Serodiscordant couples: Africa’s largest HIV at-risk group

An Interview with Dr. Susan Allen

By Simon Noble, PhD*


Dr. Susan Allen 

Susan Allen has been a lead investigator of the HIV epidemic in Africa, particularly as a pioneer in the study of HIV transmission between couples, the primary way in which HIV is spread in Africa. She established Project San Francisco in Kigali, Rwanda in 1986, one of the earliest and now longest-standing HIV cohort studies in Africa. In 1994, as a result of the genocide in Rwanda, Allen relocated the central research clinic to Lusaka, Zambia (the Zambia-Emory HIV Research Project), which has recruited the largest single-site heterosexual HIV-discordant couples (in which one partner is HIV-infected and the other is not) cohort in the world. The unified sites now employ over 150 healthcare staff and are known together as the Rwanda Zambia HIV Research Group (RZHRG); they continue to enroll and follow HIV-discordant couples.

Allen, who has MD, MPH and Diploma of Tropical Medicine and Hygiene degrees, has just taken up a new position as Professor of International Health at Emory University’s Rollins School of Public Health. Her mission is driven by the recognition that the largest at-risk group for HIV in Africa—couples—constitutes a public health crisis. Not merely a research endeavor, the project sites operate voluntary counseling and testing (VCT) centers that test and counsel couples together, providing HIV/AIDS education, free HIV testing and counseling surrounding test results, and provide free outpatient reproductive healthcare and treatment of infectious diseases for eligible study participants. Such counseling has had quantifiable results: post-counseling HIV incidence rates have been 50-70% lower than those of non-counseled discordant couples. “The principal research objectives of the RZHRG are to uncover factors relating to HIV risk and transmission, to gain a better understanding of how to prevent the spread of HIV and to improve the quality of life among HIV-infected African adults” explains Allen. She recently spoke with IAVI Reporteditor Simon Noble about HIV/AIDS in Africa and the diverse activities at her project sites there.

You've pioneered recruiting cohorts of co-habiting couples. What makes such cohorts particularly valuable?

Well, I think it's by far the best way to identify suitable cohorts for vaccine trials because, at least in Africa, the largest group at risk is couples. What few people seem to remember is that 60 or 70% of the transmissions that happen everyday in Africa occur between spouses, and so lots of people focus on higher-risk groups, like sex workers or truck drivers, and while those are certainly worthy of study they're not the largest at-risk group that account for most of the transmissions. The biology of transmission is likely to differ by both route and inoculum size, so if you were going to develop a vaccine that would work against blood-borne transmission for intravenous drug users, it might have to have different characteristics to one that would protect against a mucosal challenge. And if you were developing a vaccine that was going to protect against a mucosal challenge when the mucosa is really impaired, like in a sex worker who has lots of different partners, lots of trauma and lots of inflammation, again that's a different scenario biologically than a husband and wife.

Once we find a [vaccine] product that works we are going to start vaccinating large groups of people, so the largest at-risk groups are the ones that we should study the vaccines in; we should be doing trials in discordant couples, [which are] cohabiting couples where one partner has HIV and the other doesn't.

How do you go about recruiting these couple cohorts?

The mechanics are that you have to go out and promote couples VCT (voluntary counseling and testing), which is catching on among funding agencies and governments but is not yet established as a norm in the population. There's this awkward two-legged stool; certain groups, like pregnant women, are being targeted for VCT for prevention of transmission to their babies, but most pregnant women are married and it would be so easy when you're testing them to include the husband. But the men are left out, and the capacities of a lot of ante-natal clinics don't really allow male participation unless they have weekend programs.

We've piloted a couple of programs but it requires a concerted effort and so far those worlds, the couples VCT and the prevention of transmission from mother to infant, haven't met. So the best way we've found to do it is to go out in the community and promote couples VCT and then provide the services to couples coming in. We have our own services—our facilities are stand-alone NGO-run—and so we don't over-burden government facilities.

What kind of new strategies can be employed to get more men enrolled in VCT programs?

Any time VCT is being offered by a healthcare provider, for whatever reason—whether it's a blood donor, a sick person seeking hospital care, a TB patient, a pregnant woman—the provider has to think who might be married, which is the case in Africa for most adults; certainly most women over the age of 22, 23 are married, most men by the time they're 29 to 30. It's just a question of having the people who provide those services automatically and reflexively think, "And your partner?"

