Comorbidities such as osteoporosis and cardiovascular disease (CVD) contribute to increased morbidity and mortality in HIV-infected persons, on potent antiretroviral therapy (ART). The cross-talk between the immune and the skeletal systems has been implicated in the pathogenesis of cardiovascular and bone disease in the setting of HIV infection despite suppressive ART.1–3 The receptor activator of the nuclear factor kappa-B (NF-kB) ligand (RANKL)/osteoprotegerin (OPG) system plays a critical role in regulating bone metabolism, the immune system, and CVD. In the setting of inflammation, RANKL expression is induced in osteoblasts and synovial fibroblasts and promotes osteoclast activity.3 OPG binds RANKL, counteracts the effects of RANKL, and is induced in multiple tissues including bone.3 Both RANKL and OPG are also produced by myeloid cells, lymphocytes which are involved in inflammation, osteoclastogenesis, and atherogenesis.3 A higher ratio of circulating RANKL/OPG in the setting of inflammation promotes osteoclastogenesis and atherogenesis.3
We have previously reviewed the conflicting data on the role of the RANKL/OPG axis in HIV infection.3 In vitro and in vivo studies have suggested that both HIV-1 RNA and ART can directly and independently affect the RANKL/OPG levels. Plasma levels represent only a small fraction of the total tissue levels/activity of RANKL/OPG.3 However, there are limited data in the setting of ART initiation as to whether increased plasma levels of the RANKL/OPG axis are associated with surrogate measures of CVD and bone disease in chronic HIV infection. Elucidating the role of RANKL/OPG axis in systemic inflammation and immune dysfunction in the setting of suppressed viremia may help us understand mechanisms of HIV-related CVD and bone disease.
ACTG A5260s was a prospective substudy nested into an ART treatment study (A5257), in which participants were randomized to tenofovir disoproxil fumarate–emtricitabine (TDF/FTC) plus either raltegravir (RAL) or atazanavir/ritonavir (ATV/r) or darunavir/ritonavir (DRV/r).4–6 In this study, we found that ATV/r was associated with slower progression of carotid intima-media thickness (∆CIMT) compared with DRV/r or RAL4 and that markers of inflammation and immune activation inconsistently and partially decreased after these regimens.5 We also found that bone mineral density (BMD) losses were lowest among those receiving 96 weeks of RAL compared with protease inhibitors (PIs).6
In this article, we describe for the first time in the setting of a large randomized ART initiation trial, the effects of different successful ART regimens on plasma levels of RANKL, OPG, and the ratio (RANKL/OPG3) on surrogate markers of CVD and bone disease. In view of limited data that suggest favorable effects of integrase inhibitors compared with PIs on markers of inflammation,7–9 we hypothesized that RAL would increase plasma OPG and reduce the plasma RANKL and RANKL/OPG ratio compared with both PIs.
Study Design and Participants
A5260s included 328 HIV-infected, ART-naive adults with no CVD or diabetes mellitus. The design of this substudy (clinicalTrials.gov identifier: NCT00851799) has been previously reported4,6 and was approved by the Institutional Review Boards at all participating institutions; all participants provided written informed consent. To avoid the confounding effects of viremia on plasma levels of RANKL and OPG, the analysis population was restricted to a subset of 234 (71%) virologically suppressed participants (no ART interruptions >7 days, HIV-1 RNA <50 copies per milliliter by study week 24 and thereafter).
Measurement of biomarkers, CIMT, and BMD (total, lumbar spine, hip).
