ASHM Report Back

Clinical posts from members and guests of the Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine (ASHM) from various international medical and scientific conferences on HIV, AIDS, viral hepatitis, and sexual health.

The Conference opened with a broad based plenary looking at the new landscape in HIV, often referred to as the HIV Testing 101 Workshop. This is a two hour session which will be on line shortly and really is an excellent overview. It starts out with a glossary of terms and then moves through technology; performance; programs; surveillance and the relationship between laboratory and strategy.

I strongly recommend that anyone setting out into the world of testing watch this session. The slides will all be up on the website some time after the conference and we will advise when this happens.

The USA has recently introduced a change algorithm for HIV diagnostic testing. This raises practical issues for laboratories. But an equally important issue for this conference is how laboratories support initiatives to increase testing (and timeliness of testing) and improve the care continuum.

Details can be found on the website  

Key HIV Testing Issues

Key issues in this meeting are how to get testing done early enough and also how to use the best test on an early-after-exposure sample. This will likely play out over the next few days. Clearly the cognitive distance between the laboratory and the clinic is narrowing here. Labs are trying to play a role in the clinical improvements that are sought in reducing the time between exposure and testing. Yet with the increase of self testing, and large scale community clinics with the capacity to perform more complex tests, the laboratory is coming much closer to the community.

With this comes the big question for me: How does one get this information to the person needing testing, at the time that they need it? The Achilles heal in any algorithm would seem to be the differentiation of the population upon which it is performed.

Joanne Stekler (Seattle) discussed this in the breakfast session today. Indicating that the greatest variation between yield on different tests is how differentiated the sample is.  Population-based screening is low yield in low prevalence settings and yield rises dramatically when more targeted testing is performed.

Increased infectivity during seroconversion and early in infection mean it is vitally important to get people to test during this period. Though this has not been discussed here yet, the role of PEP in this context should be reconsidered.


Day 2 of IAS2015, I attended a very full lunch poster discussion session entitled HIV Testing: The Gateway for Everything.

Sheri Lippman presented two posters. The first presented results from a cluster randomized controlled trial comparing the ‘one man can’ campaign in South Africa which aimed to engage men into testing through community mobilization. Community mobilization was found to be associated with higher testing uptake, though not equally for all CM domains. The main three domains which had the greatest impact were consciousness, concerns and collective action.

Sheri’s second presentation was on a pilot study examining feasibility and acceptability of self-testing in transgender women in San Francisco. 50 HIV-ve trans women were provided home HIV test kits and asked to utilized the tests once a month for three months, behavioral data was also collected at baseline and three months. 96% of study cohort had multiple recent partnerships and 80% had engaged in sex work, so as with other transgender communities, a pretty high risk population. Most found the test easy to use and would recommend to their friends (>90%), 68% would use the test again. The main reasons provided for not using again were around gaining access to counseling services that facility-based testing provided, and most participants had tested with a partner or a friend. A marketing strategy which provided two home-based tests aimed at testing with a partner could be an efficacious method for targeting this hard-to-reach population.

Laura Derksen from the London School of Economics presenting results from a cluster-randomized trial examining methods to reduce stigma in a community based setting in Malawi. 122 Malawi villages were targeting (60 intervention villages and 62 control) which reach two thirds of the target 15-49yrold population. In the control villages information was provided on the benefits of ARV including prolonging life and reversal of AIDS. While in the intervention villages, in addition to the control information, information regarding reduced likelihood of transmission in partnerships where the HIV+ve partner properly adhered to ARV was also propagated. The overall concept being that if they could reduce stigma, by showing that individuals who tested were in fact safer sexual partners then rates of testing would improve. The percent of the population having testing for HIV post-exposure was found to be  60% higher in the intervention arm compared to the control arm, and this was consistent for both genders.

