Vaccine trials – where next?

New efficacy trials of HIV vaccines based on what has been learned in the RV144 vaccine trial in Thailand and the HVTN 505 trial in the US are going forward in heterosexual people in South Africa and will probably also take place in men who have sex with men (MSM) in Thailand, the 13th AIDS Vaccine conference heard last week in Barcelona. These large trials will not start until 2016 or 2017 but there are many more immediate, smaller trials of other vaccines happening that will largely measure immunogenicity (the capacity of a vaccine to stimulate an immune response).

Following on from RV 144

One trial that has already been completed, and whose initial results are currently being evaluated, is a follow-on from RV144 called RV305. In this study, 162 volunteers who received vaccine in the RV144 trial were given two boosts, six months apart, of the original RV144 regimen – either of HIV genes inside a viral ‘vector’ called MVA (ALVAC), an HIV protein called gp 120 (AIDSVAX), or both. A second study, RV306, will give 360 volunteers new to vaccines the original RV144 vaccine regimen but then add in a boost from either or both of the components a year later. A third, RV328, will give AIDSVAX alone, not in the expectation of any efficacy (as this vaccine failed in the original VAX trials) but to look anew at the body’s immune responses to it.

In RV305, 45 people per arm of the study received one of the three booster regimens and nine people in each arm received a placebo boost for comparison. The antibody response to HIV in RV144 faded quite quickly and now, eight years after originally receiving vaccine, less than 2% of the amount of IgG antibody (the useful type) generated originally by the vaccine remained in the blood. One boost of AIDSVAX restored antibody levels to the peak value seen in RV144, and one of AIDSVAX and ALVAC to three times the original level (ALVAC had little effect by itself).

Six months later, these levels had declined by 90% and the second boost did not have as big an effect, though with both components it restored them to the original RV144 peak level – and at these peak levels the antibody levels were capable, in the original trial, of cutting infections by two-thirds, albeit briefly. However, presenter Merlin Robb said that the speed with which these antibody responses declined shows that while RV144 may have been a useful proof of concept, it would require modification to be a practicable vaccine. In addition, the boosts raised the levels of possibly unhelpful IgA antibody by a similar amount (see this report): this means that the immune responses seen might not work for everyone. The response of the second component of the immune system, the T-lymphocytes, was also weak: there was some response in CD4 cells to recipients of both vaccine components, but very little in the CD8 cells.

Other trials in Thailand have been assessing various different combinations of vectors, DNA vaccines, and protein vaccines. They have confirmed that in this kind of regimen at least, the regimen must contain a protein like AIDSVAX in order to generate significant antibody responses: HIV genes by themselves, whether ‘naked’ or wrapped in a vector, don’t seem to do the trick.

A further series of trials in South Africa and hopefully in Thailand aim to build on the RV144 vaccine. The Pox Protein Public-Private Partnership (P5) is a consortium founded in 2010 after the RV144 result to build on its success.

Two follow-on trials are planned in South Africa. The first, HVTN 702, is an efficacy trial which will use a vector similar to ALVAC and a protein boost similar to AIDSVAX but both will be more finely tuned to key immunogenic genes/proteins in clade C, the most common HIV subtype in southern Africa. This is planned as a large efficacy trial and aims to enrol 5000 volunteers during 2016.

The second trial, HVTN 701, is a “discovery” trial with an adaptive design: different candidates will be tested on small groups of volunteers and those who produce the largest and most useful immune responses will be moved forward into progressively larger trials. Glenda Gray, principal investigator of the Phambili trial, commented that RV144’s success was so unexpected that there were no stocks of AIDSVAX left and the trials would have started sooner had not the investigators had to start from scratch in building the most suitable protein. The new vaccine trial series will be called “Uhambo”, a word meaning ‘opportunity’ in Shona and ‘journey’ in Xhosa.

In Thailand, progress from RV144 has been slow owing to difficulties in finding funders to back a new trial in MSM, who are seen as not such a large market as heterosexual men and women in Africa. In August, however, a new AIDS Vaccine Efficacy Consortium (AVEC) was announced, having secured funding from the Thai government, to pursue similar ‘RV144-plus’ vaccines in gay men and other vulnerable populations there. AVEC hopes to start a large efficacy trial in 2017.

More trials in Africa

Another series of trials is taking place in east Africa. HIVIS 06 was an investigatory trial using three doses of lengths of HIV DNA followed by two doses of a vector vaccine using the MVA virus similar to ALVAC. Immune responses were relatively modest, although boosted in some volunteers by a third MVA-vaccine dose, and although relatively modest, this immune response did seem to be long-lasting, detectable three years later. The larger TAMOVAC trial (named because it is taking place in Tanzania and Mozambique) will use a simplified regimen. Preliminary results from Mozambique indicate that useful immune responses can be obtained by using a needle-free injection system called Zetajet that ‘sprays’ the vaccine into the skin, with better cellular penetration.

This Zetajet technique generated a CD8 response of nearly 1000 so-called ‘spot forming units’ (SFUs) per million blood plasma T-cells. SFUs are one of the simplest ways to measure immunogenicity: they measure the proportion of cells that are producing the cytokine (chemical signal) interferon-gamma.

Are we too slow and cautious in HIV vaccine research?

