February 9, 2009
My name is Salim Abdool Karim. I am Pro Vice Chancellor at the University of KwaZulu-Natal in Durban, South Africa, and Professor of Epidemiology at Columbia University in New York. I am speaking today about the results of the HPTN 035 microbicide trial.
This trial involved 3,099 women in five countries, four of which are in Africa. The fifth site was in the United States, in Philadelphia [Pennsylvania]. Women were enrolled and randomized to one of four arms: They received either PRO 2000 gel, BufferGel, placebo gel or no gel. The placebo and the no-gel arms are the control arms. The two active products were being tested for their efficacy against HIV infection.
Salim Abdool Karim, M.D., Ph.D.
In the study, we followed women and provided them with the study gel, and with condoms, at monthly visits. We also tested them for HIV infection regularly.
In terms of the study outcomes: We found that the incidence of HIV infection was 3.9 per 100 woman-years in the placebo gel arm. It was 4 per 100 woman-years in the no-gel arm. However, most interesting result was the incidence of HIV infection in the PRO 2000 Gel arm of 2.7 per 100 woman-years. This translates to a 30% reduction in HIV infection in those women who used PRO 2000 gel compared to placebo gel.
When we analyze these data in greater detail, what we find is that in those women who used more of the gel, they had a larger protective effect, going as high as 44%. And in those women who used very few condoms and lots of gel, the effect is even larger, going up to 78%. These provide corroborative evidence to suggest that PRO 2000 may be a promising microbicide.
The BufferGel showed no effect on HIV infection, contraception or any of the sexually transmitted infections we studied. PRO 2000 was not contraceptive and had no effect on sexually transmitted infections [other than HIV].
Are these results the most successful of any microbicide so far?
Absolutely. I did my first microbicide trial in 1994, on what was then known as nonoxynol-9 foam. Fifteen years later, we have the first promising result in a microbicide trial. It's been a long road to get to this point. For me, it's very promising, very exciting, because we have had so many studies that have shown harm, or failed to show any benefit, or have had to stop early for a range of reasons. So, yes: This is the most promising of all the microbicide studies that we have to date.
But to my ears, 30% reduction does not sound very high.
Absolutely. Thirty percent reduction within our trial is not statistically significant. In other words, there's a slightly higher than 5% chance that, in fact, there's no effect. That's how we look at data of this nature within a trial.
If we just look at the 30% effect, it's not very compelling. However, we have to look at: 30% compared to what? In reality, if we take a situation where a woman is married, is trying to have children; and if we think about what the HIV prevention options are for that woman, who is already faithful and wants to have this man's children, is HIV negative, and is scared of getting HIV; well, the answer is, we have nothing to offer that woman. In that group of women, we have zero protection to offer her. She can't use a condom. So, a microbicide with 30% protection in that situation is better than nothing. Although our study is not conclusive, it provides a sense that there's potential here.
There is another study that is ongoing on PRO 2000, being done by the British Medical Research Council. If that study shows this product is beneficial, then we can say that there is potential benefit in a group for which we have nothing else to offer.
Were there any toxicities?
We looked at toxicities in great detail: There was a Phase 2 component to the study, in which we looked at systemic toxicities in great detail. Then in the Phase III study, we similarly measured a whole range of adverse events. What we found, in short, is that there were absolutely no concerns raised in the genital tract, where we would want to look at a range of things like bleeding, changes in vaginal discharge and so on. We had no adverse events that were higher in the active arms compared to the placebo.
As far as systemic toxicities were concerned: similarly, no differences across the arms. We also looked at social harms: similarly, no differences. So what we have, in effect, is a product that is very safe.
The consistency of the gel: Is it like K-Y Jelly?
Absolutely; you have hit it on the head. It's actually modeled on K-Y Jelly. The placebo is actually a variant of K-Y Jelly, and the gels themselves are designed to be very similar to the placebo gel. So they all look like K-Y Jelly -- they are all clear, colorless. They have very good viscosity, so they are excellent lubricants. You can't make out the difference when you look at all the gels in the study.
So they are designed to be used with a condom?
They are designed to be used with condoms or without condoms; it doesn't matter whether you're using a condom with it, or not. It doesn't affect condoms. It doesn't degrade them or impact on them in any way whatsoever.
It just protects the women more if a condom is used?
