Development and validation of a liquid chromatographic-tandem mass spectrometric assay for the quantification of cabotegravir and rilpivirine from dried blood spots

Abstract

Background: Dried blood spots (DBS) have been utilized as a blood plasma alternative for therapeutic drug monitoring and pharmacologic analysis. There are analytical and physiochemical considerations in bridging drug concentrations from plasma to DBS. Recently, the long-acting antiretroviral cabotegravir (CAB) has been approved for HIV prevention, and a co-packaged regimen of long-acting CAB and rilpivirine (RPV) has been approved for HIV treatment. Measurement of these drugs in blood collected as DBS may offer increased capacity and flexibility in translational applications.

Methods: Whole blood was spiked with CAB and RPV and spotted on DBS cards. Following extraction and addition of isotopically labeled internal standards, samples were subjected to liquid chromatographic-tandem mass spectrometric (LC-MS/MS) analysis. The method was validated according to regulatory recommendations, and the assay was evaluated in remnant samples from an HIV prevention trial in which paired DBS and plasma samples were collected.

Results: DBS CAB and RPV concentrations were linear from 25 to 20,000 ng/mL and 2-2500 ng/mL, respectively. Precision, accuracy, and matrix effect results were acceptable. DBS RPV demonstrated stability under all tested conditions; DBS CAB showed mean biases of - 23.5% when stored at room temperature for 36 days, and - 18.0% at 40 °C and 100% humidity for two days. DBS measurements for CAB and RPV were an average 54.0% and 14.1% lower, respectively, as compared to paired plasma samples. Derived conversion factors of 1.79 and 1.16 were applied to DBS CAB and RPV measurements, respectively, to estimate plasma concentrations. Estimated plasma CAB and RPV concentrations showed mean biases of 2.2% and 0.6%, respectively. In a CAB clinical trial, application of the conversion factor resulted in agreement between estimated plasma CAB concentrations from DBS and plasma CAB concentrations (y = 1.08x - 79.2, r = 0.932; mean bias of -3.2%; 95% CI: -48.2% to 41.9%).

Conclusions: We developed and validated a novel LC-MS/MS assay for the quantification of CAB and RPV from DBS, and identified conversion factors to estimate plasma concentrations from spotted blood.

Keywords: Cabotegravir; DBS; HIV; LC-MS; Mass spectrometry; Rilpivirine.

Figure 1.

Structures and molecular weights (M.W.) of (A) cabotegravir (CAB; C₁₉H₁₇F₂N₃O₅) and (B) rilpivirine (RPV; C₂₂H₁₈N₆). Asterisks (*) denote positions of isotopically labeled atoms for internal standards.

Figure 2.

Representative chromatograms of the low QC for (A) CAB, (B) CAB-IS, (C) RPV, and (D) RPV-IS extracted from whole blood spotted as DBS. CAB was spiked at 75 ng/mL and RPV was spiked at 6.00 ng/mL.

Figure 3.

Scatter plots with Passing-Bablok regression and Bland-Altman percent bias plots for (A, B) DBS and plasma CAB; (C, D) estimated plasma and plasma CAB; (E, F) DBS and plasma RPV; (G, H) estimated plasma and plasma RPV. Conversion factors of 1.79 and 1.16 were used to estimate plasma CAB and RPV concentrations, respectively, from DBS. For scatter plots, dotted red lines are identity lines. Solid blue lines indicate the Passing-Bablok fit; 95% CI are shaded in blue and flanked by dashed red lines. Regression parameters are found in Table 5. For Bland-Altman plots, the solid blue line is the mean difference between methods; the dashed red lines reflect the 95% CI. Mean biases for DBS CAB, estimated plasma CAB, DBS RPV, and estimated plasma RPV are −54.0%, 2.2%, −14.1%, and 0.6%, respectively.

Figure 4.

CAB correlation analysis from clinical trial samples. Scatter plots with Passing-Bablok regression for (A) DBS and plasma CAB (slope: y = 0.60x – 29.0, r = 0.932) and (B) estimated plasma and plasma CAB (slope: y = 1.08x − 9.2, r = 0.932). (C) Bland-Altman plot for CAB with the application of the 1.79 conversion factor to DBS measurements. The mean percent bias was −3.2% (95% CI: −48.2% to 41.9%).