Bringing science and development together through news and analysis

  • Even failed experiments can be important ones


Disappointing results of clinical trials of a promising pair of new AIDS vaccines have highlighted the dilemmas faced by those determining strategy in this field. Fortunately, reactions on all sides indicate a willingness to use science as the principle basis of their decisions.

Two weeks ago, a prominent Kenyan newspaper carried a news story with the headline 'AIDS vaccine trials flop'. The following day, the same newspaper carried a second news story, this time with the headline 'Search for AIDS cure "not a flop"'. At first sight, the two headlines appear contradictory. In fact, between them they convey accurately the complex and sometimes tortuous path of progress towards what one leading authority in the field has called "the hardest challenge we've ever faced in terms of a vaccine."

Both headlines refer to the news, released to a conference in Lausanne, Switzerland, that a pair of vaccines intended to boost the body's immune response to the AIDS virus had failed to demonstrate a sufficiently promising effect in preliminary trials in Kenya, the United Kingdom and Uganda. The failure was all the more disappointing because the vaccines — based on the discovery that some sex workers in Nairobi appear to have a natural immunity to the disease — promised a defence against the subtype of the virus that is particularly prevalent in Africa.

The first news story was a straightforward account of the announcement of the outcome of the trials, describing the inevitable disappointment among both the researchers involved and others who had been hoping to see a significant step towards controlling a disease that continues to decimate much of the African continent. One blow to such individuals is that, as a result, the main sponsor of the research into the two vaccines — the International AIDS Vaccine Initiative (IAVI) — is expected to shift its support to other candidates.

The second story was based on an interview with one of first Kenyan volunteers to sign up for the trials. Pamela Mandela, a Kenyan doctor, pointed to the positive results it had provided, in terms of demonstrating the apparent safety of the potential vaccines. And she also emphasised the fact that setting up and carrying out the trials had provided several African countries, particularly Kenya, with both the experience and infrastructure that are vital for new trials of other candidates in the future.

Taken together, the two news stories capture an important truth about AIDS vaccine research: that progress in the field is going to be a protracted and iterative process as investigators discover slowly what does and doesn't work in people. This is especially true given that the ultimate goal is to develop a vaccine that is protective against HIV infection — and not merely successful in either triggering or strengthening a relatively low level of immunity.

Product of partnership

The two vaccines that failed to achieve this in the latest tests have been developed in a partnership between the University of Oxford (supported by the UK Medical Research Council), the University of Nairobi (supported by the Kenyan AIDS Vaccine Initiative), subsequently joined by the Uganda Virus Research Institute. Known as DNA.HIVA and MVA.HIVA respectively, they have been designed not to stimulate the production of antibodies to attack the virus — the standard approach of many vaccines — but to produce what called a 'cell-mediated response' to HIV.

This strategy is based on attempts to find ways of stimulating the human body to kill the cells that become infected with HIV to prevent further spread in the body. The first of the two vaccines, known as the 'prime', consists of a copy of one of HIV's nine genes stitched into a ring of DNA taken from a bacteria (the gene contains genetic sequences of subtype A of the virus, which accounts for most infection in East Africa). The second, the 'boost' vaccine, involves inserting the same gene into a modified vaccinia Ankara (MVA) virus, a smallpox virus that is unrelated to HIV.

The hope had been that, when delivered one after the other to enhance each other's effectiveness, the two vaccines would stimulate the body into rejecting the HIV virus. Such a result has already been successfully achieved in monkeys – one of the main reasons for the high hopes that they had raised. But tests carried out since 2000 on 439 volunteers in Kenya, South Africa, Switzerland, Uganda and the United Kingdom, have now shown a similar immune response in, at most, only 25 per cent of those receiving the vaccines. Furthermore, the effects appear to die away relatively quickly.

For the researchers involved, the result is clearly a significant setback. But they insist that it does not necessarily mean that the approach they are pursuing is doomed to failure. Rather, they are now seeing whether a different delivery technique — still using the same genetic sequence to produce the immune response, but with a modified delivery system — will prove to be any more effective.

