Scene South Sudan, with 82, Cote d’Ivore, with

Scene
setting

Human immune deficiency (HIV) is a
chronic sexually transmitted disease that infects a person for life and leads
to acquired immunodeficiency syndrome (AIDS) if left untreated (5). HIV is
spread via blood, semen, breast milk, and vaginal and rectal fluids. The
disease weakens the immune system by attacking it, leaving those infected more
prone to disease (5).  Though HIV was
once a death sentence, anti-retroviral therapy (ART) has allowed HIV patients
to keep the disease under control and live otherwise normal lives (9).
ART is proven to reduce HIV transmission from one sexual partner to another by
93%, and the use and availability of ART in low and middle-income countries has
saved 5.2 million people from AIDS related death (2,3). Unfortunately, ARTs have
not been as effective in sub-Saharan Africa, compromising 70% of the global
burden of HIV infection (7). As of 2013, the highest death rates per 1000
people living with HIV were in Central African Republic with 91, South Sudan,
with 82, Cote d’Ivore, with 75, Cameroon with 72, and Chad, with 71 out of 1000
living with HIV (4). Though ART has reduced
instances of HIV and death in these five nations, they have almost ten times
higher death rates than high income countries, showing that there is a need for
further intervention (4).

In their study,
Use of a Vaginal Ring Containing Dapivirine for HIV-1 Prevention
in Women, Baeten
et al. indicate that young sub-Saharan women shoulder most of the HIV-1 related
burden (1). While ARTs are
proven to reduce HIV transmission, the authors cited three articles in which
ARVs were not effective in preventing HIV-1 incidences among African women, hypothesized
to be due to low adherence to daily or coital antiretroviral gels, films, or
pills (1). This subgroup of women may have difficulty remembering to use these
ARV measures, or may be prohibited from using these technologies by their
partners due to cultural beliefs or taboos. Given that the greatest mode of HIV-1 transmission in sub-Saharan
Africa is unprotected heterosexual intercourse, it is clear that young women in
sub-Saharan Africa require an innovative intervention to lower their chance of
HIV-1 infection (3).

Baeten et al. hypothesized that using
vaginal rings for ARV administration would provide longer term protection, thus
making adherence easier for these young women, ultimately decreasing new HIV-1
incidences (1). They decided to administer dapivirine via the vaginal rings, as
it is a topical antiretroviral medication that blocks HIV-1
reverse-transcriptase enzymes, preventing a broad range of HIV-1 subtypes from
multiplying in the body and causing infection (11). Dapivirine was chosen as it
decreased susceptibility to HIV-1 in previous studies without exposing the entire
body to the drug (1).

 

Need
for study

According to the WHO, 20% of new global
HIV cases were in women between the ages of 18 and 25, meaning that a fifth of new
HIV incidences occurred in only 11% of the adult population in 2015 (6). That same
year, young women in sub-Saharan Africa suffered from 56% of new HIV cases globally
among all adults (6). This report demonstrates
the gender gap in HIV infection, while identifying sub-Saharan women as an
especially vulnerable group.

As African women are the most affected
by the HIV epidemic, Baeten et al’s study addresses the need to decrease HIV incidence
by protecting this at-risk population (1). Though socioeconomic factors such as
insufficient education, lack of quality healthcare, poverty, and violence drive
the high HIV incidence among women in sub-Saharan Africa, it is clear that the HIV
epidemic needs to be controlled epidemiologically (6). Previous
studies have proven that monthly HIV and STI testing of women and their partners,
risk-reduction counselling, and free condoms were not enough to protect against
a positive HIV diagnosis (1). The need is to increase adherence to ART via a
longer acting option so women would not need to remember to take a daily pill or
apply a gel every time they have intercourse (1). As vaginal rings release
medication over a longer period of time, they offered promise in this regard (1).

The authors conducted a phase 3,
randomized, double-blind, placebo–controlled trial of dapivirine vaginal rings
in African women to determine whether vaginal ring offered simpler, long term
protection against HIV in comparison to a placebo ring (1). Every month, participants
were provided with new rings, condoms, clinical consultations, HIV-1 and STI tests
for themselves and their partners, safety monitoring, and individual adherence
counseling (1). It was important for the authors to study whether vaginal
rings, in conjunction with these other interventions, had their intended effect
of increasing adherence to ART while still providing care and information to
each participant.

 

 Key details of the study and analysis

The authors followed sexually active,
non-pregnant women with a negative HIV diagnosis between the ages of 18 and 45
years in Malawi, South Africa, Uganda, and Zimbabwe (1). At baseline, 5516
women underwent HIV-1 screening, and the 2629 women who tested negative for
HIV-1 were enrolled in the study. Baeten et. al then assigned 1313 women to the
dapivirine group and 1316 women to the placebo (1). Within each country, half
the participants were randomly assigned to use the vaginal ring containing 25
mg of dapivirine while the other half was allocated a placebo ring. The rings
were indistinguishable, enabling both participants and researchers to remain
blinded to the condition assigned to each participant (1). Women were taught to
insert and remove the rings, and were provided a new ring at each monthly appointment,
along with HIV-1 risk reduction counselling and free condoms.

