A mathematical model derived from current knowledge
about the efficacy of various different prevention strategies has found that,
based on these data, the risk of HIV transmission from a person living with HIV to an HIV-negative
partner in a serodifferent couple could still be substantial over a ten-year
The authors state they undertook their research because
people in serodifferent couples (often referred to as serodiscordant) require guidance about the likely impact of emergent
prevention strategies (such as treatment as prevention and pre-exposure
prophylaxis) alone, or in combination with each other, on the risk of HIV
They therefore estimated the risk of sexual transmission of
HIV over one-year and ten-year periods for gay and heterosexual couples. The
risk for heterosexual couples was modelled separately according to whether the
male or female partner was living with HIV. The authors, publishing in the
online edition of AIDS,
emphasise that their model does not say what the actual risk will be. “This model was not designed to
predict actual transmission risk for real-world serodiscordant couples over the
course of a multiyear relationship,” they say. “Our intent is to emphasize how
risk accumulates over time under various strategies and show the relative
differences between strategies.”
Antiretroviral therapy is the intervention most likely to
reduce the risk of sexual transmission of HIV in such couples but even with
antiretroviral therapy for the partner living with HIV, the model computes that
the ten-year risk of transmission in gay couples is 25%, with a 2% risk for
The only way of reducing transmission risk further was to
use an unrealistic combination of prevention interventions.
“Modest HIV transmission probabilities per sex act
translate into substantial cumulative risks over time,” comment the authors.
“In serodiscordant couples, particularly those practicing anal sex, some
strategies (including consistent condom use) may not provide sufficient levels
of protection over an extended time when used alone.”
The study has a number of limitations. The authors acknowledge
that they did not factor in whether the partner on HIV therapy achieved viral
load suppression. They also used pessimistic estimates of the efficacy of
various prevention strategies.
The model assumes the following reductions in risk: 80%
with consistent condom use; 54% from circumcision of the male partner in a
heterosexual couple; 73% for PrEP in heterosexual couples; 44% for PrEP in gay
couples; and 96% from antiretroviral therapy when used by the partner with HIV.
The model also assumes that male circumcision reduces the
risk of HIV by 73% for the HIV-negative partner in a gay relationship where the
partner is exclusively insertive (top) over the whole ten-year period but even
over one year surveys suggest that no more than one-in-five HIV-negative men
maintain this role exclusively and one in seven are exclusively receptive:
almost all studies show that circumcision has no protective effect in gay men
The model also assumes that couples have penetrative sex
six times per month. Gay couples had three episodes of receptive anal sex and
three episodes of insertive anal sex.
A substantial risk of HIV transmission remained when
couples relied on any single prevention strategy.
For gay men relying on condom use the risk of HIV
transmission is 13% over one year, which adds up to 76% over ten years. When
antiretroviral therapy is the sole prevention method used there is a 3% risk of
transmission over one year, equating to a 25% risk over ten years.
A strategy of antiretroviral therapy with condoms, PrEP,
circumcision and no receptive anal sex for the HIV-negative partner was needed
to reduce the one-year risk to 0.1% and the ten-year risk to 1%.
For heterosexual couples with an HIV-negative male partner,
the transmission risk with consistent condom use was 1% over one year and 11%
over ten years. The one- and ten-year risks with antiretroviral therapy alone
were 0.2% and 2%, respectively. The risk associated with PrEP alone was 2% over
one year and 15% over ten years. Combining HIV therapy, condoms, PrEP and
circumcision reduced the risk to 0.01% over one year and 0.1% over ten years.
In heterosexual couples where the female partner was
HIV negative, consistent condom use alone resulted in a 1% one-year
transmission risk and a 11% ten-year risk. HIV therapy alone was associated
with a 0.2% and a 2% risk over one and ten years, respectively. A combination
of antiretroviral therapy, condoms and PrEP reduced the one- and ten-year risks
to 0.05% and 0.5%, respectively.
It is important to note that this model uses the most
pessimistic assumptions about efficacy. The 44% efficacy of PrEP in gay men,
for instance, was based on a study in which 50% of participants turned out not
to have taken PrEP at all, and some
more recent PrEP studies have found higher levels of sufficient adherence.
The 96% efficacy for HIV therapy is based on the HPTN052 study in which the one
transmission from a partner on treatment came from someone who had only just
started treatment and was not virally suppressed. And some analyses of condom
use suggest that efficacy can be improved with behavioural support.
The model used may be based on data that are already out of
date. Recently, the PARTNER study showed no HIV
transmissions in couples – gay and straight – when the HIV-positive partner was
taking treatment and had an undetectable viral load. The results so far
predict that in the most pessimistic likely scenario, the ten-year risk of transmission
via anal sex is 10%.