Vaccine research: back to the drawing board?

This article is more than 20 years old. Click here for more recent articles on this topic

The major presentations on HIV vaccines at the Eleventh Conference on Retroviruses and Opportunistic Infections in San Francisco were dominated by one question. Is it reasonable to press ahead with large-scale clinical trials, such as the one now underway in Chon Buri and Rayong provinces, Thailand? Or should the emphasis and funding for vaccine research shift more strongly back to basic immunology and experiments in animal models?

In a plenary on Wednesday morning, Dr Susan Buchbinder of the San Francisco Department of Public Health set out the case for clinical trials. Without ongoing work with human volunteers, translating any positive findings that might emerge from laboratory work into clinical benefits would be long-delayed. Much still needs to be learned about patterns of risk behaviour among trial volunteers and potential vaccine recipients. There had to be a plausible case that a vaccine might work, but that couldn't be the only criterion on which to go ahead. Given present uncertainties, however, there might now be a stronger case for smaller exploratory trials in high-risk populations (so-called ‘Phase IIB’ trials) rather than large Phase III trials powered to detect low levels of efficacy.

The Thai trial has now recruited 1,000 of the 16,000 volunteers sought. The researchers – in Thailand and at the US National Institutes of Health - remain convinced that the canarypox candidate used (ALVAC vcp1521) is the ‘best in its class’. It will be years before other vaccine candidates have comparable safety and immunogenicity data, and none of the leading ones are based on Thai ‘subtype E’ strains of HIV. Although VaxGen’s AIDSVAX B/E is useless on its own, it might still perform in combination, for example, through antibody-dependent cellular cytotoxicity. (This hard-to-study mechanism may allow antibodies that are not otherwise effective against the virus to trigger the destruction of virus-infected cells.) There are positive spin-offs, for example, in local community involvement and mobilisation for HIV prevention. Above all, perhaps, the changing course of the Thai epidemic means that if this opportunity is missed, it may not be possible to run another comparable trial in the country.



A pill or liquid which looks and tastes exactly like a real drug, but contains no active substance.


In HIV, different strains which can be grouped according to their genes. HIV-1 is classified into three ‘groups,’ M, N, and O. Most HIV-1 is in group M which is further divided into subtypes, A, B, C and D etc. Subtype B is most common in Europe and North America, whilst A, C and D are most important worldwide.


A variant characterised by a specific genotype.



A molecule on the surface of some white blood cells. Some of these cells can kill other cells that are infected with foreign organisms.


A strategy of administering one vaccine dose (or one type of vaccine) to elicit certain immune responses, followed by or together with a booster, a second vaccine dose (or second type of vaccine). The prime-boost strategy may be used to strengthen the initial immune response or to elicit different types of immune response.

Dr Buchbinder referred frequently to a symposium held the previous afternoon, at which the outcome of the two VaxGen trials had been presented and discussed in some detail and in which two leading scientists outlined some of the problems with current vaccine candidates.

Dr Dean Follman from the National Institute for Allergy and Infectious Disease, Bethesda, presented an independent analysis of the VAX004 trial, which took place at 61 sites across the USA, Canada, Puerto Rico and Amsterdam, Netherlands. This study explored the possibility, based on an early ‘subgroup analysis’, that the vaccine might have been effective for a minority of trial participants, and in particular those who were non-white or, more specifically, black. Several possible explanations for such an association were considered, and carefully dismissed.

The most interesting association was between levels of antibody response to the vaccine and the risk of HIV infection. This clearly existed, among those who were vaccinated, yet people who received the vaccine were at no less risk of HIV than people who received the placebo. It seems that immune response to the vaccine is a marker for a natural ability to fend off infection, which was equally present among placebo recipients. (For the record, levels of antibody among black volunteers were only marginally higher than among white volunteers.)

On the other hand, using an experimental method it was clear that such an apparent association could very easily happen by chance, if an equally small subgroup – one in 12 (non-white) or one in 15 (black) were picked at random from the whole group of volunteers. This explanation is therefore the best available.

