Prediction of Human Pharmacokinetics of Ulixertinib, a Novel ERK1/2 Inhibitor from Mice, Rats, and Dogs Pharmacokinetics
Abstract
Background and Objectives Ulixertinib (BVD-523) is a novel and selective reversible inhibitor of ERK1/ERK2. The primary objectives of the study were to evaluate the pharmacokinetics of ulixertinib in mice, rats, and dogs followed by prediction of human pharmacokinetic profile by allometric equations with/without correction factors.
Methods Oral and intravenous pharmacokinetic profiles of ulixertinib were generated in mice, rats, and dogs. The human intravenous pharmacokinetics profiles [volume of distribution (Vss) and clearance (CL)] were predicted employing simple allometry and using correction factors [maximum life span potential (MLP) and brain weight (BW)]. Pharmacokinetic data obtained from dogs were used to simulate human oral profile [area under the curve (AUC) and maximum plasma concentrations (Cmax)].
Results Post-intravenous administration the CL was mod- erate in dogs (15.5 mL/min/kg) and low in mice (6.24 mL/ min/kg) and rats (1.67 mL/min/kg). Vss was 0.56, 0.36, and 1.61 L/kg in mice, rats, and dogs, respectively. The half- life (t½) of ulixertinib ranged between 1.0 and 2.5 h across the animal species. Following oral administration ulixer- tinib attained maximum concentration in plasma (Tmax) within 0.50–0.75 h in mice and rats, indicating that absorption was rapid; however, in dogs, Tmax attained at 2 h. Absolute oral bioavailability in mice and rats was[92%; however, in dogs, it was 34%. By different allometric approaches, simple method and brain weight correction factor shown clear improvement in the predic- tion efficiency of allometric scaling for Vss (1.34–1.70 L/kg) and CL (4.18–6.09 mL/min/kg), respectively, com- paring with the MLP method and simple method for CL. Similarly, simulation of oral human profile was attained from scaled values and dog data to predict reported human profile (AUC and Cmax).
Conclusions The derived pharmacokinetic parameters (AUC and Cmax at 600 mg dose) and simulated plasma concentration–time profiles of ulixertinib in humans were predicted with good confidence by allometric approach.
The extracellular-signal-regulated kinases, ERK1 and ERK2 (ERK1/2), play a critical role in the RAS/RAF/ MEK-signaling pathway, that controls several fundamental cellular processes, driving proliferation, differentiation, and cell survival [1]. Sustained activation of ERK1/2 was reported necessary for G1–S-phase progression and linked to induction of positive regulators of the cell cycle and inactivation of antiproliferative genes [2]. RAS/RAF/ MEK-signaling pathway is frequently activated by muta- tions in upstream targets such as BRAF, RAS, and receptor tyrosine kinases [3, 4] and clinical efficacy of BRAF and MEK inhibitors confirms that targeting RAS/RAF/MEK pathway has therapeutic potential and great promise [5]. Most of the resistance mechanisms to BRAF and MEK inhibitors ultimately lead to increase in phosphorylation of ERK1/2 suggesting the importance of this node in the RAS/RAF/MEK pathway even in the resistance setting [6, 7]. RAS-activating mutations have been reported in about 90% of pancreatic carcinomas, followed by colon carcinomas (50%), lung cancers, and myeloid leukemia cases (30% each) [8]. Therefore, inhibition of ERK1/2 offers a promising strategy to address both innate and acquired resistance to BRAF and MEK inhibitors in vari- ous solid tumors. Ulixertinib (BVD-523, VRT752271; Fig. 1) is a first-in-class novel small molecule, which potently and selectively inhibits ERK1/2 kinases in a reversible, ATP-competitive fashion. Ulixertinib inhibits tumor growth in vivo in BRAF-mutant melanoma and colorectal xenografts as well as in KRAS-mutant colorectal and pancreatic models [9]. In clinical studies, ulixertinib was well tolerated by patients with advanced solid tumors [10]. In an oral Phase-I dose escalation study (having nine doses) with an end point to determine the dose limited toxicities (DLT), maximum tolerated dose (MTD) along with pharmacokinetic profile and preliminary efficacy assessment, it was administered in a dose range of 10–900 mg twice daily. Ulixertinib showed that linear pharmacokinetics up to 600 mg (twice daily); this was found to be MTD [10] and recommended Phase-II dose [11].
