‘Therapeutic’ HIV vaccine leads to higher viral load, less time off treatment

In another blow to HIV vaccine development, a study published in the journal AIDS reports that a therapeutic vaccine given to people already living with HIV actually increased viral loads and reduced time off treatment - the opposite of what had been hoped. Higher doses of the vaccine - ALVAC 1452 - produced worse effects.

These results are especially puzzling and disappointing as studies with a similar therapeutic vaccine, sometimes combined with the immune booster interleukin-2, have had the opposite effect. In a study by the French vaccine pioneer Yves Levy, for instance, the viral load in subjects given the similar ALVAC 1433 vaccine ended up ten times lower than in subjects given placebo, and subsequent time off antiretroviral therapy (ART) was increased by 40% - see this report.

The authors of the current study say their results may be down to bad luck; a post-hoc analysis, for instance, found a higher proportion of “slow progressors” in the placebo group and of “fast progressors” in the vaccine group. But other initial analyses suggest that instead of stimulating lots of anti-HIV CD8 cells, which would kill HIV-infected cells, the vaccine mainly stimulated the formation of HIV-specific CD4 cells – which would only serve as targets for more HIV replication.

Whatever the reasons for this result, the authors compare it, in terms of significance, with the failure of the Merck ad5 preventative vaccine in the STEP study. Furthermore, ALVAC 1452 used a different vector – the shell of an inactive virus containing HIV proteins – from the Merck vaccine, which means that the failure of the latter may not be due to its vector.

The ALVAC 1452 vaccine consists of five pieces of HIV protein taken from the gp41, p55, protease, reverse transcriptase and nef components of HIV, wrapped inside the shell of a non-disease-causing canarypox virus.

In the study, 22 patients with CD4 counts of over 350 cells/mm³ were given four doses of ALVAC 1452 at weeks zero, four, eight and 20. A further 22 individuals were given three doses at weeks four, eight and 20. An additional 22 patients were provided with a placebo - eleven were given four doses, and eleven were given three doses.

Four weeks after the last dose of vaccine, 56 of the 66 patients stopped antiretroviral therapy (the other ten chose to stay on it). HIV treatment was then resumed if a patient’s CD4 count fell below 250 cells/mm³ or 50% below baseline, or if the viral load rose above 50,000 copies/ml. Subjects were followed until week 96 by which time 90% of those given four vaccinations were back on anti-HIV drugs, as were 71% of those given three doses. This compared to 59% of those given the placebo.

The vaccine produced a significant immune response to HIV in both dosing regimens. This involved a 40% increase in T-lymphocytes with anti-HIV immune responses in the patients given four doses of the vaccine and a 32% increase in the patients given three doses. There was no increase in T-lymphocytes with anti-HIV immune responses in the patients provided with the placebo.

The patients were well balanced in terms of age, sex, time since HIV diagnosis and lowest-ever CD4 count (CD4 nadir). In the final multivariate analysis, the only two factors associated with having to restart ART were CD4 nadir (60% higher chance of having to restart with a lower-than-average nadir, p= 0.002) and being vaccinated. Having three shots of vaccine increased the probability of antiretroviral resumption 2.7-fold and four shots more than fourfold (p=0.013).

But the results of the study may have been influenced by an important difference in the baseline characteristics of the patients. The last viral load taken before patients originally started HIV therapy was 125,000 copies/ml in those given four vaccine doses and 63,000 copies/ml in those given three, compared with 40,000 in placebo recipients. The researchers also found that three patients receiving placebo had HLA variants B27 or B57 in their genes, which confer slower progression to AIDS, and eight vaccine recipients had HLA B35, which is associated with faster progression.

However the researchers conclude that the immune response to the vaccine was at least partially responsible for higher viral production – and will be keen to find out why.