HSV-2 suppression with aciclovir and HIV transmission: further conflicting data from randomised studies

In a series of oral presentations at the XVII International AIDS Conference last week, investigators presented data from several trials on whether continuous use of aciclovir to suppress HSV-2 infection could curb HIV transmission. Disappointingly, two randomised controlled clinical trials showed no reduction in HIV infection rates due to aciclovir treatment - but two studies of episodic treatment of genital herpes showed a reduction in HIV shedding.

A wealth of evidence has established that the risk of sexual transmission of HIV is significantly greater when either partner is infected with herpes virus 2 (HSV-2). This is likely due to HSV-2 lesions being more frequent and severe in HIV-positive persons, leading to increased HIV shedding, and to HSV-2 causing genital ulcers that increase susceptibility to HIV infection in HIV-negative persons. Hence, several researchers have been investigating using aciclovir as HSV-suppressive therapy in the hope of reducing HIV transmission to and from people with HSV.

Further data from the largest trial of HSV-2 suppression to date, HPTN 039, were presented by Jorge Sanchez of Lima, Peru on behalf of a broad international research team. Initial findings from this study were presented at the Fifteenth Conference on Retroviruses and Opportunistic Infections earlier this year (see report here). This placebo-controlled, randomised controlled trial (RCT) of HIV-negative, HSV-2-positive participants used daily, standard-dose aciclovir to suppress HSV-2, and studied the resultant effect on genital ulcers and new HIV infection rates.

This trial recruited two widely different groups of participants: 1355 men who have sex with men (MSM) from sites in Peru, 459 MSM from sites in the US, and 1358 women from Zimbabwe, Zambia, and South Africa, for a total of 3172 participants. All were HIV-negative, seropositive for HSV-2, and had had at least one episode of unprotected sex (anal or vaginal) in the past three months.

Participants were randomised to aciclovir 400 mg twice daily (n=1581) or placebo (n=1591). They received study medications and risk reduction counseling at monthly visits for 12 to 18 months, HIV testing every three months, and had genital ulcers tested for HSV DNA. Study retention and adherence rates were excellent. Syphilis, gonorrhoea and chlamydia were tested for at baseline and treated as needed.

The participants receiving aciclovir had fewer genital ulcers and lower rates of HSV in their genital ulcers, but these results varied between groups. Overall genital ulcers by exam and history were reduced by about a third in the aciclovir arm, with a relative risk [RR] of 0.63, but the best results were seen in the US men (RR=0.51) and the poorest in the African women (RR=0.68), with Peruvian men intermediate (RR=0.61). Aciclovir was more effective at reducing genital ulcer episodes in which HSV-2 was detected in the ulcers, with the same geographic differences: RR was 0.36 overall, 0.12 in the US, 0.38 in Peru, and 0.41 in Africa. Reduction in HSV quantity in ulcers was only observed among US MSM.

Most significantly, however, aciclovir treatment had no significant impact on rates of new HIV infections. HIV incidence was 3.9/100 person-years in the aciclovir arm (with 75 events) and 3.3/100 person-years in the placebo arm (with 64 events), for a non-significant hazard ratio [HR] of 1.16 (95% confidence interval [CI], 0.83 to 1.62; p=0.39). There was no significant difference in HR by past genital ulcer history, risk group, or region.

The investigator concluded that in this, the largest-ever trial of HSV-2 suppression, "suppressive aciclovir therapy led to a significant reduction in incidence of genital ulcers (47% less ulceration and 63% less HSV-2+ genital ulceration), but did not reduce HIV acquisition among high-risk HSV-2 seropositive MSM and women."

Dr. Sanchez proposed several tentative explanations, suggesting that HSV-2 may simply be a risk marker rather than an actual causative risk factor for HIV infection – an explanation, however, that he dismissed as unlikely. Other speculative possibilities were that the aciclovir intervention may have been appropriate but not potent enough, and/or that we may have underestimated how much of a role HSV-2 may play in increasing HIV transmission. He remarked that "ultimately, the best intervention is to prevent HSV-2 acquisition through an HSV vaccine."

