Denosumab biosimilar in postmenopausal osteoporotic women: A randomized, assessor-blind, active-controlled clinical trial.

OBJECTIVE: The study assessed the efficacy, safety, pharmacokinetic (PK), and immunogenicity profiles of denosumab-biosimilar and denosumab-reference in postmenopausal osteoporotic women from India.
MATERIALS AND METHODS: In this randomized, assessor-blind, active-control, multicenter trial, 114 patients were randomly allocated to receive denosumab-biosimilar (n = 58) or denosumab-reference (n = 56) at a subcutaneous dose of 60 mg every 6 months, for a year. Vitamin D and oral calcium were given daily. Lumbar spine bone mineral density (BMD) change was the primary end point.
RESULTS: Of 114 randomized patients, 111 (denosumab-biosimilar, n = 56; denosumab-reference, n = 55) completed the study. All 114 patients were part of safety and immunogenicity analyses, 110 (denosumab-biosimilar, n = 56; denosumab-reference, n = 54) were part of efficacy analysis, and 20 (denosumab-biosimilar, n = 10; denosumab-reference, n = 10) were part of PK analysis. The bone mineral density (BMD) (lumbar spine) percent change at 1 year with denosumab-biosimilar and denosumab-reference (7.22 vs. 7.62; difference:-0.40; 95% confidence interval: -5.92, 5.12) showed no statistically relevant difference. Likewise, alkaline phosphatase (bone-specific) and PK parameters also did not show statistically relevant differences. Adverse events were reported in 44.83% of patients on denosumab-biosimilar versus 33.93% of patients on denosumab-reference; most events were mild or moderate and not related to the study drugs. No patients showed anti-denosumab antibody positivity.
CONCLUSIONS: Denosumab-biosimilar and denosumab-reference showed biosimilarity in osteoporotic postmenopausal women. Availability of denosumab-biosimilar provides a treatment alternative for patients.

Introduction

Osteoporosis affects 200 million people globally[1] and 75 million people in the developed world.[2] It is estimated that there are nearly 46 million women aged more than 50 years with osteoporosis in India.[3] Thus, osteoporosis is a global health concern. An American clinical practice guideline recommends different bisphosphonates and denosumab as the first-line therapy for postmenopausal osteoporotic patients having high risk to fracture.[4] Other pharmacological agents used for osteoporosis include estrogen, ibandronate, raloxifene, calcitonin, and teriparatide.[4] All of these drugs are generally prescribed with oral supplementation of calcium and Vitamin D3.

The fully human monoclonal antibody denosumab binds to receptor activator of nuclear factor kappa-B ligand (RANKL) and prevents RANK receptor activation. Block in RANKL/RANK interaction inhibits formation, function, and survival of osteoclasts and thus decreases cortical and trabecular bone resorption.[5] The FREEDOM study (international, randomized, placebo-controlled trial) conducted in 7808 osteoporotic postmenopausal osteoporotic women showed that denosumab 60 mg decreased the risk of new radiographic vertebral, hip, and nonvertebral fracture by 68%, 40%, and 20%, respectively, at 36 months.[6] In the only published study in the Indian population, which was a double-blind, multicenter, placebo-controlled study in 225 Indian postmenopausal women, denosumab increased bone mineral density (BMD) and decreased the levels of markers of turnover of bone over 6 months.[7] In addition, denosumab increased BMD compared to bisphosphonates in osteoporotic postmenopausal women.[89]

Biosimilars are expected to match with biological reference products for their critical quality attributes.[1011] They are expected to be cheaper, thereby improving patient access. Intas Pharmaceutical Ltd., India, has developed a denosumab-biosimilar, a biosimilar to denosumab-reference (the reference denosumab product Prolia® [Amgen Inc., USA]). The biosimilarity of denosumab-biosimilar to the reference product was established through a stepwise approach of physicochemical and biological characterization, nonclinical, and clinical studies. This trial compared the efficacy, safety, pharmacokinetic (PK), and immunogenicity profiles of denosumab-biosimilar to denosumab-reference in osteoporotic postmenopausal Indian women.

