Dr. Elias John Elenjickal is a Nephrology Research Fellow and PhD. student at the McGill University Health Centre in Montreal, Canada. He completed his medical degree and internal medicine residency at Goa Medical College, and his nephrology training at Dayanand Medical College, Ludhiana—both in India. He is currently on sabbatical from Christian Medical College (CMC), Vellore, the largest not-for-profit tertiary care institution in South India, where he has served as both a clinician and clinical researcher. Dr. Elenjickal’s research focuses on improving access to evidence-based cardiovascular therapies for patients with advanced chronic kidney disease, a high-risk group frequently underrepresented in clinical trials. At McGill, he is actively involved in multiple clinical trials that aim to address this critical evidence gap. His long-term vision is to return to India to build a sustainable research ecosystem by mentoring early-career investigators, strengthening local research capacity, and leading collaborative clinical trials relevant to resource-constrained settings. Outside of medicine, Dr. Elenjickal draws strength and joy from his family, who remain his greatest source of support and inspiration. Dr. Elenjickal is a 2025-26 AJKD Editorial Intern.

 

Osteoporosis is a condition characterized by low bone mineral density (BMD), poor bone quality is associated with a high risk of fractures. The World Health Organization defines osteoporosis based on a BMD T score ≤ – 2.5, which corresponds to a BMD value of at least 2.5 standard deviations below the average value for healthy individuals aged 20-29 years of the same sex. Based on data from the third National Health and Nutrition Examination Survey (NHANES III), about 11% men and 24% women with osteoporosis have advanced chronic kidney disease (CKD). Patients with CKD have a 2.5-fold higher risk for fractures, which increases to 4-fold in patients receiving kidney replacement therapy (KRT). Recognizing that fracture risk in CKD is determined not only by reduced bone mineral density, but also by CKD-specific impairments in bone turnover, mineralization and microarchitecture, the 2025 KDIGO Controversies Conference introduced the concept of CKD-associated osteoporosis as a distinct form of osteoporosis encompassing both traditional osteoporosis and renal osteodystrophy.

Anti-resorption therapy remains the cornerstone treatment for most patients with normal- or high-turnover osteoporosis, whereas anabolic agents such as teriparatide may be more appropriate for patients with low-turnover disease.  Antiresorptive therapy primarily includes bisphosphonates (intravenous or oral) and receptor activator of nuclear factor kappa B ligand (RANKL) inhibitors (denosumab). Bisphosphonates have high affinity for the bone hydroxyapatite crystals. After binding, they remain in the bone for years, thus inhibiting osteoclast migration and maturation. All bisphosphonates are predominantly excreted unchanged by the kidneys. Their use in CKD is limited because the reduced renal clearance may cause increased skeletal accumulation and over suppression of bone turnover, thus increasing the risk of adynamic bone disease. In addition, bisphosphonates may cause acute kidney injury, particularly in the elderly and those with pre-existing CKD.  Thus, bisphosphonates are generally avoided in patients with advanced CKD (estimated glomerular filtration rate [eGFR] <30 ml/min/1.73 m²), though off label use continues mainly in the management of hypercalcemia.

In contrast, denosumab is a fully humanized monoclonal antibody with high affinity to the human RANKL. By preventing interaction between RANKL and RANK on the osteoclast surface, denosumab inhibits osteoclastic bone resorption. Unlike bisphosphonates, denosumab does not accumulate in the bone and is cleared by the reticuloendothelial system, with minimal renal elimination. These pharmacokinetic properties make denosumab an ideal alternative to bisphosphonates in patients with CKD (Table 1).

Table 1: Comparison of anti-resorption therapies (bisphosphonates vs. RANKL inhibitors)

Anti-resorption therapy	Bisphosphonates (Zoledronate, Alendronate, Risedronate, Ibandronate)	RANKL inhibitors (Denosumab)
Mechanism of action	Inhibition of osteoclast migration and maturation by binding to hydroxyapatite crystals.	Monoclonal antibody against RANKL which prevents RANKL-RANK interaction, thereby inhibiting osteoclast function and survival
Pharmacokinetics	High affinity for the bone tissue   Exposure linearly related to dose   Can be detected in plasma up to 28 days post-dose (for zoledronate).   Does not undergo hepatic metabolism with no active or inactive metabolites.   Undergoes renal elimination (39% of intact zoledronate detected in urine in 24 hours)   Terminal half-life- 146 hours (for zoledronate)	Does not incorporate into the bone   Follows nonlinear, dose-dependent pharmacokinetics   Can be detected in plasma up to 9 months post-dose   Cleared by the reticuloendothelial system with minimum renal excretion   Terminal half-life- 5-10 days
Dose and frequency of administration in osteoporosis	Zoledronate: 5 mg IV infusion, once a year Alendronate: 70 mg orally once weekly Risedronate: 35 mg orally once weekly Ibandronate: 3 mg IV infusion, every 3 months or 150 mg orally once monthly	60 mg SC injection once every 6 months
Adverse events	Renal impairment (AKI) with fast rate of infusion Hypocalcaemia Acquired Fanconi syndrome Musculoskeletal pain Osteonecrosis of the jaw Atypical femoral fractures	Hypocalcaemia (severe) Osteonecrosis of the jaw
Limiting factor in patients with advanced CKD	Renal impairment Adynamic bone disease	Hypocalcemia

AKI, acute kidney injury; IV, intravenous; RANK, receptor activator of nuclear factor kappa B; RANKL, receptor activator of nuclear factor kappa B ligand; SC, subcutaneous.

