HIV Undetectable Does Equal Uninfectious: The Swiss Statement and the Vindication of Pietro Vernazza
On Jan. 30, 2008, Pietro Vernazza, M.D., was on vacation with his wife, Eva, in the Swiss Alps when his phone rang.
It rang again when they got back to their hotel room, and then again the next day. The whole week was taken up with one call after another. When he returned home a few days later, he appeared on Swiss national radio. Days after that, he began what turned into a worldwide tour of public health and professional organizations. By October, he was at the U.S. Centers for Disease Control and Prevention (CDC), answering the same questions he'd been answering for nearly a year: What were you thinking? How could you say that? Where's the data?
What he'd said, in a statement in the Bulletin of Swiss Medicine, was this: If you're a person living with HIV on consistent antiretroviral treatment (ART), if your viral load has been undetectable for at least six months, if continuing testing shows that your viral load continues to be undetectable, and if you don't have other sexually transmitted infections (STIs), you don't need to use condoms during sex.
You cannot, the statement said, pass on HIV.
This was the Swiss Statement, a public policy salvo lobbed into the heart of the HIV world -- one that prompted a swift and intense backlash. People called it premature. People accused Vernazza of getting ahead of the science. Some demanded he produce data to prove there had been no transmissions. Others simply called him irresponsible.
But it's not 2008 anymore.
"The Swiss Statement, when it first came out, I thought it was outrageous," said Seth Kalichman, Ph.D., an HIV researcher at the University of Connecticut. "The Swiss Statement isn't outrageous anymore."
Indeed, nearly nine years later, it's more than a statement. It's a fact, backed up by gold-standard studies released at the Conference on Retroviruses and Opportunistic Infections (CROI) and the International AIDS Conference (IAC).
"The evidence base has grown," said Jared Baeten, M.D., Ph.D., a University of Washington researcher who has been studying HIV prevention for more than a decade. "Now it backs up the Swiss Statement."
This is the story of how the Swiss Statement went from policy pariah to documented fact. And it is the story of the vindication of Pietro Vernazza.
Entering the Antiretroviral Treatment Era
The concept of lower levels of virus in the blood leading to lower rates of transmission is a truism in biology.
But in 1993, we were still two years away from the U.S. Food and Drug Administration approving the first protease inhibitor drugs that ushered in the new era of antiretroviral treatment. Americans with HIV were dying, often quickly, and there was no way to stem the tide of the virus in the blood and cells of people who'd acquired it.
Back then, Vernazza -- at the time a post-doctoral fellow at the University of North Carolina (UNC) -- wasn't concerned with lowering transmission rates. He was studying something far simpler: whether you can determine if a man living with HIV is infectious by measuring his semen.
But then, protease inhibitors came out, and scientists started to ask a different question: If treatment could reduce viral load -- that is, the amount of HIV in the blood and genital tract -- might there be literally nothing for them to pass on? Might they be functionally uninfectious?
In a 1999 paper in the journal AIDS, HIV researchers at UNC, including Vernazza, Joseph Eron, M.D., Susan Fiscus, Ph.D., and Myron "Mike" Cohen, M.D., tried to figure out the likelihood that someone living with HIV would pass on the virus during sex.
It was a puzzle. Not every form of transmission was as effective at spreading the virus, the team wrote in AIDS. Mother-to-child transmission was highly efficient and so was blood transfusion. But sex? That, the UNC researchers said, "is less efficient and highly variable."
The challenge wasn't just that some people seemed to replicate HIV more slowly than others, meaning there was less of it in their body. It was also that some HIV treatments seemed to work better than others at suppressing viral load. Some, such as zidovudine (AZT, Retrovir), had been found to reduce mother-to-child transmission of HIV even when treatment started late in pregnancy -- a promising development. But other drugs weren't powerful enough to reduce transmission, the paper said. And the data that did exist seemed to indicate that drugs that could lower the virus in the blood weren't always successful in doing so in semen and vaginal fluid.
"Patients need to be carefully informed about the significance of treatment induced reduction of genital shedding of HIV," the paper stated. "Reduced infectiousness does not equal lack of transmission and more importantly, not every antiviral treatment does result in reduced infectiousness. In order to be effective on a public health basis, reduced infectiousness must be coupled with continued safer sex practices."
In other words, condoms. In practice, what this ended up meaning, Vernazza said recently, is that doctors didn't talk at all about the power of treatment to reduce transmission.
