0% respectively, while inter-assay precision and accuracy ranged from 7.5 to 12.1% and 3.9 to 0.4%, respectively. The PD samples (ex vivo whole blood) were stimulated with interleukin-6 (IL-6) or thrombopoietin (TPO) to activate the JAK/STAT pathway, the blood cells lysed, and the total cell extracts were analyzed for levels of phosphorylated STAT3 (pSTAT3) using a specific enzyme-linked immunosorbent assay (ELISA). This assay was performed under a specific standard operating procedure (SOP). The details for the assay procedure and validation results have been described elsewhere.6,9 For each sample analyzed, duplicate PD analyses were performed and the average value was reported. For each subject, the percent inhibition of pSTAT3 was calculated by comparing pre-dose values obtained before the first dose with values obtained at different times after dosing. Pharmacokinetic and Pharmacodynamic Analysis Per study protocol, scheduled times were used for PK analysis, except when the difference between the actual time and nominal time was to be greater than 5 minutes for samples collected up to 4 hours after dose and greater than 15 minutes for samples collected more than 4 hours after dose; in those cases, actual time were used for the PK analysis. The vast majority of samples were collected on the scheduled times with only a few samples slightly outside of the pre-defined windows (corrected to real collection times). Standard non-compartmental PK analysis (NCA) methods were used to analyze the baricitinib plasma concentration data using WinNonlin version 5.0.1 (Pharsight Corporation, Mountain View, CA, USA).
1357
Shi et al
The relationship of baricitinib PD and concentration was evaluated using a sigmoid Imax/IC50 model with the following equation: I ¼ Imax Cg =ðCg þ IC50 g Þ where I is the pSTAT3 inhibition by baricitinib expressed as percent decrease of pSTAT3 level compared to the baseline value observed at pre-dose (t ¼ 0), C is the observed plasma concentration of baricitinib, Imax is the maximum inhibition (100%) theoretically archived at C ¼ 1, IC50 is the baricitinib concentration at which I ¼ 50%, and g is the Hill Coefficient. The PK–PD curve fitting was performed using Prism 5.0 (GraphPad Software, La Jolla, CA, USA), with the boundary for I set between 0 (at C ¼ 0) and 100% (at C ¼ 1) and all data points were included as observed. To estimate baricitinib cumulative PD effect, the area under the mean pSTAT3 inhibition–time curve from 0 to 24 hours (AUCE0–24h) was calculated at the PK steady-state using NCA with the linear trapezoidal rule. PD data (at later time points) showing negative pSTAT3 inhibition values were assigned a value of zero for AUCE0–24h calculation. The 24-hour average PD effect (Iavg) was defined as AUCE0–24h/24. The NCA portion of PD analysis was performed using WinNonlin version 5.0.1. Statistical Evaluation For the single and multiple dose escalation studies, the dose-proportionality of baricitinib Cmax (maximum concentration) and AUC (area under the concentration– time curve) were evaluated using a power-function regression (e.g., AUC0–t ¼ a Doseb), and the 90% Confidence Interval (CI) of b is calculated. For the food effect study, the log-transformed PK parameters were compared between the treatments (reference ¼
fasted dosing). The adjusted least square geometric mean ratios (GMRs) and the corresponding 90% CI for Cmax and AUC was calculated using SAS Proc Mixed (REML) for a crossover study design with the fixed factor for treatment, sequence, and period and a random factor for subject nested within sequence. All statistical analyses were performed using SAS version 9.1 (SAS Institute, Inc., Cary, NC). PD variables were summarized using descriptive statistics and graphics.
Results Study Population and Safety Analysis The baseline demographics of the study participants are summarized in Table 2. Out of the 27 subjects who participated in Part 1 of Study 1 (the dose escalation phase), 24 (88.9%) subjects completed all planned study procedures. Among the 3 subjects who prematurely discontinued the study, 2 (7.4%) subjects withdrew consent, and 1 (3.7%) subject was withdrawn at the Sponsor’s discretion due to abnormal hemoglobin values although the degree of this abnormality was not considered to be clinically significant by the investigator; this subject received a placebo and a single dose of 5 mg baricitinib in Periods 1 and 2, respectively (see Table 1). Twenty-six treatmentemergent adverse effects (TEAEs) were reported for 17 (63.0%) subjects during the dose escalation phase. There was no increase in the overall incidence of TEAE with increasing dose and the frequency of TEAEs was similar following dosing with placebo compared to the active doses. The most frequently reported TEAEs for all subjects (placebo or baricitinib treatment), which occurred in 2 (7.4%) subjects each during the dose escalation phase, were catheter site hemorrhage at the venipuncture site, back pain, somnolence, and contact
Table 2. The Baseline Demographic Characteristics of the Study Populations Study 1 Part 1 (n ¼ 27) Sex, n (%) Male Female Race, n (%) White Black Asian American Indian or Alaska Native Other Mean age, y (range) Mean weight, kg (range) Mean BMI, kg/m2 (range) BMI ¼ body mass index.