Perhaps the most relevant to the Australian setting was a poster presented by David Katz, which examined self-testing as a method to increase overall testing frequency among high-risk MSM in Seattle. 230 HIV-ve MSM were randomized to have access to free HIV self-tests versus standard HIV testing for 15 months. The primary endpoint was the comparison of HIV testing frequency, secondary endpoints included non-inferiority in regards to behavioral markers of HIV risk acquisition. The mean number of test in the self-testing arm was 5.3 (4.7-6.0) compared to 3.6 (3.2-4.0) in the control arm, which was statistically significant (p<0.0001). Non-inferiority bounds were met for risk acquisition, which included difference in frequency of bacterial STI diagnosis at 15 months, likelihood of non-concordant anal intercourse at 3 months, and number of male non-concordant partners. However confidence intervals were wide for the secondary endpoints and the study not really adequately power to examine these associations.  

And finally the last two posters were presented by Sue Napierala Mavedzenge and Pius Tih Muffih respectively. Sue presented results of a feasibility study which examined the reliability of self-testing in both rural and urban settings in Zimbabwe. There was high sensitivity and specificity of interpreting self-test kits in both rural and urban settings, however slightly lower sensitivity in the rural setting, which was likely a result of lower literacy in this group. Some practical issues to encourage interpretation of the test (such as increased window size) were discussed.  Pius Tih Muffih provided the results of a very interesting study which examined integrating partner notification into Option B+ in Cameroon. Results looked promising with 823 women testing positive providing information on 840 partners of which 693 were notified, of whom 421 were tested for HIV and 139 new HIV+ve cases identified and linked to care. They had not observed any backlash in terms of violence to women as a result of the notification but this was highlighted as an issue which needs to be carefully monitored.






This workshop is being run adjacent CROI to facilitate participation. John Mellors stressed the ongoing importance of resistance testing and subsequent drug selection so as not to render treatment scale up futile. The focus of the meeting has also shifted to being largely focused on resistance as it impacts middle and low income settings. But there were considerable references back to high income countries, particularly in relation to resistance development in the context of PrEP.

The opening presentation from Shannon Hader reviewed the current position of resistance testing in the global environment of scale-up of ART and meeting the 90:90:90 testing, treatment and viral suppression targets. NOW was seen as the time where, through a shifting focus toward quality improvement and funding high impact programs, these targets might be realised by 2030. The contracting funding climate was what was seen as driving this need to get more bang from prevention and treatment buck.

Reducing cost from not unnecessarily changing to 2nd line therapy was seen as something resistance testing could inform. Likewise making sure that individuals starting PrEP are in fact HIV negative was seen as essential and something which could be determined by earlier use of viral load.

Neil Parkin questioned the constant need for gold standard, when such aims might mean that no change is achieved. This notion was revisited throughout both days. Neil was specifically looking at the efficacy of dry blood spot viral load and resistance testing. It was recognized the DBS uses a very small sample size, thus hindering detection at lower concentrations (largely because whole blood, unlike plasma, introduces other confounding material into the sample). But this was pitched against the collection and transport benefits of DBS sample collection. In some places conventional samples cannot be collected, transported or analysed due to cost and complexity.

Suzanne Crowe provided an overview of current point of care test development for CD4 and viral load. Suzanne defined PoCT not as the technology, but the extent to which a technology could be used easily and reliably in close proximity to the patient. Like a number of other speakers, Suzanne identified the need for training. And operator proficiency as one of the largest sources of performance variation between any of the tests. The ongoing need for CD4 in the context of viral load was discussed and CD4 was seen as having a continuing place if not a rosy future.

The later discussions focused on the clinical impact transmitted resistance and the role resistance might play in the roll-out of PrEP. Interesting here was the level of discussion about clinical and preventative applications. This was great to see as previously I have witnessed a big disconnect between the science and its implementation. Dan Kuritzke questioned whether transmitted AZT resistance really meant anything in the context of AZT-free treatment regimens. Likewise there was speculation about what was driving the identification of resistant virus among 13-19 yr old MSM in large scale surveillance studies from USA. This was thought to probably be behavioural and possibly associated with amphetamines and frequency of sex and sex partners.