There seems to be considerable evidence, for example, that the main problem may lie with the DNA vaccine, a conclusion that would correlate with similar disappointments elsewhere with vaccines that seek to promote immune responses by using disease-related genes inserted into 'naked' strips of DNA. Hopes remain that the MVA vaccine can be significantly improved. And both candidate vaccines are still being studied as a way of slowing down the progression of the disease in those already infected, i.e. as a therapeutic, rather than preventative, measure.

Pragmatic response

All this is as it should be. Developing an effective vaccine against HIV remains a major scientific, as well as technical, challenge. For a long time, researchers have been stressing that we understand very little about what is needed to provoke a strong response against HIV; what exactly is it in the genetic make-up of the Nairobi sex-workers, for example, that gives them their natural immunity to the virus?

One of the current major stumbling blocks is a lack of knowledge about what are called 'correlates of protection'. Animal studies do not provide sufficient data about this, making it even more essential to carry out preliminary human trials (as was being done with DNA.HIVA and MVA.HIVA) before even being able to decide on whether to move forward with a particular vaccine candidate.

Furthermore, even if the most recent trials have not come up to expectations, this has not dealt a fatal blow to the idea that the most promising strategy for tackling HIV is through triggering a type of white blood cell — known as CD8 T-cells — that can kill HIV-infected cells before the virus spreads further. This approach forms the basis of most second-generation vaccines now in clinical trials in North America, Europe, Africa and Asia. And the two vaccines for which disappointing results were reported in Lausanne represent only one possible way of achieving this.

Others may eventually prove to be more promising. This was the judgement that IAVI scientists had to make. The poor results reported in Lausanne appear to have made this decision relatively straightforward; the organisation has developed the rule that a candidate vaccine must stimulate immunity in at least 60 per cent of patients in a trial before being considered viable for further investment.

The DNA.HIVA/MVA.HIVA pair has clearly fallen well below this level (and even though the results reporting in Lausanne were only preliminary, there is little reason to believe that the final results will be any different). IAVI's decision to cancel new trials of the vaccines that it had been planning in Rwanda and the Netherlands is therefore just as sensible — given the substantial financial and resource commitments that would have been involved — as the decision by the scientists involved in developing the vaccines to continue to work on them.

Keeping the science in perspective

Both sides are aware of the ill-feeling that has been generated in the HIV/AIDS community by the VaxGen trials carried out in Thailand. In these, thousands of individuals were recruited into Phase 3 trials when there was already clear evidence that the level of anticipated success was unlikely to be sufficiently high to give any confidence in the vaccine. Another such trial is currently ongoing. Evidence of the commercial and political pressure that was applied for the trials to proceed, even in the face of strong scientific evidence challenging their likely value and suggesting that they would be a waste of time, has provided an important lesson for all those involved in this field.

Hopefully those engaged in the current round of trials will avoid any such trap. The price, of course, is that, by sticking to evidence and refusing to get carried away by unsubstantiated hopes, researchers will have to learn to live with disappointments that, given the nature of the threat they are seeking to combat, are always likely to be even more bitter than in other fields of science (where failed experiments tend to be seen as an inevitable part of the scientific process).

In the long-term, however, only a vaccine strategy that is guided by science in both its short- and medium-term decisions is likely to succeed. This may itself be an uncomfortable lesson for those involved to accept; African politicians who are demonstrating their commitment to combating the disease are among the front ranks of those who had been hoping desperately for a more positive result from the latest trials. But, when the price of going too far down the wrong track is so high, African governments are also coming to learn that even failed experiments can remain important ones.

We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit SciDev.Net — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on SciDev.Net” containing a link back to the original article.
  4. If you want to also take images published in this story you will need to confirm with the original source if you're licensed to use them.
  5. The easiest way to get the article on your site is to embed the code below.
For more information view our media page and republishing guidelines.