HIV coinfection, reinfection and superinfection

1 September 2003. Related: Conference reports, Hepatitis coinfection, HIV prevention and transmission, Drug resistance, Intl Drug Resistance Workshop 12 Mexico 2003.

Simon Collins, HIV i-Base

Several abstracts provided additional data on cases of HIV superinfection or coinfection. The term coinfection is usually preferred when there is evidence that the initial infection occurred with two or more different viral strains at the same time, or before an immune response to the first virus has developed. (the latter is sometimes called serial infection).

The term superinfection refers to instances when a second distinct virus infects an individual after they have developed an immune response to the first. The clinical concerns regarding superinfection are essentially twofold. Firstly the second virus may be more virulent and fitter than the initial strain. This may lead to a more rapid disease progression than might otherwise occur with the first virus only. Secondly, it is possible that the second strain may harbour drug resistant mutations, which may compromise the recipient’s future or current treatment options.

On a very basic level the plausibility of coinfection or superinfection is evidenced by the very wide genetic variability of HIV and the frequency of recombinant viral forms in existence. (For viral recombination to occur at all infection of a single cell by two different viruses is required). However, detection of superinfection and coinfection is not straightforward and is an extremely labour intensive scientific process. In fact most cases have only been detected when individuals have been part of intensive primary infection studies. It is unlikely that superinfection will be the focus for large-scale studies and this has perhaps driven some of the scepticism of reporting and discussion.

Perrin and colleagues from the Swiss HIV Cohort Study followed five IVDUs who were either coinfected or superinfected with distinct viral types to study the persistence of different viral strains within a single individual over time. Three patients were coinfected with two different strains at the time of primary infection (sub-type B and Circulating Recombinant Form-11 (CRF-11)) and two patients initially infected with sub-type B and were later superinfected with CRF-11. All patients had been identified from their IVDU cohort, and cases of superinfection had been detected following unexpected clinical events.

The three coinfected patients were followed for 14, 20 and 24 months respectively and subtype specific PCR continued to detect both viruses over this period. Two of these individuals had viral loads >400,000 copies/ml.

The two cases of superinfection were both originally infected with subtype B and had previously controlled their HIV without treatment, maintaining CD4 counts >500 cells/mm3 and viral loads <50 copies/ml for three and five years respectively. Superinfection with CRF-11 in these two cases was associated with high viraemia, rapid CD4 drop and acute retroviral syndrome. Interestingly CRF-11 was the only detectable virus shortly after the time of superinfection and during subsequent monitoring, although both viruses remained detectable in proviral DNA.

Palmer and colleagues provided further details on a patient infected with multidrug resistant virus that was detected during primary infection and reported at the 9th CROI (Daar et al, abstract 96). Within two months of infection, viral load dropped to <1000 copies/ml without treatment but four months after this rebounded to 10,000 copies and at this time point showed no evidence of resistant mutations.

Phylogenetic analysis showed two distinct subtype B viruses at different times after the initial infection. At month one this was entirely resistant virus (0.025% viral diversity) with all sequences containing 69SS insertion and K103N. Subsequent samples at months five, 13 and 17 showed contained a different viral strain and were wild-type with regard to drug resistance associated mutations. Single genome sequencing showed the wild type virus to be almost homogeneous (0.007% sequence diversity) indicating very recent infection (diversity increased to 0.062 and 0.18% at month 13 and 17 respectively).

Readers interested in tracking these reports will be interested to know of a further two abstracts presented at the IAS meeting.

Burger and colleagues reported the case of a Kenyan woman who was infected prior to 1986 with subtype A virus. Complete RNA sequences from 1995 and 1997 were subtype A/C recombinants and heteroduplex tracking assays were unable to find evidence of subtype C in the 1986 samples.

Manigart and colleagues reported four cases of coinfection from a cohort of 147 commercial sex workers in Burkina Faso, two of which showed two distinct phylogenetic populations existing. Retrospective analysis of stored samples showed that each patient acquired a second virus at the same time that they experienced increases in plasma viraemia.

Although this study commented that superinfection is not an uncommon phenomenon, given the multiple opportunities for exposure, it is also surprising that it was detected as such a low level in the cohort as a whole.

Numerous posters on molecular epidemiology also documented geographical prevalence and development of both new and already recognised populations of recombinant virus within the diversity of HIV infections, including vertical transmission of dual infection.

Comment

These additional cases of superinfection add to the already published literature on this subject (Jost et al NEJM: September 2002; Altfeld et al. Nature. November 2002; Koelsch et al. AIDS May 2003).

They prove beyond doubt that superinfection with a second strain of HIV can occur with detrimental consequences to the individual affected. Several questions remain unanswered. Will superinfection with drug resistant HIV always lead to treatment failure in a person well controlled on drugs? Secondly, just how common is superinfection in day-to-day practice? A recent article published by Gonzalez et al in JID suggests this is a relatively rare event. However in the absence of the further clarification it is important to at least advise patients on the potential risks of superinfection.

References:

Unless stated otherwise, references refer to the XII International HIV Drug Resistance Workshop, Los Cabos, Mexico, 10-14 June 2003; published as part of Antiviral Therapy Volume 8 Issue 3.

Palmer S et al – Population genetics in HIV-1 superinfection. Abstract 62.
Perrin L et al – Co- and super-infection: persistent replication of both HIV-1 strains? Abstract 63.
Burger H, Fang G, Kuikero C et al – Recombination following superinfection by HIV-1. 2nd IAS Conference, Paris. 13-16 July 2003. Abstract 71.
Manigart O, Courgnaud V, Sanou O et al – HIV-1 superinfections in a cohort of commercial sex workers in Burkina Faso as assessed by a novel autologous heteroduplex mobility procedure, ANRS 1245 study. 2nd IAS Conference, Paris. 13-16 July 2003. Abstract 72.