Pharmacokinetics and tolerability of fedratinib, an oral, selective Janus kinase 2 inhibitor, in subjects with renal or hepatic impairment
Abstract
Purpose Fedratinib is an oral, selective Janus kinase 2 inhibitor that is approved in the United States for the treatment of patients with intermediate-2 or high-risk myelofibrosis. Pharmacokinetics and tolerability of fedratinib in subjects with renal impairment (RI) and hepatic impairment (HI) were evaluated in two separate studies.
Methods In the renal study, male and female subjects with stable, chronic mild, moderate, and severe RI, as well as those with end-stage renal disease, were included. The hepatic study included subjects with stable, chronic mild HI. Both were phase 1, multicenter, open-label, single-dose studies, and included matched healthy subjects. Subjects received a single oral dose of fedratinib 300 mg on day 1, were discharged on day 4, returned for clinical visits on days 5–12, and had their end- of-study visit between days 14 and 16.
Results Thirty-six and 17 subjects were included in the renal and hepatic studies, respectively. In the renal study, fedratinib area under the plasma concentration–time curve from time 0 to infinity (AUCinf) was 1.9- and 1.5-fold higher in subjects with severe and moderate RI, respectively, than in matched healthy subjects. In the hepatic study, fedratinib AUCinf did not appreciably differ between subjects with mild HI and matched healthy subjects. Overall, most treatment–emergent adverse events were gastrointestinal and mild.
Conclusion Mild RI and HI do not necessitate fedratinib dosage adjustments. Subjects with moderate RI should be monitored (with dosage adjustments made as necessary), whereas those with severe RI should receive a daily dose of 200 mg, reduced from the indicated dose of 400 mg.
Keywords : Fedratinib · Pharmacokinetics · Tolerability · Renal impairment · Hepatic impairment
Introduction
Fedratinib is an oral kinase inhibitor with activity against both wildtype and mutationally activated Janus kinase 2 that is approved in the United States for the treatment of patients with intermediate-2 or high-risk primary or secondary myelofibrosis (MF) [1]. The pharmacokinetic (PK) pro- file of fedratinib has previously been evaluated in patients with MF as well as in healthy subjects [2–5]. Findings from these studies demonstrate that fedratinib undergoes rapid absorption, reaching maximum plasma concentration (Cmax) approximately 3 h postdose. The PK parameters of fedratinib are minimally affected by food intake, although coadministration with a high-fat meal can reduce gastrointestinal toxicities [5].
Fedratinib is metabolized by cytochrome P450 (CYP) 3A4, CYP2C19, and flavin-containing monooxygenase 3 [1]. In a human mass balance study, a single dose of radiola- beled fedratinib was primarily eliminated in the feces (77%), with 5% eliminated in the urine, suggesting that the liver may have a predominant role in the clearance of fedratinib via hepatic metabolism and/or biliary excretion [Ogasawara et al. Submitted]. Furthermore, fedratinib is highly plasma protein bound (92%) [1]. Both hepatic and renal impairment are known to decrease protein binding and thus may affect fedratinib PK parameters, such as volume of distribution and terminal half-life (t1/2z) [6–9]. Therefore, although hepatic pathways predominate in the clearance of fedratinib, both renal and hepatic function may affect fedratinib PK. Moreo- ver, MF is associated with both hepatic and renal dysfunc- tion [10–18], underscoring the importance of assessing the impact of these impairments on the PK profile of fedratinib. Given the aforementioned and in accordance with the US Food and Drug Administration (FDA) pharmaceutical indus- try guidance—which recommends PK studies in (1) patients with hepatic impairment if hepatic metabolism and/or excre- tion account for a substantial portion of the elimination of a parent drug or its active metabolites and (2) patients with renal impairment if the drug is administered chronically [19, 20]—2 open-label, single-dose studies were undertaken to investigate the PK and tolerability of fedratinib in subjects with renal impairment (NCT01763190; henceforth referred to as “the renal study”) and in subjects with hepatic impair- ment (NCT01762462; henceforth referred to as “the hepatic study”). The results are presented for subjects with mild, moderate, or severe renal impairment; subjects with mild hepatic impairment; and matched healthy matched subjects with normal renal and hepatic function.