Clearly resistance remains very important in guiding clinical choices and in mapping transmissions. Just how these two functions can dovetail was also topical. Is it ethical to link clusters and transmissions to programs to get people on treatment? And if you have information about where infections are occurring is it ethical to not use that information? Not having HIV transmission as a crime was seen as enabling this information to be used for clinical and scientific purposes, rather than punishment. Tulio de Oliveira identified this as a facilitator of clinical and surveillance programs in South Africa.

Many of the speakers from this workshop are also presenting at CROI so no doubt much more on this topic. 


Tagged in: croi2015

The discussion generated by this, quite complex topic, was stimulating. CDC reported that a dried blood spot sample gave comparable performance to a liquid serum/plasma sample, an obvious practical benefit in much of the world. They were reporting on performance for one of the available tests trying to infer recent infection rates, the Limiting-Antigen Avidity EIA. Then 2 other presenters raised issues about the rate of miscalculation of this recently licensed  LAg test, and the evidence that Clade D infection really confuses it. Finally, Alex Welte expressed his groups reservations about relying on any one of the 5 available tests yet, saying that viral suppression, by ART or being an elite suppressor, were 2 major causes of error. Issues then discussed were whether there was any agreement about what should be defined as the cutoff time for a recent infection, whether there should be an agreed viral load limit to incorporate in an algorithm, what would be an acceptable level of "False Recent Rate", how great the lag time would be between the current recent infection rate, and Incidence calculation "look back" rate, and whether combining 2 of the assays would make results reliable. Although work on improving these current tests and algorithms is proceeding in a somewhat cooperative way, new bio marker research is also being funded hoping to find a single novel marker to overcome current limitations. Bottom line, we could probably use these tools now if we didn't need to believe they were totally accurate. Abstracts  #1005,#1006, #1007, #1008

Tagged in: CROI2014

HIV Testing and Monitoring the Epidemic: New Tools for Patients and Populations. Session 51 Symposium

This was one of the conference’s last sessions and was very well attended.

HIV self-testing: Opportunities and Challenges. Abstract 162

Dr Julie Myers, New York City Department of Health

This was an important talk given that increased access to HIV point-of-care testing is one of the key aims of the Melbourne Declaration. It is worth watching on the CROI webcast.

Last year the FDA approved the Alere Determine HIV Combo home self-testing kit, also known as Orasure.

Dr Myers provided the results from a recent 5,000+ patient trial of the Orasure rapid HIV home testing kit, which was conducted across 20 sites in the USA.

These data were given to the FDA Advisory Committee last year by the product sponsor but have not yet been published.

In this study they found that 82% of people using the test were from high HIV prevalence areas.

5,055 patients were evaluable for test results. The overall prevalence was 2.12%. See slide below from Dr Myer’s talk for further results of this study.

Note that the sensitivity is only 91.67%. Of note when trained professionals use this test its sensitivity is over 98%. 

As we know this product went on to become licensed in the USA. Of note, people who use the kit have 24-hour phone access for support and for referral to sites for confirmatory testing.


There are few post marketing data available but there have been some published data to indicate that users of the test find it is ‘easy to use’.

How homebased self-testing tests are being used

Dr Myer then discussed the way that the tests are being used including by those who are at risk and by potential sexual partners. Dr Myers reviewed some recent publications that showed that rapid tests are being used to test prospective sexual partners and acquaintances and that no sex occurred with partners who tested positive. The numbers in these studies were fairly small however.

Concerns noted by Dr Myers

1. Propensity for self-harm and violence

Similarly although there are no data to support ongoing concerns that HIV home self-testing kits will lead to an increased risk for self-harm and suicide and inter-partner violence Dr Myers noted that more work and study needs to be undertaken in this area.

2. Lost opportunities

In addition Dr Myers noted that home self-testing kits mean there is an increased chance for patients to lose the opportunity to have additional STI testing and to be referred for other appropriate services.

