There are a number of advantages and disadvantages of this approach. Typically, a PI-sparing regimen substitutes one of the NNRTI class of drugs [delavirdine (Rescriptor®), efavirenz (Sustiva®), or nevirapine (Viramune®)] for a protease inhibitor, while still using two of the nucleoside class of drugs (NARTIs), such as AZT and 3TC or ddI and d4T. Another option, which spares both the PI and the NNRTI classes, creates a three-drug treatment regimen made up of three NARTI drugs, or two NARTIs and a member of yet another class of drug, such a hydroxyurea.
A logical case can be made for both of these class-sparing approaches. It is now possible to achieve good suppression of virus in a majority of people just starting treatment without using either a protease inhibitor or a NNRTI. But it is far from clear which of the many possible combinations or types of regimens will actually help an HIV-infected person live the longest and most comfortable life. Just because a class-sparing regimen can suppress virus adequately, even for a year or more in a majority of people, doesn't necessarily mean that such regimens are the optimal methods of treatment or that they are fully equivalent to a protease-containing regimens.
The idea of substituting a NNRTI drug for a protease inhibitor in a typical three-drug combination in initial therapy offers a few theoretical advantages. First, it may allow a person to avoid or at least delay particular side effects usually associated with protease inhibitors. If the regimen can be as effective in suppressing virus as a PI-based regimen, then it might make sense to delay exposing people to protease inhibitor side effects. The limitation to this, however, is that the NNRTI drugs used instead have side effects of their own. Moreover, it is not yet entirely clear whether their use will always prevent the more troubling side effects of combination therapy, such as lipodystrophy (see "Lipodystrophy" in this issue). At least one NNRTI, like almost all protease inhibitors, can raise cholesterol and triglyceride levels. All NNRTIs, in varying degrees, can produce a troublesome rash, and one produces unique and often disturbing effects in the central nervous system and brain. In short, replacement of the PI with a NNRTI does not eliminate side effects, but quite likely changes them and, for some, reduces their severity.
A second motive for using PI-sparing regimens is to use a regimen that suppresses HIV for a prolonged period of time but does not contribute to the development of resistance to protease inhibitors. This would make it possible for people to effectively use a PI later, if the first combination regimen fails. Proponents argue that by sequencing the NNRTI and PI classes of drugs in this fashion, therapy overall will remain effective for a longer period. Not everyone agrees, however.
Critics point out that the most effective second or third line therapy for use after a protease inhibitor regimen fails is a new regimen that includes both a previously unused protease inhibitor and first time use of a NNRTI drug. If the patient has already used an NNRTI to the point of failure due to resistance, then this option is closed off for them, leaving them with few if any combinations likely to be effective when protease inhibitors begin to fail. Additionally, critics point out that the NNRTIs require only one mutation to render that entire class of drug ineffective. For instance, many believe (and some new data support) that when a person fails a NNRTI-containing regimen, such as AZT + 3TC + efavirenz, they will be resistant to both 3TC and the NNRTI. They counter that when a person fails a protease inhibitor-containing regimen, they may only be resistant to one of the nucleoside drugs and still are sensitive to the protease inhibitor for later use.
Either way, a person has only one shot at using the NNRTI-based regimen. There are no data at all that proves using it early or using it later results in the best durability for therapy overall. Both arguments about when to use the NNRTI regimen are theoretically valid, but neither is backed by results of research from well-designed studies. Simply put, anyone's view about when best to use this class of drugs is merely a matter of opinion.
Finally, some researchers argue in favor of using both the PI class of drugs and the NNRTI class, along with a NARTI or two, all together, concurrently in hopes of achieving the most intense and longest lasting viral suppression possible. The counter-argument, of course, is that when such a regimen fails, the patient has virtually nowhere left to turn, possibly having developed resistance to all classes of drugs at once. Again, there are no data showing that any of the three approaches leads to the longest possible life span or even the longest possible suppression of HIV.
The most recent theory of class-sparing regimens argues that if preserving the protease inhibitor for later use is a good idea, then why not spare both the PI and the NNRTI class, if that can be done without sacrificing effective viral suppression. This line of thinking leads to the use of three-drug regimens made up entirely of NARTI-class drugs. The most successful of these so far has included AZT and 3TC plus the new, highly potent NARTI, abacavir (Ziagen®). It is not as clear whether similar results can be achieved with all possible combinations of NARTIs, since abacavir represents newer and more potent generation of this class. There are a number of similarly potent new NARTIs in clinical studies. Another possibility is to combine two NARTI-class drugs plus an "outsider" drug like hydroxyurea (HU, Hydrea®), which doesn't fit into any of the common classes but has been shown to have good anti-HIV activity. Recent studies have shown results ranging from good to very good from such combinations, though longer-term data are still lacking.
What's to be gained by such an approach? In theory, it might be possible to extend the overall durability of treatment by using the available drug classes in three effective stages, applied sequentially. An example of this might begin therapy with a three-drug NARTI or NARTI plus hydroxyurea regimen, followed by a regimen made up of two NARTIs and a NNRTI, followed ultimately by a PI-based regimen. Believers suggest this might lead to the most extended period of effective therapy. Such an approach would also delay the various side effects associated with NNRTIs and PIs for later, while hopefully preserving long-term viral control.