Materials and methods
Participants
In the renal study, male and female subjects with stable, chronic renal impairment aged 18–75 years with a body mass index (BMI) of 18.0–34.9 kg/m2 were eligible for enrollment. Subjects aged > 75–79 years were eligible with approval from the study sponsor’s medical monitor. Per regulatory guidance at the time of protocol approval, renal impairment in this study was initially classified according to estimated creatinine clearance (CrCl) as calculated by the Cockcroft–Gault formula, with the following cutoffs: mild (CrCl, 50–80 mL/min), moderate (CrCl, 30–50 mL/min), and severe (CrCl, < 30 mL/min). The data were later reana- lyzed, with renal impairment categories reclassified per cur- rent regulatory guidance [20] as mild (CrCl, 60–89 mL/min), moderate (CrCl, 30–59 mL/min), severe (CrCl, 15–29 mL/ min), and end-stage renal disease (ESRD; CrCl, < 15 mL/ min). In the hepatic study, male and female subjects with sta- ble chronic liver disease with a Child–Pugh classification score of 5–9, who were aged 18–75 years and had a BMI of 18.0–34.9 kg/m2, were eligible for the study. Mild, moderate, and severe hepatic impairment were defined as Child–Pugh scores of 5–6, 7–9, and ≥ 10, respectively. Documented evi- dence of cirrhosis on physical examination was required, such as jaundice, palmar erythema, clubbing of fingers, ascites, and muscle wasting. This study included only sub- jects with mild hepatic impairment. Although inclusion of those with moderate hepatic impairment was planned, the FDA mandated a clinical hold on 15 November 2013 for all ongoing fedratinib studies due to the identification of a signal of potential Wernicke encephalopathy. Thus, mod- erately hepatically impaired subjects were not included in this study. In November 2016, Impact Biomedicines, Inc, acquired fedratinib from Sanofi and worked with the FDA to lift the clinical hold on 18 August 2017. A study in subjects with moderate or severe hepatic impairment is forthcoming (NCT03983161). Both studies also included matched healthy subjects (as certified by comprehensive clinical assessment) with nor- mal renal or hepatic function who were aged 18–75 years and had a BMI of 18.0–32.0 kg/m2. These subjects were matched to those with renal or hepatic impairment by sex (male or female), age (≤ 50 or > 50 years), and body weight (within 15% of the matched renally or hepatically impaired subject); in the renal study, healthy female subjects matched to those with severe renal impairment were required to be within 15% of the average weight of all female subjects with severe renal impairment. Exclusion criteria for both studies included clinically relevant uncontrolled diseases or signs of acute illness, consumption of citrus fruits, or citrus juices within 5 days before inclusion, and concomitant use of agents known to be at least moderate inhibitors or induc- ers of CYP3A4.
Study design and treatment
Both trials were phase 1, multicenter, open-label, single- dose studies (Supplementary Fig. 1). The studies included screening (from 21 to 2 days prior to inclusion), treatment (comprising 1 treatment day, 4 days of institutional stay, and 5 days of clinical visits), and end-of-study visit (14–16 days after fedratinib administration). Subjects began the institu- tional stay on day − 1, were administered a single oral dose of fedratinib 300 mg on day 1, and were discharged on day 4. Subjects returned for clinical visits on days 5–12 and for the end-of-study visit between days 14 and 16.
A single oral dose of fedratinib 300 mg was administered to subjects at approximately 8:00 AM on day 1 under fasted conditions (ie, no food for ≥ 10 h) with 240 mL of noncar- bonated water. After fedratinib administration, subjects were not allowed to eat food for 2 h or drink for 1 h. Fedratinib was provided as 100 mg capsules.
The dose of 300 mg was chosen for a variety of rea- sons. The maximum tolerated dose of fedratinib in patients with MF was 680 mg [2]; a single 680-mg dose in healthy subjects was also tolerated except for a high incidence of gastrointestinal adverse events (AEs) [4]. Additionally, the observed increase in the geometric mean of fedratinib area under the plasma concentration–time curve (AUC) from time 0 to the last time point with a measurable plasma con- centration (AUClast) with a 500- vs 300-mg single dose was 3.7-fold [4], providing a sufficient safety margin in renally and hepatically impaired subjects.
