Incidence and predictors of hypocalcemia in end-stage renal disease patients on denosumab therapy: A systematic review and meta-analysis

Abstract

BACKGROUND

Denosumab inhibits the receptor activator of nuclear factor kappa-ligand. It markedly increases bone mineral density and has been proven to reduce the risk of fractures. However, numerous adverse effects, notably hypocalcemia, are prevalent in patients with end-stage renal disease (ESRD).

AIM

To analyze the incidence and predictors of hypocalcemia caused by denosumab compared to control in patients with ESRD.

METHODS

We conducted this study in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. PubMed, Scopus, Cochrane Central, and EMBASE were systematically searched from inception through March 2024. All original studies investigating the effects of denosumab on patients with ESRD compared to control were extracted. The primary outcomes of our study were the incidence of mild, severe, and very severe hypocalcemia. Secondary outcomes included serum levels of intact parathyroid hormone, calcium, and phosphate. The results were pooled and analyzed using a random-effects model.

RESULTS

Seven articles comprising 3240 patients were included in our study. Patients treated with denosumab had a significantly increased incidence of mild hypocalcemia [risk ratio (RR): 2.79; 95% confidence interval (CI): 0.99-7.91; P = 0.05; I² = 37%] and of very severe hypocalcemia (RR: 9.58; 95%CI: 1.58-57.98; P = 0.01; I² = 49%). However, an increase in the occurrence of severe hypocalcemia was non-significant (RR: 4.23; 95%CI: 0.47-38.34; P = 0.20; I² = 96%). Alternatively, denosumab showed a significant decrease in serum intact parathyroid hormone [mean difference (MD): -433.20, 95%CI: -775.12 to -91.28, I² = 98%, P= 0.01], while there was a non-significant decrease in phosphate (MD: -0.47, 95%CI: -1.35 to 0.41, I² = 88%, P = 0.30) and calcium levels (MD: -0.33, 95%CI: -0.95 to 0.29, I² = 94%, P = 0.29).

CONCLUSION

Our study demonstrated that denosumab is significantly associated with mild and very severe hypocalcemia in patients with ESRD making it necessary to detect and prevent this side effect of treatment.

Key Words: Denosumab; End-stage renal disease; Hypocalcemia; Parathyroid hormone; Dialysis

Core Tip: Denosumab is a monoclonal antibody that inhibits receptor activator of nuclear factor kappa-Β-ligand and prevents bone resorption, making it useful for treating osteoporosis. Although denosumab has an acceptable safety profile, it is known to cause hypocalcemia in patients with renal insufficiency making it contraindicated in patients with hypocalcemia. Such patients may experience muscle and joint pain. To date, very few randomized controlled trials have explored the safety of this drug in patients with end-stage renal disease or on dialysis. This meta-analysis thus appraised the safety profile of denosumab in patients with end-stage renal disease.

INTRODUCTION

Osteoporosis is characterized by brittle bones, which are susceptible to fracture owing to low bone mineral density. This condition is more common in older adults, especially postmenopausal women, and develops progressively over the years[1]. Genetic factors, hormonal fluctuations, certain medications such as chronic corticosteroid use, low levels of calcium and vitamin D, and a sedentary lifestyle are contributory factors[1-3].

End-stage renal disease (ESRD), the advanced form of chronic kidney disease (CKD), is also associated with an increased risk of fragility fractures. ESRD is indicated by an estimated glomerular filtration rate (eGFR) of less than 15 mL/min[4-6]. Osteoporosis is considered a significant complication of CKD[4,5]. Abnormalities in mineral metabolism emerge from impaired kidney function in ESRD, resulting in elevated phosphate and decreased calcium levels[6].

In order to mitigate osteoporosis secondary to ESRD, denosumab, an inhibitor of receptor activator of nuclear factor kappa-B-ligand, is useful in increasing bone mineral density and reducing the risk of fractures in patients with ESRD[7]. Changes to calcium and phosphate levels in these patients can be hindered by preexisting abnormalities in mineral metabolism as a complication of ESRD. This can be further explained by the high incidence of hypocalcemia in patients receiving denosumab for ESRD and CKD treatment[8]. Hypocalcemia can precipitate symptoms including muscle spasms, peripheral numbness, disorientation, forgetfulness, swallowing difficulties, irregular heart rhythm, brittle nails, hair thinning, and decreased bone density leading to osteoporosis[9-11].