"We had prevention campaigns and ads all over the streets, on TV, saying condoms, condoms, condoms," Vernazza said. "That was a campaign from the '80s that had just been maintained over all these years."
But by 2003, Vernazza was starting to feel differently.
By this time, he had finished his post-doc work in North Carolina and had taken a job as the chief of the Division of Infectious Diseases at the Cantonal Hospital in St. Gallen, Switzerland. There, he was working with a very specialized group of people living with HIV: straight people on treatment who wanted to have children with their HIV-negative partners.
A survey of all the patients who were part of the Swiss HIV Cohort Study, an ongoing national study of Swiss people living with the virus, found that about one in five or six were not using condoms. And that was only what people were willing to tell their physicians; an anonymous questionnaire sent out later found substantially higher rates of condomless sex.
Even if only one in four people with HIV on treatment were having sex without condoms, they should be seeing their partners' acquiring HIV, Vernazza said. But they weren't.
"Out of 8,000 patients, we did not see reports of transmissions to a partner," he said. "So we knew the risk must be low. And when we interviewed these HIV-discordant couples, very uniformly they told us about their assumption of a very high risk of transmission despite very long and stable HIV-RNA suppression in the infected partner. Because of the wrong information they had, they did not conceive children even though they wanted to."
For Vernazza, it was an ethical dilemma. In the world of medical ethics, it's called equipoise: If a clinician sees two options as equally valid, he is ethically obligated to offer all treatments to his patients and let the patients decide. For example, if the couples you're working with really want to have children and using condoms is prohibiting this, and if treatment is preventing the partners who are living with HIV from passing on the virus, the couples might need to be told. To do otherwise, he said, would unnecessarily force a hardship on those couples' relationships.
He knew talking to patients would go against everything public health campaigns said about preventing the spread of HIV. When I spoke to Vernazza in 2011 for my book Positively Negative, he told me, "There's a belief that [we] should not tell."
Indeed, in practice, people living with HIV were getting the opposite of equipoise. Doctors were telling patients the risk of transmission was real, even though, Vernazza said, "none of us in the whole medical field had ever seen a single case of transmission under these circumstances." Add in the fact that other public health recommendations -- such as those regarding the negligible risk of HIV transmission via kissing -- were based on far smaller samples of people (100, by Vernazza's estimate) and Vernazza said he felt compelled to speak up.
"We felt we needed to tell them: 'Listen, the risk of transmission is not as high as you assume,'" he told TheBody.com. "In fact, it's probably zero."
He started to roll out the options with his patients that year and the next -- offering, in addition to timed, condomless sex for conception, an HIV prevention pill before and after sex for the HIV-negative partner (an approach now called pre-exposure prophylaxis or PrEP). The couples who took him up on it were carefully monitored, including counseling and treatment for STIs and regular viral load monitoring.
Many had babies. No one got HIV.
By 2008, he and his colleagues on the Swiss Federal AIDS Commission, of which he was now the president, were ready to roll this out as policy for the rest of Switzerland's HIV doctors.
He got together with physician groups, public health policy officials and legal experts and drafted guidance meant for Swiss physicians only. It laid out the current research: two observational studies, along with the mathematical model he had contributed to at UNC, and the experience of the Swiss HIV Cohort. On Jan. 30, 2008, it was published in the Bulletin of Swiss Medicine:
An HIV-infected person with potent [antiretroviral therapy, defined as "combination therapy against HIV which reduces the viral load below the limit of detection"] is not sexually infectious, i.e. does not transmit [HIV] via sexual contacts as long as the therapy is practiced consistently and monitored regularly by the treating physician; the viral load on [antiretroviral therapy] has been below the limit of detection for at least six months; [and] no infections with other STIs are present. Under these circumstances, potent [antiretroviral therapy] therefore definitely prevents HIV transmission as safely as condoms.
Vernazza and his wife went on vacation. The phone began to ring and didn't stop. Vernazza was, he said, "completely overwhelmed" by the response.
Almost all of it was negative.
Show Me the Data
Kalichman remembers where he was when he saw the Swiss Statement that January in 2008. He was in his office at the University of Connecticut, where he does behavioral research on people living with HIV who also use drugs or have multiple partners. He said he grabbed his desk phone and dialed up the World Health Organization (WHO).
"What do you think is going on here?" he remembered asking them.
Kalichman, a 2005 Distinguished Scientist from the Society for Behavioral Medicine, just knew it would be misinterpreted by the masses.