17 (63.0) 10 (37.0) 20 (74.1) 6 (22.2) 1 (3.7) – – 26.8 (19–54) 72.5 (50.9–98.9) 24.5 (19.2–30.5)
Study 2 Part 2 (n ¼ 26)
24 (92.3) 2 (7.7) 17 (65.4) 8 (30.8) 1 (3.8) – – 28.2 (19–46) 77.8 (59.1–102.1) 24.8 (19.7–29.6)
(n ¼ 93)
61 (65.6) 32 (34.4) 72 (77.4) 15 (16.1) – 1 (1.1) 5 (5.4) 30.6 (18–55) 76.8 (51.1–106.6) 25.5 (18.2–32.0)
1358 dermatitis. Two AEs (both mild in severity) considered possibly related to treatment by the Investigator were reported for 2 (7.4%) subjects, including 1 event of somnolence for one subject following placebo treatment and one subject following the 5 mg fed treatment. Out of the 26 subjects who participated in Part 2 of Study 1 (the food effect phase), 24 (92.3%) subjects completed all planned study procedures. Two (7.7%) subjects prematurely discontinued the study: 1 (3.8%) subject did not complete the study due to loss of contact, and 1 (3.8%) subject was withdrawn due to a low grade elevation of ALT which was not considered an AE by the investigator. Six TEAEs were reported for 6 subjects (23.2%) during the food-effect phase. AEs of constipation, upper respiratory infection, chlamydia urethritis, decreased appetite, headache, and contact dermatitis were reported for 1 subject each. Three AEs assessed by the Investigator as possibly related to treatment (all mild in severity) were reported for 3 (11.5%) subjects, including 1 AE each of constipation, headache, and decreased appetite. Out of the 93 subjects who participated in Study 2, 79 (84.9%) subjects completed all study procedures. Among the 14 (15.1%) subjects who prematurely discontinued the study, 9 (9.7%) subjects withdrew consent, 1 (1.1%) subject was discontinued by the Investigator for cause unrelated to the study, and 4 (4.3%) subjects were withdrawn due to AEs (1 AE each subject). Of the 4 AEs, 2 AEs were assessed to be unlikely related to the study drug and 2 AEs (elevated bilirubin and flu-like symptom, both mild in severity) were assessed to be possibly related to the study drug. One subject with a history of migraine headaches discontinued the study due to a serious AE of complicated migraine (severe and judged unlikely related to study medication) approximately 2 days after receiving a single dose of 20 mg INCB028050. No trends or clinically relevant changes were noted in mean clinical laboratory, vital sign, or ECG data following dosing across all cohorts, with the exception of an expected decrease in mean absolute neutrophil count (ANC) after dosing that reached a nadir approximately 8 hours after dosing then returned to predose levels by approximately 16–24 hours postdose, consistent with a neutrophil margination effect.9 Similar but less marked changes were seen in white blood cell count (WBC). An expected dose-dependent decrease in absolute reticulocyte count (ARC) was apparent with some evidence of compensation in the 10 mg QD dose group dosed for 28 days (maximal mean decrease of 22% in ARC on Day 8, with a median decrease of 4% on Day 28 for the 10 mg QD group). Thirteen AEs of neutropenia were reported for 12 subjects (10 events assessed as mild in severity and 3 events assessed as moderate in severity); all were resolved at study conclusion with several resolving during ongoing dosing and resulting in few dose holds.