The use of denosumab in ESRD patients and its adverse effects such as hypocalcemia have not been established in the existing literature with conflicting evidence being reported by published studies. Therefore, we conducted this meta-analysis to evaluate the safety of denosumab treatment and the associated risk for hypocalcemia in patients with ESRD with recently published large-scale studies to aid clinicians in prescribing denosumab to patients with ESRD.

MATERIALS AND METHODS

This review was conducted in concordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses[12].

Search strategy

An extensive search of PubMed, Scopus, EMBASE, and Cochrane Library was conducted from inception through March 2024. There were no limitations regarding language, location, author, year of publication, or any other aspect. The terms applied for the search were: (1) Denosumab; (2) Hypocalcemia; (3) Dialysis; (4) End stage renal disease; and (5) ESRD. A more detailed search strategy is given in Supplementary Table 1. Moreover, to identify grey literature, ClinicalTrials.gov, Medrxiv.org, and Google Scholar were searched. Additionally, we carried out a comprehensive review of the references cited in our study to acquire relevant articles.

Study selection

Two independent investigators (Fahim MAA and Salman A) screened the titles and abstracts of the articles that remained after duplicate removal through the EndNote Reference library (Version X7.5; Clarivate Analytics, Philadelphia, PA, United States), with a full-text review conducted to determine their relevance. After this, a third reviewer (Siddiqui AH) was sought to resolve any discrepancies. Additionally, the reference lists of included articles were manually examined to identify relevant studies.

Inclusion and exclusion criteria

The eligibility criteria for studies included were: (1) Population: Adults > 18 years of age with ESRD (eGFR < 15 mL/min) or on dialysis; (2) Intervention: Denosumab; (3) Control: Any; and (4) Outcomes: Any of our primary or secondary outcomes. Conference abstracts, letters, case reports, and studies containing inadequate original data for further analysis comprised our exclusion criteria.

Study outcomes

Our primary outcomes were incidence of mild (< 8.0 mg/dL), severe (< 7.5 mg/dL), and very severe (< 6.5 mg/dL) hypocalcemia. Our secondary outcomes consisted of changes in calcium level (mg/dL), changes in phosphate level (mg/dL), and changes in intact parathyroid hormone (iPTH) levels (pg/mL).

Data synthesis and analysis

Published data was analyzed using RevMan (version 5.4.1; Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2020), and the results were pooled as risk ratio (RR) or mean differences (MD) with their respective 95% confidence interval (CI). Comprehensive Meta Analyst (version 3.7) was used for all meta-regression analyses. All the outcomes were analyzed using the random effects model owing to the diversity of articles included. The significance of the results obtained was authenticated via a forest plot. A P value lower than 0.05 was considered significant. Heterogeneity across the pooled studies was examined using the Higgins I² statistics. A value of I² = 25%-50% was considered mild, I² = 50%-75% was moderate, and I² > 75% was significant heterogeneity. A sensitivity analysis was conducted for all results that presented significant heterogeneity.

Data extraction

Study and baseline characteristics were extracted onto an Excel sheet and verified by two authors (Fahim MAA and Salman A) with a third author (Siddiqui AH) being consulted on disagreement. Data from all studies included was extracted and organized according to author name, year of publication, study design, population type and size, denosumab dosage and route of administration, control drug and dosage, outcomes of each study, follow-up duration, general patient characteristics (age, sex, and body mass index), diabetes mellitus, baseline serum calcium (mg/dL), phosphate (mg/dL), 25-hydroxyvitamin D (ng/dL), iPTH (pg/mL), and alkaline phosphatase (IU/L) values, use of vitamin D analogues, and primary and secondary endpoints. Additionally, data was interconverted from medians to means and from various other units to the aforementioned units.

Quality assessment

The Cochrane Collaboration Risk of Bias Tool 2.0 was used to evaluate the quality of the included randomized clinical trials[13]. The Cochrane Collaboration Risk of Bias Tool 1.0 was used for the evaluation of non-randomized open-label clinical trials[14]. The Newcastle Ottawa Scale was used to assess bias in the included retrospective observational cohort studies[15] with the Newcastle Ottawa Scale assessment for case-control studies being used for the quality assessment of case controls[15]. Lastly, the National Institute of Health tool was used to assess the quality of the case series[16].

