Acceptability of a Dapivirine Gel Administered Rectally to HIV-1 Seronegative Adults (MTN-033)

Abstract

Rectal microbicide (RM) researchers have increasingly been interested in using antiretroviral (ARV) drugs in the form of a sexual lubricant as a form of HIV prevention. We triangulated quantitative and qualitative assessments to evaluate participants’ acceptability of 0.05% dapivirine gel administered via two separate modalities (a rectal applicator and an artificial phallus for use as a coital simulation device) in two different randomization groups: Sequence A (Applicator/Coital Simulation Device) or Sequence B (Coital Simulation Device/Applicator) as part of a Phase I trial (N=14) among a group of men who have sex with men (MSM) randomized using a 1:1 ratio from the Microbicide Trials Network (MTN)-033 trial. Overall, participants reported favorable acceptability of the gel (n=11; 78.6%) the same or more at the end of the study compared to when they started the study. Additionally, when discussing their preferred administration modality, they noted that both methods have positive qualities but also potential areas of improvement. Our findings underscore the need to create multiple delivery methods for a future microbicide gel (i.e., with and without the need for an applicator) and highlight the importance of offering MSM choices in how biomedical HIV prevention strategies are delivered.

INTRODUCTION

Background

In 2018, 69% of all new HIV infections in the United States resulted from sexual contact among men who have sex with men (MSM)(Division of HIV/AIDS Prevention, 2021). Researchers and advocates have called for both short- and long-acting biomedical prevention options that would enable individuals to choose the most suitable behaviorally congruent prevention method. Rectal microbicides (RM) are one of the biomedical strategies being developed to reduce the risk of HIV or other sexually transmitted infections (STIs)(McGowan, 2014). Since the onset of the RM development agenda, researchers have explored the safety and efficacy of RM formulated as gels given that they could serve as a sexual lubricant and be most behaviorally congruent with anal intercourse practices (Carballo-Dieguez et al., 2008) (Cranston et al., 2017). However, drug concentrations in various RM gel candidates used in Phase I trials to date have not always achieved blood and tissue levels thought to be adequate for protection against HIV (based on in vitro testing).

In most RM trials to date, participants have been asked to insert a RM candidate using an applicator, as opposed to administering it as a topical lubricant, to ensure accurate and standardized amounts of the administered RM product. Unfortunately, RM applicators – whether originally designed for vaginal use or adapted for rectal use – have encountered mixed acceptability due to their design, portability, and assembly(Bauermeister et al., 2016a; Carballo-Diéguez, Giguere, Dolezal, Bauermeister, Leu, Valladares, Rohan, et al., 2014) and have confounded participants’ overall evaluation of the RM product being tested within these trials. In a RM trial among young MSM, Bauermeister et al.(Bauermeister et al., 2016b) found that participants’ willingness to use the gel in the future decreased if they reported greater dissatisfaction with the applicator over a 12-week period of use, regardless of how acceptable they perceived the gel to be as a product. Taken together, these findings suggest that the need for an applicator as a method of delivery might hinder a RM’s goal to serve as a behaviorally congruent product.(Frasca et al., 2017) At present, however, it remains unclear whether participants’ acceptability of a RM product would differ if its mode of administration were akin to a sexual lubricant (i.e., use of the gel directly onto a phallus or rectum without an applicator).

Researchers have posited that it may be difficult to achieve sufficient drug levels in rectal tissue and systemic circulation when applying a RM gel as a sexual lubricant. Therefore, MTN-033 was designed as a Phase I, randomized, open label, single-site RM trial to evaluate the pharmacokinetics (PK) and behavioral acceptability of dapivirine (DPV) gel when administered into the rectum using two distinct modes of delivery: as a pre-filled applicator (2.5g of DPV gel), and as a sexual lubricant (up to 10g of DPV gel) to be used with an artificial phallus, referred to as a coital simulation device (CSD), as a way of simulating receptive anal intercourse (RAI).

