CROI: Recent infection responsible for one-quarter of HIV transmission amongst London's gay men

This article is more than 16 years old. Click here for more recent articles on this topic

Recent HIV infection appears to be responsible for at least a quarter of onward HIV transmission according to two studies presented to the recent Fifteenth Conference on Retroviruses and Opportunistic Infections in Boston. In addition, one of the studies – focused primarily on gay men in London – suggests that rapid onward transmission may have driven the substantial increase in HIV diagnoses seen in gay men in the UK between 1999 and 2003.

Contact tracing vs. phylogenetic analysis

Understanding who is most likely to transmit, and how sexual networks interact with each other is important if we are to understand how to curb the HIV epidemic. In the past, epidemiologists have used traditional contact tracing – where individuals recently diagnosed with HIV name the partner(s) they believe have exposed them to the virus – although, contact tracing has its limitations.

A recent study (Resik, 2007) that examined the genetic relatedness of two transmission networks in Cuba previously established by contract tracing found that between 60% and 70% of presumed transmission events were rejected – i.e., subtypes did not match or else phylogenetic analysis found no relationship between two individuals in the network.

The authors of that study argued that their “analysis indicated that contact tracing, when combined with time delays in diagnosis and in sample collection, should be taken with caution.”


phylogenetic analysis

The comparison of the genetic sequence of the virus in different individuals in order to determine the likelihood that two or more samples are related. This involves creating a hypothetical diagram (known as a phylogenetic tree) that estimates how closely related the samples of HIV taken from different individuals are. Phylogenetic analysis is not a reliable way to prove that one individual has infected another, but may identify transmission clusters, which can be useful for public health interventions.


Studies aim to give information that will be applicable to a large group of people (e.g. adults with diagnosed HIV in the UK). Because it is impractical to conduct a study with such a large group, only a sub-group (a sample) takes part in a study. This isn’t a problem as long as the characteristics of the sample are similar to those of the wider group (e.g. in terms of age, gender, CD4 count and years since diagnosis).

drug resistance

A drug-resistant HIV strain is one which is less susceptible to the effects of one or more anti-HIV drugs because of an accumulation of HIV mutations in its genotype. Resistance can be the result of a poor adherence to treatment or of transmission of an already resistant virus.

resistance testing

Laboratory testing to determine if an individual’s HIV strain is resistant to anti-HIV drugs. 


In HIV, different strains which can be grouped according to their genes. HIV-1 is classified into three ‘groups,’ M, N, and O. Most HIV-1 is in group M which is further divided into subtypes, A, B, C and D etc. Subtype B is most common in Europe and North America, whilst A, C and D are most important worldwide.

However, they also argued that phylogenetic analysis is not without its own set of limitations. Since HIV is constantly – and rapidly – mutating, they note, the further away from the transmission event that the sample under analysis is taken, the less likely an accurate picture of genetic relatedness will be. Also, the accuracy of phylogenetic analysis depends on the number of geographically and epidemiologically relevant samples used in the analysis.

Finally, they noted, “the fact that the phylogenetic signal might fade when sampling is performed at considerable times after transmission has important implications for forensics analyses that use phylogenetic inference to test transmission hypotheses.”

Using phylogenetic analysis for tracing sexual networks

Dr Gareth Hughes, presenting on behalf of his colleagues from the University of Edinburgh and the UK Collaborative Group on HIV Drug Resistance, brought up the Resik study at the start of his presentation on using phylogenetic analysis to better understand the dynamics of the HIV epidemic among gay men, and other men who have sex with men (MSM), in London between 1997 and 2003.

He said that phylogenetic analysis had found major discrepancies between contact networks and transmission networks, but that previous studies had not used enough samples to be able to more accurately describe the nature and structure of transmission networks.

To remedy this limitation, he and his colleagues used stored blood samples taken for routine drug resistance testing from a large HIV clinic in London to provide, as Dr Hughes put it, “a dense sampling of the UK HIV epidemic” between 1997 and 2003.

The sample included total of 2126 patients, of whom 75% were gay men or other MSM. They examined the oldest available stored blood sample, in order to get as close to the transmission event as possible.

