A lively debate on the use of triple nucleoside reverse transcriptase regimens (NRTIs) ended the first day of the Third International AIDS Society Conference on HIV pathogenesis and treatment in Rio de Janeiro. The debate, chaired by the UK's Professor Brian Gazzard, focused squarely on whether the public health approach to antiretroviral therapy in resource-limited nations - which have large groups of populations that are not well-served by the current World Health Organisation (WHO) guidelines suggesting that first-line regimens are NNRTI-based (e.g. those co-infected with TB, women of childbearing age, those with HIV-2) - may lead to them receiving second-class anti-HIV therapy compared with well-resourced nations if WHO were to recommend triple NRTIs in their revised guidelines, due out this later this week.
The data, the data, the data
Roy Gulick, Associate Professor of Medicine at Weill Cornell Medical College in New York argued that all the data presented on triple NRTIs from well-resourced nations over the past two years show that, whilst they may have less drug interaction issues than NNRTI- or protease inhibitor (PI) -based highly active antiretroviral therapy (HAART), they are also less potent when it comes to antiretroviral activity.
He also suggested that Charles Gilks' own DART study, which will present 48-week results during Wednesday's late-breaker sessions, was not convincing. Citing 24-week results presented at CROI in Boston earlier this year, he said that intent-to-treat, missing equals failure analysis of 200 treatment-naïve Ugandans, 68% of whom are women, with a median baseline viral load of 334,000 acopies/ml and a CD4 cell count of 100 cells/mm3 where 68% achieved a viral load below 400 copies/ml and 51% achieved a viral load below 50 copies/ml, showed a response that was clearly inferior to NRTI- or boosted PI-based HAART.
He concluded that although not all triple NRTI-regimens are alike, even the most potent ones are only comparable to unboosted indinavir, nelfinavir and amprenavir regimens, none of which are currently recommended in well-resourced nations, and inferior to efavirenz-based regimens.
Public health approach
Professor Gilks, an architect of the World Health Organisation's `3 x 5` ;programme, argued that individualised patient care, intensive high-technology laboratory monitoring and a choice of 19 or more available antiretrovirals gave well-resourced nations the luxury of being able to discard triple NRTIs. However, the public health approach to antiretroviral treatment in resource-limited nations, which requires a population-based delivery of a fixed-dose standard of care focused on maximising survival with three available drug classes in two, or at most, three regimens, needs to strongly consider using triple NRTI regimens.
Without viral load monitoring, treatment decisions in resource-limited nations must be based on clinical and immunological markers, he argued. This may result in high-level resistance, but if only one class is used up, this still leaves a highly potent combination of a boosted PI with an NNRTI for second-line therapy, which may be the only other chance an individual has to control their HIV infection in their lifetime.
He admitted, however, that there is limited experience in using triple NRTIs in resource-limited nations, other than his own DART study, and that the two most potent components, abacavir and tenofovir, are not available as generics. He added that he is also concerned about abacavir's hypersensitivity and tenofovir's and bone mineralisation side-effect profiles, but concluded that although there is no ideal regimen, triple NRTIs have many advantages in the public health approach. "We need to examine them more closely and we need to use them more widely," he told the audience.
Inherently flawed
Joep Lange, Professor of Medicine at the Centre for Poverty-Related Communicable Diseases, University of Amsterdam, supported Roy Gulick's argument that there is overwhelming evidence of the inferiority of using triple NRTIs, from both controlled clinical trials and observational studies. He argued that efavirenz can be used with rifampicin - the only affordable TB treatment in resource-limited nations - without a dose adjustment from 600mg to 800mg. A study presented in Bangkok last year found a comparable virologic response rate on 600mg efavirenz compared with 800mg when co-administered with rifampicin, and "so in a resource-limited setting 600mg, efavirenz seems appropriate in patients being treated with rifampicin."
He also underplayed concerns over efavirenz-related birth defects during the first trimester of pregnancy as "a phenomenon that is enhanced by an utterly reprehensible and stupid US culture of litigation." And he argued that since DART is a single-arm study, it proves very little. He also pointed out that NRTIs cause mitochondrial toxicity, leading to lipoatrophy and lactic acidosis. He concluded by suggesting that triple NRTI regimens are inherently flawed because both NRTIs and tenofovir "need to be phosphorylated intracellularly and this process is not equally efficient in all cell types." Therefore, triple NRTI regimens do not exert triple drug pressure at the level of every HIV-infected cell and recent data from an animal model looking at tenofovir pre-exposure prophylaxis (PREP) suggest that this drug is no exception to the rule.
Global perspective
Dr Diane Havlir, of the University of California, San Francisco, supported Professor Gilks by arguing that emerging data suggest that select triple NRTI (AZT, 3TC and either abacavir or tenofovir) regimens are safe, simple to administer, associated with "respectable" reductions in viral load and CD4 increases, and have a favourable drug interaction profile. "From a global perspective," she said, "these combinations may represent some of the best options in some rather large populations," including those co-infected with, or at high risk for, HIV and TB; women of child bearing age, or pregnant or breast-feeding; and those with HIV-2.
Returning to the issue of women, who make up "half the world's HIV population", she argued that many women have no access to birth control, and may therefore not be able to control when they become pregnant or indeed know that they are pregnant during the first trimester. For women in this position, being on efavirenz-based HAART is severely problematic. When the same women need to take HAART to prevent mother-to-child transmission (MTCT), nevirapine use is problematic due to severe, life-threatening and sometimes fatal liver toxicity in women with CD4 counts over 250 cells/mm3, she added. Triple NRTIs may also be a good option for chronic treatment in women who have previously received single dose nevirapine to prevent MTCT and are now resistant to NNRTIs, she argued.
Do the least harm
During the lively audience debate that ensued, it was suggested that if NNRTIs couldn't be used, and boosted PI's need to be spared for second-line (or were not available due to a lack of cold-chain storage for ritonavir or Kaletra) then quadruple NRTIs ought to be considered, since emerging data suggests that AZT, 3TC, abacavir and tenofovir together are as potent as currently-approved first-line regimens. "We are stuck with imperfect drugs," Charles Gilks said when asked to summarise his position, "and a huge challenge is how to use them. 'First do no harm', and triple nukes are a simpler and safer strategy for many people."
"To do the least harm as possible," countered Joep Lange, in his summary, "you make the calculation by efficacy, and you are going to do that by avoiding triple NRTIs."
Triple nukes better than no treatment
Professor Gazzard summed up the debate by acknowledging that although triple NRTIs might not be attractive to those in well-resourced settings, "everyone in this audience who was convinced that triple nukes should be discarded has a moral obligation to go away from here and make sure that three million people are treated by 2005, because a much worse option than triple nukes is no treatment at all."