November 11, 2009
Eric Daar: We now consider that to be another thing to talk to our patients about in situations where they are concerned about potentially transmitting HIV to an HIV-negative partner (for example, in the cases where someone is in a stable relationship with a discordant partner). We certainly don't suggest they shouldn't be using condoms and other means to prevent HIV transmission, but it's reasonable to explain to them that there is at least inferential data, and hopefully soon there will be randomized control trial data, showing that being on therapy may reduce the risk of transmission in the event of things like a broken condom.
Bonnie Goldman: Let's turn now to the challenge of selecting a regimen. How do experienced doctors choose a regimen? Can you walk us through your thinking, Dr. Daar?
Eric Daar: We have lots of great choices, and we have large, randomized control trials now for virtually all of our primary options, with at least two years of follow-up that demonstrate their overall safety and efficacy.
I think in the past, and even today, the standard regimens have usually included two nucleosides [NRTIs] with either a protease inhibitor or an NNRTI [non-nucleoside reverse transcriptase inhibitor]. Based upon tolerability and the efficacy data, we now have a variety of choices amongst the protease inhibitors, including many once-a-day protease inhibitors that will create once-a-day regimens with fixed-dose combination nucleosides.
|This chart was adapted from the Panel on Antiretroviral Guidelines for Adults and Adolescents initial combination regimens for the antiretroviral-naive patient issued from the U.S. Dept of Health and Human Services, in November 2008.|
In the current DHHS [U.S. Department of Health and Human Services] guidelines, the preferred PIs have been:
So we have these four PI-based regimens and a variety of quite effective alternative regimens.
Then, from the NNRTI perspective, the preferred option has been efavirenz [EFV, Sustiva, Stocrin], primarily because, I think, of the overwhelming efficacy data and tolerability, with nevirapine [NVP, Viramune] as an alternative.
I think it's reasonable to understand that there are some differences between these regimens. A simple, obvious one for our patients is the pill burden.
Efavirenz can be part of a fixed-dose combination [i.e., Atripla] with one pill once a day, which is very appealing and has been shown to be highly efficacious, but it has unique neurologic side effects that patients need to be aware of. If the side effects occur, patients need to understand that they will usually get better. But if they don't improve, certainly a medication switch can occur.
And there are issues for women of childbearing potential, because efavirenz is a category D drug in pregnancy. So, in women who are pregnant, or women who might become pregnant, we need to have that in mind.
In addition to the unique side effects and issues surrounding pregnancy, the other unique factor with NNRTIs versus PIs is the risk of developing resistant virus if virologic failure should occur.
Certainly, with efavirenz, when people experience virologic failure, resistance is a likelihood -- albeit virologic failure is quite rare if they take their drugs consistently.
In contrast to efavirenz, many of the boosted PIs are dosed more than once a day. They range in pill count from three (for example, atazanavir/ritonavir) to four or five. So it's a little bit more in the way of pills. The side effects with PIs aren't neurologic; they are generally more gastrointestinal [GI]. All of these drugs can be associated with some dyslipidemia, as can the NNRTIs.
The trade-off, I think, is different toxicities and different pill counts. But the resistance part is also really compelling, in that when you use a boosted PI -- if things don't go right -- the likelihood of developing primary protease mutations is extraordinarily low.
To me, those are the major trade-offs between the first-line regimens we choose. I would share those with any given patient and, based upon that, allow them to be a part of the decision as to what they do next.
Certainly, in those patients where I'm particularly concerned about their ability to either adhere consistently with therapy or come back for follow-up visits consistently, I may be more inclined to push them towards a boosted PI because if they do disappear, or stop taking their meds for any one of a number of reasons, the likelihood of selecting for resistance is smaller. But again, that's part of the individualization.
The last drug that I think is worth mentioning, though it's not currently listed as a preferred option in the DHHS guidelines, but it has an FDA [U.S. Food and Drug Administration] indication now, is the first-in-class integrase inhibitor raltegravir [RAL, Isentress], which is given twice a day. So it's a nucleoside combination with one pill, twice a day of raltegravir.
We now have two years of follow-up from the pivotal, treatment-naive trial of raltegravir, showing that raltegravir was as efficacious as efavirenz and probably better tolerated.11
Certainly, from a lipid perspective, raltegravir seems to be associated with limited side effects -- at least with the amount of follow-up available. But like efavirenz, if virologic failure occurs -- which was quite rare -- resistance to the integrase inhibitor happens.
So it's a new drug in a new class that can be used as first-line treatment, and while it has an indication to be used in that way, one should recognize that there is somewhat less data for it than other approved, first-line antiretrovirals.
Those are the kinds of things that I walk through with patients when thinking about the variety of first-line options.
I always tell patients, though: The most important thing is that, whatever we do, we need to continue to interact. If they are having any problems taking the therapy for any reasons -- whether it's toxicity, dosing schedule or social factors like homelessness, disclosure issues, and substance or alcohol abuse -- they need to let me know so that we can make adjustments, as necessary.
Bonnie Goldman: Dr. Hawkins?
Trevor Hawkins: Eric has done a good job of reviewing the current choices. I think what really drives the choice these days is tolerability.
A couple of old ideas have gone by the wayside. One was that if you had a patient presenting with a low CD4 and a high viral load, you should use a boosted PI. We have seen that efavirenz has done as well as, or better than, any challenger, in patients with both low CD4 counts and with high viral load.12 So that idea that you can't use a non-nuke [NNRTI] in patients in those situations is no longer acceptable.
The second thing that we always used to do was keep a PI in your back pocket so you could sequence PI failure. As Eric said, study after study has shown that when patients who are taking their first boosted PIs experience virologic failure, they do not fail with primary PI mutations. And that, I think, has caused a huge shift.
Given all of this, and given that the current recommended regimens all work pretty well -- if patients will take them -- then the issues of tolerability and adherence really dominate choice.
Within the PI class, for instance, certainly in my practice, there's a shift towards using either atazanavir or darunavir as the preferred PI, partly because they are once a day.
Fosamprenavir and Kaletra can be, and lopinavir [LPV] can be, used once a day. However, they seem to be associated with high rates of GI intolerance. Diarrhea, for instance, in the CASTLE study was lower with atazanavir than with lopinavir.13 Similarly, in the ARTEMIS study, lower rates of diarrhea were seen with darunavir than with lopinavir, as well as lower triglyceride increases.14
These things are, in fact, what really drive choices these days. Because we know that the best regimen is the one that your patients will take. If it's more tolerable, if it's fewer pills, then they are likely to take it better.
Of course, there are one hundred and one other issues, such as mental illness and substance abuse. Eric had mentioned homelessness, lack of food, disclosure issues -- all these things that are really, really important in determining how well our patient's going to take any regimen. What we can do in terms of choosing the regimen is to choose the regimen that's likely to be best tolerated.
Eric Daar: I agree. It's sort of an interesting phenomenon: how things have changed; how we focused so much early on on efficacy. We still do; it's just, there are so many efficacious options that it is ultimately tolerability. The challenge for us is that we can't guarantee how well any one patient will tolerate a specific regimen.
For example, a fixed-dose combination with efavirenz is a very popular first-line regimen [i.e., Atripla]. We certainly use it frequently in our clinic. The overwhelming majority of patients, despite warnings about neurologic toxicities, tolerate it great. They experience no side effects at all, or the side effects are minimal and they go away, but for others, they don't.
Same thing with regimens like boosted atazanavir. It is an extremely well-tolerated regimen, but some people -- not very many, but some -- will develop hyperbilirubinemia and jaundice, and they will need to switch. So you just need to make sure that patients are prepared to work with you, regardless of what happens.