Antiretroviral therapy (ART) naive individuals with non B HIV-1 subtypes commencing treatment for the first time are suppressing viral load as quickly as individuals with subtype B, according to data derived from the UK HIV Drug Resistance Database presented to the Fourteenth Annual Conference of the British HIV Association (BHIVA) taking place in Belfast this week.
Surprisingly, this retrospective cohort study of more than 2000 patients found that individuals with subtypes A and C had a statistically significant shorter time to an undetectable viral load (defined as below 50 copes/ml) than those with subtype B, even after adjusting for baseline viral load and other variables, although the investigators did not consider this to be clinically significant.
Few studies have previously analysed responses to ART according to HIV subtype, and those that have, grouped together non-B subtypes when comparing them to subtype B. Given the increasing diversity of HIV-1 subtypes circulating in the United Kingdom – and globally – the investigators sought to assess whether response to first-line ART differed by subtype and did so by separately analysing the most frequently found subtypes in the UK.
The study included all ART-naive adults (over 16 years) with a baseline resistance test who were commencing ART (defined as at least three antiretrovirals) for the first time, and who had more than twelve months of follow-up data available. Individuals with intermediate or high-level resistance to one or more of their first-line ART drugs were excluded from the analysis.
In addition to CD4 cell count responses, time to an undetectable viral load and time to viral load rebound (defined as two consecutive measurements over 1000 copies/ml or one viral load measurement over 1000 copies/ml following a treatment change) were analysed.
Of 2116 eligible patients, 1550 (73%) were infected with subtype B; 272 (13%) had subtype C; 66 (3%) had subtype A; 57 (3%) had the recombinant virus CRF_AG; 41 (2%) had subtype D; and 130 (6%) patients had a variety of other subtypes.
A total of 89% of individuals with subtype B achieved an undetectable viral load by twelve months, with a median time to suppression of three months.
In contrast, 94% of those with subtype C and 97% with subtype A achieved an undetectable viral load by twelve months, with median time to suppression of 2.3 months and 2.1 months, respectively.
The other subtypes analysed had similar responses to subtype B: 89% with the recombinant AG subtype; 90% with subtype D; and 91% with other subtypes, achieved an undetectable viral load after twelve months, with a median time to suppression of 2.8 months, 2.6 months and 2.9 months, respectively.
In multivariate analysis, adjusted for age, clinical centre, calendar year, ART regimen, viral load and CD4 count, the investigators found that there was a statistically significant association between subtype and time to achieving an undetectable viral load, with more rapid suppression observed for subtypes A (Hazard Ratio 1.7; p=0.02) and C (HR 1.2; p=0.04) compared with subtype B.
However, Dr Linda Harrison of the MRC Clinical Trials Unit, presenting, said that although the differences were statistically significant, given that the difference between the fastest and slowest median time to suppression was a matter of weeks there appeared to be “little clinical difference” between the subtypes.
In multivariate analysis adjusted for baseline viral load and CD4 count, age, clinical centre, ART regimen, calendar year and time to initial undetectable viral load, of the 20% of individuals who rebounded from an undetectable viral load, those with subtype C appeared to rebound sooner than those with subtype B (HR 1.4; p=0.05).
However, after further sensitivity analysis that teased out the real possibility of adherence difficulties, the investigators concluded that this was “probably driven by adherence rather than [other causes of] virological failure.”
The investigators found no difference in CD4 responses between subtypes at three months after ART initiation (p=0.47) or subsequently (p=0.68), although individuals with subtype B had higher CD4 counts at baseline compared with individuals with other subtypes.
A notable limitation of this analysis was that adjusting for ethnicity or risk group could not be done due to their strong association with subtype. For example, 86% of individuals with subtype B were white and 90% were gay men, or other men who have sex with men, whereas the vast majority of individuals with non-B subtypes were of African ethnicity and/or heterosexual.
Dr Harrison concluded by saying that response to first-line ART appears to be excellent in individuals with all HIV-1 subtypes, but that since the confounding effects of ethnicity and HIV risk group could not be excluded, further studies are needed to explore the influence of ethnicity and risk group, as well as adherence.
Harrison L et al. Effect of HIV-1 subtype on virological and immunological responses to first-line HAART. 14th Annual Conference of the British HIV Association (BHIVA), Belfast. Abstract 10, 2008.