Five Years With Dolutegravir Plus Lamivudine as a Switch Strategy

Much More Than a Positive Finding

Arturo Ciccullo, MD; Vanni Borghi, MD; Andrea Giacomelli, MD; Maria Vittoria Cossu, MD; Gaetana Sterrantino, MD; Alessandra Latini, MD; Andrea Giacometti, MD; Andrea De Vito, MD; William Gennari, MD; Giordano Madeddu, MD; Amedeo Capetti, MD; Gabriella d'Ettorre, MD; Cristina Mussini, MD; Stefano Rusconi, MD; Simona Di Giambenedetto, MD; Gianmaria Baldin, MD


J Acquir Immune Defic Syndr. 2021;88(3):234-237. 

In This Article


We analyzed 785 patients: 554 were men (70.6%), with a median age of 52 years [interquartile range (IQR) 45–58 years], a median time from HIV diagnosis of 14.8 years (IQR 8.0–22.0 years), and a median time of ARV exposure of 11.5 years (IQR 5.8–18.6 years). One hundred thirty-two patients (16.8%) had at least 1 previous AIDS-defining event, and 175 (22.3%) experienced at least 1 VF, whereas 282 (35.9%) were previously prescribed a 2DR. Thirty-three (4.2%) had a previously detected M184V resistance mutation. Complete patients' characteristics are shown in Table 1.

During 1992.6 Patient-Years of Follow-Up (PYFU), we observed 18 VF, with a rate of 0.9 VF per 100 PYFU. The median time to VF was 25.8 months (IQR 14.1–47.4 months); 10 VF (55.5%) occurred in the first 48 weeks of follow-up. Nine of the18 patients experiencing a VF discontinued study regimen, wherease the others maintained DTG + 3TC: all the patients experiencing VF reachieved virologic control subsequently. Moreover, none of the patients experiencing VF developed resistance mutation to either nucleoside reverse transcriptase inhibitors or INIs after failure, including patients with a previously detected M184V mutation. The estimated probability of maintaining virological suppression was 98.5% (SD ± 0.5) at 48 weeks, 97.7% (SD ± 0.6) at 96 weeks, 96.9% (SD ± 0.8) at 144 weeks, and 96.4% (SD ± 0.9) at 240 weeks. In a multivariate regression analysis, a non-B HIV subtype (vs subtype B, aHR 31.5, 95% CI: 2.0 to 488.9, P = 0.014) and a previous VF (aHR 24.0, 95% CI: 1.2 to 475.7, P = 0.037) resulted predictors of VF, after adjusting for nadir CD4+ cell count and peak HIV-RNA.

In appropriate survival analyses, we observed a higher probability of VF in PLWHIV with a peak HIV-RNA >500.000 copies/mL compared with others (log-rank P = 0.002), in PLWHIV with a previous VF compared with those who had never experienced a VF (log-rank P = 0.031), and in PLWHIV with a HIV subtype non-B compared with those with a subtype B (log-rank P = 0.014) Figure 1). We did not observe differences in probability of VF in PLWHIV with a M184V resistance mutation (log-rank P = 0.689); however, in a dedicated analysis, in patients with time of virological suppression < 88 months, the M184V mutation was a predictor of VF (aHR 11.62, 95% CI: 1.15 to 117.57, P = 0.038).

Figure 1.

Kaplan–Meier survival analyses. A, Stratified for Zenith HIV-RNA; (B) stratified for previous virological failures; (C) stratified for the presence of the M184V mutation; and (D) stratified for HIV subtype.

During 2014.7 PYFU, we censored 150 TD, with a rate of 7.5 TD per 100 PYFU. The median time to TD was 28.6 months (IQR 14.3–47.6 months). The estimated probability of maintaining study regimen was 89.0% (SD ± 1.1) at 48 weeks, 82.9% (SD ± 1.4) at 96 weeks, 79.7% (SD ± 1.6) at 144 weeks, and 74.3% (SD ± 2.2) at 240 weeks. In our cohort, the main reasons for TD were toxicity in 54 (6.9% of total population) cases (21 for neuropsychiatric events, 11 for GI toxicity, 9 for renal toxicity, and 13 for other/unspecified toxicity), simplification to STR in 13 (1.7%) cases, weight gain in 8 cases (1.0%), death (unrelated to HIV/AIDS) in 6 cases (0.8%), VF in 9 cases (1.1%), and other/unknown in 60 cases (7.6%).

In a specific subanalysis, probability of TD after neuropsychiatric events was 9.6% at week 48 and 13.3% at weeks 96 and 144; HCV coinfection resulted the sole predictor (aHR 3.90, 95% CI: 1.05 to 14.52, P = 0.043). Among the 21 discontinuations caused by neuropsychological events, 9 were due to insomnia, 6 to headache, 5 to mood disorders, and 1 was due to the sudden onset of nightmares.

As far as immunological parameters, we observed a significant increase in CD4+ cell count at 48 weeks (median +27 cell/mm3, P < 0.001), 96 weeks (median +30 cell/mm3, P = 0.001) and 144 weeks (median +11 cell/mm3, P = 0.038). A median increase of +49 cell/mm3 was observed after 240 weeks, although it resulted nonsignificant (P = 0.073).