A number of other sequences are also possible, such as starting on three NARTIs followed by a PI plus NNRTI plus NARTI combination. In effect, this approach argues that the class of drugs used isn't as critical as once thought, and that it is really the total number of drugs used in combinations that is responsible for their effectiveness.
Each class of drugs has its strengths, weaknesses and side effects, but what matters most is always using at least three drugs together, of which at least one or two should be drugs considered to be "highly active." Again, critics of this approach point out that there are only 5 NARTI drugs presently available and they have some degree of cross-resistance (where resistance to one drug results in resistance to another in the same class) to each other. As a result, they feel that it may be difficult to put together a potent follow-up treatment regimen.
Proponents of new class-sparing regimens base their claims of legitimacy on the belief that the "sparing" regimen will work as well as a PI regimen, at least for the majority of people starting therapy for the first time. This is fairly well established for some of the "sparing" regimens, but not as well for others. However, no such regimen has yet demonstrated the long-term durability proven for protease inhibitor regimens. This doesn't mean they have failed in this regard, but only that the studies have not yet run for as long as the best protease inhibitor studies (now out to three years for small groups).
There are at least theoretical reasons to question whether NNRTI combinations will last as long, but there are no hard data either way. NARTI combinations do not share this theoretical weakness, but are generally not perceived to be quite as potent overall.
One other issue clouding the comparison of PI regimens and PI-sparing regimens is a unique, recent research finding. People on protease inhibitors tend to experience continued benefits in terms of CD4+ cell counts and overall health well after a person develops resistance to the protease inhibitor and virus levels begin to rise. This is quite different from the experience with older drugs. In the past, when a person developed resistance to NARTI drugs such as AZT or ddI, CD4+ counts generally fell and overall health declined.
|"People should view with skepticism any claims that one approach is necessarily better than the others. We don't know what will last longest, and we don't know all the consequences of each choice."|
The fact that many people seem to have the opposite experience with protease inhibitors suggests that PIs may continue to provide some kind of benefit not yet understood. This phenomenon is referred to as a "disconnect" between a protease inhibitor's effect on HIV and its effects on the immune system. It is considered to be a good thing, but so far, there is no evidence that a similar "disconnect" happens with NNRTI-based regimens or all-NARTI regimens. So this may represent a unique benefit of PI-based therapy, one that should not be discounted too easily.
On the other side of the coin, it certainly seems that PI-based regimens present the most complex mix of side effects and drug interactions. These two must be weighed in the decision about what to use when starting therapy.
In short, there is no clear winner in this race as yet, and reasonable arguments can be made for all three approaches. People should view with skepticism any claims that one approach is necessarily better than the others. We don't know what will last longest, and we don't know all the consequences of each choice.
It is clear, however, that whatever choice is made for first line therapy it will definitely impact choices and outcomes of subsequent therapy. Therefore, it is important that such a choice be made only after full and careful examination of all the facts. The challenge is to understand the ups and downs of each approach and choose the one best suited to the individual's preferences, lifestyle and beliefs.
Class-Sparing Regimen Limitations
|Regimen||Possible Advantages||Possible Disadvantages||Drug Interaction Complications||Impact on Future Options|
|Best proven long-term viral suppression. Continued immune and general health benefits even after viral "failure."||Generally harder to use and adhere to. Long-term side effects might include (lipodystrophy, diarrhea, kidney problems, diabetes).||Mild to severe (ritonavir is the biggest problem; but its activity can also be exploited positively).||Preserves NNRTI for use in treatment failure. Usually leads to cross-resistance with other PIs.|
|Three-drug NNRTI-based regimen (PI-sparing)||Alleviates fears of PI-related side effects.* Generally easier to use and adhere to.||Not yet proven in long-term setting (limit, 1 year). No evidence of continued benefit after viral failure. Unique "neurological" side effect (feelings of disorientation) with efavirenz.||Delavirdine: effects can be exploited positively.
Efavirenz: moderate problems.
Nevirapine: minor problems.
|Preserves PIs for later use. Consumes NNRTI class, making it unavailable for use in later treatment failure.|
|Three drug NARTI or NARTI plus - regimen (NNRTI- and PI-sparing)||Saves PI and NNRTI for later use. Generally easier to use and adhere to.||Alleviates fears of PI and NNRTI-related side effects. Introduces few if any new side effects (most are present in other combinations anyway).||No major problems.||Preserves both PI and NNRTI classes for later use. May hasten resistance to other NARTIs.|
|* Note: some side effects being attributed to protease inhibitor therapy, such as lipodystrophy, have not been proven to be strictly associated with the use of protease inhibitor-containing regimens. One NNRTI, efavirenz, has been shown to result in increases in cholesterol and triglycerides, which some believe is related to the risk of developing lipodystrophy.
Back to the Project Inform Perspective April 1999 contents page.