Materials and Methods

Study participants

In this randomized, assessor-blind (investigators blinded to the efficacy data), active-controlled, multicenter, Phase 3 trial (clinical trial registry number: CTRI/2015/07/006065), patients were recruited from 14 centers across India from September 2015 to January 2017. The study protocol was reviewed and approved by the Drug Controller General of India and ethics committee. Patients provided written informed consent. The study followed Declaration of Helsinki, Guidance on Good Clinical Practice, and local regulatory specifications including Ethical Guidelines of the Indian Council of Medical Research.

Eligible patients were postmenopausal ambulatory women aged 80 years or less and T-score in the range of −2.5 to −4 in either the total hip or the lumbar spine. Major exclusion criteria were history of any metabolic bone disease, rheumatoid arthritis, malignancy, liver disease; current evidence of thyroid or parathyroid diseases; 25-(OH) Vitamin D level below 20 ng/mL; risk for osteonecrosis of jaw (risk factors associated with osteonecrosis excluded); administration history of bisphosphonates (intravenous), fluoride, strontium in the last 3 years, or oral bisphosphonates in the last 3 months; administration history of teriparatide, selective estrogen receptor modulators, tibolone, calcitonin, or drugs that can affect bone in last 6 weeks; or administration history of systemic hormone replacement therapy in the 4 weeks before randomization.

Study treatments

Patients who were concluded to be suitable for the study were randomized (1:1) to receive either denosumab-biosimilar (Intas, India) or denosumab-reference (Prolia, USA) at 60 mg by subcutaneous (SC) dose every 6 months (total two doses in a year). All patients additionally were provided oral supplementation of 400 IU of Vitamin D and 1000 mg of calcium daily.

Study endpoints

Lumbar spine BMD change in a year, compared between the treatment arms, was the primary end point. PK, immunogenicity, safety, and percent change from baseline to 3, 6, 9, and 12 months for alkaline phosphatase (BSAP) which was bone specific were the secondary end points.

BMD was measured using dual-energy X-ray absorptiometry scan at screening, 6 months, and 12 months at lumbar spine, neck of femur, trochanter, and total hip. BSAP was measured at baseline and at 3, 6, 9, and 12 months after the first dose. Personnel involved in the assessment of the efficacy endpoints were blinded to the treatment.

PK parameters of denosumab-biosimilar and denosumab-reference were assessed after the first dose. Blood was collected at predose and postdose at 12, 48, 72, 144, 192, 240, 288, 360, 528, 696, 1032, 1368, 2040, 2712, 3384, and 4032 h after the first dose to evaluate the PK profile.

Study suitability, safety, and tolerability evaluations included adverse events monitoring, physical examination, vital signs, laboratory tests (blood for biochemistry, blood for hematology, urinalysis, serology, pregnancy test, serum follicle-stimulating hormone, serum thyroid-stimulating hormone, serum calcium, serum parathyroid hormone, serum 25-OH Vitamin D3), electrocardiography, and X-ray of the lumbar spine and chest. Adverse event causality assessment was done using the World Health Organization–Uppsala Monitoring Centre causality categories.

Immunogenicity was evaluated by assessing serum for the presence of anti-denosumab antibody at baseline, 12 weeks, 24 weeks, and end of the study, i.e., 52 weeks. The serum samples were analyzed using an acid dissociation bridging ELISA method for the detection of anti-denosumab antibody.

Statistical analysis

Intent-to-treat (ITT) population included the patients who were given at least one dose of trial medication and had at least one efficacy assessment. Per-protocol (PP) population included patients without any major protocol violation affecting BMD. ITT and PP populations were used for the efficacy analysis. Safety analysis population included all patients who were given at least a dose of study medications.

