Impact of Zoledronic Acid on Bone Mineral Density and Trabecular Score Following Denosumab Discontinuation in Older Adults in Long-Term Care

Abstract

Background:

Discontinuation of denosumab can result in rebound bone loss and increased vertebral fracture risk. In residents of long-term care communities (LTCCs) with osteoporosis, there is limited data on managing the risks after discontinuation. We investigated the impact of a single dose of zoledronic acid on bone density and microarchitecture following two years of denosumab treatment.

Methods:

In an open-label, one-year extension study following a two-year double-blind, placebo-controlled, randomized clinical trial, 39 older adults aged 65 years and above, who were residents of LTCCs and participants in the PROUD (PReventing Osteoporosis Using Denosumab) trial, received a single 5 mg dose of zoledronic acid after completing four doses of denosumab 60 mg during the PROUD trial. We aim to evaluate the effects of a single 5 mg dose of zoledronic acid on bone mineral density (BMD) at the lumbar spine, total hip, femoral neck, and one-third radius, as well as on the spine trabecular bone score (TBS), over a one-year period. Additionally, we surveyed patients for fractures.

Results:

Our study included 27 women and 12 men, with a mean age of 81.5 years. Twelve months after the administration of zoledronic acid, the mean percent changes from the end of the denosumab trial showed no significant decline in any of the BMD sites in both women and men. In women, the mean percent changes were as follows: spine 0.97% (95% CI: −0.7 to 2.7, p=0.242) and total hip −0.10% (95% CI: −2.3 to 2.1, p=0.927). In men, the changes were −0.32% (95% CI: −3.7 to 3.1, p=0.832) for the spine and 1.79% (95% CI: −0.7 to 4.3, p=0.139) for the total hip. These findings indicate no evidence of rebound bone loss. In women, TBS significantly increased by 3.9% (95% CI: 0.8 to 5.8, p=0.007), suggesting improved bone microarchitecture. In men, there was a trend toward improvement in TBS, with an increase of 3.3% (p=0.0545). There were no reported fragility fractures among participants during the post-denosumab period.

Conclusions:

In residents of LTCCs with osteoporosis receiving a single 5 mg dose of zoledronic acid following two years of denosumab, we found no evidence of a loss in BMD or TBS. Further, participants experienced enhanced bone microarchitecture.

INTRODUCTION:

Osteoporosis poses a significant health challenge, particularly among older adults residing in long-term care communities (LTCCs) [1, 2]. This condition is marked by a decrease in bone mineral density (BMD) and deterioration of bone microarchitecture, leading to substantially greater risk of fractures. Hip and spine fractures, in particular, can result in severe health complications, reduced independence, and increased mortality in such a vulnerable population [1, 3]. Despite the critical nature of osteoporosis management in LTCCs, evidence on effective long-term strategies remains limited.

Denosumab, a monoclonal antibody that inhibits the receptor activator of nuclear factor kappa-Β ligand (RANKL), has proven efficacy in increasing BMD in individuals with osteoporosis [4], including residents of LTCCs, as we previously demonstrated in the PROUD (Preventing Osteoporosis Using Denosumab) trial [5]. Despite its benefits, discontinuation of denosumab can lead to rapid bone turnover resulting in rebound bone loss [6]. This rebound effect is accompanied by an increased risk of vertebral fractures, which can occur as early as 6 to 12 months after stopping the treatment, highlighting the critical need for effective strategies to mitigate risk of adverse outcomes [6, 7].

To address denosumab discontinuation-related challenges, studies are focusing on sequential treatment strategies that can sustain the gains in bone density achieved with denosumab. In individuals for whom denosumab cannot be continued for any reason, clinical guidelines recommend transitioning to a bisphosphonate such as zoledronic acid to reduce the risks of rebound bone loss and increased fracture incidence [8]. Zoledronic acid is a long-lasting intravenous bisphosphonate that binds to bone and suppresses osteoclastic activity which helps preserve the skeletal benefits gained from denosumab therapy [8, 9]. Although some studies have explored the said sequential treatment regimen, the evidence is limited and primarily derived from healthier populations, making it unclear if these results apply to residents of LTCCs.

The current study builds on the findings of the two-year PROUD trial, which demonstrated substantial improvements in BMD at key skeletal sites, such as the lumbar spine and total hip, by using denosumab in residents of LTCCs [5]. In the present extension study, we followed a subset of participants who transitioned from denosumab to zoledronic acid to determine whether a single dose of zoledronic acid could maintain the BMD gains over a 12-month period. We assessed the effects of this transition on their BMD at various skeletal sites and also evaluated their lumbar spine bone microarchitecture using the trabecular bone score (TBS). By focusing on LTCC residents, we aimed to obtain preliminary evidence to fill a critical knowledge gap and offer practical insights for optimizing osteoporosis management in a high-risk group which is often underrepresented in clinical research.