In this study, we describe MTN-033 participants’ behavioral acceptability of the DPV gel and their experiences applying the gel with a pre-filled applicator and as a sexual lubricant with a CSD. Informed by Morrow and Ruiz’s framework(Morrow & Ruiz, 2008) to assess microbicide acceptability in Phase I clinical trials, we describe RM acceptability across three domains: characteristics of the RM vehicle (e.g., formulation, texture and viscosity), features related to the RM application process (e.g., need for an applicator, ease of use, portability), and experiences after use (e.g., behavioral congruence, side effects). These domains have been linked to individuals’ willingness to use an effective microbicide in the future, further reinforcing the value of examining factors linked to product acceptability during early stages of clinical testing. Finally, we compared whether participants’ acceptability of the DPV gel differs between the two administration modalities (i.e., pre-filled applicator vs. sexual lubricant used with a CSD).

METHODS

Sample

Sixteen HIV-uninfected MSM and transgender women who have sex with men over the age of 18 were enrolled and randomized into one of two application sequences: 1) applicator then CSD or 2) CSD then applicator. Data collection took place between May 2018 and November 2018 in Pittsburgh, Pennsylvania. Participants were recruited through clinic registries, LGBTQIA+ community outreach events, and word of mouth.

Inclusion criteria included: 1) men and transgender women ages 18 years old or older; 2) able and willing to provide written informed consent; 3)HIV-1 or HIV-2 uninfected; 4) able and willing to provide adequate locator information; 5) able to return to all study visits and willing to comply with participation requirements; 6) in general good health at screening and enrollment visits; 7) have a history of consensual RAI at least once in the past year; 8) willing to not take part in other research studies involving drugs, medical devices, genital or rectal products, or vaccines for the duration of study participation, including the time between screening and enrollment; 9) willing to abstain from inserting any non-study products into the rectum for 72 hours prior to and after each study visit. The study was reviewed and approved by the Institutional Review Board (IRB) at the participating site.

Study Procedures

Participants were screened for eligibility prior to enrolling in the study. Participants returned to the clinic within the 30-day screening window (Visits 1 & 2) where they completed administrative, behavioral, and clinical and laboratory procedures. Additionally, clinical results or treatments for urinary tract infections, reproductive tract infections, sexually transmitted infections (UTIs/RTIs/STIs) or other findings were completed as clinically indicated at all visits.

Participants were randomized either to Sequence A (Applicator/CSD) or Sequence B (CSD/Applicator) and asked to use either the applicator (Figure 1) or CSD (Figure 2) at enrollment (Visit 2). For applicator administration, participants were administered approximately 2.5g (2.5 mL) of 0.05% DPV gel via a HTI pre-filled applicator into their rectum. For the CSD phase, participants were supplied with a CSD, weighing dish containing DPV gel, and other study supplies (gauze, moist towelette, gloves). Participants then applied the DPV gel to the CSD and/or anus in the way they would during sex. Then participants inserted the CSD and moved the CSD in and out (one cycle) at one cycle per second for five minutes. Once finished, participants disposed of the study supplies in a bag that was handed to the clinic staff.

Figure 1:

Study applicator for rectal gel administration

Figure 2:

Coital simulation device (Doc Johnson ®) used during rectal gel administration

Participants received verbal instructions regarding the two study sequences during the consent and study counseling process. During Visit 3, participants were administered their first dose of the study gel, returning for a follow-up visit the next day (Visit 4). After a minimum 14-day washout period, participants had their second dosing of the product (Visit 5) and follow-up visit the following day (Visit 6). At these visits, participants completed administrative, behavioral, clinic and laboratory procedures, including intensive collection of biological samples for PK, pharmacodynamics (PD) and safety analyses. Visit 7 served as the follow-up safety contact and termination visit where participants completed administrative and regulatory procedures as well as received clinical results or treatment for UTIs/RTIs/STIs or other findings. Participant reimbursement was based on local guidelines and approved by the local IRB prior to study implementation.

Participants completed a baseline computer assisted self-interview (CASI) during their enrollment visit (Visit 2) and a brief CASI on product acceptability and adherence during dosing visits (Visits 3 and 5). For the purposes of this report, we include data from participants’ baseline questionnaire (e.g., sociodemographic characteristics) and their follow up visits questionnaires (e.g., product acceptability indicators). Additionally, participants completed an in-depth video interview (IDI) with research staff after completing each mode of application (Visits 3 and 5). Participants were allowed to skip any survey question or IDI prompt that they did not wish to answer.

Survey Instruments

Sociodemographic characteristics.

Participants completed a baseline CASI where they self-reported sociodemographic characteristics, including age, sex assigned at birth and current gender, race/ethnicity, sexual identity, relationship status, employment status and educational attainment (see Table 1).

Table 1.

