News

A European study of men who have sex with men (MSM) presented at the recent Conference on Retroviruses and Opportunistic Infections (CROI 2018) shows that age makes a considerable difference to whether a person is more likely to be infected with, or to transmit, HIV.

Though it confirmed that younger gay men are more likely to be infected by men a number of years older than them, it also found that in gay men over 40, the risk was reversed: gay men who are still HIV-negative at that age are more likely to be infected by partners younger than they are, rather than by the same age or older.

The study has implications for what might be the most effective HIV prevention interventions for different age groups.

A number of studies support the idea that so-called ‘age mixing’ is an important factor in determining HIV incidence and prevalence in different communities. This makes intuitive sense, as the risk of being HIV positive rises with age, and therefore the risk of transmitting it unless people are on treatment and virally suppressed.

This makes a difference in the overall risk of HIV in the community if it is common for there to be larger differences in age between sexual partners. Most studies find it to be a major factor in young women’s huge risk of HIV in southern Africa, as another study presented at CROI confirmed.

It has also been suggested as one of the factors behind the higher HIV infection rates among black gay men in the US: some studies have shown that because US black gay men select partners largely from their own community and therefore from a smaller number of available partners, this has resulted in there being wider average age differences between partners.

The BEEHIVE (Bridging the Evolution and Epidemiology of HIV in Europe) study is a genetic study of seven different HIV cohorts in six European countries. Its primary purpose is to investigate differences in HIV virulence – whether there are any genetic factors that affect people’s viral load off treatment.

However, BEEHIVE has now collected enough HIV genetic samples not only to study pairs of people whose HIV is similar enough to suggest one transmitted HIV to another – but even in some cases to estimate the direction of infection.

A total of 2811 HIV genomes were sampled: transmission pairs were defined as pairs where at least 66% of sequentially selected samples of HIV gene sequences strongly suggested direct transmission, and where at least 50% suggested a direction of transmission. Samples came from Belgium, France, Germany, the Netherlands, Switzerland and the UK, with the biggest group from Switzerland.

In the end 167 transmission pairs were identified and age data were available for 148. Of these, 111 were male-male couples where sex was the risk factor, and in 106 the direction of transmission could be established.

The estimated mean age at infection was 35.6 among people classed as definite ‘sources’ and 36.9 among ‘recipients’ – already challenging the assumption that recipients would probably be younger. In 57% of cases the recipient was older than the source, with an average age difference of 2.16 years. However, this situation was reversed in men under 30, where on average the recipient was nearly nine years younger.

There was a clear and steady increase in the likelihood of the source being the younger partner as both partners got older. It was calculated that for every year older a recipient was, the average age of the source of their infection only grew by 0.28 years (3.36 months).

When the recipient was under 25, their source partner was on average ten years older; when 25-29, eight years older; and when 30-34, five years older. However, the average age difference between pairs aged 35-39 was zero, and after that source partners became younger than recipients; two years in men aged 40-44, and 11 years among men over 45. (Of note, there was a long ‘tail’ in this study of older recipient men; several were in their late 60s.)

To put this another way: in recipients under 25, 92% of their source partners were older than them; in 25-29 year olds, 51%; in 30-34 year olds, 47%. But in 35-39 year olds, 35% were younger; in 40-44 year olds, 52% were younger; and in men over 45, 71% were younger. The age at which the age difference “crossed” was 38.4.

It is obvious why young men are more likely to be infected by older men; there are simply more older HIV-positive men around. But why were older men more likely to be infected by younger? The researchers suggest three possible explanations. Older men may simply be less sexually active; or they may be better at maintaining condom use. Neither of these explanations seem to be borne out by other studies of condom use in gay men, however; if anything, younger gay men use condoms more.

The most likely explanation would seem to be that older gay men with HIV are more likely to be diagnosed, virally suppressed on treatment and therefore non-infectious.

The findings have implications for prevention; quick diagnosis and treatment may be the most relatively efficient way of preventing HIV transmission in men in their 30s, while men younger and older than this may, as groups, derive more benefit from  pre-exposure prophylaxis (PrEP).

Hall MD et al. Age transmissions in European HIV transmission pairs uncovered with viral sequence data. Conference on Retroviruses and Opportunistic Infections (CROI 2018), Boston, abstract 960, 2018.

View the abstract on the conference website.

Download the study poster from the conference website (PDF).

Watch the webcast of this presentation on the conference website.