RESULTS

The search initially retrieved 202 records from databases from which seven studies involving 1793 patients with ESRD or on dialysis and osteoporosis being treated with denosumab were finally included. A more detailed overview of the process is presented in Figure 1. These patients were compared with control groups constituting a number of 1447 patients. Among these studies, two were observational retrospective cohort studies[17,18], one was a case series[19], one was a case-control study[20], two were open-label clinical trial studies[21,22], and one was a randomized controlled trial[23]. The characteristics of the included studies are summarized in Table 1 with baseline characteristics included in Table 2.

Figure 1 Preferred Reporting Items for Systematic Review and Meta-analyses flowchart. LTE: Letter to the Editor.

Table 1 Study characteristics.

Ref.	Study design	Study sample	Sample size		Denosumab route of administration	Denosumab dosage	Control drug and dosage	Outcomes	Follow-up duration
			Denosumab	Control
Bird et al[17], 2024	Observational retrospective cohort study	Dialysis-dependent Medicare postmenopausal female patients aged ≥ 65 years	1523	1281	Subcutaneous	60 mg (6 monthly)	Oral bisphosphonate (alendronate, risedronate, and ibandronate) 70 mg (weekly)	Severe hypocalcemia (albumin-corrected serum calcium levels < 1.88 mmol/L, primary hospital hypocalcemia diagnosis, or emergency department hypocalcemia diagnosis); Very severe hypocalcemia (serum calcium levels < 1.63 mmol/L or emergency care)	12 wk
Cowan et al[18], 2023	Observational retrospective cohort study	Adults aged > 65 years with a new drug prescription for oral bisphosphonate or denosumab on chronic dialysis or eGFR < 15 mL/min/1.73 m2	174	85	Subcutaneous	60 mg (6 monthly)	Oral bisphosphonate (etidronate, alendronate, and risedronate)	Mild hypocalcemia (albumin-corrected serum calcium or levels < 2.0 mmol/L or ionized calcium levels < 1.0 mmol/L) within 180 d of new oral bisphosphonate/denosumab prescription; Severe hypocalcemia (albumin-corrected serum calcium or levels < 1.8 mmol/L or ionized calcium levels < 0.9 mmol/L) within 180 d of new oral bisphosphonate/denosumab prescription	180 d
Chen et al[19], 2020	Observational single-center case series	Patients with ESRD with SHPT and low bone mass undergoing dialysis	21	21	Subcutaneous	60 mg (single dose)	Conventional treatment	Changes in calcium, phosphate, and ALP levels; Changes in CAC from baseline; Adverse events	6 mo
Iseri et al[23], 2019	Randomized controlled trial	Patients diagnosed with osteoporosis undergoing hemodialysis	22	24	Subcutaneous	60 mg (6 monthly)	IV alendronate 900 mg (every 4 wk for 1 year)	Changes from baseline to 12 mo in LSBMD, BMD at other sites; Changes from pretreatment to 12 mo in BTM; Changes in serum calcium and phosphate levels from day 0 to day 14; New fractures and adverse events	12 mo
Takami et al[20], 2017	Observational retrospective case-control study	Patients with low BMD (< 70% of the young adult mean) undergoing hemodialysis	17	20	Subcutaneous	60 mg (6 monthly)	No denosumab	Changes in phosphorus, calcium, whole PTH, total ALP, and albumin levels from baseline to 12 mo; Radius BMD	12 mo
Chen et al[21], 2015	Open-label clinical trial	Patients with SHPT and low bone mass undergoing dialysis	24	8	Subcutaneous	60 mg (single dose)	No denosumab	Parathyroid gland volume; BMD; Adverse outcomes; Changes in serum calcium, phosphate, ALP, and iPTH from baseline to study completion	24 wk
Chen et al[22], 2014	Open-label clinical trial	Patients with ESRD and severe SHPT undergoing dialysis	12	8	Subcutaneous	60 mg (single dose)	No denosumab	Changes in serum calcium, phosphorus, ALP, BMD, and iPTH levels	6 mo

ALP: Alkaline phosphatase; BMD: Bone mineral density; BTM: Bone turnover marker; CAC: Coronary artery calcification; eGFR: Estimated glomerular filtration rate; ESRD: End-stage renal disease; iPTH: Intact parathyroid hormone; IV: Intravenous; LSBMD: Lumbar spine bone mineral density; PTH: Parathyroid hormone; SHPT: Secondary hyperparathyroidism.

