February 4, 2008
In this summary of her study, Connie Celum, M.D., of the University of Washington at Seattle, discusses surprisingly disappointing results from a study examining whether the use of acyclovir to treat herpes simplex virus type-2 infection can, in turn, help reduce a person's risk of HIV acquisition.
Connie Celum: This was a trial of 3,277 participants who were HIV negative, and HSV-2 seropositive, who were enrolled in a trial to ask the question: If you suppress genital herpes, using the standard dose of drugs that are commonly used (acyclovir 400-mg, twice a day), could we show that we could reduce HIV acquisition?1 We had 139 people who became infected in the trial, 75 of whom were in the acyclovir arm, 64 in the placebo arm. And the difference between the two arms was not statistically significant.
Connie Celum, M.D.
I think there were a couple of surprises that came out of the study. Even though, when you look at a trial like this and you look at its so-called integrity, in terms of did it not only meet its enrollment targets; it's really important in a study like this to also ask: Did people actually take the drug? And did we have any evidence that there was differential loss to follow by arm.
So, of the pills that were dispensed, adherence was very high: 94% of pills that were dispensed were taken. And so, at least at the present time, we can't say that adherence is a reason why it did not work. However, we will do some additional work to look at that.
There were a couple of surprises. I think that we went into the trial expecting from past studies that you might see up to an 80% reduction in genital ulcers. We saw a 37% reduction. And there were differences by region. For example, there was like a 50% reduction among gay men in the U.S., and about 32% among women in Africa.
So I think it also is important for us to consider biologic reasons why we did not see a bigger effect. And we will be exploring those. So was the drug metabolized in the way that we expected it to, from past studies? And also, just to make sure that the viruses, the herpes viruses, were susceptible to the acyclovir. So there is more work for us to be done. We just received these results very recently.
I think the question is, many people thought this was going to be a slam-dunk. If you really look at the epidemiologic data, it was very consistent, very compelling, showing a two- to three-fold higher risk of becoming HIV infected if someone had HSV-2 antibodies. So I think the question here is just: Is it that the strategy didn't work? That the standard dose of acyclovir, 400 b.i.d. twice daily, taken with high adherence, as best we can tell, based on pill count and self-report. Why didn't we have an effect on HIV at all? And why did we have a lesser effect than expected on genital ulcers? So those are really important questions.
I want to just finish by reminding people that there are two sides of the coin between the interactions between herpes and HIV, and this trial focused on a wealth of epidemiologic data and some biologic plausibility data that said herpes simplex type 2, the main cause of genital herpes, increased susceptibility to becoming HIV infected. There's also quite a substantial body of data that suggests that herpes makes people who are already HIV infected more infectious. And we have another trial ongoing, called Partners In Prevention, that is asking that question. It's a very challenging trial in 3,400 HIV discordant couples. So we do not know the answer to that question yet, and we should within a year. I just want to make it really clear that there are two different ways in which herpes may affect HIV transmission and acquisition. This study asked about acquisition. And within a year, we'll have data on HIV transmission, as well as whether herpes influences HIV disease progression. So stay tuned.
Reporter #1: Dr. Celum, did you do blood levels to see if they really took the medications, as opposed to took the pills?
Connie Celum: We have not. And part of the reasons for not having done that is that acyclovir has a short half-life of about four hours. And so, really, we would have a lot of imprecision in knowing much about adherence. But it would basically tell us: Did they take the last dose at the time they should have prior to this visit? Which is not, in our opinion, all that informative.
Reporter #2: Connie, do we know anything about host variability, genetic variability, in terms of both vulnerability to HSV and metabolism of acyclovir?
Connie Celum: My colleague, Anna Wald, is in the audience. And if you're interested in this particular area of genetics and HSV-2 expression, I think there's some data suggesting toll-like receptors are important, in terms of frequency of reactivation. But if you're asking about, do we know sort of pharmacogenetics of acyclovir; I think that is not, to my knowledge, an area that anyone has explored. Nor has there really been enough characterization of absorption in different populations.
What we do know, just from the data from prior PK studies, is that acyclovir is not very well absorbed. Orally, it's 15% to 20% absorption. And there is variability between persons. But in large studies of herpes suppression, even with that variability, you can achieve an 80% reduction in genital ulcers.
Reporter #3: Dr. Celum, on the study, there were different arms in Peru and Africa, and in the United States. Were your results consistent across all the arms? And how consistent were they?
Connie Celum: It's in the abstract. I just, for time, didn't go into it. But we looked at various aspects of homogeneity of the results, if you will, across various factors. And there really was no significant difference by gender or region, in terms of HIV acquisition. There was, as I showed in the slides, a difference in terms of incidence of genital ulcers and HSV detection, yes/no detection, as well as quantity by the different regions. So in the herpes outcomes, there were some differences by region and site.
This transcript has been lightly edited for clarity.