One hundred and fourteen patients in a 1:1 randomization ratio were considered adequate, for a two-sample pooled t -test to prove existence of any difference. It was estimated to have two-sided significance level of 0.05, assuming 0.075 as pooled standard deviation (SD) for absolute BMD (lumbar spine). This gave 80% power for a difference detection between means of 0.81 (test) and 0.85 (control). Absolute BMD was used for sample size estimation considering the pilot design of this study.

For the primary efficacy end point, ANCOVA was performed considering baseline lumbar spine BMD as covariate, as well as absolute lumbar spine BMD and average percent BMD change at the lumbar spine as variables. P value, point estimate, and 95% confidence interval (CI) were estimated for all the efficacy end points. Wilcoxon rank-sum test and two-sample t -test were performed for the comparison of PK parameters.

Results

A total of 203 patients were screened, of which 114 were randomized and dosed (58 with denosumab-biosimilar and 56 with denosumab-reference). Of these, 111 (56 with denosumab-biosimilar and 55 with denosumab-reference) completed the study and three withdrew consent after dosing. All 114 were part of safety and ITT populations. Of 111 who completed the study, 110 (56 with denosumab-biosimilar and 54 with denosumab-reference) were included in PP; one recipient of denosumab was excluded due to a major protocol deviation (missed 59 doses of calcium tablets). Subject disposition is shown in Figure 1; site-wise subject disposition is shown in Table 1. Baseline characteristics showed comparable groups [Table 2].

Figure 1

Patient disposition. ITT = Intention-to-treat, PP = Per-protocol

Table 1

Site-wise subject disposition

Site number	Total patients screened	Screen failure	Safety population	ITT population	PP population	PK population
A	43	13	30	30	30	8
B	6	2	4	4	4	1
C	24	14	10	10	10	0
D	21	6	15	15	15	5
E	9	6	3	3	3	0
F	22	6	16	16	16	6
G	10	8	2	2	2	0
H	3	2	1	1	1	0
I	1	1	0	0	0	0
J	16	7	9	9	7	0
K	25	10	15	15	13	0
L	6	3	3	3	3	0
M	10	7	3	3	3	0
N	7	4	3	3	3	0
Total	203	89	114	114	110	20

ITT=Intent-to-treat, PP=Per-protocol, PK=Pharmacokinetic

Table 2

Baseline characteristics (safety/intent-to-treat population)

	Denosumab-biosimilar (n=58)	Denosumab-reference (n=56)
Age (years)	62.6±8.73	59.7±7.59
Height (cm)	149.7±6.25	152.9±5.97
Weight (kg)	57.8±12.07	58±10.81
Race (Asian), n (%)	58 (100)	56 (100)
BMD at lumbar spine (g/cm2 )	0.78±0.08	0.79±0.08
T-score for lumbar spine	−3.15±0.66	−3.06±0.68
Serum BSAP (µg/l)	15.70±6.98	14.56±5.88
Serum PTH (pg/ml)	45.8±21.5	41.7±15.3
Serum TSH (mIU/l)	2.6±1.4	2.9±2.2
Serum 25-OH vitamin D3 (ng/ml)	60±40.5	56.4±33.6

All values are presented as mean±SD, except indicated. BMD=Bone mineral density, PTH=Parathyroid hormone, TSH=Thyroid-stimulating hormone, BSAP=Bone-specific alkaline phosphatase, SD=Standard deviation

Efficacy

As shown in Table 3, the absolute change and percent change in BMD at the lumbar spine were matching between the denosumab-biosimilar and denosumab-reference for the PP and ITT populations. Change to 6 months was also similar between the denosumab-biosimilar and denosumab-reference for the PP population (mean difference: −1.61; 95% CI: −6.48, 3.26; P = 0.5135) and ITT population (mean difference: −1.55; 95% CI: −5.69, 5.01; P = 0.9003).

