Secondary Logo

Journal Logo

Brief Report: Clinical Science

Changes in Bone Mineral Density After 96 Weeks of Treatment With Atazanavir/Ritonavir or Lopinavir/Ritonavir Plus Tenofovir DF/Emtricitabine in Treatment-Naive Patients With HIV-1 Infection

The CASTLE Body Composition Substudy

Moyle, Graeme J. MD*; Hardy, Hélène PharmD; Farajallah, Awny MD; McGrath, St John MD; Kaplita, Stephen MS; Ward, Douglas MD§

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: January 1, 2015 - Volume 68 - Issue 1 - p 40-45
doi: 10.1097/QAI.0000000000000383



An increased prevalence of osteoporosis, osteopenia, and fracture has been reported in both men and women infected with HIV. Bone disease is likely to become an increasingly important comorbidity in the aging HIV population.1–9 A meta-analysis of cross-sectional studies using dual-energy x-ray absorptiometry (DXA) to measure bone mineral density (BMD) demonstrated pooled odds ratios of 6.4 for reduced BMD and 3.7 for osteoporosis in HIV-infected vs non–HIV-infected patients.2 A large population-based study in a US health-care system showed that overall fracture prevalence was 2.87% in HIV-infected patients compared with 1.77% in non–HIV-infected patients.10 Evidence that antiretroviral therapy (ART) is implicated directly or indirectly in the development of bone disease includes BMD reduction after initiation of therapy.5,11 Traditional risk factors for osteoporosis, and HIV infection, play a role in bone demineralization.12,13

ART-associated BMD decline may arise from a number of potential mechanisms. Nucleoside analogue reverse transcriptase inhibitors-induced mitochondrial toxicity is one such mechanism, whereas urinary phosphate wasting and renal osteodystrophy may reduce BMD in tenofovir disoproxil fumarate (DF)–treated patients.14,15 Protease inhibitors have been associated with bone mineral reduction via dysregulation of osteoblasts and osteoclasts; however, the results of studies are conflicting.2,5,12,14,16,17 ART-induced BMD reductions and accelerated bone loss may also be related to immune reconstitution18,19 or increased levels of pro-inflammatory cytokines.20

The aim of the CASTLE BMD substudy was to evaluate treatment-related differences in regional BMD over 96 weeks in HIV-infected treatment-naive patients randomized to receive tenofovir DF/emtricitabine (TDF/FTC) plus atazanavir/ritonavir (ATV/r) or lopinavir/ritonavir (LPV/r).


Study Design

All participants were enrolled in the main study,21 a 96-week, open-label, randomized, international multicenter study designed to compare the antiviral activity of ATV/r and LPV/r each in combination with a fixed dose of TDF/FTC in treatment-naive HIV-infected patients (300/100 mg ATV/r once per day or 400/100 mg LPV/r twice per day, each with 300/200 mg TDF/FTC once per day). Participation in the BMD substudy was voluntary and offered to all patients randomized at sites with DXA facilities (224 patients: 125 on ATV/r and 99 on LPV/r). Only patients with paired BMD measurements at baseline and 48 weeks or at baseline and 96 weeks were included in the analysis.

The study was performed in accordance with Good Clinical Practice and the ethical principles of the Declaration of Helsinki. The protocol was approved by the institutional review board at each study site, and patients provided written informed consent before participation in the study. The trial is registered with, number NCT00272779.


Antiretroviral (ARV) treatment–naive patients with HIV-1 infection, >18 years old, and with HIV-1 RNA ≥5000 copies per milliliter and no CD4 cell count restriction were enrolled. Randomization was stratified by HIV-1 RNA at enrollment (<100,000 vs ≥100,000 copies per milliliter) and by geographic region. Other details of study methodology have been published.21

Substudy Endpoint

The primary endpoint was DXA BMD change from baseline (defined as within 30 d of starting study treatment) in trunk, leg, arm, and total body regions at 48 and 96 weeks expressed as grams per square centimeter. DXA scans were submitted to a centralized reader for interpretation. T scores were not calculated because these are validated for use in postmenopausal women and in men aged 50 years and older for the lumbar spine, total hip, or femoral neck regions.22 Given the younger average age of patients with HIV-1 infection and the fact that we did not examine lumbar spine, total hip, or femoral neck regions, the use of T scores in this study would have been inappropriate.