Descriptive statistics for evaluable participants in MTN-033 (n=14)

	M(SD)	N(%)
Age	29.43 (9.73)
Race/Ethnicity
Asian		1 (7.1%)
African American		2 (14.3%)
White		9 (64.3%)
Latinx		2 (14.3%)
Sexual Identity
Gay		11 (78.6%)
Bisexual		3 (21.4%)
Primary Sex Partner		3 (21.4%)
Employment
Full-time employment		5 (35.7%)
Part-time employment		9 (64.3%)
Highest Educational Attainment
High school		1 (7.1%)
Partial college		5 (35.7%)
College graduate		3 (21.4%)
Graduate school degree		5 (35.7%)
Currently enrolled in school		5 (35.7%)
Sexual Behavior (prior 8 weeks)
Total Number of Partners (range 0-4)	1.57 (1.02)
Any Condomless Insertive Anal Sex		3 (21.4%)
Any Condomless Receptive Anal Sex		2 (14.2%)
Lubricant Use
Frequency of lube for RAI
Never		0 (0%)
Sometimes		1 (7.1%)
Always		13 (92.9%)
Lubricant Type (check all that apply)
Saliva		0 (0%)
Silicon-based		4 (28.6%)
Water-based		14 (100%)
Oil-based		0 (0%)
Hybrid		1 (7.1%)
Satisfaction with Current Lubricant
Very Dissatisfied		1 (7.1%)
Dissatisfied		0 (0%)
Neither		2 (14.3%)
Satisfied		7 (50.0%)
Very Satisfied		4 (28.6%)
Lube Consistency
Very Liquid		0 (0%)
Somewhat Liquid		0 (0%)
No preference		5 (35.7%)
Somewhat thick		2 (14.3%)
Very thick		7 (50.0%)
Lube region
Directly on your partner’s penis		11 (78.6%)
Around your anus		13 (92.9%)
Inside your rectum		8 (57.1%)
On the outside of the condom		13 (92.9%)
Use of sex toys to stimulate yourself prior to receptive sex		11 (78.6%)
Use of toys to apply lubricant		5 (35.7%)
Times of re-application
Never		1 (7.1%)
Once		7 (50.0%)
Twice		3 (21.4%)
3 times or more		3 (21.4%)
Lubricant interrupt sex?
It does not interrupt sex		3 (21.4%)
It interrupts sex but does not bother me		10 (71.4%)
It interrupts sex and bothers me		1 (7.1%)

Sexual practices.

Participants were also asked to report the number of sexual partners with whom they had sex in the prior eight weeks. Additionally, participants were asked about their sexual behaviors related to oral sex, RAI, insertive anal intercourse, and insertive vaginal intercourse (if applicable).

Lubricant Use.

Participants were asked to describe their self-reported lubricant use. Items included frequency of commercial sexual lubricant use for RAI, reasons for not using a commercial sexual lubricant, type of lubricant use, satisfaction with their current sexual lubricant, lubricant characteristics (e.g., flavor, color, scent, viscosity, and dispensation method), application characteristics (e.g., location of application, who applies the lubricant, how the lubricant is applied, reapplication frequency, and how bothersome and/or disruptive reapplication is).

Product Acceptability.

As part of their follow up visit surveys and their IDIs, participants were asked a series of prompts aligned with Morrow and Ruiz’s microbicide acceptability framework.(Morrow & Ruiz, 2008) Participants were asked about vehicle-related characteristics (e.g., feeling of gel inside their body), application-related characteristics (e.g., process of application, ease of application, application problems experienced), and use-related characteristics (e.g., ease using the gel, problems using product, perceived side-effects post-use). Participants also shared their overall acceptability of the gel, recommendations for improvement, and whether their acceptability of the gel had changed since the beginning of the study.

Future Use.

At Visit 5, participants were asked to indicate their likelihood of using a rectal gel microbicide in the future relative to condoms. Participants were asked “which would you prefer to use—the study gel, the male condom, or both?” and could answer using one of the following responses: “Gel”, “Condom”, “Neither”, “Both”, and “No preference”.

Qualitative Exit Interview

After completing their first and second dose administration visit (Visit 3 and 5) behavioral CASIs, participants completed a video IDI with trained qualitative interviewers located at a partner research institution in a different city. Only the audio portion of the IDIs was recorded. Interviewers had prior experience conducting qualitative interviews, were trained in the study-specific procedures, and completed mock interviews prior to the start of the data collection.

