July 2, 2007
Table of Contents
Part I: Strategies for First-Line Therapy
HIV-infected patients and their health care providers now have an impressive array of antiretrovirals from which to choose for first-line therapy. After several years without significant modification, the list of preferred regimens in the U.S. Department of Health and Human Services (DHHS) guidelines for the initial therapy of HIV-infected adults has been expanded beyond lopinavir/ritonavir (LPV/r, Kaletra)- and efavirenz (EFV, Sustiva, Stocrin)-based regimens (Table 1).1 The addition of ritonavir (RTV, Norvir)-boosted atazanavir (ATV, Reyataz) and fosamprenavir (FPV, Lexiva, Telzir) to the long, resolute list of recommended therapies reflects not only the emergence of data from clinical trials comparing these agents, but was also driven by the difficult-to-ignore popularity of these antiretrovirals for the initial treatment of HIV infection.
This blessing of options, however, can quickly become a curse of confusion when there are so many choices available that the chooser becomes paralyzed by indecision -- witness Medicare Part D, 401K plans and toothpaste (tartar control, whitening, original flavor, baking soda, mint or all the above?).
This review will attempt to provide some relief by examining the relative virtues and deficiencies of the DHHS' preferred first-line agents. Two case studies follow the review to help provide guidance as to the practical application of this data.
|Column A||Column B|
|Atazanavir + ritonavir (AIII)||Tenofovir/emtricitabine (AII)|
|Fosamprenavir + ritonavir twice daily (AII)||Zidovudine/lamivudine (AII)|
|Lopinavir/ritonavir twice daily (AII)|
|Atazanavir (unboosted) (BII)||Abacavir/lamivudine (BII)|
|Fosamprenavir (unboosted) (BII)||Didanosine + lamivudine (BII)|
|Fosamprenavir + ritonavir once daily (BII)|
|Lopinavir/ritonavir once daily (BII)|
From the October 10, 2006 DHHS Guidelines
Rating Scheme for Clinical Practice:
The first decision to be made when selecting among antiretroviral therapies is whether to craft a combination that includes a protease inhibitor (PI) or a non-nucleoside reverse transcriptase inhibitor (NNRTI). For most patients, this choice is further reduced to choosing between either a ritonavir-boosted PI-based regimen or an efavirenz-based regimen. The added potency and drug exposure afforded by the pharmacological boosting of PIs with ritonavir makes unboosted-PI use an approach that should generally be reserved for select patients who are unable to tolerate or unwilling to take ritonavir and need a PI.
Among the NNRTIs, nevirapine (NVP, Viramune) remains an alternative to efavirenz, but its use is limited by a number of challenges -- including CD4+ cell count influenced toxicity -- and, therefore, it has been relegated to serve as an efavirenz understudy.
