Has HIV got nastier?

HIV may have evolved into a more aggressive strain in the early years of the epidemic. Is it still happening? Asks Gus Cairns

A study published in this month’s Communicable Infectious Diseases journal looks, on the face of it, worrying. “Has HIV become more virulent?” it asks, and decides that it has.¹ What this means is that researchers found that, in people recently infected with HIV, the first CD4 count measured after diagnosis has over time declined by over 100 cells/mm³, from an average of 632 cells/mm³ in the late 1980s to 514 cells/mm³ today.

They also found that the proportion of people with a first CD4 count of below 350 cells/mm³ (and therefore recommended to start HIV treatment straight away) had gone up, from one in eight patients then, to one in four now. And the proportion with CD4 counts below 200 cells/mm³, – so immediately in danger of AIDS-related illness - had gone up from one in 50 patients to one in 20.

The implications, the authors said, was that HIV was learning to adapt to human immune defences. Might it get even better at doing so in the future? What does this imply for the figure of ten or so years often quoted as the time between HIV infection and developing AIDS - might it be getting shorter? And what would this imply for the number of people in the world who need to be on HIV therapy?

The problem with establishing if there has been a change in HIV’s effect on CD4 counts is that many other factors might influence it. One obvious factor is whether the interval between infection and diagnosis has changed. Another might be if the race, age or gender profiles of the patients studied have changed over time. Black people tend to have slightly lower CD4 counts for their stage of infection than white people; older people have lower CD4 counts than younger people; and women have lower viral loads than men. If the patient group had changed, so might the initial CD4 count. Different patient groups also tend to have different types of HIV: some, like subtype D, common in east Africa, seem to be more aggressive.

The study solved this problem by looking at a group of people who all had an HIV test when they became members of the patient group studied, and then had regular HIV tests at least every four years thereafter - namely US army recruits. A test is mandatory if you apply to join the US Army (and you won’t be accepted if you test positive). Serving soldiers are now re-tested every two years.

The researchers estimated the time of infection as the time midway between the last negative test and the positive test. They then took the first CD4 count measured after the positive test. The average estimated time between HIV infection and testing positive was estimated as 17.5 months and this actually declined to 16 months by the end of the study. So the lower CD4 counts observed were not due to later testing.

Relative to the years 1985 to 1990, the average CD4 count post-diagnosis was 65 cells/mm³ lower in the early 1990s and about 104 lower post-1996. Other immune indicators declined too. The CD4 percentage (the proportion of lymphocytes, a type of white blood cell, that are CD4 cells, often seen as a longer-term and less variable measure of immune fitness) also declined from 30% in the late 1980s to 27% post-1996. So did the other kind of T-cells, the CD8 cells.

The other factors associated with a lower initial CD4 count were a longer time between infection and diagnosis, a higher viral load, being non–white and being older. But even though, for instance, the proportion of black recruits increased over the time of the study, this could not explain the decline in CD4 cells seen.

So should we all worry that HIV has for some reason started producing faster declines in T-cells? Probably not. There was no significant difference in CD4 counts between soldiers diagnosed between 1997 and 2001 and those diagnosed since 2002. Whatever caused HIV to apparently get nastier in pre-HAART days, the start of HIV therapy seems to have halted that.

What might be going on? The researchers hypothesise that, in pre-treatment days, the ongoing battle between HIV and the body’s immune system caused the virus to evolve into a more virulent strain. More often the opposite happens: diseases become less virulent as they go along – one example is syphilis - because the mutations (changes) they have to make to escape the body’s vigilance make them less efficient at reproducing, just as some kinds of drug-resistance mutations do.

In the case of HIV, however, the opposite may have happened: only the fittest viruses produce high enough viral loads to make transmission frequent; it’s a kind of ‘bottleneck’ that only the most forceful squeeze through. However by bringing viral replication down to near-zero in more patients, HIV therapy has made it less likely that fitter strains of HIV will keep arising.

Although a number of studies have found similar results, not all have. One from Belgium2 measured HIV virulence in a very different way by taking varied samples of HIV from the early days of the epidemic and comparing their viral fitness with more recent samples by seeing how much virus was produced by test-tube cell cultures. It found that HIV had become less virulent over time. Other studies have found no change.

The recent study also, obviously, only looked at HIV in US patients. In another article,³ the authors of the Belgian study note that globally, HIV may be becoming a weaker virus because the strain that infects more than half the people in the world, subtype C, is in fact a more sluggish and less virulent type. This might partly explain some of the unexpected declines in HIV prevalence seen in some African countries. Another theory is that HIV has done a culling job on human beings, weeding out the most unfit first: declines seen in prevalence and apparent infectivity in Africa and elsewhere may be due to HIV picking off the most vulnerable members of the population first.⁴ The more robust specimens left may be better able to weather low CD4 counts.

What should we conclude from all this? HIV never ceases in its capacity to throw surprises at us, and because it reproduces a million times faster than human beings, it can change its nature very fast, as we have seen in drug resistance. So we can’t afford to be complacent. At present, however, there is no sign that a strain of ‘super’ HIV is waiting for us round the corner.