To date, there is no publication describing the disposi- tion of ulixertinib in various preclinical species; hence, a series of preliminary pharmacokinetic experiments in Balb/ C mice, Sprague–Dawley rats, and Beagle dogs were per- formed to characterize and understand pharmacokinetic parameters and absolute oral bioavailability of ulixertinib across the species. Subsequently, these findings were uti- lized to predict its human pharmacokinetic profiles employing allometric scaling approach.
2 Materials and Methods
2.1 Materials
Ulixertinib (purity[99%) was purchased from MedKoo Biosciences, Inc., NC, USA. HPLC grade acetonitrile, formic acid, and methanol were purchased from Rankem, Ranbaxy Fine Chemicals Limited, New Delhi, India. All other chemicals/reagents were of research grade and used without further purification. Dimethylsulfoxide (DMSO), Solutol, dipotassium ethylenediaminetetraacetic acid (K2-.EDTA), Tween-80, and phenacetin were purchased from Sigma Aldrich, St. Louis, USA. Methyl cellulose was purchased from SDFCL Chemicals, Mumbai, India. Absolute ethyl alcohol was purchased from Changshu Chemicals, China.
2.2 Formulations
For oral administration, ulixertinib was formulated as a suspension with Tween-80 and 0.5% methyl cellulose and it was administered at 10 mL/kg volume across the species. The intravenous formulation was a solution prepared using 10% DMSO, 10% Solutol:absolute ethyl alcohol (1:1, v/v) and 80% normal saline. The volume of administration for intravenous was 10 mL/kg for mice and 2 mL/kg for rats and dogs.
2.3 Animal Experiments
All the mice and rat experiments were approved by Insti- tutional Animal Ethical Committee (IAEC) of Jubilant Biosys (IAEC/JDC/2017/121) nominated by CPCSEA (Committee for the Purpose of Control and Supervision of Experiments on Animals). Male Sprague–Dawley rats (* 7–8 weeks; body weight: 200–220 g) and Balb/C mice (* 6–10 weeks; body weight: 28–30 g) were procured from Vivo Biotech, Hyderabad, India.
IAEC of Palamur Biosciences, Telangana, India (1312/ PO/RcBiBt/S/L/09/CPCSEA) approved studies conducted in dogs. Male Beagle dogs (* 1–1.5 year; body weight: 10–12 kg) were housed in Palamur Biosciences Private Limited animal house facility in a temperature (18–28 °C) and humidity (30–70%) controlled room and fed with Pedigree standard pellet feed and water ad libitum for 1 week before using for experimental purpose. For all the experimental works, animals were kept for 12 h overnight fasting, and during this time, they were allowed to take water ad libitum. Feed was provided 4 h post-dose and water was allowed ad libitum.
2.4 Pharmacokinetic Studies
Oral bioavailability of ulixertinib was evaluated in male Balb/C mice, Sprague–Dawley rats, and Beagle dogs. Fasted mice (* 4 h) and rats (overnight * 12 h) were divided into two groups (mice: 12/group; rats: 4/group); however, in cases of dogs (n = 2; fasted for * 12 h overnight), the same dogs were dosed after 1 week wash- out period (cross over design). In each species, Group-1 animals received ulixertinib through intravenous route at a dose of 1 mg/kg, whereas Group 2 received ulixertinib through oral gavage at 10 mg/kg dose. Serial blood sam- ples [100 lL in case of mice (sparse sampling; n = 3 at each timepoint) and rats and 500 lL in dogs) were col- lected (from retro-orbital plexus for mice and rats, and jugular vein for dogs) at pre-dose, 0.12 (intravenous only) 0.25, 0.5, 1, 2, 4, 8, 10 (oral only), and 24 h. Blood samples were collected in tubes containing K2. EDTA as the anti- coagulant and centrifuged for 5 min at 14000 rpm in a refrigerated centrifuge (Biofuge, Heraeus, Germany) maintained at 4 °C for plasma separation and stored frozen at – 80 ± 10 °C until analysis.
