Evaluation of HIV-1 DNA resistance evolution in highly treatment-experienced and multi-resistant individuals under suppressive antiretroviral therapy: a longitudinal study from the PRESTIGIO Registry

Abstract

Background: This study aimed to clarify whether resistance detected in HIV-1 DNA might evolve in virologically suppressed highly treatment-experienced (HTE) individuals with multidrug resistance (MDR).

Methods: Twenty-three virologically suppressed HTE MDR individuals from the PRESTIGIO Registry with two longitudinal samples available under virological suppression at two different time points (T0-T1) were analysed. HIV-1 DNA levels were quantified using droplet digital PCR, and resistance was assessed through next-generation sequencing (NGS) set at 5%. Mutational load was also evaluated.

Results: At T0, individuals had been virologically suppressed for a median time of 3 years (IQR 3-5) under a salvage regimen, mostly containing dolutegravir (95.7%) and/or darunavir (69.6%). The median HIV-1 DNA level was 2588 copies/106 CD4+ cells at T0 and remained stable at T1 (2322 copies/106 CD4+ cells; P = 0.831). Individuals with at least ≥3-class resistance in HIV-1 DNA were 20 (87.0%) at T0 and 18 (78.2%) at T1 (P = 0.607). In those receiving NNRTI-sparing treatment (52.2%), the number of NNRTI major resistance mutation (MRM) significantly decreased over time (T0, 2 [1-3]; T1, 0 [0-1]; P = 0.027). No significant temporal differences in the number of PI, NRTI and integrase strand transfer inhibitor (INSTI) MRM were found. Specific MRM, such as M184V, decreased over time, particularly in individuals who were receiving a 3TC-/FTC-sparing salvage regimen or with a T0 mutational load of <1000 copies/106 CD4+ cells.

Conclusions: Over a year, HIV-1 DNA MRM generally remained unchanged in suppressed HTE MDR people with HIV (PWH) except for a significant decline in M184V and a reduction of NNRTI resistance in the absence of NNRTI pressure.

Figure 1.

Overview of MRM prevalence detected through HIV-1 DNA NGS from longitudinal PBMC samples of HTE MDR PWH. Bar plots report prevalence of MRMs according to HIVdb ver 9.8 as detected through NGS (set at 5% cut-off) at T0 (white bars) and T1 (black bars) longitudinal samples. (a–d) MRM to PI, NRTI, NNRTI and INSTI, respectively. Only statistically significant differences (P value < 0.05, according to McNemar test for matched pairs) are indicated by an asterisk.

Figure 2.

Burden of M184V and M41L detected through HIV-1 DNA NGS from longitudinal PBMC samples of HTE MDR PWH. (a) Spaghetti plot of M184V mutational load levels from T0 to T1. Black lines represent individuals under 3TC or FTC pressure; grey lines represent individuals not currently receiving 3TC or FTC. The table below indicates median (IQR) levels according to treatment. (b) Boxplot representing T0 mutational load levels according to clearance or maintenance of M184V at T1. (c) Spaghetti plot of M41L mutational load levels from T0 to T1. Black lines represent individuals under NRTI pressure; grey lines represent individuals not currently receiving NRTI. The table below indicates median (IQR) levels according to treatment. (d) Boxplot representing T0 mutational load levels according to clearance or maintenance of M41L at T1. P value according to Wilcoxon test for matched pairs.