Protocol for Routine Therapeutic Drug Monitoring?

Angela D.M. Kashuba, BScPhm, PharmD, DABCP

Disclosures

June 12, 2001

Question

Several patients have inquired about therapeutic drug monitoring (TDM) for their antiretroviral therapy. Based on currently available data, what would be a "reasonable" protocol for the use of TDM in routine clinical practice?

Corklin R. Steinhart, MD, PhD

Response from Angela D.M. Kashuba, BScPhm, PharmD, DABCP

One of the fundamental principles of pharmacology is that there exists a quantifiable relationship between drug concentrations in the body and efficacy, toxicity, or both. TDM is used to adjust doses or dosing intervals for drugs that, in the majority of individuals, have a defined concentration range for maximal efficacy and/or minimal toxicity. This technique is used routinely for aminoglycoside antibiotics, anticonvulsants, and certain antiarrhythmics.

Currently, clinical data are insufficient to warrant the use of antiretroviral TDM in the routine clinical management of HIV-infected individuals. However, there are some specific situations in which TDM may be of value.

Response from Angela D.M. Kashuba, BScPhm, PharmD, DABCP

It has been estimated that approximately 35% of patients may have suboptimal protease inhibitor (PI) concentrations, and of those, 50% will experience virologic failure.[1] With the knowledge that a standard dose of medication may not be appropriate for all patients, the role of measuring specific antiretroviral exposure to promote durable suppression of viral load and prevent the development of viral resistance has become an area of intense investigation. Dose-response and concentration-response relationships have been identified primarily for PIs and nonnucleoside reverse transcriptase inhibitors (NNRTIs), although some data exist for the nucleoside analogues. For most agents investigated, there appears to be a positive relationship between some pharmacokinetic measure of exposure (eg, Cmin, AUC) and viral load suppression. Concentration-response relationship data have been generated for zidovudine, lamivudine, indinavir, saquinavir, ritonavir, nelfinavir, amprenavir, and efavirenz.[2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19] However, many of these studies have enrolled small numbers of antiretroviral-naive patients with diverse disease stages and background antiretroviral medications. Therefore, concentration-response relationships defined in one particular investigation may not be optimally extrapolated to an individual patient.

In addition, large intraindividual variability in concentration due to food effects or medication adherence may make concentrations a poor predictor of response. The assumption that short-term behavior (as reflected in a single antiretroviral concentration) will predict long-term behavior (ie, months of antiretroviral exposure) has not yet been verified. Finally, validated assays with rigid quality-control standards for all PIs and NNRTIs are not widely available for clinical use.

Response from Angela D.M. Kashuba, BScPhm, PharmD, DABCP

Preliminary information from 2 large, prospective TDM studies (PharmAdapt and Athena) were presented recently in abstract form with conflicting results. The PharmAdapt study examined the effect of PI concentration monitoring on viral outcome in patients failing their current antiretroviral therapy.[20] These investigators found no significant difference in 12-week plasma HIV-1 RNA response in 180 patients randomized to treatment modification based on either viral genotyping or viral genotyping plus adjusted PI doses based on Cmin at week 4. This study had a number of limitations that proscribe a definitive conclusion, including suboptimal target PI concentrations, delayed dosage adjustments at week 8, and a suboptimal 12-week endpoint.

The Athena study is examining the effect of PI concentration monitoring on viral outcome in 600 treatment-naive patients; 6- and 12-month data for indinavir (55 patients randomized to TDM vs no TDM) and nelfinavir (92 patients randomized to TDM vs no TDM) have been presented. Using intent-to-treat analyses, these investigators demonstrated that patients in both the indinavir[21] and nelfinavir[22] TDM groups had improved treatment outcomes at 6 and 12 months, with significantly higher proportions of TDM patients having plasma HIV-1 RNA levels less than 500 copies/mL at 12 months. In the indinavir TDM group, this was attributed to fewer treatment discontinuations due to toxicity, and in the nelfinavir TDM group, this was attributed to fewer treatment discontinuations due to virologic failure.