The Journal of Clinical Pharmacology / Vol 54 No 12 (2014)
To summarize the safety analysis for the two clinical studies, the study drug was generally safe and well tolerated with no deaths or no treatment related serious adverse event (SAE) reported. Pharmacokinetic Profiles The PK profiles of baricitinib were comparable between the single and multiple dose administrations, and shown in Figure 1 are the steady-state PK profiles following the last dose of the 10-day QD dosing regimens (the profile from 28-day dosing cohort was similar and not shown). Following fasting, oral dose administration, baricitinib was rapidly absorbed, typically attaining the peak concentrations (Cmax) in plasma within 1.5 hours postdose, and subsequently, baricitinib plasma concentrations declined in an apparent bi-exponential fashion, with a geometric mean terminal-phase disposition t1/2 of about 8 hours in the healthy subjects. Based on baricitinib trough plasma concentration time course (data not shown), the PK steady-state was attained with multiple dosing within 48 hours (or 6x half-life) after the first dose. There were no significant differences in observed PK parameters on Day 10 (or Day 28) versus Day 1 for all regimens. The average accumulation index (GMR) for Cmax and AUC0–24h was 1.08 and 1.13, respectively, for the QD dosing regimen (steady-state vs. first dose). There was no evidence to suggest that baricitinib pharmacokinetics was time-dependent throughout the repeat dosing. The observed minimal systemic drug accumulation was consistent with the estimated disposition t1/2. Following the oral administrations, the variability observed in baricitinib plasma exposures was moderate. For Cmax, the overall intersubject percent coefficient of variation (%CV) was 35.3% and 34.6%, respectively, in Study 1/Part 1 (ascending single dose study) and Study 2 (the QD cohorts), and the overall intrasubject %CV was 14.4% in Study 1/Part 1. For AUC0–1, the overall
Figure 1. Baricitinib steady-state mean plasma concentrations in healthy subjects following the last dose in the 10-Day dosing regimen. Error bars represent standard deviations.
1359
Shi et al
decreased the geometric mean Cmax by approximately 29%, the changes were thought unlikely to have any clinical significance.
intersubject and intrasubject %CV was 29.6% and 4.4%, respectively, in Study 1/Part 1. For steady-state AUC0–24h following repeat QD dosing, the overall intersubject %CV was 20.2% in Study 2.
Pharmacodynamics Baricitinib demonstrated dose and time-dependent inhibition of cytokine-induced pSTAT3, with maximal inhibition of pSTAT3 occurring 1–2 hours after administration for all doses, coincident with the observed Cmax. Levels of pSTAT3 generally returned to predose control baseline by 16–24 hours in all single dose and QD dose cohorts examined. Similar magnitudes of pSTAT3 inhibition were observed using either IL-6 or TPO as the cytokine stimulus. The extent of pSTAT3 inhibition observed in fed versus fasted individuals was similar although the maximal inhibition occurred at a slightly later time point (3–4 hours post dose) following administration with the meal, consistent with delayed absorption. The PD profile of baricitinib was similar following the single and multiple dose administrations with no evidence of hysteresis, and therefore the data were pooled from both the studies for analysis of relationship between baricitinib PD and plasma concentration. A sigmoid Imax/IC50 model fitted to the combined PK–PD data (Figure 2) indicates that cytokine-induced pSTAT3 can be inhibited by baricitinib with an ex vivo IC50 value of 90 nM (95% CI: 84–97 nM) and a Hill Coefficient (g) of 0.87 (95% CI: 0.81–0.94). WBC and differential, including absolute ANC and absolute lymphocyte count (ALC), was also serially determined following dosing in both the single dose and repeat dose studies with similar results. ANC decreased in a dose-related manner reaching a nadir at 8 hours following dosing and returning to baseline 12–24 hours post-dose. Similar changes were seen in the multiple dose study with a return to baseline ANC within 24 hours following the final dose of study medication on Day 10 or Day 28 of repeat dosing (data not shown). In contrast ALC tended to increase peaking 6 hours after dosing and returning toward normal by 24 hours after dosing (Figure 3).
