Validation and application of a quantitative LC-MS/MS assay for the analysis of first-line anti-tuberculosis drugs, rifabutin and their metabolites in human breast milk

Breast milk is the preferred method of infant nutrition. Breastfeeding infants born to mothers treated for TB may be at risk of drug toxicity through breast milk exposure, or potentially be vulnerable to select for drug resistance with low level drug exposure. Except for isoniazid, the quantification of first-line TB drugs including rifabutin in breast milk has not been previously described and will provide much-needed insight to TB drug exposure in breastfeeding infants. We developed and validated a novel method to quantify several first-line TB drugs and their major metabolites in breast milk. Accuracy and precision were assessed during three consecutive, independent validation batches over a calibration range of 0.300–30.0 µg/mL for isoniazid and ethambutol, 0.150–15.0 µg/mL for acetyl isoniazid, desacetyl rifampicin, rifampicin, and pyrazinamide, 0.0150–1.50 µg/mL for rifabutin, and 0.00751–0.751 µg/mL for deacetyl rifabutin in breast milk. The method was reproducible for all analytes when using breast milk from six different sources and was not influenced by matrix effects with a mean regression precision (CV(%)) ranging between 1.0 and 2.8. The average recovery of analytes from the matrix was 76.7–99.1%, with a CV(%) between 0.4 and 4.4, while the average process efficiency was between 74.4 and 93.1% with a CV(%) between 1.9 and 8.3. Although only acetyl isoniazid, isoniazid, ethambutol, and pyrazinamide were successfully assayed in breast milk, samples taken from mothers treated for rifampicin-resistant TB and the inclusion of all first-line TB drugs, including rifabutin in the assay development and validation process will allow future quantification of these analytes in breast milk.