Dapivirine PrEP vaginal ring shows only limited PrEP protection against HIV in African women
Simon Collins, HIV i-Base
A slow-release monthly vaginal ring containing dapivirine was reported to reduce rates of HIV infection compared to placebo in two large randomised studies. Overall protection however appears limited.
These results were presented at a press conference two days before studies were due to be presented in detail. This considerably limited the ability to report detailed results as other dapivirine studies were embargoed until later in the week.
Although both presenters concluded that their results were positive, the ring showed little or no protection for the youngest women who were at highest risk and are in greatest need. Although the ring was designed with the hope that it would minimise the risk of low adherence, many women did not use the ring consistently during the studies.
Both studies – MTN-020/ASPIRE and IPM 027 – were double-blind placebo controlled phase 3 studies, and even in the context of good adherence, efficacy fell well below 100%. The results showed that dapivirine might only have limited potential for reducing HIV incidence on a population level and might be much less effective for someone wanting optimal protection.
One of the studies, MTN-020, was simultaneously published in the New England Journal of Medicine, and this paper is the main focus of the report below. 
Results from MTN-020/ASPIRE were presented by Jared Baeten from the University of Washington, Seattle. This study screened 5516 women and enrolled 2629 women aged 18-45 years at 15 sites in Malawi, South Africa, Uganda, and Zimbabwe. [1, 2]
Baseline characteristics included median age 26 years (IQR 22 to 31 years), with 95% having at least secondary school education (level not specified). Nearly all women had one main partner (99.5%) had a primary partner and 41% were married, almost half (45%) had an independent income. Women were sexually active (mean 26 times in previous 3 months, +/-24). Only 17% had more than one partner in the previous 3 months, almost 60% reported using a condom the last time they had sex, 6% reported transactional sex and only 2% reported anal sex (for which a vaginal ring would offer no protection.
Retention in the study was reported as good with women attending 91% of expected monthly visits.
Adherence in the study was more difficult to report. The main assessment by 3-monthly drug level testing reported adherence at 80%. However, the effective design of the ring ensured that target levels are reached within 8 hours of the ring being inserted. Although the ring was designed as an intervention to insert once a month, the ring seems to have often been taken out between clinic visits.
Interpreting these results is difficult. Actual adherence would be overestimated – and true efficacy therefore underestimated – if the ring was only inserted the day before a clinic visit.
Secondary adherence measures included remaining drug levels in returned rings, but this is the subject for future analyses.
For the primary endpoint of new infections, during 4280 patient years of follow-up (PYFU), with median follow-up of 1.6 years (IQR 1.1 to 2.3 years), 168 women became HIV positive: 71 in the active group vs 97 in the placebo group. This resulted in incidence rates of 3.3 vs 45 per 100 years of follow up, respectively. The relative reduction overall in HIV incidence from dapivirine compared placebo was 27% (95%CI: 1 to 46, p=0.046), reaching statistical significance.
Two further post hoc sub-group analyses were also reported showing higher efficacy.
- A 37% reduced risk (95%CI: 12 to 56, p=0.007) was reported after excluding two sites with the lowest rates of retention and adherence. In these two sites, infections occurred more frequently in women in the active (n=17) vs placebo group (n=12).
- A 56% reduced risk (95%CI: 31 to 71, p=0.0003) when restricting results to women older than 21 years of age (as adherence was lower in women aged 18-21).
Safety results looked similar between the two groups with 14% of people in each group reporting a primary safety endpoint. Serious events were reported by 4% of women in each group, with 2% in each group having grade 4, 12% reported grade 3, and <1% deaths (n =4 and 3). None of these events were attributed to the ring or active drug.
It is notable that NNRTI drug resistance in women who became HIV positive during the study was not higher in the active group: 8/68 (12%) in the active vs 10/96 (10%) in the placebo group, p=0.80).
IPM 027 study
Results from the IPM 020 study were presented by Annalene Nel from the International Partnership for Microbicides (IPM). 
IPM 027 enrolled 1959 women (1762 in South Africa and 197 in Uganda) aged 18 to 45 who were randomised 2:1 to dapivirine:placebo ring. The two-year study was stopped early following a recommendation from the Data and Safety Monitoring Board (DSMB) due to higher than expected HIV incidence.
Limited data on baseline demographics include mean age at enrolment of 26 years, 59% completed secondary school and 91% were mothers. Nearly all women (>98%) had a primary partner, 96% reported having sex weekly over the previous three months.
During 2805 patient years of follow up (761 women completed two years), 133 women became HIV positive: 77 vs 56 in the active vs placebo groups. This produced incidence rates of 4.08 vs 6.10 per 100 person-years.
