(M1330-06-33) Plasma and Liver Pharmacolipidodynamics of Gefitinib after Intravenous Administration to Mice Using LC-MS/MS and MS Imaging

Purpose: Understanding off target pharmacology and idiosyncratic toxicity is critical to successful drug development and mitigation of late-stage compound attrition. Gefitinib is an anilinoquinazoline thymidylate kinase inhibitor (TKI), selective for the epidermal growth factor receptor (EGFR) used in the treatment of non-small cell lung cancer [1]. The purpose of this study was to investigate the effect of gefitinib on lipid abundance in the systemic system and liver of male mice, here we report the results of the study.

Methods: Gefitinib or vehicle alone was dose intravenously (IV) via tail to male C57Bl/6JRj mice (n = 10/group), at 10 mg/kg or orally at 50 mg/kg.  Blood samples were obtained over a 24 h period, via the tail vein.  The resulting plasma was extracted using organic solvent and subject to analysis using reversed – phase UPLC-HRMS in both positive and negative ion mode [2]. Liver tissue from the IV group was homogenized, extracted and analysed by discovery and targeted quantitative LC-MS/MS as well as MS Imaging. The resulting data was subject to peak detection, multivariant statistical analysis (MVA) and lipid identification.

Results: 3900 lipid features were detected in plasma, subsequent MVA showed that both the IV and PO plasma samples exhibited time-related changes in lipids abundance following dosing. The largest effects occurred immediately after gefitinib administration, followed by a gradual return to the lipid predose levels by 24 h post dose. Investigation of the lipids responsible for the variance observed in the data showed that the PI, GPI, PC, LPC, PE and TG were subject to the largest disruption with both transient increases and decreases in relative amounts seen in response to administration of the drug. The pattern of the changes in the relative abundances of those lipids subject to variation appeared to be correlated to the pharmacokinetic profile of gefitinib (and its major metabolites), with the maximum variation being observed at T_max. 292 dysregulated lipids were detected by HRMS analysis of the liver samples, MVA revealed time related changes in the lipid abundances following dosing. The lipids contributing to the observed variance included PCs, TGs, SMs, PEs, DGs, PAs, Cers’, PE, PCs, PGs, SMs, FAs. PCs reduced in abundances after drug administration, reaching a minimum 3 h post dose, before returning to their original abundances at 24 h. In contrast LPCs concentrations increased following dosing reaching a maximum 3 to 8 h post dose before steadily declining to 24 h time point. 433 lipids and gefitinib (plus metabolites) were quantified using HILIC-MS/MS analysis. Gefitinib concentrations peaked at 12 mg/g in the 0.5 h liver samples gradually reducing in subsequent samples before becoming undetectable at 24 h.  MVA showed a clear time-based trajectory from the 0.5 h group through the 1, 3 and 8 h liver samples with the 24 h livers clustering with the 3 h samples (Figure 1) for +ve and -ve ion data.  MVA showed that LPC, FFA, PGs, PCs, LPEs and PEs was significantly elevated post dose, whereas SM, GlcCer and linoleoyl-L-carnitine, TGs were dysregulated in treated mice.  PCs concentrations peaked 3-h post dose, before reducing below the predose levels at 8-h and not returning to initial levels at 24- h. The triglycerides exhibited an immediate increase following dosing reaching a maximum 0.5 h post dose, then rapidly reducing to a minimum at 1 h before increasing to a second, lower, maxima 3 h post dose. TG concentrations then steadily declined to base level at 24 h. Further data analysis indicated a potential relationship between drug plasma concentration and changes in lipid tissue levels for the 5 and 3 TGs, Figure 2. MS Imaging allowed for the detection of both dysregulated lipids and drug/metabolites with observed abundance mirroring the LC-MS data

Conclusion: Administration of gefitinib, either IV or PO, to male C57Bl/6JRj mice resulted in significant changes to the plasma and liver lipid profiles. MVA of the resulting lipidomic data, for both IV and PO, showed a time-related trajectory with major effects association with the high drug concentrations attained soon after drug administration in both plasma and liver. By 24 h post dose the effects of gefitinib on the IV plasma lipid profile had generally dissipated and had largely returned to predose conditions. The PO dose showed similar changes in lipid profiles but a return to the predose condition was not obtained by 24h post dose. Investigation of the lipids responsible for the variance revealed that the PE, PI, PA, PC and LPC profiles were the most disrupted by the administration of gefitinib, indicating a strong link between EGFR and lipid metabolism. The pattern of the changes in relative lipid abundances appeared to correlate to the pharmacokinetics of gefitinib (and its major metabolites). These observations support the concept of a distinct pharmacolipidodynamic relationship between drug exposure and plasma lipid abundance.

Acknowledgements: Andrew Leightner, Nyasha Munjoma, Lee Gethings and Robert Plumb are employees of Waters Corporation, Ian Wilson is a consultant for Waters Corporation. The animal study reported here was funded by Waters Corporation and carried out by Evotec SAS, France, a full management review of the study was performed to ensure that the design conformed to both National and EU guidelines prior to study commencement.

Scores Plot

Lipid Abundance over Time