Adherence was assessed via a plasma
dapivirine level of over 95 pg per millimeter (1). Additionally, a used ring
containing less than 23.4 mg of dapivirine objectively indicated the
participant was adherent (1).  Based on
their definition, the authors reported the rate of adherence in the study to be
over 70% (1).

Baeten et al. aimed for their test to correctly
reject false positives 90% of the time and used an end point driven study
design to record a minimum of 120 HIV-1 diagnoses to achieve this statistical
power (1). They aspired to record 60% lower incidences of HIV-1 in the
dapivirine group over the placebo (1). All participants were followed for 12
months minimum or until they tested positive for HIV-1. The results were
modelled using cox regression showing cumulative incidence to compare
differences between the dapivirine and placebo groups in each country. The authors claimed to use intention-to-treat (ITT) methods
in carrying out two analyses, one that showed cumulative incidence rates at all
15 sites, and one that excluded two sites with exceptionally low adherence (1).

Using ITT analysis, the results showed 26%
less incidence of HIV-1 in the dapivirine condition, which was significant in
all 15 sites. The 95% Confidence Interval showed HIV-1 incidence to be 1 to 46
percent lower in the dapivirine group, with a p value of 0.046. The data was then
analyzed by excluding two sites where adherence was lower than expected. After
exclusion, the authors reported 139 HIV-1 infections among 2395 participants,
54 of which were in the dapivirine group and the other 85 incidences in the placebo
group. This result showed that incidence of HIV was 37% lower in the dapivirine
condition, with a 95% confidence interval between 12 and 56% decreased
incidence in the dapivirine group with a strongly significant p value of 0.007
(1). Baeten et al. noted that the efficacy of HIV-1 protection of less than 25%
was not ruled out in either analysis (1).

