icon-    folder.gif   Conference Reports for NATAP  
  16th CROI
Conference on Retroviruses and Opportunistic Infections Montreal, Canada
February 8-11, 2009
Back grey_arrow_rt.gif
US and Canadian Studies Trace Impact of Resistance Results on Mortality
  16th Conference on Retroviruses and Opportunistic Infections, February 8-11, 2009, Montreal
Mark Mascolini
Resistance to antiretroviral therapy did not independently predict progression to AIDS or death in an eight-study AIDS Clinical Trials Group (ACTG) case-control comparison, probably because the cases had a significantly lower pretreatment CD4 count and a significantly higher pretreatment viral load [1]. However, a study of the British Columbia HIV cohort found a higher risk of death among people who did not have a resistance test before starting antiretrovirals and among those with whose physicians had less HIV experience [2].
Susan Swindells and ACTG coworkers devised a case-control study involving 104 antiretroviral-naive or experienced ACTG trial enrollees who had a new AIDS diagnosis or died during follow-up in the ACTG Longitudinal Linked Randomized Trials (ALLRT) cohort. They compared these 104 cases with 287 controls who did not die and who were matched to cases for age, gender, antiretroviral regimen, initial ACTG trial, and length of follow-up. ALLRT cohort members agree to extended follow-up after their initial ACTG trial ends. This analysis focused on progression or death at least 24 weeks after enrollment in the initial trial and excluded accidental deaths.
The ACTG team searched for resistance mutations with population genotyping in stored samples of ALLRT cohort members who participated in eight ACTG trials. (They did not report how many people were naive when they entered those trials and how many experienced.) Most case-control study members were men (87%), 50% were white, 25% black, and 21% Hispanic. Median age stood at 39 years. Cases had a significantly higher median viral load when they enrolled in their original trial (204,913 vs 56,885 copies, P = 0.03) and a significantly lower CD4 count (117 vs 235, P < 0.0001).
Median time to AIDS progression or death was 99 weeks. Only 6 of 104 cases died; the others had a new AIDS diagnosis. In a conditional logistic regression analysis, future drug options (defined in an earlier study [3]) were statistically equivalent in cases and controls (2.53 vs 2.69, P = 0.3). Neither did this analysis detect significant between-group differences in percentage with high-level resistance to nucleosides, nonnucleosides, or protease inhibitors (as defined by the Stanford algorithm). In addition, the groups did not significantly differ in mean number of classes with high-level resistance (0.58 cases, 0.48 controls, P = 0.5).
British Columbia provides universal free access to health services (from Jules: and in British Columbis bone dexas are recommended for HIV+). This analysis by Viviane Lima and colleagues involved 1820 antiretroviral-naive people who began antiretrovirals between January 2000 and June 2006 and were monitored until June 2007. Median age at study entry was 41 years (interquartile range [IQR] 35-48), median CD4 count 160 (IQR 70-240), and median viral load 100,000 copies (IQR 50,00000-100,000). While 26% of this population had a history of injecting drugs, 16% had AIDS, 47% started a ritonavir-boosted protease inhibitor, and 45% started a nonnucleoside; 61% had an adherence level above 95% during the first year of follow-up.
Before starting their first regimen, 36% of these people were offered a resistance test. During a median 38 months of follow-up (IQR 21-59), 68% of patients had no resistance test as a routine part of their care. By the end of follow-up, 15% of patients had died.
Women were 21% less likely than men to have a pretreatment resistance test, a difference that fell just short of statistical significance (odds ratio [OR] 0.79, 95% confidence interval [CI] 0.61-1.01). In contrast (and perhaps as a result of the pretreatment finding), women were 36% more likely to have a resistance testing during follow-up (OR 1.36, 95% CI 1.06-1.73). The mortality analysis showed a trend toward a higher death risk in women (OR 1.28, 95% CI 0.95-1.73).
Compared with people who started antiretrovirals in 2000-2001, those starting in 2002-2004 were twice as likely to get genotyped for resistance before treatment (OR 1.93, 95% CI 1.45-2.56), and those starting in 2005-2006 had more that a tripled chance of pretreatment resistance testing (OR 3.65, 95% CI 2.81-4.73).
Compared with people who began treatment with a CD4 count below 50, those who began with 50 to 199 CD4s had a 64% higher chance of getting a pretreatment resistance test (OR 1.64, 95% CI 1.23-2.18), and those with 200 to 349 CD4s had a 66% better chance (OR 1.66, 95% CI 1.22-2.25). Chances of getting tested for resistance before treatment did not differ significantly between the under-50 CD4 group and the over-350 group.
History of AIDS or injecting drug use had no impact on likelihood of pretreatment genotyping. But injecting drug users had twice the chance of getting an on-treatment resistance test than noninjectors (OR 2.16, 95% CI 1.74-2.69). On-treatment resistance testing became significantly less likely in 2005-2006 than in 2000-2001 (OR 0.42, 95% CI 0.33-0.54), perhaps because newer combinations were controlling HIV so much better than the 2000-2001 medleys. On-treatment genotyping rates did not differ significantly between 2002-2004 and 2000-2001.
Compared with people who had at least 95% adherence in the first year of therapy, those with 80% to 94% adherence had a 77% higher likelihood of resistance testing (OR 1.77, 95% CI 1.31-2.41), those with 40% to 79% adherence had almost a three times higher likelihood (OR 2.83, 95% CI 2.17-3.69), and those with under 40% adherence had a doubled likelihood (OR 1.91, 95% CI 1.41-2.61).
Pretreatment resistance testing had a profound impact on survival in British Columbia. The risk of death was about 40% lower in people who did get genotyped before therapy (OR 0.58, 95% CI 0.44-0.77). The investigators caution that this finding does not necessarily mean lack of pretreatment genotyping itself raises the risk of death, because failure to test before starting therapy may be a signal of less HIV experience by physicians. Indeed, analysis of physician experience per 100 HIV patients showed about a 15% to 20% lower risk of death among patients of physicians with more experience (OR 0.82, 95% CI 0.74-0.91 for pretreatment resistance testing; OR 0.84, 95% CI 0.75-0.94 for on-treatment testing).
Lima and colleagues believe their findings "suggest a close monitoring of physicians' practices to ensure all patients receive appropriate care."
1. Swindells S, Jiang H, Mukherjee L, Winters M, Bosch R, Katzenstein D. Virologic drug resistance is not associated with AIDS-defining events or mortality: an ACTG longitudinal linked randomized trials analysis. 16th Conference on Retroviruses and Opportunistic Infections. February 8-11, 2009. Montreal. Abstract 659.
(This poster is online at http://www.retroconference.org/2009/PDFs/659.pdf.) 2. Lima V, Eyawo O, Swenson L, Hogg R, Montaner J, Harrigan R. Drug resistance monitoring practices are strongly associated with mortality. 16th Conference on Retroviruses and Opportunistic Infections. February 8-11, 2009. Montreal. Abstract 666.
(This poster is online at http://www.retroconference.org/2009/PDFs/666.pdf.) 3. Jiang H, Deeks SG, Kuritzkes DR, et al. Assessing resistance costs of antiretroviral therapies via measures of future drug options. J Infect Dis. 2003;188:1001-1008.