In preparation for the IDIs, the interviewers examined participants’ CASI data across Visit 3 and 5 to identify domains that might require clarification during the IDI. A total of 31 interviews were completed (one participant did not attend visit 5) using a semi-structured interview guide that explored participants’ general study experiences, gel perceptions, gel acceptability, modality experiences, partner preferences, and anal sex experiences. Acceptability probes explored general impressions of the gel, perceived ease or difficulty of insertion and use, physical and emotional discomfort attributed to the gel, lubrication expectations, perceived ability to discuss the gel with a sexual partner, likelihood to use gel to prevent HIV, gel recommendations, and factors that would either prevent or encourage gel adherence. The audio was transcribed verbatim, de-identified, and checked for accuracy. Interviews lasted an average of 45 minutes (range = 15 to 69 minutes).

Data Analysis

We used IBM SPSS, version 27, to compute descriptive statistics (e.g., means and standard deviations, counts and percentages) from the CASI data. Of the 16 participants enrolled, one participant discontinued the trial after visit 4 and another completed the applicator-specific behavioral survey by mistake in both visits. Therefore, in the analysis, we included participants who completed the study and completed both the applicator-specific and CSD-specific CASI surveys. We present data on these 14 participants in this analysis.

We compared participants’ gel acceptability between the two forms of administration (e.g., pre-filled applicator vs CSD) using paired samples t-tests for continuous endpoints and McNemar test for binary endpoints. We did not impute a score for item-specific missing data; where appropriate, we note the subsample included in the analysis.

We used thematic analysis(Braun & Clarke, 2006) to synthesize the major domains explored in the IDI based on Morrow and Ruiz’s framework(Morrow & Ruiz, 2008), triangulating participants’ survey and qualitative data across these domains where appropriate. In preparation for qualitative analyses, the research team developed a codebook guided by the major areas of interest and probes from the interview guide. The codebook included code names, explicit definitions, and inclusion and exclusion criteria to ensure coding accuracy. To validate and finalize the codebook, three researchers independently coded three transcripts and discussed coding discrepancies to reach consensus. The codebook was then updated as needed for clarification. All 30 transcripts were coded independently by two members of the team, who then met to discuss the codes assigned to each transcript. Any discrepancies between coders were resolved with input from a third coder to reach consensus. Analysis was facilitated by Dedoose, version 7, an online web application for qualitative data analysis. Illustrative participant quotes, alongside gender identity, age, and study arms are included in italics below.

RESULTS

The mean age of the 14 evaluable participants was 29.4 years of age (SD=9.73). Most participants identified as non-Hispanic White (n=9; 64.3%), followed by African American (n=2; 14.3%), Latinx (n=2; 14.3%), and Asian (n=1; 7.1%). Most of the participants completed a college education (n=9; 57.1%) and had full-time (n=5; 35.7%) or part-time (n=9; 64.3%) employment.

Most participants identified as gay (n=11; 78.6%). Three participants (21.4%) reported having a primary sex partner at the beginning of the study. At baseline, participants reported having an average of two sexual partners in the prior eight weeks (Mean [M]: 1.57; Standard Deviation [SD]=1.02). Three participants (21.4%) reported engaging in condomless anal sex (see Table 1). Prior to participating in the study, 11 participants (78.6%) reported having used a sex toy to stimulate themselves prior to engaging in RAI. Five participants (n=5; 35.7%) reported using sex toys as a way to apply lubricant to the rectum prior to engaging in RAI.

All but one participant reported always using lubricant prior to RAI (n=13; 92.9%), with participants noting that they had used water-based (n=14; 100%), silicone-based (n=4; 28.6%), and/or hybrid-based (n=1; 7.1%) lubricant. Most participants reported feeling satisfied (n=7; 50.0%) or very satisfied (n=4; 28.6%) with their current lubricant. More than half of participants reported preferring a lubricant with a somewhat thick (n=2; 14.3%) or very thick (n=7; 50.0%) consistency; the remaining participants reported no preference (n=5; 35.7%) or preferred a more liquidy consistency (n=1; 7.1%). When asked where they applied the lubricant, most participants noted lubricating around their anus (n=13; 92.9%) or on the outside of the condom (n=13; 92.9%), directly on their partner’s penis (n=11; 78.6%), and/or inside their rectum (n=8; 57.1%). Most participants reported having to re-apply lubricant at least once (n=7; 50.0%), two times (n=3; 21.4%), or three or more times (n=3; 21.4%) during sex. More than half of participants self-reported that lubricant application interrupted sex (n=11; 78.5%), with only one participant noting that this often bothered him (n=1; 7.1%).

