Convergent Treatment Strategies

This talk was the first of two that contrasted the major point of diversion in the common approach to antivirals -- whether to use agents all directed against the reverse transcriptase enzyme or whether instead to use protease inhibitors as well. Often called "PI sparing", convergent antiretroviral approaches have several potential advantages. These were outlined by Dr. Raffi, and include:

• Randomized studies suggest equivalence of PI containing and PI sparing approaches

• Concerns for PI side effects and desire to postpone their use

• Relatively easy dosing with two of the three available nonnucleosides, with once daily dosing of just two or three pills, without food restrictions

However, there might be some concerns with this approach as well, including:

• PI containing regimens have proof they reduce AIDS illnesses and allow for a longer life; studies with other approaches have not been done to confirm this is similarly true.

• PI's can have a higher "genetic barrier" to resistance, meaning that it usually takes more than one resistance mutation to create resistance to these agents, whereas the non-nucleosides are considered to lose potency after just one mutation.

• After viral rebound, PI containing regimens show durable CD4 stability for up to two years. This has not yet been reported for other options, raising the possibility that there may be either some effect of the PI's directly on the immune system and/or on HIV. One consequence of high level resistance to the PI's may be that it weakens the ability of HIV to grow, and this preserves the CD4 count.

Whichever approach is taken, it is desirable that it is both highly effective, as well as allows for a subsequent chance(s) for viral suppression. However, given the current medications as well as those in active development, the desire for a long term strategy may be equally true regardless of which of these options (PI sparing or PI containing) is selected first.

Triple nucleoside therapy now has two studies confirm the success of this approach in most patients. There are randomized studies using the combination of stavudine (d4T), didanosine (ddI) and lamivudine (3TC). There are also studies of zidovudine (AZT), 3TC, and abacavir -- both of these triple nucleoside approaches have been compared to PI containing regimens. These studies confirm that triple nucleoside approaches are generally equivalent in rates of viral suppression to regimens containing either a PI or a nonnucleoside. However, both studies suggested a somewhat lower success rate of achieving viral suppression below 50 copies for persons who start with a viral load over 50-100 thousand, and thus the triple nucleoside approach can be considered to be best targeted for those with lower viral loads at baseline. Further, some clinicians are still awaiting lymph node data to confirm that the suppression of HIV in the blood corresponds to what is seen in the nodes. One advantage of this class is that, while there are rare cases of multinucleoside resistant HIV, use of triple regimens (if viral rebound occurs) rarely leads to broad cross resistance within the class, unlike what is expected when using nonnucleosides.

Durable suppression has been demonstrated for approaches containing nonnucleosides beyond one year as seen in a study known as INCAS, as well as DuPont's 006. The former was a study using AZT, ddI, and nevirapine (Viramune), and there are data out beyond two years that shows that those patients who initially achieved a viral load below 20 copies have a durable response. Other studies based on nevirapine have shown similar success rates, including one known as Virgo, using d4T and ddI. The advantage in these studies is that both the ddI and nevirapine can be taken just once daily, providing a relatively simple antiviral combination. Dr. Raffi presented the data from the subset of patients in these studies who started with a viral load over 100,000, and showed similar rates of success as compared to those below this level. This issue of success at higher viral loads has also been demonstrated in several regimens containing two nucleosides and efavirenz. There have been studies using efavirenz with either AZT plus 3TC, d4T with 3TC, or d4T plus ddI, and each has shown similar rates of viral suppression at both lower and higher viral loads. (This similarity in the success rate appears to differ from several studies when two nucleosides are given with one protease inhibitor. In that case there is more often a decreased viral suppression rate in persons with a viral load over 100,000 -- as a result, dual PI combinations could be recommended for this circumstance when using PI's). Regimens based on nucleosides plus efavirenz (Sustiva) have also confirmed the durability of viral suppression, as those who initially went below 50 copies were more likely to have continued suppression two years later than with two regimens based on indinavir (Crixivan).

The toxicities of nevirapine tend to include the risk of a rash in the first few weeks after starting the drug. In the Virgo study, this rash was seen in 17%, although only 8% needed to discontinue the medication for this reason. (A symposium focussing on nevirapine suggested that the use of corticosteroids when starting nevirapine would decrease this risk of rash.) The other side effect seen with nevirapine is liver toxicity, and in these studies was seen in about 8% of those starting the regimen. The toxicities of triple nucleosides were also reviewed, with a reminder of the primary side effect seen when using abacavir, specifically the hypersensitivity reaction, which in this study was reported in 6.7% of those on the drug.

For those who started therapy with a PI containing regimen but are experiencing a side effect (or concerned about future side effects), there have been a few studies that have substituted a non-PI medication for the PI. In general these studies have been successful in maintaining viral suppression after the substitution. One of the primary factors identified by Dr. Raffi as a predictor of who would be successful after a substitution are those who were antiviral naïve before starting the PI containing regimen, i.e. those who were on their first combination. When compared to those who had more antiviral experience, antiviral naïve patients had less than a 5% risk of losing viral suppression when substituting a nonnucleoside, whereas about 25% had viral rebound if they had prior antiviral experience. There are data that these substitutions will allow resolution in some of the side effects associated with PI's. The one that has been most proven is the normalization of lipids when using either nevirapine or abacavir instead of a PI. As there are few if any lipid abnormalities seen when using triple nucleosides, one study that replaced abacavir for the PI showed normalization of the lipids in the majority of those who made the switch. Similar data are available when nevirapine was used. Improvements in glucose control have also been noted. Changes in body shape have been reported but may be less evident in the first six months after such substitutions, and longer term data are needed to confirm the early reports suggesting some benefit.