The pivotal FREEDOM trial (Fracture REduction Evaluation of Denosumab in Osteoporosis every 6 Months) demonstrated that denosumab as compared to placebo significantly reduces the risk of vertebral and non-vertebral fractures in postmenopausal women with osteoporosis. Although this trial did not show any signals of hypocalcemia risk, it is important to note that very few participants with advanced CKD were included (n=73). The risk of hypocalcemia with denosumab use in CKD is biologically plausible because denosumab inhibits osteoclast mediated bone resorption, thereby reducing release of calcium from bone to circulation. In individuals with normal kidney function, a fall in serum calcium level can be compensated by increased calcitriol production, enhanced intestinal calcium absorption, and increase in parathyroid hormone secretion. However, these adaptive responses are impaired in patients with advanced CKD. Progressive kidney dysfunction is associated with reduced calcitriol synthesis, which  is partly mediated by increased secretion of fibroblast growth factor-23 (FGF-23) by the osteocytes which inhibit 1-alpha hydroxylase, the rate limiting enzyme for calcitriol synthesis. The resulting decrease in intestinal calcium absorption leads to secondary hyperparathyroidism. However, there is skeletal resistance to parathyroid hormone seen with advancing CKD which limits the ability to release calcium from bones. Hence patients with progressive CKD are increasingly dependent on ongoing bone resorption to maintain their calcium balance, rendering them susceptible to clinically significant hypocalcemia with initiation of denosumab. Thus, the potent anti-resorptive property of denosumab may have a disproportionately greater impact on calcium homeostasis within a complex physiological milieu of advanced CKD.

Over the last decade, a number of observational studies (Kunizawa et al., Sun et al., Cowan et al.) began reporting higher events of hypocalcemia with denosumab treatment in patients with CKD. However, there has never been a head-to-head comparison of denosumab and bisphosphonates in a randomized controlled trial design. Using the Medicare database from United States, Bird et al. recently conducted a target trial emulation in osteoporotic  female patients 65 years or older, who were initiated on either denosumab, oral bisphosphonates, or intravenous bisphosphonates. The primary outcome was hospital or emergency department admissions for hypocalcemia assessed in the first 12 treatment-weeks. The authors reported a higher incidence of hypocalcemia events with denosumab use in patients with CKD. This risk increased with advancing CKD stages and was highest in patients on KRT. In addition, denosumab initiation in patients with stage 4 and 5 CKD was associated with higher risks of hypocalcemia in those with underlying CKD–mineral and bone disorder (CKD-MBD). The major limitation of this study though was that both the exposure (CKD stage) and outcome (hypocalcemia events) were ascertained based on diagnostic codes, without any access to clinical or laboratory data.

A recently published article in AJKD, “Risks of hypocalcemia and other bone mineral disorders for denosumab versus zoledronate across the spectrum of kidney function: A target trial emulation”, by Xiao et al.  leveraged data from the SCREAM (Stockholm CREAtinine Measurement) project, a healthcare utilization cohort of all residents in Stockholm, Sweden (2006-2021). This is a comprehensive database with complete information on demographics, laboratory tests, healthcare use and prescriptions at all Swedish pharmacies. Following a hypothetical target trial emulation framework, the authors emulated a target trial in patients initiating either denosumab 60 mg or zoledronate 5 mg between 2011-2021. The primary outcome was hypocalcemia, defined as the first occurrence of hospitalization for hypocalcemia or ≥2 outpatient diagnosis of hypocalcemia or serum calcium <2 mmol/L or ionized calcium <1 mmol/L. Secondary outcomes included severe hypocalcemia (defined as hospitalization for hypocalcemia or serum calcium <1.75 mmol/L or ionized calcium <0.9 mmol/L), secondary hyperparathyroidism, and hypophosphatemia. The follow-up period began on the day of first dispensation of zoledronate or denosumab and was censored at the occurrence of the outcome or death or emigration or 180 days, whichever came first. The maximum follow-up was set at 180 days to align with denosumab’s dosing interval which resulted in an identical intention-to-treat and per-protocol effect.