The protocol was approved by the institutional review boards (IRBs) at all participating sites (Crescent City IRB [New Orleans, LA] and Aspire IRB [Santee, CA] for both renal and hepatic studies; RCRC IRB [Austin, TX] for the renal study). All subjects provided written informed consent. The study was designed and conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonisation Good Clinical Practice Guidelines.
Outcomes
The primary endpoints in both studies were Cmax, AUC from time 0 to infinity (AUCinf), and AUClast of fedratinib. Sec- ondary endpoints were apparent clearance, apparent volume of distribution at steady state, t1/2z, effective half-life (t1/2eff), unbound AUCinf, unbound Cmax, predicted accumulation ratio, and safety.
PK analysis
Blood samples for PK analysis were drawn prior to fed- ratinib administration, then at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12,
24, 48, 72, 96, 120, 168, 216, and 264 h postdose. Addi- tional blood samples for protein binding assessment via equilibrium dialysis to determine fedratinib unbound con- centrations were collected prior to fedratinib administration and at 1, 2, 3, 8, 24, and 48 h postdose. Validated liquid chromatography with tandem mass spectrometry was used to determine fedratinib plasma concentrations [21] (lower limit of quantification [LLOQ], 1.00 ng/mL) and fedratinib concentrations in dialysate samples following equilibrium dialysis (LLOQ, 0.100 ng/mL).
Safety
AEs were coded according to the Medical Dictionary for Regulatory Activities version 16.0. AEs that occurred during the pretreatment phase were considered pretreatment AEs. AEs that occurred or worsened during the on-treatment phase (the period between administration of fedratinib on day 1 [inclusive] and the end-of-study visit [14–16 days after the fedratinib dose, inclusive]) were defined as treat- ment–emergent AEs (TEAEs). AEs that occurred or wors- ened during the posttreatment phase (the time after the end-of-study visit) were considered posttreatment AEs. In addition to AE monitoring, safety assessments included physical examination, vital signs, electrocardiogram, and clinical laboratory tests.
Statistics
Based on the findings from a prior study evaluating the effect of food on the bioavailability and tolerability of fedratinib [5], the pooled total-subject standard deviation (SD) esti- mates were 0.382 and 0.358 for log-transformed Cmax and AUCinf, respectively. With a total of 16 subjects (8 in each impairment group and 8 matched subjects for each impair- ment group) and a true total subject SD of 0.40, the ratio of adjusted geometric means was estimated with a maximum imprecision of 35.1% with 90% assurance.
The PK analysis population included all subjects who received fedratinib and had no deviations related to fed- ratinib administration (e.g., vomiting shortly after receiving it) and for whom PK parameters were available. PK parame- ters were summarized via descriptive statistics. Mean values are presented as arithmetic means unless otherwise speci- fied. Concentration values below the plasma assay limit were treated as zero in calculating mean values. The mean values below the LLOQ were reported as LLOQ in tables and not plotted in figures if beyond Cmax. The values expressed in all tables are for the ease of presentation and are not meant to imply accuracy to more than 3 significant figures. In both studies, Cmax, unbound Cmax, AUCinf, unbound AUCinf, and AUClast were log transformed and analyzed using linear regression models with PROC MIXED (SAS Institute Inc). The safety population included all subjects who received fedratinib. Safety was evaluated by review of individual val- ues, AEs, and descriptive statistics. Demographic and base- line characteristics were summarized for the safety population via descriptive statistics.
