Osteoporosis: denosumab improves BMD in patients previously treated with bisphosphonate

Takeaway

• Treatment with denosumab improves bone mineral density (BMD) in patients previously treated with bisphosphonates and is comparable to zoledronate.
• The skeletal response in patients with chronic kidney disease (CKD) stage IV is significantly lower at the hip and is mainly associated with parathyroid hormone (PTH) concentrations.

Why this matters

• The study findings suggest that elevated PTH concentrations may impair the response to denosumab at the hip sites in patients with CKD.

Study design

• Study of 134 patients (11 males, 123 females) who had started treatment with denosumab for osteoporosis following treatment with bisphosphonates during 2013-2017.
• Patients on denosumab were divided into 2 groups based on their estimated glomerular filtration rate (eGFR) when bisphosphonates were contra-indicated:
  • group A (n=105) with eGFR >35 mL/minute and
  • group B (n=24) with eGFR <35 mL/minute.
• 95 patients had previously been on oral and 28 on IV bisphosphonate.
• Outcome: changes in BMD.
• Funding: None disclosed.

Key results

• Overall, 127 (95%) patients had sustained ≥1 fragility fractures.
• Significant increase in BMD was observed following denosumab (mean [SEM] % change at):
  • lumbar spine (LS): 6% (0.62); P<.001,
  • total hip (TH): 2.3% (0.64); P<.001 and
  • femoral neck (FN): 1.9% (0.77); P<.001.
• Changes at the TH and FN were lower in Group B vs Group A (% change at):
  • TH: 2.9% (0.72) vs −0.84% (1.28); P=.015 and
  • FN: 2.76% (0.86) vs −1.47% (1.53); P=.025.
• Changes in BMD were significantly lower at the hip sites in patients on denosumab with GFR <35 mL/minute vs those on denosumab with GFR >35 mL/minute (P<.05).     
• % change in BMD at the FN and PTH were negatively associated in the group who transitioned to denosumab (r=−0.25; P=.013).
• BMD changes were not different at 12-18 months between patients on denosumab vs zoledronate (% change at):
  •  LS: 3.97% (0.85) vs 2.6% (0.5); P=.19 and
  •  TH: 0.97% (0.58) vs 0.92% (0.6); P=.95.

Limitations

• Retrospective design.
• Absence of treatment-naïve group as control.