We have come to rely on antiretroviral regimens containing either NNRTIs, especially efavirenz, or ritonavir-boosted PIs due to their demonstrated potency. But there have been very few well-powered, prospective trials comparing these drug classes. A 2006 analysis by John A. Bartlett from Duke University and colleagues of all published clinical trials of triple antiretroviral therapy in treatment-naive patients found that ritonavir-boosted PI and NNRTI-based regimens were equally capable of reducing HIV viremia over 48 weeks; however, boosted PIs produced greater CD4+ cell count increases.2
Remarkably, only two published large clinical trials have pitted a boosted PI against efavirenz: 1) the Clinically Significant Long-Term Antiretroviral Sequential Sequencing (CLASS) study of amprenavir (APV, Agenerase) + ritonavir versus efavirenz, and 2) the AIDS Clinical Trials Group (ACTG) study A5142, a trial that compared lopinavir/ritonavir versus efavirenz.3,4
Perhaps presaging the results of study A5142, the CLASS study found that when combined with abacavir (ABC, Ziagen) + lamivudine (3TC, Epivir), efavirenz was better able to suppress HIV viremia to below 50 copies/mL compared to ritonavir-boosted amprenavir taken with the same nucleoside reverse transcriptase inhibitors (NRTIs).4
In ACTG study A5142, efavirenz plus two NRTIs was found to produce higher rates of suppression of HIV viremia to below 50 copies/mL when compared to a regimen of lopinavir/ritonavir and two NRTIs. Again, as in the Bartlett analysis of clinical trials of triple-drug combinations, CD4+ cell count gains were greater in patients who were assigned the boosted PI. This study also demonstrated the relative tolerability of these therapies, as described below.5
These and other clinical trials provide useful data that can help guide clinicians and their patients. In many cases, recent studies of HIV therapies have produced unexpected results, suggesting there is still much about even currently available antiretrovirals that we do not understand. Therefore, it is imperative that patients be made aware of opportunities to participate in HIV clinical research. Clinicians should not assume their patient is not interested in joining a clinical study and should make every effort to identify where such studies are being conducted and refer potential participants to such centers, whether at university, private offices or ACTG sites. In the United States, federally funded ACTG sites can be found at www.aactg.org/clinicaltrials_actulocator.asp. Additional study opportunities can be searched at www.aidsinfo.nih.gov/ClinicalTrials/Default.aspx?MenuItem=ClinicalTrials.
Now let's take a look at the advantages and disadvantages of selecting an efavirenz-based combination as an initial regimen.
Table 3: NNRTI-Based Regimens (1 NNRTI + 2 NRTIs)
From the October 10, 2006 DHHS Guidelines
Rating Scheme for Clinical Practice:
The dosing convenience of the first one pill, once-a-day combination to treat HIV infection is unmatched. The combination formulation of tenofovir (TDF, Viread), emtricitabine (FTC, Emtriva) and efavirenz has greatly simplified therapy -- a long-stated desire of both patients and clinicians. Convenience is only slightly affected with the use of an alternative dual-nucleoside combination, the fixed-dose combination of abacavir/lamivudine (ABC/3TC, Epzicom, Kivexa) with efavirenz, which requires two pills once a day. The fixed-dose combination of zidovudine/lamivudine (AZT/3TC, Combivir), a preferred NRTI by the DHHS, raises the pill burden ante to three pills a day and is a twice-a-day regimen. These NRTIs are discussed in greater detail below.
In addition to low pill burden and once-daily dosing, the potency of efavirenz-based regimens has enhanced the attractiveness of such combinations. As described above, in study A5142, significantly more participants assigned to efavirenz combined with two NRTIs achieved a viral load below 50 copies/mL after 96 weeks than those who were receiving two NRTIs and lopinavir/ritonavir (89% versus 77%, intent-to-treat analysis).4
In other trials, efavirenz combined with tenofovir + lamivudine, tenofovir/emtricitabine (TDF/FTC, Truvada), abacavir/lamivudine or zidovudine/lamivudine was observed to be at least as effective as any comparator regimen in so far as suppression of HIV viral load while being relatively well tolerated.6-9
The long half-life of efavirenz also is advantageous and permits not only once-daily dosing, but also likely provides some degree of forgiveness for sub-optimal adherence.10 Therefore, it is likely that viral suppression can be achieved and maintained even among patients who are not 90% adherent to efavirenz dosing.
The major disadvantages of efavirenz are well known to HIV clinicians and to many patients. The most problematic of this drug's adverse effects is the high rate of central nervous system (CNS) toxicity. Dizziness, sleep disturbance and perhaps mood disorders can develop in a significant proportion of patients who are receiving efavirenz. While these symptoms resolve for the vast majority of patients over the first weeks of therapy, there are reports of some of these effects persisting longer term.11-13 In clinical trials, however, CNS toxicity has been treatment limiting in less than 5% to 10% of participants.
