Therapeutic Drug Monitoring in the Treatment of Tuberculosis

Transcription

1 CURRENT OPINION Drugs 2002; 62 (15): /02/ /$30.00/0 Adis International Limited. All rights reserved. Therapeutic Drug Monitoring in the Treatment of Tuberculosis Charles A. Peloquin 1,2 1 Department of Medicine, National Jewish Medical and Research Center, Denver, Colorado, USA 2 Schools of Pharmacy and Medicine, University of Colorado, Denver, Colorado, USA Abstract Therapeutic drug monitoring (TDM) is a standard clinical technique used for many disease states, including many infectious diseases. As for these other conditions, the use of TDM in the setting of tuberculosis (TB) allows the clinician to make informed decisions regarding the timely adjustment of drug therapy. Such adjustments may not be required for otherwise healthy individuals who are responding to the standard, four-drug TB regimens. However, some patients are slow to respond to treatment, have drug-resistant TB, are at risk of drug-drug interactions or have concurrent disease states that significantly complicate the clinical situation. Such patients may benefit from TDM and early interventions may preclude the development of further drug resistance. It is not possible to collect multiple blood samples in the clinical setting for logistical and financial reasons. Therefore, one typically is limited to one or two time points. When only one sample can be obtained, the 2-hour post-dose concentrations of isoniazid, rifampin, pyrazinamide and ethambutol are usually most informative. Unfortunately, low 2-hour values do not distinguish between delayed absorption (late peak, close to normal range) and malabsorption (low concentrations at all time points). A second sample, often collected at 6-hour post-dose, can differentiate between these two scenarios. The second time point can also provide some information about clearance and half-life, assuming that drug absorption was nearly completed by 2 hours. TDM requires that samples are promptly centrifuged, and that the serum is promptly harvested and frozen. Isoniazid and ethionamide, in particular, are not stable in human serum at room temperature. Rifampin is stable for more than 6 hours under these conditions. During TB treatment, isoniazid causes the greatest early reduction in organisms and is considered to be one of the two most important TB drugs, along with rifampin. Although isoniazid is highly active against TB, low isoniazid concentrations were associated with poorer clinical and bacteriological outcomes in US Public Health Services (USPHS) TB Trial 22. Several earlier trials showed a clear dose-response for rifampin and pyrazinamide, so low concentrations for those two drugs also may correlate with poorer treatment outcomes. At least in USPHS TB Trial 22, the rifampin pharmacokinetic parameters were not predictive of the outcome variables. In contrast, low concentrations of unbound rifapentine may have been responsible, in part, for the worse-than-anticipated performance of this drug in clinical trials. The second-line TB drugs, including p-aminosalicylic acid, cycloserine and

2 2170 Peloquin ethionamide, are relatively weak TB drugs. Under the best conditions, treatment with these drugs takes over 2 years, as opposed to 6 to 9 months with isoniazidand rifampin-containing regimens. Therefore, TB centres such as National Jewish Medical and Research Center in Denver, CO, USA, measure serum concentrations of the second-line TB drugs early in the course of treatment. That way, poor drug absorption can be dealt with in a timely manner. This helps to minimise the time that patients are sputum smear- and culture-positive with multidrug-resistant TB, and may prevent the need for even longer treatment durations. Patients with HIV are at particular risk for drug-drug interactions. Because the published guidelines typically reflect interactions only between two drugs, these guidelines are of limited value when the patient is treated with three or more interacting drugs. Under such complicated circumstances, TDM often is the best available tool for sorting out these interactions and placing the patient the necessary doses that they require. TDM is only one part of the care of patients with TB. In isolation, it is of limited value. However, combined with clinical and bacteriological data, it can be a decisive tool, allowing the clinician to successfully treat even the most complicated TB patients. The treatment of tuberculosis (TB) differs from that of most infectious diseases in several ways. First, the patient can be latently infected with TB, exhibiting no signs or symptoms of the disease other than a positive skin test reaction. [1] This differs considerably from what is called colonisation for selected Gram-positive or Gram-negative organisms. Mycobacterium tuberculosis is never considered an innocent bystander, nor is it considered part of the normal human flora on any surface, in any organ or in any secretion. Latent TB infection is an infection that has been contained by the immune system but one that can reactivate given the right conditions. [2] Second, the treatment of active TB disease always requires the use of multiple antibacterial agents. [1] In comparison, most bacterial infections are treated with monotherapy. Third, in comparison to antibacterial agents, the mechanisms of action for several of the TB drugs are poorly understood. [3-6] Progress has been made in this area for some of the drugs, while for others, our knowledge has not improved substantially for decades. [7,8] Fourth, the treatment of TB disease can be divided into two parts, in some ways similar to cancer chemotherapy. The initial, intensive phase typically uses at least four drugs and is designed to drastically reduce the total body burden of TB. [1,9-11] Presumably, many of these organisms are extracellular (that is, not within macrophages) and are multiplying approximately once daily. Next, generally after 2 months of intensive treatment, the continuation phase is designed to eliminate the persisters. These are organisms that survive the initial phase of treatment for reasons that are poorly understood. Furthermore, it is unclear how similar or dissimilar these organisms are to those found during latent infection. [12-15] Perhaps they are particularly hardy or they have evoked some protective response that allows them to survive, perhaps at a slower rate of replication. Alternatively, they may be hiding in places where the drugs and the immune system are unable to deal with them. Other explanations also may exist. Because so little is known about these persisters, it is difficult to design optimal treatment regimens for them. Happily, the majority of patients who receive a total of 6 months of treatment for drug-susceptible TB will respond completely to treatment. [1,16] In general, more than 90%, and as high as 98% of patients, are completely cured by standard regimens. The most frequently used regimens are isoniazid, rifampin, pyrazinamide, and either ethambutol or streptomycin. [1,16,17] Isoniazid and

