Background:  Facial lipoatrophy is disfiguring, can stigmatize, and can affect self-esteem, quality of life, and adherence to ART. Small studies have found poly-L-lactic acid injections to have acceptable safety and efficacy over 48 weeks, but no study was randomized and also included objective endpoints.

Methods:  HIV⁺ adults with moderate/severe facial lipoatrophy were randomised to 4 open-label poly-L-lactic acid treatments (5 mL per cheek) every 2 weeks from week 0 (IMM; n = 51), or 4 poly-L-lactic acid treatments after week 24 (DEF; n = 50). The primary endpoint was mean change from baseline in facial soft tissue volume (upper limit at mid-orbit; lower limit at angle of mandible) by spiral computed tomography (CT). The study had 80% power to detect a 25% difference in change in facial soft tissue volume between groups at week 24 by intention-to-treat analysis.

Results:  Mean changes in facial soft tissue volume at week 24 were 0 cm³ in the IMM group and –10 cm³ in the DEF group (between-group difference, 10 cm³ [95%CI –7, +28]; p = 0.24; +3%). Relative to the DEF group, the IMM group had a significantly greater between-group change in subcutaneous soft-tissue depth just inferior to the malar bone (2.2 mm; 95%CI 1.6, 2.9; p <0.0001; +19%) and just superior to the nasolabial folds (1.0 mm; 95%CI, 0.3, 1.6; p = 0.003; +4%), but not in untreated areas (mid-orbit, p = 0.9; angle of mandible; p = 0.21). Subjective facial lipoatrophy severity assessed by patient and physician improved significantly in IMM participants (p <0.0001). There was no change in peripheral fat mass (p = 0.63), total fat mass (p = 0.29), or lean body mass (p = 0.28) by DEXA. Poly-L-lactic acid treatment adherence was 99%. Poly-L-lactic acid treatment did not significantly impact HIV viral load (p = 0.57), CD4⁺ count (p = 0.54), or ART adherence (p = 0.70). The most common procedure/product adverse events were pain or discomfort (any grade, 79%), edema (66%; ≥25 mm diameter, 29%) and erythema (56%; ≥25 mm diameter, 22%). Median duration of treatment-related adverse events was 2 (IQR 1 to 3) days. No poly-L-lactic acid–related adverse event was dose-limiting or delayed treatment. Of 4 severe adverse events reported in 3 subjects, none was attributed to poly-L-lactic acid.

Conclusions:  Although 4 facial injections of poly-L-lactic acid over 6 weeks in HIV-infected adults with facial lipoatrophy did not increase facial soft tissue volume overall, but prevented deterioration in injected areas. Poly-L-lactic acid injections were administered safely and were well tolerated. Objectively assessed relative increases in cheek subcutaneous thickness were less than reported in other studies.