HIV superinfection may cause increasing viral loads and a second seroconversion illness

Infection with a second strain of HIV (superinfection) may have medical consequences, according to a presentation at the 15^th British HIV Association (BHIVA) conference.

A small study of eight gay men with HIV who were not on treatment and had increases in their viral load found two whose viral load increases were clearly due to infections with a second strain of HIV.

In one case the patient‘s second strain of HIV was drug-resistant. He also experienced a recurrence of acute HIV symptoms which required hospitalisation for suspected meningitis and a large, though temporary decrease in CD4 count. In the other case the patient’s original strain of HIV, which was drug-resistant, was replaced by an apparently stronger non-resistant strain and his viral load increased from around 3000 to half a million. However he maintained a CD4 count over 1000 and his viral load had returned to 3000 a year later.

In this prospective study at the Royal Free and Royal London Hospitals, gay men who were diagnosed with HIV, did not start HIV treatment, continued to have unprotected anal sex after diagnosis, and had a more than threefold (0.5 log) increase in their viral load during routine follow-up appointments were asked to have their HIV genetically analysed in more detail. The eight men who met these criteria and joined the study were diagnosed between 2004 and 2008 and their average age was 30, younger than the average age for diagnosis in gay men.

An increase in the men’s viral load was noted an average of 19 months after diagnosis in these eight cases. When this was detected, the genetic make-up of the men’s HIV was re-analysed. In six cases there was no difference in the viral sequences, but in two cases the researchers found a completely different strain of virus which had ‘taken over’ from the first one. There was no overlap between the genetic sequences of the two viruses, indicating that this was not a case of recombination (two viruses combining to make a new one) but of two populations of viruses, a stronger and a weaker, co-existing.

In the first case the superinfection happened five months after the initial one. The patient was diagnosed during the acute phase of his initial infection. He had HIV seroconversion symptoms (a flu-like illness and severe headache) and a viral load of a million which subsequently declined to 40,000 copies/ml.

However five months after infection he experienced a return of the same symptoms, requiring hospitalisation and an MRI scan for suspected meningitis.

His viral load went back to 160,000 and subsequently increased further to nearly a million. His CD4 count fell temporarily from 430 to 240 cells/mm³ though it subsequently rebounded to about 390 cells/mm³. He continued to have high-risk sex and over the next six months also acquired first infections of syphilis and herpes. His second virus had two resistance mutations to nucleoside drugs.

In the second case the patient acquired a second strain three years after the first. In this case his original virus had two drug resistance mutations. When his viral load increased he was given another resistance test 3.5 years after diagnosis which showed no evidence of the mutations. Analysing previous samples showed he had acquired a superinfection six months previously.

The researchers conclude that HIV-infected patients who continue risk behaviour are at risk of superinfection “both in the early and established phases of the disease”. They recommend that all patients not on treatment who experience unexpected viral load increases should be screened for superinfection (though the kind of intensive phylogenetic screening used in this study is purely a research tool, costing £10,000 per patient).

They argue that this study adds to the case for starting HIV treatment early in patients who continue risk behaviour, both because it may cause illness and pass on drug resistance, and also because it may make people considerably more infectious.