Quantitative systems toxicology (QST) reproduces species differences in PF-04895162 liver safety due to combined mitochondrial and bile acid toxicity

Abstract

Many compounds that appear promising in preclinical species, fail in human clinical trials due to safety concerns. The FDA has strongly encouraged the application of modeling in drug development to improve product safety. This study illustrates how DILIsym, a computational representation of liver injury, was able to reproduce species differences in liver toxicity due to PF-04895162 (ICA-105665). PF-04895162, a drug in development for the treatment of epilepsy, was terminated after transaminase elevations were observed in healthy volunteers (NCT01691274). Liver safety concerns had not been raised in preclinical safety studies. DILIsym, which integrates in vitro data on mechanisms of hepatotoxicity with predicted in vivo liver exposure, reproduced clinical hepatotoxicity and the absence of hepatotoxicity observed in the rat. Simulated differences were multifactorial. Simulated liver exposure was greater in humans than rats. The simulated human hepatotoxicity was demonstrated to be due to the interaction between mitochondrial toxicity and bile acid transporter inhibition; elimination of either mechanism from the simulations abrogated injury. The bile acid contribution occurred despite the fact that the IC₅₀ for bile salt export pump (BSEP) inhibition by PF-04895162 was higher (311 µmol/L) than that has been generally thought to contribute to hepatotoxicity. Modeling even higher PF-04895162 liver exposures than were measured in the rat safety studies aggravated mitochondrial toxicity but did not result in rat hepatotoxicity due to insufficient accumulation of cytotoxic bile acid species. This investigative study highlights the potential for combined in vitro and computational screening methods to identify latent hepatotoxic risks and paves the way for similar and prospective studies.

Keywords: DILIsym; PBPK; PF‐04895162; QSP; QST; bile acid transporters; drug‐induced liver injury; mechanistic; mitochondria; species translation.

Figure 1

Examples of in vitro data collected to quantitatively characterize the relationship between PF‐04895162 and mechanisms of toxicity. (A) PF‐04895162 interaction with human BSEP in vesicular transport assays (SB‐BSEP‐HEK293), using 0.2 µmol/L taurocholate as the probe substrate. PF‐04895162 inhibited BSEP‐mediated taurocholate transport by 38% at 250 µmol/L (maximum tested concentration). (B) PF‐04895162 interaction with rat Bsep in vesicular transport assays (SB‐ratBsep‐HEK293), using 2 µmol/L taurocholate as the probe substrate. The calculated IC₅₀ value was 229.6 µmol/L. (C) Oxygen consumption rate (OCR) following 24‐hour incubation of primary human hepatocytes with PF‐04895162. Basal OCR was measured in a Seahorse XF Analyzer for 40 minutes, followed by injection of FCCP (3 µmol/L) and measurement of the increase in OCR as an indicator of spare respiratory capacity. Two independent studies were conducted, with similar results. (D) OCR following 1‐hour incubation of rat primary hepatocytes with PF‐04895162. Basal OCR and spare respiratory capacity were measured using a Seahorse XF Analyzer. Two independent studies were conducted, with similar results. Studies using 24‐hour incubation with PF‐04895162 were also conducted, with similar results

Figure 2

Comparisons of PBPK sub‐model simulations against measured data. The rat PBPK sub‐model was simultaneously optimized against multiple dosing protocols. (A) Simulated rat plasma PF‐04895162 (ng/mL) following single oral administration of 3, 30, or 100 mg/kg was compared with measured data for the same protocols. (B) Simulated rat plasma PF‐04895162 (ng/mL) for study day 28 following repeat daily dosing of 3, 30, or 100 mg/kg was compared with measured data for the same protocols. (C) Simulated human plasma PF‐04895162 (ng/mL) following optimization against data for a single 300‐mg po dose. (D) Simulated human plasma PF‐04895162 (ng/mL) from day 14, following 300‐mg BID dosing, as compared with data. The human repeat dosing simulation result did not require further changes to the human PBPK sub‐model optimized to a single 300‐mg dose. (E) Simulated rat liver concentrations on day 28 following administration of 100 mg/kg/day. (F) Simulated human liver concentrations on day 14 following administration of 300‐mg BID for 14 days

Figure 3

Simulation of PF‐04895162 (100 mg/kg/day for 28 days) in rat SimPops (n = 294) does not result in hepatotoxicity. Evaluation of drug‐induced serious hepatotoxicity (eDISH) plot for rat SimPops results, illustrating peak ALT (x‐axis) vs peak total bilirubin (y‐axis) for each individual. Each star represents peak ALT and total bilirubin for an individual rat. Vertical lines correspond to 3x ULN for ALT and 2x ULN for total bilirubin

Figure 4

Subclinical simulated indicators of mechanisms of toxicity in rat SimPops. (A) Liver CDCA‐amide across the 28‐day dosing period (100 mg/kg/day). Each line represents an individual rat. (B) Liver average ATP across the 28‐day dosing period (100 mg/kg/day)

Figure 5

Simulation of PF‐04895162 (300‐mg BID for 14 days) in human SimPops (n = 285) results in hepatotoxicity. Each star represents peak ALT and total bilirubin for an individual human subject in the eDISH plot

Figure 6

Simulated time courses for ALT in human SimPops treated with PF‐04895162 (300‐mg BID for 14 days and followed for an additional 14 days). Each line represents an individual subject

Figure 7

Comparison of time to peak ALT in human SimPops vs measured clinical data. Simulated individuals are represented in red. Clinical measurements for subjects in the Phase I study are represented in black. The median value is indicated by a line

Figure 8

Subclinical simulated indicators of mechanisms of toxicity in human SimPops. (A) Liver CDCA‐amide across the 14‐day dosing period (300‐mg BID), with 14‐day follow‐up. Each line represents an individual subject. (B) Liver average ATP across the 14‐day dosing period (300‐mg BID), with 14‐day follow‐up