Gel as Lubricant Acceptability

Participants were asked to rate how much they liked the gel overall after participating in the study. Participants self-reported liking the gel (M=4.07, SD=.73) at the end of the study using a 5-point scale (1=Disliked very much; 5=Liked very much). Most participants liked the gel the same (n=11; 78.6%) or more (n=3; 21.4%) than when they started the study. Most participants reported willingness to use half of the product amount required via an applicator (n=12; 85.7%) or as a lubricant (n=13; 92.9%). Similarly, most participants reported willingness to use double the amount of product required via an applicator (n=10, 71.4%) or as a lubricant (n=12; 85.7%). No differences in willingness to change product amounts were observed between administration modalities.

Participants overwhelmingly (75%) said that there was little to no sensation of the gel after insertion. The temperature, smell, and consistency were the major characteristics that were commented on. After insertion with the applicator, most participants noted no major sensations of the gel inside of their bodies. Eight participants said that they could not feel the gel. One was surprised about not feeling anything after seeing the amount to be applied beforehand. Of the remaining seven, most said that the product felt like other lubricants that they have used in the past. One study participant said that the gel was, “a pretty standard lube.” Most had a positive experience, with one participant noting that they did not experience any cramping or feeling of fullness. Another said they felt a slight pressure in their lower abdomen after gel insertion.

The temperature of the gel was also noted by a few participants. While the gel was administered after being in room temperature, five participants (35.7%) noted that the gel felt cold during application, which caused some minor discomfort, but that the coldness went away after about a minute. Two of these participants acknowledged that the cold feeling of the gel could have been attributed to the clinic’s cool temperature and air conditioning system.

Three participants (21.4%) suggested changes to the gel’s consistency after using the applicator and the CSD, respectively. During their interviews, participants discussed how the gel’s viscosity affected their overall opinions of the product. Six participants said that they liked the consistency, saying, “it was like, a really good lube,” and it “was just perfect.” Of the six who held positive views, they explained that the thickness of the gel made it less prone to run and make a mess. The gel was not too “liquidy;” however, other participants remarked how the thickness of the gel was not a positive trait. Five participants negatively described the gel as “chunky” or “goopy” during application and on their bodies. As one participant noted,

“The gel definitely used alone, isn’t really…I didn’t find that it was a good lubricant. I don’t know if that’s maybe because the consistency or just about the makeup of the gel in general. Even though it was using a dildo, if you were using this with another person, then I don’t know that it would be comfortable on either end. …Like I don’t know if it’s because how thick it is or maybe just the composition of it, but it was a little uncomfortable, not painful, but it was a little uncomfortable, so.” (Gay Male, Non-Hispanic White, 47 years old)

The remaining three participants held neutral opinions toward the gel. They said that it was neither great nor terrible.

“It’s been… it’s been okay. It wasn’t, it’s not the best, it’s also not the worst. For me, the worst would be if it were runny, like water, and just too sticky. This wasn’t like that at all.” (Gay Male, Non-Hispanic Black, 34 years old)

A few participants noted some leakage after inserting the gel with the applicator (n=3; 21.4%) and the CSD (n=2; 14.3%). Others also noted some soiling after inserting the gel with the applicator (n=1; 7.1%) and the CSD (n=1; 7.1%), or diarrhea after inserting the gel with the applicator (n=2; 14.3%) and the CSD (n=1; 7.1%). There were no significant differences in perceived side effects observed between modes of application.

When asked about the gel’s potential as an HIV prevention tool, participants reported that they would prefer using the gel over condoms if found effective in the future (n=6; 42.9%) or using both concurrently (n=6; 42.9%). One participant preferred the condom (n=1; 7.1%), while another had no preference (n=1; 7.1%).

