Understanding Risk: A Conversation
Editor Bob Leahy interviews CATIE’s James Wilton about the tricky topic of communicating risk in the age of undetectable viral load
Bob Leahy: James, you gave a presentation at the Gay Men’s Sexual Health Summit in Toronto last week on Understanding and Communicating Risk: Viral Load and HIV Transmission. That’s a topic that fascinates us here, and one we’ve been following on PositiveLite.com for some time. Clearly it’s important for people living with HIV to have the best possible understanding of this too. With this in mind, thank you for agreeing to talk to PositiveLite.com and helping us understand more.
I guess we should start with the basics. Tell us what an undetectable viral load means in plain language.
James: Undetectable basically means that the amount of virus (also known as the viral load) in a body fluid is below the limit that our viral load tests can detect. Viral load is regularly measured in the blood to monitor how well treatment is working for someone living with HIV. Generally, successful antiretroviral treatment can reduce the blood viral load to undetectable levels within a few months of starting. In Canada, an undetectable blood viral load normally means that there are less than 40 copies of the virus per ml of blood. Tests to detect the amount of virus in other body fluids such as semen, vaginal fluid, and rectal fluid, are not available to people living with HIV but have been developed for research purposes.
Bob: So a person with a lower viral load is likely less infectious than one whose viral load isn’t under control?
James: A lot of research shows that a lower viral load in the blood is generally associated with a lower risk of sexual HIV transmission. Although blood isn’t a fluid that’s often involved in the sexual transmission of HIV, the viral load in the blood is generally correlated with the viral load in the fluids that are, such as semen, vaginal fluid, and rectal fluid. In other words, if the viral load is controlled in the blood, it’s also generally controlled in those other body fluids. However, this isn’t always the case and some people living with HIV can have detectable amounts of virus in the genital and rectal fluids even though the viral load is undetectable in the blood. This is more common if someone has a sexually transmitted infection (STI) but can also happen when there isn’t an STI.
It’s important to note that pretty much ALL the research that has been done to date around viral load and HIV transmission has been among heterosexual couples.
Bob: I guess the big question is HOW MUCH less infectious and how we communicate that risk so that people can make decisions appropriate to their own situation? There is research from last year that made international headlines - HPTN 052 – with its conclusions that in the right circumstances, the chance of transmission in sero-discordant (heterosexual) couples was reduced by 96%. Can you comment on how important was that study in trying to understand our own risk?
James: There are two different pieces of information people living with HIV normally want to know with regards to antiretroviral treatment, viral load, and the risk of HIV transmission. The first is HOW MUCH treatment can reduce their risk of transmission, also known as the relative-risk reduction. The second is HOW LOW that risk is reduced to when they are on treatment, also known as the absolute risk of transmission. People living with HIV are often most interested in the latter; their absolute risk of transmitting HIV when they are on treatment and have an undetectable viral load.
Unfortunately, biomedical HIV prevention trials such as the HPTN 052 study are not designed to provide information on an individual’s absolute risk of HIV transmission. These trials tell us about the change in risk of HIV transmission in a population using an intervention relative to a “control” population not using the intervention, in other words the relative risk-reduction. The relative risk-reduction is important to know because it tells us how effective a strategy is at reducing the risk of HIV transmission and can be used to compare the effectiveness of one strategy to another. The 96% relative risk-reduction calculated in the HPTN052 study tells us that antiretroviral treatment is highly effective at reducing the risk of HIV transmission among heterosexual couples who are mostly having vaginal sex.
However, the relative risk-reduction is not something that an individual can use to easily assess their absolute risk of HIV transmission. It’s really difficult to quantify someone’s absolute risk of HIV transmission while on treatment because it depends on a number of different factors unique to an individual such as how often they are having sex, how often they are using condoms, how well they are adhering to treatment, if they have any STIs, and the type of sex they are having. Therefore someone’s absolute risk of HIV transmission while on treatment may be higher or lower than another individual on treatment. We really need studies that try to calculate the absolute risk of HIV transmission from a single exposure to HIV through different types of sex (when the viral load is undetectable).
