Evaluation of the impact of renal impairment on the pharmacokinetics of glasdegib in otherwise healthy volunteers

This was an open-label, parallel-group study (ClinicalTrials.gov NCT03596567). Participants from the impaired groups were recruited first, and demographics were pooled across study sites and levels of impairment to determine an average value for age and weight. Participants with normal renal function were then recruited so that each participant’s age was within 10 years and weight within 15 kg of the mean of the pooled impaired groups.

Renal function was estimated using both Cockcroft–Gault and Modification of Diet in Renal Disease (MDRD) equations: estimated glomerular filtration rate (eGFR; mL/min/1.73 m²) was obtained directly from the laboratory or calculated using the following equation from the MDRD study: 175 × (_Scr,std)^−1.154 × (Age)^−0.203 × (0.742 if female) × (1.212 if African American). Scr,std denotes serum creatinine measured with a standardized assay for serum creatinine. In terms of clearance of renally filtrated drugs, renal elimination capacity was related to absolute GFR. To use the MDRD-derived, body surface area (BSA)-adjusted value of eGFR to obtain absolute GFR (mL/min) for participant assignment to moderate or severe renal impairment groups, this value was multiplied by the individual participant’s BSA (i.e., measured BSA/1.73 m²). CrCl was also estimated from a spot serum creatinine measurement using the following Cockcroft–Gault equation: CrCl (mL/min) = [140 – age (years)] × total body weight (kg) × (0.85 for females)/72 × serum creatinine (mg/dL). The eGFR value obtained within 24 h prior to the glasdegib dose was the value used for stratification and group assignment. The CrCl value was recorded at the same time eGFR was determined.

Participants were required to have stable renal function (two serum creatinine values within 20% of each other obtained within a 2-week period) during the screening period. The eGFR value obtained within 24 h prior to the glasdegib dose was the value used for participant stratification and group assignment. The CrCl value was recorded at the same time eGFR was determined.

Although glasdegib can be taken with or without food, the study was performed under fasted conditions, to minimize potentially confounding factors. Following an overnight fast of ≥ 10 h, participants received a single, oral, 100-mg dose of glasdegib with approximately 240 mL of water in the morning on day 1. No food or drink (except water) was permitted for ≥ 4 h post dose. Each participant underwent serial blood samplings to determine plasma concentrations of glasdegib up to 120 h post dose on day 6. PK blood sampling (~ 2 mL) occurred pre-dose, and at 0.5, 1, 2, 4, 6, 10, 24, 48, 72, 96, and 120 h post dose. The eGFR value obtained on day –1 was used for PK analysis. At pre-dose, and at 1 and 2 h post dose, separate blood samples (10 mL) were collected for measurement of protein binding. Physical examinations, electrocardiogram (ECG), vital signs, and clinical laboratory tests were conducted, and adverse events (AEs) were monitored throughout the study. Participants were confined to the unit during the study until completion of PK sampling and safety assessments on the morning of day 6. Participants who withdrew were to be replaced to ensure that six PK-evaluable participants were in each group. A follow-up assessment 28–35 days after the administration of the glasdegib dose was conducted by phone.

The study protocol was approved by an independent institutional review board and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Informed consent was obtained from all individual participants included in the study.

All participants were aged 18–75 years with a body mass index (BMI) of 17.5–40.0 kg/m² and a total body weight > 50.0 kg. Participants had no clinically relevant abnormalities identified by a detailed medical history, full physical examination, 12-lead ECG, or clinical laboratory tests. Participants were excluded if they: had any condition affecting drug absorption; were renal allograft recipients; had urinary incontinence without catheterization; had congenital long QT syndrome, torsades de pointes, or clinically significant ventricular arrhythmias.

Participants with normal renal function had eGFR ≥ 90 mL/min, based on the MDRD equation, and were matched for age (within 10 years), weight (within 15.0 kg), and sex to participants in the impaired renal function groups. Participants were excluded if they: had blood pressure ≥ 140 mm Hg (systolic) or ≥ 90 mm Hg (diastolic); had QT interval > 450 ms or QRS interval > 120 ms.

The following eligibility criteria were applicable for participants with moderate (eGFR ≥ 30 mL/min and < 60 mL/min) or severe (eGFR < 30 mL/min, but not requiring hemodialysis) impairment of renal function: good general health commensurate with the chronic kidney disease population; stable drug regimen, defined as not starting a new drug or changing dosage within 7 days or five half–lives (whichever was longer) before dosing the study drug; any form of renal impairment except acute nephritic syndrome. Good general health was defined as no clinically relevant abnormalities, with the exception of hypertension, diabetes mellitus, hyperparathyroidism, and ischemic heart disease, as long as the participant was medically stable, was on a stable drug regimen, and was able to abide by the fasting conditions of the study. Participants with a history of previous nephritic syndrome, but in remission, were included. Participants were excluded if they: required hemodialysis; had blood pressure ≥ 180 mm Hg (systolic) or ≥ 110 mm Hg (diastolic); had QT interval > 470 ms or QRS interval > 120 ms; had serum albumin < 2.5 g/dL; had platelets < 60,000/μL; had hemoglobin < 8 g/dL.