Applicator Acceptability

Participants perceived that the insertion of the gel with the applicator was an overwhelmingly positive experience. In their surveys, all participants noted liking the applicator. Most participants felt that the applicator was a good size for inserting the gel, with most participants noting that the applicator’s finger grips were useful for administering the gel (M=8.62, SD=1.98) using a 10-point scale (1=Not useful at all; 10=Very useful). Participants also commented that the width and length were ideal because the applicator was comfortable and did not go too far in the anus. One participant said that they liked the size but would not make it any bigger. Several participants commented that they did not feel staff inserting the applicator because it was small. Additionally, several participants felt the applicator would be portable and easy to use because of its size. Over half of the participants (n=9; 64.3%) perceived the applicator as easy to carry. Most participants (n=13; 92.9%) reported wanting a disposable applicator in the future if it were made available. As one participant (Gay Male, Non-Hispanic White, 36 years-old) commented:

“I: So talking a little bit about gel insertion can you tell me about your experiences inserting the gel today?

P: Uh sure. That was done um flawlessly (laughter). I mean it was pretty simple. I kinda liked the fact that it looked like it was almost travel size in nature. You could definitely like take with you if you were going out, on a date, or just even having around the house so I mean the sample’s size from what I what I saw looked kinda cool.”

In addition to ideal size, many participants commented that the applicator seemed simple and easy to use. Participants mentioned that the finger markings made the applicator easier to handle and prevented it from slipping in their fingers, helping to deliver a pre-portioned amount of gel. Most participants agreed that the design was straight forward enough that there was not a lot of room for human error in the application process.

“I also wanted to add one thing. I don’t know if you can ask about this later, but I’m assuming, well obviously those ones are disposable. I like that they’re disposable because it gives you a pre-portioned amount so you don’t have to worry about people who might not be putting enough in so I think it’s good that they’re disposable and not refillable. But I mean I guess you could have a fill line or somethin’ but I don’t know —I feel like it just takes most of the human error out of it.” (Gay Male, Non-Hispanic White, 23 years-old)

Although most of the feedback related to the applicator was positive, eight participants provided negative feedback. A few participants said that using the applicator felt unnatural. One participant said the applicator was “sterile looking,” while another said it looked like a tampon. As one participant said, the tubular aspect and the “cross” that pushes the gel out is very clinical and detracts from sexual excitement. Similarly, another participant said the applicator looks very medical/scientific and makes it feel like a medical procedure or like getting an enema. One participant said that it would not feel natural to insert the gel daily and they would rather use the gel as a lube.

Coital Simulation Device (CSD) Acceptability

While no participants reported CSD-related problems in their survey responses, participants provided mixed feedback about the CSD in the IDIs. Most participants had neutral or positive feelings about the CSD and felt that it was relatively easy to use. Most participants relayed they felt comfortable using the CSD with the gel as a lubricant, and the CSD was similar to dildos they had used in the past.

“It went pretty well. I would say that the girth of the dildo was a little bit larger than what I was used to so that was a little bit different and took some getting used to, but overall it went pretty well.”(Gay Male, Non-Hispanic White, 21 years-old)

Although clinic staff discussed the purpose of using the CSD during the consent process, several participants (n=6) did not recall that using the CSD was supposed to simulate an actual sexual encounter. Instead, they believed that the CSD would be the method of application if the gel was effective in the future. Once the interviewer clarified the purpose of the CSD during the interviews, participants noted that if the DPV gel were to be used in place of lubricant, the gel may be just as effective. However, as noted below, several participants noted that using a CSD to simulate sex felt behaviorally different than sex with a partner, and may affect how people perceive the gel’s potential as a lubricant.

“So if it was like a real person, I could do it. I feel like I could put sprinkle the lube on or apply the lube, and hop on the piece, but it just felt a little, as far as the dildo was concerned and the and the exercise of simulating sex, it felt medicinal. It didn’t feel, psychologically, it didn’t feel like sex to me – it felt like a homework assignment.”(Gay Male, Non-Hispanic Black, 34 years-old)

Participants also expressed fear and concern about the amount of drug that would be required or would end up being absorbed if used in place of a sexual lubricant. A participant mentioned that they felt the CSD was not as precise as the applicator, and that the CSD spread more gel on the outside around the anus. Additionally, while the CSD provides a more poignant and intense sensation, it takes longer to fully insert and apply the gel appropriately.