Statistical Analysis

Statistical evaluation was based on observed values for as-treated patients in this substudy. Because changes in BMD typically decay exponentially over time, values were log transformed before analysis. For univariate analyses, generalized linear models were conducted and coefficients back transformed to provide estimates of mean percent changes from baseline. Treatment regimens were compared using the difference in mean percent changes from baseline to weeks 48 and 96 with 95% confidence intervals (CIs) and P values based on t tests. The effect of baseline covariates [age, gender, smoking status, body mass index (BMI) categories, specific concomitant medications known to decrease BMD (dexamethasone, methylprednisolone, prednisolone, prednisone, lansoprazole, omeprazole, pantoprazole, phenobarbital, and warfarin), CD4 count categories, HIV-1 RNA categories] on BMD changes with treatment was explored in multivariate regression models. Differences in ATV/r vs LPV/r adjusted least squares (LS) mean percent changes from baseline were calculated from the multivariate models to provide an assessment of treatment effect having controlled for potentially relevant baseline covariates. All statistical analyses were conducted using the SAS software package, version 8.2 (SAS Institute, Cary, NC). Statistical significance was determined at the α = 0.05 level, unadjusted for multiple comparisons.



Of the 878 HIV-infected treatment-naive patients treated in the main study, 224 patients (125 on the ATV/r and 99 on the LPV/r) participated in the BMD substudy. Overall, 68.3% of the substudy participants were males and 41.5% were non-Hispanic white. Median age was 35 years, median BMI was 23.3 kg/m2, median CD4 cell count was 190 cells per microliter, and median HIV-1 RNA load was 5.06 log10 copies per milliliter. Baseline characteristics were similar between treatment arms (see Table S1, Supplemental Digital Content, and in men and women between treatment arms (see Table S2, Supplemental Digital Content, A greater proportion of women were of Black or African American racial origin (25.4%) compared with men (2.6%); however, these women were equally distributed across the ATV/r (25.0%) and LPV/r (25.8%) treatment groups. The discontinuation rate of substudy participants was 14.7% overall.

Univariate Analysis of BMD Mean Percent Change From Baseline With Treatment

Paired baseline to week 48 and baseline to week 96 BMD data were available for 118 and 106 patients on ATV/r and 82 and 70 patients on LPV/r, respectively. BMD declined significantly at 48 and 96 weeks in all body regions, regardless of treatment group. No significant between-treatment differences were observed for changes in BMD in arm or leg regions; however, significant differences were observed for trunk and total body regions and are shown in Figure 1.

At week 96, trunk BMD declined to a smaller extent on ATV/r (−2.80%; 95% CI: −3.73 to −1.86) compared with LPV/r (−4.60%; 95% CI: −5.69 to −3.49), resulting in a significant ATV/r vs LPV/r difference estimate of 1.89% (95% CI: 0.37 to 3.42; P = 0.015). At week 96, total body BMD declined −2.65% (95% CI: −3.44 to −1.87) in the ATV/r group and −3.70% (95% CI: −4.50 to −2.89) in the LPV/r group, with an ATV/r vs LPV/r difference estimate of 1.08% (95% CI: −0.12 to 2.30; P = 0.077).

At 96 weeks, loss in trunk (P < 0.0001) and total body BMD (P = 0.006) was significantly more pronounced in male patients taking LPV/r than in male patients taking ATV/r. In women, between-treatment differences for both trunk (P = 0.164) and total body (P = 0.817) regions were not significant (Fig. 1).

Mean percent change from baseline and 95% CIs in trunk and total body BMD at 48 and 96 weeks in the overall patient group, in female patients, and in male patients. B/L, baseline.