In the absence of additional information on absolute risk and in the face of these uncertainties, there are still messages that we can give to individuals so they can make informed choices and keep their risk of HIV transmission as low as possible while using “treatment as prevention.” This includes using condoms correctly and as often as possible, adhering to meds, regular viral load testing and regular STI testing and treatment for STIs.
Bob: Of course that leaves gay men a little in the dark doesn’t it, because that HPTN 052 data doesn’t necessarily reflect the realities of the risk associated with anal sex, for instance?
James: There is a much larger research gap when it comes to gay men.
We really don’t know if the relative risk-reduction while on treatment will be the same for gay men as for heterosexual couples. However, researchers think that it could be similar.
Even if the relative risk-reduction is the same for gay men, the absolute risk of HIV transmission while on treatment may be higher for gay men (who are having anal sex) than for heterosexual couples (who are mostly having vaginal sex). We know that bottoming without a condom (unprotective receptive anal sex) is up to 20 times more likely to lead to HIV transmission than unprotected receptive vaginal sex. Therefore, the higher initial risk associated with anal sex may mean that the absolute risk of HIV transmission when undetectable is much higher for anal sex than for vaginal sex.
The 96% relative-risk reduction from being on treatment is equivalent to approximately a 20-times reduced risk of HIV transmission. Furthermore, when not on treatment, we know that the risk of HIV transmission through bottoming is up to 20-times higher than vaginal sex. Therefore, if being on treatment reduces the risk of HIV transmission through bottoming by 20-times, the absolute risk of HIV transmission after this reduction in risk may still be in the same range as vaginal sex when not on treatment.
This is all hypothetical and really emphasizes the need for more research
Bob: So the message here is that we need more research in to the impact of undetectable viral load on MSM, right? Is anything going on?
James: Yes, we need more research. I know of some that’s going on in Australia and the Netherlands, hopefully we will see some results at the next International AIDS Conference in Washington this summer.
Bob: Let’s talk about risk guidelines for a moment. Tell us how risk factors – percentages like 96% - are ultimately translated in to low-high risk language. What degree of certainty needs to be in place before they are formulated in this way?
James: There is no guide for translating risk-reduction percentages (relative risk reduction) into low-high risk language (absolute risks). The CAS Transmission Guidelines do not use risk-reduction percentages to determine which activities or behaviors should be placed into “high” or “low” risk categories. The CAS Transmission Guidelines were developed when our knowledge of HIV was much more limited. At the time the guidelines were developed, we knew that unprotected vaginal/anal sex was significantly more risky than oral sex and that condoms could significantly reduce that risk. It was this knowledge that formed the basis of these guidelines.
In the past decade there has been a significant amount of research emerging around the biology of HIV transmission and new HIV prevention technologies. It’s only recently that we have had to deal with these relative risk-reduction percentages and we really haven’t figured out the best way to incorporate all this information into our discussion of risk. It’s difficult because the use of “treatment as prevention” and other new prevention options have a number of caveats and uncertainties and there are still large gaps in the research.
Bob: So in the case of risk guidelines which include reference to undetectable viral load, we don’t yet have that degree of certainty? What about for heterosexual couples? Isn’t the data strong enough there for risk guidelines to be in place, based on HPTN 052, do you think?
James: There is strong evidence that being on treatment and having an undetectable viral load significantly reduces the risk of HIV transmission for heterosexual couples. However, simply saying that the risk is “low” doesn’t reflect some of the caveats and uncertainties of this approach or the research gaps that exist, particularly among gay men.