And he wasn't the only one. Public health agencies including WHO, UNAIDS and those in France and Australia roundly rejected the Swiss Statement. Anthony Fauci, M.D., director of the National Institute of Allergy and Infectious Disease (NIAID) at the National Institutes of Health (NIH) told TIME magazine that he, WHO, and the CDC were "concerned" about telling people that there was no risk.
"There is no such thing," he told the magazine, "as zero risk."
By the time the International AIDS Conference rolled around in July that year, the statement had become the subject of a special panel discussion at which Vernazza answered the questions of several scientists, among them his old UNC mentor, Mike Cohen. At the two-hour event, Vernazza apologized for the statement's misleading headline ("HIV-positive people with no other STIs and on effective antiretroviral therapy do not transmit HIV sexually"), saying he would have said that people living with HIV have "very little" chance of transmitting HIV rather than "do not" transmit it. He added that the statement was never meant to influence global policy; it was intended to guide Swiss doctors, only, in discussing sexual risk with their patients.
Individual researchers also reacted. Some, such as Sharon Hillier, Ph.D., a microbiologist at the University of Pittsburgh and a member of the program committee of the HIV Research for Prevention conference, said her first reaction to the statement was, "Yes, of course."
"To be quite honest, as a biologist, you think, 'If you're fully suppressed you can't transmit the virus,'" she said. "And so, to me, it's just logical. I didn't think it was some blinding breakthrough."
But not everyone was so sanguine.
"Being a scientist, you scrutinize, 'What is the data in support of that statement, that you essentially become non-infectious if placed on treatment?'" said Jens Lundgren, M.D., DMS.c, professor of infectious diseases and founder of the Centre for Health and Infectious Disease Research at the University of Copenhagen. "We need to separate out our personal feelings from the process of doing research, because the research has to be objective and has to be looking at both ends of the question. Therefore, opinions should be derived from the research results. You should be very concerned if you're building too strong an opinion before the research is done."
The Swiss Statement, he said, "was an opinion based on plausible but weak data."
UNC's Cohen agreed, saying he worried that a few observational studies, a mathematical model and Vernazza's clinical experience weren't enough to call for policy change.
"Policy decisions depend on high levels of evidence," Cohen told TheBody.com. "The Swiss Statement was a giant piece of deductive logic."
Others had even stronger reactions. Vernazza said at least one researcher stopped talking to him after the statement came out.
Despite the blowback, Vernazza said arguments against the Swiss Statement only bolstered his belief that he was right about transmission and infectiousness. But more than that, he said, his patients couldn't wait for the clinical trials to be completed if they wanted to have children.
"I was saying in 2008 that it will take eight years to show that we were right," he said recently. "I do not agree that we should let patients wait for this relief. And now, look -- it took eight years."
Getting the Data
One thing that's true about HIV researchers is that they fight about issues for years -- and that fight only ends when the data is in.
Take immediate versus delayed antiretroviral therapy.
"It was a huge controversy," said Lundgren. "There were strong emotions."
The emotions and shouting only cooled last year, he said, when data from the START trial proved that early testing and treatment were better for everyone.
So the fact that there were vigorous debates, even fights, about the Swiss Statement isn't a problem. It's an incitement to research.
"The only thing I know to do when I see things like this is to generate research that either refutes or supports the statement," Lundgren said. "That's what we did."
Indeed, Lundgren worked with a team of researchers, including Vernazza, on the PARTNER study, a study that followed a group of straight and gay people living with HIV and on suppressive antiretroviral therapy and their HIV-negative partners as they had condomless sex to see whether treatment really did prevent transmission.
For the scientists involved in another study called HPTN 052, the Swiss Statement didn't influence their research -- they'd begun planning the international, randomized controlled trial of HIV treatment and prevention in 2000. But it did have a big impact on it, said Mike Cohen, HPTN 052's principal investigator.
"I'll say this about the Swiss Statement," Cohen said. "It shined a light on 052."
So when the early results of HPTN 052, which followed 1,763 HIV-positive people around the world and their partners, were unblinded in 2011, they were the first concrete facts on the power of treatment to reduce transmission. And boy, did they reduce transmission: In 1.7 years, only one new HIV acquisition could be linked directly to a partner on antiretroviral therapy. For people not yet on treatment, there were 27 new linked HIV transmissions.
