CROI: Women infected with HIV subtype D progress faster despite same viral load

Analysis of women bar workers from a longstanding longitudinal study in Mombasa, Kenya has found that women infected with HIV subtype D had more than twice the risk of death than women with subtype A over six years of follow-up, according to findings presented at the Fourteenth Conference on Retroviruses and Opportunistic Infections in Los Angeles on Tuesday.

It also found a faster rate of CD4 cell depletion in women with subtype D, amounting to an average difference of 120 cells/mm³ after six years of infection.

No explanation has been found for the difference in apparent pathogenicity, though presenter Jared Baeten of the University of Washington, Seattle said that some other studies had found that subtype D more easily switched tropism to using the CXCR4 co-receptor than other subtypes.

The cohort of 1,597 initially HIV-negative women was followed from 1993 until 2004 when antiretrovirals became available. They were monitored monthly before seroconversion so that an exact date of infection could be ascertained. Their average age at study entry was 18. They were classed as sex workers because they supplemented their bar income with an average of one client per week.

During the study 218 of the women seroconverted, of whom 147 had blood samples genotyped to determine subtype.

Of these 114 (78%) had subtype A HIV-1 infection, which is the predominant type in east Africa; 21 (14%) had subtype D, which is predominantly found in Uganda and parts of Kenya; and 10 (7%) had subtype C, which is the predominant type found in southern Africa. Two clients had recombinant viruses which were excluded from the analysis.

All samples subtyped were collected within a year of infection, the average being 76 days post-infection – this was important in order to minimise possible cases of superinfection.

During the study 20 women with subtype A, seven with subtype D and three with subtype died. The researchers calculated that (in the absence of treatment) 55% of women with subtype A would still be alive 8.7 years after infection and that 50% of those with subtype D would be alive 7.7 years after infection.

Without controlling for viral load the risk of dying was 2.3 times greater for women with subtype D than subtype A. This was almost statistically significant. When viral load was controlled for, this effect was strengthened, with the relative risk of mortality increasing to 2.7. This achieved statistical significance (p=0.04).

Women with subtype C were also more likely to die; their relative risk of dying compared with subtype was 1.8, controlling for viral load. However because of the small numbers of women with subtype C this was not statistically significant (p=0.3, in other words there was a one in three chance this observation was due to random effects).

There was no significant difference in viral load in women with different subtypes; women with subtype A had an average viral load of 4.75 logs (55,000) and with subtype D 4.9 logs (80,000) when this was measured the first time after infection, and viral loads taken over time also showed no difference between subtypes.

However CD4 counts declined faster in women with subtype D, at a rate of 3.44 cells a month more in subtype D. This would lead, the researchers calculated, to an average CD4 count six years after infection of 249 in women with subtype D and 371 in women with subtype A.

Presenter Jared Baeten acknowledged that the study has limitations in the number of samples collected and in the fact that they only genotyped limited regions of the virus, so that the presence of other recombinants could not be ruled out.

He said they had found no explanation for why women with subtype D appeared to progress faster, but that one previous study had found that subtype D more easily shifted tropism so that it used the CXCR4 co-receptor on cells rather than CCR5.