Dose Proportionality In both the single and multiple dose escalation studies, baricitinib exhibited linear PK characteristics with the plasma Cmax and AUC values rising proportionally to dose. Over the 1–10 mg single-dose range, the powerfunction regression analysis produced dose-proportionality equations of Cmax ¼ 26.8 Dose0.991 (90% CI of b: 0.907–1.08) and AUC0–1¼175 Dose0.998 (90% CI of b: 0.928–1.07), and over the 2–20 mg QD dose range, the dose-proportionality equations were Cmax ¼ 25.8 Dose0.968 (90% CI of b: 0.861–1.08) and AUC0– 0.926 (90% CI of b: 0.852–1.00). In all 24h ¼ 178 Dose the above equations, the 90% CI’s for the exponent, b, of the power function (or equivalently the slope of the logtransformed equation) included the value of 1, suggesting that the baricitinib Cmax and AUC were approximately proportionally related to the dose. Renal Clearance At the steady-state following 10-day QD dosing, the average oral-dose clearance (CL/F) of baricitinib was 17.8 L/h for the 5 cohorts dosed for 10 days and 17.0 L/h for all cohorts. For the QD dosing cohorts, the average percent dose excreted over a 24 hours period (on Day 10) as unchanged parent compound was 64.1% (range: 60.3– 65.6%). The average renal clearance (CLR) of 11.8 L/h (range: 10.5–13.0 L/h) represented approximately 2/3 of the total oral-dose clearance (Table 3). Food Effect Co-administration of a high-fat, high-calorie meal did not alter baricitinib AUC (Table 4). The GMR of baricitinib AUC0–1 comparing the fed and fasted administrations was approximately 99.7%, with the 90% CI of 96–103%, meeting the bioequivalence criterion. Although the highfat meal delayed the median Tmax by 3 hours and
Table 3. Baricitinib Steady-State PK Parameters Following QD or BID Dosing for 10 Days Dose
n
Cmax (nM)
2 mg QD 5 mg QD 10 mg QD 20 mg QD 5 mg BID Average
8 8 8 8 8
45.7 136 206 415 146
(16.7) (27.1) (30.4) (22.1) (16.3) –
Tmax (h) 1.5 1.2 1.0 1.2 1.5
(1.0–2.0) (0.5–2.0) (0.5–4.0) (0.5–2.0) (0.5–3.0) –
t1/2 (h) 8.4 7.4 9.1 6.8 11
(21.1) (17.5) (25.9) (18.3) (26.9) 8.5
AUC0–t (nM*h) 312 831 1460 2490 867
(19.7) (17.4) (17.5) (23.2) (17.6) –
CL/F (L/h)
CLR (L/h)
17.3 (21.5) 16.2 (18.3) 18.4 (14.2) 21.6 (22.0) 15.5 (18.1) 17.8
11.3 (21.9) 10.5 (27.2) 12.1 (24.4) 13.0 (19.8) 12.1 (42.1) 11.8
PK parameters are reported as geometric mean (%CV) value, expect that Tmax is reported as median (range) value; Average PK value is the arithmetic mean of all cohorts. Cmax, peak plasma concentration; Tmax, time to Cmax; t1/2, elimination half-time in the terminal disposition phase; AUC0–t, area under concentrationtime curve within one dosing interval; CL/F, oral-dose plasma clearance; CLR, renal clearance.
1360
The Journal of Clinical Pharmacology / Vol 54 No 12 (2014)
Table 4. The Food Effect on Baricitinib PK Parameters Treatment
n
Fasted Fed
12 11
Cmax (nM) 127 (24.0) 90.3 (51.3)
Tmax (h) 1.0 (0.50–1.5) 4.0 (0.50–6.0)
AUC0–t (nM*h)
AUC0–1 (nM*h)
718 (24.3) 713 (24.8)
730 (24.5) 722 (24.9)
Adjusted Geometric Mean Ratios and 90% Confidence Intervals (Reference = Fasted)* 0.716 (0.572, 0.897)
1.00 (0.971, 1.03)
0.997 (0.964, 1.03)
PK parameters values are geometric means (%CV). Cmax, peak plasma concentration; Tmax, time to Cmax; AUC0–t, area under concentration–time curve estimated up to the last observed concentration time point; AUC0–1, area under concentration–time curve extrapolated to infinity. * Statistical analysis was based on data from the 11 subjects who completed the study through both periods.
Similar changes were seen in the 10-day and 28-day repeat dose studies with transient decreases in ANC and increases in ALC that reversed by 24 hours after each dose, with the exception of the BID regimens where there was not sufficient time after each dose for these parameters to return to baseline. There was no evidence to indicate the changes in ANC or ALC were a consequence of myelosuppression.
Discussion Two FIH studies of baricitinib were conducted in healthy volunteers to evaluate the PK, PD, and safety profiles following single and multiple oral dose administrations. The safety analysis showed that following the single dose administrations of 1–20 mg, the overall incidence of TEAE did not increase with increasing doses and was similarly compared to the placebo groups. Following multiple dose administrations of 2–20 mg up to 10 or 28 days, baricitinib was generally well tolerated with no SAE reported in the active dose groups (the only one SAE of migraine reported was assessed by the Investigator to be
Figure 2. The PK–PD relationship established for baricitinib using ex vivo data (n ¼ 782, R2 ¼ 0.66). The PD marker is inhibition of phosphorylated STAT3 (pSTAT3) in whole blood. The solid line and dashed lines are the best-fit curve and 95% predictive bands, respectively, using a sigmoid Imax/IC50 model, which estimated Hill Coefficient and IC50 values to be 0.87 and 90 nM, respectively.
Figure 3. Percent change in mean absolute neutrophil count (ANC, Graph A) and absolute lymphocyte count (ALC, Graph B) following single doses of baricitinib in healthy subjects. Error bars represent standard error of mean.
