Determination Of Pirfenidone And Metabolites In Rat Plasma By Coupling On-Line Fractionation With LC-MS/MS

By Ming-Luan Chen, Jeff Plomley, Milton Furtado, and Anahita Keyhani

The purpose of this study was to determine concentrations of Pirfenidone (PFD) and its three metabolites—4’-hydroxy-PFD (OH-PFD), 5-hydroxymethyl-PFD (CH2OH-PFD), and 5-carboxy-PFD (COOH-PFD)—in rat plasma using a single assay that couples LC-MS/MS with on-line trapping and fractionation. PFD and its metabolites were extracted from rat plasma by protein precipitation, and the samples were injected for on-line trapping using an Acquity BEH C18 column. Elution from the trapping column was divided into two fractions through multiple valve switching. The first fraction containing CH2OH-PFD and COOH-PFD was eluted with an isocratic mobile phase containing 10% methanol. The trapping column was then switched off-line, and CH2OH-PFD and COOH-PFD were separated using a methanol gradient from 10% to 20%. A second fraction containing OH-PFD and PFD was eluted from the trapping column and separated isocratically with 20% methanol. The analytes were detected in positive ESI-MRM mode using a SCIEX 6500+ TripleQuad, with the duty cycle optimized by period splitting.

Chromatographic separation capacity was achieved by an orthogonal trapping approach, allowing the MS duty cycle to be optimized with > 100 ms per MRM transition, resulting in signal-to-noise ratios greater than 10:1 for all analytes. The organic content was optimized for greater ionization efficiency, particularly for OH-PFD, whose detection limits were 10 times lower than those for PFD and COOH-PFD. The transfer to the analytical phase with 10% methanol also allowed for the resolution of CH2OH-PFD from an ion suppressor, which could not be eliminated using gradient elution alone. The sample run time was kept to 2.5 minutes, including in-line flushing of both the trapping and analytical columns with 100% methanol after each injection. Assay evaluation confirmed precision, accuracy, specificity, matrix effect, and matrix factor met all acceptance criteria, proving that a single assay was feasible for the quantification of all four analytes extracted from rat plasma, despite their dynamic range differing by a factor of 10.