Dr Punnee Pitisutithum from Mahidol University, Bangkok, gave the first public scientific presentation of the data from the Thai AIDSVAX trial. This recruited its volunteers from among injecting drug users at clinics run by the Bangkok Metropolitan Authority, some of whom clearly continued to inject. Reported risk behaviour declined through the trial, though it wasn't clear that this led to a measurably reduced risk of HIV infection. The vaccine clearly had no effect, either on infection with ‘subtype E’ or subtype B viruses in this population. Retention of volunteers in the trial had been excellent and unwanted effects were the same in vaccine and placebo groups. Reported ‘social harm’ from vaccine trial participation was rare, and no different for vaccine and placebo groups – none had arisen from vaccine-induced positive antibody test results.

Dr Dennis Burton of the Scripps Institute in California, who is a leader in the study of broadly neutralising antibodies and approaches to designing vaccines to generate them, reviewed progress in that field. He presented results from work on antibodies directed against parts of gp41 which are exposed in the course of fusion and entry into a cell.

At present, the levels of antibody needed to provide protection appear to be high. The normal effect of antibodies, in Dr Burton’s view, is to ‘blunt’ infection rather than to prevent it altogether, and so questions will remain as to whether this can be enough to protect against HIV. There is no evidence for synergy between different antibodies (though combinations do seem to block more virus infections than single antibodies) or between antibodies given passively and cell-mediated immunity induced separately with a vaccine.

In discussion, one of the researchers whose work had been cited by Dr Burton argued that their research had shown that antibodies transmitted from mothers to babies prior to birth are protective in at least some cases against infection with the mother’s virus.

Finally, Dr Ronald Desrosiers from the New England Primate Research Center at Harvard University spoke on why an effective HIV vaccine is currently beyond our reach. Dr Desrosiers was one of the signatories to a recent letter in the US journal Science which challenged US funding for the current Thai vaccine trial.

Dr Desrosiers is concerned that the ‘most promising’ approaches now under consideration, such as the prime-boost systems currently favoured as lead products by Merck and IAVI, have failed to protect against a monkey virus, SIVmac239, even when vaccines are based directly on that virus. He thought this was a more realistic model for HIV transmission than the most widely-used SHIV strain, SHIV89.6P, since SIVmac239 uses the CCR5 co-receptor rather than the CXCR4 co-receptor used by SHIV89.6P. In principle, SIVmac239 should be an ‘easy’ test for a vaccine approach, not a difficult one. Yet even monkeys that show strong cellular immune responses and are challenged with a relatively low number of infectious doses of the virus show no evidence of protection.

At the end of his talk, he presented a study in which some monkeys treated with a live attenuated version of the virus had been protected against SIVmac239 – and others had not. Both groups included animals with and without detectable cellular immune responses. Until such results could be understood, the meaning of any supposed ‘correlates of protection’ seen in any human efficacy trials would be unclear. However, he conceded that he had not carried out an experiment, in the monkeys that had shown protection, to test what would happen if their CD8 cells were suppressed.

(His argument may also have been undermined by a study presented the next day in an oral abstract session by T Matano et al. Containment of SIV Replication in Rhesus Macaques by Vaccine-induced Cytotoxic T Lymphocytes. This reports that several macaque monkeys were, in fact, protected against a SIVmac239 challenge using as a vaccine a prime-boost DNA-Sendai virus recombinant system expressing a gag gene. In this case a high level of CTL response was clearly correlated with protection.)

A more positive perspective on the prospects for vaccines based on CD8 responses was given in a plenary talk on Tuesday morning by Simon Mallal of the Royal Perth Hospital and Murdoch University in Western Australia. Recent research, which he presented, has shown how HIV is shaped by the need to avoid cellular immune responses in each individual and population in which the virus is established. The implication is that careful selection of antigens could still make a difference to the ability of HIV to overcome a vaccine that induced a range of relevant cellular immune responses. Perhaps there is a scientific case for clinical trials, after all.

Further information

Preventive vaccines on aidsmap

AIDS Vaccine 2004 conference site

Webcasts from the 11th Conference on Retroviruses and Opportunistic Infections

Looking forward to 2004: HIV vaccine prospects (aidsmap news)