Double trouble: South African study on dual HIV infections highlights superinfection risk

Theo Smart
11 August 2003

http://www.aidsmap.com/news/newsdisplay2.asp?newsId=2239

Researchers from the University of Cape Town have found further evidence suggesting that patients who become infected by more than one strain of HIV prior to seroconversion (antibody formation) are more likely to progress rapidly to AIDS. Earlier this year, at the Tenth Conference on Retroviruses and Opportunistic Infections in Boston, researchers from the University of Washington reported similar findings in four individuals with dual HIV infection who had progressed to AIDS or death within two years of infection. One of those patients had been from South Africa.

A researcher from the University of Cape Town, Mr. Jandre Grobler, presented data on four dual infections from a South African cohort in an oral session on Monday at the South African AIDS Conference in Durban.

Dual infection can be a consequence of either co-transmission (infection with two strains at the same time or very close to each other before seroconversion) or superinfection (when a second HIV infection occurs in an already infected person). The viruses are not always by different HIV subtypes. The two viruses may in fact be from the same subtype, though genetically highly divergent from each other.

“Detection of dual infection is becoming more common, especially in regions where multiple subtypes are circulating,” said Mr. Grobler. “Furthermore,” he added, “dual infection commonly leads to recombinant stains and the high numbers of unique recombinant forms of HIV in circulation suggest that dual infection and recombination may occur frequently.”

To determine the incidence and natural history of dual HIV infection, researchers recruited a cohort of 32 women sex workers who had all been participants in a Phase III study of nonoxynol-9 (Van Damme et al., Lancet 2002; 360: 971-7.). The sex workers were recruited at five truck stops between Johannesburg and Durban. All had received monthly monitoring during the microbicide study. This was a very high-risk cohort. In an average week prior to seroconversion, the women serviced about 15 clients — 59% of whom were reported to be positive. The women reported using condoms only about 20% of the time (Ramjee, G et al. 2001. MRC News 32:14-15.)

Twenty-eight of the women enrolled in the dual infection study within at least 6 months of seroconversion. Populations were tracked over time using the heteroduplex-tracking assay to determine the diversity of their viral populations. Data for changes in viral load, CD4 counts and viral diversity are available for 22 of the women out to 24 months.

Dual infections were detected in 12% of the cohort (4/32). In these four, dual infection occurred prior (or very shortly after) to seroconversion, and thus most likely represent cases of co-transmission rather than superinfection).

After dual infection, the viruses recombined to form new hybrid viruses in each of the dually infected patients. However, their respective viral populations displayed different evolutionary patterns. In one patient, the recombinant strain became dominant, in two others the dominant population varied, while in the fourth patient, the recombinant strain persisted as a minority strain. At present, the significance of these findings is unknown.

During primary infection, people who are infected with a single strain of HIV, the virus tends to evolve a variety of mutant populations during the first few months until the immune system recognises HIV and brings it under control (at setpoint), leaving a viral population that is once again more homogenous. However, each of the four dually infected women continued to have viral populations that were more diverse than those in the rest of the cohort.

This viral diversity was associated with an inability to achieve a low viral load setpoint. While the majority of infected women were able to establish a low viral load set point by month 12, the four dually infected women all had significantly higher viral loads than had the other subjects, ranging between ~30,000 and ~500,000 copies (p=0.0056). Such viral loads are rather high for women, and high viral loads at setpoint (or the inability to ever achieve a setpoint) have consistently been associated with faster rate of clinical progression to AIDS or death. Accordingly, in these four, one has already begun to progress clinically, while in the others both viral load and diversity continue to increase.

Dr. Jim Mullins, who spoke on viral diversity at the Conference on the following morning, and was one of the authors on the dual infection study at the Tenth Conference on Retroviruses as well, believes that these data support the University of Washington’s earlier study.

He pointed out that it isn’t clear “whether becoming dually infected causes greater damage to the immune system that leads to faster progression or whether people who are dually infected are genetically predisposed to be unable to control viral diversity. That defect may be the same reason why they become dually infected in the first place.”

However, known genetic factors predisposing patients to rapid progression have yet to be noted in these patients. Furthermore, if these patients do have a specific immunologic defect that predisposes them to becoming infected by more than one virus, why haven’t the researchers observed three, four or even more strains multiply infecting a patient? Why stop at two strains?

Although this study was in women sex workers, if further data support the findings, it may have important implications for other types of high-risk behaviour. For example, individuals who become infected during unsafe ‘sex binges’ with multiple partners (at a party, club, online or via internet dating over a weekend) could potentially have a very different clinical outcome than someone who becomes infected during the occasional unsafe sexual encounter.