Results
Baseline characteristics and disposition
The renal and hepatic studies included 36 and 17 subjects, respectively (Table 1). The mild renal impairment (n = 4) and ESRD (n = 1) groups did not have matched healthy sub- jects. The mean age of all subjects in the renal study was 60.9 years (range 31–78 years), and 55.6% were females. Subjects with severe and moderate renal impairment were older (mean age 68.3 and 66.3 years) than those with mild renal impairment (mean age 59.0 years). Subjects in the hepatic study had a mean age of 47.6 years (range 31–62 years), and most were male (58.8%). Healthy sub- jects matched to those with mild hepatic impairment were younger than subjects with mild hepatic impairment (mean age 45.4 vs 50.1 years). The majority of the subjects in both studies had a BMI < 30 kg/m2 (renal study 80.6%; hepatic study 76.5%). PK analysis: renal study In the renal study, 29 subjects were included in the PK analysis; 6 (1 subject with moderate renal impairment and 5 healthy subjects enrolled to match subjects with severe renal impairment) were excluded due to vomiting within 2 h after fedratinib dosing, and 1 healthy subject (enrolled to match a subject with severe renal impairment) did not match criteria for body weight. These subjects were excluded from the reanalysis due to the revised CrCl cutoffs as described in the Materials and Methods section. In general, fedratinib plasma concentrations over time were higher in subjects with moderate or severe renal impairment than in healthy subjects (Fig. 1a), and fedratinib AUCinf tended to decrease with increasing CrCl (Supplementary Fig. 2). Supplemen- tary Table 1 shows intercept and slope estimates for fed- ratinib PK parameters based on the regression model. Over- all, differences in t1/2z and t1/2eff between subjects with renal impairment and matched healthy subjects were not clini- cally relevant (Table 2). In subjects with severe or moderate renal impairment, the unbound fraction (fu) was 1.58% and 1.72%, respectively, and in matched healthy subjects, the fu was 2.27% and 2.26%, respectively. An analysis of vari- ance model, which included renal function group as a factor, was used to calculate the ratio of geometric means and 90% confidence intervals between subjects with renal impairment and their matched healthy subjects (Table 3). Following a single 300 mg oral dose of fedratinib, AUCinf was higher in subjects with severe or moderate renal impairment (1.9- and 1.5-fold, respectively) than in matched healthy subjects. Ratios could not be calculated for subjects with mild renal impairment or those with ESRD because these groups had no matched healthy subjects. PK analysis: hepatic study The PK analysis of the hepatic study included 16 subjects due to the exclusion of 1 matched healthy subject due to vomiting within 2 h after fedratinib dosing. Subjects with mild hepatic impairment and matched healthy subjects had similar fedratinib plasma concentration–time profiles (Fig. 1b). In subjects with mild hepatic impairment, the fu was 2.64% and in matched healthy subjects, the fu was 2.41% (Table 4). Between the groups, differences in t1/2z and t1/2eff were not clinically significant. The analysis of geometric mean ratios via an analysis of variance model demonstrated that AUCinf was not appreciably different between subjects with mild hepatic impairment and matched healthy subjects (Table 5). Fig. 1 Fedratinib plasma concentration–time profile (linear and inset with semilogarithmic scales) in a subjects with mild, moderate, or severe renal impairment (RI) and matched healthy subjects and b sub- jects with mild hepatic impairment (HI) and matched healthy subjects following oral administration of a single dose of fedratinib 300 mg. Safety: renal study In the renal study, 69.4% of subjects experienced ≥ 1 TEAE (Supplementary Table 2). Gastrointestinal disorders were the most common TEAEs (63.9%), including diar- rhea (36.1%), nausea (36.1%), and vomiting (19.4%). Most gastrointestinal TEAEs were mild in intensity, although a few were moderate. One matched healthy subject experi- enced two severe TEAEs: cytomegalovirus infection and increased aminotransferases. The increased aminotrans- ferases were judged to be due to cytomegalovirus hepatitis and unrelated to fedratinib. The cytomegalovirus infection was considered to be possibly related to fedratinib. All subjects recovered from TEAEs without sequelae. No seri- ous AEs or deaths occurred. In the hepatic study, 52.9% of subjects experienced ≥ 1 TEAE (Supplementary Table 2). The most frequently reported TEAEs were gastrointestinal disorders (47.1%), including diarrhea (29.4%) and nausea (23.5%). All TEAEs were of mild intensity, and subjects recovered without sequelae. No deaths or serious AEs were reported. No clini- cally relevant potentially clinically significant abnormalities were observed in matched healthy subjects. One subject with mild