Do you think that vaccine efficacy trials will be a possibility in couple cohorts? Is the power there to get the relevant numbers?

Oh yes, absolutely. In fact, I think they're the best cohort because the partners support one another. Even in discordant couples, the majority of them are committed to each other and they don’t separate. They do begin to use condoms after they learn their HIV status, certainly more than they did before, going from something like 0 to 90% use at blinding speed. But they don't use them perfectly so there continues to be transmission at a relatively high rate, about six to eight percent a year.

In the past when researchers first started looking at couple cohorts and the couples didn’t know they were discordant—and this is from cohorts that I wasn’t involved with—the transmission rate was about 20 or 25% a year. So you can see that just knowing has a huge impact. But you're left with this residual rate of six to eight percent, which is still high, so compared to other cohorts that could be used in vaccine trials, discordant couples are still definitely contenders. The other really key thing is that when a transmission event does happen in a vaccine trial, you have the donor whose virus you can study as well.

Depending on where you go, the proportion of the couples that you test that are discordant will differ. In Lusaka [Zambia], for example, our couples VCT centers test lots of couples and 20% of those will have one HIV-positive and one HIV-negative partner. In Kigali [Rwanda], the proportion is 10%.

Do you think discordant couples are a feasible way to specifically increase the number of women participating in trials?

Yes, that's actually what we're trying to do, both in Rwanda and Zambia, so that there’s a gender balance in trials, that there are half men, half women.

In Zambia and Rwanda specifically, what are the challenges to recruiting women to participate in trials?

The tricky part with women is that they can't be pregnant or breast-feeding to be in the trials. There's lots of willingness [to participate], and since our discordant couples include women who are HIV-negative and who have HIV-positive partners and vice versa, we can recruit from among discordant couples and have the vaccinee be either the man or the woman, depending on who the negative partner is. But the HIV-negative women have the added challenge of needing to be not pregnant or breastfeeding, so at intake you have to ask whether they would be willing to use a long-acting method of contraception, like IUD or hormonal methods, to avoid conceiving in the 12 to 18 months that they're in the trial. Of course, these would be supplementary to the barrier methods, male and female condoms, that they need to use to prevent HIV infection.

Is there any good evidence that hormonal contraceptives might have an effect on transmission susceptibility?

Not in our cohorts. In fact, if anything, it's protective for progression of disease. In the HIV-positive women in Rwanda that we have followed for 18 years now, use of injectable or oral hormonal contraceptives was associated with increased survival. And, among the HIV-negative women, use of those methods was not associated with increased acquisition [of HIV]. Now, I know different results have been found in sex workers but, again, the whole ecology of the genital tract in sex workers is completely different.

In some communities in Zambia, your work has established that marriage is a risk factor for women becoming infected with HIV. What can be done to reduce their susceptibility to infection?

The only way to do it at the moment is couples testing and counseling. If you can get it moved into the premarital realm, which is part of what we're working with the churches to do, that's the best way to do it, and premarital testing is actually becoming more of a social norm than a lot of people realize. But if it isn't done premaritally then it needs to be done amongst married couples.

Do you think there is sufficient awareness in some developed countries of the phenomenon of marriage as a risk factor?

No. I think a big part of what we're trying to do with our educational campaigns is to make people aware that just because you're married to someone doesn't mean you're protected. A lot of the prevention messages in the early days would say things like, “Stay faithful and you'll be okay,” but if the person you're faithful to has HIV and you don't then that's not going to help you. That's the big hurdle to overcome, in particular because women tend to have fewer partners outside of marriage than men. So if women believe, “Okay, if I'm faithful to my husband, I'll be okay,” that's often not the case.

What's been your experience in providing comprehensive family planning services in the context of HIV prevention, especially with the recent calls from some quarters for the promotion of the ‘ABC’ [Abstinence, Be faithful, Condomize] programs?

We've done a lot of research on combining family planning messages and HIV prevention messages in married couples, and I have to say that in married couples we set the whole ‘A’ aside, abstinence is not even on the radar in married couples. And no church would recommend it. But in terms of deciding whether to use barrier methods and/or other types of contraceptives, we promote the dual-method message—female or male condoms as a barrier method are good for the prevention of HIV and other STDs, but they're not great contraceptives. So in our counseling we recommend having added protection against pregnancy, adding IUD or Norplant or an effective long-acting method in addition to the barrier method that they use for prevention of HIV/STD.

