Why, for HIV, Undetectable = Untransmittable
In recent months, an increasing number of national and international scientists, advocates and organizations have issued statements about undetectable viral loads, HIV treatment and HIV transmission.
The most prominent comes from Prevention Access Campaign, the U.S.-based activist group that coined the "Undetectable = Untransmittable" slogan. They spell out what this means:
People living with HIV on [antiretroviral treatment] with an undetectable viral load in their blood have a negligible risk of sexual transmission of HIV.
They clarify that "negligible" means "so small or unimportant as to be not worth considering; insignificant."
Their statement continues:
Depending on the drugs employed it may take as long as six months for the viral load to become undetectable. Continued and reliable HIV suppression requires selection of appropriate agents and excellent adherence to treatment. HIV viral suppression should be monitored to assure both personal health and public health benefits.
Prevention Access Campaign's statement has been endorsed by some of the leading scientists working on this issue. They include Myron Cohen, M.D., of the University of North Carolina (the HPTN 052 study); Jens Lundgren, M.D., of the University of Copenhagen (the PARTNER study); and Pietro Vernazza, M.D., of St. Gallen Hospital (the "Swiss Statement").
The statement has also been endorsed by over 150 organizations. In the United States, endorsers include AIDS United, GMHC, Housing Works, Human Rights Campaign, Latino Commission on AIDS, National Black Justice Coalition, the Positive Women's Network - USA, the San Francisco AIDS Foundation, Southern AIDS Coalition and us -- TheBody.com. The International AIDS Society and a large number of organizations in Europe, Australia, Canada and South Africa have also given their support.
Similar language has been used by some of the world's most prominent HIV scientists. In a statement on World AIDS Day 2016, Anthony S. Fauci, M.D., director of the National Institute of Allergy and Infectious Diseases (NIAID) and Carl W. Dieffenbach, Ph.D., director, Division of AIDS, NIAID, said:
[S]tudies have proven that when an individual living with HIV is on antiretroviral therapy and the virus is durably suppressed, the risk that he or she will sexually transmit the virus is negligible.
How Can Treatment Have This Effect?
With all viral infections, the potential for passing on the infection depends on the quantity of virus in body fluids. When there's a large quantity of virus, transmission is more likely. Having a smaller (or non-existent) quantity of virus makes transmission less likely (or impossible).
HIV treatment works by stopping HIV from replicating in cells, which dramatically reduces the quantity of HIV in the body. Although treatment is unable to cure people of HIV (the virus will still be hiding out in some places, such as the gut), treatment can virtually eliminate HIV from some body fluids. Crucially, these include semen, vaginal fluids and blood.
The impact of treatment can be measured with a blood test, called a viral load, which measures the quantity of virus in blood. When the result is described as "undetectable," this means that, in the sample of blood that was analyzed, the test was unable to detect any HIV or found only a tiny trace.
After a few months of taking HIV treatment, most people with HIV have an undetectable viral load. It should stay that way so long as the pills are taken every day, without interruptions.
What Evidence Do We Have for This?
Pietro Vernazza, M.D., is one of the Swiss physicians who first delivered the "Undetectable = Untransmittable" message in the Swiss Statement of 2008.
He recently said that the evidence he finds most convincing is the simple absence of people who have acquired HIV sexually from a partner who was taking treatment and had a fully suppressed viral load.
He called this the "non-observation argument." Documented case reports of HIV transmission in those circumstances are absent from the world's medical journals. This absence was striking a decade ago, but it is even more so in the years since HIV treatment and transmission became a hot topic for scientists.
"[O]ne might expect an even higher degree of scrutiny amongst physicians to investigate and publish suspected cases compared with the years before 2008," Vernazza wrote. "Therefore, a further lack of documented cases should be even more convincing than it was in 2008."
The PARTNER Study
Some people might wonder whether such cases exist but simply haven't been noticed by doctors. To make sure that wasn't the case, scientists set up the PARTNER study. The aim was to observe and carefully measure the real life risk of HIV transmission in couples that weren't using condoms.
The researchers specifically sought out couples in which one partner was taking HIV treatment and the other partner did not have HIV. Moreover, they only enrolled couples who were already having sex without condoms.