Table 2 Baseline characteristics.

Characteristic	Subcategory	Bird et al[17], 2024	Cowan et al[18], 2023	Chen et al[19], 2020	Iseri et al[23], 2019	Takami et al[20], 2017	Chen et al[21], 2015	Chen et al[22], 2014
Sample size	Denosumab	1523	174	21	22	17	24	12
	Control	1281	85	21	24	20	8	8
Female/male	Denosumab	1523/0	141/33	18/3	9/13	0/17	19/5	5/7
	Control	1281/0	48/37	12/9	9/15	0/20	2/6	3/5
Age in yr	Denosumab	74.5 ± 6.6	79.2 ± 7.7	62.14 ± 2.50	71.3 ± 10.5	72.8 ± 9.5	58.38 ± 2.77	53.5 ± 3.8
	Control	73.8 ± 6.5	78.0 ± 7.3	54.76 ± 2.00	71.5 ± 9.3	71.2 ± 11.0	58.50 ± 2.38	N/A
Serum calcium in mg/dL	Denosumab	9.3 ± 0.5	9.48 ± 0.16	9.96 ± 0.20	9.2 ± 0.5	9.2 ± 0.5	10.08 ± 0.16	10.1 ± 0.4
	Control	9.2 ± 0.6	9.4 ± 0.19	9.73 ± 0.20	9.3 ± 0.3	9.0 ± 0.5	10.30 ± 0.37	10.1 ± 0.3
Serum phosphate in mg/dL	Denosumab	4.7 ± 1.0	14.353 ± 0.67	5.50 ± 0.32	4.9 ± 1.1	5.0 ± 1.3	5.68 ± 0.29	5.3 ± 0.3
	Control	4.7 ± 1.1	14.477 ± 0.65	5.69 ± 0.21	4.6 ± 1.2	4.8 ± 1.2	5.79 ± 0.34	6.0 ± 0.3
Serum 25(OH)D in ng/mL	Denosumab	N/A	29.22 ± 12.34	27.01 ± 2.29	22.5 ± 9.5	N/A	30.04 ± 3.08	N/A
	Control	N/A	31.18 ± 14.05	25.16 ± 2.54	18.0 ± 9.3	N/A	32.36 ± 5.68	N/A
iPTH in pg/mL	Denosumab	N/A	295 ± 319.9	1310.50 ± 108.40	127.5 ± 83.2	150.2 ± 137.02	1464.77 ± 93.17	1702.1 ± 181.9
	Control	N/A	345.5 ± 456.81	1044.74 ± 61.24	138.4 ± 88.6	165.33 ± 98.14	974.58 ± 53.76	1300.1 ± 132.1
ALP in IU/L	Denosumab	N/A	N/A	268.38 ± 30.43	244.6 ± 60.8	276 ± 129	331.67 ± 48.86	449.8 ± 94.2
	Control	N/A	N/A	119.86 ± 8.85	265.9 ± 145.0	270 ± 69	112.50 ± 12.37	330.1 ± 81.3
BMI in kg/m2, mean ± SD	Denosumab	N/A	N/A	N/A	20.5 ± 2.7	21.6 ± 2.3	53.77 ± 1.74	N/A
	Control	N/A	N/A	N/A	21.1 ± 5.3	20.8 ± 2.3	62.48 ± 3.27	N/A
Diabetes	Denosumab	N/A	111 (63.8)	3 (14.3)	9 (40.9)	9 (52.9)	N/A	N/A
	Control	N/A	40 (47.1)	2 (9.5)	11 (45.8)	11 (55.0)	N/A	N/A
Use of vitamin D analogue	Denosumab	946 (62.1)	54 (31.0)	N/A	20 (90.9)	N/A	N/A	12 (100)
	Control	850 (66.4)	23 (27.1)	N/A	19 (79.2)	N/A	N/A	8 (100)

Data are presented as n (%) or mean ± standard deviation. 25(OH)D: 25-hydroxyvitamin D; ALP: Alkaline phosphatase; BMI: Body mass index; iPTH: Intact parathyroid hormone; N/A: Not applicable.