2.5 Plasma Sample Processing and Analysis
An aliquot of 50 lL plasma sample was precipitated with 200 lL of acetonitrile:methanol (1:1, v/v) enriched with internal standard (200 ng/mL of phenacetin) and cen- trifuged at 14,000 rpm for 5 min (Eppendorff 5424R, Germany) at 5 °C. Clear supernatant (125 lL) was trans- ferred into vials and 5 lL of supernatant was injected onto LC–MS/MS system for analysis using a validated method [12]. The linearity range was 1.05–2096 ng/mL. In-study quality control (QC) samples, supplemented with concen- trations of 3.14, 1048, and 1747 ng/mL of ulixertinib, were analysed with the unknowns.Plasma samples above the high calibration standard were diluted with corresponding species blank plasma to bring the concentration within linearity range.
For plasma samples analysis, the criteria for acceptance of the analytical runs encompassed the following: (i) 67% of the QC samples accuracy must be within 85–115% of the nominal concentration (ii) not less than 50% at each QC concentration level must meet the acceptance criteria. Following completion of the analysis, both the linearity and quality control sample values were found to be within the accepted variable limits.
2.6 Prediction of Human Pharmacokinetics
The human pharmacokinetics of ulixertinib was predicted employing allometric scaling. The generated pharmacoki- netic data in mice, rats, and dogs were utilized for this scaling exercise.
2.7 Pharmacokinetic Analysis
Pharmacokinetic parameters were calculated by a non- compartmental method using the Phoenix WinNonlin 1.3 software (Pharsight, Mountain View, CA, USA). Absolute oral bioavailability (f) was calculated using the relation- ship, F = [Dose (intravenous) 9 AUC(0-?)oral/Dose (oral) 9 AUC (0-?)intravenous] 9 100.
3 Results
3.1 Pharmacokinetic Studies
The mean plasma concentration versus time profiles for ulixertinib following single intravenous and oral adminis- tration in mice, rats, and dogs are depicted in Fig. 2a–c. The relevant pharmacokinetic parameters are summarized in Table 1. Following intravenous administration, plasma concentrations of ulixertinib decreased mono-exponentially and elimination half-life was found to be 1.04, 2.52, and
1.21 h in mice, rats, and dogs, respectively. Systemic exposure (AUC0-?) at 1.0 mg/kg was found to be higher in rats (10,179 ng 9 h/mL) followed by mice (2672 ng 9 h/ mL) and dogs (1019 ng 9 h/mL). Clearance and volume of distribution were 6.24 mL/min/kg and 0.56 L/kg;
1.67 mL/min/kg and 0.36 L/kg and 15.5 mL/min/kg and 1.61 L/kg in mice, rats, and dogs, respectively.
Post-oral administration, absorption of ulixertinib from gastrointestinal tract was rapid in mice and rats as maxi- mum plasma concentrations (Cmax) in mice (7768 ng/mL) and rats (15,026 ng/mL) attained at 0.5 and 0.75 h, respectively; however, in dogs, Cmax (1442 ng/mL) was attained at 2 h (Tmax). Like in intravenous route, the AUC0-? at 10 mg/kg was found to be higher in rats (98,421 ng 9 h/mL) followed by mice (24,460 ng 9 h/ mL) and dogs (3687 ng 9 h/mL). The elimination half-life values (2.06, 2.02, and 1.29 h in mice, rats, and dogs, respectively) were more or less close to intravenous elim- ination half-life values. Absolute oral bioavailability of
Fig. 2 Plasma concentration–time profiles of ulixertinib after oral (10 mg/kg) and intravenous (1 mg/kg) administration to male a Balb/C mice (mean ± SD, n = 12), b Sprague–Dawley rats (mean ± SD, n = 4), and c Beagle dogs (mean, n = 2).