Given that in many societies men have disproportionate power in sexual relations, what is the most effective way to increase condom use?

We find that the key is having the husband and wife, or cohabiting partners, however you define the marriage, do the pre- and post-HIV test counseling together. If they get their test results in the room together and they make a plan based on the combination of their test results, then condom use is optimized. We've measured that in all kinds of objective and subjective ways.

What have been the specific concerns of community advisory boards (CABs) with regard to your couple cohorts? Are there ethical dilemmas specific to couple cohorts and their counseling? Particularly in discordant couples, I would imagine.

Our CABs have always been really supportive, I think because they realize that once you have a couple that's been tested and counseled, and they know that one has HIV and the other doesn't, you've done the very best you can in terms of counseling and support, promoting condom use, providing ongoing follow-up and so forth. And, you know, people are human, they don't always do what's best for them all the time, so it's understood that whether it’s because they might want to have a baby or some other reason, they might have sex without a condom or might not use a condom correctly occasionally. That's certainly the case for most discordant couples, there continues to be risk.

So the CABs understand that while you try to do the very best you can with counseling and you try to continually improve your counseling, you're never going to have complete protection with that approach. It's a little like seat belts—you can legislate seat belts, and 80% of people will use them just to avoid getting a ticket, but 20% still won't, even though they know that it's life-saving and they know the accident statistics. You can only go so far with trying to get people to change their behavior. So CABs are very supportive of biomedical interventions, at least in our experience, once the rationale is understood.

Are there any special challenges to counseling discordant couples? How does it differ from your regular VCT counseling?

Well, you have three scenarios when you're giving results to couples. The first and happiest one is they're both negative, and that's obviously the easiest one because then you're trying to reinforce behaviors to keep them both negative, and there's a huge sense of relief and a renewed commitment to the relationship and faithfulness. Then you have a situation where both partners are positive, and that presents a challenge because they have to think about the kids they have and the fact that they might not live to raise those kids, and planning for their own health and the family's future and all kinds of things. But obviously there’s less focus on transmission from one to the other because they both already have it.

With discordant couples, of course it’s more complex, you have to support the HIV-positive person as an individual with HIV and their health issues and concerns, whether they're symptomatic or, more usually, asymptomatic. And then there's the issue of protecting the non-infected partner from acquiring HIV, they’re more vulnerable because they're in that marriage. And if you have a discordant couple where the positive partner is the woman, then you have to deal with the added issue of, should she become pregnant, her child might acquire HIV.

We've taken an almost modular approach to counseling, there’s the basic VCT and then you add things to it. If the woman's pregnant, that's another added message. If they're coming in together as a couple, that's another added message. It becomes a very complex business in the end.

What kind of guidelines are in place to dictate how much information should be released about one partner's status to the other? For instance, one partner becomes HIV-infected, are you obliged to inform the other?

In our couples VCT centers, we have the couples sign joint informed consent, and they see that informed consent by video. So for those that struggle to read, all the information is presented visually. So all the couples that come through our center sign saying they want to be tested and counseled together, they want to receive their results together, so we don't have a problem of disclosure between partners. But we do tend to advise people to be very careful before they choose to share the results outside the couple because they have to judge their own family members and how supportive they'll be.

When a seroconversion event happens—like in the course of our prospective studies, if we have discordant couples and at some point the originally negative partner becomes positive—at that point we counsel them about the fact that their serostatus has changed and what that all means.

It strikes me, speaking to you now, that these marriages sound very strong. Is that something that's struck you?

Yes, I think a lot of things go into marriages, and by the time people come through our doors, they've been cohabiting an average of six years. They almost all have children, some sort of ceremony has been held, whether it's a civil ceremony or traditional one, there's been a bride price exchanged. A whole community has been involved in that union, and so to separate, it's not done lightly. I would say Zambian and Rwandan urban couples tend to have the most erosion, if you will, of the traditional values, so in more rural areas I think you would find even more of that. So yes, I think married couples tend to stick it out, the whole for-better-or-worse thing. I have been struck, actually, by how committed husbands and wives are to each other in the face of this in Africa.