In one of the most provocative presentations at the conference, Adrian Hill of the Jenner Institute in Oxford, UK, a scientist currently working on malaria vaccines, commented that the immune response seen so far in efficacy trials was not nearly enough to be likely to translate into protection.

HIV and malaria were both tough infections, he said: a malaria vaccine needed to create a response 500 times higher than the licensed vaccine for meningococcal meningitis, for instance, and like HIV it probably requires a combination of two different vaccines targeted at different stages of the malaria parasite’s life cycle to work properly. In early October, a malaria vaccine in Kenya had reduced infections by 67% in an efficacy trial, and the CD8 response in this vaccine was 2800 SFUs per million T-cells. In contrast, the response in the STEP trial was only 300 SFUs and in HVTN 505, 500 SFUs.

Hill was also unconcerned about the possible toxicity seen in the STEP and Phambili trials (see this report); he was not convinced that the higher rate of infections seen in vaccine recipients was anything other than a statistical fluke. He instanced several examples of malaria trials where apparent higher rates of adverse events in vaccine recipients in early trials were not confirmed by larger efficacy trials, and two examples of the opposite (where unexpected adverse effects only appeared late in efficacy trials).

Hill criticised the HIV vaccine field for being too slow and cautious.

“Most HIV vaccine candidates will fail, and you will know if your vaccine is not going to work after you measure immune responses in your first ten subjects”, he said, advocating smaller, faster studies of more candidates. In answer to a question, Hill also told Aidsmap.com that he was not convinced that the apparently positive RV144 result was anything more than a statistical variation either.

The HIV Conserv vaccine – or cure?

One vaccine candidate Hill did think was promising was a vaccine or vaccine concept called HIV Conserv, currently being taken forward by Professor Andrew MacMichael’s team at Oxford University in a series of trials called HIVCORE. This had generated an immune response as high as 5000 SFUs – ten times as potent as HVTN 505. Unsurprisingly, HIV Conserv was the second-most talked about vaccine of the conference, after Louis Picker’s CMV-based vaccine (see this report).

Tomas Hanke of Oxford University explained the concept behind HIV Conserv.

Conventional vaccines incorporating actual genes or gene sections from HIV generate ‘wasteful’ immune responses, in that the most immunogenic parts of HIV are also the most genetically variable ones. This means firstly that it is very difficult to create a vaccine that works with a large variety of different viral subtypes and secondly that the immune response is one that the virus can easily escape (mutate itself away from).

HIV Conserv is a ‘chimeric’ vaccine, meaning that, like a Frankenstein’s monster, it is stitched together from different parts of HIV that do not lie next to each other in nature. It is made up of a chain of 14 different pieces of the HIV genome, the majority from the polymerase (pol) gene, that combine immunogenicity with the property of being highly ‘conserved’. This means that they vary little between viral subtypes, because HIV finds it difficult to mutate away from them without becoming a very much less ‘fit’ and poorly-replicating virus in the process.

The first HIVCORE trial packaged the HIV Conserv gene-string in three ways – as a plasmid (a naked ring of DNA), and inside two viral vectors, an MVA virus (like ALVAC) and an adenovirus (as in STEP, Phambili and HVTN 505, except that this vector came from a chimp rather than a human adenovirus, for added safety).

The researchers gave volunteers the HIV Conserv vaccine in three different regimens: as chimp adenovirus then MVA (CM): as three shorts of DNA then chimp adenovirus then MVA (DDDCM): or at the same regimen but with MVA before the adenovirus (DDDMC).

These vaccines were highly immunogenic, generating CD8 cell responses to HIV of 5150 SFUs for the CM regimen and 5800 for the DDDCM regimen (the third regimen CD8 response was 2000 SFUs, which is still good but not as good as the others). Furthermore, the immune response seems quite persistent, with responses of 600 and 1500 SFUs after a year for the DDDMC and DDDCM regimens respectively.

Immunogenicity does not always mean efficacy; so the researchers also took immune-stimulated CD8 cells out of volunteers’ bodies and incubated them with CD4 cells and HIV of three different subtypes: the virus was stopped from infecting the CD4 cells in 15 out of 23 cases.

The HIVCORE trials will now move forward into trying further experimental regimens from the Conserv vaccine, in Oxford, London and Nairobi. Since complete protection from HIV will need an antibody response as well as a CD8 response, MacMichael’s team are working with IAVI to develop an antibody-stimulating vaccine to go alongside the Conserv vaccine.

As a CD8 vaccine, HIV Conserv could also work as a therapeutic vaccine in people with HIV, possibly as part of a cure or HIV-remission study. It could help to supress the last remnants of HIV in cells after the majority of the HIV-infected cells had been purged by immune-stimulating drugs and antiretroviral therapy (see this report).

A team based in Barcelona will be looking at the use of the CM Conserv regimen in people who started HIV therapy in the first six months after infection, to see if they might be able to stay virally undetectable when their ART is discontinued. So far 24 people have received the chimp adenovirus and, of these, eight have received the MVA virus too.

We will have data on the immunogenicity of this regimen by summer 2014. This and similar research, including the CMV vaccine too, hold out the promise that a cure for HIV and something that prevents it may look very similar; certainly, cure and vaccine research are starting to move together into a useful synergy.