There are 5,000 or so delegates registered at the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention currently taking place in Sydney. Only one of these delegates turned up for a press conference launching a new, highly accurate HIV viral load testing assay, developed specifically for resource-limited settings.
Move forward 24 hours and the importance of low cost viral load testing for millions of HIV-positive patients worldwide was made crystal clear.
Dr Steve Deeks, one of the world’s leading authorities on treating drug-resistant HIV, told a well attended session on recent developments in resistance, that intensive viral load monitoring was essential for patients taking NNRTI-based regimes – the first, and often the only treatment option in poorer countries – as failure to switch from a virologically failing regimen within six months leads to a significant and sustained risk of death.
Viral load monitoring is a routine aspect of HIV care in industrialised countries. It can be used to guide decisions about the best time to initiate antiretroviral therapy. Guidelines, such as those of the British HIV Association, recommend that everybody who is taking potent HIV therapy should have their viral load monitored at regular intervals to make sure that their virus is suppressed to undetectable levels.
But viral load testing requires considerable resources – indeed some UK NHS HIV clinics are reducing the frequency of viral load testing in treatment-naïve patients because of cost.
The situation is even more troublesome in many resource-limited settings. First, it is necessary to have access to testing equipment, which requires a sophisticated laboratory, highly trained staff and a reliable supply of chemicals and other materials used in running viral load tests based on polymerase chain reaction (PCR).
Then there is the expense of on-going, routine monitoring of patients - each individual viral load test can cost up to $80 US. Many taking HIV therapy in poorer countries therefore fail to have their viral load monitored – and stay on virologically failing regimes. And in many poor- and middle-income countries these failing regimens are usually NNRTI-based.
Even when they receive antiretroviral therapy, HIV-positive individuals in Africa and other poor- and middle-income countries still have a significantly higher risk of death in both the short and longer term than antiretroviral-treated patients in industrialised countries. This is the case even after controlling for factors such as other serious illnesses, poverty and malnutrition.
Dr Steve Deeks, of the University of California, San Francisco, suggested a possible reason for this. He highlighted data from the SCOPE cohort in San Francisco, originally published in 2000, 2002 and 2003, demonstrating that sustained CD4 cell gains are possible even in the presence of viral replication.
Patients in this cohort had been managed by Dr Deeks and colleagues using a conservative treatment strategy, reflecting the limited number of drugs available for second or third line treatment in the late 1990s. Patients remained on their initial regimen (mainly nelfinavir-based).
In patients who sustained a 2 log reduction in viral load for 96 weeks on potent therapy, CD4 cell counts had risen by a median of 200 cells/mm3 by the end of year 2.
In comparison, among 74 patients who did not sustain viral suppression, but still had viral load at least 1 log below baseline at week 96 (partial viral suppression still being sustained by their initial regimen), CD4 counts continued to rise throughout the follow-up period, although the increase was somewhat blunted during the second year, coinciding with viral rebound (+150 cells/mm3).
In 49 patients with a transient virological response (rebound within 24 weeks of starting treatment), a sustained increase in CD4 cells was still evident at the end of year 2, albeit blunted (+100 cells/mm3).
Although CD4 cell gains despite virologic failure have been reported mainly from cohorts of patients treated with protease inhibitors, Dr Deeks reminded the audience that a similar phenomenon has also been reported in NNRTI-treated patients in the ART-LINC collaboration of developing country cohort studies. Fifty-nine per cent of patients who experienced virologic failure nevertheless gained CD4 cells by month 6 of treatment (compared to 70% of PI-treated patients).
But he also presented still unpublished data from two North American cohorts demonstrating that a longer duration on failing NNRTI-based therapy is associated with a higher risk of death. The study combined 982 patients from two cohorts, Johns Hopkins University and the University of North Carolina, Chapel Hill, with 3414 person years of follow-up. 742 patients had failed a PI-based regimen and 240 had failed a non-PI-based rgeimen.
After adjusting for HIV RNA, CD4, gender, age, race, intravenous drug use, calendar date, date of first antiretroviral therapy, prior NRTI exposure, AIDS events and HIV diagnosis date, patients with a delayed switch from NNRTI-based therapy had a significantly elevated risk of death. That elevated risk became apparent within months, said Deeks, and persisted even after treatment was switched.
In contrast, the patients who remained on failing PI-based therapy showed a much slower increase in the risk of death, and the difference in risk between those who switched earlier or later was much less pronounced.
Furthermore, two previously published cohort studies have shown a similar elevation in risk of death for patients who fail NNRTI-based therapy (Hogg; McArthur).
Another factor which independently explains the greater risk of death, and its potentially rapid onset, after the failure of an NNRTI-based regimen is the background level of immune system activation. Immune activation – caused by infections and allergens – speeds CD4 cell decline, regardless of viral load levels.
Greater CD8 cell activation is associated with lower CD4 gains during PI-based therapy (Hunt), yet protease inhibitor- resistant virus is associated with lower levels of immune activation when compared to wild-type virus, suggesting that one explanation for sustained CD4 cell gains despite failing PI-based therapy is its ability to damp down immune activation, even if it cannot control HIV.
But when Deeks compared immune activation levels in his patients in San Francisco with those in Uganda, using the same laboratory and controlling for viral load and CD4 count, he found significantly higher CD8 cell activation in Ugandans (but not CD4 cell activation) (p<0.001).
The implication of these data, argued Deeks, is that some form of viral load monitoring is necessary for patients taking NNRTI-based regimens in resource-limited settings. CD4 cell counts do not provide enough information, and as data due to be presented on the final day of the conference will show, are not a reliable marker of treatment failure.
Viral load testing is becoming more accessible in resource-limited settings, and the sparsely-attended press conference at the IAS conference was told of the launch of a new test, specially designed with poorer countries in mind. The ExaVir< Load test version 3 is manufactured by Cavidi. Earlier versions of the assay are already used in many low-income countries.
The new version is 30% faster and has 50% greater capacity Although designed specifically for use in remote settings, the Exa Vir Load version 3 has been shown to be highly accurate when compared to commercial assays by independent researchers in Australia. Instead of using a PCR platform, the assay uses the ELISA system used for HIV antibody testing, and looks for reverse transcriptase, an HIV enzyme that indicates viral replication.
The assay is already in use in Kenya, Zambia, Botswana, Zimbabwe, Nigeria and China according to Cavidi, but is likely to be of greater interest once researchers have optimised its use with dried blood spots on filter paper. This will allow blood to be drawn many miles from a laboratory and sent to a central site where viral load analysis for a large number of treatment centres can be carried out.
Hopefully, the increasing availability of low-cost diagnostic tests will mean that increasing numbers of antiretroviral-treated patients in poorer countries can have the level of viral load monitoring patients in industrialised countries often take for granted.
Deeks S. When to switch? First line failure and its implications. Fourth International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention, abstract TUBS103, Sydney, 2007.