Background.There is a need to prevent or minimize bone loss associated with antiretroviral treatment (ART) initiation. We compared maraviroc (MVC)- to tenofovir disoproxil fumarate (TDF)-containing ART. Methods.This was a double-blind, placebo-controlled trial. ART-naive subjects with human immunodeficiency virus type 1 RNA load (viral load [VL]) >1000 copies/mL and R5 tropism were randomized to MVC 150 mg or TDF 300 mg once daily (1:1), stratified by VL <100 000 or ≥100 000 copies/mL and age <30 or ≥30 years. All subjects received darunavir 800 mg, ritonavir 100 mg, and emtricitabine 200 mg daily. Dual-energy X-ray absorptiometry scanning was done at baseline and week 48. The primary endpoint was percentage change in total hip bone mineral density (BMD) from baseline to week 48 in the as-treated population. Results.We enrolled 262 subjects. A total of 259 subjects (130 MVC, 129 TDF) contributed to the analyses (91% male; median age, 33 years; 45% white, 30% black, 22% Hispanic). Baseline median VL was 4.5 log10 copies/mL and CD4 count was 390 cells/μL. The decline in hip BMD (n = 115 for MVC, n = 109 for TDF) at week 48 was less with MVC (median [Q1, Q3] of -1.51% [-2.93%, -0.11%] vs -2.40% [-4.30%, -1.32%] for TDF (P <. 001). Lumbar spine BMD decline was also less with MVC (median -0.88% vs -2.35%; P <. 001). Similar proportions of subjects in both arms achieved VL ≤50 copies/mL in as-treated and ITT analyses. Conclusions.MVC was associated with less bone loss at the hip and lumbar spine compared with TDF. MVC may be an option to attenuate ART-associated bone loss. Clinical Trials Registration.NCT01400412.