To reduce the potential impact of these CNS toxicities on daily functioning, patients are instructed to take efavirenz before going to sleep. For many patients, the nighttime dosing of efavirenz-based regimens is not an issue; for others, this restriction can present challenges to drug adherence.
Rash can also develop during efavirenz therapy. While this adverse effect is rarely severe and typically does not require drug discontinuation, when efavirenz is co-administered with abacavir, the rash may be mistaken for a manifestation of an abacavir hypersensitivity reaction. This could lead to the unjustified lifelong banishment of abacavir from the patient's list of future treatment options.
Peripheral Fat Wasting
Results from study A5142 also suggest that when combined with NRTIs, particularly stavudine (d4T, Zerit) or zidovudine (AZT, Retrovir), efavirenz is more likely to produce peripheral fat wasting than lopinavir/ritonavir coupled with the same NRTIs.5 Significant fat wasting was uncommon in the trial among participants who were receiving tenofovir and efavirenz.
However, stavudine, not unexpectedly, was associated with the greatest risk of fat loss, with 51% of those on this NRTI plus efavirenz experiencing a 20% or greater loss of limb fat compared to 33% of those taking stavudine along with lopinavir/ritonavir. This degree of lipoatrophy was observed in 40% of those taking zidovudine/lamivudine and efavirenz versus 16% of patients who were on these NRTIs plus lopinavir/ritonavir. To what extent efavirenz plus zidovudine/lamivudine produces clinically evident lipoatrophy is unclear; however, given the twice-a-day dosing of zidovudine, the early hematologic complications associated with this NRTI and these body shape data, use of zidovudine/lamivudine as an initial agent has been waning.
The effect of efavirenz on lipid parameters has been even less clear. In studies 903 and 934, which pitted tenofovir against stavudine and zidovudine, respectively, efavirenz was found to increase cholesterol subsets (including high-density lipoprotein [HDL] levels) to some degree, albeit to a lesser extent when combined with tenofovir.6,7 However, the effect of tenofovir + lamivudine or emtricitabine + efavirenz on triglyceride levels in these trials was essentially nil.
In another study (BMS 034), which compared unboosted atazanavir and efavirenz and in which all trial participants were treated with zidovudine and lamivudine, there were greater increases in low-density lipoprotein [LDL] cholesterol and triglyceride levels among the patients who were receiving efavirenz compared to those who were assigned the atazanavir.14
It is well appreciated that efavirenz does not have as favorable a lipid profile as its NNRTI sibling, nevirapine. In the 2NN study, which compared these two NNRTIs, LDL cholesterol and triglyceride levels rose to a significantly greater extent during efavirenz therapy than with nevirapine.15 Further, HDL cholesterol level increases were greater with nevirapine.
Lastly, study A5142 demonstrated little difference between efavirenz and lopinavir/ritonavir vis-à-vis lipid changes during therapy.5 Therefore, efavirenz does appear to increase cholesterol subsets. The data on the effect of efavirenz on triglyceride levels remain mixed.
Viral Drug Resistance
Finally, efavirenz has a relatively low genetic barrier to resistance -- meaning drug resistance to efavirenz develops after only a few viral mutations. In patients who are sufficiently non-adherent, viral resistance emerges rapidly, rendering the drug and all other first generation NNRTIs (e.g., nevirapine and delavirdine [DLV, Rescriptor]) unable to suppress viral replication. There is also a theoretical concern that, when an efavirenz-based regimen is stopped, the long half-life of this NNRTI relative to its companion NRTIs could lead to the selection of resistant virus. It is unclear whether this is an issue when the regimen includes tenofovir and emtricitabine -- drugs that may also persist for days after discontinuation. Accumulated NNRTI-associated mutations will also reduce viral susceptibility to second generation NNRTIs, such as etravirine (TMC125) and rilpivirine (TMC278). Therefore, patients who are virologically failing an NNRTI regimen should generally have their regimen stopped or changed as soon as possible to avoid handicapping the effectiveness of the second generation NNRTIs in development.
