Men and women do equally well on HAART according to data from the EuroSIDA cohort published in the 1st April 2003 edition of the Journal of Acquired Immune Deficiency Syndromes (now available online).
Investigators compared virologic and immune response to HAART and clinical progression after commencing HAART between adult men and women.
The study included member of the EuroSIDA cohort, recruited from over 60 HIV treatment centres between 1994 and Spring 2001. Patients recruited to this study were naive to protease inhibitors or NNRTIs when they started a HAART regimen including at least two NRTIs and a PI or NNRTI. Patients were also required to have had a viral load of above 500 copies/ml and a CD4 count taken in the six months before entry to the study.
Study end-points were:
- Time to the achievement of a viral load below 500 copies/ml;
- Time to viral rebound (defined as two consecutive viral load measurements above 500 copies/ml;
- Time to achieving a a 100 cell rise in CD4 count;
- Time to progression to a first or new AIDS diagnosis;
- Time to death.
Investigators used a Cox regression analysis to provide hazard ratios comparing men and women. Both univariate and multivariate analyses were provided, the multivariate models adjusted for CD4 count, HIV viral load, previous AIDS defining illnesses, HIV treatment history, age, HIV risk group, race, haemoglobin level and geographic region.
Baseline characteristics
A total of 2547 patients (30% of the entire EuroSIDA cohort) were included in the study. Eighty percent were men, and the majority were white, although women were twice as likely to be be non-white as men (25% versus 10%).
Men were, on average, older than women (39 years versus 35 years) and had a lower average CD4 count (211 cells/mm3 versus 240 cells/mm3.Viral load was also higher at baseline amongst men and males were also more likely to have had an AIDS defining illness (26% versus 18%). Slightly more women (69%) then men (66%) had normal haemoglobin levels. Men were more likely to have never taken any antiretrovirals before (34% versus 29%). The geographic origins of both men and women were broadly similar.
HAART regimens
Similar proportions of men and women started on PI-based regimens (86%) and NNRTI regimens (14%). Amongst patients whose original regimen was PI -based, similar proportions of men and women added drugs, stopped therapy or swapped treatment. Men and women who changed from a PI-based regimen did so at a similar time.
There were however notable differences in the treatment experiences of patients taking NNRTI-based regimens, with nevirapine being the most usually prescribed drug (74% of patients). Amongst men 1.4% added a drug, 19.2% stopped and a little under 24% swapped a drug. This compared to 8.2% of women adding a drug, 39.7% stopping, and 19.2% swapping.
In both the PI and NNRTI treated patients women changed treatment sooner than males.
Outcomes
A comparable number of men (89%) and women (88%) achieved a viral load below 500 copies/ml on their first HAART regimen. In men the average time to achieving this was four months and in women, five months. On univariate analysis there was no difference in the hazard ratio between men and women for achieving this outcome. Factors found to be independently associated with an increased chance of becoming undetectable were higher baseline CD4 count, and later start of HAART. A previous AIDS defining illness and higher viral load were found to be independent predictors of not achieving an undetectable viral load. On multivariate analysis, women were found to have a non-significant 9% lower probability of achieving a viral load below 500 copies/ml.
Of the patients achieving an undetectable viral load, 33% experienced a rebound (31% men and 40% women). Rebound was associated with higher baseline CD4 count, being treatment naive, older age, and previous NRTI use. The adjusted hazard ratio showed that women had 17% increased probability of virological failure.
CD4 cell count increased by 100 cells/mm3 within nine months in both men and women. Having higher baseline viral load and CD4 cell count, and later start of HAART were all predictive of achieving an increase in CD4 cell count of 100 cells or more.
There were 201 new AIDS diagnoses during the period of the study. A little over 7% of women progressed to AIDS compared to 8.1% of men. Univariate analysis showed a similar rate of progression. A preexisting AIDS diagnosis, and higher HIV viral load were both predictive of new AIDS defining illnesses occurring after treatment with HAART. In multivariate analysis the hazard ratio for women developing AIDS was found to be slightly increased.
During the study, 133 patients died (4.7% of females and 5.4% of men). Again, univariate analysis showed that men and women had the same risk of death, but adjustment reversed and women had an adjusted hazard ratio of 1.15. Higher viral load and pretreatment with NRTIs was predictive of death.
In a discussion of their findings the investigators note that “despite females having lower viral loads, higher CD4 counts, and less likelihood of having an AIDS diagnosis at the time of HAART, there was no evidence of significant gender difference in subsequent virologic outcomes or clinical events.”
They suggest that the higher CD4 counts and lower viral loads seen in women may confer no therapeutic advantage.
The investigators note some possible limitations for their study. No evidence on adherence was gathered. Women changed therapy sooner than men, and as dosing requirements for anti-HIV drugs are not gender specific, the suggest that gender differences in blood volume and body-mass index may effect the efficacy, tolerance and toxicity of HAART. The investigators also note that no socioeconomic or educational information was collected. This has been shown in some studies to be associated with survival, and the investigators note that “gender differences in health-seeking behaviour and access to care and treatment would almost certainly impact on outcomes”.
Although the EuroSIDA investigators conclude that even larger cohort studies are needed to provide a definitive answer to the impact of gender on HIV during the HAART era, “the weight of evidence suggests that there is no significant effect of gender on virologic, immunologic, or clinical outcome.”
Further information on this website
Moore AL et al. Virologic, immunologic, and clinical response to highly active antiretroviral therapy: the gender issue revisited. JAIDS 32: 452-461, 2003.