Diane Wang

462PA POPULATION PHARMACOKINETIC (PK) ANALYSIS OF PALBOCICLIB (PD-0332991) IN PATIENTS (PTS) WITH ADVANCED SOLID TUMORS

ABSTRACT Aim: Palbociclib, a selective oral inhibitor of cyclin-dependent kinases 4 and 6, is in phase 2 and 3 clinical trials across multiple oncology indications. Objectives of this analysis were to describe the population PK of palbociclib in cancer pts, identify significant covariates, and provide individual PK predictions. Methods: This analysis was based on pooled data from 183 pts treated with palbociclib isethionate salt capsules in 3 studies. Compartmental PK models were tested with food effect in the base model. Graphic inspection and stepwise covariate modeling were used to assess covariates of interest based on the metabolic mechanism, including demographic factors, laboratory test variables, and concomitant medications. The significant covariates included in the final model were evaluated for clinical importance by calculating the magnitude of effects on PK parameters. Diagnostic plots, visual predictive check (VPC), and standardized VPC were used for model evaluation. Results: Palbociclib PK was best described by a 2-compartment model with 1st order absorption. Population typical values were estimated to be 60.2 L/h for apparent oral clearance (CL/F) and 2710 L for volume of central compartment (V2/F) (inter-patient variability: CL/F, 36.2%; V2/F, 30.2%); the absorption rate constant and absorption lag time were estimated to be 0.367 1/h and 0.658 h, respectively. The relative bioavailability and absorption lag time of palbociclib were increased by 16.0% and 28.8%, respectively, when taken with a high-fat meal versus fasted conditions. Baseline body weight (BWT) and age were significant covariates on CL/F; BWT was a significant covariate on V2/F. However, they were not considered clinically important. Other tested covariates (liver enzymes, creatinine clearance, co-administration of acid-reducing agents) were not significant covariates on relevant PK parameters of palbociclib. Model evaluation showed no bias in model predictions. Conclusions: The population PK model of palbociclib was well defined. Results support the recommendation that no dose adjustment is necessary for age, BWT, sex, pts with mild or moderate renal impairment, or pts with mild hepatic impairment. Disclosure: W. Sun: Employment: Pfizer; D.D. Wang: Employment: Pfizer Stock ownership: Pfizer.

Palbociclib plus letrozole as first-line treatment for ER-positive and HER2-negative advanced breast cancer in Chinese women: a phase 1 study

Abstract Background Palbociclib, a cyclin-dependent kinase 4 and 6 inhibitor, is approved in many countries for the treatment of ER-positive and HER2-negative advanced breast cancer in combination with endocrine therapy. We aimed to evaluate the pharmacokinetics, safety, and efficacy of palbociclib combined with letrozole as first-line treatment for ER-positive and HER2-negative advanced breast cancer. Methods This is a phase 1, open-label, single-arm study in Chinese women with ER-positive and HER2-negative advanced breast cancer. From cycle 1 day 1, patients received letrozole 2·5 mg orally once daily continuously plus palbociclib 125 mg orally once daily for 3 weeks followed by 1 week off treatment. Blood samples for pharmacokinetic evaluation of palbociclib were collected up to 120 h after a single dose of palbociclib in a lead-in phase and after palbociclib dose on cycle 1 day 21 for multiple dose pharmacokinetic evaluation. Predose blood samples for pharmacokinetic evaluation of palbociclib and letrozole were collected on days 19–21 of cycle 1 and cycle 2 day 1 (letrozole only). Disease assessments were done every 12 weeks from cycle 1 day 1. Safety was assessed per Common Terminology Criteria for Adverse Events, version 4.0. This study is registered with ClinicalTrials, number NCT02499146, and is ongoing but not recruiting. Findings As of March 16, 2016, the cutoff date, the study completed enrolment with 26 patients. All patients completed single dose pharmacokinetic evaluation and 13 completed multiple dose pharmacokinetic evaluation. After multiple doses, the steady-state palbociclib geometric mean area under the concentration–time curve from 0 to 24 h was 2005 ng × h/mL and maximum observed concentration was 112·3 ng/mL, similar to those obtained in Caucasian patients following the same dosing regimen. Comparisons of single dose and multiple dose pharmacokinetic data indicated linear pharmacokinetics of palbociclib. The geometric mean accumulation ratio of palbociclib exposure after multiple dose compared with single dose was 2·1, consistent with a terminal half-life of 23·4–27·5 h. Trough concentrations of letrozole were similar to concentrations obtained in Caucasian patients. No deaths or permanent discontinuations due to adverse events were reported. One serious adverse event occurred but was not considered to be treatment-related. The most common clustered adverse events were neutropaenia (n=20, 77%), leukopaenia (n=19, 73%), and anemia (n=11, 42%), with neutropaenia (n=15, 58%) and leukopaenia (n=8, 31%) the most common grade 3–4 adverse events. Interpretation No dose adjustment for palbociclib is needed based on the Chinese population. The toxicity of palbociclib in combination with letrozole treatment was tolerable and manageable. Funding Pfizer Inc.