Table 3

Bone mineral density (lumbar spine) at 12 months

	PP population			ITT population
	Denosumab-biosimilar (n=56)		Denosumab-reference (n=54)	Denosumab-biosimilar (n=58)		Denosumab-reference (n=56)
Absolute BMD (g/cm2 )
Mean±SD	0.84±0.122		0.85±0.133	0.83±0.123		0.85±0.132
Mean difference		0.01			0.02
95% CI		−0.05-0.04			−0.04-0.04
P		0.86			0.87
Change in BMD from baseline to 12 months (%)
Mean±SD	7.22±12.72		7.62±16.33	6.98±12.57		7.32±16.12
Mean difference		−0.40			−0.34
95% CI		−5.92-5.12			−5.69-5.01
P		0.89			0.90

PP=Per-protocol, BMD=Bone mineral density, SD=Standard deviation, CI=Confidence interval, ITT=Intent-to-treat

The percent change in BMD at the left and right hip to 12 months was similar between the denosumab-biosimilar and denosumab-reference groups for the PP population (mean difference for left hip [n = 8]: 1.44; 95% CI: −6.72, 3.84; right hip [n = 6]: 3.32; 95% CI: −9.17, 2.53) and for the ITT population (mean difference for left hip [n = 17]: −0.08; 95% CI: −3.01, 2.85; right hip [n = 11]: −1.62; 95% CI: −5.66, 2.43).

No significant difference was noticed for the percent difference in BSAP between the denosumab-biosimilar and denosumab-reference groups [Figure 2]. The percent change in BSAP was − 31.38 with denosumab-biosimilar versus − 35.13 with denosumab-reference (mean difference: 3.75; 95% CI: −8.67, 16.17; P = 0.5505) in the PP population and − 14.54 versus − 35.56 (mean difference: 21.02; 95% CI: −13.92, 55.96; P = 0.2339) in the ITT population.

Figure 2

Change in BSAP from baseline over time denosumab-biosimilar, n = 58; denosumab-reference, n = 56; BSAP = Bone-specific alkaline phosphatase

Pharmacokinetics

PK analysis was done in 20 patients (10 in each treatment groups); these patients were selected randomly at the study start from four identified centers. Denosumab serum concentration after 60 mg single SC injection was similar between denosumab-biosimilar and denosumab-reference [Figure 3]. Significant difference was not observed between denosumab-biosimilar and denosumab-reference for AUC0–t (mean ± SD: 5924825.48 ± 2261364.63 vs. 6647544.58 ± 3122017.90; P = 0.5607), Cmax (mean ± SD: 6.68 ± 2.13 vs. 7.72 ± 2.79 µg/ml, P = 0.3647), Tmax (median [range]: 218.29 [73.58–289.25] vs. 168.08 [48.13–240.83] h; P = 0.089), and t1/2 (mean ± SD: 560.04 ± 158.80 vs. 497.60 ± 61.56 h).

Figure 3

Serum denosumab concentration-time profiles (linear scale)

Safety

Sixty-four adverse events were reported in 45 (39.47%) patients during the study. Denosumab-biosimilar and denosumab-reference groups had similar adverse events incidence [Table 4]. Adverse events were mostly mild or moderate and mostly not related to the study drug. Asthenia, pyrexia, and arthralgia were the common three adverse events. There were no deaths or discontinuation due to adverse events. The only serious adverse event that was reported occurred in a denosumab-biosimilar recipient and was considered not related to the study drug; this patient went through a surgical procedure due to an injury and subsequently recovered. No patient was positive for anti-denosumab antibodies.

Table 4

Common adverse events (reported in ≥2 patients)

Events	Denosumab-biosimilar (n=58), n (%) e	Denosumab-reference (n=56), n (%) e
All	26 (44.83) 34	19 (33.93) 30
Asthenia	1 (1.72) 1	6 (10.71) 6
Pyrexia	3 (5.17) 3	5 (8.93) 6
Arthralgia	4 (6.90) 4	1 (1.79) 1
Hyperchlorhydria	3 (5.17) 3	2 (3.57) 2
Urinary tract infection	3 (5.17) 3	1 (1.79) 1
Cough	1 (1.72) 1	2 (3.57) 2
Back pain	2 (3.45) 2	1 (1.79) 1

n (%) e=Number of patients with adverse event (percentage of patients with adverse event) number of adverse events

Discussion

The efficacy, PK, and safety profiles of denosumab-biosimilar were compared with that of denosumab-reference in Indian osteoporotic postmenopausal women. The study found no significant differences in the efficacy (BMD and BSAP), PK, and safety profiles of denosumab-biosimilar as compared to that of denosumab-reference.