3.2.3 Simulation of Human Plasma Profile
The oral plasma profile of ulixertinib at 600 mg was pre- dicted employing one-compartment model. The profiles were predicted employing the two predicted human CL values (simple allometry and brain weight correction). The Vss value derived using simple allometry exercise with all three species was used in both these simulations. As mentioned in the methods section, the fraction oral bioavailability (0.34) and absorption rate constant (0.66/h) (in-house data) values derived from dog oral pharmacoki- netics were utilized in this simulation.
The predicted oral plasma profiles of ulixertinib are presented in Fig. 5. The predicted oral plasma AUC values at 600 mg dose were 3.09 and 9.44 lg 9 h/mL employing the CL predicted through simple allometry and brain weight correction, respectively. The predicted oral Cmax values were found to be 0.9 and 1.3 lg/mL with simple allometry and brain weight correction, respectively. The Cmax and AUC values (i.e., 1.3 lg/mL and 9.44 lg 9 h/ mL, respectively) obtained with brain weight corrected human clearance value were in close agreement with the reported values (Cmax: 1.21 ± 0.11 lg/mL and AUC: 7.83 ± 0.81 lg 9 h/mL) (Table 3).
4 Discussion
Preclinical pharmacokinetics of a novel target class drug has great influence on the development and investigation of potential candidates for the further drug design, which necessitates a through exploration and understanding of pharmacokinetic disposition. To the best of our knowledge, there is no preclinical pharmacokinetic data reported for ERK1/2 inhibitors in the literature. In this paper, we have generated both oral and intravenous pharmacokinetics of ulixertinib in mice, rats, and dogs and these data were used in allometric scaling to predict the human pharmacokinetic parameters, and when integrated with efficacy data in PK- PD models, can help estimate efficacious doses in humans. The oral absorption properties of ulixertinib were gen- erally very good in mice and rats, which is reflected by early Tmax (0.50–0.75 h) and very good bioavailability. However, in dogs, Tmax was delayed (2.0 h) and oral bioavailability was acceptable. In all the species, the half- life values are in very close to intravenous route half-life values. Post-intravenous administration ulixertinib showed high plasma clearance (50% of hepatic blood flow) com- pared to mice and rats (3–7% of hepatic blood flow [15]. This phenomenon observed in rats and dogs is in line with the projected clearance from in vitro hepatocyte stability experiments (7.1 and 16.4 mL/min/kg, respectively) [16]. However, disparity was observed between the in vitro hepatocytes stability and in vivo clearance in mice (41 versus 6.4 mL/min/kg, respectively [16]). This observation warrants further investigation. The volume of distribution in mice and rats was approximately equal to total body water, whereas in dogs, it was found to be slightly higher (* twofold higher than the total body water). The oral bioavailability of ulixertinib ranged from 34% in dogs (likely due to moderate hepatocyte stability compared to rodent species) to[92% in mice and rats [16]. The low plasma clearance observed in mice and rats translated into excellent oral bioavailability in these species, whereas in dogs, moderate plasma clearance translated into moderate oral bioavailability. Taken together, the bioavailability of ulixertinib appears to be driven by plasma clearance across the species.
Simple allometric scaling, without any correction factor, worked well for prediction of human volume of distribu- tion. The obtained value of 1.5 L/kg is in line with the volume of distribution values observed for ulixertinib in rodents and dogs. However, it is evident from this scaling exercise that simple allometric scaling did not work for prediction of human clearance value of ulixertinib. Underestimation of the human AUC value (3.09 versus 7.83 ± 0.81 lg 9 h/mL, predicted versus reported) with simple allometry derived human clearance value clearly indicates overestimation of this clearance value. Incorpo- ration of the fraction oral bioavailability and absorption rate constant values from rodents (instead of values from dogs) did not improve the prediction efficiency of human AUC with simple allometry (data not shown).
