South African resistance survey confirms that clade C is more likely to develop multi-drug resistance mutation

A survey of resistance among recipients of HIV therapy in South Africa has found that HIV clade C – the predominant subtype in southern Africa – is much more likely to develop the dangerous K65R resistance mutation, which confers resistance to most of the nucleoside (NRTI) drugs except AZT. Furthermore, whereas K65R is normally seen in patients failing tenofovir, abacavir or ddI, among clade C patients it was common in patients failing therapies including d4T (stavudine).

The South Africa study confirms an association between clade C virus and K65R originally suspected from a study in Botswana – see this report.

Meanwhile a UK study of resistance among patients with subtype C has found that although they had less resistance in general than patients with subtype B (the predominant one in gay men) they had higher levels of resistance to the non-nucleoside (NNRTI) drugs, efavirenz and nevirapine. This is probably due to suboptimal regimens and the use of nevirapine for prevention of mother-to-baby transmission in Africa.

Resistance in South Africa

Catherine Orrell of the Desmond Tutu HIV Centre at the University of Cape Town presented resistance test results from a cohort of patients in the pioneering HIV treatment programme at Gugulethu township. During the period studied, 2778 patients were on treatment in the cohort and 137 (about 5%) failed first-line drug therapy. While 20% of patients had one viral load over 1000 copies/ml while on therapy, an intensive adherence counselling programme enabled three-quarters of these patients to re-suppress HIV.

Two hundred and thirty patients had resistance tests, 120 drug-naive patients as part of a resistance research programme, and 110 because of virological failure. All but five had subtype C; these five, who had subtype B, were all men who had sex with men. The main first-line treatment at Gugulethu is d4t/3TC/efavirenz (67% of patients), with 14.5% receiving d4t/3TC/nevirapine and smaller numbers receiving AZT and tenofovir (the latter drug has just become available, restricted to patients with hepatitis B or those who develop lactic acidosis due to d4T).

Resistance in the drug-naive patients was rare; four had NRTI mutations, one K65R and three Y118I, an AZT-associated resistance mutation that normally only reduces response to the drug when other mutations develop. Two had the common resistance mutations to the NNRTI drugs Y181C and G190A. No-one had clinically significant protease inhibitor resistance.

Resistance among patients failing therapy was inevitably much more common. Ninety-six patients (87% of all failing therapy) had mutations to the NNRTI drugs while 86 patients (78% of those failing) had the M184V 3TC resistance mutation, which arises very easily but appears either to be rarely transmitted or quickly suppressed in people who acquire it, as it impairs viral fitness.

Patients who had both NNRTI and M184V mutations were also likely to have the thymidine analogue mutations (TAMs) that confer resistance to AZT and d4T while those who had NNRTI or M184V mutations alone were unlikely to develop TAMs. However while the prevalence of other mutations did not increase over time, the prevalence of TAMs did, being seen in one in five of failing patients at the start of the study and in one in three by the end.

Ten patients had the K65R mutation. This was described as ‘unexpected and worrisome’ by Orrell, because no patient with K65R had received tenofovir, abacavir or ddI. Although in other HIV subtypes K65R can cause low-level resistance to d4T, it is rarely caused by d4T treatment. The patients who developed the K65R mutation were also likely to have TAMs too, effectively ruling out the use of the whole NRTI class in future treatment. Such patients would benefit solely from the protease inhibitor in any second-line regimen. Orrell concluded by saying that viral load monitoring, done every six months at Gugulethu, was important and that “we could even do low-cost genotyping”.

Professor Mark Wainberg, the Canadian HIV expert who originally spotted the higher frequency of K65R in clade C patients, commented from the audience that Orrell’s study showed that “we should just stop using d4T”, but Orrell commented that d4T is “the cheapest drug we have and it’s not going to go away”.

UK non-B resistance

Daniella Chilton of the UK Medical Research Council presented data on primary (transmitted) drug resistance in different subtypes in the UK, using resistance tests done on drug-naive patients from the UK’s Drug Resistance Database. She compared resistance in patients with subtype B, in subtype C and in patients with neither B nor C (these other subtypes were lumped together, as they are much less common in the UK). Non-B patients were by no means solely African or black; 54% were of black ethnicity, 15% white and 31% other; 43% were African and 44% female.

In general resistance in people with non-B subtypes was lower than in subtype B, as is to be expected because patients who transmit subtype B virus are likely to have had longer suboptimal treatment histories. The prevalence of any resistance in naive patients declined from 10% in 2001 to 3.5% in 2006 in non-B patients. In patients with subtype B it peaked at 14% in 2002 then declined to 10% in 2006.

However among people with confirmed primary resistance, resistance to more than one drug class was more common in patients with subtype C and more common still in non-B non-C patients. Fifteen per cent of clade B patients with confirmed resistance had resistance to more than one class, 20% of clade C patients, and 35% of those with the other subtypes, though only 5% of patients with any subtype had resistance to three or more classes.

This was driven by much higher rates of NNRTI resistance in non-B subtypes. Among those with confirmed resistance, 38% of subtype B patients had NNRTI resistance, 72% of subtype C, and 62% of non-B non-C. In contrast somewhat more subtype B patients had nucleoside (NRTI) resistance.

“This is evidence for NNRTI resistance in developing countries and the legacy of NRTI monotherapy in developed ones,” commented Chilton.