Recently, the issue of TDM has become more complex with the integration of viral susceptibility and pharmacokinetic exposure measures into a new pharmacodynamic index (Cmin/IC50) called the inhibitory quotient (IQ).[23,24,25] This type of index has been successful in predicting efficacy of other anti-infectives such as quinolones (AUC/MIC) and aminoglycosides (Cmax/MIC), and most likely will be the best parameter to use for individualized antiretroviral dosing in patients with predominant non-wild-type viral isolates. However, the influence of active metabolites, synergy or antagonism with concomitant antiretroviral therapy, the need to obtain a patient-specific viral phenotype, and the lack of specific target values prevent the routine use of this index in the clinical setting.

A large number of unanswered questions remain regarding the use of antiretroviral TDM, including:

  • what is the "therapeutic range" for antiretrovirals?

  • does the "therapeutic range" change with changing background antiretroviral therapy?

  • is the "therapeutic range" defined for single-PI therapy still valid for dual- or triple-PI therapy?

  • does the "therapeutic range" change with different viral populations?; and

  • what is the true cost-effectiveness of TDM?

A number of planned and ongoing investigations have been designed to answer these questions, with data becoming available over the next 1-2 years.

Response from Angela D.M. Kashuba, BScPhm, PharmD, DABCP

Although the lack of target values for antiretroviral concentration or IQ and the lack of clinical outcome data prevent the routine use of TDM for managing HIV-infected individuals, there are certain scenarios -- discussed below -- in which TDM could be of benefit. First, however, it should be noted that some of these scenarios involve comparison of a measured concentration with a measure of central tendency (mean or median) for Cmin published in the individual product monographs (using pharmacokinetic data from HIV-infected volunteers). These approaches are most suited to antiretroviral-naive individuals, or those currently receiving their first antiretroviral regimen. Target concentrations for antiretroviral-experienced individuals with some degree of antiretroviral resistance are unknown, and most likely vary between individuals (as illustrated in the PharmAdapt study). In these situations, a viral phenotype may be helpful, although protein-binding issues and the lack of IQ target values make these situations more difficult to manage. All individuals must be highly adherent (> 95% of the doses taken) and at steady-state conditions (ie, after at least 14 days of therapy). In addition, blood samples should be obtained at the end of the dosing interval, as close to Cmin as possible to enable comparison to product monograph concentrations.

With these caveats in mind, specific scenarios in which TDM may be considered are as follows:

  1. Subgroups of patients, such as pregnant women and children, may have altered and highly variable pharmacokinetics, and may be at increased risk for suboptimal antiretroviral exposure. Although target antiretroviral concentrations in these situations have not been determined, adjusting drug doses to achieve Cmin values within 30% of the mean/median product monograph concentrations that have been shown effective in clinical studies of adults or children may increase the likelihood of adequate antiretroviral exposure.

  2. Highly adherent individuals placed on standard doses of antiretroviral agents who have poor initial plasma HIV-1 RNA responses, or who have an initially adequate but transient HIV-1 RNA response, may benefit from TDM. In these situations, Cmin values may assist in determining whether individuals may have genetically determined high hepatic metabolic rates, poor absorption, or as-yet unidentified drug interactions that are preventing adequate drug exposure. Again, adjusting antiretroviral doses to achieve Cmin values within 30% of the mean/median value documented in the product monograph as being effective may increase the likelihood of adequate drug exposure and antiretroviral response.

  3. Individuals who have responded well to an antiretroviral regimen but require new drug therapy -- or wish to begin a nutraceutical regimen/herbal supplement -- that has unknown or theoretical drug interaction potential (eg, anticonvulsants) would benefit from TDM. Establishing a "baseline" level of antiretroviral exposure before any new therapy has been added, and comparing this to the steady-state condition after the new therapy has started, may detect a drug interaction in sufficient time to prevent therapeutic failure or toxicity by adjusting antiretroviral doses accordingly.

  4. Several of the proposed once-daily PI regimens provide Cmin values similar to those obtained with currently approved dosing strategies for these agents, as opposed to the increased concentrations seen with ritonavir-enhanced twice-daily dosage regimens. In these circumstances, individuals who use once-daily regimens may benefit from TDM by ensuring that adequate concentrations of antiretrovirals are achieved throughout the entire dosing interval.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....