3 Therapeutic Drug Monitoring in TB 2171 rifampicin (rifampin) are the heavy weights in these regimens. [10,11] Isoniazid causes the greatest early reduction in organisms and this can be measured over 2 days as a rapid decline in the colony forming units (cfu) found in the sputum of the patient. Rifampicin has the unique properties of a rapid onset of action once in contact with M. tuberculosis, and the ability to kill M. tuberculosis under a variety of conditions, including intracellularly (along with isoniazid) and during the persister phase. Both drugs show some dose response, although with standard doses this is most relevant for rifampicin. [11] The typical adult rifampicin dose of 600mg appears to be at the low end of the effective range and there is ample reason to believe that higher daily doses would be more effective. [6,11,18] Higher intermittent doses (twice or once weekly) have been studied extensively but these intermittent regimens are associated with an unacceptably high rate of flu-like symptoms. Fortunately, higher daily doses of rifampicin are not associated with this flu-like syndrome. [1,16,18-20] Theissueofhighdose daily rifampicin merits additional study. [11] Given such high success rates, what could be the role of therapeutic drug monitoring (TDM) in the treatment of TB? This paper explores those circumstances where the risk of treatment failure or toxicity is increased, and where higher than usual doses may be appropriate. 1. The Otherwise Healthy Patient with Tuberculosis Patients with uncomplicated, pulmonary TB usually respond to standard treatment. [1,16] Here, the primary concern is compliance with the regimen. Directly observed treatment (DOT) is a key component of TB treatment. [16,21] With DOT, a person associated with the public health department watches the patient take their medications, and confirms that the tablets and capsules were swallowed. This simple and cost-effective procedure can be done at the TB clinic, at the patient s home or at any meeting place agreed upon by the two parties. Because even short course TB treatment is 6 months, the potential for patient default is high. However, it is incumbent upon the healthcare providers, not the patient, to recognise and plan for this problem. With successfully delivered TB treatment, roughly 80% of patients will be sputum smear and culture negative by 2 months. [16,22] Of course, the culture results take up to 6 weeks to become available, so it is often the smears that provide the first clues to the success of TB treatment. Most patients are sputum smear negative by 2 weeks, and certainly 3 weeks of treatment, when they receive the standard regimen. Those patients who are slower to respond may be more likely to have unsuccessful treatment. [23] Such slow responders may be candidates for TDM, as discussed by several authors. [6,24-26] Some clinicians prefer to observe the patient and to continue with the standard doses of the TB drugs. Others prefer to check serum drug concentrations early during treatment, hoping to head off the potential development of drug-resistance. A key component of the decision-making process is the clinical status of the patient. In this section, otherwise healthy patients are described. If the patient is making steady progress (fevers have ended, weight is stable if not increasing, no more night sweats, diminished cough), then careful observation of slow converters may be reasonable. However, the author would advocate checking serum concentrations in patients who are slow to respond clinically or in those whose condition is worsening. 2. Performing Therapeutic Drug Monitoring The general approach for TDM in various patient populations is the same. Additional details are provided in sections 3 and 4 for specific types of patients and for each drug. Since the trough concentrations for many of the TB drugs are below the limit of detection for the assays, and because the peak concentration may be more important for several of these drugs, a 2-hour post-dose sample can be collected to estimate the peak concentration. For a few drugs, such as ethambutol and rifabutin,

4 2172 Peloquin a 3-hour time point may approximate the peak better. Given the variability of oral absorption, a single time point may miss the actual peak concentration. Therefore, a second sample, typically 6-hours post-dose, allows one to capture information on the rate and completeness of absorption. It also can provide information regarding the elimination of drugs that have short half-lives, such as isoniazid and rifampicin, provided that absorption was completed by 2-hours post-dose. The normal pattern for serum concentrations is that the 2-hour value is substantially higher than the 6-hour for isoniazid and rifampicin. Should the 2-hour and 6-hour values be roughly the same, perhaps somewhat below the expected ranges, or should the 6-hour values be higher than the 2-hour, delayed absorption is most likely. In these situations, it is possible that the actual peak occurred between the two blood draws. Here, one may recommend that the patient take the drugs on an empty stomach, which clearly is preferred for isoniazid and rifampicin. Finally, should both values be well below the expected ranges, malabsorption is most likely, and consideration should be given to using higher doses of the drugs. For most of the TB drugs, relationships between the dose (and by way of extension, the serum concentrations) and toxicity are not present. Exceptions include daily pyrazinamide and hepatotoxicity(morecommonwithdailydosesabove40 mg/kg), ethambutol and ocular toxicity (more common with daily doses above 30 mg/kg), and cycloserine (more common with serum concentrations above 35 µg/ml). It is important to bear these relationships in mind when increasing doses. However, in the face of malabsorption, systemic toxicity is unlikely, as the drugs are not reaching the circulation. It is quite reasonable to use higher than standard doses in the face of documented malabsorption. Serum concentrations can be rechecked after the dose increases to verify achievement of the desired concentrations. The concentrations required for effective therapy are only partially known. These are described in section 4 for each drug. Detailed pharmacokinetic-pharmacodynamic data from human studies are lacking for the TB drugs. Precise targets for peak serum concentrations (C max ) relative to the minimal inhibitory concentration (MIC) [C max : MIC], or time above MIC, are not available from human studies. In general, the author targets the concentrations achieved in healthy volunteers under controlled, phase I study conditions. These normal ranges are predictably tolerated and represent the best-case scenario for the standard doses of the drugs. It is well known that the standard doses are generally effective, so these concentrations also should be effective. Although lower concentrations may be associated with effective therapy in some patients, there is no clear guidance on how low can you go. So, once the decision is made to use TDM, it seems logical to aim for the normal ranges. If you can achieve such concentrations, then it would appear unlikely that poor drug absorption is the primary reason for the poor clinical response, and a search for other reasons should be initiated or continued. In rare situations, with desperately ill patients, higher concentrations than normal may be warranted. Because there are no published data to guide these rare decisions, clinical judgement is the only guide available. Samples should be promptly centrifuged, and the serum harvested and frozen as soon as possible after the blood draw. This is critical for isoniazid and ethionamide, since these two drugs are not stable at room temperature in whole blood or in serum. Rifampicin is somewhat more stable, and most of the other TB drugs are stable for 24 hours at room temperature in whole blood or in serum. Assays should be very specific for the drugs of interest. In general, high-performance liquid chromatography (HPLC) and gas chromatography (GC) are preferred. Commonly used detection systems (ultraviolet or fluorescence detection for HPLC, mass spectrometry for HPLC or GC) work well, provided that extensive interference checks are performed prior to validating the assays. Guidance is available from the College of American Pathologists, and from the US Food and Drug Ad-

5 Therapeutic Drug Monitoring in TB 2173 ministration, regarding proper laboratory techniques. 3. More Complicated Patients The recent US Public Health Service (USPHS) TB Study 22 compared the standard regimen, including twice-weekly isoniazid plus rifampicin in the continuation phase, with once-weekly isoniazid plus rifapentine in the continuation phase. [23] Isoniazid plus rifampicin proved to be somewhat more effective, although the failure and relapse rates were higher than expected in both regimens. Patients with bilateral pulmonary disease, cavitary lesions, low bodyweight and, in particular, those patients who were White, were more likely to have a poor outcome. These patients should be followed carefully and, if they are slow to respond, TDM may be helpful in ensuring that drug delivery is adequate. [27,28] It is very clear from the discussion thus far that the treatment of TB is multi-factorial and that serum concentrations represent only one of several factors that need to be considered. [6,29,30] The clinician needs to consider the condition of the patient, the extent of disease, the susceptibility of the organisms once it becomes available, and the rapidity of the clinical and bacteriological response. Using all of these factors, clinicians can decide if knowledge of the serum concentrations will allow them to intervene on behalf of the patients. As with TDM for any drug, the clinician must decide the patient s need for drug. [31] Desperately ill patients may require higher-than-usual doses in order to achieve the desired effect. In the face of dire consequences, the clinician and patient may be forced to accept a higher risk of toxicity (depending on the particular drug) in order to save the life of the patient. With TB, such situations may include TB in the intensive care patient, TB meningitis and miliary TB. [1,16] Other factors may weaken the relationships between the calculated pharmacodynamic parameters and in vivo activity in selected clinical cases. If the organisms are in areas where some of the drugs do not penetrate well, such as in the cerebrospinal fluid, aqueous humour or bone, then the pharmacodynamic relationships may not hold up. [6,32,33] Also, in selected areas of infection, such as abscesses and empyemas, the ph may decrease or the oxygen content may decrease. These factors could reduce the activity of certain drugs, including the aminoglycosides, as has been demonstrated for bacterial infections. Presumably, the activity of streptomycin would be reduced in the treatment of tuberculous empyema, while the activity of pyrazinamide may be enhanced by the low ph. Detailed studies of the penetration of the TB drugs into such areas are lacking, so the current approach is to use the same regimen as indicated for more typical lesions, and to follow the patient carefully for clinical and bacteriological response. In some circumstances, including TB meningitis, treatment duration may be extended by 3 to 6 months. This is explained further in the American, Canadian and British TB treatment guidelines. [17] 3.1 Patients with Diabetes Mellitus Patients with diabetes are more likely to experience gastrointestinal (GI) problems, including gastroparesis. [34] This can produce either delayed absorption of drugs or malabsorption. Much of the available data are anecdotal and not suitable for statistical analysis. In USPHS TB Study 22, patients with diabetes had somewhat lower concentrations than those without diabetes. However, given the study size and variability within all groups, the differences were not statistically significant. Because patients with diabetes frequently have other medical conditions, including coronary artery disease and chronic renal failure, some patients with diabetes are particularly fragile. For these complicated patients, it may be prudent to check serum concentrations early to ensure adequate therapy. [35] The approach is as described in section Patients Co-Infected with HIV HIV-positive patients are more likely to progress to active disease if exposed to or previously infected with TB. [1] Active TB is an AIDS-defin-