Quality assessment

In the included studies all randomized and non-randomized controlled trials were revealed to have a low risk of bias. Additionally, all cohort and case-control studies included in the analysis were also revealed to have a low risk of bias as they achieved scores of eight. Moreover, all case series received a quality rating of ‘Good.’ A detailed summary of quality assessment is presented in Supplementary Figures 1 and 2, Supplementary Tables 2-4.

Primary outcomes

Six studies reported mild hypocalcemia in patients with ESRD post-denosumab administration. Overall, the analysis suggests that there is a statistically significant association and a higher likelihood of developing mild hypocalcemia after denosumab therapy when compared to the control group (RR: 2.79; 95%CI: 0.99-7.91; P = 0.05; I² = 37%). Six studies reported severe hypocalcemia in patients with ESRD post-denosumab administration. The analysis suggests that there is no significant difference in the occurrence of severe hypocalcemia between denosumab and control groups with high inter-study heterogeneity (RR: 4.23; 95%CI: 0.47-38.34; P = 0.20; I² = 96%). Three studies reported severe hypocalcemia in patients with ESRD post-denosumab administration. The analysis indicated that denosumab is associated with a significantly higher likelihood of very severe hypocalcemia compared to the control group. (RR: 9.58; 95%CI: 1.58- to 57.98; P = 0.01; I² = 49%) Forest plots for primary outcomes are given in Figure 2.

Figure 2 Forest plots for primary outcomes. A: Mild hypocalcemia; B: Severe hypocalcemia; C: Very severe hypocalcemia. CI: Confidence interval.

Secondary outcomes

Three studies were evaluated to assess the changes in serum calcium levels after denosumab treatment in patients with ESRD showing a non-significant decrease in serum calcium levels when compared to the control group. (MD: -0.33; 95%CI: -0.95 to 0.29; P = 0.29; I² = 94%). Three studies reported changes in iPTH levels. The pooled analysis indicated that the iPTH levels decreased significantly in patients after denosumab use, however there was high inter-study heterogeneity (MD: -433.20; 95%CI: -775.12 to -91.28; P = 0.01; I² = 98%). Three research studies documented alterations in phosphate levels throughout denosumab therapy. The overall effect indicated no significant difference in serum phosphate levels between denosumab and control groups with high inter-study heterogeneity (MD: -0.47; 95%CI: -1.35 to 0.41; P = 0.30; I² = 88%). Forest plots for secondary outcomes are given in Figure 3.

Figure 3 Forest plots for secondary outcomes. A: Change in serum calcium; B: Change in serum intact parathyroid hormone; C: Change in serum phosphate. CI: Confidence interval.

Sensitivity analysis

A sensitivity analysis was performed for outcomes with significant heterogeneity. In the severe hypocalcemia outcome and change in serum phosphate outcome, leave-one-out sensitivity analysis based on removing outliers resulted in a significant reduction in heterogeneity as shown in Supplementary Figure 3 (RR: 4.23; 95%CI: 0.47-38.34; I² = 96% to RR: 13.82; 95%CI: 5.98-31.93; I² = 16%) and Supplementary Figure 4 (MD: -0.47; 95%CI: -1.35 to 0.41; I² = 88% to MD: -0.09; 95%CI: -0.67 to 0.48; I² = 21%), respectively. In the case of change in serum calcium and change in serum iPTH outcomes, leave-one-out sensitivity analysis was able to show only a slight reduction in heterogeneity evident in Supplementary Figure 5 (MD: -0.33; 95%CI: -0.95 to 0.29; I² = 94% to MD: 0.00; 95%CI: -0.27 to 0.28; I² = 59%) and Supplementary Figure 6 (MD: -433.20; 95%CI: -775.12 to -91.28; I² = 98% to MD: -635.30; 95%CI: -765.98 to -504.62; I² = 85%), respectively.

Meta-regression

We assessed age, sex, serum calcium, serum phosphate, and iPTH as possible covariates for having an impact on the effect sizes for severe hypocalcemia. We found severe hypocalcemia to be significantly associated with serum phosphate (Coeff: -0.3634, P = 0.0000) only (Supplementary Table 5 and Supplementary Figures 7-12).