METHODS:

Study Design:

The design of the parent trial (NCT02753283 on ClinicalTrials.gov) has been previously published [5]. The PROUD trial was a two-year, double-blind, placebo-controlled, gender-stratified randomized clinical trial conducted in LTCCs. It enrolled 201 women and men aged 65 years or older who met the criteria for osteoporosis pharmacological treatment [10]. Notably, participants with multimorbidity, dysmobility, and cognitive impairment were included. All study visits were conducted in the participants’ residential facilities. The aim of the PROUD trial was to evaluate the efficacy of denosumab in improving bone density among a high-risk population of residents in LTCCs. After the baseline assessment, participants were randomized to receive either denosumab 60 mg injections every six months or a matching placebo. They also consumed at least 1200 mg of calcium, from diet or supplements, and a minimum of 800 IU of vitamin D supplements daily. Follow-up visits occurred at 6, 12, 18, and 24 months. A significant challenge encountered during the parent trial was the onset of the COVID-19 pandemic in the United States, which impeded our ability to enroll new patients and conduct follow-up visits.

Upon completion of the PROUD trial, all participants were offered a single dose of zoledronic acid; however, only 39 of the participants in the denosumab arm agreed to receive one dose of intravenous (IV) zoledronic acid 5 mg, administered between 6 to 9 months after their fourth (final) denosumab injection who constitute our study sample. Their bone mineral density (BMD) at various skeletal sites and spine trabecular bone score (TBS) were monitored for 12 months following zoledronic acid infusion (Figure 1). Additionally, participants were surveyed for fractures every six months, and during the extension study period. The study was reviewed and approved by the University of Pittsburgh’s Institutional Review Board (IRB).

Figure 1.

Study Design and Timeline.

Measures:

Bone Mineral Density (BMD) (g/cm²):

We assessed BMD at multiple skeletal sites, including the lumbar spine, total hip, femoral neck, and the one-third radius. These measurements were obtained using dual-energy X-ray absorptiometry (DXA) with a Hologic Discovery densitometer (Hologic, Bedford, MA) that was housed in a mobile unit [11]. By utilizing a mobile DXA unit, we were able to bring the equipment directly to the participants, minimizing the need for travel and thereby enhancing the study’s feasibility and participant compliance. We also employed the same technician during both the main and extension trials, minimizing any between-technician variability. For the total hip, with a precision of 1.2%, the least significant change (LSC) was calculated to be 3.32%. For the femoral neck, with a precision of 1.9%, the LSC was 5.26%. Additionally, for the PA spine, with a precision of 1.5%, the LSC was 4.16%.

Trabecular Bone Score (TBS):

TBS was obtained from the DXA images of the lumbar spine. TBS serves as an analytical method that evaluates the lumbar spine’s two-dimensional DXA images to create three-dimensional representations of the bone’s microarchitecture [12]. We used Lumbar spine L1 to L4 to measure TBS. In contrast to standard BMD measurements, TBS offers insights into bone quality by analyzing the bone’s internal structure with parameters that are independent of BMD, thereby allowing for a more comprehensive assessment of fracture risk [13,14]. A TBS above 1.31 generally indicates normal bone microarchitecture, whereas scores between 1.24 and 1.31 suggest partially degraded microarchitecture, and values equal or lower than 1.23 are indicative of degraded bone microarchitecture [12].

Fragility Fractures:

Every six months, all participants were surveyed about any clinical fragility or nontraumatic fractures defined as those occurring after falls from a standing or sitting height.

Functional Vulnerability Assessment:

The Fried Frailty Index and a comorbidity questionnaire collectively assess functional vulnerability by evaluating physical frailty and overall disease burden. The Fried Frailty Index examines key frailty indicators [15], while the comorbidity questionnaire covers 18 medical conditions across eight physiologic domains, providing a comprehensive view of factors impacting health and physical performance [16]. Patients with higher frailty scores, typically a score of 3 or above, and those with more comorbid conditions are at increased risk for adverse health outcomes, including falls, hospitalization, and mortality [15, 16].

Statistical Analysis:

Analyses for men and women were conducted separately. Baseline characteristics of participants who received the extension treatment (one dose of zoledronic acid) after denosumab were summarized using appropriate descriptive statistics. Outcomes representing bone density and microarchitecture were summarized at the end of the parent trial period and 12-months after the zoledronic acid infusion, as well as percent change and its statistical significance using a paired samples t-test.