Overall, dapivirine reduced the risk of infection by 31% (95% CI: 0.90 to 51.5%; p=0.040) compared to placebo. When analysed by age there was no significant benefit for women <21 years: 15% (95%CI: -60% to +59%). In women aged >25 years, the reduction was 37.5% (95% CI: 3.5 to 59.5%).
Although a trend to higher protection correlated with higher adherence (measured by lower levels of drug in the returned rings), this was one of several subsequent post hoc analyses.
Similar product-related side effects (approximately 87%, with <5% grade 3-4), were reported for active and placebo groups, including serious side effects. This included similar reports in both groups of unusual bleeding between periods, discomfort and pain.
Four additional studies on the 25 mg dapivirine ring were presented at CROI 2016.
Two posters reported PK, safety and generally high acceptability from a US 12-week, phase 2a study (MTN-024/IPM 031) in 96 post-menopausal women randomised 3:1 to active vs placebo ring. [4, 5]
Mean age was 57 years (range 46-65); 66% were white, 31% were black, and 3% were of other race. Side effects (grade 2 or higher) were reported in 8% vs 13% in the active vs 13% groups respectively (p=0.68). One grade 3 vaginal pain was related to the device.
Adherence reported by drug levels in the returned rings was consistent with use through the month (median 21.1 mg, with 4 mg associated with sustained use during one month). Adherence by self report included 73% of women using the ring throughout the study and 91% reported never removing the ring for more that 12 hours.
Six women reported that at some point the ring fell out and 26 reported partial slippage, mainly due to bowel movements; 18 women removed the ring at some point due to discomfort, worries, or to clean the ring. High rates of acceptability were reported by the majority of women.
Two posters included cost effectiveness and population effect modelling that generally relied on higher adherence and efficacy levels than were seen in the above studies. At the highest modelled efficacy (75%) it was predicted that the ring might reduce 4-7% of infections in South Africa. [6, 7]
Dapivirine was originally identified by Tibotec in the late 1990s (with a development name TMC-120). In 2004, when an alternative NNRTI was selected for development as treatment, dapivirine was made available for public research as a microbicide for use in low income countries in a royalty free licensing agreement. Development was led by the International Partnership for Microbicide who plan to apply for regulatory approval based on these studies. 
While demand for effective new prevention options are clearly needed, the most disappointing outcome from both studies is an inability to know with any confidence the likely level of protection in the context of perfect adherence – especially as comparative oral PrEP using tenofovir/FTC sets the bar high at close to 100%.
Although the results with dapivirine cannot rule out a population benefit (linked to high background incidence and available options), low levels of efficacy are not helpful for individual protection.
Both studies are now planning to ask all women to continue using the dapivirine ring in an extended open label follow-up phase.
Using an NNRTI as the active compound was complicated by the lack of an animal model that could prove efficacy, relying on in vitro and other animal data mainly focused on tissue penetration.
While 100% high efficacy was demonstrated for tenofovir/FTC in macaque studies long before human studies were planned, the lack of a SHIV model means that phase 3 studies were used to provide first proof of principal results in women.
The ring itself is an impressive product with rapid drug deliver to steady state in tissue and steady drugs levels for 4 to 12 weeks, although there is significant intra- and inter-patient variability.
It is unclear why the ring was either difficult or not acceptable to use in practice in the African studies.
Unless stated otherwise, all references are to the Programme and Abstracts of the Conference on Retroviruses and Opportunistic Infections, 22-25 February 2016, Boston, USA.
- Baeten JM et al for the MTN-020/ASPIRE Study Team. Use of a vaginal ring containing dapivirine for HIV-1 prevention in women. NEJM, 22 February 2016. DOI: 10.1056/NEJMoa1506110. Free full text.
- Baeten JM et al for the MTN-020/ASPIRE Study Team. A phase III trial of the dapivirine vaginal ring for HIV-1 prevention in women
. CROI 2016, Boston. Oral abstract 109LB.
- Nel A et al for the IPM 027/Ring study research center Teams. Safety and efficacy of dapivirine vaginal ring for HIV-1 prevention in African women
. CROI 2016, Boston. Oral abstract 110LB.
- Chen BA et al. Safety and pharmacokinetics of dapivirine vaginal rings in postmenopausal US women. CROI 2016, Boston. Oral abstract 872.
- van der Straten A et al. Adherence and acceptability of a dapivirine vaginal ring in postmenopausal US Women. CROI 2016, Boston. Oral abstract 873.
- Glaubius R et al. Dapivirine vaginal ring preexposure prophylaxis for HIV prevention in South Africa. CROI 2016, Boston. Oral abstract 1057.
- Smith J et al. Cost-effectiveness of the intravaginal dapivirine ring: a modeling analysis. CROI 2016, Boston. Oral abstract 1058.
- IPM press statement. Two large studies show IPM’s monthly vaginal ring helps protect women against HIV. (22 February 2016).