Similarly disappointing results of another trial conducted in Tanzanian women between 2004 and 2007 were presented in the same session by Deborah Watson-Jones of the London School of Hygiene and Tropical Medicine. The first analysis of this trial was presented at last year’s International AIDS Society conference in Sydney (a href="/page/1427879/" >see report here). This trial studied 821 female bar, hotel and other facility workers, 16 to 35 years of age, in Tanzania. All were HIV-seronegative but HSV-2 seropositive.

The participants were randomised to aciclovir 400 mg twice daily (n=400) or placebo (N=421), and followed every three months for twelve to thirty months (depending on date of enrolment). All were also given treatment for other sexually transmitted infections (STIs), condoms, and voluntary HIV counselling and testing.

A total of 659 of the women (80%) completed follow-up. Sixty-three of these women (8%) HIV seroconverted during the course of the study. HIV incidence was 4.27 per 100 person-years overall, with no impact of aciclovir treatment on HIV incidence (RR=1.01; 95% CI, 0.61 to 1.66; p non-significant).

Multivariate analysis of other risk factors found that the women were more likely to become HIV-positive if they were younger (16-to-19-year-olds were four times as likely to become infected as women over thirty: adjusted RR [ARR]=4.02; 95% CI, 1.7 to 9.7; p=.01), had a paying sexual partner in the past three months (ARR=1.82; 95% CI, 1.1 to 3.1; p=.03), drank alcohol (10-29 drinks/week: ARR=3.00; 95% CI, 1.5 to 6.0; for ≥30 drinks/week: ARR=4.39; 95% CI, 1.7 to 11.3; p<.001), received any injections outside the study clinic within the previous three months (ARR=3.45; 95% CI, 1.6 to 7.3; p=.005), or recently had gonorrhoea (ARR=3.62; 95% CI, 1.6 to 8.1; p=.007).

Dr. Watson-Jones concluded that, despite the lack of protective effect against HIV found for aciclovir, STI control was still a crucial issue, and that "prevention interventions need to highlight risks associated with alcohol use, and to target young women."

In questions after the session, Dr. Watson-Jones reported that median adherence to aciclovir was roughly 90%, but that it was not certain whether adherence could have affected the outcomes. She did offer the opinion that "aciclovir's short half-life and twice-daily dosing has limited the power of these proof of concept trials… if we could do it again we would prefer to study valaciclovir," due to its longer half-life and easier (once-daily) dosing.

A higher dose of aciclovir has also been proposed as a means of improving the efficacy of HSV-2-suppressive treatment in people with HIV infection, as has episodic management of genital lesions in preference to indefinite treatment.

A study in Malawian men and women found mixed effects of higher-dose aciclovir treatment on HIV shedding, with fewer participants showing HIV in the semen and in HSV lesions. Yet there was no effect on HIV shedding in the vaginal tract or on plasma levels. This study did not measure rates of new HIV infections.

Sam Phiri of Kamuzu Central Hospital, Malawi presented data from this study which used short-course, twice-daily doses of 800 mg (rather than 400 mg) aciclovir in Malawian men and women with genital ulcerative disease. This randomised, double-blind, placebo-controlled trial was conducted at Kamuzu Central Hospital STI clinic in Lilongwe, Malawi. The objectives were to analyse the impact of aciclovir treatment on genital ulcer healing, and on the quantity of lesional, genital, and plasma HIV RNA.

The trial enrolled 422 patients (294 men and 104 women) with genital ulcerative disease. Overall, 60.9% and 71.7% had antibodies to HIV-1 and HSV-2 respectively. The median CD4 cell count was 233 cells/mm³ and mean plasma HIV-1 RNA was 4.80 log₁₀copies/mL in HIV-positive participants.

Over half the participants (244) were positive for both HIV and HSV-2. Most ulcers (67.1%) were due to HSV-2 (80% in women, 63% in men). All study participants received syndromic management for their ulcers (immediate penicillin injections plus oral ciprofloxacin) plus either aciclovir 800mg twice daily for five days (n=208 overall, 123 in co-infected), or placebo (n=214 overall, 121 in co-infected).

