• Inhibitory Quotients versus Drug Levels
• More Pharmacology: QD versus BID Kaletra
• Drug Interactions

Therapeutic drug monitoring (TDM) for antiretrovirals remains one of the most interesting and controversial topics in HIV pharmacology. Several presentations at the 9th CROI reflected recent progress in thinking about how to make TDM clinically useful.

Inhibitory Quotients versus Drug Levels

Most TDM studies have examined the association between a patient’s drug concentrations and the likelihood of deriving viral load benefit from that drug. Virologists have long argued that this only works for treatment-naïve patients with drug-sensitive virus. Different rules about drug levels must apply to patients with complex treatment histories and drug-resistant HIV.

Three studies presented in an oral abstract session predicted treatment outcome by using a ratio of patients’ drug concentrations and their phenotypic sensitivity. The inhibitory quotient, or IQ, is a previously described ratio of the patient’s drug trough concentration (C_min) and the drug concentration required to inhibit that patient’s virus by 50% in vitro (IC₅₀; see table, above). Two other IQ’s were presented: A virtual IQ (VIQ) and a normalized IQ (NIQ).

In an analysis of ACTG Protocol 359, Courtney Fletcher and colleagues from the University of Minnesota examined saquinavir (SQV) trough concentrations and measured SQV IC₅₀’s in 32 heavily treatment-experienced patients taking SQV plus ritonavir (RTV) or nelfinavir (NFV) as part of salvage therapy [Fletcher, CV, et al., Abstract 129]. Although SQV levels alone correlated poorly with virologic response, the SQV IQ (the ratio of trough to phenotypic IC₅₀) more strongly predicted early treatment response. The mean IQ was 201 in those with viral loads below baseline at week 4, but only 17 in those with viral loads at or above baseline.

IQ correlated with viral load response out to week 12, but by week 16 the correlation was no longer significant. This may be because patients were heavily pre-treated and destined to fail anyway, especially given the unconventional regimen, which included delavirdine (DLV) and adefovir. It should also be noted that the investigators were looking at only one drug in a 4-drug regimen.

In a second study with 24 heavily pretreated patients, investigators from Vancouver and Toronto measured the association between virologic outcome and the “virtual” IQ (VIQ) in patients treated with a salvage regimen containing lopinavir/ritonavir (LPV/RTV) plus amprenavir (APV) [Phillips, Abstract 130]. They measured LPV and APV troughs, and then measured IC₅₀ using the “virtual” phenotype, which estimates IC₅₀ from genotype. After at least three months of treatment, the median VIQ for APV was 5.2 in patients with a viral load of <50 c/mL, versus 0.9 in those with detectable viral loads. Similarly for LPV, median VIQ was 12.3 for those with undetectable viral loads versus 1.8 for those with detectable viral loads. Twelve of twelve patients who had a greater than one log drop in viral load on LPV had a VIQ >5.0, suggesting that this might be a reasonable target for TDM.

A third study conducted in Italy looked at the LPV IQ in 52 patients failing a previous PI-containing regimen, using a parameter called the normalized IQ (NIQ) [Castagna, Abstract 128]. In this study, the NIQ was calculated by comparing the patient’s VIQ (using the Virco VirtualPhenotype assay) to a “population” VIQ, which is the mean LPV trough (C_min) from previous large studies in patients given the same dose of LPV/RTV, divided by the mean IC₅₀ in a population of viruses deemed sensitive to LPV (based on the VirtualPhenotype assay). The advantage of this assay is that it compares the patient’s IQ to drug-sensitive patients with “normal” trough concentrations. The disadvantage is that the NIQ involves somewhat arbitrary decisions about what is a “normal” comparative patient population, and what constitutes sensitive or resistant virus. In this study, NIQ was more predictive of viral load outcome at 48 weeks than resistance testing alone, and was in fact more predictive of viral load outcome than any other parameter except baseline viral load (p<0.001, mixed effects model). When patients were divided up into quartiles, those with an NIQ of <0.6 had a median viral load drop of 0.7 logs at week 48, compared to a 2.8 log drop in those with an NIQ of >14.5. Interestingly, in this study, LPV trough concentrations were not associated with treatment outcome at all.

Taken together, these studies suggest that TDM strategies in treatment-experienced patients will need to factor in resistance as well as drug concentrations. This is probably not surprising, but these studies provide some initial guidelines for what IQs (or VIQs or NIQs) should look like.

All three of these studies were small, and one of the studies used a different method to correct for protein binding of the involved drug. However, as was pointed out, if the protein binding correction is consistent within an assay, it becomes a constant and is therefore only relevant if comparing one company’s assay with another’s. Finally, none of these studies was a prospective TDM trial; they all examined the correlation of baseline IQ to outcome without adjusting anyone’s drug dose (or changing regimen) to try and achieve a higher IQ. Two of the three studies found that drug concentrations alone–the basis for traditional TDM–did not predict treatment outcome, suggesting that TDM might be a waste of time in treatment experienced patients.

