Vaccine did not prevent HIV infection: non-significant increase in infections in vaccine recipients
The US National Institute of Allergy and Infectious Diseases (NIAID) has announced that it is discontinuing the HVTN 505 HIV vaccine trial. This trial, which started in July 2009, has involved 2504 gay and transgender volunteers in 19 US cities. Since the successful conclusion of the RV144 vaccine trial in September 2009, HVTN 505, as a randomised, placebo-controlled phase IIb trial, has been the only ongoing HIV vaccine trial large enough to be a true test of vaccine efficacy.
NIAID stopped administering injections when the trial‘s independent data and safety monitoring board (DSMB) found during a scheduled interim review that there was no sign that the vaccine regimen was preventing HIV infection, nor any sign that it was reducing viral load among vaccine recipients who became infected with HIV.
The DSMB found that there were actually more HIV infections in volunteers receiving vaccine than placebo, but it is important to emphasise that this difference was not statistically significant and may have been due to chance. Statistically speaking, the vaccine had zero efficacy.
The HVTN 505 study was testing an investigational ‘prime-boost’ vaccine regimen developed by NIAID’s Vaccine Research Center. It involved a series of four injections. The first two, at the start of the study and four weeks later, consisted of a length of DNA – artificial genetic material – that ‘coded’ for proteins found on the surface and inside the HIV virus. The idea was to sensitise the immune system to the specific HIV genetic sequences.
The third injection, at eight weeks, involved a vector. This means the same HIV genetic material was wrapped inside the shell of a different virus, an adenovirus, one of the types that cause common colds. In this case the viral shell was altered so that it could not cause illness. The idea of a vector is that it causes a ‘fake infection’: the viruses can carry the genetic material through the cellular membrane and into the interior of immune-system cells. The two investigational vaccines tested in HVTN 505 cannot cause HIV infection because neither contains live or weakened versions of HIV.
The reason behind a prime-boost design is that it is thought to be the best safe way to stimulate both branches of the adaptive immune system: antibodies, which stop viruses getting into cells in the first place, and CD8 cells or cytotoxic T-lymphocytes (CTLs), which kill off virus-infected cells. Researchers hoped that if a prime-boost vaccine were successful, it might prevent infection altogether in the majority of people, but in the minority who were still infected, it might kill off enough virus-infected cells to permanently contain HIV replication and produce a consistently low HIV viral load.
The fourth injection, at 24 weeks, involved an injection of the viral vector alone, without any HIV genetic material. This was to gauge the level of immune response to the adenovirus shell rather than to the HIV material it contained. This is important because in one of the previous vaccine efficacy trials, the STEP study, the vaccine actually made people with high levels of pre-existing immunity to the adenovirus vector more, rather than less, vulnerable to HIV. In the case of HVTN 505, volunteers were required to have no pre-existing immunity to ad5, the adenovirus vector used.
In its April 22 interim review, the DSMB looked at volunteers who were diagnosed with HIV infection after having been in the study a minimum of 28 weeks and found that 27 HIV infections had occurred among the vaccine recipients and 21 among placebo vaccine recipients. Twenty-eight weeks was chosen because by this time the vaccine, if it worked, would have stimulated a sufficiently strong protective immune response. Including volunteers who had become infected less than 28 weeks after enrolment, there were 41 cases of HIV infection in volunteers receiving vaccine regimen and 30 cases in those receiving placebo.
Additionally, the DSMB found that viral load among the 30 volunteers who acquired HIV infection at least 28 weeks after entering the study, and who had been followed for at least 20 weeks after diagnosis, was no lower in vaccine than in placebo recipients. Study volunteers are being asked to report to their specific clinic sites over the next few weeks to find out whether they received the investigational vaccines or placebo. Individuals who became HIV-infected during the trial were referred to local services for appropriate medical care and treatment.
The HVTN 505 study will continue follow-up with study participants to further evaluate the trial data, and especially to see if the greater number of vaccine recipients who were infected is in any way significant.
- For more information about the HVTN 505 study, please see the updated Questions and Answers page here.