The pharmacokinetic effect of clarithromycin on the AUC0-12h of linezolid in
multidrug-resistant and extensively drug-resistant tuberculosis patients

The prevalence and mortality of multidrug-resistant and extensively
drug-resistant tuberculosis (MDR- and XDR-TB) are high. The WHO recommends
treatment of MDR- and XDR-TB with a combination of at least two
anti-tuberculosis drugs of Group 5 drugs, e.g. linezolid (LIN) and
clarithromycin (CLA). LIN is a promising drug for the treatment of MDR- and
XDR-TB because of the high in vitro activity against M tuberculosis (MIC of
0.125-0.5mg/L) but its use is limited due to its toxicity and high cost. Dose
reduction has been evaluated in TB patients in an attempt to reduce toxicity to
allow prolonged treatment for 18-24 months. A case series demonstrated CLA
significantly increases LIN AUC0-12h, requiring LIN dosage reduction. CLA, a
potent inhibitor of p-glycoprotein (P-gp) and a well known cytochrome P450
(CYP) 3A4 inhibitor, possibly increases LIN AUC0-12h through a P-gp mediated
interaction, since an in vitro study by Wynalda et al suggest that LIN is not
metabolized by CYP enzymes. In patients, combining LIN and RIF resulted in
decreased LIN serum levels. LIN is possibly a substrate for a P-glycoprotein
mediated interaction. Inhibition of the P-gp efflux pump in the intestinal
membrane could result in an increased P-gp substrate concentration, and
possibly an increased LIN AUC0-12h. We observed increased LIN serum levels in
three cases after combining LIN and CLA of which we described one in a case
report. . Based on the above the hypothesis is formulated that Llinezolid is a
substrate for P-gp mediated interactions.

Patients might benefit from quantifying the interaction between linezolid and
clarithromycin, the subject of this study. When quantified this interaction
could eventually prevent toxicity, such as time- and dose dependant
myelosuppression. Prevention of toxicity could result in not having to cease
one of the few anti-TB drugs that are still effective against MDR/XDR-TB. It
could possibly lead to reduced LIN therapy costs by adjusting LIN dosages in
advance. Finally patients might benefit from a combination of LIN and CLA in
the treatment of MDR/XDR-TB due to possible synergistic activity as shown in in
vitro experiments in different Mycobacteria strains.

We suggest to conduct a prospective pharmacokinetic study in MDR- and XDR-TB
patients to quantify the above described interaction between LIN and CLA.

The primary objective is to quantify the LIN AUC0-12h increase after addition
of 250 mg, or 500 mg CLA compared to baseline (0 mg CLA). The secondary
objectives are to describe the LIN and CLA pharmacokinetic parameters and to
assess short-term safety and tolerability when combining LIN with CLA by
monitoring adverse events (AEs), i.e. gastro-intestinal effects,
hyperlactatemia, haematological abnormalities (thrombocytopenia or anaemia) and
neuropathy. Also to describe observational pharmacokinetic parameters of
anti-TB drugs that are co-administered as part of the continued standard care.

An open label, prospective, single-centre, observational pharmacokinetic
drug-drug interaction study.

Patients will receive 300 mg LIN orally twice a day during six weeks. At week 1
the steady state area under the curve (AUC) 0-12h will be calculated by
obtaining samples via an intravenous catheter at 0, 1, 2, 3, 4, 8, and 12 hours
after LIN dose. At week 1, also a sample will be subtracted in order to
determine P-gp polymorphism, if the subject decided to participate in this part
of the study. At this point, 250 mg CLA once a day will be added to the
therapy. At Week 3, AUC 0-12h will be measured again. Simultaneously to the
seven blood samples, seven saliva samples will be collected if the patient
decided to participate to this optional part of the study. Subsequently, the
CLA dose will be increased to 500mg once daily. At Week 5, AUC 0-12h will be
measured. During a wash-out period of one week the patients will receive LIN,
but no CLA. After one wash-out week, at Week 6, a trough LIN serum
concentration will be measured.

The AUC0-12h will be measured thrice consisting of 4 mL blood samples at 0, 1,
2, 3, 4, 8, and 12h after LIN dosage. Each AUC0-12h will be subtracted via one
intravenous catheter. Two trough samples of 4 mL will be obtained. No extra
visits are needed since the study is conducted with hospitalized patients.
During the study, the patients will continue to receive standard care. There
will be minimal physical and/or physiological discomfort associated with
participation, due to drawing a limited number of blood samples (4 times 7 x 4
mL and two trough samples of 4 mL each).

If the patient decided to participate in the optional part of the study seven
saliva samples will be collected. In order to do this the subject will have to
keep a piece of cotton in his/her mouth during a few minutes. There are no
risks involved in the collection of saliva. The burden for the subjects is
negligible.

Patients could experience adverse events, but these are expected to be limited
due to the short and relatively low exposure to CLA and LIN. Based on previous
cases, we anticipate the LIN 300 mg AUC0-12h to increase approximately 220%
after addition of CLA 500 mg per day, thus becoming comparable to the AUC0-12h
of LIN 600 mg, the recommended dose by the Summary of Product Characteristics
of LIN. CLA is generally well tolerated when used in dosages up to 2000 mg per
day, whereas in this study a maximum dose of 500 mg CLA per day will be
administered. The most commonly occurring adverse reactions, with the highest
frequency at dosages of 4000 mg per day, are gastrointestinal disturbances.