icon-folder.gif   Conference Reports for NATAP  


Back grey_arrow_rt.gif
First-Line Responses Similar With Integrase Inhibitors and NNRTIs in UK
  16th European AIDS Conference, October 25-27, 2017. Milan
Mark Mascolini
Virologic failure rates proved similar in antiretroviral-naive people starting an integrase inhibitor and in those starting a nonnucleoside (NNRTI) in a 2012-2014 analysis of 4938 people in the United Kingdom (UK) [1]. People starting a boosted protease inhibitor (PI) had an 87% higher risk of virologic failure than those starting an NNRTI.
In 2012 British HIV treatment guidelines added integrase inhibitors to the list of potential first-line agents, which already included NNRTIs and PIs. Researchers from University College London and other centers conducted this study to compare virologic failure rates in antiretroviral-naive people starting a combination including one antiretroviral in one of these three classes from January 2012 and June 2014.
The research team defined virologic failure as consecutive viral loads above 50 copies at least 6 months after starting treatment. Follow-up continued until a regimen change (unless involving agents within the same antiretroviral class), date of last clinic visit or death, or December 31, 2014, whichever came first.
The analysis included 4938 people from the UK CHIC observational cohort, 55.4% of whom started an NNRTI, 36.7% a PI, and 7.9% an integrase inhibitor. The NNRTI group included a higher proportion of men (83%) than the integrase group (79%) or the PI group (72%) (P < 0.0001). The integrase cohort had a higher proportion of whites than the PI or NNRTI groups (62% versus 56.6% versus 58.5%, P = 0.03). Age averaged 39 in the integrase group and 37 in the other two groups (P = 0.0018). Pretreatment viral load did not differ significantly between the three groups (median about 63,000 copies).
The most frequently prescribed antiretrovirals in the three groups were raltegravir (5.4% overall), efavirenz (51.3% overall), and darunavir (20.4% overall) or atazanavir (14.9% overall). The proportion of people using a first-line NNRTI dropped from 62% in 2012 to 44% in 2014. Over those same years, PI use rose from 34% to 41% and integrase inhibitor use tripled from 5% to 15%.
Through a median follow-up of 11.4 months (interquartile range 6.0 to 20.9), 386 of 4938 people (7.8%) had virologic failure. A similar proportion (7%) dropped out of regular care. Kaplan-Meier analysis figured a significantly shorter time to virologic failure with PIs than with integrase inhibitors or NNRTIs (P < 0.0001). Eighteen months after treatment began, about 10% of the integrase group and 10% of the NNRTI group had virologic failure, compared with about 20% in the PI group.
Multivariable Cox regression analysis determined that risk of failure with an integrase inhibitor did not differ substantially from risk with an NNRTI (hazard ratio [HR] 1.10, 95% confidence interval [CI] 0.59 to 2.04, P = 0.77). But risk of failure proved almost twice higher with a PI than with an NNRTI (HR 1.87, 95% CI 1.49 to 2.35, P < 0.0001). These analyses adjusted for HIV acquisition risk group, pretreatment viral load, nucleos(t)ide background, and year treatment began, but not race. In sensitivity analyses, the impact of treatment group on virologic failure remained significant with different virologic failure definitions.
The UK CHIC investigators suggested the higher failure rate with PIs may reflect adherence challenges that this analysis did not measure or control for. Another possibility is that early channeling bias resulting in use of integrase inhibitor in specific contexts favored that class in these years before wide use of integrase inhibitors.
1. El Bouzidi K, Jose S, Phillips A, et al. Integrase strand transfer inhibitors compared to other agents used in first ART regimens and the risk of virological failure in people living with HIV in the United Kingdom, 2012-2014. 16th European AIDS Conference. October 25-27, 2017. Milan. Abstract PE9/6.