6 2174 Peloquin ing illness. Patients with HIV infection represent a continuum of relatively healthy to seriously ill patients. They may have other concurrent opportunistic infections, they may be taking a large number of drugs and they may have various forms of malabsorption. [36-38] Therefore, it is not surprising that some patients with AIDS malabsorb TB medications. There have been a number of studies, with conflicting results. [39-53] Some of the studies used minimal numbers of samples and so it is possible that delayed absorption was misidentified as malabsorption in some cases. Data from North America has tended to show a greater difference in TB drug serum concentrations between HIV-positive versus HIV-negative patients than studies conducted in Asia or Africa. It is not known if this is a coincidence or if this suggests an unidentified co-factor that is present in North American patients. As with patients with diabetes in USPHS TB Study 22, the HIV-positive patients had somewhat lower concentrations but the differences were not statistically significant. Four HIV-positive patients relapsed with rifampicin-resistant TB and the one patient for whom blood samples were available had very low serum concentrations of isoniazid. Unfortunately, samples were not available for the three other patients. Such development of drugresistant TB has been described previously in AIDS patients. Therefore, as with other complicated patients, AIDS patients with TB should be followed carefully. Some clinicians will elect to watch and wait, while others will check the serum concentrations early in an attempt to head off problems. [6,29,51] AIDS patients with TB have an additional huge problem that of drug-drug interactions. [51,54-57] In particular, decisions regarding the selection of rifamycin and the concurrent use of HIV drugs that inhibit or induce hepatic microsomal enzymes must be made carefully. In the author s opinion, TDM should be used to verify the adequacy of both the TB drug doses and the HIV drug doses, especially the protease inhibitors and non-nucleoside reverse transcriptase inhibitors (NNRTI). [58-62] Both diseases require long durations of treatment and it seems to make sense that you should use the correct doses for such critical therapy. Also, when one considers the cost of the HIV drugs, TDM appears to be a relatively inexpensive way to verify the correct doses. The approach for monitoring the TB drugs is as described in section 2. For the protease inhibitors and NNRTIs, a trough value appears to be most valuable for assessing the adequacy of the dose and frequency. Peak concentrations can provide additional information, including that related to certain toxicities, such as indinavir-associated nephrolithiasis and ritonavir-associated neuropathies. An expanded discussion on this topic has been published recently. [62] 3.3 Patients with Renal Failure Patients with renal failure warrant special attention. Chronic renal failure compromises the immune system. Patients receiving dialysis are far morelikelytogoontohaveactivetbdiseasethan those with normal renal function, whether they are latently infected or newly exposed to TB. [63-66] Patients with renal failure comprise three groups: those with poor renal function who are not on dialysis, those receiving haemodialysis and those receiving peritoneal dialysis. All of these patients are at risk of accumulating some of the TB drugs, possibly leading to adverse drug reactions (ADR). [33,67,68] In particular, ethambutol and streptomycin require renal elimination, and their accumulation may be associated with ocular toxicity (ethambutol) or oto- or vestibular toxicity (streptomycin). Reducing the frequency of dose administration of ethambutol and streptomycin can reduce the risk of these toxicities. In general, ethambutol should not be given more than three times per week (after dialysis), and streptomycin should be given two or three times per week. Neither drug is well cleared by dialysis and streptomycin, in particular, does not have a hepatic component to its clearance. Among the second-line TB drugs, cycloserine relies on renal clearance and the accumulation of this agent can lead to central nervous system

7 Therapeutic Drug Monitoring in TB 2175 (CNS) toxicities. Unlike ethambutol and streptomycin, cycloserine is cleared by haemodialysis. Furthermore, certain metabolites, including pyrazinoic acid, 5-hydroxypyrazinoic acid, and acetyl-para-aminosalicylic acid also require renal clearance. [32,33] The toxicity of these metabolites is not known nor is their contribution to the toxicity profile in combination with their parent drugs. So, these factors must be considered when administering these agents. In general, for patients with renal failure, we recommend giving standard doses (those typically given daily) no more than 3 times weekly for ethambutol, aminoglycosides and pyrazinamide. The approach to TDM is similar to that described in section 2. One difference may be the addition of a post-dialysis sample to verify the trough concentration and the potential for drug accumulation. Because of drug redistribution postdialysis, it may be desirable to wait approximately an hour post-dialysis before collecting the trough concentration. Specific data on post-dialysis rebound of serum concentrations for the TB drugs are not available. 3.4 Patients with Hepatic Dysfunction Isoniazid, rifampicin, pyrazinamide, ethionamide and p-aminosalicylic acid (PAS) are predominantly cleared hepatically. [32,33] Unfortunately, the degree of compromise in hepatic drug clearance cannot be predicted based only on the serum tests of liver function (AST, ALT, bilirubin, etc.). Furthermore, patients with either hepatic or renal dysfunction often experience nausea and vomiting. This can produce the dual problems of malabsorption and reduced drug clearance. Therefore, it is entirely reasonable to check serum concentrations in patients with significant renal or hepatic dysfunction, so that each patient gets the correct, individualised dose. The approach would be the same as for patients without hepatic disease, described earlier in this section. 4. Specific Drugs Table I shows the usual serum concentrations associated with standard doses of the TB drugs. Additional detail for each drug can be found in the section for each drug below (4.1 to 4.10) and in the references cited. 4.1 Isoniazid Isoniazid is generally absorbed quickly from the GI tract, with C max occurring 1 to 2 hours postdose when isoniazid is given on an empty stomach. [32,33,69,70] Food delays and reduces the absorption of isoniazid. In particular, high-fat meals cause a 51% drop in C max. Therefore, isoniazid shouldbegivenonanemptystomach.ifonlyisoniazid is being measured, 1- and 4-hour concentrations may effectively pick up C max and most potentially delay absorption. Since isoniazid routinely is given with other agents that are somewhat more slowly absorbed, a 2- and 6-hour post-dose sampling strategy seems reasonable. In the USPHS TB Study 22, low isoniazid plasma concentrations were associated with treatment failure and relapses in those patients who received once-weekly isoniazid and rifapentine. [71] Therefore, patients who receive sparse regimens such as this appear to be good candidates for TDM. As noted in section 2, the TB literature does not provide exact guidance regarding dose adjustment. Therefore, we pursue the normal ranges for the TB drugs. With isoniazid, we target a peak concentration of 3 to 5 µg/ml after a 300mg daily dose and 9to15µg/ml after a 900mg biweekly dose. We generally recommend a dose increase if the peak is less than 2 µg/ml for the 300mg dose and less than 7 µg/ml with the 900mg dose. Isoniazid is hepatically cleared and generally dose adjustment is not required in patients with renal failure. Some patients may experience CNS or peripheral neuropathies, and in these selected patients, less frequent dose administration of isoniazid, especially after 2 weeks of treatment, may be appropriate. Patients with renal failure should receive pyridoxine 10mg every day to prevent these toxicities.