“…but I wasn’t really sure but I think I put enough—feels fine. I was thinking: should I apply more on? But I was like, no I don’t think so. I think it there was enough. That was one question I had: how much is enough? How much is not enough? Is it more is just about using up the lubricant? I wasn’t really sure. But otherwise it was fine, I was doing it, and then it was in there, and I was like, ‘Great.’”(Gay Male, Non-Hispanic Asian, 34 years-old)

Preferred Method of Administration

Participants were asked to rate their perceived ease of use of the gel after each administration using a 5-point scale (1=Very difficult; 5=Very easy). Paired comparisons of perceived ease did not differ across application method (Applicator: M=4.64, SD=.50; CSD: M=4.43, SD=.65, t(13)=1.39, p=.19). After using the study gel with both methods, participants were asked to report which method of application they preferred during their IDIs. Seven participants (50%) said that using the gel as a lubricant was their preferred method of application. There were several reasons for their preferences. Many said that they preferred use as a lubricant because using the applicator would be too troublesome.

“I felt like, if I was actually having intercourse using the gel for the applicator, I would kinda have to stop whatever we’re doing and you know and either fill the applicator or use it. It would just be like an uncomfortable break in the sexual activities. Using it as just lubricant was easy, simple. It was something I knew how to do.” (Gay Male, Latino, 19 years old)

Participants that preferred using an applicator to administer the study gel offered a variety of reasons for their preferences. One of the most common reasons was that the applicator would be quicker. Additionally, one participant said they felt the applicator was more precise than the gel as lubricant and that the applicator has more potential for discretion, so it can be used in a wider variety of situations. A few participants said that the applicator’s small size makes it easier and preferable to insert. Also, many participants said that they could more easily integrate the applicator into their usual sex preparation.

“…and, that would be a really great incentive for me to take care of myself, protect myself if I knew that I could keep that in my pocket until it’s time and throw it away. There’s nothing that anybody sees. There’s no mess. It’s just very easy one and done.” (Gay Male, Non-Hispanic Black, 39 years-old)

Two participants acknowledged that the application method would depend on which was most effective, but that they would prefer using the gel as regular lube.

“I mean the gel as a lubricant worked fine on the dildo. I would have to say whichever is the more effective way…I don’t know if you’re inserting the gel inside of you, you’re still using a lubricant beyond that for sex, so… how that works between the two, I’m not sure. You’d still have to use something else. You wouldn’t just say, “Okay, I’m gonna put this the lubricant inside me and that’s all we’re going to use.” I don’t think that the gel would suffice for that.”(Gay Male, Non-Hispanic White, 45 years old)

One participant showed no preference toward the applicator or gel as lubricant They said that the applicator was easy to use, but it might be hard to time their use of it in the future relative to the time of potential exposure, so using the gel as a lubricant may be better.

“I don’t know if I could make an actual preferential statement. I like the applicator because it’s really easy. But depending on whether you need to dose yourself within like a certain window or something before potential exposure, the gel as a lubricant might be easier. But, but you know, you never know. I don’t know how it will be used as like an actual medication. If it becomes an actual medication…I can’t really say like my choice right now, but I mean if I had to choose and they were both easy to use, I would probably choose the lubricant rather than the applicator.” (Gay Male, Non-Hispanic White, 23 years old)

DISCUSSION

Assessing acceptability together with determining the clinical safety profile of products is increasingly recognized as a crucial step in early phases of microbicide trials. These data can facilitate decisions related to the product’s formulation, its delivery system, or mode of use prior to further safety and efficacy trials. (Carballo-Dieguez et al., 2008; Carballo-Dieguez et al., 2007; Carballo-Diéguez, Giguere, Dolezal, Bauermeister, Leu, Valladares, Rohan, et al., 2014; Gorbach et al., 2014; Mantell et al., 2005; van der Straten et al., 2012) Achieving consistent and correct use among the product’s consumers will require researchers to develop products perceived to be desirable and acceptable, both within the context of clinical trials and in the real world. As a secondary endpoint of the trial’s findings, we employed a mixed-methods analysis to understand participants’ acceptability of two delivery modalities of the DPV gel. Consistent with Morrow and Ruiz’s framework(Morrow & Ruiz, 2008), we ascertained key factors associated with product acceptability, including gel-specific traits (e.g., satisfaction with gel, formulation side effects), applicator-specific characteristics (e.g., ease and comfort of use, ease carrying applicator), and contextual factors associated with its use (e.g., sexual satisfaction when RM is used in anal sex). Overall, our findings suggest that participants perceived the DPV gel acceptable across both modes of administration (i.e., a pre-filled applicator and as a lubricant inserted while using a CSD). In fact, participants noted that any awkwardness related to the trial’s application methods was secondary to their potential contribution to advance an efficacious RM product.

