Relative Potency of the HMG-CoA Reductase Inhibitors (a.k.a. Statins)
|Drug||Daily Dose||Relative Potency*|
|Fluvastatin (Lescol)||60 mg||1|
|Pravastatin (Pravachol)||60 mg||2|
|Lovastatin (Mevacor)||20 mg||3|
|Simvastatin (Zocor)||10 mg||6|
|Atorvastatin (Lipitor)||5 mg||12|
* Relative to effect of 60 mg fluvastatin (=1)
Source: ACTG A5047 study team
Many HIV-positive people are currently treated with statins to manage lipid elevations. Understanding the pharmacologic interactions between statins and HIV drugs is essential. Some statins have been implicated more than others in their potential for interactions with antiretroviral therapies.
Implementing lifestyle modifications, such as smoking cessation, is extremely important. Dietary adjustments are also an efficient way to lower cholesterol levels and can lead to reductions of up to 10-20%. To reduce cholesterol levels even more significantly, the use of a statin to lower LDL cholesterol levels may be needed.
Hydrochlorothiazide (a diuretic and antihypertensive drug, trade name Hydrodiuril) and atorvastatin should control blood pressure and LDL cholesterol, but myalgias and muscle weakness are not uncommon after a short time. Serum LFTs and CPK values should be monitored to see if either is elevated -- because the statin has the potential to cause muscle toxicity. Stopping statins in this case would be recommended. Switching to fenofibrate (TriCor) can reduce very low-density lipoprotein (VLDL) values, but it is not as effective in reducing low density lipoprotein cholesterol (LDL-C) elevations.
Some key points to bear in mind:
Simvastatin and lovastatin are lactone pro-drugs, which have to be converted into a hydroxy-acid form in order be more lipophilic and active. Rosuvastatin (Crestor) is the newest in the class (and the subject of much media attention regarding the risk of kidney toxicity). Statins are metabolized by CYP450.
Simvastatin is metabolized to simvastatin-acid significantly through the CYP3A pathway. The inhibition of CYP3A4, as occurs with many of the HIV protease inhibitors, can lead to fairly significant increases in simvastatin-acid levels -- and undesirable toxicities. A study investigating the impact of ritonavir on the metabolism of the statin drugs showed a remarkably high accumulation of simvastatin-acid levels, with increases of around 3,000%.
Pravastatin, by contrast, is metabolized differently and involves multiple oxidation pathways through the CYP450 system -- but, importantly, not CYP3A. So blood levels of pravastatin are actually reduced (by some 50%) when co-administered with ritonavir.
The two active metabolites of atorvastatin are generated by CYP3A, and so the inhibition of CYP3A4 will again (as is the case with simvastatin) lead to increased blood levels of atorvastatin. The key difference here is that blood levels of the active metabolites actually decrease when the CYP3A system is inhibited, so the overall increase in total active atorvastatin is not that large: about a 2-fold increase with ritonavir.
With nelfinavir, the observed effect is similar to that seen with ritonavir: atorvastatin levels are increased and simvastatin levels are increased significantly. Data with lopinavir/r (Kaletra) also show increases in atorvastatin levels -- by up to 5-fold. (Although in this study only the unchanged atorvastatin levels were examined and not the total active levels of atorvastatin. Kaletra did not show any significant changes with pravastatin.)
Other studies have shown that pravastatin exposure is reduced by 50% with the use of ritonavir and saquinavir, by 40% with efavirenz and by 50% with nelfinavir. Efavirenz has been shown to be a potent inducer of simvastatin metabolism, leading to reductions in exposure of 60% and of atorvastatin by 30%. Significant reductions in simvastatin, atorvastatin and pravastatin levels result in a slowing down of the LDL-cholesterol reduction in the presence of efavirenz.
A simple analysis of the data available on statins would suggest that pravastatin and fluvastatin are safe to use with CYP450 3A4 inhibitors, although there could be more data on fluvastatin. The efficacy of pravastatin may be somewhat compromised, as its metabolism is induced. Atorvastatin, however, should be used with caution, as drug-drug interactions with commonly used antiretrovirals could result in dangerously high atorvastatin levels when combined with protease inhibitors -- and uselessly low levels when combined with efavirenz (and very likely nevirapine, although data are difficult to come by).
Since drug levels of simvastatin and lovastatin are severely affected by inhibitors of the CYP3A4 system (e.g., the protease inhibitors), these statins should not be co-administered with the PIs.
Data for rosuvastatin are virtually non-existent, and so use of this drug (Crestor) concomitantly with HAART would best be avoided for now. Concerns about muscle damage (rhabdomyolysis), and its resultant effect on the kidneys, are particularly worrisome.
There are currently no interaction data for the statins and nevirapine. Until there are, the interaction effects should be suspected to be similar to those of efavirenz.
Tenofovir undergoes a complex metabolic process, leading to the possibility of significant inter-patient variability in drug levels and therefore risk of adverse events and antiviral response. Tenofovir has also been shown to increase the plasma concentrations of ddI -- although the mechanism of this interaction is as yet unclear. The active form of nucleoside analogue drugs such as AZT, ddI, 3TC and nucleotide analogues, such as tenofovir, is the phosphorylated derivative produced inside the host cell. These "anabolites" remain inside the cell and are de-phosphorylated there. Importantly, the rate of de-phosphorylation within the cell may be different from the rate of disappearance (half-life or t1/2) of the nucleoside which circulates in the plasma.
