Determining the feasibility of utilizing the microbicide applicator compliance assay for use in clinical trials

Assessing the Reporting of Adherence and Sexual Activity in a Simulated Microbicide Trial in South Africa: An Interview Mode Experiment Using a Placebo Gel

“…There has been some concern expressed about the precision of the applicator test in measuring gel insertion because of a reliance on the naked eye for identification. But even if accuracy in this study were lower than the 93-95% in the clinical trial site validation study in South Africa or the 97% in the controlled clinical site validation study at a hospital in New York [28], the expectation is that it would be similar across interview modes, thereby not affecting the estimate of the difference between modes in reported and tested applicators.…”

“…1 Granular Food Dye. The applicator test, which has been shown to be effective at detecting insertion regardless of whether applicators have undergone long term storage and/or been extensively handled, is reported to have 97.5% sensitivity and 96% specificity [27,28]. The presence of semenogelin, a protein produced by seminal vesicles that causes seminal fluid to form a coagulum subsequent to ejaculation in the vagina indicates that a woman has been exposed to a partner's ejaculate in the previous 48 h [29] and thus is a biomarker of sex without a condom.…”

Adherence in the CAPRISA 004 Tenofovir Gel Microbicide Trial

“…While returned applicators are a more direct, objective measure of gel use in the previous month, this method is still dependent on participant behavior (31) and does not account for sexual frequency, possible gel dumping or that participants could forget to return applicators. To overcome some of these limitations, data from self-report were combined with the objective applicator count data to obtain a more accurate applicator-based adherence measurement (primary adherence measure) for the preceding 30 days.…”

Microbicide safety and effectiveness: an overview of recent clinical trials

“…There were 3.3 HIV infections per 100 person-years (134 new cases) among women randomized to Carraguard, compared with 3.7 per 100 personyears (151 new cases) on placebo, a difference that did not reach statistical significance in the ITT analysis. Although 96% of participants reported using a gel with last sex, researchers utilized a staining method of verifying vaginal applicator use [39][40][41] and calculated that only 10% of women used the study gel 100% of the time, 20% used it on around 75% of occasions, and 30% used it on less than 25% of occasions [42]. Overall, the researchers estimated that only 44% of sex acts involved use of study gel.…”

“…Trials of microbicides have until now mostly relied on self-reported adherence estimates, which are influenced by recall and social desirability biases [21 ]. More objective measures are being developed [21 ] to corroborate self-reports of risk behavior and product use, including biomarkers of exposure to semen [27 ,72] such as prostate specific antigen and Y-chromosome DNA testing and an assay to evaluate whether applicators have been exposed to the vagina [39][40][41]. With many of the newer ARV-based microbicides, adherence can be monitored by measuring ARV levels in genital tract tissues [27 ], but the utility of this method will depend on factors such as product half-life and timing of last dose and study visit.…”