We definitely need the latest science to be incorporated into guidelines so people are getting accurate information on “treatment as prevention” and this information is accompanied by the appropriate messages to keep this risk as low as possible. This information is already starting to show up in different guidelines, including treatment guidelines which are suggesting that physicians discuss the role of treatment as prevention with patients. Mostly people are considering undetectable viral load as an additional strategy for HIV prevention, along with regular condom use. However, the HPTN 052 study was only released last year and there is still a lot we don’t know. We are still trying to figure out how to communicate this information and incorporate it into different guidelines.
Bob: You’re probably aware that all this is a bit frustrating for some poz folks. Community leader and POZ magazine founder Sean Strub, for instance questions (Five Things about HIV They’re not Telling You) the risk associated with undetectable viral load and what we are being told. Strub says “We have neglected to recognize the extent to which a person who is on treatment and undetectable is rendered non-infectious.” How fair a statement do you think is that?
James: Well first of all we need to avoid using the term non-infectious. There is a general consensus that the risk of HIV transmission is not eliminated when the viral load is undetectable.
Among people who are well versed with the research, I think most agree that being on antiretroviral treatment and having an undetectable viral load significantly reduces the risk of HIV transmission for heterosexual couples.
I don’t think that information is being withheld, we just don’t have a consensus yet on what we should be saying. For most people, the research that we have only provides partial answers to the key questions and this information is difficult to communicate accurately because of the caveats and uncertainties. There are no simple messages yet that applies to everyone. There is the potential for a lot of misunderstanding to occur which could have negative consequences. A major concern is that people may switch from the correct and consistent use of condoms to a strategy that is less protective.
However, I do think we need to acknowledge that not everyone consistently uses condoms (for a variety of reasons) and these individuals need accurate messages on other ways to reduce their risk of HIV transmission, including “treatment as prevention.”
Bob: Strub also talks about the relative risk associated with undetectable viral load and the use of condoms, saying that undetectable may afford the greater protection of the two. When I interviewed Dr Julio Montaner he said much the same thing “I think you should be fully comfortable with advising fully suppressed individuals on HAART that they are as well protected as when using condoms, if not better protected” is what he said to me. What do you think of these comparisons?
James: I think we need to be careful when we make comparisons to condoms because these are two very different strategies and both have their own caveats.
We also have to make sure we aren’t always pitting condoms against “treatment as prevention” and creating an either/or situation. Both can fail to prevent transmission in their own ways and using both in combination may add an extra “backup” layer of protection.
We know that if a condom is used consistently and correctly (and the condom doesn’t break, slip or leak), then the risk of HIV transmission is pretty close to zero because no exposure to HIV can take place. HIV cannot pass through the material that is used to make condoms. Of course, condoms are not without their own caveats. Condoms aren’t always used consistently and there are lots of ways in which condoms can be used incorrectly. Also, we know that a condom can break even if it’s used correctly.
The use of “treatment as prevention” is different and there are more uncertainties. Unlike condoms which prevent an exposure occurring in the first place, “treatment as prevention” aims to reduce the risk that an exposure leads to infection. Since an exposure is occurring, other factors that influence the risk of HIV transmission from an exposure also come into play and may decrease the effectiveness of this strategy. For example, we know that tearing and inflammation, anal sex, and other STIs can increase the risk of HIV transmission from an exposure and therefore may increase the risk of transmission when the viral load is undetectable.
Also, with “treatment as prevention” you are relying on the viral load in the body fluids to be undetectable. However, it’s difficult to know what the viral load in the blood is at any given time and it’s even more difficult to know what the viral load is in the genital and rectal fluids. In addition, undetectable doesn’t mean that there is no virus, so there is still HIV present that could lead to transmission. All these uncertainties make it very difficult to know if, and how well, this strategy will work.
So which is more effective: Condoms or “treatment as prevention”? It really depends on the individual, their risk factors and how well they are able to use the prevention strategy. We know that both are highly effective in reducing the risk of HIV transmission through vaginal sex if used consistently and correctly. In this case, some people may find the consistent and correct use of one option easier than the other and therefore that option may be more effective for them.
