High-level science at its best says Anna Poppa of the 11th Retrovirus
Conference in San Francisco
In the HIV conference calendar, America’s annual Conference on Retroviruses and Opportunistic Infections (CROI) consistently attracts the highest quality scientific presentations, often with less fanfare than that accompanying larger meetings.
For example, though most people cite the 1996 Vancouver International Aids Conference as the meeting that broke combination therapy’s impact on Aids, the first reports were made at that year’s CROI in Washington.
A highlight of this year’s 11th CROI in San Francisco (8-11 February) was the unfolding story of drug combinations consisting solely of NRTIs - the nucleoside reverse transcriptase inhibitors like AZT, ddI and 3TC, to name just three.
In 2003, a series of triple NRTI combinations had performed badly for people taking their all-important first HIV treatment regime. There’s been a tendency to lump all these trials together when in fact there are some important distinctions.
A trial last year comparing Trizivir®, the all-in-one-tablet formulation of AZT/3TC/abacavir, with two efavirenz-containing combinations, reported poorer viral responses in people receiving Trizivir alone, with 61 per cent of people achieving viral loads under 50 on Trizivir compared with 83 per cent on efavirenz.
But markedly worse failure rates, as high as 91 per cent in one study, were noted in three trials presented at CROI where people new to HIV treatment took either 3TC/tenofovir/abacavir, or 3TC/tenofovir/ddI.
This failure rate is even worse than that seen during early pilot studies of tenofovir, when the drug was taken by itself, as monotherapy. Clearly something else is going on when these particular drugs are taken together.
We are already aware that HIV drugs can raise or lower each other’s levels in the body in ways we might not have predicted. Could this have been the cause of failure with these triple nuke combinations?
According to a study at CROI called TONUS, the answer to this question appears to be no. Volunteers new to therapy took 3TC/tenofovir/abacavir, and noted the same pattern of frequent early viral load failure as two earlier studies had reported. However, TONUS also tested participants’ drug levels to see if an interaction was causing one or more drug to fall below its active concentration. Tests revealed that drug concentrations were adequate in the majority of study volunteers.
The most likely culprit is what’s termed ‘a low genetic barrier to resistance’. What this means is that it’s easy for HIV to develop resistance to the components of these combinations, as your HIV only needs to acquire two single mutations that confer resistance to 3TC and tenofovir to render these regimes ineffective.
Tests show that these two mutations (codenamed respectively M184V and K65R) were present at the time of patients’ viral load failure at a relatively high frequency.
The pattern of resistance mutations seen in people who fail on triple-nuke combinations containing AZT is different. You don’t see so much K65R, because if you develop resistance to AZT, it makes it more difficult for HIV to develop K65R too.
This stuff is complicated, but has important implications for the selection of initial HIV treatment combinations.
The K65R mutation is seen more commonly in people who’ve taken HIV treatment than it was a few years ago, and this is probably because AZT is used less often in first-line therapy than it was before the availability of tenofovir and abacavir, which (along with ddI) can both give rise to resistance via the K65R mutation. While tenofovir, abacavir and ddI can all be taken once-daily, AZT remains a twice-daily treatment, and has lost favour in part for this reason, but if it continues to be left out of nucleoside combinations, the likelihood is that we’ll continue to see much more K65R.
In addition there’s an important message here about the risks of taking any HIV treatment combination before it’s been properly studied within a trial. If you take a combination that mixes inadequate potency with a low genetic barrier to resistance, you’ll have a recipe for treatment failure even if you’re taking three drugs.
We now have a number of once-daily HIV treatments, with the promise of more in future. At CROI, a study comparing the protease inhibitor (PI) Kaletra® (lopinavir/ritonavir) taken once-daily compared with the licensed twice-daily dosing regime found no difference between the two options after 48 weeks of treatment. This study used the newer NRTI backbone of tenofovir/FTC, also taken once-daily.
Once-daily Kaletra did cause more diarrhoea than twice-daily, and it’s also important to note that this study was performed in people who hadn’t taken other HIV treatments previously. If you’ve already taken a PI, then it’s probably safer to stick with twice-daily dosing of Kaletra for now, as it may guard better against PI resistance you may have acquired from your earlier treatment.
A separate trial looked at PIs boosted with ritonavir for people who’d already taken a PI-containing combination. There were no differences in viral load response between those taking atazanavir/ritonavir or lopinavir/ritonavir (Kaletra). However combining saquinavir and atazanavir, which in theory should boost each other, did less well.
Atazanavir, recently licensed in the UK for use in people who are treatment-experienced, is notable for having little effect on blood fats (lipids), a problem which can accompany other PIs.
Pharmaceutical companies are often criticised for being slow to look for variations in their drugs’ effects in different communities, such as in women compared to men, or in people of African rather than European ancestry.
A study looking at the relationship between genetic factors and the side effects of the non-nucleoside (NNRTI) drug efavirenz noted some important differences between black African and white participants.
Efavirenz passed through the body more slowly in many of the African volunteers. This was because a genetic variation in the enzyme that processes the drug was more common in African participants than white. The result is that efavirenz reached much higher levels in the blood, producing more frequent side effects in black African volunteers.
Perhaps the most compelling study involving nevirapine, the other licensed NNRTI, was a Thai study in which women who had received a single dose of the drug in labour, later received a nevirapine-containing HAART regime for treatment of their HIV infection.
Single-dose nevirapine, a cheap and easy-to-take option, has helped to reduce mother-to-child transmission in the developing world. But its success has been tempered by a good proportion of the women acquiring NNRTI resistance as a consequence of protecting their babies from HIV infection.
The Thai study confirmed this fear. Mothers who had received nevirapine at childbirth did less well on nevirapine-containing HAART than mothers whose preventative treatment involved AZT.
A further presentation was a large-scale comparison of whether a planned Caesarean section rather than normal vaginal delivery reduced mother-to-child infection.
A study from CROI suggests that Caesarean sections have no impact on mother-to-child transmission - and can therefore be avoided - in women who are taking antiretrovirals and have a viral load below 1,000 copies when they give birth.
But the study did find a benefit from Caesarean section in women with a viral load below 1,000 copies who were taking just a single drug, and in women with higher viral loads, regardless of the treatment they were taking.
Finally, a study from the US showed that ‘rapid tests’ for HIV can be used in pregnant women at the time of labour, allowing preventative measures to be offered to women who turn out to have HIV, even at this late stage.
Rapid tests that can give results in around 20 minutes are under-used in the UK at present. This study highlights one reason why this needs to change.
Email: annapoppa@tiscali.co.uk