RESULTS:

Participants included 27 women and 12 men, with an average age of 81.5 years. 69.2% of participants were frail or prefrail. They had comorbidity in over 3 physiologic systems on average (Table 1). In women, percent changes in measures of bone indices from end of the denosumab trial were as follows: lumbar spine BMD 0.97% (95% confidence interval (CI): −0.7 to 2.7, p=0.242), total hip BMD −0.10% (95% CI: −2.3 to 2.1, p=0.927), femoral neck BMD 1.71% (95% CI: −0.3 to 3.7, p=0.099), one-third radius BMD −1.04% (95% CI: −2.6 to 0.5, p=0.186) and TBS 3.91% (95% CI: 0.8 to 5.8, p=0.007) (Table 2, Figure 2). In men: lumbar spine BMD −0.32% (95% CI: −3.7 to 3.1, p=0.832), total hip BMD 1.79% (95% CI: −0.7 to 4.3, p=0.139), femoral neck BMD 1.52% (95% CI: −3.6 to 6.6, p=0.505), one-third radius BMD 1.38% (95% CI: 0.3 to 2.4, p=0.015), and TBS 3.33% (95% CI: −4.0 to 13.0, p=0.054) (Table 2, Figure 2). After the denosumab phase of the trial ended, we did not receive any reports of fragility fractures among our participants.

Table 1:

Baseline Participant Characteristics

Characteristic	Women (N=27)	Men (N=12)	All Participants (N = 39)
Age, years	81.4±1.5	81.7±2.7	81.5±1.3
Race
White	26 (96.3)	12 (100.0)	38 (97.4)
Black	1 (3.7)	0 (0.0)	1 (2.6)
Other	0 (0.0)	0 (0.0)	0 (0.0)
BMI, kg/m2	29.4±1.3	27.7±1.5	28.9±1.0
GFR, mL/min/1.73m2	56.4±1.5	56.6±2.0	56.4±1.2
Vitamin D, ng/mL	51.1±3.1	42.8±3.4	48.6±2.4
Fried Frailty Index
Non-frail	7 (25.9)	5 (41.7)	12 (30.8)
Pre-frail	20 (74.1)	7 (58.3)	16 (41.0)
Frail	0 (0.0)	0 (0.0)	11 (28.2)
Comorbidity Index (0–8)a	3.3±0.2	3.4±0.4	3.4±0.2
BMD, g/cm2
Lumbar spine	1.052±0.026	1.202±0.076	1.098±0.031
Total hip	0.788±0.022	0.910±0.041	0.824±0.021
Femoral Neck	0.685±0.021	0.752±0.036	0.705±0.019
One-third Radius	0.588±0.014	0.722±0.030	0.629±0.017
TBSb	1.317±0.027	1.311±0.030	1.315±0.020

Data are expressed as mean ± standard error or absolute number (with percentage). Abbreviations: BMI =Body Mass Index, BMD = Bone Mineral Density, TBS = Trabecular Bone Score

^a.lower comorbidity index is better

^b.TBS ≥ 1.31 = normal bone microarchitecture, 1.24 to 1.31 = partially degraded microarchitecture, ≤ 1.23 = degraded bone microarchitecture

Table 2.

Twelve Months Post-Zoledronic Acid Changes in Bone Mineral Density and Trabecular Bone Score

Bone Index	Women				Men				All Participants
	Month 27 Mean±SE	Month 39 Mean±SE	Percent Change Mean±SE	p-Value	Month 27 Mean±SE	Month 39 Mean±SE	Percent Change Mean±SE	p-Value	Month 27 Mean±SE	Month 39 Mean±SE	Percent Change Mean±SE	p-Value
Lumbar Spine BMD (g/cm2)	1.137±0.033	1.190±0.036	0.97±0.80	0.242	1.306±0.077	1.289±0.098	−0.32±1.47	0.832	1.189±0.035	1.221±0.039	0.57±0.71	0.426
Total Hip BMD (g/cm2)	0.819±0.021	0.822±0.028	−0.10±1.03	0.927	0.940±0.044	0.981±0.055	1.79±1.08	0.139	0.856±0.022	0.871±0.029	0.49±0.79	0.545
Femoral Neck BMD (g/cm2)	0.707±0.021	0.717±0.024	1.71±0.98	0.099	0.775±0.036	0.786±0.050	1.52±2.17	0.505	0.727±0.019	0.738±0.023	1.65±0.93	0.087
One-third Radius BMD (g/cm2)*	0.602±0.015	0.605±0.018	−1.04±0.76	0.186	0.736±0.031	0.744±0.038	1.38±0.44	0.015	0.643±0.017	0.647±0.021	−0.30±0.58	0.613
TBS	1.349±0.024	1.423±0.021	3.91±1.25	0.007	1.353±0.033	1.376±0.028	3.33±1.08	0.054	1.350±0.019	1.409±0.019	3.79±1.01	0.001

Abbreviations: BMD = Bone Mineral Density, TBS = Trabecular Bone Density, SD = Standard Error, BL = Baseline

^*Finding significantly different between men and women (p=0.010)

Figure 2.

