icon-folder.gif   Conference Reports for NATAP  
  9th International Congress on Drug Therapy in HIV Infection
November 9-13, 2008
Back grey_arrow_rt.gif
Failure Rate Twice Higher With Second-Line Versus First-Line HAART
  9th International Congress on Drug Therapy in HIV Infection, November 9-13, 2008, Glasgow
Mark Mascolini
People starting second-line antiretroviral therapy after failure of first-line therapy at a London HIV clinic had a twice higher failure rate with the second regimen [1]. A higher CD4 count when the second regimen began lowered the risk of failure, and a higher viral load at that point raised the risk of second-line failure. Although this study from the Royal Free Hospital offers probably the most robust analysis of second-line failure to date, the second regimens analyzed date back to the early 2000s, so the results probably do not hold true for the diverse and potent rescue regimens available today.
The Royal Free analysis began with 1876 people who started combination therapy in 1996 or later and never took single- or double-nucleoside therapy. Defining virologic failure as a viral load above 400 copies after more than 4 months of continuous therapy, the investigators counted 483 virologic failures (26%). Of the 483 people with virologic failure, 166 (34%) met entry criteria for this analysis: (1) started one or more new protease inhibitors (PIs) or nonnucleosides and/or two different nucleosides, (2) had a viral load above 400 copies, (3) started second-line therapy from 1999 onward, and (4) had at least one follow-up viral load. The Royal Free team defined second-line failure as two consecutive viral loads above 400 copies more than 4 months after starting second-line therapy.
Comparing the first-line and second-line groups, Colette Smith and colleagues found equivalent proportions of men (74% and 74%), gay men (53% and 51%), and whites (59% and 55%). Median age measured 36 years in the first-line group and 37 in the second-line group. Median viral load was substantially higher before the first regimen (5.1 log) than before the second (4.3 log). Median (and interquartile range [IQR]) start dates for the first regimen were January 2002 (May 1999 to March 2004) and for the second regimen December 2002 (August 2000 to July 2004). Thus none of these people could have started a second-line regimen with tipranavir, darunavir, or any antiretroviral approved after those two PIs.
Kaplan-Meier analysis tracked equivalent rates of a discontinuation or switch of any antiretroviral during 24 months of first- or second-line therapy. Rates were also similar for discontinuation or switch of a PI or nonnucleoside. But 24 months after starting a first- or second-line regimen, Kaplan-Meier analysis estimated about a doubled risk of second-line compared with first-line failure.
Median CD4 count 1 year after starting second-line therapy stood at 374 (IQR 248 to 558), a 102-cell increase over the count when the second regimen began. But 19% of the second-line group had a lower CD4 count after 1 year of second-line therapy than when they started their second regimen, and at that point 12% had a CD4 count lower than their count before first-line therapy.
Every 100-cell higher CD4 count when second-line therapy began independently lowered the risk of second-line failure more than 25% (hazard ratio [HR] 0.73, 95% confidence interval [CI] 0.56 to 0.96, P = 0.03). The change in CD4 count from the pretreatment level to the level before starting second-line therapy did not independently affect virologic failure. Every 1-log (10-fold) higher viral load at the start of second-line therapy more than doubled the risk of virologic failure (HR 2.56, 95% CI 1.33 to 4.94, P = 0.005). But the viral load change from before treatment to before the second-line regimen did not affect virologic failure with the second-line regimen. Nor did having AIDS affect response to second-line therapy.
Compared with people who took only one new (previously unused) drug in their second-line regimen, those who took two new drugs had almost a 70% lower risk of second-line failure (HR 0.31, 95% CI 0.13 to 0.75, P = 0.01), and those who took three or more new drugs had almost a 75% lower risk of second-line failure (HR 0.26, 95% CI 0.12 to 0.60, P = 0.01).
Although rescue regimen have improved dramatically since people in this cohort started second-line therapy, these results support the clinical impression that several virologic, behavioral, and socioeconomic factors may favor a pattern of repeated treatment failure after first virologic failure.
1. Smith CJ, Lampe FC, Youle M, Johnson MA, Phillips AN. Treatment discontinuation and virological failure amongst HIV-positive individuals starting second-line combination antiretroviral therapy. 9th International Congress on Drug Therapy in HIV Infection, November 9-13, 2008, Glasgow. Abstract O333.