Integrase inhibitors, like many antiretroviral medications, work by binding to one of several specific enzymes that HIV uses when copying itself. The current approved HIV drugs target and form a tight bond with HIV's reverse transcriptase and protease enzymes. Once bound to the drug, the enzyme is unavailable to anything around it, including HIV, hindering HIV's ability to make copies of itself.
In the same way, integrase inhibitors bind to integrase, the enzyme HIV uses to insert its newly made genetic material (DNA) into a CD4 cell's DNA. Their development marks an important frontier in HIV research, since integrase is one of the only enzymes that hasn't yet been successfully targeted. After ten years of attempts to design an integrase inhibitor with little success (including L-870810, zintevir, and S-1360), research presented at the 13th Conference on Retroviruses and Opportunistic Infections (CROI) finally shows two candidates, MK-518 and GS-9317, that could be the first successful integrase inhibitors.
One of the investigators, Beatriz Grinsztejn, reported that after four months, 65-85% of the people in the trial had their HIV viral load drop to below 400. Furthermore, between 56 and 72% of those taking MK-518 had their viral load drop below 50 -- an impressive result, and quite promising for people with highly resistant virus. During this same time period, people taking the 600 mg and 400 mg doses saw their CD4 counts increase by 100 to 110 cells on average, and those on the 200 mg dose on average had a 30-cell increase. While it will be useful to see what the results look like at the end of the six-month trial, these findings show that, at least initially, MK-518 is quite effective at lowering viral load in people who have high levels of resistance to multiple HIV drugs.
Understandably, this report generated enthusiasm about MK-518's prospects as a potential integrase inhibitor. Previous studies showed that other HIV drugs most likely won't interact with MK-518, and that it can be taken with or without food. While many HIV drugs use the protein CYP3A4 to get broken down in the body; MK-518 does not. This means that HIV drugs such as protease inhibitors most likely won't interact with MK-518. It's also unlikely that we'll need to add low-dose Norvir (ritonavir) to MK-518 to create simpler dosing. It also looks as if MK-518 will be able to be taken with or without food, and without any dietary restrictions.
As far as side effects go, so far they are few and mild. In Grinsztejn's study, nausea was the most commonly reported side effect (11% among those on the 600 mg dose, and 5% among those on the 400 and 200 mg doses); however, just as many people taking placebo also reported nausea. In addition, 2-10% of patients taking MK-518 reported diarrhea, fatigue, headache, or itching (pruritis). But these same side effects (with the exception of pruritis) were also reported in 2-9% of those taking placebo. Also, 8% of those on the 200 mg and 600 mg doses reported higher than normal bilirubin levels, which sometimes signal liver problems. In addition, 5% of those on the highest dose had increased amylase levels (which can signal pancreas problems).
The final results of this study are promising, but they only begin to provide the understanding we need about MK-518 in order to fully discern how useful it would be as an antiretroviral. The good news is that a larger and longer phase III study has recently opened, and will hopefully answer these questions.
After ten days, there was an overall reduction in viral load of 1.44 to 2 logs among the various doses, which means that, on average, people who had a viral load of 100,000 would have dropped to 3,571-1,000 after 10 days. Those on placebo, on the other hand, reported only a 0.26 log drop, which would mean a much smaller drop, down to 50,000 copies.
While this shows that GS-9137 may have a strong antiviral effect, the study's main purpose was to look at safety and side effects. Overall, GS-9137 was well tolerated in this short study and no one withdrew due to side effects. The side effects reported included fatigue, diarrhea, headache, and nausea, and were mild to moderate. Individuals taking 400 mg twice-daily also reported an increase in triglyceride levels, and those taking 50 mg boosted with Norvir and those on placebo reported an increase in amylase levels. These lab abnormalities were moderate to severe but are not conclusive due to the small number of people in the study.
Unlike MK-518, GS-9137 is broken down through the CYP3A pathway in the liver. This means that it is more likely to interact with other HIV drugs; either the dose of GS-9137 or the other drugs might need to be adjusted. For example, levels of GS-9137 have already been shown to rise 20 times higher when taken with Norvir, which shows there may be some benefit to taking GS-9137 with low-dose Norvir. Also, in order to be broken down properly, GS-9137 needs to be taken with food.
In addition, it looks as though GS-9137 stays in the body longer than many other HIV drugs (as long as nine hours), which promises less frequent dosing. At this early stage, it looks like there shouldn't be any cross-resistance between GS-9137 and MK-518, which means that if someone's virus stops responding to one of the drugs, the other one might still work. But these preliminary findings tell us little about how well GS-9137 will work or its safety in larger groups of people. A larger and longer phase II study is set to open in May and should provide us with many more answers.
Donna M. Kaminski is the former Associate Director of Treatment Education at ACRIA and currently a first-year medical student.