icon-folder.gif   Conference Reports for NATAP  
  HIV DART 2008
Rio Grande, Puerto Rico
December 9-12, 2008
Back grey_arrow_rt.gif
Dropout Rate in Study of US Minority Women Near 25% at 24 Weeks
  HIV DART 2008, December 9-12, 2008, Rio Grande, Puerto Rico
Mark Mascolini
Almost one quarter of antiretroviral-experienced women enrolled in the GRACE trial of darunavir/ritonavir dropped out of the study by week 24, but usually not because of side effects and in only one case because of virologic failure. GRACE (Gender, Race, and Clinical Experience) aimed to enroll minority women in the United States and Puerto Rico and succeeded in recruiting a population that is nearly two thirds African American and nearly one quarter Hispanic. The dropout rate suggests the difficulties in treating a poor, often marginalized population, even in a wealthy country. (from Jules: and the diffiulty of completing a study in the patient population, particularly when there are so many more treatment options available today as compared to years ago without having to remain in a study).
The primary endpoint of GRACE is the difference in virologic response (under 50 copies) between women and men at week 48. Enrollees could have hepatitis B or C infection if the condition was clinically stable and did not require treatment during the trial. GRACE excluded people who had already used darunavir, etravirine, enfuvirtide, tipranavir, integrase inhibitors, or CCR5 antagonists. All enrollees had taken a regimen including a protease inhibitor (PI) or a nonnucleoside for at least 12 weeks.
All study participants began twice-daily darunavir/ritonavir plus a background regimen selected by the investigator. The trial provided etravirine (the newest nonnucleoside), emtricitabine, tenofovir, fixed-dose tenofovir/emtricitabine, and zidovudine but did not mandate their use. Participants could not take enfuvirtide, tipranavir, integrase inhibitors, or CCR5 antagonists.(from Jules: perhaps some women dropped out to access raltegravir).
A planned 24-week interim analysis of this 65-site study involved 203 of 429 enrolled people who completed 24 weeks of treatment or discontinued therapy. The 24-week population included 180 women and 23 men. Their age averaged 42 years, 64% were African American, and 21% were Hispanic. Among women, pretreatment viral load averaged 4.67 log (about 45,000 copies) and median pretreatment CD4 count measured 206 (range 1 to 868).
Most women--107 of 180 or 59.4%--had already taken two or more PIs. Of 173 women whose background regimen was analyzed, 29 (17%) had no active background drugs, 45 (26%) had one active drug, and 99 (57%) had two or more active drugs, not counting etravirine. Fifty-nine women (33%) entered GRACE after a treatment interruption of 4 or more weeks.
At the 24-week point, 44 women (24.4%) and 2 men (8.7%) had dropped out of the trial. Fourteen women (32% of dropouts) and 1 man quit because of side effects, and 1 woman stopped because of virologic failure. The remaining dropouts were attributed to loss to follow-up (11 women, 25% of dropouts), protocol violations (5 women, 11%), withdrawn consent (5 women, 11%), and other reasons (8 women, 18%).
A time-to-loss-of-virologic-response analysis figured a 50.5% sub-50-copy response rate for women at week 24 (91 of 180 women). An analysis that excluded women who dropped out for reasons other than virologic failure figured a 65.5% sub-50-copy response rate (91 of 139 women). In a noncompleter-equals-failure analysis, the average CD4 count among women rose by 86 cells through 24 weeks. The investigators did not figure response rates separately for men because of their low number. Eighty-two of 203 people (40%) in this analysis took etravirine, but the investigators did not dissect responses according to etravirine use or number of active drugs in the background regimen.
Thirty women (17%) had a serious adverse event, including 8 with pneumonia. None of the 23 men in the interim analysis had pneumonia. The most common grade 2 to 4 adverse events possibly or probably related to study drugs in women and men were nausea (5.9%), diarrhea (5.4%), rash (3.0%), weight gain (3.0%), dizziness (2.0%), and dyspepsia (painful or difficult digestion) (2.0%). Rates of these problems did not differ between women and men. The most frequent grade 2 to 4 lab abnormality was hyperglycemia (in 16% of women and 19% overall) and total cholesterol (in 12% of women and 13% overall). Grade 2 to 4 AST elevations affected 6% of women (6.3% overall), and ALT elevations affected 4% of women (6% overall).
The investigators concluded that 24-week virologic response rates and adverse event incidence in GRACE generally reflect those in earlier trials of darunavir/ritonavir in treatment-experienced people. Although entry criteria included "ability to comply with protocol requirements," that did not prevent the 30 dropouts for reasons other than toxicity or virologic failure. Carmen Zorrilla (University of Puerto Rico) noted that the target population shoulders a heavy burden of sociologic problems and comorbidities such as mental illness. In their effort to recruit high proportions of women and minorities, trial planners sought out several sites with little experience in clinical trials. Perhaps that inexperience translated into a diminished ability to retain patients in the study. It would also be interesting to see how many people who signed up for GRACE during a drug holiday quit the study before week 24.
1. Zorrilla C, Currier J, Squires K, Bridge DA. Safety, tolerability and efficacy of darunavir/ritonavir in treatment-experienced women with HIV infection: interim analysis of GRACE (Gender, Race, And Clinical Experience). HIV DART 2008, December 9-12, 2008, Rio Grande, Puerto Rico. Abstract 34.