The low genetic barrier to resistance of efavirenz is evident in study A5142. Of the 60 patients who experienced study defined virologic failure while receiving efavirenz and two NRTIs, 33% of the 33 participants with samples available for genotypic resistance testing had evidence of efavirenz resistance.4 The ease with which efavirenz resistance develops has public health implications as the viral mutations associated with resistance to the NNRTI class can be transmitted to others and then persists in the infected individual, potentially reducing the effectiveness of NNRTI therapy in that individual. In urban settings in the United States and Europe, NNRTI resistance mutations have been detected in individuals with chronic HIV infection who are naive to antiretroviral therapy, and, increasingly, baseline NNRTI resistance is being detected.16,17 Therefore, although a convenient and potent agent, efavirenz has some important tolerability and resistance issues that must be appreciated by the clinician and patient when regimens containing this drug are being considered.
The potency and convenience of PIs have increased as a result of pharmacological boosting with ritonavir. Boosting with low doses of ritonavir has resulted in the once- or twice-a-day administration of antiretrovirals in this class and has also permitted the achievement of PI drug levels that exceed the concentration needed to inhibit viral replication, even in virus that may harbor some PI resistance.
Table 4: PI-Based Regimens (1 or 2 PIs + 2 NRTIs)
From the October 10, 2006 DHHS Guidelines
Rating Scheme for Clinical Practice:
The prolonged half-life and antiretroviral potency of boosted PIs may also lend agents of this class a degree of adherence flexibility. However, unlike the NNRTIs, viral drug resistance to ritonavir-boosted PIs is very uncommon during treatment failure. In contrast to the efavirenz resistance seen among patients with virologic failure while on efavirenz + two NRTIs, no resistance to lopinavir was detected among the 52 ACTG study A5142 participants who experienced virologic failure while on lopinavir/ritonavir + two NRTIs and who had specimens available for resistance testing.4 Similarly, an absence of PI resistance despite virologic failure has been observed in trials of all other ritonavir-boosted PIs.18-20
Of the boosted PIs listed as preferred for initial therapy by the DHHS, disparate pill counts, dosing frequencies, toxicities and cost can also distinguish among these agents. In treatment-naive patients atazanavir can be co-administered once daily, as can fosamprenavir -- although twice-a-day dosing of fosamprenavir remains the preferred dosing recommended in the DHHS guidelines.
Lopinavir/ritonavir, the only boosted PI that is co-formulated with ritonavir, can be taken once daily, although tolerability issues and variability in drug concentrations among patients who are administered this drug once daily have tempered widespread enthusiasm for this approach and, according to the guideline committee, once daily is not a preferred dosing schedule.
Recently, lopinavir/ritonavir and fosamprenavir + ritonavir were compared in the KLEAN (Kaletra versus Lexiva with Epivir and Abacavir in ART-Naive Patients) study, which is a large, well-powered clinical trial during which all patients also received abacavir/lamivudine.19 This study demonstrated near identical virologic and immunologic performances by these two regimens.
In addition, toxicity rates were also extremely similar across the study arms in the KLEAN study, with diarrhea being the most commonly reported symptom (11% - 13%). Increases in LDL and HDL cholesterol, as well as triglyceride levels, were also indistinguishable across the study arms.
There are less data regarding ritonavir-boosted atazanavir in treatment-naive patients. In BMS 034 (see above), unboosted atazanavir was found to be non-inferior to efavirenz, when each was combined with zidovudine/lamivudine.14 The BMS 089 study of atazanavir with and without ritonavir boosting, along with extended release stavudine and lamivudine, found that the rate of virologic failure was higher with unboosted atazanavir, but tolerability and lipids favored the unboosted arm.20 Remarkably, despite the popularity of the combination of tenofovir/emtricitabine and boosted atazanavir, to date no large comparative trial has included this particular combination.
