icon-    folder.gif   Conference Reports for NATAP  
  1st International Workshop
on HIV and Women,
January 10-11, 2011
Washington, DC
Back grey_arrow_rt.gif
Striking Regional Antiretroviral Response
Differences Found in Canadian Women

  1st International Workshop on HIV and Women, January 10-11, 2011, Washington, DC
Mark Mascolini
Virologic response and rebound rates differed significantly in Canada's three most-populous provinces, even after statistical correction for certain factors that affect those outcomes [1]. Results raise questions about how demographic and other factors vary from one region to another in high-income countries, and how those variations may affect virologic response.
The study involved women starting their first antiretroviral combination after 1999 in three provinces--British Columbia (BC), Ontario, and Quebec. All study participants had at least one viral load measurement after starting treatment, and all women were at least 19 years old. The investigators did not describe proportions of aboriginal women (who make up a large part of the HIV population in BC) compared with women of African or Caribbean descent (who account for a large proportion of HIV-positive women in Ontario and Quebec).
CANOC is Canada's largest HIV cohort, collecting data from nine centers. Notably, the British Columbia contingent includes all antiretroviral-treated women in the province who met study entry criteria, while the Ontario and Quebec groups did not. Antiretroviral distribution and coverage plans also differ from one province to the next. The investigators defined virologic response as two consecutive viral loads below 50 copies. They defined rebound as a viral load above 1000 copies after an undetectable load.
The analysis included 874 women, 389 (45%) of them from British Columbia, 271 (31%) from Ontario, and 214 (24%) from Quebec. Study participants had a median age of 36 years (interquartile range [IQR] 30 to 43), median pretreatment CD4 count of 199 (IQR 110 to 290), and median pretreatment viral load of 50,000 copies. Ninety-one women (10%) had AIDS when they enrolled in CANOC, 45% started a nonnucleoside-based regimen, 280 (32%) started a ritonavir-boosted protease inhibitor (PI), 160 (15%) started an unboosted PI, and 42 (3%) started a nucleoside-only regimen.
Half of the women were seronegative for hepatitis C virus (HCV), 250 (29%) were HCV seropositive, and the rest had an unknown HCV status. HCV positivity was more prevalent in BC (44%) than in Ontario (13%) or Quebec (20%) (P < 0.001). A significantly higher proportion of BC women (41%) had a history of injection drug use than did women in Ontario (11%) or Quebec (14.5%) (P < 0.001). Women in BC were more likely to begin treatment with a nonnucleoside (49%) than were women in Ontario (41%) or Quebec (42%) (P < 0.001). Also, women in BC were significantly less likely to have fewer than three viral load tests yearly (31%) than were women in Ontario (40%) or Quebec (36%) (P < 0.001). Median follow-up time across the three provinces was 39 months (IQR 20 to 63).
Multivariate analysis considering history of injection drug use, pretreatment viral load, viral load testing rate, and initial third antiretroviral determined that, compared with women in BC, those in Ontario had a 25% better chance of reaching an undetectable viral load (adjusted hazard ratio [AHR] 1.25, 95% confidence interval [CI] 1.00 to 1.67, P = 0.048). Chances of viral suppression did not differ significantly between BC and Quebec in this analysis. The same multivariate analysis found almost a 50% better chance of virologic response among women who started with efavirenz than among those starting nevirapine (AHR 1.48, 95% CI 1.15 to 1.91, P = 0.002). Starting treatment more recently and starting with a lower viral load also favored virologic suppression.
A second multivariate analysis considering history of injection drug use, pretreatment CD4 count, rate of viral load testing, and initial third antiretroviral determined that, compared with women in BC, women in Ontario had more than a 60% lower risk of virologic rebound (AHR 0.38, 95% CI 0.21 to 0.067, P < 0.001) and women in Quebec had about a 50% lower rebound risk (AHR 0.51, 95% CI 0.30 to 0.87, P = 0.013). A history of injection drug use more than doubled the risk of virologic rebound (AHR 2.25, 95% CI 1.52 to 3.35, P < 0.001). Compared with women starting nevirapine, those starting lopinavir/ritonavir had about a 50% lower rebound risk (AHR 0.49, 95% CI 0.25 to 0.97, P = 0.039) but those starting nelfinavir had more than a doubled rebound risk (AHR 2.39, 95% CI 1.48 to 3.88, P < 0.001). Rebound risk did not differ between people starting nevirapine and efavirenz. Rate of viral load testing and year of starting therapy had no impact on rebound risk.
The researchers noted that defining differences between HIV-positive women from different regions "is an important first step in developing targeted prevention and management programs." The findings also underline the potential regional differences in antiretroviral response within the same country, even after accounting for variables such as injection drug use, pretreatment virologic and CD4 variables, and type of regimen. But the investigators pointed out that they did not have data on adherence, ethnicity, and pregnancy.
1. Cescon AM, Palmer AK, Chan K, et al. Regional differences in demographics, antiretroviral use, and virological response to therapy among HIV-positive women in a multi-site Canadian cohort. 1st International Workshop on HIV and Women. January 10-11, 2011. Washington, DC. Abstract O_02.