Plasma RANKL and OPG were determined at study entry before ART initiation and after 48 and 96 weeks on treatment, as previously described.6 CIMT and BMD were determined before ART initiation and after 48 (CIMT), 96 (CIMT, BMD), and 144 weeks (CIMT). Primary results and methodologies are reported elsewhere.4,6
Pairwise treatment group comparisons of changes from baseline were assessed by Wilcoxon rank sum tests. Associations between on-treatment levels of the biomarkers and BMD change (% change from baseline to week 96) were evaluated with linear regression models, adjusted for randomized treatment, stratification factors (screening viral load and 10-year Framingham risk score), and baseline biomarkers levels. Consistent with the study primary analysis, associations between biomarkers and CIMT were examined using mixed effects linear regression models with random intercept and slope and unstructured covariance matrix on the random effect, adjusted for baseline CIMT, time (years), and baseline Framingham risk score and HIV-RNA stratification factors.4,6 All treatment group comparisons were assessed with a 2.5% alpha level; all other comparisons used a 5% level, without adjustment for multiple comparisons. All analyses were performed with SAS, version 9.4 (SAS Institute, Cary, NC).
Baseline demographic characteristics of the 328 participants from the A5260s study were previously described.10,11 The 234 (220 in the current substudy) participants included in the successfully treated population had similar baseline demographic characteristics across treatment groups.5 Briefly, 90% were men, 48% white, with a median age of 36 years, CD4+ cell count 338 cells/mm3, and HIV-1 RNA 4.6 log10 copies per milliliter. Levels of RANKL, OPG, and RANKL/OPG per timepoint are presented in Table 1. No notable differences in distributions are seen across treatment groups (Supplemental Digital Content Tables 1 and 2, http://links.lww.com/QAI/B136, Supplemental Digital Content Fig. 1, http://links.lww.com/QAI/B136).
Changes in Plasma RANKL and OPG
Consistent with the declines in markers of inflammation,5 decreases in plasma RANKL from baseline were observed at week 48 that were sustained to week 96 among all participants (Table 1) and among all treatment groups (Supplemental Digital Content Table 1, http://links.lww.com/QAI/B136). Specifically, levels of RANKL at 96 weeks were on average 50% lower than baseline measures. Increases (approximately at least 10% higher than baseline measures) in plasma OPG from baseline were noted only at week 96 among all participants (Table 1) and among all treatment groups (Supplemental Digital Content Table 1, http://links.lww.com/QAI/B136).
Associations Between RANKL, OPG, RANKL/OPG Ratio, BMD, and CIMT
In baseline OPG-adjusted models, there was a trend to higher OPG levels at 48 and 96 weeks being associated with spine (but not hip) BMD loss (P = 0.07). Examining OPG levels on study, without baseline adjustments, higher levels of OPG at week 48 and 96 were associated with a larger decrease in spine BMD [1.04% (P = 0.039) and 1.27% (P = 0.034)]. Associations were not observed between levels of RANKL and RANKL/OPG and lumbar spine and total hip BMD (P > 0.36) (Table 2). No associations between CIMT and levels of RANKL, OPG, RANKL/OPG at 48 weeks were apparent (P > 0.35) (Table 2).
In this prospective study of ART-naive participants who achieved virologic suppression after initiation of TDF/FTC along with either RAL, ATV/r, or DRV/r, plasma RANKL and RANKL/OPG ratio were not associated with BMD or CIMT change. Higher OPG level on ART was associated with greater spine BMD loss in models that did not adjust for baseline OPG level and marginally in baseline-adjusted models. RAL did not seem to have a more favorable effect on increasing OPG and decreasing RANKL and RANKL/OPG ratio compared with the PIs. To our knowledge, this is the largest prospective study that explored associations of plasma RANKL and OPG with markers of CVD and bone disease. Overall, given previous in vivo and in vitro data that support a major role of the RANKL-OPG axis in pathogenesis of CVD, bone disease, and HIV infection and the fact that plasma levels represent only a small fraction of the total tissue levels/activity of RANKL/OPG,3 these results suggest that plasma levels of the RANKL-OPG axis may not be the best surrogate markers3,12 to predict bone disease and CVD in chronic treated HIV infection.