8 2176 Peloquin Table I. Pharmacokinetic parameters of the antituberculosis drugs Drug Usual adult dose Usual serum C max Usual serum t max Normal serum t1 2 Isoniazid 300mg qd 3-6 µg/ml h Polymorphic: 900mg biwk 9-18 mg/ml fast: 1.5h slow: 4h Rifampin 600mg qd 8-24 µg/ml 2h 3h Rifabutin 300mg qd µg/ml 3-4h 25h Rifapentine 600mg qd 8-30 µg/ml 5h 15h Pyrazinamide 25 mg/kg qd µg/ml 1-2h 9h 50 mg/kg biwk µg/ml Ethambutol 25 mg/kg qd 2-6 µg/ml 2-3h Biphasic: 50 mg/kg biwk 4-12 µg/ml A: 2-4h B: 12-14h Streptomycin 15 mg/kg qd µg/ml a h IM 3h 25 mg/kg biwk µg/ml a End infusion IV p-aminosalicylic acid 4000mg bid µg/ml (granules) 4-8h (granules) 1h Clofazimine 100mg qd µg/ml 2-7h Biphasic: A: 10d B: several wks a Calculated C max using linear regression to 1 hour post IM dose or end of IV infusion. Streptomycin range also applies to amikacin, kanamycin, and capreomycin when given at the same doses. bid = twice daily; biwk = twice weekly; C max = peak serum drug concentration; IM = intramuscular; IV = intravenous; qd = every day; t1 2 = serum half-life; t max =timetoc max. 4.2 Rifampicin (Rifampin) Rifampicin absorption is potentially the most variable among the TB drugs. [32,33,69,72-74] C max is reduced and the time of maximum concentration (t max ) delayed by high-fat meals, so the drug should be given on an empty stomach whenever possible. The formulation of rifampicin with isoniazid and pyrazinamide modestly reduces rifampicin absorption, and in the Aventis product, this generally is compensated for by the slightly higher dose typically given. [75] Some formulations across the globe have been more problematic, prompting a WHO testing program for the these fixed-dose combination products. [76] As noted above, rifampicin clearly exhibits a dose response and higher doses (1200 to 1800mg daily) should be studied in the context of current four drug regimens to see if more patients can be rendered sputum smear and culture negative before they are placed on intermittent regimens. [11,18] A 2- and 6-hour sampling strategy captures most C max values (typically at 2 hours) and potentially delayed absorption for rifampicin. The latter time point allows one to make clinical decisions regarding dose. Delayed but normal concentrations do not require dose adjustment, but malabsorption, especially C max values less than 4 µg/ml, should prompt a dose increase. This allows one to better take advantage of the dose response of rifampicin. Furthermore, every day rifampicin administration rarely exhibits dose-related toxicity, with most ADR being idiosyncratic. In contrast, doses above 900mg and given once or twice weekly were associated with higher rates of the flu-like syndrome. This appears to be associated with anti-rifampicin antibodies that build up when the drug is given intermittently. [16,19,77] This generally can be avoided with every day or everyother-day dose administration, or potentially by using another rifamycin that has a longer elimination

9 Therapeutic Drug Monitoring in TB 2177 half-life. Like isoniazid, rifampicin is predominantly cleared by the liver so dose adjustment is not needed in patients with renal failure. In most patients, we target a peak rifampicin concentration of8to24µg/ml after a 600mg dose. We generally recommend a dose increase if the peak is less than 6 µg/ml. 4.3 Rifabutin Rifabutin can be used instead of rifampicin in situations where hepatically based drug interactions will be particularly problematic. [20,51,55,56] This includes patients with AIDS who also are receiving protease inhibitors or NNRTIs. Other situations also may qualify for this approach, such as in patients with difficult-to-control seizures or serious cardiovascular disease. Such patients are probably receiving drugs that are hepatically cleared. The introduction of rifampicin could seriously compromise the control of these other conditions, as rifampicin is the most potent hepatic enzyme inducer routinely used in clinical practice. Rifabutin is approximately 40% as potent an inducer, so while this cannot be ignored, such adjustments to the doses of concurrent medications are feasible. This approach is not without cost. First, rifabutin is partially cleared by cytochrome P450 (CYP) 3A4, and the active 25-desacety metabolite is significantly cleared by this enzyme. [20] Therefore, rifabutin also is subject to induction or, more commonly, inhibition of clearance caused by coadministered drugs. Because some of the toxicity of rifabutin appears to be dose and concentration dependent, increased serum rifabutin concentrations can result in leucopenia, skin discolouration, arthralgias and anterior uveitis. In addition to clinical and haematological observation, serum rifabutin concentrations can be used to prevent the development of these toxicities. The usual t max of rifabutin is 3 to 4 hours, and a second concentration 6 or 7 hours post-dose can help to detect delayed absorption. In general, the peak concentration should fall between 0.3 to 0.9 µg/ml. When the peak is less than 0.2 µg/ml, consideration can be given to increasing the dose. Although the precise concentration associated with toxicity probably varies from patient to patient, a dose decrease may be indicated when peak concentrations exceed 1.0 µg/ml. The decision to change the dose should take into account the clinical status of the patient, their need for drug and any evidence of adverse reactions, in addition to the serum concentrations. TDM for the concurrent medications will facilitate their dosage adjustments. 4.4 Rifapentine Rifapentine is an interesting rifamycin. [20,78,79] It has serum concentrations similar to or slightly higher than rifampicin, and an active 25-desacetyl metabolite like rifampicin and rifabutin. The halflife of rifapentine is 15 hours, which is intermediate between the very short half-life of rifampicin and the very long half-life of rifabutin. Rifapentine typically reaches maximum serum concentrations around 5 hours post-dose. The hope is that rifapentine can be used in the continuation phase of TB treatment, reducing the number of patient contacts from two to one per week. It appears that three problems need to be solved in order to make this a reality. [80] First, rifapentine is very highly protein bound, which helps to explain its long half-life. Unfortunately, this also provides less free drug to interact with mycobacteria. USPHS TB Study 25 assessed higher rifapentine doses (900 and 1200mg) and these larger doses were safely used. [81] Since rifapentine has the same mechanism of action as rifampicin, these higher doses should produce better dose- and concentration-related bactericidal effects. The second problem to be addressed is that isoniazid has a very short half-life, and once weekly isoniazid plus rifapentine produces long periods of time when only rifapentine is present in the blood and, presumably, at the site of infection. This is precariously close to giving monotherapy, which is not acceptable in the treatment of active TB disease. Among the standard TB drugs, ethambutol and pyrazinamide both have longer terminal elim-