There has been a longstanding concern that PIs can cause unwanted changes in body fat, particularly increases in abdominal fat. However, studies that have rigorously and objectively evaluated changes in adipose tissue depots during antiretroviral therapy have demonstrated consistently similar gains in abdominal fat with PI- and NNRTI-based regimens. Therefore, fat accumulation cannot be considered a distinct disadvantage of PIs.
Indeed, in two studies, A5142 and M03-613, which compared efavirenz plus zidovudine/lamivudine with an induction maintenance strategy of lopinavir/ritonavir and the same NRTIs, both found that abdominal fat increased to the same extent in patients treated with lopinavir/ritonavir and those treated with efavirenz.5,21
Further, as described above, in study A5142, lopinavir/ritonavir was associated with less limb fat loss than efavirenz. In the BMS 034 trial comparing efavirenz versus unboosted atazanavir, no differences in body shape changes were found between these agents when combined with zidovudine/lamivudine.14 Therefore, recent data suggest that abdominal fat increases are likely universal during HIV therapy with currently available regimens and are not specific to any one drug or drug class.
The use of ritonavir to boost PIs is a double-edged proposition. It accounts for much of the therapeutic success of this drug class, but also its greatest liabilities.
Tolerability has been the most significant concern regarding boosted-PI therapy, with most of the attention focused on gastrointestinal and metabolic adverse effects. And, to a large extent, much of the tolerability issues related to boosted PIs come at the hand of ritonavir.
Because even low doses of ritonavir can prompt gastrointestinal problems and because HIV-infected people now expect that treatment will result in decades of survival, there is impatience regarding such adverse events that impact quality of life.
As was demonstrated in the KLEAN study discussed above, over 10% of participants had moderate to severe gastrointestinal complaints while they were on ritonavir-boosted fosamprenavir and lopinavir, and more mild gastrointestinal effects are even more common.19 Available data indicate that gastrointestinal unpleasantness is less of a problem with unboosted PIs.20
Atazanavir, and to a lesser extent the PI relic indinavir (IDV, Crixivan), can produce increased levels of unconjugated bilirubin in the blood. The severity of hyperbilirubinemia is dose related and ritonavir boosting of atazanavir heightens the risk of this adverse effect. For most persons, jaundice is mild and not treatment limiting, but for some scleral icterus becomes intolerable. An additional gastrointestinal consideration when considering atazanavir is its interaction with proton pump inhibitors (PPIs). As atazanavir requires an acidic gastric environment to be absorbed, agents such as PPIs that severely raise gastric PH should not be administered with atazanavir.
Another thing that should be kept in mind is that as a potent inhibitor of the hepatic cytochrome P450 isoenzymes responsible for the metabolism of a vast number of medications, ritonavir often interacts with commonly used drugs, both prescription and illicit. In general, ritonavir will increase the level of the companion substance. There's a long list of drug-drug interactions with various medications, such as methadone (Dolophine, Methadose), fluconazole (Diflucan), diazepam (Diazepam Intensol, Valium), oxycodone (Endocodone, OxyContin, Oxydose, OxyFast, OxyIR, Percolone, Roxicodone), paclitaxel (Onxol, Taxol), sildenafil (Viagra) and even St. John's wort. Clinicians and patients must be aware of the potential for such interactions and take care whenever taking other drugs with ritonavir.
With the increase in price of ritonavir, the cost of boosted PIs (except, of course for lopinavir/ritonavir, which is co-formulated) skyrocketed. For patients with insurance plans that require self-pay of a proportion of prescription costs, health care organizations not covered by the Public Health Service pricing and private insurers, the effect of the price increase has been dramatic. Compounding this situation is the high cost of some companion PIs, such as atazanavir and fosamprenavir. For many people receiving HIV therapy in the United States through public or comprehensive private insurance, the cost of antiretrovirals is not a concern. But for many others, expense is a major factor when selecting among HIV regimens and the overall burden of drug cost on HIV health care in this country is substantial.