Multivariate Analysis of BMD Mean Percent Change From Baseline With Treatment and Adjusted for Baseline Covariates

For both trunk and total body regions, variables significantly associated with changes in BMD were treatment, baseline CD4 count, and age (Table 1). The loss of BMD at 96 weeks in patients receiving LPV/r was greater by 2% (trunk) and 1.3% (total body) than that observed in patients receiving ATV/r. The loss of BMD at 96 weeks in patients with low baseline CD4 counts (<50 cells per microliter) was greater by 2.7% (trunk) and 2.5% (total body) than that observed in patients with high baseline CD4 counts (≥200 cells per microliter). BMD declined faster with advancing age; for every 1 year increase in age, BMD declined by an additional 0.1% (trunk) and 0.2% (total body) at 96 weeks. Taking all covariates into account, the ATV/r vs LPV/r difference in adjusted LS mean percent change at 96 weeks for trunk BMD was 2.00% (95% CI: 0.52 to 3.45; P = 0.008) and for total body BMD was 1.24% (95% CI: 0.13 to 2.35; P = 0.029). For the total body region, low baseline BMI category (<22 kg/m2) was associated with a numeric 1.3% decrease in BMD at 96 weeks compared with the medium baseline BMI category (22 to <27 kg/m2) (P = 0.054).

Multivariate Regression Analysis of Effect of Treatment and Baseline Covariates on Percent Changes From Baseline in Trunk and Total Body BMD at 96 Weeks


In this study, patients initiating treatment on both ritonavir-boosted protease inhibitors (PI) regimens had declines in BMD ranging from 1% to 5% from baseline to weeks 48 and 96, consistent with earlier findings showing an estimated 2%–6% reduction in BMD within the first 2 years of ART initiation.8,11,23 The majority of the change in BMD was observed over the first 48 weeks. This leveling-off over time is consistent with data from ARV initiation studies and from longitudinal studies of treatment-experienced patients showing stability in BMD over time.4,24,25 The clinical significance of these changes in BMD is unclear.23,26

At week 96, compared with the LPV/r group, the ATV/r group was associated with smaller declines in trunk and total body BMD, with multivariate-adjusted LS mean difference estimates of +2.00% and +1.24%, respectively. In univariate analyses, men receiving LPV/r showed significantly greater losses in trunk and total body BMD than men receiving ATV/r; there were no between-treatment differences in demographic or baseline characteristics that could potentially explain these differences in men. In contrast, women did not show significant between-treatment differences in the rate of BMD loss for any region. In multivariate analyses, gender differences were not evident; rather, low baseline CD4 cell count (<50 cells per microliter)27 and increasing age were associated with reductions in trunk and total body BMD at 96 weeks, irrespective of treatment. For the trunk region, there was a trend toward greater BMD loss in patients with low baseline BMI (<22 kg/m2).

