 |
|
|
| |
Infection With Resistant Virus More Than Doubles Risk of Virologic Failure
|
| |
| |
XVIII International AIDS Conference, July 18-23, 2010, Vienna
Mark Mascolini
Primary infection with HIV bearing one or more resistance mutations more than doubled the risk of first-line virologic failure in a 10,000-person cohort study [1]. Among people with one or more mutations but predicted by the Stanford system to be taking a fully active regimen, those taking a nonnucleoside (NNRTI)-based regimen rather than a protease inhibitor (PI) regimen tended to have a higher failure rate.
Antiretroviral treatment guidelines in Europe and the United States call for genotyping to uncover transmitted resistance mutations before a person starts antiretroviral therapy. But the standard genotyping assays used cannot detect resistance mutations that make up a small part of a person's viral population. This study tried to account for that lack of assay sensitivity by dividing cohort members into three groups: (1) people with no evidence of resistance mutations (as defined by the World Health Organization surveillance list), (2) people with one or more transmitted resistance mutations but predicted by the Stanford algorithm to be taking a fully active regimen, and (3) people with one or more resistance mutations predicted to be taking a regimen compromised by those mutations.
The 10,485 people studied came from four large cohorts (CASCADE, COHERE, EuroSIDA, and PENTA-EPPICC), and they could be any age. Everyone began combination antiretroviral therapy after January 1, 1998, was taking two nucleosides with either an NNRTI or a boosted PI, and had at least one plasma sample available for standard genotypic resistance testing.
Genotyping revealed that 953 people (9.1%) had been infected with virus harboring one or more resistance mutations. The Stanford system classified 476 people (4.5%) as receiving a fully active regimen despite their resistance mutation(s) and 477 (4.6%) as having at least low-level resistance to one or more of the drugs they were taking.
Defining virologic failure as two consecutive viral loads above 500 copies after 6 months of first-line treatment, the investigators devised a multivariate model to predict failure risk. That model considered gender, age, pretreatment viral load and CD4 count, year treatment began, previous AIDS diagnosis, HIV transmission risk group, and country of origin.
People with at least one transmitted resistance mutation and Stanford-predicted resistance to one or more drugs in their regimen had a 2.6-fold higher risk of virologic failure than did people who had no detectable resistance mutations (95% confidence interval [CI] 2.0 to 3.4, P < 10(-4). People with one or more resistance mutations but with a Stanford-predicted fully active regimen did not differ from people with no detectable mutations in risk of virologic failure.
Among people with one or more detectable mutations but taking a regimen predicted to be fully active, those taking an NNRTI regimen had a higher risk of failure than those taking a PI regimen. This risk difference fell just short of statistical significance (P = 0.08). That finding may mean some people with one or more detectable mutations before starting an NNRTI regimen had additional mutations undetectable by standard genotyping that rendered their regimen ineffective.
The investigators concluded that infection with resistant virus raises the risk of first-line virologic failure if the regimen is not tailored to be active against detected resistance mutations. They proposed that "in regions where genotypic testing is not routinely available but high prevalence of transmitted drug resistance is suspected, first-line regimens containing [a] boosted PI should be discussed."
Reference
1. Wittkop L, on behalf of the EuroCoord-CHAIN Joint Project Team. Impact of transmitted drug resistance (TDR) on virological and immunological response to initial combination antiretroviral therapy (cART) - EuroCoord-CHAIN joint project. XVIII International AIDS Conference. July 18-23, 2010. Vienna. Abstract THLBB108.
From Jules: after the talk Dan Kuritzkes commented at the microphone that results seen in NNRTI receiving patients confirm the fear that many clinicians have had that if you find resistance to one drug there may be undetected resistance to another drug and that giving a NNRTI in that setting would be giving an insufficiently potent regimen, your data very nicely supports that hypothesis", in response the presenter said "when transmitted drug resistance is present probably there are some minority mutations present that could cause some effect on the prescribed regimen".
Slides with audio are also available in english (click here to view it)
|
| |
|
|
|
|
|