3. False reassurance

A major concern is that some people will be falsely reassured that they are HIV negative when they have false-negative results.


Cost is a real concern because the kit costs US$40 and those most at risk of HIV are those who are most financially disadvantaged. Only 17-18% of people in the US and Europe would be willing to pay this amount for a kit in studies cited by Dr Myers.

5. Risk compensation

There are not yet any data on whether home HIV self-testing kits are associated with risk compensation- meaning an increase in risk-taking.

Dr Myers noted an important study presented at CROI 2013 Abstract#1064 by Katz et al where modeling done around MSM in Seattle suggested that switching from clinic to home-based testing using Orasure would lead to an increase in HIV transmission.

6. Linkage to care

There are not data to show whether use of these tests will lead those people who test HIV positive to seek appropriate medical care. Although 88% of people in the self-test clinical trial presented to the FDA last year said that they would ‘definitely get follow-up’ this remains an important area to monitor and study.

7. Resource constrained settings

Limits of supply chain and also quality assurance.

 During question time Dr Myers got hammered by two audience members who were particularly concerned about how little data there are available about whether home-testing would improve linkage to care in the USA. However, as another audience member noted, the US linkage to care using clinic based testing is pretty poor! This is also an issue that Australia must address.


Point of care assays for immunological and virological monitoring of HIV disease. Abstract 163.

Dr Ilesh Jani, Mozambique

Dr Jani gave a fantastic talk and overview about point of care testing for HIV diagnosis and monitoring regarding POCT for CD4 cell counts and HIV viral load including for children.

Amongst other things he discussed implementation issues and noted:

1. POCT is not error proof. It IS error prone in all phases of testing

2. Implementation of point of care tests is not always done efficiently.

In Mozambique his team's data showed that despite having POCT CD4 instrument on site, they lost 24% of people BEFORE people got tested with the POCT CD4 test. They only managed to test 50% of people on the same day as their HIV diagnosis. Median time to get a CD4 count was 3 days (Jani et al 2011

One key point that I took home from him was that whole blood viral load testing is likely to become common in the future. In patients whose plasma HIV viral is undetectable they may still have a detectable HIV viral load in their whole blood test. He explained that we don’t really know what this means clinically and we don’t know the correlation between plasma and whole blood HIV viral load testing


Accurate cross-sectional incidence testing. Abstract 164

Dr Laeyendecker, NIAID and Johns Hopkins

Dr Laeyendecker gave a masterful talk on this subject, which was valuable for a naïf such as myself. It’s a must watch talk for those who wish to understand how to optimize testing algorithms to assess HIV incidence in populations

Key points include (1) that there is a difference between how HIV clade D versus clades A and C perform in these testing algorithms wherein the tests overcall the prevalence of clade D infections and (2) that multi-assay algorithms perform best for HIV incidence testing in stable, expanding and waning HIV epidemics.


Viral load measures: patients, populations and interpretations. Abstract 165

Dr Irene Hall, CDC

This talk was good. The main point I learnt was that community viral load refers to the viral load of those people who have been diagnosed with HIV. Population viral load refers to all people in a population who are HIV infected including those who are unaware of their diagnosis. This is an important distinction because community viral load does not accurately reflect HIV transmission potential in the community.

Dr Hall noted an excellent oral presentation given at CROI 2013 Abstract 96 which reported on the first study in Swaziland of their HIV population viral load. This household-based counseling and testing study evaluated 18,169 individuals. 5802 individuals were identified as being HIV+ and, of these, only 67% knew that they were HIV+ and, of these, only 50% reported current ART use. A high viral load, defined as >50,000 copies/ml was found in 35% of the total population, (cue heartsink feeling here).

Overall an excellent session.



Tagged in: CROI2013
Twitter response: "To protect our users from spam and other malicious activity, this account is temporarily locked. Please log in to to unlock your account."