An additional concern for all boosted PIs, save lopinavir/ritonavir, is the sensitivity of the ritonavir gel-caps to heat. This is an important issue for patients living in warmer climates who have no access to refrigeration; although a more heat tolerant formulation of ritonavir is under development.
Co-formulated NRTIs now dominate the NRTI market. On the preferred list of the DHHS recommendations are the co-formulated versions of tenofovir/emtricitabine and zidovudine/lamivudine. However, use of zidovudine/lamivudine has been decreasing. As described above, the twice-a-day dosing, the association with fat wasting and the other side effects, as well as the advent of once-daily combination antiretrovirals, are largely responsible for reduced reliance on this agent.
The dust has largely settled in the battle of the NRTIs, which began with the ascendance of tenofovir, and each agent has now found a fairly stable niche. Tenofovir/emtricitabine and abacavir/lamivudine offer both potency and the convenience of once-daily dosing of a single tablet.
Table 5: Dual-Nucleoside Options as Part of Initial Combination Therapy
From the October 10, 2006 DHHS Guidelines
Rating Scheme for Clinical Practice:
The best data to date suggest that changes in renal function during tenofovir therapy are modest and typically of limited clinical significance.22 However, concomitant therapy with other drugs that can be nephrotoxic and failure to appropriately dose adjust for creatinine clearance can trigger a decline in renal function.
Abacavir is not included in the DHHS' list of preferred NRTI regimens; however, it remains an important alternative to other NRTIs. Fear of hypersensitivity has been a perennial drag on abacavir use. Although the reaction occurs in less than 10% of those administered the drug, patients may find the dire warnings regarding this complication off-putting and clinicians may find the counseling that must accompany the prescription of this NRTI burdensome. However, emerging data suggesting that abacavir hypersensitivity reactions are uncommon in persons of African ancestry and that HLA-B57 typing may predict who will not develop the reaction offer an opportunity for a potential resurgence of this NRTI.23,24
The antiretroviral landscape has and will continue to shift. New agents are being developed to treat HIV infection and several of these are being studied as initial regimens. The CCR5 antagonist maraviroc (UK-427,857; with the expected brand name of Selzentry) theoretically may be better suited for earlier infection, when CCR5 tropic virus predominates, than for a more advanced disease state, when CXCR4 and dual/mixed virus emerges.
Likewise, raltegravir (MK-0518; with the expected brand name of Isentress), an inhibitor of HIV integrase, is being studied as part of a ritonavir-free first-line regimen, while elvitegravir (GS 9137), another drug of this class, is a bit further behind in development and will require ritonavir boosting. Rilpivirine is also not as far along in development as either maraviroc or raltegravir, but there are plans for it to be tested in clinical trials of treatment-naive patients.
Lastly, darunavir (TMC114, Prezista), a ritonavir-boosted PI approved for use in treatment-experienced patients, is under investigation as an initial therapy at a reduced once-daily dose. Whether any of these drugs will be added to or supplant any of the existing preferred agents remains to be seen.
Given the variety of antiretroviral choices available to patients starting HIV therapy, we've entered an era that permits the customization of treatment to meet individual needs and preferences. Clinicians need to weigh the advantages and disadvantages of potential regimens when selecting HIV treatment for a particular patient. Any of the so-called preferred regimens are very likely to suppress viral load and increase CD4+ cell counts. Therefore, tolerability, convenience and cost remain to distinguish between regimens.
As illustrated in the case studies, there is no one regimen that fits all. A potential drawback of one regimen may be of minimal consequence to one patient, but can be a deal-breaker for another patient. Ultimately, the clinician needs to be familiar not only with the medications used to treat HIV infection, but also with the individual patient for whom therapy is prescribed.
Please read both the Strategies for First-Line Therapy and the Case Studies sections of this article.
Please note: Knowledge about HIV changes rapidly. Note the date of this article, 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 article.