What animals say about vaccines

Dr Norman Letvin of Harvard Medical School gave an overview of HIV-related vaccine research on monkeys on Friday 7 September at the AIDS Vaccine 2001 conference in Philadelphia.

In recent years, US animal studies of vaccines have increasingly used a virus called SHIV 89.6P to test HIV vaccine strategies. This virus combines elements of a monkey virus (an SIV) with the envelope protein of HIV itself. It has been made pathogenic by repeated transmission from one monkey to another. Whereas the parent virus, SHIV 89.6 is very slow to cause disease, SHIV 89.6 P is consistently fierce in the rate at which it progresses, with most infected monkeys dying in a few months, unless protected with a vaccine or antiviral drugs.

Dr Letvin explained that these features of SHIV 89.6P mean that valid answers to questions about vaccine efficacy can be obtained using small numbers of monkeys and in a short period of time. The variability and longer time course of disease caused by most SIVs would require much larger studies and would seriously limit the number and variety of vaccines that could be tested. Some studies should continue to be done using other viruses, but it was quite reasonable to standardise on SHIV 89.6P.

Outside the conference, there has been criticism, notably in a Newsday article by Laurie Garrett which quoted several eminent vaccine researchers saying that SHIV 89.6P is too far removed from the natural pattern of HIV or SIV infection to give a valid guide to the effectiveness of vaccines against HIV. They suggest it is too easy to get protection against SHIV 89.6P and that studies should be repeated using other virus challenges to be sure of the validity of these studies.

Dr Letvin answered these criticisms by showing that several groups have been able to replicate with SIV the same kind of immune system control that is achieved over SHIV 89.6P when vaccines induce strong and broad killer T-cell (CTL) responses against HIV. Basically, the initial viral load peak is suppressed, the viral load “set point” is greatly reduced and T-cells do not decline in the way they do during HIV or SHIV disease. He said, emphatically, that we should now move to test such vaccines in full-scale clinical trials.

Harriet Robinson’s studies

By far the most exciting, thorough and clinically suggestive series of animal vaccine studies have been carried out at Emory University, Atlanta, and the Yerkes Primate Research Center, by a team led by Dr Harriet Robinson who gave an overview of their work on Thursday 6 September. This group reported several findings at this meeting which extend their previously published work showing long-lasting protection against disease in vaccinated monkeys exposed intrarectally to SHIV 89.6P a full seven months after the last vaccination they were given.

Dr Robinson's group is now designing versions of their vaccines for human use, based on subtype B (which remains the commonest virus in the USA, Latin America and Western Europe) and a recombinant AG virus (which would be the best match for those circulating in much of West Africa).

The strategy used by Dr Robinson is very similar to that now entering clinical trials in Oxford and Nairobi (see separate story), with a DNA prime followed by a recombinant MVA boost, both incorporating DNA versions of SHIV genes. One difference is that Dr Robinson is using three genes, namely gag, pro and env whereas the initial Oxford-Nairobi trials use only gag. A poster (Amara) showed that the inclusion of env was necessary to get efficient control of infection in the monkeys. This is obviously why the planned “RENTA” vaccine now being designed by the Oxford-Nairobi team is to include part of the env gene.

A second poster (Buge) addresses another important question. Given that env is included in the DNA/MVA vaccine used, would it help to boost by injecting another vaccine based on the envelope protein and designed to stimulate antibodies? The envelope vaccine used in this study induced a strong antibody response although not a very strongly neutralising antibody response, which is important to bear in mind. What was found was that the extra vaccine actually diminished the effectiveness of the DNA/MVA vaccine rather than enhancing it. If VaxGen’s AIDSVAX product turns out to be even partially effective, an immediate question would be whether to give that vaccine to people taking part in trials of vaccines designed to stimulate CTL responses, such as the planned Oxford-Nairobi trials. This study should make people think twice before doing that.

Making Links to Therapy

A poster from PowderJect Vaccines, a UK-US biotechnology company and vaccine manufacturer (Fuller) addresses in monkeys another very interesting clinical question, which is whether people who become infected with HIV after having received a vaccine that stimulates CTL responses will respond better or worse to subsequent therapy than positive people who did not receive such a vaccine.

The monkey study used DNA vaccination before exposure to a pathogenic SIV and antiretroviral treatment with PMPA (a version of tenofovir) started two weeks after infection. Some of the monkeys were given DNA vaccination while on treatment. What was shown was that the best response (i.e. lower viral load) was in monkeys vaccinated both before and after infection. The implication is that therapeutic immunisation will be more effective in people who already have a range of relevant CTL responses.

In this monkey study, the virus used to challenge was heterologous (i.e. had differences from) the vaccine strain, but there must obviously have been some level of similarity. The big question will always be: how similar must the vaccine and the infecting virus be, for vaccination to be of real clinical value?