However, since these were samples taken for resistance testing – and during the study period resistance testing was used primarily for treatment-experienced patients – even the earliest sample could have been taken years – or even decades – after transmission took place. As Resik and colleagues have noted, this somewhat reduces the likelihood of highly accurate phylogenetic analysis.

The investigators analysed the part of the sequences least likely to have been affected by drug resistance mutations (third codon positions of full protease and partial reverse transcriptase sequences), comparing all 2126 sequences with each other.

They found that 483 patients had closely related sequences (i.e. were less than 5.4% different), of which 402 were subtype B and 362 of these were in gay men and other MSM.

Within the 402 closely related subtype B sequences, they found nine clusters (closely related sequences) consisting of six or more individuals.

The role of primary and early chronic HIV infection

Using a method known as a ‘relaxed molecular clock’ in order to estimate the time (measured in six month periods) between transmission events, they further analysed six clusters that included ten or more individuals (two clusters with ten individuals, and one cluster each with twelve, 18, 20 and 30 individuals). These six clusters included about 25% of the cohort with related virus.

The median estimated time between acquisition of HIV and onward transmission was found to be fourteen months, but around 25% of HIV transmission events were estimated to have occurred within six months of infection – i.e. during primary infection – and more than 90% of all HIV transmission events were estimated to have occurred within 42 months of infection – i.e. during early chronic, and most likely undiagnosed and/or untreated, HIV infection.

A similar study from Switzerland, presented as a poster, and which used stored drug resistance samples from a cohort previously reported here, found that at least 30% of recent HIV infections between 1995 and 2005 formed phylogenetic clusters, suggesting that primary infection accounts for a large proportion of new infections.

By comparison, an arguably much more rigorous 2007 phylogenetic study of recent seroconverters from Quebec, Canada, found that primary HIV infection accounted for half of all onward transmission in the province.

More infection or more testing?

When the investigators related these timed transmission events to actual calendar years, they found that over 65% of transmissions were estimated to have occurred within a five year period, 1995-2000. “Given the expected lag between infection and diagnosis,” noted Dr Hughes, “this corresponds well with the observed increases in HIV diagnoses [among gay men and MSM] in the UK.”

However, this contradicts findings from the UK’s Health Protection Agency (HPA) which examined HIV diagnoses amongst gay men and other MSM in the UK between 1997 and 2004 and which concluded that the increase in HIV diagnoses seen since 1997 reflected an increase in HIV testing rather than a rise in the number of men infected each year.

When asked about this during the question and answer session that followed, Dr Hughes admitted that this, too, was a possibility, and that “this study was really undertaken as a way to show that this methodology can work and the type of information it can provide.”

Dr Hughes went on to show to that when they expanded the dataset to over 8000 patients from all over London, all of the previously seen clusters expanded, and six linked together to form one large group. In fact, the largest cluster in the single clinic study (30 patients) grew to over 500 individuals in the pan-London dataset.

He concluded that since their estimated dating of transmission events coincides with an increase in HIV diagnoses for gay men and other MSM in the UK, their analysis suggests that rapid transmission may drive sub-epidemics of HIV amongst gay men and other MSM in the UK.

Privacy and criminal liability concerns

During the question and answer session, Dr Hughes was also asked whether it was possible to identify the “index patient” – the person thought to be responsible for introducing a particular strain into the network.

Given concerns over issues of consent and privacy and the use of stored blood samples as evidence in criminal prosecutions for ‘reckless’ HIV transmission, Dr Hughes’ answer was somewhat reassuring.

He said that since all samples were completely anonymised, it was not possible to identify individuals used in this research, and besides, he added, phylogenetic analysis is unable to show direction of transmission.

He told the conference that he and his colleagues now plan to expand the analysis to include more than 25,000 patients throughout the UK, using sequences obtained from the UK Drug Resistance Database in order to “gather more information about more recent dynamics of the epidemic.”


Hughes G et al. Recent phylodynamics of the HIV epidemic among MSM in the UK Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 13, 2008.

Resik S et al. Limitations to contact tracing and phylogenetic analysis in establishing HIV type 1 transmission networks in Cuba. AIDS Research and Human Retroviruses 23 (3); 347-356, 2007.

Yerly S et al. The contribution of individuals with recent infection to the spread of HIV-1 in Switzerland: a 10-year survey. Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 512, 2008.