967 
First-line PI-containing Regimens Enhance Decreased Bone Mineral Density Greater than NNRTI-containing Regimen in HIV-1-infected Patients: A Substudy of the HIPPOCAMPE–ANRS 121 Trial
C Duvivier1,2,3, S Kolta4, L Assoumou2,3, J Ghosn1,2,3, S Rozenberg5, R Murphy1,2, C Katlama1,2,3, Dominique Costagliola*2,3, and the ANRS 121 Study Group
1Hosp Pitie-Salpetriere, Paris, France; 2Univ Pierre and Marie Curie, Paris, France; 3INSERM UMR S720, Paris, France; 4Hosp Cochin, Univ Paris-Descartes, Paris, France; and 5Hosp Pitie-Salpetriere, Paris, France
Background: No randomized study has evaluated the effect
of different antiretroviral (ARV) regimens on the evolution of bone mineral
density (BMD) at specific sites. Our objective was to evaluate the change in
bone mineral density (BMD, g/cm²) at hip and lumbar spine in patients
initiating ARV with 1 of the following regimens: NNRTI/protease inhibitor (PI)
(a NRTI-sparing regimen), PI/2NRTI, or NNRTI/2NRTI.
Methods: ARV-naïve patients enrolled in the Hippocampe-ANRS 121 study
were randomized (2:1:1) to receive either a NNRTI (efavirenz [EFV] or nevirapine
[NVP]) + a PI (lopinavir [LPV]/ritonavir [r] or indinavir [IDV]/r), or PI/2NRTI,
or NNRTI/2NRTI. In a substudy, BMD at hip and at lumbar spine were evaluated at
baseline and week 48 by dual X-ray-absorptometry, with a central reading
laboratory.
Results: We enrolled 71 patients in the substudy, 36 in the NNRTI/PI,
19 in the PI/2NRTI, and 16 in the NNRTI/2NRTI arms. Baseline characteristics
were (median [IQR]): male (77%), age 40 years (33 to 49), 69% Caucasian, 58%
smokers, body mass index 23 kg/m2 (21 to 24), time since HIV diagnosis
6 months (2 to 37), CD4 count 219 cells/mm3 (144 to 285), plasma
HIV-1 RNA 5.1 log10 copies/mL (4.4 to 5.5). In the arms with NRTI, >90%
of patients received zidovudine (ZDV)/lamivudine (3TC). At baseline, median T score
was ‑0.14 (‑0.98 to 0.47) at lumbar spine and ‑0.01 (‑0.88
to 0.83) at hip. Before starting ARV, osteopenia was present in 31% of patients
including 22% at lumbar spine and 22% at hip and 2 patients had osteoporosis.
At week 48, there was a significant reduction in BMD at both sites for
all-regimens, with a mean change in BMD between baseline and week 48 of –4.14±3.91%
at lumbar spine and ‑2.79±4.67% at hip (all p ≤0.001). The decrease
of BMD at lumbar spine was significantly worse in the NNRTI/PI arm (–4.44±3.43%)
and in the PI/2NRTI arm (–5.81±4.51%) compared to the NNRTI/2NRTI arm (–1.48±2.88%)
(NNRTI/PI vs NNRTI/2NRTI p = 0.007, PI/2NRTI vs NNRTI/2NRTI p = 0.001).
Conclusions: BMD was impaired in 31% of patients
before starting any ARV suggesting a causative role of HIV. After 1 year, the
decrease in lumbar spine BMD was more pronounced in patients receiving either
PI-containing regimens compared to NNRTI/2NRTI. BMD at specific sites should be
monitored during lifelong ARV therapy.
|
