September 13, 2009
Listen to Audio (15 min.)
Please note: These files can be quite large. Allow some time for them to download.
There's nothing like hearing the results of studies directly from those who actually conducted the research. In this interview, you'll meet one of these impressive HIV researchers and read her explanation of a study she presented at the 49th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2009).
I'm Kimberly Smith from Rush University Medical Center in Chicago. I am at this year's ICAAC presenting an analysis from the GRACE (Gender, Race And Clinical Experience) Study. The data that I'm presenting is outcomes by race at week 48.1
Kimberly Smith, M.D., M.P.H.
Just to remind people a little bit about what GRACE was: GRACE was a multi-centered, open-label, phase 3 study that was done primarily in the U.S., but also there were a few sites in Canada and Puerto Rico. It was a study that looked at treatment-experienced patients, but the thing that was unique about this study was that it targeted women and people of color. If you look at the demographics of who was enrolled, it was about 70% people of color in this trial. What I'm presenting here today is a race analysis.
What we show here in this analysis was that black patients did not respond as well as Caucasian or Hispanic patients. That was particularly true when you look at black women. The response rate for black women was about 46%, and the response rates for Caucasian women was more like 60% and for Caucasian men, again, more like 60%.
Could you define "response?"
Viral load undetectable at a level of less than 50.
How does that compare to other populations?
The thing that's unique about the GRACE Study is that we don't have a good comparator, because this is the first treatment-experienced study to have a large proportion of people of color in it to be able to answer this question. We have lots of treatment-naive studies. The HEAT study, which I presented at IAS [International AIDS Society], had a good proportion of people of color.2 We showed that that difference existed there.
This is the first treatment-experienced study to actually show that difference because it was the first one to enroll a large number of patients.
So the response rates weren't as good for black patients. That was true whether you looked at the whole TLOVR [time to loss of virologic response] analysis or even if you looked at excluding people who discontinued for reasons other than virologic failure. If you took out people who discontinued and all that, the difference was still present.
But there were a couple of things that stood out. As I mentioned already, black women didn't respond as well. People keep asking me why I think that that is. I really think that there are a lot of factors that we can't necessarily measure.
I think that patients in this population and HIV-infected patients in the United States that are black are more likely to be poor and are more likely to have other competing challenges and priorities. They may be still challenged by issues of substance use. There's just a long list of things that you can't necessarily get at in clinical trial.
What this showed is that even in a situation where folks have access to medical care and access to drugs -- because all the drugs were provided in this study -- unless we deal with some of the other challenges that exist for people that make it hard for them to continue in a study, we can't get their response rates up to what is the case in the general population.
To me, I think the take-home message is that we need to start to design studies that identify individuals that may have challenges and intervene for those individuals before they end up with virologic failure, so that we can get all the populations up to what I think is the standard.
The standard for treatment-naive trials is to get 80%, 85% of the population undetectable. If we can raise that bar for black patients, then we raise the bar for everybody else. I think that that should be our new challenge -- i.e., try to get everybody up to a standard as opposed to drug A versus drug B. It should be about getting everybody to what I think now is the standard of care.
Do you think that other studies have missed discovering the bad response rates in this population because they never enrolled them in the studies and this issue was there all along?
In treatment-experienced studies, you rarely see a proportion of African-American patients that's more than 10% or 12%. Some of it is that the African-American populations enter into HIV treatment later, so there are fewer of them who have multi-drug resistance available to recruit into some of the treatment-experienced trials. So that may have been some of the driving force for that.
For the treatment-naive trials, I do think that the drug companies are now being challenged to enroll a more diverse population. When they're doing that, then we're able to have enough patients in the study to be able to see those differences.
But this is not a new finding. The reality is that we've had at least three studies in the ACTG [AIDS Clinical Trials Group] that have shown that blacks didn't respond as well as whites -- and these were all naive studies.3,4 The HEAT study I just presented, again, showed the same thing.2
What's unique about those studies is they did have 30% or 35%, 40% people of color, so you had enough people to be able to do the comparison. When you do that, then you're able to see what differences exist.
I think now we don't need more studies that show that the differences exist. We need studies that actually seek to get rid of the difference. How do we get rid of the difference? We know that it's there. How do we intervene?
Are there no genetic issues that are responsible for the lower response rates? Is it all life issues?
In this particular study, there's nothing that suggests that it's a drug-driven issue. The same thing was true in the HEAT study. The HEAT study looked at two different regimens. The response rate was lower for blacks with both of those regimens.
In this study, the only drug that they got that was across the board in all the patients was darunavir. The other studies of darunavir haven't shown a racial difference in response rate. I don't think that it's necessarily a drug-driven thing or toxicity-driven thing. When we look at the toxicities by race, we don't see any differences.
I think that this is primarily due to drop out, which clearly was shown to be higher for black patients, and lower adherence, which we did show. There was about a 10% difference in the proportion of individuals who had greater than 95% adherence when you compared blacks to Caucasians. If you take those two things -- adherence and drop out -- those are primarily the things that I think are leading to the difference. We have to address why people drop out and why there is lower adherence. We need to have strategies to try to fix those things.
Wasn't this trial unique in that there was greater support for adherence?
It was unique in the sense that it really went out of its way to try to attract women and had supports in place for women. There was a lot of advertisement, a lot of support. People literally got birthday cards sent to them on their birthday. They made a point of having women-centered advertising.