Overall, 85% of the ulcers were healed after fourteen days. However, aciclovir did not significantly speed ulcer healing compared to placebo, either in the overall group (RR=1.02, 95% CI, 0.93 to 1.15) or in the co-infected individuals.

Ulcer swabs, and cervical swabs or semen, were collected from HIV-1 positive participants. Among HIV-1/HSV-2 dually seropositive patients, aciclovir reduced the number of people in whom HIV-1 RNA could be detected in semen (RR=0.6; 95% CI, 0.4 to 0.8; p=.01) and in genital lesions (adjusted RR=0.6, 95% CI, 0.4 to 0.9; p=.05). However, there was no significant change in the actual mean measured quantity of HIV RNA. Nor did aciclovir have an impact on frequency or quantity of cervical HIV-1, or on plasma HIV viral load.

Phiri concluded that "aciclovir had little impact on ulcer healing rates," but that there was a "reduction in frequency of lesional and seminal HIV-1 RNA … suggest[ing] that herpes therapy may reduce genital HIV-1 transmission."

Another randomised study of episodic treatment, carried out among men with genital ulcer disease in South Africa, using an aciclovir dose of 400mg three times a day for five days, found an improved rate of wound healing and reduced lesional shedding of HSV-2, as well as reduced HIV shedding.

The double-blind, placebo-controlled trial randomised 615 men with genital ulcer disease to receive a five day course of aciclovir - 400mg three times a day – or a placebo.

The primary endpoint of the study was to assess the effect of aciclovir on the healing of an HSV-2- related ulcer. Studies in HIV-negative patients have shown that episodic treatment is primarily effective in treating initial episodes of genital herpes; treatment must be started early for aciclovir to have a significant effect in reducing the length of subsequent episodes of genital herpes. On average, episodic treatment reduces the duration of ulceration and shedding by one to two days.

In this study the men were followed up for a month to assess the healing of ulcers and to collect blood and ulcer swabs to test for HIV. All were being treated in three clinics in South Africa.

The median age of the participants was 29 years: 63% were HIV-positive and 70.6% were seropositive for herpes simplex virus 2 (HSV-2) (the study recruited men with genital ulcer disease in order to mimic the clinical experience of syndromic management of genital ulcer disease; WHO recommends presumptive aciclovir treatment of genital ulcerative disease where local prevalence of HSV-2 is greater than 30%). In 21% of cases no causative organism for the genital ulcer could be identified.

Aciclovir was shown to improve ulcer healing in those who were HIV-positive - 59% of these men had healed ulcers by day seven compared to 41% of those taking placebo (p=0.003) (ARR 1.4 (95% CI 1.1 – 1.8, p=0.001), but did not significantly affect the rate of ulcer healing in HIV-negative men. Among the HIV-positive men with CD4 counts below 200, ulcer healing was no more likely within seven days with aciclovir treatment than with placebo, but among men with CD4 counts above 200, it was significantly faster (60.9% vs 39%, ARR 1.6, 95 CI% 1.1 – 2.2, p=0.003).

The same overall pattern held true when the researchers analysed wound healing according to the median duration of time to wound healing rather than the proportion of wounds healed within seven days. Among HIV-positive men overall, the time to wound healing was reduced by 50%.

But the researchers were also interested in whether aciclovir had any impact on shedding of HIV from the ulcers.

The drug was shown to reduce HIV ulcer shedding at day seven with 25% having detectable virus on aciclovir compared to 38% on placebo, p=0.05). Mean ulcer HIV RNA levels at days 7 and 14 were also significantly reduced. HSV-2 shedding was significantly reduced at day 7 but not at day 14.

Adding aciclovir to the syndromic management of genital ulcerative disease should be recommended for all HIV-positive individuals, the study investigators recommend, and HIV testing should be strongly encouraged whenever individuals present with a sexually transmitted infection. Management of genital ulcer disease is a potentially important entry point to bring people into HIV care, and for achieving HIV prevention gains, concluded Sarah Hawkes of the London School of Hygiene and Tropical Medicine, presenting the study.