hepatic impairment had asymptomatic decreases in white blood cell count (2.90 × 109/L) and neutrophil count (0.70 × 109/L) on day 3 of the study; these were judged as not clinically significant. Both counts were recovered by the end of study. No changes in alanine aminotransferase or aspar- tate aminotransferase (AST) levels were observed during the study, although three subjects with mild hepatic impairment had increases in both levels at baseline. Discussion Fedratinib PK and tolerability were evaluated in subjects with renal impairment and those with hepatic impairment in two respective studies. Fedratinib Cmax and AUCinf increased by approximately 1.5-fold in subjects with moderate renal impairment and by nearly twofold in those with severe renal impairment compared with healthy subjects with normal renal function. The fu was higher in healthy subjects with normal renal function than in subjects in any renal impair- ment group. Between subjects with mild hepatic impairment and matched healthy subjects with normal hepatic function, neither fedratinib exposure nor fu was appreciably different. Across all subjects, most TEAEs were gastrointestinal and mild in nature, and no deaths or serious AEs were reported. These findings align with what has been previously reported with fedratinib in patients with MF and comor- bid kidney or liver impairment. In a population PK analysis of patients with MF, CrCl was identified as a statistically significant covariate for fedratinib apparent clearance [21]. When compared with patients with normal renal function, patients with mild (defined as CrCl ≥ 60 and < 90 mL/min), those with moderate (defined as CrCl ≥ 30 and < 60 mL/min) and those with severe (defined as CrCl ≥ 15 and < 30 mL/ min) renal impairment had fedratinib plasma AUC expo- sure increases of 10%, 37%, and 59%, respectively. Due to the small number of patients with severe renal impairment (N = 3), extent of the impact of severe renal impairment on fedratinib PK in patients with MF should be interpreted with caution. Although any conclusion about tolerability in a chronic dose setting cannot be directly based on a single dose study, notably, in the phase 3 JAKARTA trial, the fre- quencies and distributions of TEAEs were generally simi- lar in patients with MF with normal renal function (n = 57) and those with mild or moderate renal impairment (n = 128) [Bristol Myers Squibb. Data on file]. In context with the findings reported in this manuscript, these results demon- strate that the increase in fedratinib exposure in subjects with mild or moderate renal impairment is not clinically meaningful. However, the exposure increase in subjects with moderate renal impairment suggests that patients with pre- existing moderate renal impairment require more thorough safety monitoring (with dose adjustments if necessary) than those with mild renal impairment or normal renal function. Moreover, given the more marked exposure increases in sub- jects with severe renal impairment, a dose reduction from the currently approved dose of 400 mg daily to 200 mg daily is warranted, with additional dose modifications made as required with safety and efficacy monitoring. In the previously noted population PK analysis, neither mild (total bilirubin ≤ upper limit of normal [ULN] and AST > ULN or total bilirubin 1–1.5 × ULN and any AST) nor moderate (total bilirubin > 1.5–3 × ULN and any AST) hepatic impairment were statistically significant covari- ates affecting fedratinib PK [21]. Furthermore, in the phase 3 JAKARTA study, the frequencies and distributions of TEAEs were generally similar across baseline hepatic func- tion groups (normal hepatic function, n = 126; mild or mod- erate hepatic impairment, n = 66) [Bristol Myers Squibb. Data on file]. The findings of the present study, together with these analyses, demonstrate that fedratinib does not require dose modifications for patients with mild hepatic impairment.
Although the mass balance study indicated that urinary clearance is a minor pathway of fedratinib (5%) [Ogasawara et al. Submitted], higher fedratinib exposure was observed in subjects with moderate or severe renal impairment vs matched healthy subjects or those with mild renal impairment, and fedratinib AUCinf tended to increase with decrease of increased fedratinib exposure in subjects with renal impairment.
In summary, fedratinib does not require dosage adjust- ments in patients with mild renal or mild hepatic impair- ment. Patients with moderate renal impairment should be monitored and dose adjustments implemented if necessary, whereas patients with severe renal impairment should be administered a dose of fedratinib 200 mg daily, reduced from the currently approved dose of 400 mg daily. The findings of these studies will help clinicians make treatment deci- sions when prescribing fedratinib for their patients with MF who may have or may develop comorbid renal and hepatic dysfunction.