Case of MDR reinfection with MDR virus

Simon Collins, HIV i-Base

Incidents of HIV reinfection or superinfection are generally difficult to isolate and report as the structure for patient management are not geared to be able to look for this. When cases are identified (see report on last years meeting in HTB 4/7, Aug/Sep 2003) they tend to be picked up by clinicians who notice unexpected virological rebound or increases, that are linked to a personal patient history, and then confirmed by technology that is not generally easily accessible to practicing clinicians.

In one case of multiple drug resistant (MDR) transmission with low level viremia reported by Brenner and colleagues from McGill University AIDS Centre, Montreal, superinfection occurred with a second heterologous MDR strain. [1] This was confirmed by phylogenic and clonal analysis of viruses from the index case and source partners. In cell culture, both of these transmitted MDR species showed reduced replication capacity relative to the WT virus isolated from the first source partner after treatment interruption.

This report came from a larger study looking at transmission of resistance in the Montreal Primary Infection Cohort, which included wild-type infection in 15 patients, resistant infection in ten patients and nine patients infection with MDR virus.

Longitudinal follow up of this group over 2-6 years showed that mutations in patients infected with resistant virus (other than M184V) generally persisted over time, including those patients with MDR infection. This was in contrast to reversion to wild-type virus, in a number of source partners for these MDR infections, following a treatment interruption.

The global history of HIV clades and recombinant viruses clearly shows the epidemiological significance of recombination. Several presentations at the meeting looking at recombination including a study from Frank Maldarelli and colleagues [2]. This study estimated similar rates of recombination in treatment naïve and experienced patients and suggested that at relatively low rates of viral load of 3000 copies/mL this pool of infected cells is sufficiently large to permit frequent recombination.

Although a fine point, it is clinical important to realise that, for example a strain with M184V mutation from earlier treatment history combines with, for example K103N resistant strain from later use of NNRTI, to form a dominant strain that contains both 184V and 103N. A crude estimate from sequential samples from a recently infected patient was that >20% of infected cells in vivo are likely to be infected with two or more proviruses.


The growing collection of documented case reports or reinfection or ‘superinfection’ clearly establishes this as certain possible risk.

This doesn’t mean that HIV-positive people have to use condoms every time they have sex, but it does mean that other risk factors are important when both partners are HIV-positive – especially when they have different treatment histories and resistance patterns.

Risk is a continuum. For example, any additional risk is likely to be extremely low for monogamous couples, who both have undetectable viral load, and the same treatment and resistance history. Higher viral load, as with all cases of infection, is likely to be an independently increase the risk.

Someone who is treatment naïve, or currently on successful treatment could jeopardise future health and treatment following reinfection from a partner who has any drug resistance, however extensive, and it is plausible that risk of reinfection is likely to increase with traditionally associated factors: viral load, mode of exposure, concurrent sexually transmitted infections etc.


  1. Brenner BG, Turner D, Wainberg MA et al. Natural history of transmitted drug resistant and wild-type infections and superinfection in the Montreal Primary HIV-1 Infection (PHI) Cohort. XIII Intl HIV Drug Resistance Workshop, 2004. Abstract 90. Antiviral Therapy 2004; 9:S101.
  2. Maldarelli F, Kearney M, Coffin J et al. HIV-1 populations are large, highly diverse, and characterised by frequent recombination in drug-naïve and drug-experiencecd individuals. XIII Intl HIV Drug Resistance Workshop, 2004. Abstract 45. Antiviral Therapy 2004; 9:S54.

Links to other websites are current at date of posting but not maintained.