Dual class drug resistance present in two-thirds of South African HIV cohort on failing ART

Almost two-thirds of individuals failing first-line antiretroviral therapy (ART) in a KwaZulu-Natal cohort had resistance to drugs from two classes, and one-third had at least one mutation that could reduce response to the entire nucleoside analogue class, researchers from the South African Resistance Cohort Study Team report in the May 15^th edition of Clinical Infectious Diseases.

The commonest mutations were associated with non-nucleoside reverse transcriptase inhibitor (NNRTI) and nucleoside reverse transcriptase inhibitor (NRTI) classes of HAART drugs, with resistance to 3TC most frequent within this class.

Previous results from this study were presented at the Fourteenth Conference on Retroviruses and Opportunistic Infections in 2007.

The study findings highlight the controversy over how to monitor for the failure of antiretroviral treatment and when to switch patients experiencing treatment failure when viral load is not readily available and second-line treatment can cost five to ten times as much as first-line treatment.

Current World Health Organization guidelines for antiretroviral treatment in resource-limited settings suggest that if viral load testing is not available - as is the case in most of sub-Saharan Africa - treatment should be switched to a second-line regimen if an individual develops new symptoms of serious immune suppression (WHO stage 3 or 4 symptoms), or if the CD4 cell count declines by 50% from its peak or 33% within the past six months (after a confirmatory test).

Indeed, a recently published computer modelling study suggests that although viral load testing may prevent the development of resistance, switching treatment in response to viral load test results rather than CD4 cell counts or new symptoms would not result in significantly improved survival, due to the assumption in the model that drug resistance would not substantially undermine the effectiveness of second-line treatment.

But some, particularly in the United States, are critical of the WHO approach.

In an accompanying editorial also published in the May 15^th edition of Clinical Infectious Diseases, Davey Smith and Robert Schooley of the University of California San Diego comment: "Failing to use laboratory tools that monitor treatment sucess is like running with scissors; it is all quick and easy until someone falls down...If a choice must be made between monitoring viral load or CD4 cell count during HAART, we believe it would be more useful to monitor viral load than CD4 count."

However, patients in this study were identified as having failing treatment through viral load testing, suggesting that rates of thymidine analogue resistance could have been much higher if viral load monitoring had not been available.

After the failure of first-line treatment in resource-limited settings international and national guidelines recommend switching to a protease inhibitor-based regimen in order to have the benefit of a new class of drug.

Thymidine analogue resistance is critical to the chances of successful second-line therapy in resource-limited settings due to the limited range of affordable nucleoside analogues available for combination with a protease inhibitor. However, knowledge of patterns of resistance to first-line drugs in resource-limited settings is still limited.

South African Resistance Cohort Study

In KwaZulu-Natal, HAART use increased by more than six-fold during the period 2002-2006. A team of US and South African investigators addressed this issue by assessing the prevalence of resistance mutations in HIV-1-infected patients following failure of their first HAART regimen.

The study sites were two referral centres for HAART in KwaZulu Natal. Study participants were HIV-1-infected patients aged 18 years or older who experienced virologic treatment failure after 24 weeks of their first HAART.

HAART consisted of either first- or second-line regimens or some combinations of drugs.

The first-line regimen was stavudine plus lamivudine and either efavirenz or nevirapine; the second-line regimens included zidovudine plus didanosine and lopinavir-ritonavir fixed-dose combination. Second-line regimens were given to patients who had previously received a sub-optimal non-HAART regimen or were intolerant to first-line regimens.

Virologic failure was defined as an HIV-1 RNA level of more than 1,000 copies/ml. Genotypic resistance testing was performed on plasma virus samples from all patients who experienced virologic failure. Drug resistance mutations and deletions in the reverse transcriptase and protease genes were identified by DNA sequencing.

A number of other laboratory parameters were measured at enrolment. These included CD4 counts, complete blood counts, HIV-1 RNA levels, haemoglobin, and liver and kidney function tests. Clinical and demographic data were also obtained.

In all, a total of 124 antiretroviral-treated adults who experienced virologic failure were enrolled between January 2005 and August 2006. The predominant viral subtype was HIV-1C. Virus samples from 83.5% of participants carried one or more significant drug resistance mutations. Dual-class drug-resistant virus was present in 64.3 % of participants while triple-class drug resistant virus was present in 2.6 %.

The commonest mutation was M184V/I which was observed in 64.3% of patients; K103N was present in 51.3%, and V106M was present in 19.1% of the patients. In 39.1% of patients the M184V/I and K103N mutations were both present. Thymidine analogue and protease resistance mutations were found in 32.2 % and 4.4 % of patients, respectively.

Thymidine analogue mutation patterns

Resistance researchers classify thymidine analogue mutations according to the distinctive pattern of mutations that emerges, with different drugs and viral sub-types tending to favour different pathways. The TAM-1 pattern, which tends to confer a degree of resistance to all NRTIs once it emerges, was present in 7% of patients. The TAM-2 pattern, which confers resistance only to d4T and AZT, was present in 19% of patients. Mutations associated with both patterns were present in 6.1% of patients.

Thirteen per cent of patients had at least three thymidine analogue mutations, indicating the potential for reduced effectiveness of abacavir and tenofovir in second-line treatment.

The authors describe the prevalence of TAMs as quite low.

Factors associated with drug resistance

Regression analysis was carried out to identify factors that are associated with drug resistance mutations. None of the clinical and laboratory variables were significantly associated with drug-resistant infections in a multivariate statistical model.

The observed predominance of NNRTI drug resistance mutations and the paucity of protease inhibitor drug resistance mutations reflected the viral subtype and the class of drugs used in first-line regimens.