Raltegravir continues to perform well at four years

People taking the HIV integrase inhibitor raltegravir (Isentress) continued to show good viral suppression with few side-effects at 192 weeks after starting treatment for the first time, researchers reported yesterday at the 13^th European AIDS Conference in Belgrade.

Jürgen Rockstroh from the University of Bonn presented findings from a pre-specified subgroup analysis of the Phase III STARTMRK trial. This study compared first-line therapy with 400 mg twice-daily raltegravir vs 600 mg once-daily efavirenz (Sustiva, Stocrin), both taken in combination with coformulated tenofovir/emtricitabine (Truvada).

The primary analysis at 48 weeks showed that raltegravir was comparable to efavirenz in overall efficacy but caused fewer side-effects, especially central nervous system symptoms. Raltegravir also produced slightly larger CD4 cell gains.

STARTMRK was designed to follow participants on blinded therapy for five years¾longer than most randomised trials. Exploratory analyses of patient subgroups were scheduled for weeks 156, 192 and 240. Participants were sorted according to demographic factors such as sex, age and race/ethnicity, HIV subtype (B or non-B), baseline viral load, CD4 cell count and hepatitis B or C coinfection.

The study randomly assigned 281 people to the raltegravir arm and 282 to the efavirenz arm. The two groups were well-matched for demographic factors and health status. Most (about 80%) were men and the median age was about 37 years. About 40% were white, nearly one-quarter were Hispanic, and the remainder were evenly divided between black, Asian and "multiracial". About one-quarter had hepatitis C and 7% had hepatitis B.

Participants had relatively advanced HIV disease. The mean baseline CD4 cell count was approximately 220 cells/mm³ and about 10% started treatment with CD4 cell counts less than 50 cells/mm³. They were about evenly divided between those who started with viral load above and below 100,000 copies/ml. About 80% had clade B HIV, the rest had non-B types.

A total of 58 people (21%) quit taking raltegravir during the study: five due to lack of efficacy, eight lost to follow-up and 13 due to adverse events. There were 85 discontinuations (30%) in the efavirenz arm: eight due to lack of efficacy, 17 lost to follow-up and 26 due to adverse events. This left 223 raltegravir recipients and 197 efavirenz recipients in the 192-week analysis.

In a non-completer equals failure analysis, 76% of raltegravir recipients and 67% of efavirenz recipients achieved viral load below 50 copies/mL. In an "observed failure" analysis - meaning people who discontinued therapy due to poor efficacy were counted as treatment failures, but those who quit for other reasons were excluded - the corresponding rates were 91% and 85%, respectively.

Proportions of people with undetectable viral load remained generally stable overall from about week 16 onward. But by week 192, raltegravir pulled ahead of efavirenz. Although this study was designed to evaluate non-inferiority rather than superiority, raltegravir appeared to work slightly better than efavirenz at four years (in statistical terms, the confidence interval for the difference in efficacy was above zero). CD4 cell gains baseline remained a bit higher in the raltegravir arm, 360 vs 301 cells/mm³ over baseline, respectively.

Response to treatment in the two groups did not differ according to sex, age, race/ethnicity, hepatitis coinfection status or HIV clade. There was no difference in efficacy in those people who started therapy with viral load above 100,000 copies/ml, but Rockstroh noted that raltegravir worked "especially well" for people with lower levels and suggested better efficacy was probably driven by this group. CD4 cell gains were also similar across subgroups.

Raltegravir continued to be well-tolerated during extended follow-up. The overall rate of drug-related side-effects was lower for raltegravir recipients compared with efavirenz recipients (50% vs 80%, respectively); there were few serious clinical adverse events or deaths in either arm. No signal of raltegravir liver toxicity was observed during long-term observation.

The researchers concluded that in the STARTMRK trial of previously untreated patients, raltegravir "demonstrated consistent virologic and immunologic efficacy" relative to efavirenz across groups with varying demographic and prognostic factors.

Rockstroh JK et al. Long-term efficacy of raltegravir (RAL) or efavirenz (EFV) combined with tenofovir (TDF) and emtricitabine (FTC) in treatment-naive HIV-1-infected patients: wk-192 subgroup analysis from the STARTMRK trial. 13^th European AIDS Conference, PS1/1, Belgrade, 2011.