They collected data on 888 couples who had sex without condoms (or pre-exposure prophylaxis [PrEP]) when viral load was undetectable (below 200 copies/mL). There were 58,000 unprotected penetrative sex acts, around one-third in gay couples and two-thirds in heterosexual couples.
And how many people acquired HIV from their partners?
Zero. Not one HIV-negative partner contracted HIV from a positive partner. No transmissions took place within these couples.
What's more, all of the factors that someone might reasonably suggest could make transmission more likely, in fact, did not make the slightest bit of difference.
Are women more vulnerable than men? Zero transmissions in both genders. Is anal sex riskier than vaginal sex? Zero transmissions for both. Is it riskier to be the bottom than the top? Zero transmissions for bottoms and for tops. Do sexually transmitted infections increase the risk? Zero transmissions in people with sexually transmitted infections.
We should add that 11 of the HIV-negative partners did acquire HIV, but from somebody else. Careful analysis showed that in all 11 cases, the virus was genetically different enough from their partners' virus to indicate that they must have contracted it from another person -- from outside their main relationship.
Reporting their results in the prestigious Journal of the American Medical Association last year, the researchers summed it up: "The estimate of the overall transmission rate, and the transmission rate for anal sex, was zero."
With the consequences of 58,000 unprotected sex acts already recorded, the findings are robust. In science, having a larger number of observations or study participants generally boosts the certainty with which scientists can state their conclusions. (In statistics, this is expressed through something called the 95% confidence interval).
However, because the first phase of the PARTNER study had fewer gay couples than heterosexual couples, researchers have less data on anal sex than vaginal sex. To remedy this, a second, ongoing phase of the study is collecting more data on only gay couples. There's no indication that the researchers expect this to reveal any cases of transmission during anal sex when viral load was undetectable. It's more likely to increase the certainty and scientific rigor with which the researchers can state that, with undetectability, the transmission rate during anal sex is zero.
The HPTN 052 Study
The other key scientific study is known as HPTN 052. This randomized controlled trial (perhaps the most reliable scientific method) recruited 1763 mostly heterosexual couples with mixed HIV status. The HIV-positive partners were randomly allocated either to start HIV treatment immediately or to delay treatment until their CD4 count was considerably lower.
Again, the researchers were interested in how many HIV-negative partners wound up acquiring HIV from their HIV-positive partner. The direct comparison of couples who had immediate access to HIV treatment and couples who had to wait for it tells us about the difference that providing HIV treatment makes.
Forty-six individuals acquired HIV from their primary partner during the trial, three in the immediate-treatment arm and 43 in the delayed-treatment arm. This amounts to 93% fewer transmissions in the immediate-treatment group.
But what about those three people who picked up HIV from his or her partner, even though they had been prescribed HIV treatment?
In each case, the HIV-positive partner did not have an undetectable viral load. The transmission occurred either within the first three months of taking treatment (before viral load had been durably suppressed) or after HIV treatment had failed and viral load had risen to detectable levels. "No linked infections were observed when HIV-1 infection was stably suppressed by [antiretroviral treatment] in the index participant," Myron Cohen, M.D., and colleagues reported in New England Journal of Medicine last year.
So, while the 93% figure is some way off 100% effectiveness, it relates to the reduction in transmissions when HIV treatment is provided early, not the reduction when HIV treatment results in an undetectable viral load.
How to sum up this scientific evidence? Demetre Daskalakis, M.D., M.P.H., assistant commissioner for the Bureau of HIV/AIDS at the New York City Health Department put it like this:
Several strong lines of evidence indicate that consistent viral suppression is truly HIV prevention. Given these empiric findings, New Yorkers living with HIV who take their antiretrovirals consistently and have suppressed viral loads should be confident that their risk of HIV transmission is negligible.
According to Matthew Hodson of NAM aidsmap:
This understanding transforms the way that HIV is considered with enormous implications for what it now means to live with HIV and the best ways to prevent it. ... The fear of catching HIV from a sexual partner fuels HIV stigma, which is why it's so important that the 'undetectable equals untransmittable' message is heard and understood.