Everybody focused on trying to make sure that their needs were met. I think some sites were more successful at doing that than other sites. That's how we did a good job of enrolling this study. There's no question. It enrolled very rapidly. Industry has in general been nervous about enrolling studies with a large proportion of people of color, large proportion of women, because they thought they would never fill and it takes too long. This study enrolled very rapidly. The reason is that they made those extra efforts to try to attract women and make them feel like this was a study designed for them.
Isn't one of the dangers of that that there's some bias in the study? The researchers were trying to enroll people who wouldn't normally be good candidates for a trial because they might not be adherent or might have other competing issues. Could there be a bias in the results?
Sure, you could argue that. At some institutions, folks assume that someone's not going to be good for a study on the basis of them being a woman that's a single mother with kids, or being an African-American patient with a history of drug use. People make assumptions.
But when you look at the studies that I've looked at, whether or not those are predictors of adherence, they're not. We don't do a good job of predicting ahead of time who's going to adhere. I do think that studies that try to enroll a diverse population -- that represent what our clinics look like -- really are going to be more accurate at telling us what our challenges are in getting people to undetectable. If we only enroll what we believe to be the ideal candidates for clinical trials, then sure, we'll know what the response rates are in those folks, but they won't at all be reflective of what's going on in real life.
Isn't that what's been going on, the recruitment of ideal candidates?
I think that that has happened in some settings. If you enroll somebody in a study, you want them to stay. Maybe this study will have a negative effect in that, because we showed that African Americans were more likely to discontinue, folks won't want to enroll African Americans in their study. I hope that that's not the case. I hope that what happens is that we recognize that we need to put some supports in place to try to make sure that patients don't discontinue. Maybe if they can't make their visit, we find a way to go get them and bring them for their visit.
I hope that what we'll learn from this is not that we don't want to enroll people of color into these trials because they bring down our response rates. I hope what we'll learn is that we need to do some of the extra things in order to try to bring the response rates of everybody up. I keep saying to folks, if we bring up the response rates for the toughest-to-treat folks, then ultimately that brings up the response rates for everybody.
Are people convinced?
[Laughs.] Is this a real uphill battle then?
It is. I mean it's a challenge. Myself and other colleagues have for a long time been pushing to get more studies like this done. Now, I think the next push is that we design studies that intervene for folks who don't do as well. If I see that someone doesn't reach a certain threshold early on (e.g., at week 2 or week 4, they haven't had the viral load response that I'd expect by that time), then I should do something. I can call them up or have a nurse or a peer or someone call them up and say, "I don't think you're taking all your medicines. Tell me why. Is it that you're having a side effect? Is it that you got a copay and you couldn't get your medicines until a week late? What is it?" Find out what it is and then try to fix the problem, rather than do what we traditionally do in studies, which is enroll somebody and then look at them at week 16 to see if they've responded like we want. If they haven't, they're a failure. If they have, then they're a success. That's not good enough anymore.
Do you think there's money for this kind of support?
I was talking to Jules Levin. I've talked to folks from the NIH [U.S. National Institutes of Health]. I've talked to all of the people at all of the industries and said, "We have to invest in successful responses. It's not just about drug A versus drug B. It's about everybody responding in a similar way. How do we get the folks with the most challenges up to the standard? If we can do that, that raises the bar for everybody."
Right. Don't you think that this study is providing a window into the lives of HIV-positive, African-American women in the U.S.? Women who, for instance, are not doing well or are dying in the South. We're really not reporting on that.
If you look at any of the data on who's dying from HIV nowadays, there's no question that blacks are more likely to die, black women in particular. They're more likely to be diagnosed late. That's shown in this study as well.
Black folks were likely to have more advanced disease than the other individuals in this study. Yes, I think this is a microcosm of what's going on in the larger society. What we know is that if you have virologic failure, you're more likely to go on to have a higher rate of mortality. If we can't get people suppressed, we can't keep them in studies. In many cases what that means is that we can't keep them in care. If we can't keep them in care, they're not going to do well.
You can ask any provider and they will tell you that nowadays we don't see that many deaths, but the deaths that we do see are in people that we have a hard time keeping in care. We have a hard time keeping them on their medicines.
It's very unusual to see people die with an undetectable viral load these days. It happens, but it's rare. More often than not, it's people who can't stay on a study or can't stay in the clinic. They can't take their medicines consistently. Either they're pulled away by issues of substance use or other challenges that exist in their lives, but those are the folks that are dying.
Do we need another study to find out what are the particular issues that these women are struggling with?
I don't think we need another -- we've got studies. From the cohort studies -- like the WIHS [Women's Interagency HIV Study] analyses -- you can see the challenges that exist in these people's lives. One of the WIHS studies showed that the average income for the women in that cohort was $4,500 a year. Forty-five hundred dollars a year! That, by itself, is a huge challenge. That means that the priorities in your life are not just about getting your pills swallowed. It's about making sure you have food on the table, making sure you got a place to live, taking care of your kids.
There are so many other things. There is much more depression seen in HIV-infected women compared to those who are not infected. All of those issues I think are ones that we've documented in a number of trials. Now we have to decide how we intervene to try to get past those issues.
OK. Thanks very much.
This transcript has been lightly edited for clarity.
|Please note: Knowledge about HIV changes rapidly. Note the date of this summary's publication, and before treating patients or employing any therapies described in these materials, verify all information independently. If you are a patient, please consult a doctor or other medical professional before acting on any of the information presented in this summary. For a complete listing of our most recent conference coverage, click here.|
The content on this page is free of advertiser influence and was produced by our editorial team. See our content and advertising policies.