The normalization of clearance by the MLP or brain weight is a mathematical manipulation that may not be related to the physiology of the species used in the scaling. MLP and BW correction factors were proposed in allo- metric scaling by Boxenbaum [17] and their work also demonstrated that MLP and BW corrections improved the accuracy of human CL prediction when the allometry power exponent b was higher than 0.80–0.90 [14, 18]. Incorporation of brain weight correction, as suggested by ‘‘rule of exponents’’ [14], has shown clear improvement in the prediction efficiency of allometric scaling, which is evident from the closeness of predicted AUC and Cmax values to the observed values in the clinic (Table 2). Fur- thermore, the predicted human clearance value with brain weight correction (i.e., 5.14 mL/min/kg) is in a very good agreement with the projected clearance value (i.e., 5.0 mL/ min/kg) [16] from the human the in vitro hepatocyte sta- bility experiment. The simulated plasma concentration versus time profiles of ulixertinib obtained employing both predicted human plasma clearance values (simple allome- try and with brain weight correction) are represented in Fig. 5. The predicted human AUC and Cmax values post- incorporation of the human clearance value from brain weight correction (i.e., 9.44 lg 9 h/mL and 1.3 lg/mL, respectively) are also in close agreement with the observed values (7.83 ± 0.81 lg 9 h/mL and 1.21 ± 0.11 lg/mL, respectively) in the clinic. Authors would like to mention that incorporation of absorption rate constant and fraction bioavailability values from rodents (i.e., * 3.6/h and * 0.95, respectively, unpublished data) in this pre- diction exercise resulted in the over estimation of human AUC and Cmax values obtained at 600 mg dose. This observation, taken together with the result obtained post- incorporation of values from dogs, indicates that disposi- tion of ulixertinib in humans could be similar to that observed in dogs. However, this aspect warrants further investigation.
In summary, allometric scaling approach with inclusion of apt correction factors was found to be a feasible approach for the prediction of human pharmacokinetics of ulixertinib. Recent publications have indicated that allom- etry-based predictions within twofold prediction error can be useful in establishing human relevant values [19, 20]. Hence, this approach can be considered as a viable option for prospective scaling of both intravenous and oral phar- macokinetics of newer ERK1/2 inhibitors.
5 Conclusion
In conclusion, pharmacokinetics of ulixertinib was asses- sed in mice, rats, and dogs post-administration by oral and intravenous routes. The derived pharmacokinetic parame- ters of ulixertinib and simulated plasma concentration–time profiles in humans were predicted with good confidence by allometric approach. We believe that our exercise will be used prospectively for the prediction of human pharma- cokinetic parameters of newer analogs in ERK1/2 inhibi- tors before dosing for the first time to cancer patients.
Compliance with Ethical Standards
Funding This study did not receive any funding.
Conflict of interest All the authors have no conflict of interest to declare.
Ethical approval All procedures performed were in accordance with the ethical standards of the institutional animal ethics commit- tee and were approved by Institutional Animal Ethical Committee (IAEC) of Jubilant Biosys (IAEC/JDC/2017/121) nominated by CPCSEA (Committee for the Purpose of Control and Supervision of Experiments on Animals).
References
1. Akinleye A, Furqan M, Mukhi N, Ravella P, Liu D. MEK and the inhibitors: from bench to bedside. J Hematol Oncol. 2013;6:27.
2. Meloche S, Pouyssegur J. The ERK1/2 mitogen-activated protein kinase pathway as a master regulator of the G1- to S-phase transition. Oncogene. 2007;26(22):3227–39.
3. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417(6892):949–54.
4. Fernandez-Medarde A, Santos E. Ras in cancer and develop- mental diseases. Genes Cancer. 2011;2(3):344–58.
5. Flaherty KT, Infante JR, Daud A, Gonzalez R, Kefford RF, Sosman J, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med. 2012;367(18):1694–703.
