Yet another seemingly promising HIV vaccine has failed in clinical trials. According to Anthony Fauci, the head of the National Institute of Allergy and Infectious Diseases at the United States National Institutes of Health, which was conducting the trial, a vaccine is ‘essential to end the global pandemic’. But while the latest failure is a disappointment, it should come as no surprise.

To understand why, it is useful to go back to the beginning. Just over 12 years ago, two studies involving a vaccine candidate known as MRK-Ad5 were halted. The failure was comprehensive: the studies—STEP (which enrolled men and women in the Americas, the Caribbean, and Australia) and Phambili (including men and women in South Africa)—found that MRK-Ad5 failed to protect subjects against HIV infection. Worse, there was evidence that it may have increased the chances of acquiring HIV, the virus that causes AIDS.

The next year, however, hopes were again raised, as another clinical trial—known as RV 144—seemed to show modest positive results in Thailand. Yet the conclusion that the vaccine worked was based on somewhat dubious statistical analysis.

In fact, of the 125 (initially HIV-negative) trial participants who became infected with HIV (out of a total of 16,402), 74 had received the placebo and 51 had received the vaccine. That amounts to an efficacy rate of 31.2 percent—a good starting point, but nowhere near the level required to tackle a public-health challenge of the magnitude of Sub-Saharan Africa’s HIV epidemic.

A closer look at the results indicates that they are even weaker than they seem. As the former Harvard Medical School professor Ronald Desrosiers observed in 2017, the HIV-acquisition curve for the placebo group is nonlinear, with a sudden increase in acquisition among the placebo recipients within the first year of the trial.

That ‘anomalous’ increase accounts for ‘most or all of the difference in acquisition’ compared to the vaccine recipients. In other words, the fact that more placebo recipients contracted HIV had nothing to do with them being less protected than those who had received the vaccine.

Moreover, Desrosiers argued, if the vaccine did have a protective effect, the vaccinated individuals who became infected would have lower viral loads than their unvaccinated counterparts. That was not the case.

Nonetheless, some researchers considered the results promising enough to proceed with the recent South Africa trial, HVTN 702, which tested a vaccine candidate based on the RV 144 candidate, but adapted to the HIV strain that is most prominent in the country. After 18 months, 129 of the 5,400 participants who received the vaccine had become infected, compared to 123 of those who received the placebo.

This outcome was a major blow to the millions of Africans who hoped that researchers were finally approaching a long-term solution to the AIDS epidemic. But the results of the Thailand trial were never strong enough to justify such a large—and expensive—clinical trial.

This is not to say that trial results have to be overwhelmingly positive to merit further study. But, as Desrosiers also emphasises, spending hundreds of millions of dollars to manufacture and test products with little reasonable hope of efficacy is wasteful, at best. More comprehensive basic and preclinical research would enable scientists to identify approaches with a far better chance of success.

In the case of RV 144, a smaller intermediary trial could have enabled researchers to determine whether it was worth investing in another large-scale trial—at much lower cost. The leftover money could have been channelled toward other mitigation strategies for HIV/AIDS, including basic research.

Fauci is right: a vaccine is essential to end the HIV epidemic in Sub-Saharan Africa and beyond. But it is a long-term solution, and we don’t yet have the knowledge to merit highly expensive, large-scale clinical trials. Small-scale human trials, which can be scaled up once they have been convincingly shown to work, are simply more appropriate at this point.

In the meantime, vulnerable or infected people need other kinds of support, from education to treatment. As exciting as they may seem, pie-in-the-sky ideas—or even ‘intelligent’ guesses—are not sufficient reason to sacrifice the health and wellbeing of people today.

—Project Syndicate