Background:  Options for patients (patients) with peripheral lipoatrophy are limited. The MITOX and TARHEEL studies demonstrated that switching from a thymidine analog partially reversed lipoatrophy. ACTG 5125s showed that switching to a nucleoside reverse transcriptase inhibitor (NRTI) sparing regimen increases peripheral fat. We evaluated both interventional approaches and no treatment change.

Methods:  Patients at 15 ACTG sites receiving thymidine analogs stavudine (d4T) or zidovudine (ZDV) containing regimens with HIV RNA ≤ 500 copies/mL and clinical evidence of peripheral lipoatrophy were prospectively randomized to:  switch thymidine analog to abacavir (ABC); discontinue all ART and switch to lopinavir/ritonavir (LPV/r) plus nevirapine (NVP), a regimen without NRTI; or delay switching for 24 weeks. Centrally analyzed single-slice CT of mid-thigh and abdominal fat, metabolic and virologic/ immunologic parameters were measured at baseline and week 24 as the primary endpoint. At week 24, the delayed switch group began their intervention. All patients were followed 48 weeks post-intervention.   

Results:  Of the 101 patients enrolled (85% men, 69% white), 77 switched immediately and 24 delayed.  Median age was 46 years, CD4 = 611 cells/mm³, viral load < 200 copies/mL = 96%; 76% were on d4T and 24% were on ZDV.  Baseline median (IQR) subcutaneous thigh fat was 18.9 (8.3 to 29.2) cm², subcutaneous abdominal adipose tissue (SAT) 74.2 (44.6 to 122.5) cm², visceral adipose tissue (VAT) 116.3 (69.8 to 176.8) cm², and VAT:TAT (total adipose tissue) ratio 0.58 (0.45 to 0.71). The table shows median percentage changes at 24 weeks. In sum, at 24 weeks subcutaneous thigh fat increased in the LPV/r+NVP group. There were significant SAT increases and VAT:TAT decreases for both interventions and a decrease VAT for ABC. Between group comparisons were significant for SAT and VAT:TAT. There was a significant increase in CD4 for LPV/r+NVP.

Change within arm p = 0.06*, p < 0.01**, p = 0.04***, p = 0.03^#;  Difference between arms p = 0.008^†, p < 0.001^††   

Conclusions:  In patients with lipoatrophy, switching d4T or ZDV to a nonthymidine analog or changing to a NRTI-sparing regimen is associated with significant improvements in SAT, VAT, and VAT:TAT while maintaining virologic control and improving CD4 with NRTI-sparing. Further follow-up is needed to identify long-term effects.

Keywords: lipoatrophy; nucleoside analogs; metabolic complications