Participants expressed liking the DPV gel equally or more than they expected after using it in the study. In discussing the properties of the gel, participants often compared the DPV gel to their current lubricant product. These comparisons underscore the need for a RM product that resembles or supersedes current lubricant products in the market as a strategy to optimize behavioral congruence between the RM and participants’ lubricant use experiences. For example, many participants noted that their current sexual lubricant is thinner than the DPV gel. Participants noted that a thin lubricant often led to reapplying it several times during their sexual encounters. In reflecting about the DPV gel, some participants noted that the study gel’s viscosity and thickness might alleviate the need to reapply lubricant during sex and help reduce the feeling that they were interrupting sex. From a drug development perspective, these findings highlight the need to develop and test a RM gel that can be administered perianally prior to penile penetration or via reapplication directly on their partner’s penis during receptive sex.

While participants’ satisfaction with the DPV gel indicates its promise for use by individuals engaging in RAI in the future, participants’ preferred mode of administration in the future – either through applicator or as a lubricant – generated mixed reviews. There was high acceptability regarding the ease of use for both administration processes; however, when asked to consider their preferred method of application if an effective RM gel was available in the future, participants had diverse opinions about their preferred administration process. For example, some participants preferred the applicator method as it eased any concerns regarding having to measure that they had the right amount of drug needed to offer protection, whereas others preferred the CSD method as it aligned with how they use lubricants when having RAI. It is important to note that some participants expressed confusion regarding the role of the CSD in the trial during the in-depth interviews. While study staff discussed the purpose of using the CSD during the consent process, no written educational counseling materials were offered to participants to explain the rationale for using the CSD in the study. Therefore, participants’ misunderstanding of the CSD’s purpose may have influenced their comparisons between the two modes of administration. Moreover, while the CSD offered a mechanical way to simulate the gel as a lubricant during insertion, other participants noted that the CSD could not simulate the interpersonal and psychological aspects of RAI with a partner. Taken together, these findings are consistent with prior RM studies noting that the gel’s acceptability may be hindered in real life contexts by users’ acceptability of the administration process(Bauermeister et al., 2016a; Carballo-Dieguez et al., 2014) and partner dynamics(Giguere et al., 2016). Dyadic research(Mitchell, Lee, et al., 2016; Mitchell, Sophus, Lee, & Petroll, 2016) examining the potential utility of the DPV gel as a prevention strategy, inclusive of how both partners perceive its administration with and without the applicator, may be warranted.

Several limitations to this study deserve mention. First, as this was a Phase I safety trial, a small sample of low-risk individuals was invited to participate. Individuals who participate in higher-risk behaviors may have different needs, perceptions of the rectal gel being studied and opinions regarding application methods. For example, given that most of our participants reported using sex toys prior to RAI, it is unclear if these experiences made them more willing to use a CSD as part of this clinical trial. Second, participants were asked to abstain from receptive anal sexual activity or inserting any non-study products into the rectum for 72 hours before and after each study visit. Given the influence that partners may play in sexual decision-making, future acceptability research of the DPV gel should include dyadic perspectives from the receptive and insertive partners. Assessing the experiences of partners across both modes of application (i.e., pre-filled applicator and as a lubricant) may inform the viability of each as participants engage in RAI. Third, our findings regarding the DPV gel may not be generalizable to other microbicide gel candidates as each might have rheologic and PK parameters that affect users’ acceptability of them. Finally, given our small sample size, acceptability data cannot be generalized to other populations. Therefore, future research examining RM gel acceptability and adherence remains vital as new pre-exposure prophylaxis drugs are tested as potential HIV prevention methods.

Taken together, these findings echo previous work(Carballo-Diéguez, Giguere, Dolezal, Bauermeister, Leu, Valladares, Frasca, et al., 2014; Cranston et al., 2017; McGowan et al., 2016) highlighting the support and acceptability for a RM gel and the importance of designing a variety of biomedical HIV prevention strategies that meet the diverse needs of individuals engaging in RAI. Given our findings provided mixed results regarding the ideal RM modality, future trials should continue to assess drug delivery modalities in multiple configurations. Future research examining the efficacy of the DPV RM gel and its feasibility and potential for HIV prevention among key populations is warranted.