10 2178 Peloquin ination half-lives and might be better choices than isoniazid based on this criterion. Among the second-line agents, cycloserine has the longest halflife and might be useful except for its relatively high frequency of CNS ADR. [1,6] Therefore, cycloserine is not likely to emerge as a new first-line TB drug. Finally, the new fluoroquinolones may provide an answer. Moxifloxacin and gatifloxacin, as well as the older levofloxacin, all have considerable anti-tb activity (perhaps comparable to ethambutol) and have half-lives that are at least two times longer than that of isoniazid. [33,82-84] Clinical trials are being designed to investigate these agents as companion drugs for rifapentine in the continuation phase of treatment. Other studies aim to determine their sterilising effect in combination with first-line agents, isoniazid, rifampicin, ethambutol and pyrazinamide. Ideally, their anti- TB activity will allow for a shortened duration of treatment. Further research should confirm their roles in the treatment of isoniazid-resistant and/or multi-drug resistant (MDR)-TB. [85-87] Finally, as described, rifapentine was least effective in patients with extensive disease and high numbers of residual organisms at 2 months. Therefore, intensification of the initial phase of treatment is required in order to allow for once-weekly continuation phase treatment. As far as drug interactions are concerned, rifapentine is much closer to rifampicin than rifabutin as an inducer of hepatic microsomal enzymes (about 85% as potent as rifampicin). [20] Therefore, it is not an ideal choice for patients at risk of serious drug interactions. We generally target a C max of 8 to 30 µg/ml after a 600mg dose of rifapentine, similar to rifampicin. We would consider recommending a dose increase if the peak is less than 6 µg/ml. 4.5 Pyrazinamide Pyrazinamide is perhaps the most reliably absorbed of the TB drugs. [69,88] Pyrazinamide usually reaches C max 1 to 2 hours post-dose and, with its long half-life, is present in the serum for many hours. Therefore, the 2 and 6 hour sampling strategy for isoniazid and rifampicin also works well for pyrazinamide. Pyrazinamide and its biological effects can be used as markers of compliance with therapy. Patients who are taking their TB medications, including pyrazinamide, will have elevated uric acid serum concentrations and will have pyrazinamide in the serum throughout a 24-hour dose administration interval. Concentrations will be easily measurable from 2 to 12 hours post-dose, and usually can be measured at 24 hours. If pyrazinamide is not present, and if the uric acid is not elevated, then it is very unlikely that the patient is taking the medications. While DOT usually suffices to ensure drug ingestion, some patients have been known to place the tablets and capsules under their tongues until they are no longer observed, or even to vomit them up after swallowing them. While it is difficult to imagine why someone would do this, the author and colleagues have witnessed this behaviour. Pyrazinamide and its metabolites inhibit the secretion of uric acid in the kidneys. [89] It also is possible that they inhibit the secretion of other drugs, although this has not been carefully studied. It is possible that the much higher than expected intolerance of pyrazinamide and levofloxacin as an empirical treatment for latent MDR-TB is due to such an interaction. [90] For pyrazinamide, we target a C max of 20 to 40 µg/ml after a 25 mg/kg daily dose and40to60µg/ml after a 50 mg/kg biweekly dose. We generally recommend a dose increase if the peak is less than 75% of the desired range. 4.6 Ethambutol Ethambutol reaches C max 2 to 3 hours post-dose, so the above 2- and 6-hour sampling strategy is usually adequate to detect ethambutol delayed or malabsorption. [91] Should the patient also be taking rifabutin and, in the case of M. avium infection, clarithromycin, a 3- and 6- to 7-hour post-dose sampling strategy may be preferred. [92] The reliance of ethambutol on renal elimination was described in section 3.3, so the use of ethambutol in patients with renal failure is best accompanied by TDM. Also, frequent documentation of visual acu-

11 Therapeutic Drug Monitoring in TB 2179 ity (Snellen chart) and red-green colour discrimination (Ishihara plates) is highly recommended. The patient should be instructed to stop taking ethambutol and to call the primary physician should they notice any change in vision, such as a new found difficulty in reading the newspaper. For ethambutol, we target a C max of 2 to 6 µg/ml after a15to25mg/kgdailydoseand4to12µg/ml after a 50 mg/kg biweekly dose. We generally recommend a dose increase if the peak is less than 2 µg/ml with daily doses or less than 4 µg/ml with biweekly, as ethambutol is a fairly weak anti-tb drug. 4.7 Streptomycin Because streptomycin is given intramuscularly or intravenously, malabsorption generally is not a concern. Standard doses of 15 mg/kg intramuscularly produce C max values of 35 to 45 µg/ml. [33,93] It appears that repeated intramuscular injections may toughen the tissue and alter the absorption of streptomycin over time, and this may blunt the C max. Renal elimination is responsible for the removal of streptomycin from the body. The adequacy of the C max, and potential delays in elimination, can be detected using the 2- and 6-hour sampling strategy. This works equally well for both intramuscular and intravenous administration, because adequate time is allowed for absorption from the injection site or for post-infusion distribution. Sampling too soon after an intravenous infusion of large aminoglycoside doses has been shown to falsely elevate the calculated C max value when a one-compartment model is used. [94] Other aminoglycosides, including kanamycin and amikacin, display similar pharmacokinetics to streptomycin, as well as similar toxicities. [33] Capreomycin is a polypeptide, not an aminoglycoside, but it also resembles the aminoglycosides in the above two characteristics. Therefore, these observations are applicable to these agents also. For streptomycin and the other injectable anti-tb drugs, we target a C max of 35 to 45 µg/ml after a 15 mg/kg daily dose and 65 to 80 µg/ml after a 25 mg/kg biweekly dose. We generally recommend a dose increase if the peak is less than 75% of the desired range. Again, at least 1 hour should pass from the end of an intravenous infusion until collection of the peak concentration sample, in order to avoid falsely elevated concentrations during the distribution phase. 4.8 P-Aminosalicylic Acid PAS was a mainstay in the treatment of TB into the 1960s. [10] Along with isoniazid and streptomycin, it was a first-line agent for TB. However, it was plagued by poor GI tolerance. Ethambutol was later shown to be approximately equivalent to PAS in potency, and generally better tolerated than PAS when ethambutol was used at dosages of 25 mg/kg/day or less. Therefore, PAS was replaced by ethambutol as a primary TB drug. However, because of the relative lack of use of PAS over the past 3 decades, most isolates of TB remain susceptible to it. Thus, PAS has experienced a renaissance in the management of patients with MDR- TB. Furthermore, the granule form of PAS is much better tolerated than the plain tablets of PAS. [95-97] Currently, the granule form of PAS is the only form available in several countries, including the US. PAS granules are enteric coated and sustained release. Therefore, samples for C max should be collected approximately 6 hours post-dose. Because only PAS, and not acetyl-pas, has antimycobacterial activity, we try to administer PAS in a way that keeps inhibitory concentrations in the blood for most of the dose administration interval. [96] On the basis of our review of the literature, there does not appear to be a good reason to avoid PAS in patients with renal failure who require the drug. [68] For PAS, we target a C max of 20 to 60 µg/ml after a 4g dose and this usually occurs 4to6hourspost-dosewhenthegranulesofPAS are used. With regular tablets of PAS, the peak occurs by 2 hours in most patients. We generally recommend a dose increase if the peak is less than 10 µg/ml.