Impact of Acid-Reducing Agents on the Pharmacokinetics of Palbociclib, a Weak Base With pH-Dependent Solubility, With Different Food Intake Conditions

Palbociclib free base capsule is a weak base drug with highly pH‐dependent solubility. In vitro and in vivo studies evaluated the impact of acid‐reducing agents on exposure of palbociclib and determined whether the impact, if any, can be mitigated by food. A drug‐drug interaction study (study 1) was conducted first under fasted conditions and showed that coadministration of multiple doses of the proton‐pump inhibitor rabeprazole substantially reduced palbociclib mean area under the concentration‐time curve from time 0 to infinity and maximum observed plasma concentration by 62% and 80%, respectively. In vitro assessment suggested that the presence of bile salt mixed micelles to mimic the fed state can significantly enhance the solubility of palbociclib. Subsequently, study 2 was conducted under fed conditions and demonstrated that coadministration of rabeprazole decreased palbociclib maximum observed plasma concentration by 41% but had limited impact on area under the concentration‐time curve from 0 to infinity (13% decrease). This study also showed that the histamine‐2 receptor antagonist famotidine and local antacid with staggered dosing had no impact on palbociclib exposure under fed conditions. Food intake effectively mitigated the impact of acid‐reducing agents on palbociclib exposure. Palbociclib free base capsule should be taken with food, and acid‐reducing agent use does not need to be avoided.

Abstract LB-229: A phase 1 open-label fixed-sequence two-period crossover study of the effect of multiple doses of tamoxifen on palbociclib (PD-0332991) pharmacokinetics in healthy male volunteers

Introduction: Tamoxifen and one of its primary active metabolites, 4‑hydroxy‑tamoxifen, are known to induce CYP3A4 in vitro, and in vivo coadministration of tamoxifen with letrozole and anastrozole (both CYP3A4 substrates) has resulted in decreased exposures (AUC) of each by 37% and 27%, respectively. The primary route of oxidative metabolism of palbociclib (PD‑0332991) is through CYP3A4, thus coadministration of tamoxifen and palbociclib could result in decreased exposure of palbociclib. The primary objective of this study is to estimate the relative exposure of single 125 mg palbociclib oral doses alone and in the presence of steady‑state concentrations of tamoxifen and its active metabolites (4‑hydroxy‑tamoxifen, N‑desmethyl‑tamoxifen, and endoxifen (N‑desmethyl‑4-hydroxytamoxifen)) in healthy male subjects in order to inform appropriate palbociclib dosing when administered in combination with tamoxifen in the upcoming Phase 3 PENELOPEB study. Methods: This was an open label, 2‑period fixed‑sequence study of the effect of multiple oral doses of tamoxifen on palbociclib pharmacokinetics in 25 healthy male volunteers. Each subject received a single 125 mg oral dose of palbociclib alone, followed by a single 125 mg dose of palbociclib on Day 22 of a 27-day tamoxifen loading dose regimen (60mg QD for 4 days, followed by 20mg QD for 23 days). Serial blood sampling for palbociclib pharmacokinetics was performed up to 144 hours post-palbociclib dose in both periods. Tamoxifen pre-dose blood sampling was performed on specified visits to document achievement of steady-state tamoxifen and metabolite concentrations. Plasma concentrations of palbociclib, tamoxifen, 4‑hydroxy‑tamoxifen, N‑desmethyl‑tamoxifen, and endoxifen were each measured using validated LC/MS/MS (liquid chromatography coupled with mass spectrometry) methods. Palbociclib plasma pharmacokinetic (PK) parameters were estimated using standard non-compartmental methods. Discussion: There were no significant changes in the palbociclib plasma PK parameters AUCinf (area under the curve from time zero to infinity) and Cmax (maximal plasma concentration) in the presence of steady-state tamoxifen when compared to palbociclib alone. Tamoxifen, 4‑hydroxy‑tamoxifen, N‑desmethyl‑tamoxifen, and endoxifen all achieved steady-state prior to administration of palbociclib in period 2. In general, palbociclib and the combination of palbociclib and tamoxifen were well tolerated with no reported serious adverse events; 1 subject discontinued due to generalized itchiness which resolved with a single dose of fexofenadine. Citation Format: Justin T. Hoffman, Melissa O9Gorman, Cho-Ming Loi, Anna Plotka, Tanya Boutros, Leonid Kirkovsky, Corrado Gallo Stampino, Diane Wang. A phase 1 open-label fixed-sequence two-period crossover study of the effect of multiple doses of tamoxifen on palbociclib (PD-0332991) pharmacokinetics in healthy male volunteers. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr LB-229. doi:10.1158/1538-7445.AM2014-LB-229

