A Dutch man has contracted a non-drug­-resistant strain of HIV while, according to numerous sources of evidence, adhering well to the daily regimen of Truvada (tenofovir disoproxil fumarate/emtricitabine) as pre-exposure prophylaxis (PrEP). His case has puzzled experts because the only two cases of PrEP failure documented thus far, both reported within the past year, involved rare drug-resistant strains of the virus that apparently evaded the two drugs in Truvada.

Researchers are able only to speculate as to how this man could have contracted HIV. They theorize that his “remarkably high” number of sex partners and condomless sex episodes may have contributed to this aberrant outcome.

Elske Hoornenborg, MD, an infectious disease specialist the Public Health Service Amsterdam presented findings from the case study in a poster presentation at the 2017 Conference on Retroviruses and Opportunistic Infections (CROI) in Seattle.

Elske Hoornenborg of the Public Health Service Amsterdam with her poster presentation at CROI 2017 in SeattleBen Ryan

Seeking to assuage anxieties that may arise from the news of this case, Hoornenborg told POZ at CROI, “Internationally, tens of thousands of people are using PrEP. And we see that PrEP offers a high degree of protection against HIV. It’s not 100 percent. So we’ll have to acknowledge that. It also doesn’t protect against STDs. So we think it’s clever to combine different prevention methods—PrEP being one of them, condoms and other risk-reducing behaviors being other ones.”

Researchers estimate that PrEP reduces the risk of HIV by 99 percent when taken daily as prescribed. In fact, due to a high level of “dosing forgiveness,” taking just four tablets per week still likely confers maximum protection (although taking PrEP that infrequently is not recommended).

The Dutch case study involves a 50-year-old man who has sex with men (MSM) who began taking daily PrEP as part of the Amsterdam PrEP study, known as AMPrEP, of which Hoornenborg is the head. Upon his enrollment in the study, he tested HIV negative according to a fourth-generation HIV antigen/antibody test and an HIV RNA test.

The fourth-generation test looks both for HIV antibodies, which typically take a few weeks to develop after infection, and the p24 antigen, which normally appears earlier than antibodies and disappears after a few weeks. An HIV RNA test looks for the virus’s genetic material and is even more sensitive to very early infections.

Counts of the number of Truvada tablets that the man had remaining when he returned to each study visit, as well as daily diary information he logged, indicated he was adhering well to the daily Truvada regimen. Dried blood spot testing taken at the six- and eight-month points during his time on PrEP confirmed a high level of adherence. Dried blood spot tests have the capacity to essentially look back in time and determine whether someone has been adherent to PrEP during the previous eight weeks or so.

The man tested negative for HIV according to fourth-generation tests administered one, three and six months after starting PrEP.

He kept a diary of his sexual activity on a mobile phone application. During months one through eight of his time on PrEP, he reported a respective 75, 56, 56, 50, 38, 49, 66 and 12 anal sex partners. (The eighth month constituted only 20 days.) During each of these months, he reported having condomless anal sex on a respective 21, 12, 13, 15, 15, 19, 17 and 3 days. According to his reports, during the days he had condomless anal sex, the median number of partners he had on those days was 3 during the first month, 4.5 during the second, and 4, 4, 2, 3, 5 and 5 for each following month, respectively.

According to computer-assisted questionnaires the man completed, during his first 12-week period on PrEP, he had 90 sexual partners with whom he engaged in condomless anal sex and 100 total condomless anal sex acts. During the second 12-week period, the corresponding figures were 51 and 100, respectively. There were no data on the subsequent 12-week period of his PrEP taking, which was ultimately cut short after he showed signs of HIV infection.


While he was on PrEP, the Dutch man was twice diagnosed with rectal gonorrhea and once with rectal chlamydia.

He reported using recreational drugs during sex, including crystal meth, cocaine, GHB, mephedrone and ketamine.

After he had been taking PrEP for eight months, the man developed symptoms of a fever and painful urination and difficulty urinating, known as dysuria. A test taken at that time indicated he likely had very recently contracted HIV. Strangely, the results of a fourth-generation HIV test were positive for antibodies but not the p24 antigen, the opposite finding expected from a new infection.