12 2180 Peloquin 4.9 Cycloserine Cycloserine remains a second-line TB drug because of its frequent CNS effects. [1,16,33,98] Most commonly, patients complain of an inability to concentrate or of lethargy. These complaints appear even with serum concentrations at the low end of the normal range (20 to 35 µg/ml). More serious CNS toxicity may be associated with elevated serum concentrations, although literature detailing these effects are hard to find. In the experience of the author and colleagues, seizures induced by cycloserine are very rare and the author has never witnessed one. Food modestly decreases the absorption of cycloserine, so it is best to give this drug on an empty stomach. [99] Antacids and orange juice have little effect on the absorption of cycloserine and the t max is usually approximately 2 hours post-dose. Delayed absorption can be detected with a 6-hour sample, while a 10-hour sample may be preferred for estimating the elimination half-life. For cycloserine, we target a C max of 20 to 35 µg/ml after a 250 to 500mg dose. We generally recommend a dose increase if the peak is less than 15 µg/ml and a dose decrease is the peak is above 40 µg/ml. Given the longer half-life of cycloserine, it may be best to allow 3 to 4 days of administration before collecting blood samples. This will allow for the achievement of steady-state prior to sampling Ethionamide Ethionamide is a very difficult drug to take. [100] It frequently causes GI upset far more annoying than that experienced with macrolide antibiotics. The pattern of absorption can be erratic, possibly due to changes in GI motility associated with the nausea it induces. Although 2 hours is a typical t max, delayed absorption is fairly common. [101] The 2- and 6-hour sampling strategy works reasonably well for this drug. We target a C max of1to5µg/ml after a 250 to 500mg dose. We generally recommend a dose increase if the peak is less than 1 µg/ml, since ethionamide is a weak anti-tb drug. 5. Other Drugs Levofloxacin and other fluoroquinolones, including ciprofloxacin and sparfloxacin, have been used for the treatment of MDR-TB, with newer agents awaiting further study. [87] Because the mechanism of action against TB is the same as that against aerobic Gram-negative bacilli, it is reasonable to expect a dose response with the fluoroquinolones against TB. This has lead us to the practice of giving fairly large doses of levofloxacin (750 to 1000mg) every day. The associated C max values at about 2 hours post-dose are 8 to 12 µg/ml. This should produce a C max : MIC of approximately 10, which approaches the target ratio for Gram-negative bacilli. Clinical trials are needed to validate and improve upon this approach. Levofloxacin relies on renal elimination, so patients with renal dysfunction should receive the drug every day or less often. [84] A second sample 6 to 10 hours post-dose allows for estimation of the elimination half-life. As a class, fluoroquinolones can cause CNS excitation, including caffeine-like effects and insomnia, and this sometimes is evident in our patients. Tendonitis is seen in some patients and, rarely, Achilles tendon rupture has occurred. However, compared to the frequency of ADR seen with other second-line agents, such as cycloserine and ethionamide, the fluoroquinolones are well tolerated. Clofazimine is a weak anti-tb agent and it is used sparingly in the treatment of MDR-TB, when no other options exist. A 2-hour post-dose sample can verify that the drug is being absorbed, with most patients showing 0.5 to 4.0 µg/ml in the serum. Given its very long tissue half-life, second samples tend to say more about distribution than elimination of clofazimine. [33,102] Over time, dry skin and skin discolouration also become clinical clues as to the absorption of the drug. However, because it takes weeks for this colour to fade after drug discontinuation, these clinical signs are not very useful measures of regular compliance with the regimen.

13 Therapeutic Drug Monitoring in TB Conclusions TDM is never a substitute for sound clinical judgment nor is it a substitute for directly observed treatment, when indicated. [30] However, TDM is a useful tool in a variety of clinical situations. It allows one to determine the extent to which the failure to achieve the desired clinical and bacteriological outcomes can be attributed to the inadequate dose administration of the drugs. [30] It also can resolve issues of drug-drug interactions before the patient experiences failure, relapse or toxicity. In these ways, TDM is a very powerful ally in the management of complex clinical cases. [30] Acknowledgements The author has provided no information on sources of funding or on conflicts of interest directly relevant to the content of this review/study. References 1. Peloquin CA, Ebert SC. Tuberculosis. In: DiPiro JT, Talbert RL, Yee GC, et al., editors. Pharmacotherapy: A pathophysiologic Approach. 4th ed. Stamford (CT): Appleton & Lange, 1999: American Thoracic Society. Targeted tuberculin skin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med 2000; 161: S Verbist L. Mode of action of antituberculous drugs: I. Medikon 1974; 3: Verbist L. Mode of action of antituberculous drugs: II. Medikon 1979; 3: Winder FG. Mode of action of the antimycobacterial agents and associated aspects of the molecular biology of the mycobacteria. In: Rat-ledge C, Stanford J, editors. The Biology of Mycobacteria: Vol 1. Physiology, Identification, and Classification. London: Academic Press, 1982: Peloquin CA. Using therapeutic drug monitoring to dose the antimycobacterial drugs. Clin Chest Med 1997; 18: Heym B, Saint-Joanis B, Cole ST. The molecular basis of isoniazid resistance in Mycobacterium tuberculosis. Tuber Lung Dis 1999; 79: Somoskovi A, Parsons LM, Salfinger M. The molecular basis of resistance to isoniazid, rifampin, and pyrazinamide in Mycobacterium tuberculosis. Respir Res 2001; 2: Mitchison DA. Basic mechanisms of chemotherapy. Chest 1979; 76: Fox W, Ellard GA, Mitchison DA. Studies on the treatment of tuberculosis undertaken by the British Medical Research Council Tuberculosis Units, , with relevant subsequent publications. Int J Tuberc Lung Dis 1999; 3: S Mitchison DA. Role of individual drugs in the chemotherapy of tuberculosis. Int J Tuberc Lung Dis 2000; 4: Hernández-Pando R, Jeyanthan M, Mengistu G, et al. Persistence of DNA from Mycobacterium tuberculosis in superficially normal lung tissue during latent infection. Lancet 2000; 356: Bishai WR. Rekindling old controversy on elusive lair of latent tuberculosis. Lancet 2000; 356: Flynn JL, Chan J. Tuberculosis: latency and reactivation. Infect Immun 2001; 69: Orme I. The latent tuberculosis bacillus (I ll let you know if I ever meet one). Int J Tuberc Lung Dis 2001; 5: Iseman MD. A clinician s guide to tuberculosis. Philadelphia: Lippencott Williams & Wilkins, 2000: American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children. Am J Respir Crit Care Med 1994; 149: Kreis B, Pretet S. Two three-month treatment regimens for pulmonary tuberculosis. Bull Int Union Tuberc 1976; 51: Girling DJ. Adverse effects of antituberculous drugs. Drugs 1982; 23: Burman WJ, Gallicano K, Peloquin CA. Comparative pharmacokinetics and pharmacodynamics of the rifamycin antibiotics. Clin Pharmacokinet 2001; 40: Burman WJ, Dalton CB, Cohn DL, et al. A cost-effectiveness analysis of directly observed therapy vs. self-administered therapy for treatment of tuberculosis. Chest 1997; 112: British Thoracic Association. A controlled trial of six months chemotherapy in pulmonary tuberculosis. Br J Dis Chest 1981; 75: Vernon A, for the TB Trials Consortium. TBTC Study 22 (Rifapentine Trial): Preliminary Results in HIV-negative Patients [abstract]. Am J Respir Crit Care Med 2000; 161: A Kimerling ME, Phillips P, Patterson P, et al. Low serum antimycobacterial drug levels in non-hiv-infected tuberculosis patients. Chest 1998; 113: Yew WW. Therapeutic drug monitoring in antituberculosis chemotherapy. Ther Drug Monit 1998; 20 (5): Mehta JR, Shantaveerapa H, Byrd RP, et al. Utility of rifampin blood levels in the treatment and follow-up of active pulmonary tuberculosis patients who were slow to respond to routine directly observed therapy. Chest 2001; 120: Weiner M, Khan A, Benator D, et al. Low isoniazid levels are associated with tuberculosis treatment failure or relapse with once-weekly rifapentine and isoniazid [abstract]. Proceedings of the 97th American Lung Association / American Thoracic Society International Conference; 2001 May 18-23; San Francisco (CA). Am J Respir Crit Care Med 2001; 163: A Peloquin CA, Benator D, Hayden K, et al. Low rifapentine, rifampin, & isoniazid plasma levels are not predicted by clinical and demographic features [abstract]. Proceedings of the 97th American Lung Association / American Thoracic Society International Conference; 2001 May 18-23; San Francisco (CA). Am J Respir Crit Care Med 2001; 163: A Peloquin CA. Tuberculosis Drug Serum Levels (letter). Clin Infect Dis 2001; 33: Peloquin CA. Pharmacological issues in the treatment of tuberculosis. Ann N Y Acad Sci 2001; 953: Jelliffe R. Goal-oriented, model-based drug regimens: setting individualized goals for each patient. Ther Drug Monit 2000; 22: Holdiness MR. Clinical pharmacokinetics of the antituberculosis drugs. Clin Pharmacokinet 1984; 9: Peloquin CA. Antituberculosis drugs: pharmacokinetics. In: Heifets L, editor. Drug susceptibility in the chemotherapy of mycobacterial infections. Boca Raton: CRC Press; 1991: Steil CF. Diabetes mellitus. In: DiPiro JT, Talbert RL, Yee GC, et al., editors. Pharmacotherapy: a pathophysiologic approach. 4th ed. Stamford (CT): Appleton & Lange, 1999:

14 2182 Peloquin 35. Bashar M, Alcabes P, Rom WN, et al. Increased incidence of multidrug-resistant tuberculosis in diabetic patients on the Bellevue Chest Service, Chest 2001; 120: Kotler DP, Gaetz HP, Lange M, et al. Enteropathy associated with the acquired immunodeficiency syndrome. Ann Intern Med 1984; 101: Gillin JS, Shike M, Alcock N, et al. Malabsorption and mucosal abnormalities of the small intestine in the acquired immunodeficiency syndrome. Ann Intern Med 1985; 102: Bartlett JG, Belitsos PC, Sears CL. AIDS enteropathy. Clin Infect Dis 1992; 15: Berning SE, Huitt GA, Iseman MD, et al. Malabsorption of Anti-tuberculosis medications by a patient with AIDS [letter]. N Engl J Med 1992; 327: Peloquin CA, MacPhee AA, Berning SE. Malabsorption of antimycobacterial medications. N Engl J Med 1993; 329: Gordon SM, Horsburgh CR, Peloquin CA, et al. Low serum levels of oral antimycobacterial agents in patients with disseminated Mycobacterium avium complex disease. J Infect Dis 1993; 168: Colborn D, Lewis R, Narang P. HIV disease does not influence rifabutin absorption [abstract no. A42]. Proceedings of the 34th Interscience Conference on Antimicrobial Agents and Chemotherapy; 1994 Oct 4-7; Orlando (FL) 43. Patel KB, Belmonte R, Crowe HM. Drug malabsorption and resistant tuberculosis in HIV-infected patients [letter]. N Engl J Med 1995; 332: Peloquin CA, Nitta AT, Burman WJ, et al. Incidence of low anti-tuberculosis drug concentrations in patients with AIDS. Ann Pharmacother 1996; 30: Choudhri SH, Hawken M, Gathua S, et al. Pharmacokinetics of antimycobacterial drugs in patients with tuberculosis, AIDS, and diarrhea. Clin Infect Dis 1997; 25: Sahai J, Gallicano K, Swick L, et al. Reduced plasma concentrations of antituberculous drugs in patients with HIV infection. Ann Intern Med. 1997; 127: Peloquin CA. Serum Concentrations of the Antimycobacterial Drugs. Chest 1998; 113: Taylor B, Smith PJ. Does AIDS impair the absorption of the antituberculosis agents? Int J Tuberc Lung Dis 1998; 2: Peloquin CA, Berning SE, Huitt GA, et al. AIDS and TB drug absorption [letter]. Int J Tuberc Lung Dis 1999; 3: Jaruratanasirikul S. The pharmacokinetics of oral rifampicin in AIDS patients. J Med Assoc Thai 1998; 81: Burman WJ, Gallicano K, Peloquin CA. Therapeutic implications of drug interactions in the treatment of HIV-related tuberculosis. Clin Infect Dis 1999; 28: Perlman DC, Remmel R, Brundage R, et al., for the ACTG 309 Protocol Team. Pharmacokinetics of antituberculous agents in persons with HIV-related tuberculosis [abstract no. 619]. Proceedings of the Infectious Diseases Society 37th Annual Meeting; 1999 Nov 18-21; Philadelphia (PA). Clin Infect Dis 1999; 29: Narita M, Hisada M, Thimmappa B, et al. Tuberculosis recurrence: multivariate analysis of serum levels of tuberculosis drugs, human immunodeficiency virus status, and other risk factors. Clin Infect Dis 2001; 32: Narita M, Stambaugh JJ, Hollender ES, et al. Use of rifabutin with protease inhibitors for HIV-infected tuberculosis patients. Clin Infect Dis 2000; 30: CDC. Prevention and treatment of tuberculosis among patients infected with human Immunodeficiency Virus: principles of therapy and revised recommendations. Morb Mortal Wkly Rep 1998; 47 (RR-20): CDC. Updated guidelines for the use of rifabutin or rifampin for the treatment and prevention of tuberculosis among HIVinfected patients taking protease inhibitors on nonnucleoside reverse transcriptase inhibitors. Morb Mortal Wkly Rep 2000; 49: Finch CK, Chrisman CR, Baciewicz AM, et al. Rifampin and rifabutin drug interactions: an update. Arch Intern Med 2002; 162 (9): Durant J, Clevenbergh P, Garraffo R, et al. Importance of protease inhibitor plasma levels in HIV-infected patients treated with genotypic-guided therapy: pharmacological data from the Viradapt Study. AIDS 2000; 14: Piscitelli SC. The value of drug levels: the plot thickens. Medscape.com 2001 [online]. Available from URL: http: //id.medscape.com/medscape/cno/2001/retro/story.cfm?.story_id=2055 [Accessed 2002 Sep 3] 60. Acosta EP, Kakuda TN, Brundage RC, et al. Pharmacodynamics of Human Immunodeficiency Virus Type 1 Protease Inhibitors. Clin Infect Dis 2000; 30 Suppl. 2: S Angel JB, Khaliq Y, Monpetit ML, et al. An argument for routine therapeutic drug monitoring of HIV-1 protease inhibitors during pregnancy. AIDS 2001; 15: Back D, Gatti G, Fletcher C, et al. Therapeutic drug monitoring in HIV infection: current status and future directions. AIDS 2002; 16 Suppl. 1: S5-S Cuss FMC, Carmichael DJS, Linington A, et al. Tuberculosis in renal failure: a high incidence in patients born in the third world. Clin Nephrology 1986; 25: Tokars JI, Miller ER, Alter MJ, et al. National surveillance of dialysis associated diseases in the United States, AS- AIO J 1998; 44: Cengiz K. Increased incidence of tuberculosis in patients undergoing hemodialysis. Nephron 1996; 73: Chia S, Karim M, Elwood RK, et al. Risk of tuberculosis in dialysis patients : a population-based study. Int J Tuberc Lung Dis 1998; 2 (12): Malone RS, Fish DN, Spiegel DM, et al. The effect of hemodialysis on isoniazid, rifampin, pyrazinamide, and ethambutol. Am J Respir Crit Care Med 1999; 159: Malone RS, Fish DN, Spiegel DM, et al. The effect of hemodialysis on cycloserine, ethionamide, para-aminosalicylate, and clofazimine. Chest 1999; 116: Peloquin CA, Jaresko GS, Yong CL, et al. Population Pharmacokinetic Modeling of Isoniazid, Rifampin, and Pyrazinamide. Antimicrobial Agents Chemother 1997; 41: Peloquin CA, Namdar R, Dodge AA, et al. Pharmacokinetics of Isoniazid Under Fasting Conditions, with Food, and with Antacids. Int J Tuberc Lung Dis 1999; 3: Weiner M, Khan A, Benator D, et al., and the TB Trials Consortium. Low Isoniazid (INH) Levels Are Associated with TB Treatment Failure/Relapse with Once-Weekly Rifapentine (RPT) and INH [abstract]. Am J Respir Crit Care Med 2001; 163: A Acocella G. Pharmacokinetics and metabolism of rifampin in humans. Rev Infect Dis 1983; 5: S Cavenaghi R. Rifampicin raw material characteristics and their effect on bioavailability. Bull Int Union Tuberc Lung Dis 1989; 64: Peloquin CA, Namdar R, Singleton MD, et al. Pharmacokinetics of Rifampin Under Fasting Conditions, with Food, and with Antacids. Chest 1999; 115: McEvoy GK, editor. AHFS Drug Information. Bethesda (MD): American Soc Health-Systems Pharmacists, 2002:

15 Therapeutic Drug Monitoring in TB Ellard GA, Fourie PB. Rifampicin bioavailability: a review of its pharmacology and the chemotherapeutic necessity for ensuring optimal absorption. Int J Tuberc Lung Dis 1999; 3: S Kucers A, Bennett N. The use of antibiotics. 4th ed. Philadelphia: J.B. Lippencott Company, 1987: Nahata MC, Temple ME. Rifapentine: Its role in the treatment of tuberculosis. Ann Pharmacother 1999; 33: Tam CM, Chan SL, Kam KM, et al. Rifapentine and isoniazid in the continuation phase of a 6-month regimen. Interim report: no activity of isoniazid in the continuation phase. Int J Tuberc Lung Dis 2000; 4: Mitchison DA. Development of rifapentine: the way ahead. Int J Tuberc Lung Dis 1998; 2: Bock N, Sterling T, Pachucki C, et al. for the TB Trials Consortium. Tolerability of once-weekly rifapentine 900 mg plus INH vs once-weekly rifapentine 600 mg plus INH during continuation phase treatment of pan-susceptible tuberculosis in HIV-negative adults [abstract + poster]. Am J Respir Crit Care Med 2001; 163: A Nightingale CH. Moxifloxacin, a new antibiotic designed to treat community-acquired respiratory tract infections: A review of microbiologic and pharmacokinetic-pharmacodynamic characteristics. Pharmacotherapy 2000; 20: Perry CM, Barman Balfour JA, Lamb HM. Gatifloxacin. Drugs 1999; 58: Fish DN, Chow AT. The clinical pharmacokinetics of levofloxacin. Clin Pharmacokinet 1997; 32: Berning SE, Madsen L, Iseman MD, et al. Long-term safety of ofloxacin and ciprofloxacin in the treatment of Mycobacterial infections. Am J Respir Crit Care Med 1995; 151: Peloquin CA, Berning SE, Madsen L, et al. Ofloxacin and Ciprofloxacin in the Treatment of Mycobacterial Infections: Development of Resistance and Drug Interactions. J Infect Dis Pharmacother 1995; 1: Berning SE. The role of fluoroquinolones in tuberculosis today. Drugs 2001; 61: Peloquin CA, Bulpitt AE, Jaresko GS, et al. Pharmacokinetics of Pyrazinamide Under Fasting Conditions, with Food, and with Antacids. Pharmacotherapy 1998; 18: Weiner IM, Tinker JP. Pharmacology of pyrazinamide: Metabolic and renal function studies related to the mechanism of drug-induced urate retention. J Pharmacol Exp Ther 1972; 180: Horn DL, Hewlett Jr D, Alfalla C, et al. Limited tolerance of ofloxacin and pyrazinamide prophylaxis against tuberculosis [letter]. N Engl J Med 1994; 330: Peloquin CA, Bulpitt AE, Jaresko GS, et al. Pharmacokinetics of ethambutol under fasting conditions, with food, and with antacids. Antimicrobial Agents Chemother 1999, 43; Peloquin CA. Mycobacterium avium complex infection: pharmacokinetic and pharmacodynamic considerations that may improve clinical outcomes. Clin Pharmacokinet 1997; 32: Zhu M, Burman WJ, Jaresko GS, et al. Population pharmacokinetics of intravenous and intramuscular streptomycin in patients with tuberculosis. Pharmacotherapy 2001; 21: Demczar DJ, Nafziger AN, Bertino JS Jr. Pharmacokinetics of gentamicin at traditional versus high doses: implications for once-daily aminoglycoside dosing. Antimicrobial Agents Chemother 1997; 41: Peloquin CA, Henshaw TL, Huitt GA, et al. Pharmacokinetic evaluation of p-aminosalicylic acid granules [published erratum appears in Pharmacotherapy 1994; 14: (4) P-2]. Pharmacotherapy 1994; 14: Peloquin CA, Berning SE, Huitt GA, et al. Once-daily and twice-daily dosing of p-aminosalicylic acid (PAS) granules. Am J Respir Crit Care Med 1999; 159: Peloquin CA, Zhu M, Adam RD, et al. Pharmacokinetics of p-aminosalicylate under fasting conditions, with orange juice, food, and antacids. Ann Pharmacother 2001; 35: Berning SE, Peloquin CA. Antimycobacterial agents: Cycloserine. In: Yu VL, Merigan TC, Barriere S, White NJ, editors. Antimicrobial chemotherapy and vaccines. Baltimore (MD): Williams and Wilkins, 1998: Zhu M, Nix DE, Adam RD, et al. Pharmacokinetics of cycloserine under fasting conditions, with orange juice, food, and antacids. Pharmacotherapy 2001; 21: Berning SE, Peloquin CA. Antimycobacterial agents: Ethionamide. In: Yu VL, Merigan TC, Barriere S, White NJ, editors. Antimicrobial chemotherapy and vaccines. Baltimore (MD): Williams and Wilkins, 1998: Auclair B, Nix DE, Adam RD, et al. Pharmacokinetics of ethionamide under fasting conditions, with orange juice, food, and antacids. Antimicrobial Agents Chemother 2001; 45: Garrelts JC. Clofazimine: a review of its use in leprosy and Mycobacterium avium complex infection. DICP Ann Pharmacother 1991; 25: Correspondence and offprints: Dr Charles A. Peloquin, Pharmacokinetics Laboratory, Room D-106, National Jewish Medical and Research Center, 1400 Jackson Street, Denver, CO 80206, USA. peloquinc@njc.org