Actually, our problem is, our study can't address that question. Our study can't address what the additive benefit is of PRO 2000 compared to condoms. What we do know is that we can compare women who have used condoms and PRO 2000 versus [women who have used] condoms and placebo. What we show, in that instance, is that there's a protective benefit of using PRO 2000. But it's not PRO 2000 and condoms versus [no PRO 2000 and] condoms; it's PRO 2000 and condoms versus placebo gel and condoms.
How much of the substance must a woman use to have an effect?
In our study, we advised women to use a single dose of 3.5 mL PRO 2000 gel up to an hour before she has sex.
How is it inserted?
It's in a single-use applicator. It's got a barrel, and it's pre-filled with the requisite volume. It's got a plunger that you insert, and then the applicator itself actually has a smooth, rounded edge, so a woman can insert it into the vagina without any trauma. It's actually quite a simple, painless procedure.
Once the barrel is inserted into the vagina, they push the plunger down, and that ejects the gel out of the barrel into the vagina.
Is there any way to ensure coverage? How does a woman make sure she's gotten it everywhere within the vagina?
The beauty of the vagina is that it does that for you. Because the vagina is actually a virtual space -- in other words, it's not real space -- what happens is that, in our studies of magnetic resonance imaging [MRI], we've looked at the spread of the gel. Most of the gel goes and sits in the posterior fornix. That's just on the outer side of the cervix. Then it tracks so that it is actually present all along the vagina. So when you're doing MRI scanning, there's a thin foam of gel all along the vagina. It self-spreads: You just walk about a bit, and it spreads itself in the vagina.
Is there a way to make it more effective?
I don't know if there's a way. Part of doing the analyses, now that the study is over, is to see what more can we learn, to do things better next time round. We're talking about perhaps formulations of the gel which might have a different effect, or maybe combinations of this product with something else. To do better than the 30%, we would have to look at different strategies, as well.
How long before intercourse must a woman use it?
We advise women to use a single applicator up to one hour before sex, and it's just one dose. There's a one-to-one correlation: You have sex once, you use the gel once. If you have sex twice in the night, you use it twice.
How much would this cost?
We have no idea; it's never been made to commercial scale. As it stands right now, to make this chemical, it's a pretty straightforward process. There's no real expensive, complicated process to make it.
Is it being directed at the developing world? I think in the developed world, 30% protection against HIV would not be considered very impressive.
I think you've touched on an issue: There might be a different benchmark for something that would be considered useful, for example, in the United States, than in South Africa. We designed the study to be able to answer the question: Is this product good for the whole world? That's why we did the study not only in Africa; we also did the study in the U.S., because we want this product to be available in the U.S. If it's good enough for African women, it's good enough for women in the U.S.
However, that's a more complicated regulatory issue. We would hope it would be licensable. We don't have enough data from our study, but with additional studies, this product could hold much potential.
How does this study compare with other microbicide studies that are ongoing?
At this point, there are only two microbicide studies underway; efficacy studies against HIV infection. One is the British study, which is studying the same product, half of the same PRO 2000 gel. It's a study in 9,000 women, and it's being done in several countries in Africa. It's looking at whether PRO 2000 is able to prevent HIV infection. Its results will become available towards the latter part of this year.
Another study is under way of an antiretroviral microbicide using a drug called tenofovir [TDF, Viread] in a gel. That study involves 900 women and is being done in South Africa. We expect that trial to report its results in the middle of next year.
I think, from what I've seen so far, antiretroviral use is a little bit more successful at preventing HIV as a microbicide.
Well, at this point, only in animals. There were no human studies at all. The CAPRISA 004 study of tenofovir gel will be the first microbicide gel with an antiretroviral that will report a result [in humans]. Until then, we have no idea.
In animal models, they are amazingly protective. We have not seen those kinds of data for protection, either from vaccines, or other microbicidal products. Certainly, in animal data, it's promising.
But there are problems, maybe, because of resistance issues.
Lots of issues.
And toxicity issues.
Oh, I can't even start to tell you the complexity. Giving an antiretroviral drug to a healthy woman is not something you can do simply or easily. It's something that needs careful consideration.
So what are the next steps for PRO 2000?
For us, the next steps are engaging with the company, Indevus, that owns this product, to look at how they want to move forward with the development plan for this product, and to await the results of the British MRC trial. That is really where we see this going.
Thank you very much.
It's a pleasure. Good luck.