6. Johnson GL, Stuhlmiller TJ, Angus SP, Zawistowski JS, Graves LM. Molecular pathways: adaptive kinome reprogramming in response to targeted inhibition of the BRAF-MEK-ERK pathway in cancer. Clin Cancer Res. 2014;20(10):2516–22.
7. Warthaka M, Adelmann CH, Kaoud TS, Edupuganti R, Yan C, Johnson WH Jr, et al. Quantification of a pharmacodynamic ERK end point in melanoma cell lysates: toward personalized precision medicine. ACS Med Chem Lett. 2014;6(1):47–52.
8. Manning AM, Davis RJ. Targeting JNK for therapeutic benefit: from junk to gold? Nat Rev Drug Discov. 2003;2(7):554–65.
9. Germann U, Furey B, Roix J, Markland W, Hoover R, Aronov A, et al. The selective ERK inhibitor BVD-523 is active in models of MAPK pathway-dependent cancers, including those with intrinsic and acquired drug resistance. Cancer Res. 2015;75(15). https:// doi.org/10.1158/1538-7445.AM2015-4693 (Abstract 4693).
10. Infante JR, Janku F, Tolcher AW, Patel MR, Sullivan RJ, Fla- herty K, et al. Dose escalation stage of a first-in-class phase I study of the novel oral ERK1/2 kinase inhibitor BVD-523 (ulixertinib) in patients with advanced solid tumors. J Clin Oncol. 2015;33(15):2506 (Abstract 2506).
11. Sullivan RJ, Infante RJ, Janku F, Wong DJL, Sosman JA, Keedy V, et al. First-in-class ERK1/2 inhibitor Ulixertinib (BVD-523) in patients with MAPK mutant advanced solid tumors: results of a Phase I dose-escalation and expansion study. Cancer Discov. 2017. https://doi.org/10.1158/2159-8290.CD-17-1119.
12. Kumar R, Suresh PS, Rudresh G, Zainuddin M, Dewang P, Kethiri RR, et al. Determination of ulixertinib in mice plasma by LC–MS/MS and its application to a pharmacokinetic study in mice. J Pharm Biomed Anal. 2016;125:140–4.
13. Li BT, Janku F, Patel MR, Sullivan RJ, Flaherty K, Iannoti E, et al. First-in-class oral ERK1/2 inhibitor Ulixertinib (BVD-523) in patients with advanced solid tumors: Final results of a phase I dose escalation and expansion study. J Clin Oncol. 2017;35(15):2508 (Abstract 2508).
14. Mahmood I, Balian JD. Interspecies scaling: predicting clearance of drugs in humans. Three different approaches. Xenobiotica. 1996;26(9):887–95.
15. Davies B, Morris T. Physiological parameters in laboratory ani- mals and humans. Pharm Res. 1993;10(7):1093–5.
16. Vinod AB, Police A, Hiremath RA, Raj A, Suresh PS, Devaraj VC, et al. Preclinical assessment of ulixertinib, a novel ERK1/2 inhibitor. ADMET DMPK. 2017;5(4):212–23.
17. Boxenbaum H. Interspecies scaling, allometry, physiological time, and the ground plan of pharmacokinetics. J Pharmacokinet Biopharm. 1982;10(2):201–25.
18. Feng MR, Lou XC, Brown RR, Hutchaleelaha A. Allometric pharmacokinetic scaling: towards the prediction of human oral pharmacokinetics. Pharm Res. 2000;17(4):410–8.
19. Paine SW, Menochet K, Denton R, McGinnity DF, Riley RJ. Prediction of human renal clearance from preclinical species for a diverse set of drugs that exhibit both active secretion and net reabsorption. Drug Metab Dispos. 2011;39(6):1008–13.
20. Sinha VK, Vaarties K, De Buck SS, Fenu LA, Nijsen M, Gilissen RA, et al. Towards a better prediction of peak concentration, volume of distribution and half-life after oral drug administration in man, using allometry. Clin Pharmacokinet. 2011;50(5):307–18.