The same day, he tested negative for HIV RNA through an assay that had a detection threshold of a viral load of 50. A Western blot HIV test conducted at that time showed that he had only antibodies to the p160 viral antigen, a result that usually comes up later in the course of an infection, after other HIV-specific antibodies arise.

The researchers even went so far as to conduct a highly sensitive assay known as a nested pol PCR test, which looks for both DNA and RNA of HIV.

“The bottom line is we couldn’t detect virus in any way” at that early stage, Hoornenborg said.

As a result of these findings, Hoornenborg and her colleagues were concerned that if the man continued taking Truvada, he might develop drug resistance, so they took him off PrEP. They then monitored him at weekly intervals.

Three weeks after the man tested positive for HIV antibodies, tests showed he had a viral load of 40,000. Because of the high viral load test result, the researchers started the man on combination antiretroviral (ARV) therapy to treat his HIV. He started taking Truvada with Norvir (ritonavir)-boosted Prezista (darunavir) and Tivicay (dolutegravir). Within a month, he had an undetectable viral load.

This man’s case is particularly unique, as well as confounding, because drug-resistance tests indicated that his virus was a wild-type virus, meaning that it did not have detectable resistance mutations. Specifically, his virus showed no mutations associated with tenofovir or emtricitabine, the two drugs in Truvada, or with any other ARVs, for that matter.

The viruses contracted by the two other MSM for whom daily PrEP failed were each highly drug-resistant. (It is particularly rare for HIV to be resistant to both tenofovir and emtricitabine.) This phenomenon likely lowered Truvada’s power to fight these particular infections.

The puzzling question remains as to how HIV managed to infect the Dutch man, not to mention why initial signs of his infection followed such an unusual pattern according to numerous HIV diagnostic tests.

Hoornenborg theorized that it is possible that the man was repeatedly exposed to HIV, which, kept relatively in check by Truvada, remained in a localized infection in his rectum for a time without spreading throughout the body. Then perhaps a drop in Truvada concentration in the rectum, maybe occurring without a corresponding drop in blood levels, allowed a window for HIV to thrive and establish an infection throughout the body. Because there are only two HIV drugs in Truvada as opposed to the three or four needed to successfully treat the virus, PrEP may be unable to thwart such a systemic infection.

The peculiar opening round of HIV test results may have been a reflection of the immune system responding to that initial, ongoing localized infection. Typically, a standard HIV test would not detect a localized infection; it would find signs of the virus only after it spread throughout the body. One recent study found that, at least among monkeys (that were not on PrEP), HIV spreads out of its initial infection site in the reproductive tract as quickly as within one day.

“PrEP has prevented thousands of infections,” said Robert M. Grant, MD, MPH, a professor at the University of California San Francisco School of Medicine who was the head of the global iPrEx study that first proved PrEP’s efficacy among men who have sex with men in 2010. “The report from Amsterdam is one of only a few people worldwide who have become HIV positive while taking PrEP.  His viral load was undetectable until his PrEP was stopped.”

Robert Grant of UCSF addressing CROI 2016 in BostonCourtesy of Benjamin Ryan

Grant intimated that it is possible that a drug-resistant strain of HIV may initially have infected the Dutch man and then receded in the face of thriving wild-type virus by the time the investigators tested his virus for drug resistance. “The virus that eventually appeared was sensitive to drugs; such drug-sensitive virus strains readily overgrow drug-resistant strains, which may have been present initially,” Grant said.

The drug-resistant infections in the other two men for whom PrEP failed did not follow such an evolution, however.

“The good news is that the person is now doing very well on therapy,“ Grant said, ”and his infection was caught early because he was in a PrEP program.”

In their poster at CROI, Hoornenborg and her coauthors concluded, “This case underscores the importance of counseling and monitoring PrEP users, including frequent HIV testing. Furthermore, we should be alert for atypical seroconversion resulting in indeterminate HIV [antigen/antibody] test [results] in individuals on PrEP.”

To read the conference abstract and access the poster, click here.

To read a POZ feature on how PrEP is failing to reach African-American MSM, click here.