Plasma PK samples were analyzed for glasdegib concentrations at Covance Bioanalytical Services (Shanghai, China) using a validated, sensitive, and specific high-performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS) method. Plasma samples were stored at –70 °C until analysis and assayed within the 575 days of established stability. Calibration standard responses were linear over the range of 3–3000 ng/mL using a weighted (l/concentration²) linear regression. The lower limit of quantification (LLQ) for glasdegib was 3 ng/mL.

Assay accuracy, expressed as the inter-assay percentage relative error (%RE) of the mean glasdegib quality control (QC) sample concentrations ranged from –1.7% to 3.0% for the low (9 ng/mL), mid (100 ng/mL), and high (2250 ng/mL) QC samples. Assay precision, expressed as the inter-assay percentage coefficient of variation (%CV) of the mean glasdegib QC sample concentrations, was ≤ 6.0% across the low, medium, and high concentrations.

Plasma samples to measure the fraction of glasdegib not bound to human plasma proteins underwent equilibrium dialysis with phosphate-buffered saline (PBS). Dialyzed plasma and dialyzed PBS were diluted with non-dialyzed PBS and non-dialyzed plasma, respectively, to generate plasma:PBS mixed matrix samples that were analyzed using a validated, sensitive, and specific HPLC–MS/MS method to measure glasdegib concentrations. The fraction of glasdegib not bound to plasma proteins was calculated as the ratio of the glasdegib concentration in dialyzed PBS to the sum of glasdegib concentrations in dialyzed plasma and dialyzed PBS. Plasma samples were stored at –70 °C and dialyzed within the 575 days of established stability in plasma. Post-dialysis plasma:PBS samples were stored at –20 °C and assayed within the 63 days of established stability in plasma:PBS. Calibration standard responses were linear over the range of 1–1000 ng/mL using a weighted (l/concentration²) linear regression. The LLQ for glasdegib was 1 ng/mL.

Assay accuracy, expressed as the inter-assay  %RE of the mean glasdegib QC sample concentrations, ranged from –0.8% to 2.0% for the low (3 ng/mL), low–mid (40 ng/mL), mid (400 ng/mL), and high (800 ng/mL) QC samples. Assay precision, expressed as the inter-assay  %CV of the mean glasdegib QC sample concentrations, was ≤ 3.6% across the low, low–mid, mid, and high concentrations.

The primary objective of the study was to estimate the effect of renal impairment on the PK of a single, oral, 100-mg dose of glasdegib under fasted conditions. The PK concentration population was defined as all participants enrolled and treated who had ≥ 1 concentration. The PK parameter analysis population was defined as all participants enrolled and treated who had ≥ 1 of the PK parameters of primary interest, which were the PK parameters selected as primary endpoints. Using non-compartmental analysis of plasma concentration–time data, PK parameters assessed were: AUC from time 0 to the time of the last quantifiable concentration (AUC_last); unbound AUC_last; AUC_∞; unbound AUC_∞; CL/F; unbound CL/F; C_max; unbound C_max; fraction of unbound drug in plasma; time to C_max (t_max); apparent volume of distribution of total drug (V_z/F); unbound V_z/F; and half-life (t_½).

All participants who received ≥ 1 dose of study medication were included in the safety analyses and listings. As a secondary objective, the safety and tolerability of glasdegib were assessed by AE monitoring and changes in clinical laboratory results, ECGs, and physical examination findings. AEs were graded according to the Medical Dictionary for Regulatory Activities version 21.0.

A total of approximately 18 participants (approximately six per group) were to be enrolled into the study, based on EMA recommendations [14]. Glasdegib exposure in normal participants was the reference treatment. One-way analysis of variance was used to compare the natural log-transformed AUC_∞ and C_max for each of the renal impairment groups (Test) with the normal group (Reference). Estimates of the adjusted mean differences (Test–Reference) and corresponding 90% confidence intervals (CIs) were obtained from the model. The adjusted mean differences and 90% CIs for the differences were exponentiated to provide estimates of the ratio of adjusted geometric means (Test/Reference) and 90% CIs for the ratios. Linear regression was used to analyze the potential relationship between appropriate PK parameters (CL/F or unbound CL/F) and renal function (CrCl and eGFR). Estimates of the slope and intercept, together with their precision (90% CI) and the coefficient of determination were obtained from the model. All other data were summarized descriptively, unless otherwise stated. Analyses were performed using SAS version 9.4.

A separate post hoc analysis of efficacy and safety was performed on patients from the BRIGHT 1003 MDS & AML clinical study (Phase Ib and II). All patients received glasdegib 100-mg QD plus low-dose cytarabine 20-mg BID subcutaneously [15, 16]. Patients were assigned as having normal renal function or mild or moderate renal impairment per the National Kidney Foundation, Kidney Disease Quality Outcomes Initiative criteria.