Abstract P5-19-16: Effect of palbociclib concentration on heart rate-corrected QT interval in patients with cancer

Background: Palbociclib is a selective cyclin-dependent kinase 4/6 inhibitor that blocks G1/S cell cycle progression. The current population analysis assessed the effect of palbociclib exposure on QT interval and heart rate (via evaluation of RR interval) in cancer patients. Methods: Plasma palbociclib concentration (C) and electrocardiogram (ECG) data were pooled from 3 studies of once-daily oral palbociclib: (Study 1) phase 1 dose-escalation study in advanced cancer (Schedule 1: 25, 50, 75, 100, 125, and 150 mg on a 3 wk on/1 wk off cycle [3/1]; Schedule 2: 100, 125, 200, and 225 mg [2/1]); (Study 2) phase 2 study in mantle cell lymphoma (125 mg [3/1]); (Study 3) phase 1/2 study in advanced breast cancer (Cycle 1: 125 mg [2/1]; Cycle 2+: 125 mg [3/1] + letrozole 2.5 mg daily). Blood samples were collected predose, after first palbociclib dose, and at steady state, including the anticipated time of maximal palbociclib concentration. Triplicate (∼2 min apart) ECGs were obtained at baseline; 3 h postdose (Study 1); predose on Day (D) 1 and D21 (Study 2); predose on D1 and predose, 2, 4, 8, 24, 48, and 96 h postdose on D14 of Cycle 1, predose and 4 h postdose on D1 and D14 of Cycle 2, and end of treatment (Study 3, phase 1); and predose on D1 of Cycle 1&3, D14 of Cycle 1&2, and end of treatment (Study 3, phase 2). Baseline singlet ECG data (from time closest to first palbociclib dose) were used to estimate a study-specific heart rate correction factor [β], using a linear mixed effect model of log(QT) vs log(RR/1000) with intersubject variability only on intercept. The averaged triplicate ECG data with time-matched palbociclib concentration data were used to explore RR–C and corrected QT interval (QTc)–C relationships. A linear mixed effects model was used to assess RR–C and QTc–C with intersubject variability on both intercept and slope; sex was tested as a covariate on QTc interval intercept. Results: 184 patients supplied 569 matched pharmacokinetic and ECG assessments. Estimated β values for Studies 1, 2, and 3 were 0.367, 0.369, and 0.363, respectively. Compared with QT correction by Fridericia (QTcF) and Bazett (QTcB) methods, the study-specific correction (QTcS) best minimized the apparent QT–RR correlation and therefore was selected for use in subsequent analyses. Palbociclib had no effect on RR interval in RR–C analysis but increased QTcS in a concentration-dependent manner. The average (90% CI) QTcS increase at the mean and median maximum steady-state concentrations for patients on palbociclib 125 mg (Cmax,ss = 107 and 112 ng/mL, respectively [Study 3]) were 5.60 (2.48–8.72) and 5.88 (2.61–9.16) ms. Sex was not a significant covariate for intercept by analysis of variance. A similar effect of palbociclib on QTcF was observed. Conclusion: In a pooled population analysis of patients with cancer, palbociclib had no concentration-dependent effect on heart rate. There was a slight positive linear relationship between palbociclib concentration and QTcS; however, the upper bound of the 1-sided 90% CI for the increase in QTcS at Cmax,ss did not exceed the threshold of 10 ms. Therefore, QT prolongation is not a major safety concern for palbociclib at the 125-ng/mL recommended therapeutic dose. Citation Format: Jenny Zheng, Michael Amantea, Diane Wang. Effect of palbociclib concentration on heart rate-corrected QT interval in patients with cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P5-19-16.

