 |
|
|
| |
HIV incidence in users of HIV preexposure prophylaxis
in Australia: a whole-of-population analysis
|
| |
| |
Infrequent or Rare PrEP Tied to Higher HIV Incidence in National Australian Analysis
CROI 2024 (Conference on Retroviruses and Opportunistic Infections), March 3-6, 2024, Denver
Mark Mascolini
Compared with people who used HIV preexposure prophylaxis (PrEP) on more than 60% of days, those who used it on fewer than 60% of days had a two-thirds higher incidence of HIV infection and those who tried PrEP only once had almost a 5-fold higher HIV incidence in a 5-year country-wide Australian analysis [1]. Hepatitis C virus (HCV) infection boosted HIV incidence almost 9-fold.
Australia boasts high PrEP access since a government-subsidized PrEP program began in April 2018. More than 70,000 people in a national population of 25 million have started PrEP in Australia. Most PrEP users are gay or bisexual men who have sex with men (MSM).
Nick Medland (Kirby Institute, Sydney) and coworkers at other centers wanted to estimate how effective PrEP is when scaled up over the long term. The challenge they faced was how to measure HIV incidence outside of planned studies. For example, what could they compare population-wide incidence to?
Certain factors in Australia's HIV plan, Medland and colleagues figured, could help fashion a whole-population PrEP analysis: First, a government-supported program makes PrEP highly accessible—more than 90% of PrEP users rely on the government plan. Second, almost all MSM who use PrEP in Australia get tested for HIV regularly—99% in the past year. Third, if Australians do acquire HIV infection, 95% start antiretroviral therapy (ART) within 6 weeks.
The researchers decided they could use delinked government prescription data for PrEP and ART to estimate (1) HIV incidence in people who used PrEP in the 5 years from April 2018 to March 2023, (2) predictors of HIV acquisition, and (3) population effectiveness of PrEP comparing regular PrEP users to those who used PrEP less frequently or only once. They scrutinized prescription data for PrEP, ART, and HCV infection to the end of the study period in March 2023 then added 6-month observation period up to September 2023.
Because of known high HIV testing rates in MSM, the investigators assumed that people who took PrEP got tested for HIV twice a year or when PrEP is prescribed. They further assumed that if a person had not begun ART in the 6-month observation period following the 5-year study span, that person did not have HIV infection at the end of those 5 years. Because Australians begin ART rapidly after becoming infected, Medland and colleagues figured starting ART meant a person just got diagnosed with HIV. The imputed date of acquiring HIV fell at the midpoint between 30 days before ART began and the most recent PrEP prescription or 6 months earlier.
The researchers used date and quantity of PrEP pills prescribed to rank PrEP use as (1) only a single dispensed PrEP supply, (2) more than one PrEP supply but fewer than 60% of days covered, or (3) 60% or more days covered. They considered the first group to be eligible for PrEP at some point and to have a contemporaneous ongoing HIV risk. The investigators used Poisson regression to determine unadjusted and adjusted HIV incidence and incidence rate ratios.
This exercise yielded 66,206 people dispensed PrEP, 207 of whom (0.30%) acquired HIV infection. Median age was slightly but significantly older in all people who got PrEP than in people who got PrEP but got infected with HIV (33 vs 32 years, P = 0.042).
Among the 66,206 people prescribed PrEP, 19.0% got only one PrEP supply, 53.5% got more than one supply but had fewer than 60% of days covered, and 27.5% got more than one supply and had 60% or more days covered. Respective proportions among people who used PrEP and acquired HIV were 30.4%, 54.6%, and 15.0% (P < 0.001).
Overall HIV incidence came to 1.07 cases per 1000 person-years (95% confidence interval [CI] 0.93 to 1.23), meaning 1 of 1000 people acquired HIV every year. People who got only one PrEP supply had higher HIV incidence (2.61 per 1000 person-years, 95% CI 2.04 to 3.33) than people who got more than one supply but had fewer than 60% of days covered (0.99 per 1000 person-years, 95% CI 0.83 to 1.2) or people who had 60% or more days covered (0.56 per 1000 person-years, 95% CI 0.39 to 0.8).
People 18 to 29 years old had higher HIV incidence (1.33 per 1000 person-years, 95% CI 1.07 to 1.64) than people 30 to 39 years old (1.03 per 1000 person-years, 95% CI 0.81 to 1.31) or people 40 or older (0.86 per 1000 person-years, 95% CI 0.66 to 1.12). Compared with people not treated for HCV infection (1.01 cases per 1000 person-years, 95% CI 0.88 to 1.16), people treated for HCV had a higher HIV incidence (9.83 cases per 1000 person-years, 95% CI 5.71 to 16.93).
Adjusted analyses determined that, compared with people who had 60% or more days covered by PrEP, those who got more than one PrEP supply but had fewer than 60% of days covered had a 1.66-fold higher HIV incidence (adjusted incidence rate ratio [aIRR] 1.66, 95% CI 1.11 to 2.49, P = 0.01) and those who got only one PrEP supply had 4.7-fold higher HIV incidence (aIRR 4.71, 95% CI 2.97 to 7.46, P < 0.001). When Medland and colleagues used the group that got only one PrEP supply for comparison, people who got more than one supply but had fewer than 60% of days covered had 62% lower HIV incidence, and people with 60% or more days covered by PrEP had 79% lower HIV incidence.
Compared with people 40 or older, those 18 to 29 had about a 50% higher HIV incidence (aIRR 1.56, 95% CI 1.11 to 2.21, P = 0.011) and people taking versus not taking HCV therapy had almost a 9-fold HIV incidence (aIRR 8.70, 95% CI 4.86 to 15.5, P < 0.001).
Medland and colleagues believe their findings underscore the need to extend more support for people who try PrEP but don't return for prescription refills and people who use PrEP less often than frequent users. The study also highlights the need to refine PrEP uptake and persistence programs for high-risk groups like people with HCV infection and people younger than 30.
Reference
1. Medland NA, McManus H, Bavinton B, et al. HIV incidence in users of HIV preexposure prophylaxis in Australia: a whole-of-population analysis. CROI 2024 (Conference on Retroviruses and Opportunistic Infections), March 3-6, 2024, Denver. Abstract 166.
|
| |
|
|
|
|
|