The substudy was not designed or powered to assess the impact of specific ARV components of each regimen on BMD; however, several explanations can be offered to potentially account for the observed differences in regional BMD loss. First, the BMD loss observed may have been related to treatment differences in T-cell activation and reconstitution secondary to the decrease in viral replication after the introduction of ART.19 Although the finding on multivariate analysis that low baseline CD4 count was associated with greater BMD loss at 96 weeks lends weight to this hypothesis,27 the statistical interaction between treatment and low baseline CD4 count was not tested. However, it is unlikely that treatment differences in immune reconstitution and inflammatory cytokine production were responsible for the ATV/r vs LPV/r differences in BMD because, in a further sub-analysis of CASTLE data, week 48 interleukin-6 and tumor necrosis factor alpha values were slightly higher in the ATV/r group.28 Second, body weight and body fat composition are known to influence bone metabolism in normal populations,29 and these factors are of particular importance among patients with HIV-1 infection who may be underweight or develop fat redistribution syndrome.30 In a further sub-analysis of CASTLE data, ATV/r-treated patients within the lowest BMI category (<22 kg/m2) showed a significant increase in both visceral and subcutaneous adipose tissues compared with LPV/r patients and also a significant increase in trunk:limb fat ratio.31 Thus, a greater “return to health” phenomenon in ATV/r-treated patients compared with LPV/r-treated patients may have contributed to the between-group differences in change in BMD at 96 weeks. Third, in previous clinical trials, patients with HIV-1 infection have experienced greater decreases in hip and spine BMD while on TDF compared with patients on stavudine14 or abacavir.8 In the A5224s substudy, TDF/FTC with ATV/r was associated with significantly greater bone loss in the spine, but not in the hip, than TDF/FTC with efavirenz.23 Increased bone turnover and BMD loss observed in regimens containing a ritonavir-boosted PI combined with TDF may be related to increased tenofovir exposure in the presence of low-dose ritonavir because the latter has been shown to inhibit the renal clearance of tenofovir leading to an increase in both plasma and intracellular tenofovir concentrations.23 Given that the total daily dose of ritonavir with LPV/r was double that with ATV/r, an increase in TDF-mediated BMD loss in the LPV/r could potentially explain the between-group differences in change in BMD at 96 weeks in this study. However, although some data have suggested increased tenofovir exposures with LPV/r,32,33 ATV/r,34 or ritonavir35 coadministration, data from a separate CASTLE intensive pharmacokinetic substudy in patients with HIV-1 infection did not show increases in tenofovir exposures with either LPV/r (n = 19) or ATV/r (n = 17) administration.36 Moreover, the TELEX II study, in which steady-state tenofovir pharmacokinetics were assessed before and after reducing ritonavir boosting of a fosamprenavir/TDF/FTC regimen from 200 to 100 mg once daily, demonstrated a significant 30% increase (not decrease) in tenofovir exposures.37 Finally, recent in vitro studies demonstrated that LPV/r, and to a lesser extent ATV/r, reduced the proliferation of human bone marrow mesenchymal stem cells and thus their capacity to differentiate into osteoblasts, a phenomenon that was reversed by pretreatment with pravastatin.38

Limitations of this analysis include the small sample size, the lack of baseline vitamin D or parathyroid hormone assessments, and other potential unmeasured baseline factors that may have contributed to the observed between-treatment differences. These unmeasured baseline factors included level of physical activity, alcohol consumption, and comorbid medical conditions influencing BMD, such as rheumatoid arthritis. Although the use of some concomitant medications potentially affecting BMD (mainly systemic steroids and proton pump inhibitors) was adjusted for in the multivariate analysis, information on other potentially relevant concomitant medications, such as bisphosphonates or calcium supplements, was lacking.


BMD loss occurred in both ATV/r- and LPV/r-treated patients, consistent with studies showing BMD declines of 2%–6% within the first 2 years of ART initiation.10 In this subpopulation, BMD declined to a greater extent in patients receiving LPV/r compared with ATV/r. Other factors associated with BMD decline included increased patient age and more advanced HIV disease (lower baseline CD4 counts). The reasons for differences in BMD losses observed between regimens containing a boosted PI in combination with TDF remain unclear, are likely to be multifactorial, and warrant further investigation.