Along those lines, there was more bad news for TDM at the 9th CROI with release of the results of the GENOPHAR Study [Bossi, Abstract 585-T]. This prospective French trial randomized 134 patients failing therapy to a genotypically guided salvage regimen with or without the addition of TDM information to guide dosing. At the end of 24 weeks, 52% of genotype-only patients and 60% of genotype+TDM patients had a viral load of <400 c/mL, a difference that was not statistically significant.

Unfortunately, GENOPHAR had many of the same design flaws as PharmADAPT, another prospective TDM trial with negative results that was presented at the 8th CROI. Neither study modified regimens based on pharmacokinetic results until week 8, and both were conducted in a heavily pretreated population with poor prognosis and possibly poor adherence from the outset.

The addition of IQ testing to TDM strategies may make us smarter in the future about how we manage our patients, especially in salvage. But whether we are smart enough to make TDM work to improve outcomes in the clinic will require more examination.

[Return to top]

More Pharmacology: QD versus BID Kaletra

LPV/RTV (Kaletra) has become a popular PI for both initial and salvage therapy, owing in large part to its potent inhibition of most HIV isolates and its high IQ (see above). LPV/RTV is currently dosed twice-daily, although the long apparent half-life of LPV in the presence of RTV suggests its possible use as a once-a-day drug. The first pharmacokinetic and efficacy data on qd LPV/RTV compared to the twice-a-day drug were presented at the 9th CROI and looked promising.

Rick Bertz and colleagues from Abbott Laboratories presented data showing that six capsules of Kaletra (800/200 mg LPV/RTV) given once-daily produced a 24-hour area under the concentration-time curve (AUC₂₄) similar to that produced by three capsules (400/100 mg) given every 12 hours [Bertz, Abstract 126]. However, the qd regimen produced a trough (C_min) at the end of 24 hours that was about half that seen with the bid regimen.

IQs were still high (40 on average with the qd regimen and 84 with the bid regimen), but there was more variability in troughs with the qd than with the bid regimen. Two patients in the qd arm had IQs of <10, while the lowest IQ with the bid regimen was 36.

Despite these differences, treatment outcome after 48 weeks in 38 treatment-naïve patients was similar: 79% of bid patients had a viral load <50 c/mL versus 74% treated with qd [Abstract 409-W]. Larger and longer comparisons of qd versus bid LPV/RTV are still needed, especially in treatment-experienced patients, for whom these results might
be irrelevant.

[Return to top]

Drug Interactions

• Amprenavir, Ritonavir, and Efavirenz: These three drugs are sometimes given together in salvage regimens. Investigators from GlaxoSmithKline, Johns Hopkins, Buffalo, and Albany examined the interaction between the APV prodrug GW433908 at a dose of 700 mg bid with efavirenz (600 mg qd) and RTV. The addition of efavirenz did not reduce APV concentrations if subjects were taking either 100 or 200 mg RTV bid. However, this regimen was associated with increases in both cholesterol and fasting triglycerides at weeks 2 and 4 in these 32 healthy seronegative volunteers [Wire, Abstract 431-W].
  
• 
  
• Amprenavir and Delavirdine: Recent cross-sectional studies have found that APV can reduce LPV concentrations by 25% or more, suggesting that APV might serve as a cytochrome P450 3A4 inducer for some drugs. A study from Denmark [Justesen, Abstract 442-W] showed that APV 600 mg bid reduced the DLV (600 mg bid) steady-state AUC by 60%, and reduced the DLV trough by almost nine-fold. This suggests that APV can significantly induce DLV metabolism, and the investigators recommend that APV and DLV not be given together at these doses.
  
• 
  
• Ritonavir and 2D6: Ritonavir is a potent inhibitor of drug metabolism mediated by the cytochrome P450 3A4 enzyme. It is also a moderate inhibitor of the cytochrome P450 2D6 isoform in vitro. 2D6 is an enzyme involved in the metabolism of a number of anti-depressant and psychoactive substances, and the clinical significance of RTV’s potential to inhibit 2D6 has been of some concern. Investigators from Abbott Laboratories found that LPV/RTV, which contains 100 mg RTV per dose, did not significantly alter the pharmacokinetics of desipramine (100 mg) in healthy volunteers [Preston, Abstract 433-W]. Desipramine is an antidepressant and a pure 2D6 substrate and is a good indicator of the potential of RTV to act as a clinically significant 2D6 inhibitor. This study suggests that a 100 mg bid dose of RTV (or RTV/LPV) can be co-administered safely with most 2D6 substrates.

20020501
JH20020503

©1997,1998,1999,2000,2001, 2002. The Johns Hopkins University AIDS Service, Division of Infectious Diseases. Permission to use and reproduce portions of this newsletter is hereby granted provided that author and publication are fully credited and both copyright and permission notice appear with reprinted material. Inquiries may be directed to Sharon McAvinue, Managing Editor. Website: Johns Hopkins AIDS Service.

AEGiS is made possible through unrestricted grants from Boehringer Ingelheim, iMetrikus, Inc., the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2002. This material is designed to support, not replace, the relationship that exists between you and your doctor.

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 2002. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.