The differential effects of integrase inhibitors compared with PIs on RANKL/OPG remain unclear.3 We did not find any benefit of RAL over PIs on reducing RANKL or RANKL/OPG ratio during the first 96 weeks of successful treatment. A strength of our study is that, in contrast to previous RAL switch or intensification-related studies, previous ART does not confound our data because we enrolled only ART-naive participants and followed them for 96 weeks, a longer duration than previous studies. However, changes in levels of RANKL and OPG were also measured in the setting of initiation of different ART regimens during a period where there may be major changes in inflammation and immune activation that may also affect the RANKL/OPG axis.3 Thus, the specific effect of ART on the RANKL/OPG axis may best be examined in HIV-uninfected persons receiving ART for HIV prevention or in HIV-infected persons on longer term potent ART who switch ARTs.
Consistent with the hypothesis that the RANKL/OPG axis contributes to HIV-related bone disease, we found that participants with higher plasma OPG after 48 weeks of ART had more spine BMD loss. Increased levels of OPG in HIV-infected individuals may reflect a compensatory mechanism to downregulate increased bone resorption.3 However, like others, we did not find an association between plasma RANKL/OPG ratio with measures of bone disease.13
Reports of associations between circulating levels of OPG or RANKL with CVD in HIV-uninfected subjects have been conflicting.3 Despite evidence from retrospective studies that serum RANKL was negatively associated with the coronary artery plaque burden and calcium score in HIV-infected individuals on ART compared with controls,12 similar to our data, in a smaller prospective study of HIV-positive participants on ART, there were no significant associations of serum RANKL, OPG, and RANKL/OPG with CIMT.14 In the largest study to date that has investigated the associations of plasma levels of RANKL and OPG with subclinical atherosclerosis in chronic HIV infection, only higher OPG (but not lower RANKL) concentrations were associated with the presence of coronary artery calcium, mixed plaque, and coronary stenosis >50% in HIV-infected men.15 Thus, plasma OPG rather than plasma RANKL levels may be a better surrogate marker for HIV-related morbidity.
Our study had several limitations that have previously been described.5 Briefly, these include limited power to detect effect sizes with adjustment for multiple biomarker comparisons, selection bias of A5260s participants when restricting to the cohort of virologically suppressed individuals on potent ART, and inclusion of mostly men, which may limit generalizability of our findings. Finally, the relatively younger age of our study population without clinical CVD may have limited our ability to study the cross-talk between RANKL/OPG, CVD, and bone disease.
In conclusion, in this prospective study, plasma levels of the RANKL and RANKL/OPG were not associated with CIMT and BMD change after 96 weeks among treatment-naive individuals initiating and remaining on successful ART regimens of TDF/FTC with RAL, ATV/r, or DRV/r. Our data suggest that higher levels of plasma OPG may predict BMD loss in ART-treated HIV-infected individuals. Given that plasma levels of RANKL and OPG may not reflect local tissue expression, and in view of the role of RANKL/OPG axis in vascular calcification and bone disease,3 these results highlight the need to further understand the role of the RANKL/OPG axis at the tissue (arteries, bone, and lymphocytes) level. Larger prospective studies that will also explore tissue levels of RANKL, OPG, and CVD endpoints such as plaque may more definitely address the role of RANKL/OPG axis in HIV-related CVD and bone disease. Elucidating the role of the RANKL/OPG axis in HIV infection may lay the foundation for novel strategies to prevent future bone or CVD in this population.
ACTG 5260s Team Members: H. Ribaudo, H. Hodis, C. Godfrey, B. Jarocki, A. Benns, K. Braun. The authors thank the staff and patients from the following hospitals who participate in ACTG (in alphabetical order): Beth Israel Deaconess Medical Center, Brigham and Womens Hospital, Case University CRS, Duke University Medical Center, Harbor-UCLA Medical Center, Houston AIDS Research Team CRS, John Hopkins Adult AIDS CRS, Metrohealth, New Jersey Medical School, New York University HIV/AIDS CRS, Northwestern University, Rush University Medical Center ACTG, The Ohio State University, The Ponce De Leon Center CRS, UCLA Care Center, UCSF AIDS CRS, University of Cincinnati, University of Colorado, University of North Carolina AIDS CRS, University of Pittsburg CRS, University of Rochester ACTG AIDS Care, University of Southern California, University of Washington, Vanderbilt Therapeutics CRS, Washington University.