We also assessed age, sex, serum calcium, serum phosphate, iPTH, and alkaline phosphatase as possible covariates for having an impact on the effect sizes for mild hypocalcemia. We found mild hypocalcemia to be significantly associated with sex (Coeff: 0.0336, P = 0.0421), age (Coeff: -0.1003, P = 0.0290), and serum phosphate (Coeff: -0.1784, P = 0.0124) (Supplementary Table 6 and Supplementary Figures 13-19).

DISCUSSION

Osteoporosis is one of the common age-related challenges, being of considerable importance in postmenopausal females and other older adults. Denosumab is clinically implicated in reducing the fracture risk and increasing bone mineral density among patients with ESRD. However, the adverse effects of the drug are not clearly demonstrated. Therefore, we conducted this systematic review and meta-analysis, which showed a statistically significant association of denosumab use with mild hypocalcemia, very severe hypocalcemia, and decreased serum iPTH. However, there were no significant differences between the control group and ESRD group participants for severe hypocalcemia and changes in serum calcium and phosphate levels following denosumab administration. The meta-regression findings indicated that age, sex, and serum phosphate were potential covariates that significantly influenced the incidence of mild hypocalcemia in patients with ESRD on denosumab treatment. Serum phosphate was also found to be a significant covariate for the incidence of severe hypocalcemia.

Denosumab, a human monoclonal antibody, exerts its action by suppressing the maturation and function of osteoclasts, decreasing bone resorption. The inhibition of receptor activator of nuclear factor kappa-B-ligand and subsequent reduction in bone resorption leads to clinically effective improvement in bone mineral density and skeletal-related events[24]. Denosumab-induced hypocalcemia can be explained by a decline in calcium levels in the blood secondary to a reduction in bone resorption. Moreover, denosumab causes impairment in the renal production of calcitriol, resulting in the development of hypocalcemia[21,25,26]. While preventing osteoclast-mediated bone resorption, denosumab disrupts the PTH-driven maintenance of calcium levels in the blood, resulting in hypocalcemia. The simultaneous increase in PTH at non-skeletal sites lowers the serum phosphate levels[27]. Notably, among patients with late-stage CKD, the bones become the primary source of serum calcium levels. This is further supported by the notion that patients with late-stage CKD on dialysis and a regular diet with calcium supplements exhibit calcium efflux from the skeletal system[28].

In our focused population comprising patients with advanced CKD, iPTH can be utilized as a useful bone turnover biomarker at the population level. Extremely high concentrations of iPTH are indicative of high-turnover bone pathology, very low levels of iPTH represent low-turnover bone disease[29]. In this study, denosumab administration was significantly associated with a decrease in iPTH levels. A retrospective analysis conducted by Tsvetov et al[30] concluded that pretreatment creatinine and albumin-adjusted serum calcium levels were the strongest predictors of denosumab-induced hypocalcemia in postmenopausal females with osteoporosis. The rise in the rates of hypocalcemia was found to be parallel to the decline in eGFR. Therefore, serum calcium monitoring for early identification of severe hypocalcemia is recommended in high-risk individuals. The retrospective multivariate analyses suggest CKD and baseline hypocalcemia as risk factors for the development of denosumab-induced hypocalcemia[31]. However, Saito et al[32] reported that renal dysfunction did not play a significant role in the risk of denosumab-induced hypocalcemia.

Our study demonstrated a significant association between the incidence of denosumab-induced mild hypocalcemia and age, sex, and serum phosphate. Serum phosphate was also found to be a significant covariate in the development of severe hypocalcemia in patients with ESRD treated with denosumab. Several studies have investigated the role and statistical significance of different risk factors in the development of denosumab-induced hypocalcemia. Sex does not play a predictive role in the incidence of denosumab-induced hypocalcemia[30,33]. However, denosumab-treated males are found to be significantly older compared to denosumab-treated females, with the former having more advanced disease and lower eGFR[33]. Conversely, Spångeus et al[34] identified males, severe renal failure, and pretreatment denosumab as important predictive factors. While diabetes is not explicitly investigated in existing studies for its correlation with the risk of denosumab-induced hypocalcemia in patients with ESRD, higher rates of baseline diabetes are observed in patients with lower eGFR[18].

