PK, PD, and TB
Posted by Dr. Chris Bode on January 16, 2012
I’m always looking for examples of the distinction between pharmacokinetics (PK) and pharmacodynamics (PD), partly to blog about and partly as new material for the undergrads I lecture to in the Pharmaceutical Product Development program at West Chester University. I came across an interesting one recently, although maybe not exactly PK vs. PD…more of an indirect connection.
In a paper in the December 15, 2011 issue of The Journal of Infectious Diseases (http://jid.oxfordjournals.org/content/204/12/1951.full.pdf+html), researchers torpedoed a widely held and highly influential assumption in the field of tuberculosis (TB) therapy. The assumption is that poor compliance by TB patients is responsible for the rise of multidrug-resistant TB. This has led the World Health Organization (WHO) to require that healthcare workers directly witness each patient taking every dose of a combination of drugs over several months of therapy, a practice that is more expensive than the drugs themselves in many parts of the world. However, multidrug resistant strains of TB continue to emerge in spite of >98% adherence to the direct-observation regimen.
The researchers used in vitro, hollow-fiber systems inoculated with M. tuberculosis, including a drug-resistant strain in some cases. The cultures were maintained for 4 or 8 weeks, with antibiotics (isoniazid, rifampin, and pyrazinamide) administered daily by programmable syringe pumps (and gradual dilution with fresh culture medium to model PK clearance). Antibiotic non-adherence (varying degrees and different patterns) was modeled by skipping the drug infusion on some days. The emergence of drug-resistant bacteria was monitored by sampling the cultures throughout the study. Therapeutic failure (defined as incomplete killing of bacteria) occurred only with >70% non-adherence, but the multidrug-resistant strain never reached 1% of the total bacterial population under any conditions. Thus, non-adherence was not a sufficient condition for emergence of multidrug-resistant TB.
To test an alternative hypothesis, the researchers performed Monte Carlo simulations to determine the effect of inter-subject PK variability on efficacy (bacterial killing) and the emergence of drug resistance. Based on previous reports of the variability in the PK of rifampin and isoniazid in a large population of patients, the authors concluded that a small percentage of patients clear rifampin rapidly enough that they are effectively treated with isoniazid monotherapy, which is known to promote the emergence of multidrug-resistant TB. Thus, they hypothesize that PK variability, not non-adherence, is to blame for multidrug resistance among TB patients. Furthermore, they propose that monitoring the blood levels of antibiotics in patients (enabling appropriate adjustment of doses) would be more effective than direct observation of therapy. I’m not sure whether monitoring blood levels would cost less than direct observation or even be feasible in many remote areas, but it was interesting to see the importance of PK recognized in an unexpected circumstance.
Back to PK vs. PD: in the sense that any drug that you don’t take can’t be effective, one could consider non-adherence an aspect of PD. As for PK, rest assured that Absorption Systems (www.absorption.com) knows it inside and out…PK is our business, and our portfolio of preclinical ADME services is both broad and deep. Contact us for help with your ADME challenges.