ARTEMIS: 192-week efficacy and safety of once-daily darunavir/ritonavir (DRV/r) vs lopinavir/r (LPV/r) in treatment-naïve HIV-1-infected adults

Background

ARTEMIS was a Phase III, randomised, open-label study assessing efficacy and safety of DRV/r 800/100mg qd versus LPV/r 800/200mg total daily dose (qd or bid) in treatment-naïve HIV-1-infected adults. At 96 wks, DRV/r demonstrated non-inferiority and superiority to LPV/r in virological response. Wk 192 results are reported.

Methods

Patients stratified by baseline (BL) viral load (VL [HIV-1 RNA] < or ≥100,000 copies/mL [cpm]) and CD4 cell count (< or ≥200 cells/mm³) were randomised 1:1 to DRV/r qd or LPV/r. Primary efficacy parameter: non-inferiority (≤ –12%) of DRV/r to LPV/r in virological response (VL <50 cpm, ITT-TLOVR). DRV/r superiority ( ≤ 0%) was assessed if non-inferiority was demonstrated.

Results

689 patients (30% female; mean BL VL 4.85 log₁₀ cpm; median CD4 225 cells/mm³) were randomised. Overall, significantly more DRV/r than LPV/r patients had VL <50 cpm at Wk 192, confirming DRV/r qd non-inferiority (p<0.001) and superiority (p=0.002) (Table 1). In patients with virological failure (VF; TLOVR non-VF censored) no developing primary PI mutations were identified in either arm; all VFs with paired BL/endpoint phenotypes that were susceptible at BL to amprenavir, atazanavir, indinavir, lopinavir, saquinavir or tipranavir remained susceptible after treatment.

Table 1

Conclusions

DRV/r qd demonstrated sustained efficacy with non-inferiority and superiority to LPV/r over 192 wks. Development of resistance was low in both arms. DRV/r was associated with smaller median increases in total cholesterol and triglycerides than LPV/r, and a lower incidence of grade 2–4 diarrhoea.