While we observed a statistically significant association between serum phosphate levels as a potential covariate in mild and severe hypocalcemia, the sensitivity analysis revealed significant heterogeneity for changes in serum phosphate levels and severe hypocalcemia. A leave-one-out sensitivity analysis demonstrated a significant reduction in heterogeneity. However, only a slight reduction was observed when leave-one-out sensitivity analysis was performed for changes in serum calcium levels and iPTH. Despite recruiting a very large cohort of patients, Cowan et al[18] failed to undertake group matching, resulting in statistically significant differences in the baseline variables of the denosumab and control group participants. Serum phosphate heterogeneity in the associations of denosumab use and severe hypocalcemia prompted the removal of the study from the meta-analysis in order to yield more valid findings[35]. Dietary factors, age, genetics, sex, and season all tend to have a significant influence on serum phosphate levels[36]. Phosphate control is inherently impaired in patients with ESRD without any remaining renal function, stimulating hyperphosphatemia. However, other patient-related factors should be taken into account[37]. The existence of age and sex differences in the levels of serum and phosphate levels also provides a valid reason for the studies to utilize matched control and intervention group data in order to yield reliable findings[38].

In this study, we conducted a comprehensive systematic review and meta-analysis of incidence and predictors of hypocalcemia in patients with ESRD on denosumab therapy. While our study was able to identify clinically relevant findings and delineate age, sex, and serum phosphate as predictors of mild hypocalcemia associated with denosumab use in patients with ESRD, there are a few limitations. The high inter-study heterogeneity observed in the analysis can be attributed to methodological variation across the studies, predominantly the differences in the study design. The retrospective cohort by Cowan et al[18], which was identified as the source of heterogeneity as per the sensitivity analysis, did not account for the significant baseline differences between the denosumab and bisphosphonate groups. Due to the small numbers of patients with low eGFR and significant baseline differences between those prescribed denosumab and bisphosphonates, the study did not employ matching or weighing techniques to balance the groups. Due to incomplete population coverage, as the study excluded approximately 39% of the Ontario population due to their residence outside of the hospital, it led to a loss of 142000 patients, possibly introducing bias. Moreover, a significant proportion of new denosumab users (approximately two-thirds) did not have calcium levels monitored post-initiation. This lack of comprehensive monitoring may have resulted in underreporting of hypocalcemia cases, especially in those who develop symptoms. The study also could not account for over-the-counter calcium and vitamin D3 use, which may impact the risk of hypocalcemia. Moreover, the lack of control groups and the potential for selection bias in the included case series may also account for the high inter-study heterogeneity[20]. The cohort study by Takami et al[20] was also identified as a source of heterogeneity in our paper and had the following limitations: It had a small sample size and lacked a randomized controlled group, which limits its ability to generalize findings and assess the safety and efficacy of denosumab comprehensively. The lack of bone mineral density measurements at multiple sites (e.g., lumbar spine, total hip, femoral neck) in the referenced study meant that our analysis was limited to the distal third of the radius. This site-specific focus may have contributed to variability in understanding the broader effects of denosumab on bone health.

Our study findings may contribute to the improvement of the management of patients with ESRD undergoing denosumab therapy. The identification of the incidence and the predictive factors of hypocalcemia defines several future implications across the different key areas. One of these is the development of risk stratification tools, which facilitate the recognition of high-risk patients with ESRD who have a greater likelihood of developing hypocalcemia. Statistically significant predictors such as age, baseline osteoporosis, and baseline diabetes mellitus can inform strategies to optimize denosumab therapy for patients with ESRD and prevent the development of hypocalcemia post-denosumab treatment. The study findings may also instigate the development of clinical trial protocols, which are more focused on denosumab treatment in patients with ESRD. In addition to these prospects, the research findings are suggestive of patient monitoring and improving patient care by minimizing the number and severity of adverse effects.

CONCLUSION

Denosumab, used as a treatment for ESRD, can lead to either mild, moderate, or severe hypocalcemia post-administration. The results show a non-significant decrease in serum calcium and phosphate levels but a statistically significant decrease in serum PTH levels with denosumab administration. Therefore, it is important to provide integrated care, calcium and vitamin D supplementation, and CKD-mineral bone disorder optimization. Serum calcium monitoring is recommended for early detection of severe hypocalcemia in high-risk patients.