That's a 96% drop in infectiousness. (But that doesn't mean that someone on suppressive antiretroviral therapy has a 4% chance of passing on the virus. The initial study didn't investigate whether the person whose partner acquired HIV from them had a suppressed viral load; that came later.)
Unlike Vernazza, who had total confidence in the power of treatment to reduce transmission, Cohen had not been as sure what his study would find. There's a myth, he said, that the 052 team was confident from the start. Not so, he said -- not at all.
"My colleague Marybeth [McCauley] was literally shaking [when it became clear that the study's oversight board was going to make a recommendation that would affect the future of the study, in 2011]," he said. "We absolutely thought the study had failed."
But it hadn't. That year, Science named HPTN 052 its Breakthrough of the Year.
That was followed, in 2014, by the preliminary results of the PARTNER study, which reported similarly spectacular outcomes. Out of 48,000 instances of condomless sex between people on suppressive antiretroviral therapy and their HIV-negative partners, there was not a single linked case of HIV. That means that, if people did acquire HIV, it was genetically different enough from their partners' virus to indicate that they got it from someone else.
This was particularly important because PARTNER, unlike HPTN 052, only studied condomless sex, not HIV transmission with condom use. More than 90% of the couples in HPTN 052 reported that they used condoms, leading some researchers, including Lundgren, to say that the benefit of treatment for reducing transmission was probably even higher. Plus, PARTNER included a large number of gay couples, showing that treatment also prevents transmission via anal sex.
Then, to top off this series of positive results, Cohen's team reported out the final results of HPTN 052 in July 2015. They were even better than the preliminary results: Of the new cases of HIV in the study, all were linked to people who'd been on treatment for less than six months or who hadn't yet started treatment -- in other words, to people who didn't meet the Swiss Statement criteria.
"The real point, to us, is that because of the results of 052 and PARTNER's work with gay couples, it becomes increasingly compelling," Cohen said, "that if people are sufficiently suppressed, transmission becomes exceedingly rare."
"The data," he said, "support strongly that the Swiss Statement is correct."
But just because 052 and PARTNER have validated Vernazza's 2008 statement doesn't mean resistance to the undetectable equals uninfectious concept has evaporated.
When asked whether they would sign on to the Swiss Statement if it were released today, only Hillier and Jared Baeten, principal investigator of the Partners PrEP trial, said unequivocally that they would. Lundgren said he would, but only with the caveat that we still need more information about treatment's impact on transmission via anal sex -- which the final results PARTNER will address.
Cohen would sign on, too, he said, now that we have the data from HPTN 052 and other studies. But scientists still need to understand why the virus continues to shed into the genital tract, even when people are on antiretroviral therapy, he said -- something that he said could both affect the risk of transmitting the virus and provide information leading to a cure.
"As long as you stay virally suppressed, it's durable prevention," Cohen said. "But if you stop being virally suppressed, transmission becomes possible again. So we don't believe we've achieved some sustained prevention in the absence of antiretroviral treatment."
And you could stop being virally suppressed for reasons that have nothing to do with whether you take your meds every day, said Kalichman, who works primarily on adherence myths and treatment beliefs (such as the belief that antiretrovirals cure HIV rather than suppressing the virus as long as you continue to take them).
"The big ringer in this is genital tract inflammation," he said. "That's why, for example, in Vernazza's research that the Swiss Statement was based on, the people in those studies, the men, were tested for STIs and clinically cleared for STIs every three months or something. That's also true on 052."
To him, the message isn't that treatment makes one uninfectious, but that treatment along with regular viral load tests showing an undetectable viral load and excellent sexual health care every three to six months makes one uninfectious. It's a system of care, not a magic pill.
"If you just assume that if you're adherent, then you have an undetectable viral load in your genital tract, you're wrong," he said.
Maybe that's why some public health policy people are still hesitant to endorse the Swiss Statement. Take Eugene McCray, M.D., director of the CDC's Department HIV/AIDS Prevention. He said in an email to TheBody.com that we've come a long way since the Swiss Statement. But he wasn't willing to go so far as to say that undetectable equals uninfectious.
"HIV treatment is a powerful prevention tool," he said, "but [it] alone is not sufficient to stop HIV transmission."
That hesitance may be showing up in medical publishing, too. Soon after Lundgren and the PARTNER team presented their data on condomless sex, suppressive antiretroviral therapy and HIV transmission at CROI in early 2014, they submitted the data to medical journals. Several passed. Finally, the Journal of the American Medical Association (JAMA) published the paper in April 2016.