1361
Shi et al
unlikely related to baricitinib). As dose limiting toxicity was not observed, the maximum tolerated dose (MTD) was not determined in the two studies conducted. Baricitinib appeared to be safe and well tolerated with an overall safety profile supporting its further clinical development. The transient decrease in ANC seen with other JAK inhibitors was confirmed with baricitinib. This appeared to be linearly related to dose, and is unlikely to represent bone marrow suppression given the rapid return to baseline following dosing for up to 28 days. There was also a transient increase in ALC, which has been seen for other JAK inhibitors not as well documented (Incyte data on file). It is also of note that the decreases in reticulocyte counts seen appeared to partially reverse in the 10 mg QD dose group which was dosed for 28 days. This suggests that with prolonged daily dosing of baricitinib, tachyphylaxis may occur for the effects seen on hematopoiesis. Baricitinib PK profile following multiple-dose administration was similar to that following single-dose administration, and the modest increase in plasma exposures observed was consistent with linear systemic drug accumulation. There was no evidence in these studies to suggest baricitinib induced its own clearance or saturated its elimination pathway (mainly renal excretion). Of note, the mean terminal phase t1/2 of approximate 8 hours was estimated in healthy subjects using a noncompartmental PK method. It is anticipated that the value of t1/2 may increase in disease states associated with reduced renal function. The value may also change when a compartmental PK analysis method is used and the terminal phase is better characterized. Consistent with the PK data, the PD profiles (using pSTAT3 as the PD biomarker) showed the pharmacological activity of baricitinib was maintained upon repeat dosing. The observed PK and PD response appeared to be directly linked and there was no evidence to suggest any PD hysteresis. The PD IC50 value from analyzing the ex vivo whole blood samples was comparable to that determined in vitro.
Conclusion An immediate release formulation of baricitinib, a new and potent JAK1/2 inhibitor, was studied in two Phase 1 studies. The investigational drug demonstrated a favorable pharmacokinetic and pharmacodynamics profile suitable for once-daily administration. Baricitinib exhibited linear
and time-independent PK, and its pharmacological activity was maintained with repeat dosing. All doses of baricitinib were safe and well tolerated, and the overall safety profile of this potent JAK1/2 inhibitor supports its further clinical development.
Conflict of Interest Baricitinib is being co-developed by Eli Lilly and Company and Incyte Corporation, and all authors of this manuscript are employees of Incyte Corporation.
Author Contributions JGS conducted PK/PD data analysis. JGS, WVW, and SY prepared the manuscript. WVW, NGP, EGM, XC, and SY designed study protocol and/or conducted safety data analysis. FL and TE performed GLP bioanalysis for baricitinib concentrations in plasma and urine samples. YL and PAS performed ex vivo pSTAT3 analysis. References 1. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov. 2003;2:473–488. 2. Klareskog L, Catrina AI, Paget S. Rheumatoid arthritis. Lancet. 2009;373:659–672. 3. Rindfleisch JA, Muller D. Diagnosis and management of rheumatoid arthritis. Am Fam Physician. 2005;72:1037–1047. 4. Ghoreschi K, Laurence A, O’Shea JJ. Janus kinases in immune cell signaling. Immunol Rev. 2009;228:273–287. 5. Kremer JM, Cohen S, Wilkinson BE, et al. A Phase IIb dose-ranging study of the oral JAK inhibitor tofacitinib (CP-690,550) versus placebo in combination with background methotrexate in patients with active rheumatoid arthritis and an inadequate response to methotrexate alone. Arthritis Rheum. 2012;64:970–981. 6. Fridman JS, Scherle PA, Collins R, et al. Selective inhibition of JAK1 and JAK2 is efficacious in rodent models of arthritis: preclinical characterization of INCB028050. J Immunol. 2010;184:5298–5307. 7. Greenwald M, Fidelus-Gort RK, Levy R, et al. A randomized, doseranging, placebo-controlled study of INCB028050, a selective JAK1 and JAK2 inhibitor, in subjects with active rheumatoid arthritis. Presentation at the annual scientific meeting of American College of Rheumatology (ACR), Atlanta, GA, USA, Nov. 6–11, 2010. 8. Keystone E, Taylor P, Genovese M, et al. 12-Week Results of a Phase IIb Dose-Ranging Study of LY3009104 (INCB028050), an Oral JAK1/JAK2 Inhibitor, in Combination with Traditional DMARDs in Patients with Rheumatoid Arthritis. Late-breaking oral presentation (Abstract LB0005). European League against Rheumatism's (EULAR) Annual European Congress of Rheumatology, Berlin, Germany, June 8, 2012. 9. Shi JG, Chen X, McGee RF, et al. The pharmacokinetics, pharmacodynamics, and safety of orally dosed INCB018424 phosphate in healthy volunteers. J Clin Pharmacol. 2011;51:1644–1164.