A total of 5085 new users of denosumab and 3599 new users of zoledronate were included in the study. Of these, 25% had CKD defined as eGFR <60 ml/min/1.73 m². As per Swedish guidelines, zoledronate is the recommended first line therapy for osteoporosis, preferred for its single annual administration which improves compliance. Denosumab is prescribed only in patients with intolerance or contraindications to bisphosphonates, such as CKD. Not surprisingly, denosumab users were older, had a lower eGFR, and had a higher prevalence of comorbidities and fractures. The authors then efficiently used inverse probability of treatment weights to adjust for a comprehensive 66 baseline covariates. After weighting, all covariates were well balanced.  Over the follow-up of 180 days, 103 patients experienced hypocalcemia, of which 37 events were considered severe. Most of these events occurred in the first 14 days. Compared to zoledronate, treatment with denosumab was associated with higher risks of hypocalcemia (hazard ratio [HR] of 2.10; 95% confidence interval [CI] of 1.28-4.33), absolute risk difference (RD) of 0.72% (95% CI, 0.26-1.18), hypophosphatemia [HR 1.96 (95% CI, 1.30-3.38), absolute RD 0.85% (95% CI, 0.33-1.38)] and secondary hyperparathyroidism [HR 2.85 (95% CI, 1.94-5.48), absolute RD 1.33% (95% CI, 0.91-1.85)]. This risk was further exaggerated in patients with CKD (Table 2). Sensitivity analysis using oral bisphosphonates as an active comparator or excluding patients with advanced CKD (eGFR < 30 ml/min/1.73 m²) or using quantitative bias analyses to account for ascertainment bias due to higher rates of laboratory monitoring in denosumab users showed consistent results.

Table 2: Target trial emulation studies comparing the risks of hypocalcemia with denosumab and bisphosphonates in patients with chronic kidney disease

Target trial emulation study	Bird et al.	Xiao et al.
Database	Medicare fee-for-service database, USA	SCREAM project, Sweden
Inclusion criteria	Women ≥ 65 years initiating first-line treatment for osteoporosis between 2012-2020	Men and women ≥ 50 years initiating denosumab vs. zoledronate between 2011-2021
Comparison arms	Denosumab vs. oral bisphosphonates vs. IV bisphosphonates	Denosumab vs. zoledronate
Primary outcome	Hospital or emergency department admissions for hypocalcemia	Hypocalcemia, defined as the first occurrence of hospitalization for hypocalcemia or ≥2 outpatient diagnosis of hypocalcemia or serum calcium <2 mmol/L or ionized calcium <1 mmol/L
Duration of follow-up	12 weeks	180 days
Analytic approach	ITT	ITT= per-protocol
Total number of patients	361453 new users of denosumab 829044 new users of oral bisphosphonates 160413 users of IV bisphosphonates	5085 new users of denosumab 3599 new users of zoledronate
Main findings	Cumulative risk for emergently treated hypocalcemia with denosumab was 13-fold higher than with oral bisphosphonates and 4.3-fold higher than with IV bisphosphonates   Risk of emergently treated hypocalcemia increased with worsening CKD stage and with presence of CKD-MBD in denosumab users as compared to those on bisphosphonates	Denosumab users had higher risks of hypocalcemia, hypophosphatemia and secondary hyperparathyroidism   Adverse events with denosumab were higher in patients with CKD as compared to those with normal kidney function
Strengths of the study	Large database with linkage to multiple datasets with information on covariates, hospitalization and drug prescription	Outcomes were defined based on diagnostic and laboratory data, thus improving the validity   Robust sensitivity analyses confirming the primary findings
Limitations	Exposure and outcome definitions of CKD stage, CKD-MBD status and hypocalcemia were based on codes, without any access to eGFR, laboratory or clinical data.   Only acute symptomatic hypocalcemia could be captured by this study	Missing laboratory data at baseline (multiple imputation analysis)   Ascertainment bias due to higher rates of serum calcium monitoring among denosumab users (quantitative bias analysis)   Fewer patients with CKD stage-4 (2%)

CKD-MBD, Chronic kidney disease-mineral bone disorder; eGFR, estimated glomerular filtration rate; ITT, intention to treat; IV, intravenous; SCREAM; Stockholm CREAtinine Measurement project; USA, United States of America

This study beautifully captures the intricacies of kidney-bone-physiology axis, demonstrating that despite its favourable pharmacokinetic profile in CKD, denosumab carries a higher risk of hypocalcemia. CKD is as much a disorder of physiology as it is of filtration and the impact of antiresorptive therapies must be interpreted within this complex and fragile  mineral homeostatic milieu. This study also further strengthens the KDIGO recommendations of close monitoring of calcium, phosphate, and parathyroid hormones after initiation of denosumab. In addition to laboratory monitoring, the authors emphasize clinical vigilance for symptoms like muscle cramps or paresthesia, especially in the first 2 weeks, when the risk appears to be the highest, with consideration for pre-emptive administration calcium and vitamin D supplementation. Not surprisingly, the FDA has now added a black box warning of increased risk of severe hypocalcemia with denosumab use in patients with advanced CKD.

Although randomized controlled trials are the gold standard for assessing the effect of therapeutic interventions, most studies are not primarily powered to detect adverse events. Target trial emulation, when conducted in a rigorous manner, enables causal interpretations in populations that are under-represented in clinical trials, such as advanced CKD. It allows us to perform head-to-head comparisons of available interventions such as denosumab vs. bisphosphonates, which is more clinically meaningful than comparisons to placebo, and also enables evaluation of rare outcomes of interest such as hypocalcemia. This study by Xiao et al. is the prefect application of a target trial design to answer a clinically relevant question.