Percent change in Bone Mineral Density (BMD) ± standard error at various skeletal sites. The vertical dotted line marks month 27, which is the baseline for the extension study in which participants received one dose of zoledronic acid 5 mg. All participants received denosumb over the first 24months of the study.

DISCUSSION:

Discontinuation of denosumab without appropriate transition therapy is associated with rebound bone loss and an increased risk of vertebral fractures [6, 7]. This is a significant concern in managing osteoporosis, particularly in populations that are already at high risk for fractures, such as older residents of LTCCs. Our study addresses the critical gap in osteoporosis management by demonstrating that there is no evidence of a rebound decline after a single 5 mg dose of zoledronic acid following two years of denosumab treatment in LTCC residents aged 65 years and older. In women, zoledronic acid not only maintained the BMD gains achieved with denosumab but also unexpectedly enhanced bone microarchitecture, as evidenced by improvements in the TBS. In men, the treatment effectively preserved BMD, with a significant increase observed in the one-third radius BMD. Additionally, there was a trend toward improvement in the trabecular bone score, however, this was not statistically significant.

The efficacy of zoledronic acid in maintaining BMD following the discontinuation of denosumab appears to vary based on the duration of prior denosumab therapy. Our results are consistent with other short-term studies. For example, Anastasilakis et al. observed similar outcomes in a cohort of 27 community-dwelling postmenopausal women. They received denosumab for two years, followed by a single 5 mg infusion of zoledronic acid, which maintained BMD at the lumbar spine and femoral neck for at least one year [17]. Similarly, a study by Tsai et al. involving 53 postmenopausal women with osteoporosis treated with denosumab for one year showed that zoledronic acid maintained BMD across all skeletal sites [18]. In contrast to the findings of short-term studies, including ours, longer-term studies have demonstrated different outcomes. Anastasilakis et al. showed that patients treated with denosumab for more than four years experienced more pronounced rebound bone loss, with zoledronic acid being less effective at maintaining BMD compared to those with shorter treatment durations [19]. Similarly, in a study by Sølling et al., involving 61 men and postmenopausal women above 50 years treated with denosumab for a mean duration of 4.6 years, 5 mg of zoledronic acid did not fully prevent the loss of BMD in patients discontinuing denosumab [20].

In terms of TBS, our findings demonstrate a significant improvement in bone microarchitecture in women. While evidence on the effects of zoledronic acid on TBS remain limited, our results contrast with other studies, such as Sølling et al., which reported TBS stabilization rather than improvement following the transition to zoledronic acid [20]. A potential explanation for our findings may lie in the unique characteristics of our study population. Residents in LTCCs typically face distinct challenges, such as accelerated bone turnover and greater prevalence of comorbidities affecting bone density and microarchitecture. Such factors may create a more dynamic environment for bone remodeling, allowing for more pronounced TBS gains following effective antiresorptive treatment.

Our study has several strengths. It is a prospective study, and by focusing on residents of LTCCs, a group at high risk for fractures and often excluded from clinical trials, we started to address an important gap in long-term osteoporosis management for a vulnerable population. The inclusion of men, who are usually underrepresented in osteoporosis studies, further enhances the generalizability of our findings. Our study has limitations as well. The small sample size, partly due to challenges posed by COVID-19, limits our statistical sensitivity. Additionally, the absence of data on bone turnover markers prevents us from gaining a deeper understanding of the biochemical processes involved, which could offer valuable insights into how responses differ based on the duration of previous denosumab treatment. The long-term care population is heavily skewed towards women, and the lesser number of men results in lesser statistical sensitivity for eliciting findings in men. However, this limitation is also not unique to our study. Finally, due to the racial demographics of the study region, representation of minorities was limited as is the generalizability of findings.

In conclusion, our study demonstrates preliminary evidence that in residents of LTCCs with osteoporosis, a single 5 mg dose of zoledronic acid effectively maintains BMD gains and potentially enhances bone microarchitecture following two years of denosumab treatment. The findings suggest that zoledronic acid can be a valuable option for transitioning patients off denosumab, particularly in this high-risk population. However, further prospective studies are needed to explore the impact of zoledronic acid in individuals who have undergone longer-term treatment with denosumab. Additionally, more research is needed to explore the effects of zoledronic acid on TBS, as well as to understand how frailty and comorbid conditions may influence outcomes when transitioning therapies.