A total of 18 participants were assigned and received glasdegib treatment between May 17, 2018 and September 19, 2018 at two centers in the United States. Six participants were assigned to each of the three groups. All participants completed the treatment and were included in the PK analysis. 11/18 participants were male and the majority were white (88.9%). Mean (range) age was 64.6 years (52–75) (Table 1). Mean (range) weight and BMI were 85.3 kg (60–104) and 30.2 kg/m² (22–36), respectively. Thirteen participants received concomitant drug treatments during the study, the most frequent of which were acetylsalicylic acid and amlodipine (n = 6 each). Administration of concomitant medications that could impact the PK of glasdegib was not allowed during the conduct of the study.

The median plasma glasdegib concentration–time profiles across renal function groups are shown in Fig. 1. Following a single, oral, 100-mg dose of glasdegib, median t_max was 2.0 h in both impairment groups and 1.5 h in the normal group, with the same range in all groups (1.0–2.0 h) (Table 2). Longer t_½ mean values were observed in participants with moderate (22.7 h) and severe (25.0 h) renal impairment compared with those with normal renal function (16.8 h).

In a regression analysis using both CrCl and eGFR, a decrease in CL/F was observed with decreasing renal function (Fig. 2). Glasdegib mean oral clearance (CL/F) values decreased by approximately 50.0% in both renal impairment groups compared with the normal group: 5.1 L/h and 5.2 L/h for the moderate and severe renal impairment groups, respectively, compared with 10.4 L/h for the normal renal function group.

Glasdegib plasma exposure was approximately twofold higher in participants with moderate and severe renal impairment compared with participants with normal renal function: adjusted geometric mean AUC_∞ ratios (90% CI) for renal impairment versus normal function were 204.8% (142.3–294.7%) and 202.4% (145.9–280.9%) for participants with moderate and severe renal impairment, respectively (Table 3). Mean C_max values were 1.4-fold and 1.2-fold higher for participants with moderate and severe renal impairment, respectively, compared with those with normal renal function: adjusted geometric mean C_max ratios (90% CI) for renal impairment versus normal function were 136.8% (96.7–193.3%) and 120.1% (76.9–187.6%) for participants with moderate and severe renal impairment, respectively. Variability in glasdegib exposure based on geometric  %CV ranged from 23% to 37% for AUC_∞ and 28% to 50% for C_max.

Glasdegib plasma protein binding was > 90.0% and the plasma-free fraction of glasdegib was not altered by renal impairment. Glasdegib protein binding was not saturable at the exposure range tested in this study, and in line with the observation that the protein binding in vitro was independent of glasdegib concentration over the tested range of 1–10 µM (1 µM = 375 ng/mL). The post-dose protein binding samples collected at 1 and 2 h post dose (expected t_max of glasdegib; see Table 2) were similar within each participant. The mean of the two time points within each participant was used to calculate individual unbound PK parameters. The geometric means for the fraction of unbound drug in plasma values were similar between groups: 7.5%, 7.2%, and 7.3% for normal, moderate, and severe groups, respectively (Table 3). Variability in unbound glasdegib exposure based on geometric  %CV ranged from 25% to 46% for unbound C_max and 19% to 42% for unbound AUC_∞.

Of 100 patients (n = 16 in Phase Ib; n = 84 in Phase II) with AML and MDS in BRIGHT 1003 MDS & AML, 39 and 45 had mild and moderate renal impairment, respectively, and 16 had normal renal function. The AE profiles were similar between patients with mild and moderate renal impairment and patients with normal renal function (Supplementary Table 1). Additionally, median overall survival (80% CI) was 6.9 (4.4–9.9) months for the mild renal impairment group, 7.5 (5.0–11.1) months for the moderate renal impairment group, and 7.4 (4.0–12.7) months for the normal renal function group (Fig. 3).

Glasdegib was well tolerated, with five all-causality treatment-emergent AEs (TEAEs) reported in three participants. There were two TEAEs reported by one participant with severe renal impairment (infusion-site phlebitis and reversible airways obstruction), both of which were not considered to be treatment-related by the investigator. TEAEs of constipation and tremor were reported in one participant with moderate renal impairment, of which tremor was considered to be treatment-related by the investigator. One TEAE of musculoskeletal pain was reported by a participant with normal renal function, and was considered by the investigator to be treatment-related. All TEAEs were considered mild in severity and resolved by the end of the study.

No participants discontinued from the study. There were no deaths, serious AEs, severe AEs, or medical errors reported in this study. The most frequently observed laboratory abnormalities (abnormal baseline) were: blood urea nitrogen > 1.3 × upper limit of normal in one participant with moderate renal impairment and two participants with severe renal impairment; urine protein ≥ 1 in one participant with moderate renal impairment and two participants with severe renal impairment; and leukocyte esterase ≥ 1 in one participant with normal renal function and two participants with severe renal impairment. None of the laboratory test abnormalities, or changes in vital signs or ECG values were considered clinically significant by the investigator.