That's 30 months from presentation to publication and 14 months for JAMA to review and publish the article.
It's a lag time that Lundgren said is "an unusually long time for an editorial process of a submission." And indeed, Kalichman, who edits the journal AIDS and Behavior, said that while every journal's process is different, 11 months was the longest it had taken his journal to review and publish.
And though Lundgren said he doesn't often think about the delay anymore, he added that "the long time from first presentation at CROI to eventual publication has created some degree of frustration within the wider scientific community, which didn't understand why the paper was still not published."
Vernazza, who was on the executive board of the PARTNER study and is listed as one of its co-authors, suspects that the same attitude that informed responses to the Swiss Statement continues to color responses to the hard data.
"It wasn't that the data was bad; it was because [journal editors] didn't like to show the evidence," said Vernazza. "It was unacceptable, this process -- which tells you there were other forces that tried to push this publication back. That's my assumption based on experience."
Howard Bauchner, editor of JAMA, did not respond to an email seeking comment.
A New Consensus
In July, not long after Cohen presented the final results of HPTN 052 at IAC, the advocacy group Prevention Access Campaign came out with this statement:
People living with HIV on [antiretroviral therapy] with an undetectable viral load in their blood have a negligible risk of sexual transmission of HIV. 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.
The statement includes a note that an undetectable viral load only prevents HIV transmission -- not the transmission of other STIs or pregnancy -- and that people "having sex with multiple partners or in a non-monogamous relationship ... might consider using condoms to prevent other STIs." It also includes a definition of the word negligible: "so small or unimportant as to not be worth considering; insignificant."
The signers of the new statement? Leading public health officials in New York City, Australia and Canada, but also Myron Cohen from UNC, Jens Lundgren at the University of Copenhagen and, of course, Pietro Vernazza, who is listed in PAC's press release as "author, Swiss Statement 2008, Update 2016."
Kalichman, in an email to TheBody.com, called it "Swiss Statement 2016."
They're not alone.
In September, the directors of all U.S. state-based HIV programs, under the auspices of the National Alliance of State and Territorial AIDS Directors, signed on to the statement, which places condomless sex with someone who has an undetectable viral load in the same negligible risk category as spitting.
Then, at the 2016 National Ryan White Conference on HIV Care and Treatment, no less than the head of the NIH, Carl Dieffenbach, Ph.D., echoed the statement, saying, "Once you begin therapy, with full virologic suppression, you are not capable of transmitting HIV to a sexual partner. ... With successful [antiretroviral therapy], that individual is not infectious."
Even Richard Wolitski, Ph.D., acting director of the Office of HIV/AIDS and Infectious Disease Policy at the U.S. Department of Health and Human Services has praised the new consensus statement.
And the NIH is sending another signal, too, about the role of condoms in the lives of people living with HIV and on suppressive antiretroviral therapy. NIAID's Fauci -- the same man who said in 2008 that the Swiss Statement concerned him -- told TheBody.com that, in retrospect and with the data from HPTN 052, the Swiss Statement "turned out to be rather accurate."
"My major statement to you is that the data we've accumulated from the time the Swiss Statement came out to today, now several years later," he said, "actually confirms and verifies that someone with an undetectable viral load on [antiretroviral therapy] has a very low chance of transmitting HIV to someone else."
When asked whether he would endorse the Swiss Statement, including that people could have sex without condoms if they meet Swiss Statement criteria, Fauci said now it's up to the person living with HIV and his or her partner.
"It depends on how much risk someone is willing to accept," he said. "If people are OK with a very, very low risk that's not zero, they can take that risk. There are also going to be people who recognize that, you know, it's possible to get a blip [of detectable viral load] once in a while and who want that very, very low risk to be as close to zero as possible. Then they can go the extra step and use a condom."
Vernazza stands by the release of the Swiss Statement. Maybe back then, he says, it would have been more accurate to say that we'd learned that people living with HIV and on stable HIV treatment with suppressed viral loads for more than six months "are very unlikely to transmit" HIV, rather than that they "cannot transmit" the virus. But he doesn't regret the controversy -- or starting the discussion. And this year, when the Prevention Access Campaign created a consensus statement saying that the risk of HIV transmission from those described in the Swiss Statement is actually negligible to non-existent, Vernazza came on as one of the first endorsers.
"Perhaps," he said, "the issue would never have gained this importance if I'd phrased it less provocatively."