Abstract CT419: A phase 1 randomized, open-label, 2-sequence, 2-period crossover study of the effect of multiple doses of palbociclib (PD-0332991) on midazolam pharmacokinetics in healthy women of non-childbearing potential

Introduction: Palbociclib (PD‑0332991) is an oral cyclin‑dependent kinase (CDK) 4/6 inhibitor currently in Phase 2 and Phase 3 clinical trials in multiple oncology indications. In vitro, palbociclib and its active lactam metabolite (PF‑05089326) caused time‑dependent inhibition of CYP3A-mediated midazolam 1’‑hydroxylase and testosterone 6β‑hydroxylase activities. The primary objective of this study is to investigate the effect of multiple doses of palbociclib on the pharmacokinetics of a single oral dose of the sensitive CYP3A4/5 probe substrate midazolam in healthy women of non-childbearing potential. Methods: This was an open label, randomized, 2‑period 2‑sequence study of the effect of multiple oral doses of palbociclib on midazolam pharmacokinetics in 26 healthy female volunteers of non-childbearing potential. Simulations were performed with SimCYP, using the in vitro CYP inhibition data and human pharmacokinetic data for palbociclib from Study A5481001 as inputs, to optimize the duration of dosing and washout periods. Each subject received two treatments (Treatment A = 2 mg oral dose of midazolam alone, Treatment B = 2 mg oral dose of midazolam on Day 7 of an 8-day 125 mg QD palbociclib dose regimen) in a crossover scheme. Subjects were randomized to one of two sequences (Sequence 1 = A→B, Sequence 2 = B→A). In Sequence 2 there was a washout period of no less than 14 days after the last administered dose of palbociclib to ensure that there was no carryover effect of CYP3A inhibition into Period 2. Serial blood sampling for midazolam pharmacokinetics was performed up to 36 hours post-midazolam dose in both periods. Palbociclib pre-dose blood sampling was performed on specified visits to document achievement of steady-state concentrations. Plasma concentrations of midazolam and palbociclib were each measured using validated LC/MS/MS (liquid chromatography- tandem mass spectrometry) methods. Midazolam plasma pharmacokinetic (PK) parameters were estimated using standard non-compartmental methods. Discussion:In general, palbociclib reached steady-state levels prior to co-administration of midazolam, allowing for a worst case assessment of the potential palbociclib-mediated time-dependent CYP3A inhibition. Midazolam geometric mean Cmax and AUCinf values increased 37% and 61%, respectively, when midazolam was co-administered with multiple doses of palbociclib as compared to its administration alone. These results indicate that palbociclib is a weak CYP3A inhibitor following daily 125 mg dosing at steady-state according to the FDA classification. In general, the combination of palbociclib and midazolam were well tolerated with no reported serious adverse events or treatment discontinuations. Reversible neutropenia was reported in 15 out of the 26 subjects. Citation Format: Justin T. Hoffman, Anna Plotka, Melissa O9Gorman, Cho-Ming Loi, Leonid Kirkovsky, Corrado Gallo-Stampino, Diane Wang. A phase 1 randomized, open-label, 2-sequence, 2-period crossover study of the effect of multiple doses of palbociclib (PD-0332991) on midazolam pharmacokinetics in healthy women of non-childbearing potential. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr CT419. doi:10.1158/1538-7445.AM2014-CT419