However, efficacy of HIV-1 protection
differed significantly based on age, which Baeten et al. determined via a
post-hoc analysis (1). They found dapivirine increased HIV-1 protection by 61%
as compared to placebo in women aged 25 years or older (95% CI, 32 to 77; P<0.001). For women under 25, dapivirine had an efficacy of 10% (95% CI, ?41 to 43; P = 0.64) (1). To better analyze the effect of age, the authors stratified participants into 3 age groups: 18 to 21, 22 to 26, and 27 to 45. Among the 18 to 21 year olds, 44 out of 451 were diagnosed with HIV-1 in the placebo group, yielding an incidence of 5.4 out of 100 person years (1). Within the 22 to 26 year olds, 51 out of 752 participants in the placebo group were diagnosed with HIV-1. The incidence rate was 6.1 per 100 person years (1). Finally, 27 to 45 year olds had 44 out of 1192 participants diagnosed with HIV-1 in the placebo group with an incidence rate of 3.0 out of 100 person years (1). These results imply that dapivirine was effective in reducing new cases HIV-1 in women over 25, but was not effective in reducing HIV-1 incidences in younger women, as shown by the non-significant p value of 0.64. Additionally, HIV-1 incidence rates were lowest in women between 27 and 45, indicating the intervention was most successful in this demographic.   Study critique Limitations                                                                                While adherence was assessed via a plasma dapivirine level above 95 pg per millimeter, this level is usually achieved after using a dapivirine vaginal ring for 8 continuous hours (1). As women were instructed to wear the ring for a month, they cannot be truly adherent if they wear the ring for a shorter time (1).  As the authors' objective was to find a way to administer ART in a way that young women will adhere to, it is important that they actually ensure whether or not their target population adhered to the vaginal rings and found them to be a more desirable form of ART over gels, pills, or films. Thought the authors reported that 70% adhered to wearing the ring based on their definition, they can only speculate whether these women truly wore the ring for the whole month. Baeten et al. seemed to modify their results to show as strong an effect as possible. Originally, the null hypothesis was set as dapivirine being no more than 25% effective in reducing incidence of HIV-1. The alternative hypothesis was that dapivirine would have an HIV-1 incidence difference up to 60% lower in the dapivirine group. The authors aimed to rule out effectiveness below 25%, however, they changed the parameters of statistical analysis when dapivirine reduced HIV-1 incidence by only 26% (1). In order to obtain significant results, 0% was used as the standard reference parameter. This difference of 26% was only moderately significant in all 15 sites based on the p value of 0.046, meaning this result could be due to chance. Using ITT analysis while excluding participants in due to noncompliance is a limitation of the study. In ITT analysis is the gold standard and includes all participants within the condition to which they were randomly assigned, including the noncompliant, in order to maintain randomization (9). Though Baeten et. al wrote that the majority of participants met the criteria for objective adherence, they analyzed their results while excluding two sites, because they believed participants in these places to be non-compliant with little discussion as to their reasons for thinking so (1). Analyzing participants based on whether they followed the parameters of their assigned condition is known as per-protocol analysis (9). As the authors originally set out to determine whether dapivirine vaginal rings would be more effective and make compliance easier for women in sub-Saharan Africa, it seems unethical for the authors to pick and choose data to analyze under the guise of ITT analysis when it is more in line with per-protocol analysis. The per-protocol analysis showed HIV incidence to be 37% lower in the dapivirine condition, with a p value of 0.007, a strongly significant result. The confidence interval showed that dapivirine protection was between 12% to 56% lower in dapivirine compared to the placebo, which still includes the original null hypothesis value of 25% lower HIV-1 incidences in the dapivirine condition. Though there was a protective effect due to dapivirine in the study, it seemed as if the authors were statistically manipulating data to show a stronger effect. Though Baeten et al. discussed younger women having lower compliance rates and higher incidences of HIV-1 in their introduction, they did not stratify for age until after the study was over. The median age for the 2629 women between ages 18 and 45 was 26, meaning half the participants were younger than this age while half were older (1). Thus, there were many more younger participants in this study, making the fact that women between 18 and 21 having a 27% more HIV-1 incidences in the dapivirine group even more striking, as this age groups makes up a large portion of the study population (1). This result was moderately significant with a p value of 0.45, showing that vaginal dapivirine rings were not an effective treatment to reduce HIV in young sub-Saharan women between 18 and 21 (1). The authors stated that the lack of protection within the 18 to 21 year age group could be due to behavioral and biological factors combined, citing low adherence and potentially more HIV susceptible genital tracts (1). Seeing as the women were provided monthly support, STI testing, and free condoms along with the dapivirine rings, it is hard to imagine what more could be done to increase adherence in sub-Saharan Africa, especially as Baeten et al. did not explain potential biological factors in detail.  Though only 2% of women reported engaging in anal sex within the last 3 months at baseline, it is important to keep in mind that younger women may have higher rates of HIV because they are engaging in anal sex without reporting it due to fear of judgement and social stigmatization (1).  Topical dapivirine in the vagina would not be able to prevent HIV infection transmitted in the rectum (1). If this is the case, the vaginal ring's concentrated protection is not beneficial to young women, who would need a systemic ART.   Strengths             The authors used good randomization techniques in order to decrease confounding factors. Baeten et al. used block randomization to allocate research condition while stratifying by country, ensuring that confounding factors were reduced to a minimum (1). A strength of this study was demonstrating that incidence of HIV-1 was significantly lower in older women who adhered to the dapivirine study condition (1). Dapivirine vaginal rings were effective in women over 21 years of age, with results showing 56% less HIV-1 incidences as compared to the placebo. A p value of 0.001 shows the result is strongly significant. Additionally, 27 to 45 year olds had the lowest cumulative incidence rate of the three age groups with incidence of 3 out of 100 person-years. Conversely, 18-21 year olds were the only group in which the cumulative incidence rate of HIV-1 in placebo was lower than in the dapivirine condition, meaning less people died on placebo. Though Baeten et al. did not succeed in finding an intervention to help younger women in sub-Saharan Africa reduce incidence of HIV-1, they may have found a potential intervention to decrease HIV-1 transmission in older women. As many other studies could not yield any protective effects, a 26-37% decrease in HIV incidence is a promising start (8).   Next Steps While this study shows dapivirine releasing vaginal rings increases efficacy in HIV-1 protection in women over 21, who were much more adherent to the rings than younger women, there is still a public health need to develop an intervention to decrease HIV-1 incidence in young sub-Saharan women. Clearly a vaginal ring would not be clinically feasible for the young women who did not utilize them even with additional support. Policy makers and scientists should focus on finding a long acting, systemic ARV to protect young women against all methods of HIV transmission, as it remains a pertinent public health challenge. The women in this study should be followed up for qualitative interviews or questionnaires regarding their experience with vaginal rings to determine if they were comfortable, culturally appropriate, easy to use, and whether there were side effects. Women could also report their partners perspective on the vaginal rings, as a non-supportive partner may have contributed to decreased adherence. The results of these interviews could help researchers understand why younger women did not adhere to the rings as much as older women, ultimately leading to the development of a more clinically significant intervention. While vaginal rings were not proven to be clinically significant in helping younger women prevent HIV-1, scientists should continue to study vaginal rings potential to deliver ART as an option for more adherent women. Results from Baeten et al's study shows promise of longer term, efficient protection against HIV-1 for the population of older women, which may prove to be more economically feasible and clinically more significant in this demographic.   Though more work needs to be done to protect young women living in sub-Saharan Africa, this study provided evidence of dapivirine protecting women over 25 with longer term HIV-1 protection. Though the authors may not have achieved their goal in decreasing HIV-1 incidences in younger women, they have developed a promising intervention that may ultimately decrease the number of women living with HIV-1 in the future. Policy makers and scientists should work together in developing dapivirine rings as a new intervention, while further exploring new potential interventions targeted towards the young women in sub-Saharan Africa.