1. Vikulina T, Fan X, Yamaguchi M, et al. Alterations in the immuno-skeletal interface drive bone destruction in HIV-1 transgenic rats. Proc Natl Acad Sci U S A. 2010;107:13848–13853.
2. Brown TT, Tassiopoulos K, Bosch RJ, et al. Association between systemic inflammation and incident diabetes in HIV-infected patients after initiation of antiretroviral therapy. Diabetes Care. 2010;33:2244–2249.
3. Kelesidis T, Currier JS, Yang OO, et al. Role of RANKL-RANK/osteoprotegerin pathway in cardiovascular and bone disease
associated with HIV infection. AIDS Rev. 2014;16:123–133.
4. Stein JH, Ribaudo HJ, Hodis HN, et al. A prospective, randomized clinical trial of antiretroviral therapies on carotid wall thickness. AIDS. 2015;29:1775–1783.
5. Kelesidis T, Tran TT, Stein JH, et al. Changes in inflammation and immune activation with atazanavir-, raltegravir-, darunavir-based initial antiviral therapy: ACTG 5260s. Clin Infect Dis. 2015;61:651–660.
6. Brown TT, Moser C, Currier JS, et al. Changes in bone mineral density after initiation of antiretroviral treatment with tenofovir disoproxil fumarate/emtricitabine plus atazanavir/ritonavir, darunavir/ritonavir, or raltegravir. J Infect Dis. 2015;212:1241–1249.
7. Asmuth DM, Ma ZM, Mann S, et al. Gastrointestinal-associated lymphoid tissue immune reconstitution in a randomized clinical trial of raltegravir versus non-nucleoside reverse transcriptase inhibitor-based regimens. AIDS. 2012;26:1625–1634.
8. Hatano H, Strain MC, Scherzer R, et al. Increase in 2-long terminal repeat circles and decrease in D-dimer after raltegravir intensification in patients with treated HIV infection: a randomized, placebo-controlled trial. JInfectDis. 2013;208:1436–1442.
9. Martinez E, D'Albuquerque PM, Llibre JM, et al. Changes in cardiovascular biomarkers in HIV-infected patients switching from ritonavir-boosted protease inhibitors to raltegravir. AIDS. 2012;26:2315–2326.
10. Brown TT, Chen Y, Currier JS, et al. Body composition, soluble markers of inflammation, and bone mineral density in antiretroviral therapy-naive HIV-1-infected individuals. J Acquir Immune Defic Syndr. 2013;63:323–330.
11. Stein JH, Brown TT, Ribaudo HJ, et al. Ultrasonographic measures of cardiovascular disease risk in antiretroviral treatment-naive individuals with HIV infection. AIDS. 2013;27:929–937.
12. Hwang JJ, Wei J, Abbara S, et al. Receptor activator of nuclear factor-kappaB ligand (RANKL) and its relationship to coronary atherosclerosis in HIV patients. J Acquir Immune Defic Syndr. 2012;61:359–363.
13. Seminari E, Castagna A, Soldarini A, et al. Osteoprotegerin and bone turnover markers in heavily pretreated HIV-infected patients. HIV Med. 2005;6:145–150.
14. Kelesidis T, Kendall MA, Yang OO, et al. Perturbations of circulating levels of RANKL-osteoprotegerin axis in relation to lipids and progression of atherosclerosis in HIV-infected and -uninfected adults: ACTG NWCS 332/A5078 Study. AIDS Res Hum Retroviruses. 2013;29:938–948.
15. Ketlogetswe KS, McKibben R, Jacobson LP, et al. Osteoprotegerin, but not receptor activator for nuclear factor-kappaB ligand, is associated with subclinical coronary atherosclerosis in HIV-infected men. J Acquir Immune Defic Syndr. 2015;70:362–369.