1. Bolland MJ, Wang TK, Grey A, et al.. Stable bone density in HAART-treated individuals with HIV: a meta-analysis. J Clin Endocrinol Metab. 2011;96:2721–2731.
2. Brown TT, Qaqish RB. Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review. AIDS. 2006;20:2165–2174.
3. Zhao H, Goetz MB. Complications of HIV infection in an ageing population: challenges in managing older patients on long-term combination antiretroviral therapy. J Antimicrob Chemother. 2011;66:1210–1214.
4. Cassetti I, Madruga JV, Suleiman JM, et al.. The safety and efficacy of tenofovir DF in combination with lamivudine and efavirenz through 6 years in antiretroviral-naive HIV-1-infected patients. HIV Clin Trials. 2007;8:164–172.
5. Duvivier C, Kolta S, Assoumou L, et al.. Greater decrease in bone mineral density with protease inhibitor regimens compared with nonnucleoside reverse transcriptase inhibitor regimens in HIV-1 infected naive patients. AIDS. 2009;23:817–824.
6. Hansen AB, Obel N, Nielsen H, et al.. Bone mineral density changes in protease inhibitor-sparing vs. nucleoside reverse transcriptase inhibitor-sparing highly active antiretroviral therapy: data from a randomized trial. HIV Med. 2011;12:157–165.
7. Rivas P, Gorgolas M, Garcia-Delgado R, et al.. Evolution of bone mineral density in AIDS patients on treatment with zidovudine/lamivudine plus abacavir or lopinavir/ritonavir. HIV Med. 2008;9:89–95.
8. Stellbrink HJ, Orkin C, Arribas JR, et al.. Comparison of changes in bone density and turnover with abacavir-lamivudine versus tenofovir-emtricitabine in HIV-infected adults: 48-week results from the ASSERT study. Clin Infect Dis. 2010;51:963–972.
9. van Vonderen MG, Lips P, van Agtmael MA, et al.. First line zidovudine/lamivudine/lopinavir/ritonavir leads to greater bone loss compared to nevirapine/lopinavir/ritonavir. AIDS. 2009;23:1367–1376.
10. Triant VA, Brown TT, Lee H, et al.. Fracture prevalence among human immunodeficiency virus (HIV)-infected versus non-HIV-infected patients in a large U.S. healthcare system. J Clin Endocrinol Metab. 2008;93:3499–3504.
11. Brown TT, McComsey GA, King MS, et al.. Loss of bone mineral density after antiretroviral therapy initiation, independent of antiretroviral regimen. J Acquir Immune Defic Syndr. 2009;51:554–561.
12. Amorosa V, Tebas P. Bone disease and HIV infection. Clin Infect Dis. 2006;42:108–114.
13. Walker Harris V, Brown TT. Bone loss in the HIV-infected patient: evidence, clinical implications, and treatment strategies. J Infect Dis. 2012;205(suppl 3):S391–S398.
14. Gallant JE, Staszewski S, Pozniak AL, et al.. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA. 2004;292:191–201.
15. Woodward CL, Hall AM, Williams IG, et al.. Tenofovir-associated renal and bone toxicity. HIV Med. 2009;10:482–487.
16. Borderi M, Gibellini D, Vescini F, et al.. Metabolic bone disease in HIV infection. AIDS. 2009;23:1297–1310.
17. Pan G, Yang Z, Ballinger SW, et al.. Pathogenesis of osteopenia/osteoporosis induced by highly active anti-retroviral therapy for AIDS. Ann N Y Acad Sci. 2006;1068:297–308.
18. Franco JM, Rubio A, Martinez-Moya M, et al.. T-cell repopulation and thymic volume in HIV-1-infected adult patients after highly active antiretroviral therapy. Blood. 2002;99:3702–3706.
19. Ofotokun I, Weitzmann MN. HIV and bone metabolism. Discov Med. 2011;11:385–393.
20. Arora S, Agrawal M, Sun L, et al.. HIV and bone loss. Curr Osteoporos Rep. 2010;8:219–226.
21. Molina JM, Andrade-Villanueva J, Echevarria J, et al.. Once-daily atazanavir/ritonavir versus twice-daily lopinavir/ritonavir, each in combination with tenofovir and emtricitabine, for management of antiretroviral-naive HIV-1-infected patients: 48 week efficacy and safety results of the CASTLE study. Lancet. 2008;372:646–655.
