Comment on: Development and validation of a reversed-phase high-performance liquid chromatography assay for polymyxin B in human plasma

Abstract

Sir, We read with great interest the article by Cao et al. on the development of a reversed-phase HPLC assay for polymyxin B in human plasma. The research group from the Facility for Anti-Infective Drug Development and Innovation (FADDI) should be commended for their sustained contribution to understanding the pharmacology of polymyxins over the years. However, before this assay can be adopted widely for pharmacokinetic studies, several issues need to be discussed. First, we corroborate with Cao et al. that the prevalence of multidrug-resistant Gram-negative bacteria is on the rise and polymyxin B is increasingly used as the last viable therapeutic option. The pharmacokinetics of polymyxin B remains poorly understood; the most commonly used dosing regimens are often based on convention rather than being supported by pharmacokinetics and pharmacodynamics. Commercial formulations of polymyxin B contain multiple components. Current quality control procedures focus on overall antimicrobial activity, rather than on the proportion of each component. Therefore, manufacturing lot-to-lot variation in the composition of the polymyxin B components may exist. In addition, patients with multidrugresistant bacterial infections are often given other antimicrobial agents concurrently. Consequently, we agree that using a microbiological assay to detect polymyxin B in clinical samples may not be very satisfactory. Our group has also been working on the pharmacokinetics and pharmacodynamics of polymyxin B over the years. The complete assay is still under development, but several key points have been learnt. Polymyxin B1 was found to be the major component among more than 30 other components (e.g. polymyxin B2, B3 etc.) present in the complex. Using our HPLC-MS assay, a typical chromatogram of polymyxin B (USP) detecting several of these components is shown in Figure 1. In contrast, the chromatography used by Cao et al. (Figure 2 of the article) may not have the desired resolution to distinguish these closely related polymyxin B components. Without an optimal chromatographic separation, using mass-charge ratios to confirm the identity of various polymyxin B components may also be misleading. As shown in Figure 1, polymyxin B1 and B5 have an identical mass-charge ratio, as well as polymyxin B2 and B3. Second, as analytical grade forms of each of these components are presently not commercially available, a relative method of quantification was used. In view of the possible lot-to-lot variations in the relative composition of various polymyxin B components, for the assay to be useful to quantify polymyxin B concentration in a patient, the same mixture of known total quantity (formulation batch) should be used as the reference standard. We are unsure if this is practical as: (i) different patients may be given different manufacturing lots of polymyxin B; and (ii) the same patient may also be given different manufacturing lots of polymyxin B during a course of therapy. As a result, multiple lots of reference standards may have to be used in a study. Finally, it is unclear to us if each polymyxin B component has different antimicrobial activity, pharmacokinetic property and propensity for toxicity. Consequently, the validity of correlating the sums of polymyxin B1 and B2 peaks to ‘total polymyxin B’ may be questionable. By characterizing the pharmacokinetics of total polymyxin B concentrations over time, a more rapid clearance of one component could have been compensated by a more gradual clearance of its counterpart. Until the antimicrobial activity and pharmacokinetics of polymyxin B1 and B2 can be proved to be identical, it may be prudent to characterize them as separate entities. With a better understanding of its pharmacokinetics, we are hopeful that the dosing regimens of polymyxin B could be designed more rationally in the future to improve patient outcomes and avoid toxicity. There is still much to be learnt about polymyxin B. It is our expectation that this communication will encourage clinicians and other investigators to examine these