Novel processes and adaptive design

To achieve this, we may need to think in terms of a series of vaccines to be developed and tested, in parallel and then against each other, possibly in combinations, and successively improved on over many years.

This would probably require changing from a sequential model of vaccine development, to a so-called adaptive design for trials.¹

In conventional drug development, candidate molecules are initially evaluated for biological activity, then screened for safety in animal trials and then in small Phase I trials involving tens of people. Candidates that remain are then moved into Phase II trials, usually involving a few hundred people, which establish potency and dose ranges and further evaluate safety. More recently, larger Phase IIb trials have been established that are not powered to demonstrate efficacy exactly, but which rule out candidates with little efficacy and indicate the likely range of efficacy in successful ones.

Finally large Phase III trials establish efficacy. In the study of drugs to treat illness, Phase III trials may require a few thousand patients or less, but prevention trials, where the majority of participants will remain uninfected, may need tens of thousands of participants to demonstrate an effect.

The US HIV Vaccine Trials Network has calculated that in a population with a 2% risk of HIV infection every year, it should be possible to get answers from a trial that recruited 5000 people over twelve months and followed them for another four years. This would show if the vaccine did, in fact, have an impact on HIV infection rates – and would be large enough to tell the difference between a vaccine that was 30% effective and one that was 60% effective. Two per cent a year, however, is typical only of highly at-risk populations, and trials would need to be bigger if conducted amongst the general population in all but the highest-prevalence countries.

In adaptive design, rules are pre-set for Phase II trials which allow for an early stop for the trial if an independent panel of experts, who have access to unblinded data, sees a strong signal of efficacy or a strong indication that there is no effect, or a negative one. They also allow for the trial to be adaptive to strengthen the possible efficacy observed. Such adaptations could include refining the selection of recruits, adding a booster vaccination, changing the vaccine formulation, or even vaccinating the placebo group to establish before-and-after efficacy.

The other benefit of adaptive trial design is that it may enable us to identify something so far sorely lacking in HIV trials design, namely correlates of protection. What this means is that, while we have a number of assays that can measure different kinds of immune response in the cells of people given the vaccine, we have almost no knowledge of what kinds of immune response actually translate into protection against HIV, other than a general indication that vaccines have to include protection against HIV’s envelope surface protein. This has meant that the candidates actually selected for the Phase III trials have so far essentially emerged by chance, governed as much by the funding capacity of individual research bodies as by rational selection. Adaptive trial design could allow us to correlate observed immune responses with efficaciousness much more quickly, thus narrowing down the selection of candidates to move into Phase III trials much more rationally.         

The adaptive and iterative process needed to develop an HIV vaccine will require:

• an unprecedented degree of co-operation and co-ordination between research teams, some of them commercial competitors;
• extended collaborations between countries with the technical resources to develop the vaccines and those with the largest populations affected by HIV;
• private sector expertise in manufacturing and production of vaccines, underpinned by public financial and legal guarantees where market mechanisms fail to secure the investment to take products forwards;
• community education and mobilisation to enable evaluation to proceed in an ethically acceptable way, with backing from governments and international institutions.

For the communities worst affected by HIV and AIDS, whose members must be involved in any programme to evaluate preventive vaccines, there are increasingly complex medical and social issues to be addressed. At a community level, any vaccine will have to be evaluated and used in combination with other treatment and prevention strategies. It is essential that this happens in ways that reinforce those other strategies and do not undermine them.