Development and validation of a liquid chromatography tandem mass spectrometry assay for the analysis of bedaquiline and M2 in breast milk

Abstract

Objective: To develop and validate an assay for the analysis of bedaquiline and its M2 metabolite in human breast milk.

Methods: The analytes were extracted using solid phase extraction following protein precipitation. Quantification was performed with liquid chromatography coupled with tandem mass spectrometry. Chromatographic separation was achieved using gradient chromatography on a Poroshell 120 SB-C18 analytical column at 40 °C, with a flow rate of 350 µL/minute and a total run time of eight minutes. An AB Sciex 3000 mass spectrometer with electrospray ionization in the positive mode was used for detection, employing multiple reaction monitoring scan mode. Bedaquiline-d6 and M2-d3-¹³C were used as internal standards.

Results: Calibrations curves for bedaquiline and M2 exhibited quadratic (weighted 1/x concentration) regressions over the respective concentration ranges of 0.0780 to 5.00 µg/mL and 0.0312 to 2.00 µg/mL. Inter- and intra-day validation accuracies ranged between 96.7 % and 103.5 % for bedaquiline, and 104.2 % to 106.5 % for M2, with a coefficient of variation below 9.2 % for both compounds.

Conclusion: The developed assay demonstrated selectivity and robustness, enabling differentiation between bedaquiline and M2 within the context of endogenous compounds from six separate lots of breast milk samples. Successful application was observed in the analysis of breast milk samples sourced from patients treated for multidrug-resistant tuberculosis within a clinical study setting.

Keywords: Bedaquiline; Breast milk; Drug resistance; LC-MS/MS; Tuberculosis.

Fig. 1

Fragment mass spectra of bedaquiline (A) indicating the precursor ion at 555.2 and the most prominent product ion at 58.2, and M2 (B) indicating the precursor ion at 541.2 and the most prominent product ion at 480.0.

Fig. 2

Representative calibration curves for bedaquiline (A) and M2 (B) in breast milk. The calibration curves were fitted using a quadratic regression model weighted by 1/x with concentration ranges of 0.0780 to 5.00 µg/mL and 0.0312 to 2.00 µg/mL for bedaquiline and M2, respectively. The regression equation is as follows f(x) = ax2 + bx + c; bedaquiline: y = -0.402 x2 + 13.4 x  + 0.107 (r = 0.9997) and M2: y = 0.123 x2 + 11.1 x  + 0.0366 (r = 0.9993).

Fig. 3

Overlay example of LLOQ and blank chromatograms (A – bedaquiline, B – M2) of one breast milk lot. The LLOQ is shown in blue, and the blank is in red. Monitored for BDQ and M2 mass transitions 555.1 to 58.2 and 541.1 to 480.3, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

A zoomed-in view of a chromatogram showing the separation of the analytes: M2 elutes at 4.05 min along with its internal standard, M2-d3-¹³C; bedaquiline and its internal standard, bedaquiline-d6 elute at 4.15 min. The M2 quantifier and qualifier are represented in blue and red, respectively; green represents M2-d3-¹³C. Bedaquiline quantifier and qualifier are represented in grey and light blue, respectively; purple represents bedaquiline-d6. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)