Resistance in archived virus dissimilar to circulating virus in patients on failing treatment

A small Italian study has found that, in people on failing antiretroviral regimens, viral mutations associated with drug resistance can vary significantly between circulating virus in the blood plasma and “archived” provirus in peripheral blood mononuclear cells (PBMCs). The study was published in the April 15^th edition of the Journal of Acquired Immune Deficiency Syndromes.

HIV variants present in the plasma do not necessarily reflect the entire variety of HIV subtypes present in any individual. Latently HIV-infected peripheral blood mononuclear cells (PBMCs) provide a reservoir in which drug-resistant virus may persist for long periods of time. Depending on treatment history, this archived virus has been found to contain different variants than circulating virus.

In this study, researchers analysed samples taken during routine genotypic resistance testing at the University “La Sapienza”, Rome, Italy after the year 2000. Samples were collected from 95 treatment-experienced people (mean treatment duration, seven years) whose treatment was virologically failing. Median CD4 cell counts were 287 cells/mm³and the median viral load was 4.4 log₁₀ copies/ml. Six patients had been exposed to nucleosides (NRTIs) only, nineteen had received NRTIs and non-nucleosides (NNRTIs), and thirty-nine had been exposed to all three major drug classes – NRTIs, NNRTIs and protease inhibitors (PIs). Genotypic profiles of viral RNA from blood plasma and viral DNA from PBMCs were compared using standard resistance assays (TruGene HIV-1 Genotyping and OpenGene automated DNA sequencing, both from Bayer).

Of the 95 patients, 85 (89%) showed resistance mutations in both plasma RNA and archived DNA. (Two had mutations in RNA only, and eight did not show any resistance mutations.) Of the 85 in whom the circulating and archived mutations could be compared, 62 (73%) were dissimilar: either the number or type of observed mutations, or both, were not the same in circulating RNA and archived DNA. In 51% of these 62 discordant samples, a greater number of mutations was found in plasma RNA; in 37%, more were found in PBMCs.

NRTI and PI mutations were very common – more than 47% of the samples had at least five NRTI and at least five PI mutations. The most frequently detected mutations were the 3TC-associated M184V (76%), PI mutations L63P (76%) and L10F (61%), and thymidine analogue mutations T215Y/F/S (70%), M41L (66%), D67N (51%), and L210W (47%). Most NRTI mutations and primary PI mutations were found in both compartments (circulating and archived). NNRTI mutations were less common overall; the majority of samples had two, one or no NNRTI mutations.

Five patients underwent treatment interruptions for at least three months, and samples were available from before and after the interruptions. For people with archived resistance mutations in PBMC DNA, the number of resistance mutations in plasma RNA did not decrease during or after treatment interruption. However, when the dominant archived virus was wild-type, mutations in plasma RNA decreased during treatment interruption as wild-type virus replaced the HIV population in the blood plasma.

The researchers note that their study was based on a routine assay that does not detect minor viral variants. However, they state that “in patients receiving extensive antiretroviral treatment and with therapeutic failure … the PBMC compartment does not necessarily reflect the plasma compartment”, possibly implying that “PBMCs may constitute a reservoir for drug-resistant variants”. They also note that mutations conferring varying degrees of fitness might persist for different lengths of time: “[in] this regard, it is worth noting that the [AZT-associated] K70R mutation was the only mutation found more frequently in PBMCs than in plasma.”

Previous studies have found considerable agreement between viral populations in the two compartments. The authors note that the numbers studied here were small, and that they defined viral discrepancy to include quite small differences (one mutation). Overall, they believe their study “reinforces the data obtained by other authors and strongly indicates that in patients who have received extensive treatment, PBMC DNA represents a reliable guide to drug resistance.”