The tenofovir half-life (circulating in plasma) is approximately 17 hours, while that of its dephosphorylated anabolite can be as long as 50 hours -- more than adequate PK to supports its once-daily dosing. Intracellular phosphorylation (of tenofovir but also of all the nucleoside analogues as well) is difficult to measure. If the dephosphorylated anabolite is the important measure, might we be overdosing, for example, tenofovir at once-a-day? Has this been looked at?
The use of antiretroviral drugs in pregnancy is complicated by factors like hormonal and metabolic changes as well as the need for maternal and fetal safety. Use of therapeutic drug monitoring (TDM) can provide guidance for clinicians in the management of patients who have either sub-optimal or supra-optimal drug levels.
Example: a pregnant woman initially showed a good response to HAART despite low nelfinavir plasma levels, but one day her undetectable viral load broke through. The levels of nelfinavir are probably relevant here -- so the frequency of viral load monitoring should be increased. Increasing the nelfinavir dose might help to raise the blood levels of nelfinavir. Doing both (repeating the viral load and increasing the dose) would be most appropriate.
Nelfinavir plasma levels may be low during pregnancy. It is unknown whether these low levels are associated with sub-optimal antiviral response. With nelfinavir, a TID (three times a day) regimen would perhaps result in more reliable drug levels (although one retrospective ACTG study found this not to be the case.) Would prospective pharmacokinetic studies on the use of nelfinavir in pregnant women be helpful here?
Some clinicians would recommend increasing the frequency of viral load monitoring and perform therapeutic drug monitoring as soon as possible. Two weeks after starting treatment, therapeutic drug monitoring should be undertaken and the interpretation of that result should be used (along with the virologic response) to adjust the treatment.
The care of patients with prior experience of multiple antiretroviral agents and a limited number of options for switching presents challenges. The understanding of significant mutations that confer drug resistance and the impact of cumulative changes in the viral mutations can help guide therapeutic choices.
In the presence of many NAMs (nucleoside analogue mutations) plus the M184V, it is unlikely that any NRTI change will have much effect. With four or more TAMs (thymidine analogue mutations), there is little impact of new NRTIs, based on both expert opinion and the recent GSK 30009 debacle (abacavir plus 3TC plus tenofovir is a nix!). The co-administration of tenofovir and ddI produces an increase in plasma ddI concentrations, and tenofovir plus lopinavir/r increases tenofovir levels.
The ability of the protease inhibitors to inhibit viruses with accumulating resistance mutations is a relative phenomenon. Viruses may not be fully susceptible but may still retain some activity. We need to understand the impact of resistance and to combine what may be partially active drugs in a regimen that confers the greatest cumulative activity. By integrating pharmacologic and virologic interventions, we can enhance protease inhibitor activity and improve therapeutic outcomes.
Pharmacologic principles are based on risk-benefit ratios. Another ACTG study addresses the question of whether increasing protease inhibitor levels has an impact on viral suppression. Balancing the potency of the regimen needs to be looked at with any increase in toxicity: will patient adherence be compromised? A small pilot study compared the use of four Kaletra tablets (total 533 mg lopinavir and 133 mg ritonavir) to three Kaletra tablets, twice daily, but with the inclusion of 200 mg of additional ritonavir (total 400 mg lopinavir and 300 mg ritonavir). As might be expected, study outcomes showed an increased incidence of adverse events -- both gastro-intestinal events and lipid elevations -- among those in the group that got the lower dose of lopinavir but higher (over twice as much) dose of ritonavir.
In another similar study which enrolled individuals with multiple PI resistance mutations, the protease regimen was also increased to four (instead of three) Kaletra pills (533 mg lopinavir and 133 mg ritonavir), twice daily. At least in terms of target drug levels, the higher dose was better: the chance of reaching target concentrations of 5,500 ng/mL (which is predictive of a longer-term antiviral response) was reported to be higher in the four tablet group.
By now, there have been fairly definitive studies documenting the antagonistic interaction between lopinavir/r and amprenavir. When these drugs are used in combination, blood levels of the two principal component drugs (lopinavir and amprenavir) are lowered to dangerously ineffective levels. Although avoiding this combo is the most straightforward way to go, therapeutic drug monitoring might be useful in guiding dosage in situations where few other therapeutic choices are available.
Lopinavir/r interaction data for the newer version of amprenavir (fosamprenavir or Lexiva in the U.S. -- and for some reason Telzir in Europe) are still hard to come by. Although amprenavir and fosamprenavir are not exactly bio-equivalent, they are chemically close enough to each other to raise a warning flag. Interactions-wise, there has been little difference between the two. Given the complex interactions, if this combination must be used, therapeutic drug monitoring is probably a good idea to correct for any imbalances in the already adjusted dosing.
For Rob's data-rich version complete with tables and charts (both for the statins and the tenofovir half-life numbers), please check out our Web site: www.aidsinfonyc.org/tag.
Back to the TAGline August 2004 contents page.