Using a more sensitive technical approach, Ruff et al. reported

in nine children that archival wild-type HIV-1 persisted in a replication-competent form in resting CD4 T cells despite up to 10 years of continuous antiretroviral exposure [4]. Interestingly, among the 86 virologically controlled patients enrolled in the ANRS 106-WINDOW trial, 31% had cellular CT99021 nmr HIV DNA mutations, which were associated with longer of treatment duration exposition [14]. Firstly, the lower resistance found in DNA could be a consequence of the fact that sequence amplifications in DNA failed for 24% and 8% of RT and PR sequences, respectively. Secondly, it is important to underline the point that our analysis is based on routine assays, such as population-based sequencing, which does not detect minor variants, and that total DNA was extracted from whole blood. Cloning or single

genome sequencing would probably have detected more archived mutations [5, 15, 16], as would sequencing of DNA from selected latent T cells, the main cellular HIV reservoir [4]. However, these methods are too costly and complex for routine clinical use. It should also be noted that we analysed the cumulative number 3 MA of RNA mutations. As recently reported, a single plasma genotypic test tends to underestimate the level of resistance in heavily pretreated patients with a history of treatment failures [17]. Undetected drug-resistant minority variants are known to persist when antiretroviral therapy is discontinued or changed, and the rapid reappearance of these mutations on treatment resumption has been extensively documented [18-21]. This difference in resistance mutations between the plasma and cellular compartments suggests that plasma viruses at the time of treatment failure are enriched for resistant viruses, leading to a better Baricitinib capture of the resistant genotypes than in the mononuclear cells which may have a large library of archival variants, the bulk of which may be lacking resistance mutations. Verhofstede et al. suggested that the probability of finding a resistant variant within the cellular reservoir depended, at least in part, on the period

during which this variant was able to replicate [5]. Thus, delays in changing a failing therapeutic regimen may favour mutant archiving. Interestingly, a longitudinal analysis has shown that resistance mutations emerge in plasma HIV-1 more than 1 year before they are found in peripheral blood mononuclear cells (PBMCs) [2, 22]. In contrast, because genetic changes in cellular proviruses occur more slowly than in plasma viruses, which are more sensitive to selective pressure, mutations persisted longer in proviral DNA [6, 11, 14]. Another hypothesis is that, in patients on effective HAART, cells infected by archived resistant provirus could be diluted by more recent uninfected cells and therefore be less readily detectable.