February 26, 2002
David Back hails from Liverpool, as, he reminded us, did the Beatles. He suggested that his review might seem like a "magical mystery tour" into the sometimes-arcane world of HIV pharmacology. Instead, it was a tour de force. Given that this was the last talk of a long day, beginning 12 hours after the first session of the day, this was fortunate.
In his introduction of the talk, Dr. Charlie Flexner stated that he knew of no area in HIV more controversial than the issue of therapeutic monitoring. That is very true, but Professor Back left us with a message that pharmacologic monitoring is, in fact, likely to be vital, although much work remains to be done.
In general, Back argued, monitoring drug levels in blood with routine measurements, (therapeutic drug monitoring or TDM) is considered when there is a clear relationship between drug level and the drug's efficacy and toxicity. This relationship, he argued, exists for both protease inhibitors and NNRTIs.
Another criteria to take into consideration is that there is a large variation between patients in the drug levels they achieve with a standard, "one-size-fits-all" dose. This is very true for both protease inhibitors and NNRTIs -- even among healthy adults, where drug levels may vary by up to ten fold despite taking the same dose. When one considers the difficult issues of pediatrics, pregnancy, kidney disease and complicated drug regimens, this is also particularly true. Lastly, therapeutic drug monitoring may be helpful when there is a narrow "window" between the blood level of a drug that is effective and the level that is toxic.
Thus, it would seem that therapeutic drug monitoring is perfect for both NNRTIs and protease inhibitors. In fact, we routinely use therapeutic drug monitoring to monitor the use of drugs that do not appear to be as vital or difficult to use, including antibiotics, drugs for seizures, antidepressants and immunosuppressants. The consequences of misusing antiretrovirals are very serious indeed, so this might seem to be a no-brainer. However, things are not so simple, or there might be no need for this talk under the heading "Controversies in Antiretroviral Therapy."
Much of the controversy (which sounds much better with the British pronunciation "conTRUHvusy") revolves around how to define the target levels. Several complex methods have been proposed and debated, including the "inhibitory quotient" or IQ (see "The Inhibitory Quotient (IQ) for Saquinavir (SQV) Predicts Virologic Response to Salvage Therapy"), "virtual inhibitory quotient" or VIQ (see "The Use of Virtual Inhibitory Quotient (VIQ) in Antiretroviral (ART)-Experienced Patients Taking Amprenavir/Lopinavir Combinations"), and now the "normalized inhibitory quotient" or NIQ (see "The Normalized Inhibitory Quotient (NIQ) of Lopinavir Is Predictive of Viral Load Response over 48 Weeks in a Cohort of Highly Experienced HIV-1-Infected Individuals"). These are variations on the idea of dividing the patient's actual blood level by the amount of drug that inhibits their own virus in a phenotype test. The term IQ -- meaning inhibitory quotient -- seems designed to make us think that someone involved in this debate has a high IQ. There are many ways to calculate IQ depending on how one deals with protein binding and whether one uses IC50 or IC90. Sometimes it seems that the debate is more about which method makes one drug look better than which method makes sense and works.
Dr. Back did not shy away from this debate, but he suggested we may be making the issue too complicated. He advocated beginning with a minimalist approach similar to what we do with other drugs. In this approach, we simply need to determine minimum and maximum target levels for a drug in routine use. For resistant virus and drug development, some form of IQ probably will be needed.
Dr. Back then recited the arguments against therapeutic drug monitoring. Several involve adherence: Could a low drug level simply mean the patient has missed doses? How do we deal with protein binding? When do we draw levels and how do we make sure we know when the last pills were taken? Who will interpret the levels and provide expert advice? Will outcome actually improve?
Another major limitation is that there is limited prospective trial data to support therapeutic drug monitoring. One study, Pharmadapt, did not show a benefit, while another, Athena, did. Both studies suffered from major design limitations, which limited the ability to show a benefit.
Other issues include cost, the need for yet more blood tests, and the quality of the test results. So far, the cost of a single blood measurement is far cheaper than a CD4 count or viral load test, but the costs can add up rapidly. No patient wants more needle sticks. Perhaps most importantly, it is important that the labs participate in international quality assurance programs so that we know the results we get back are accurate. In the current U.S. situation, it is hard to know whether the labs that are offering the test commercially are offering truly accurate tests. A study by Dr. David Berger in the Netherlands documented a wide range of accuracy among labs.
In rescue or third-line therapy, we often rely on boosted protease inhibitors such as lopinavir/ritonavir (Kaletra) or ritonavir/saquinavir. Increasingly, we are combining up to three protease inhibitors. Some have argued that the booster effects raise levels so high that monitoring levels is unnecessary. Back showed data making it clear that this may not be true.
With complex and under-studied regimens, drug interactions may be particularly troublesome. One issue, for example, is the use of lopinavir/ritonavir with amprenavir. Dr. Back showed data from six studies of this combination, including three at this meeting ("Pharmacokinetic Interaction between Lopinavir/r and Amprenavir in Salvage Therapy," "Pharmacokinetic Parameters and Virological Response to the Combination of Lopinavir/Ritonavir (LPV/r) and Amprenavir (APV) in HIV-Infected Patients with Multiple Treatment Failures: Week-6 Results of Puzzle 1-ANRS Study," "A Prospective Study of Deep Salvage Therapy with Lopinavir/r, Amprenavir, and NRTIs: Final 24-Week Data, Pharmacokinetics, and Association of Drug Levels/Drug Susceptibility With Virologic Response") that do not give exactly the same answer.
He concluded what he called the "Hard Day's Night" (this symposium took place from 7 to 9 p.m. on Tuesday night; by the way, I am not responsible for any Beatles puns) with suggestions for possible uses for therapeutic drug monitoring while the data are still emerging. These include complex regimens, drug failure, pediatrics, pregnancy, liver or kidney disease, and in the setting of unusual toxicity.
He ended with more Beatles puns: Should we "Let It Be" or can "We Work It Out"? We will need to work "Eight Days a Week," however. Thanking his collaborators, Back reminded us that he, like the rest of us, "Gets By With a Little Help From His Friends."
|Please note: Knowledge about HIV changes rapidly. Note the date of this summary's publication, and before treating patients or employing any therapies described in these materials, verify all information independently. If you are a patient, please consult a doctor or other medical professional before acting on any of the information presented in this summary. For a complete listing of our most recent conference coverage, click here.|