Ultra-sensitive test reveals highly variable patterns of NNRTI resistance decay

There is considerable variability in the rate of decay of resistance mutations after stopping non-nucleoside reverse transcriptase inhibitor (NNRTI)-based treatment, according to a study published in AIDS. Using a highly sensitive test to detect resistant viruses, the study’s investigators showed that the levels of NNRTI-resistant HIV can decay rapidly or persist for many years.

However, none of the patients in the study showed a complete clearance of NNRTI-resistant virus after failure of an NNRTI-based treatment regimen. This confirms that patients who fail NNRTI-based treatment or take single-dose nevirapine (Viramune) to prevent mother-to-child tranmission of HIV may be at risk of treatment failure if they start another NNRTI-based treatment combination in the future.

Resistance to the NNRTIs efavirenz (Sustiva) and nevirapine can occur after the development of a single mutation in the virus’s gene for reverse transcriptase, notably the K103N mutation. Development of this mutation leads to resistance to both drugs in the class.

Standard resistance assays can only detect resistance mutations that make up around 25% or more of the population of HIV in a sample of blood. However, it is possible that resistant viruses making up less than 25% of the HIV population can be selected in the presence of an NNRTI. This can result in the resistant variants becoming increasingly prevalent until HIV levels rebound.

Investigators in the United States wished to examine the development and persistence of the K103N mutation in HIV-positive patients. To detect mutant variants that were low in prevalence, they developed and validated an assay called the ‘real-time reverse transcriptase polymerase chain reaction’, which can detect the mutation down to a prevalence of 0.1%.

They used the assay to follow the evolution of K103N in 18 patients with HIV. Twelve of the patients had never taken antiretroviral therapy, while six had taken an NNRTI but experienced treatment failure.

Nine of the treatment-naïve patients had no evidence of K103N. However, three patients had increases in the prevalence of the virus above the background level of 0.029%, but the investigators found that these increases were transient after testing samples taken from the patients at different times.

“Most of the samples had mutant frequencies statistically indistinguishable from assay background, implying a steady-state level of K103N less than 0.03%,” the investigators write.

In the six patients who had taken NNRTIs, the investigators did not see a persistent pattern of K103N emergence and decay. Five of the patients’ HIV was predominantly K103N at the time of NNRTI failure, while the sixth had K103N at a prevalence of around 5%.

However, following discontinuation of NNRTI therapy, the rate of decay of K013N varied widely. One patient had almost 100% K103N virus for almost six years, another showed a rapid decay from 100 to 3% over six months, and the remaining patients had intermediate rates of decay of K103N.

Standard genotyping failed to detect the presence of resistant virus in some patients who had K103N detected using the sensitive test.

“Our results indicate that acquired drug resistance can … persist for more than five years after cessation of therapy and may not be detected by standard genotyping,” conclude the investigators. “The clinical significance of the persistent variants is not defined, but preliminary evidence suggests that minor drug-resistant variants that are missed by standard genotyping can lead to failure of subsequent treatment regimens.”

This assay is unlikely to be introduced into HIV treatment due to its cost. However, it will allow research scientists to investigate the relationship between mutations present at low levels and the risks of HIV treatment failure.

“The method described here will help to define the clinical significance of low-frequency drug-resistant HIV-1 variants in treatment-naïve and treatment-experienced patients,” the investigators conclude. “Additional studies of the frequency and variation of pre-existing or persistent drug-resistant variants using this method should provide further insights into mechanisms of evolution of drug resistance.”

The investigators also developed their assay further, to allow them to distinguish two different forms of the K103N mutation, termed AAC and AAT. These two codes refer to different changes in the structure of the virus’s genetic material, but both produce the same alteration in the structure of the reverse transcriptase protein.

Although the AAC mutation is more common than AAT, two of the six patients had relatively high levels of AAT during their treatment. In one patient, AAC and AAT were of approximately equal prevalence during nevirapine therapy, but AAC became predominant after a change from nevirapine to efavirenz.

The second patient showed a switch from AAC to AAT, which occurred at the same time as the development of the M184V mutation, which results in resistance to 3TC (lamivudine, Epivir). After stopping 3TC treatment, AAC became the more predominant form of the K103N mutation again, leading the investigators to suggest that the M184V mutation and the AAT version of the K103N mutation may be linked.

The researchers explain that the prevalence of the AAC mutation in most patients may be due to a higher mutation rate for AAC, the structure of the virus’s genetic material or a preference for the AAC mutation by the cell’s protein-producing machinery. However, the two forms of the mutation are expected to have the same impact on the effectiveness of antiretroviral treatment regimens.