Simultaneous determination of LY3214996, abemaciclib, and M2 and M20 metabolites in human plasma, cerebrospinal fluid, and brain tumor by LC-MS/MS

Abstract

A sensitive and rapid liquid chromatography tandem mass spectrometry (LC-MS/MS) method was established for the quantification of total and unbound concentrations of LY3214996, an extracellular signal-regulated kinase inhibitor; abemaciclib, a cyclin-dependent kinase 4/6 inhibitor; and abemaciclib active metabolites, M2 and M20, in human plasma, brain tumor, and cerebrospinal fluid samples. The method was validated over a concentration range of 0.2-500 nM within a total run time of 3.8 min using isocratic elution on a Kinetex™ F₅ column. Detection was performed on a Sciex QTRAP 6500+ mass spectrometer employing multiple reaction monitoring mode under positive electrospray ionization. The intra- and inter-batch accuracy as well as the precision of the method for all matrices was within ±20% and ≤20% at the lower limit of quantification, and within ±15% and ≤15% for other quality control levels for all analytes. The unbound fractions of drugs and metabolites in spiked and patient samples were determined using an optimized equilibrium dialysis. The validated method was successfully applied in a phase 0/2 clinical trial to assess the central nervous system penetration of LY3214996 and abemaciclib.

Keywords: Abemaciclib; Brain tumor penetration; Equilibrium dialysis; LC-MS/MS; LY3214996.

Graphical abstract

Fig. 1

Chemical structures of LY3214996, abemaciclib, and abemaciclib metabolites M2 and M20 as well as their related stable isotope-labeled internal standard (IS).

Fig. 2

Representative product ion scan spectra of (A) LY3214996, (B) abemaciclib, (C) abemaciclib metabolite M2, and (D) abemacilclib metabolite M20 along with their related IS.

Fig. 3

Representative ion chromatograms of (A) blank plasma samples, (B and C) plasma samples spiked with standards LY3214996 (m/z 454.1 → 367.1), M20 (m/z 523.3 → 409.2), M2 (m/z 479.2 → 393.0), and abemaciclib (m/z 507.3 → 393.0) at (B) lower limit of quantification (LLOQ), and (C) upper limit of quantification (ULOQ), and (D) plasma samples spiked with IS, i.e., LY3214996-IS (m/z 458.1 → 371.1), M20-IS (m/z 531.3 → 409.2), M2-IS (m/z 486.2 → 400.1) and abemaciclib-IS (m/z 512.3 → 393.0).

Fig. 4

Fraction unbound of drugs and metabolites determined in (A) pooled human plasma and brain homogenate spiked with analytes and (B) plasma and glioblastoma (GBM) samples of selected patients. The 96-well microdialysis plates were used to determine the unbound concentration of analytes in the samples at an optimal 16 h equilibrium time. GBM (NE) and GBM (E) are gadolinium contrast non-enhancing and enhancing regions of GBM, accordingly, which were resected 7–9 h after the last dose administration. (A) The experiments were performed at least three times with each sample in triplicate at both 20 and 200 nM analyte concentrations. (B) The fraction unbound values were determined in five patient samples. Results are presented as mean ± SD.

Fig. 5

Median (A) total (K_p) and (B) unbound (K_p,uu) tumor-to-plasma partition coefficients of LY3214996, abemaciclib, and abemaciclib metabolites M2 and M20 in gadolinium contrast non-enhancing (GBM (NE)) and enhancing (GBM (E)) regions of GBM, as determined in five patients. Colored bar represents the median value of tumor-to-plasma partition coefficients in five patients, while the open circle represents the individual patient result.