‘Undetectable’ HIV means ‘untransmissible’ confirmed in larger sex study

One maxim of current HIV management is U=U; Undetectable equals Untransmissible, and it received new backing for the specific question of whether HIV treated to an undetectable level can transmit from a man who engages in condomless anal sex to an uninfected man. The answer was it effectively could not.

In nearly 77,000 sexual encounters of this type performed during a median 1.6 years of follow-up of each participating couple, no episodes resulted in HIV transmission. When researchers combined the new findings with numbers from a smaller, earlier study they ran of similar male couples the statistics showed a worst case of at most one HIV transmission for every 435 person years of follow-up, or “effectively zero”.

“The evidence is very robust. Transmission simply does not occur”. It was cautioned, however, that the finding occurred in couples where the HIV-infected partner consistently had a serum viral load of less than 200 copies of HIV RNA/mL when tested at the start and the end of follow-up and at various times in between. The researchers censored data from couples when the infected partner had a detectable viral load.        

The PARTNER-2 (Partners of People on ART – A New Evaluation of the Risks) study followed the PARTNER-1 study that had examined the same question of HIV transmissibility when undetectable during sexual encounters but with a much smaller number of men who have sex with men.

PARTNER-2 enrolled 783 male couples from any of 12 European countries that contributed 1,596 years of couple follow-up. During the median 1.6 years of follow-up per couple, one or both partners had a different sexually transmitted disease 23%-27% of the time, and 37% of the HIV-negative partners had condomless sex with a different partner. The couples averaged 43 sexual encounters a year that produced an overall total of 76,991. During follow-up, 15 of the HIV-negative partners become HIV positive, but in all 15 cases the source of the infection was genetically proven to have come from someone other than the index partner.

Based on the number of sex events studies it was calculated the 95% confidence interval around the zero transmissions they observed during gay male sex events with undetectable HIV in both PARTNER-2 and PARTNER-1. The upper bound of this confidence interval was a rate of 0.23 transmissions per 100 person years of follow-up.

In other words, the scope of the two combined studies cannot statistically rule out a worst-case scenario of one transmission for every 435 person years of follow-up, or for every 18,696 sex events of the type studied. However, “we’ve made the confidence much more robust. We can’t say its zero risk, but it’s effectively zero, “You could keep studying this forever” to further rule out the risk for transmission of undetectable HIV, “but I think the question is now settled.