The percent change in BMD (lumbar spine) was 7.22 with denosumab-biosimilar and 7.62 with denosumab-reference for PP population; these values are similar to the values (BMD change: approximately 6%–7%) reported in previous studies with denosumab in osteoporotic postmenopausal women.[612] The BMD at 6 months was 5.85 with denosumab-biosimilar and 7.40 with denosumab-reference; these values are also similar to the values (BMD change: approximately 4–6) reported in previous studies.[6712] The percentage change in BSAP at 3, 6, 9, and 12 months ranged from − 30 to − 60 in both the treatment groups. Previous studies with denosumab 60 mg report similar values, ranging from − 40 to − 60.[1213]

After single SC dose of denosumab-biosimilar or denosumab-reference, Cmax of 6–7 µg/ml was noted in 7–9 days, with a half-life of 21–23 days. After Cmax, serum levels decreased in a period of 5 months. These values are similar to the values reported in the denosumab summary of product characteristics (Cmax: 6 µg/ml; Tmax: 10 days; t1/2: 26 days; elimination time: 3 months).[5]

The most common adverse events (≥1/10 patient) reported in the literature with denosumab are extremity pain and musculoskeletal pain.[5] Musculoskeletal adverse events (asthenia and arthralgia) were common in the present study too, with an incidence rate of 15.52% for denosumab-biosimilar and 5.36% for denosumab-reference. Apart from musculoskeletal events, the adverse events of significance for denosumab include hypocalcemia, skin infection, jaw osteonecrosis, femur atypical fractures, diverticulitis, and drug-related hypersensitivity reactions.[5] None of these events occurred in the present study. Moreover, no patient developed anti-denosumab antibodies during the study. Previous studies reported the anti-drug antibody incidence rate to be <1% with 5-year treatment with denosumab, and there was no evidence of PK differences, toxicity, or difference in clinical response.[5]

Overall, results of the study are in line with the findings of previous denosumab (Prolia) studies. Moreover, our study demonstrates similarity of denosumab-biosimilar with denosumab-reference for efficacy, PK, and safety in osteoporotic postmenopausal women. Many patients having osteoporotic fractures do not receive treatment,[14] which indicates that healthcare researchers and providers need to provide more attention to this commonly observed clinical condition.

Small sample size (n = 114) is a limitation of this study besides relatively short duration of the study (1 year) and unblinded safety assessment. A larger sample size can potentially detect clinically meaningful differences between the test and reference products in a more precise manner. A longer treatment duration and follow-up are essential to identify adverse events such as osteonecrosis of the jaw and hypersensitivity reaction. Blinded safety assessment would have avoided any possible bias that may have occurred in this study. Notably, a Phase 4 study with denosumab-biosimilar has been initiated in osteoporotic postmenopausal women with 5-year treatment duration, which will provide long-term safety data in the Indian population.

Conclusions

In this study, BMD of the lumbar spine improved with denosumab. The efficacy of denosumab-biosimilar is similar to that of denosumab-reference. Moreover, the percent change from baseline in reduction in BSAP was also similar between the treatment groups at 3, 6, 9, and 12 months. Single-dose PK parameters of denosumab-biosimilar and denosumab-reference were similar. Both study drugs were well tolerated, with no clinically significant difference between denosumab-biosimilar and denosumab-reference.

Financial support and sponsorship

Intas Pharmaceuticals Ltd., India, funded this study.

Conflicts of interest

There are no conflicts of interest.