So I understand that you're currently involved with IAVI in setting up the adeno-associated virus AIDS vaccine candidate Phase I trial in Zambia and South Africa.

That's what we're hoping to do, we just had a consensus conference in Zambia which went well, and we're now going to submit the protocols for formal review to the ethics committee and the other regulatory committee which they call the Pharmacy and Poisons Board in Zambia. These trials follow on from the trials in Europe, in Belgium and Germany, and the proposed trials will test the immunogenicity and the optimal dosing schedule of the adeno-associated vaccine candidate.

Could you talk about Project San Francisco, and how AIDS vaccines are being brought into that program?

Project San Francisco is our program in Kigali, it’s now 18 years old. We started out with pregnant women, like a lot of people do, and they said to us, "Would you test our husbands?" and it just sort of evolved—that interested us in couples and we realized that discordant couples were really the main risk group and should be our focus of interest. Then we had to temporarily leave Rwanda in '94 because of the genocide, and we set up an analogous project in Lusaka, which is now 10 years old, called the Zambia-Emory HIV Research Project, ZEHRP. The two projects together are under an umbrella that we call the Rwanda/Zambia HIV Research Group, RZHRG. We are staunch in our opposition to acronyms that mean anything (laughs). We’re now working to set up AIDS vaccine trials under that umbrella group; the common theme between the two countries is the couple cohorts. The vaccine products that will be tested will probably be different because in Rwanda it's predominantly HIV clade A that’s circulating, whereas in Zambia it's clade C, but the methodology and the risk groups will be the same.

What other types of studies are coming out of these cohorts?

Our two biggest focuses have always been prevention of transmission and the study of the natural history of disease; once somebody is infected, what happens? Eric Hunter, my husband, is studying the genetics of the virus and looking at the characteristics of the virus that is transmitted, or at least the virus that establishes infection, because these viruses are likely to be very important from a vaccine perspective. That work was published in Science a couple of months ago. Dick Kaslow and James Tang, who are immunogeneticists, have been looking at the genetics of the host. They've published a great deal on the genetic markers associated with living a long time once you have HIV, versus succumbing quickly to the infection, developing disease and dying. Just recently they had a paper in The Lancet looking at host genetics in HIV transmission in our Zambia discordant couples. The bottom line there is that the more similar husband and wife were with respect to the genetics that code for immune responses, the more likely it was that virus was transmitted from one to the other because being genetically similar means you have similar immune responses, at least with respect to the aspects that they studied.

We've also just done some behavioral studies this summer with a student from Emory, Joyce Au, where we're providing ARVs (antiretrovirals) to our sick patients in Rwanda with Global Fund support. We’ve just completed a survey of the partners in discordant couples who are sick and getting ARVs. Actually, that reminds me—in every case, their drug buddy was their spouse, who was helping them remember to take the drugs and so forth. We asked them, “So how has this affected your condom usage? Do you feel that you still need condoms? Are you okay now and you don't need them?” And 100% of the people that we interviewed said, “No, no, condoms are more important than ever because the last thing I'd want to do is transmit a resistant virus to my spouse.” So they’re extremely sophisticated in their thinking about this.

What do you consider should be the key goals for HIV/AIDS prevention in the next couple of years, and perhaps in the more medium term?

Well, applying behavioral strategies that we know work; in sub-Saharan Africa, couples testing and counseling. I think that should be a number-one promotional strategy, and CDC has adopted it as a centerpiece, even though they don't have much practical experience with it they have agreed that it's important. And to get more people on the bandwagon, that the largest risk group on the planet is cohabiting couples in sub-Saharan Africa, and the only thing that is known to work with them is couples testing. So that's number one.

And then the next step is vaccines, which I see as the most feasible biomedical prevention/intervention on the horizon, despite the obstacles.

What kinds of things are you working on now and what do you think are the most exciting areas of HIV/AIDS prevention?

Right now, my grants are behavioral in nature; promoting couples VCT, combining VCT messages or prevention messages with family-planning messages. Our cohorts act as a source of samples for studies like Eric's looking at the virus and Dick Kaslow's looking at the genetics of the host, and other investigators who are interested in immune responses. And then we're gearing up for vaccines, which is, I'd say, the most exciting thing we're doing right now.

*Simon Noble is editor of the IAVI Report