22. The International Society for Clinical Densitometry. 2013 ISCD Official Positions—Adult. Available at: Accessed August 08, 2014.
23. McComsey GA, Kitch D, Daar ES, et al.. Bone mineral density and fractures in antiretroviral-naive persons randomized to receive abacavir-lamivudine or tenofovir disoproxil fumarate-emtricitabine along with efavirenz or atazanavir-ritonavir: Aids Clinical Trials Group A5224s, a substudy of ACTG A5202. J Infect Dis. 2011;203:1791–1801.
24. Mondy K, Yarasheski K, Powderly WG, et al.. Longitudinal evolution of bone mineral density and bone markers in human immunodeficiency virus-infected individuals. Clin Infect Dis. 2003;36:482–490.
25. Nolan D, Upton R, McKinnon E, et al.. Stable or increasing bone mineral density in HIV-infected patients treated with nelfinavir or indinavir. AIDS. 2001;15:1275–1280.
26. Mundy LM, Youk AO, McComsey GA, et al.. Overall benefit of antiretroviral treatment on the risk of fracture in HIV: nested case-control analysis in a health-insured population. AIDS. 2012;26:1073–1082.
27. Grant PM, Kitch D, McComsey GA, et al.. Low baseline CD4+ count is associated with greater bone mineral density loss after antiretroviral therapy initiation. Clin Infect Dis. 2013;57:1483–1488.
28. Moyle G, Hardy H, Uy J, et al.. Changes in biomarkers in HIV-1-infected treatment-naive patients treated with tenofovir DF/emtricitabine plus atazanavir/ritonavir or lopinavir/ritonavir for 96 weeks: the CASTLE biomarker substudy. Antivir Ther. 2014. [Epub ahead of print].
29. Reid IR. Relationships between fat and bone. Osteoporos Int. 2008;19:595–606.
30. Sharma A, Tian F, Yin MT, et al.. Association of regional body composition with bone mineral density in HIV-infected and HIV-uninfected women: women's interagency HIV study. J Acquir Immune Defic Syndr. 2012;61:469–476.
31. Moyle GJ, Hardy H, Farajallah A, et al.. Comparison of body composition changes between atazanavir/ritonavir and lopinavir/ritonavir each in combination with Tenofovir/Emtricitabine in antiretroviral-naive patients with HIV-1 infection. Clin Drug Investig. 2014;34:287–296.
32. Kearney BP, Mathias A, Mittan A, et al.. Pharmacokinetics and safety of tenofovir disoproxil fumarate on coadministration with lopinavir/ritonavir. J Acquir Immune Defic Syndr. 2006;43:278–283.
33. Kiser JJ, Carten ML, Aquilante CL, et al.. The effect of lopinavir/ritonavir on the renal clearance of tenofovir in HIV-infected patients. Clin Pharmacol Ther. 2008;83:265–272.
34. Bristol-Myers Squibb Company. Prescribing Information for Reyataz (Atazanavir). Available at: Accessed June 24, 2014.
35. Baxi SM, Greenblatt RM, Bacchetti P, et al.. Common clinical conditions—age, low BMI, ritonavir use, mild renal impairment—affect tenofovir pharmacokinetics in a large cohort of HIV-infected women. AIDS. 2014;28:59–66.
36. Zhu L, Liao S, Child M, et al.. Pharmacokinetics and inhibitory quotient of atazanavir/ritonavir versus lopinavir/ritonavir in HIV-infected, treatment-naive patients who participated in the CASTLE Study. J Antimicrob Chemother. 2012;67:465–468.
37. Parks DA, Jennings HC, Taylor C, et al.. Steady-state amprenavir, tenofovir, and emtricitabine pharmacokinetics before and after reducing ritonavir boosting of a fosamprenavir/tenofovir/emtricitabine regimen from 200 mg to 100 mg once daily (TELEX II). HIV Clin Trials. 2009;10:160–167.
38. Hernandez-Vallejo SJ, Beaupere C, Larghero J, et al.. HIV protease inhibitors induce senescence and alter osteoblastic potential of human bone marrow mesenchymal stem cells: beneficial effect of pravastatin. Aging Cell. 2013;12:955–965.

atazanavir; ritonavir; lopinavir; bone mineral density; body mass index

Supplemental Digital Content

© 2015 by Lippincott Williams & Wilkins