Protease inhibitor drug suppresses HIV in semen more slowly than drugs from other classes

Protease inhibitors may not be the best class of drug for people newly diagnosed with HIV to start treatment with, if they wish to quickly reduce their risk of passing HIV on to others, a Spanish study has found.

The protease inhibitor darunavir (Prezista) took much longer to fully suppress HIV in the semen of twelve male patients than did drugs of the non-nucleoside reverse transcriptase inhibitor (NNRTI) class or the integrase inhibitor class, the study from the University of Seville found.

The study recruited 36 gay men with HIV who were not yet taking antiretroviral therapy (ART). They all started first-line therapy containing tenofovir/emtricitabine (Truvada). For the third drug in their treatment combination, they were randomised into three groups of twelve who were given either darunavir boosted by ritonavir, rilpivirine (Edurant, also in Eviplera/Complera) or elvitegravir boosted by cobicistat (Vitekta, more usually in Stribild).


integrase inhibitors (INI, INSTI)

A class of antiretroviral drugs. Integrase strand transfer inhibitors (INSTIs) block integrase, which is an HIV enzyme that the virus uses to insert its genetic material into a cell that it has infected. Blocking integrase prevents HIV from replicating.

boosting agent

Booster drugs are used to ‘boost’ the effects of protease inhibitors and some other antiretrovirals. Adding a small dose of a booster drug to an antiretroviral makes the liver break down the primary drug more slowly, which means that it stays in the body for longer times or at higher levels. Without the boosting agent, the prescribed dose of the primary drug would be ineffective.

first-line therapy

The regimen used when starting treatment for the first time.

statistical significance

Statistical tests are used to judge whether the results of a study could be due to chance and would not be confirmed if the study was repeated. If result is probably not due to chance, the results are ‘statistically significant’. 


In HIV, an individual who is ‘treatment naive’ has never taken anti-HIV treatment before.

The small study found that while HIV was fully suppressed in semen within twelve weeks of starting therapy in participants taking elvitegravir or rilpivirine, in participants taking darunavir, a seminal viral load was still detectable in 42% of the participants.

After 24 weeks, all the 36 participants in the study had undetectable viral loads in semen except one taking darunavir (8.3% of darunavir participants).

The pattern in blood viral loads was different; here 100% of participants taking elvitegravir had fully suppressed viral loads by week twelve but one-third of participants taking either rilpivirine or darunavir still did not have fully suppressed viral loads.

This study appears to indicate that seminal viral load takes considerably longer to fall to undetectable levels in people taking darunavir, and maybe other protease inhibitors, than it does in people taking third drugs of the other two classes.

It also implies that people may be infectious for longer. Conventionally, the World Health Organization defines the viral load below which infection is very unlikely as 1500 copies/ml in blood. The infectiousness limit in semen is less well characterised, but while viral load in semen had fallen to below 1500 copies/ml within a week of starting therapy in people taking elvitegravir or rilpivirine, it took two weeks to fall below this level in people taking darunavir. A seminal viral load of 100 copies/ml was reached on average by people taking the first two drugs four weeks after starting elvitegravir or rilpivirine, but only between weeks eight and twelve in people taking darunavir.

The study was conducted by the University Hospital of Virgen del Rocio in Seville in Spain. The participants were all gay men with CD4 counts over 200 cells/mm3 and viral loads off treatment between 1000 and 100,000 copies/ml.

There were slight differences between the randomised groups at the start of the study, though its small size means these differences were not statistically significant. The people taking darunavir were on average older than the people taking the other two drugs (42 versus 33 and 32). They also had somewhat lower CD4 counts (466 cells/mm3 versus 517 and 659 cells/mm3 in participants taking elvitegravir and rilpivirine respectively). After 24 weeks on therapy, however, greater CD4 increases in people with lower CD4 counts meant these differences were less marked (719, 753 and 656 cells/mm3 in people taking elvitegravir, rilpivirine and darunavir respectively).

Participants randomised to take darunavir also had somewhat higher seminal viral loads at baseline (14,800 copies/ml versus 7600 and 6200 copies/ml in people taking elvitegravir and rilpivirine respectively). They also had lower blood viral loads than the other two groups. This meant that the ratio of seminal viral load to blood viral load was higher (0.95 in the darunavir-randomised group versus 0.46 for elvitegravir and 0.30 for rilpivirine).

As stated, however, these differences were not statistically significant. What was significant was that by week twelve all the participants taking elvitegravir or rilpivirine had undetectable viral loads (defined as <30 copies/ml) in semen, but only 58% of the participants taking darunavir. Also significant was the fact that all the participants taking elvitegravir had an undetectable viral load in blood by this time (defined as <20 copies/ml) but only 67% on the other two drugs.

Viral load in semen fell considerably more slowly in people taking darunavir than in people taking the other two drugs. After just one week on therapy, viral loads in blood had fallen 1.97 log10 copies (95-fold) in people taking elvitegravir and 1.91 log10 copies (81-fold) in people taking rilpivirine but only 1.22 log10 copies (17-fold) in people taking darunavir.

In semen, the corresponding falls were 28-fold on elvitegravir, 19-fold on rilpivirine, but only 3.4-fold on darunavir.

This would imply that after a week on therapy, the average blood viral loads in patients on elvitegravir, rilpivirine and darunavir would be 179, 258 and 929 copies/ml respectively, and the seminal viral loads 271, 326 and 4353 copies/ml respectively.

In terms of drug levels, while drug levels measured in semen were more than twice the EC90 (the level required to reduce HIV replication by 90%) in the case of elvitegravir and rilpivirine, they were only above twice the EC90 in 34% of patients on darunavir.

Drug levels tend to be higher in blood than semen with most HIV drugs, and most of the protease inhibitors have poor seminal penetration. There was only 10% as much darunavir in the semen of patients on this drug as there was in their blood, whereas there was 19% as much rilpivirine and 40% as much elvitegravir.

There was no correlation between drug levels in the blood and levels in the semen in individual participants.

Once participants had fully suppressed viral loads, the drugs all seemed to be as efficient as each other in maintaining viral suppression. There were two ‘blips’ in participants taking elvitegravir, both at week 18 (to 144 and 125 copies/ml); one in a participant taking rilpivirine (to 190 copies/ml at week six) and one in a participant taking darunavir (to 95 copies/ml at week eight).

This is the first randomised study comparing the effects of different drug classes on viral load in the semen of drug-naive men with HIV. While it was already known that viral load tends to drop more slowly in semen, which was also seen in this study, it was thought that only the integrase inhibitors cause a faster-than-average drop. This study unexpectedly showed that a drug of the second-generation of NNRTIs, rilpivirine, suppressed HIV in semen just as fast and there is no reason to assume that its closely related sister drug etravirine (Intelence) would be different.

CD4 rises were similar in all the drugs studied so if the main consideration is health rather than infectiousness, speed of viral suppression matters less, but if this matters to the patient, e.g. because they have HIV-negative partners, then protease inhibitors would appear to be not such a good choice.


Gutierrez-Valencia A et al. Viral kinetics in semen with different antiretroviral families in treatment-naïve HIV-infected patients: a randomised trial. Clinical Infectious Diseases, e-pub ahead of print, 2017. doi: 10.1093/cid/cix358. Abstract here.