However, if used consistently and correctly, condoms are the still most reliable and effective strategy available because they prevent an exposure from occurring in the first place and there are fewer uncertainties and caveats associated with condoms compared to the use of “treatment as prevention.” Also, condoms can lower the risk of HIV transmission to the same level for anal and vaginal sex while the risk of HIV transmission may be higher for anal sex than for vaginal sex while using “treatment as prevention.”
In the end, HIV prevention needs to help individuals adopt strategies to reduce their risk of transmission that are appropriate to their individual circumstances and the acceptable level of risk they and their partners are willing to take. We definitely need guidelines for people who want to use “treatment as prevention” and are willing to accept the risk that comes along with its uncertainties. These guidelines need to include important messages that can help a person keep the risk of HIV transmission as low as possible while using “treatment as prevention”.
Bob: I want to turn to what we know about semen. Almost every prevention message you see discussing undetectable warns that undetectable viral load in the blood doesn’t necessarily translate to undetectable viral load in the semen. But isn’t it true that in more cases than not there is that correlation?
James: Many people who have an undetectable viral load in the blood also have an undetectable viral load in the semen and other bodily fluids. However, studies suggest that this isn’t always true. The percent of people in these studies who have an undetectable viral load in the blood, but a detectable viral load in the semen, ranges widely, from 3% to 48%. Similar studies suggest that the same also applies to vaginal and rectal fluid.
We really need more research to gain a better understanding of how common this is, and why it happens, among people living with HIV who are undetectable in the blood.
Bob: How about the amount of viral load in the semen? Isn’t it true that where it IS detectable in the semen but not in the blood, the viral load in the semen isn’t typically very high, the research seems to indicate, and thus not very infectious?
James: Most research has looked at the association between the risk of HIV transmission and the viral load in the blood, not the viral load in other fluids. This means we don’t really know what a “high” viral load in semen (or other bodily fluids) is in terms of infectiousness.
We do know that, in some cases, the amount of virus in the semen (among people who are undetectable in the blood) can be quite a lot higher than undetectable, over 5000 copies/ml. This difference may be quite significant in terms of HIV transmission but we don’t really know and need more research in this area.
Bob: Do you think we are moving towards a place in time when we will see risk guidelines which take in to account the impact of undetectable viral load? Any guesses when that might be?
James: I definitely think we need to move in that direction. I know of a lot of organizations in Canada that are reviewing the evidence and discussing what their key messages need to be around viral load and risk of HIV transmission. In terms of guidelines from Public Health, I am not sure when those will come. It’s difficult because there are lots of research gaps and the research is still emerging quickly.
Bob: In the meantime it’s being argued before the Supreme Court that a person with undetectable viral load translates to extremely low risk of transmission. Do you think that has the potential to confuse people living with HIV? If so what can be done to end that confusion?
James: The Supreme Court is considering what constitutes a “significant risk” of HIV transmission under criminal law. We need to keep in mind that criminal negligence is a serious charge and the burden of proof is different than it is for public health messages.
We know that the evidence shows that the risk of heterosexual HIV transmission is significantly lowered when someone is on antiretroviral treatment and has an undetectable viral load. Whether this risk is lowered to below what the law defines as a “significant risk” to be considered criminally negligent is up to the Supreme Court.
The court’s decision shouldn’t change the prevention messages we give to people living with HIV who want to use “treatment as prevention.” Regardless of the court’s decision, we still need to inform individuals that treatment does not eliminate risk, there are a number of caveats associated with this approach, and there are certain things an individual can do to keep this risk as low as possible. We will also, of course, need to inform people living with HIV who want to use this as a prevention strategy about the law and any changes that happen with the Supreme Court decision.
Bob: James, thank you so much for talking to us.
James is the coordinator of the Biomedical Science of HIV Prevention Project at the Canadian AIDS Treatment Information Exchange (CATIE) where his work focuses on the biology of HIV transmission and new HIV prevention technologies.