icon- folder.gif   Conference Reports for NATAP  
 
  4th IAS (Intl AIDS Society) Conference on HIV Pathogenesis, Treatment and Prevention
Sydney, Australia
22-25 July 2007
Back grey_arrow_rt.gif
 
 
 
Twice-Daily Maraviroc Compared to Efavirenz in Treatment Naive
 
 
  4th IAS Conference on HIV Pathogenesis, Treatment, and Prevention
July 22-25, 2007
Sydney, Australia
 
Mark Mascolini, Jules Levin
 
Note from Jules: the primary endpoint for this study was non-inferiority of Maraviroc compared to efavirenz as judged by both <400 and <50 copies/ml. The study found Maraviroc met the criteria for being non-inferior to efavirenz when evaluated by the <400 c/ml criteria but was not non-inferior when evaluated by the <50 copies/ml criteria. See below for exact data.
 
The primary endpoint for this is non-inferiority of Maraviroc to efavirenz as evaluated by both <400 copies/ml and <50 copies/ml. The study found Maraviroc was non-inferior to efavirenz based on the <400 copies/ml criteria but Maraviroc was not non-inferior by the <50 copies/ml criteria. The study stratified by Northern and Southern Hemisphere, and it appeared that Maraviroc did perform as well in the Northern Hemisphere countries but not in the southern Hemisphere Countries, and speculation why is below but Pfizer will examine this to better understand differences in responses between Hemispheres [1]. People randomized to twice-daily Maraviroc gained more CD4 cells than those randomized to efavirenz. Argentina, Australia, and South Africa represented the Southern Hemisphere countries. An analysis by country may identify if one particular country affected the analysis.
 
MERIT investigators recruited previously untreated people with CCR5-using HIV, a viral load at or above 2000 copies, and no evidence of resistance to efavirenz, AZT, or 3TC. They randomized enrollees to take 300 mg of maraviroc once daily, 300 mg of maraviroc twice daily, or 600 mg of efavirenz once daily, each with Combivir (AZT plus 3TC). An independent review board closed the once-daily maraviroc arm earlier last year, so this analysis compares only twice-daily maraviroc and efavirenz.
 
Treatment groups had similar pretreatment median CD4 counts (241 for maraviroc and 254 for efavirenz), and viral loads in the two groups both averaged close to 4.9 log (just under 100,000 copies). The study had two primary endpoints: proportion of patients with a viral load under 400 copies and proportion under 50 copies at 48 weeks.
 
Study statisticians determined that maraviroc would be noninferior to efavirenz if the lower bound of the 97.5% confidence interval around the 48-week difference between study arms was above -10%. That condition held true for the difference in proportion of people taking maraviroc versus efavirenz who reached a load below 400 copies (9.5%) (Table). But it did not hold for the slightly larger difference between proportions who attained a load below 50 copies (-10.9%). So the MERIT team concluded that maraviroc is not non-inferior to efavirenz by the 50-copy measure at 48 weeks.
 

Between-1.gif

Through 48 weeks of follow-up, more people had to quit maraviroc than efavirenz because of lack of efficacy (11.9% versus 4.2%), while fewer people stopped maraviroc because of toxicity (4.2% versus 13.6%). Overall, similar proportions discontinued maraviroc (26.9%) and efavirenz (25.2%).
 
CD4 gains averaged 170 with the maraviroc regimen through 48 weeks versus 144 with efavirenz. Although that difference proved statistically significantly, some may question whether it is clinically meaningful since the average CD4 counts after 48 weeks came to 411 with maraviroc and 397 with efavirenz.
 
Grade 3 or 4 adverse events were more common with efavirenz than maraviroc, though overall rates of grade 3 or 4 liver enzyme elevations proved similar in the two groups. More malignancies developed in people taking efavirenz than in those taking maraviroc. Overall rates of adverse events (91.9% maraviroc, 94.2% efavirenz) or serious adverse events (11.3% maraviroc, 12.7% efavirenz) were similar for the two treatment groups.
 
Two factors appeared to drive the worse sub-50-copy response in the maraviroc armÑhigher pretreatment viral load and southern versus northern hemisphere. As the following table shows, sub-50 rates with the two regimens proved similar for people who began therapy with fewer than 100,000 HIV RNA copies, but worse with maraviroc for people who started with a higher load. Similarly, 48-week 50-copy response rates proved almost equal with the two regimens in northern-hemisphere countries, but worse south of the equator.
 

response-2.gif

Michael Saag (University of Alabama at Birmingham), who presented the results, said investigators would try to figure out why southern countries did worse with maraviroc. He did not think adherence explains the difference, because the double-blind, double-dummy randomization should make adherence a nonissue. He suggested three possibilities for the difference--sicker patients in southern countries, different responses with different HIV-1 subtypes, and a difference in accuracy of coreceptor test results.
 
If statisticians had used a 12.5% bound on the confidence interval to rate noninferiority of the two regimens, as some other noninferiority trials do, maraviroc would have proved noninferior to efavirenz. Saag explained that regulators called for the 10% lower bound used in MERIT.
 
Reference
1. Saag M, Ive P, Heera J, et al. A multicenter, randomized, double-blind, comparative trial of a novel CCR5 antagonist, maraviroc versus efavirenz, both in combination with Combivir (zidovudine/lamivudine), for the treatment of antiretroviral naive subjects infected with R5 HIV-1: week 48 results of the MERIT study. 4th IAS Conference on HIV Pathogenesis, Treatment, and Prevention. July 22-25, 2007. Sydney. Abstract WESS104.