HORIZON-PFT

Published

Black DM, et al. "Once-Yearly Zoledronic Acid for Treatment of Postmenopausal Osteoporosis". The New England Journal of Medicine. 2007. 356(18):1809-22.
PubMed • Full text • PDF

Clinical Question

Among postmenopausal women with osteoporosis, does annual zoledronate infusion reduce the rate of vertebral, hip, and other fractures compared to placebo?

Bottom Line

When compared to placebo, annual zoledronate significantly reduces the rate of vertebral, hip, and other fractures among postmenopausal women with osteoporosis.

Major Points

Osteoporosis affects 7% of those age 50-59, 12% of those age 60-69, 26% of those age 70-79, and 35% of women age ≥80 years. Osteoporotic fractures are a major cause of disability for women and contribute $14 billion in annual costs in the US. The time following a fracture is associated with excess risk in mortality. Bisphosphonates inhibit bone reabsorption and increase bone mineral density (BMD), and the oral bisphosphonates alendronate and risendronate reduce the rate of fractures among individuals with osteoporosis. Adherence to these daily or weekly oral regimens can be poor, and a less frequent dosing schedule was desirable.

The HORIZON-PFT study randomly assigned nearly 4,000 older postmenopausal women with osteoporosis to either zoledronate 5 mg IV once annually, or to placebo, for 2 years. All patients received calcium and vitamin D. Patients were stratified according to baseline receipt of osteoporosis therapy. Primary endpoints included new vertebral fracture and hip fracture through study month 36. During the 3-year study period, zoledronate was associated with a 70% reduction in risk of vertebral fracture (3.3% zoledronate vs. 10.9% placebo) and hip fracture by 41% (1.4% zoledronate vs. 2.5% placebo). Safety signals included acute kidney injury (creatinine increase of >0.5 mg/dL) which was more common with zoledronate than placebo (1.3% vs. 0.4%) but generally transient, as well as atrial fibrillation which was more common with zoledronate than with placebo (50 vs. 17 patients); osteonecrosis of the jaw was seen in 2 patients, one in each group. The authors conclude that zoledronate was effective, well tolerated, and safe.

Guidelines

Endocrine Society Clinical Practice Guidelines (2020, adapted)

• In postmenopausal women at high risk of fracture, we recommend initial treatment with bisphosphonates (alendronate, risedronate, zoledronate, and ibandronate) to reduce fracture risk.
• In postmenopausal women with osteoporosis who are taking bisphosphonates, we recommend that fracture risk be reassessed after 3 to 5 years, and women who remain at high risk of fractures should continue therapy, whereas those who are at low to moderate risk of fractures should be considered for a bisphosphonate holiday.

Design

• Multicenter, randomized, double-blind, placebo-controlled trial
• N=7,765 postmenopausal women with osteoporosis
  • Zoledronic acid (n=3,889)
  • Placebo (n=3,876)
• Setting: 60 study centers internationally
• Enrollment: 2002-2003
• Follow-up: 3 years
• Analysis: Not specified, presumably intention-to-treat
• Primary outcomes:
  • New vertebral fracture among patients not taking concomitant osteoporosis medication
  • Hip fracture among all participants

Population

Inclusion Criteria

• Postmenopausal woman
• Age 65-89 years
• One of the following:
  • T score less than -2.5 at femoral neck, with or without vertebral fracture
  • T score less than -1.5 with ≥2 mild vertebral fractures or 1 moderate vertebral fracture on imaging

Exclusion Criteria

• Prior use of parathyroid hormone, sodium flouride, or strontium
• Anabolic steroids or growth hormone in prior 6 months
• Systemic corticosteroids in prior 12 months
• Ca >2.75 or <2.00 mmol/L
• CKD with CrCl <30 mL/min or 2+ protein on urinalysis

Baseline Characteristics

From the placebo group.

• Demographics: Median age 73 years, Europe 50%, North America or Oceania 20%, Latin America 16%, Asia 14%
• Body measurements: BMI 25 kg/m²
• T-score at femoral neck: Less than -2.5 71%, -2.5 to -1.5 28%, greater than -1.5 1%
• Bone mineral density (g/cm²): Femoral neck 0.53, total hip 0.65, lumbar spine 0.79
• Prior vertebral fracture: None 36%, 1 28%, ≥2 36%
• Prior meds: HRT 21%, bisphosphonates 14%, calcitonin 11%, SERMs 11%, other 1.1%
• Among the 21% of women in stratum 2, concomitant meds in >5%: Raloxifene 42%, calcitonin 17%, conjugated estrogens 12%, estradiol 8%

Interventions

• Participants were divided into a stratum:
  • Stratum 1 - Not taking an osteoporosis med at baseline
  • Stratum 2 - Taking an osteoporosis med at baseline
• Randomized to a group:
  • Zolendronate: 5 mg IV over 15 mins every 12 months (at day 0, 12 months, and 24 months)
  • Placebo
• All patients received oral calcium 1-1.5 g/d and vitamin D 400-1200 IU/d.

Outcomes

Comparisons are zoledronate vs. placebo. Outcomes are for all participants except where specified.

Primary Outcomes

New vertebral fracture

Among patients not taking concomitant osteoporosis medications (ie, stratum 1).

3.3% vs. 10.9% (RR 0.30; 95% CI 0.24-0.38; P<0.001)

Hip fracture

Among all patients randomized.

1.4% vs. 2.5% (HR 0.59; 95% CI 0.42-0.83; P=0.002)

Secondary Outcomes

Nonvertebral fracture

8.0% vs. 10.7% (HR 0.75; 95% CI 0.64-0.87; P<0.001)

Any clinical fracture

8.4% vs. 12.8% (HR 0.67; 95% CI 0.58-0.77; P<0.001)

Clinical vertebral fracture

0.5% vs. 2.6% (HR 0.23; 95% CI 0.14-0.37; P<0.001)

≥2 morphometric vertebral fractures

Among patients not taking concomitant osteoporosis meds (i.e., stratum 1) with a baseline radiograph and ≥1 follow-up xray, and a fracture.

0.2% vs. 2.3% (RR 0.11; 95% CI 0.05-0.23; P<0.001)

Subgroup Analysis

Not reported.

Adverse Events

Any

95.5% vs. 93.9% (P=0.002)

The authors note that the higher rate with the study medication was primarily driven by post-infusion symptoms.

Serious: 29.2% vs. 30.1% (P=0.40)

Death: 3.4% vs. 2.9% (P=0.27)

Leading to discontinuation: 2.1% vs. 1.8% (P=0.41)

Osteonecrosis of the jaw: None reported.

Most symptom within 3 days of infusion

Pyrexia: 16.1% vs. 2.1% (P<0.001)

Myalgia: 9.5% vs. 1.7% (P<0.001)

FLI: 7.8% vs. 1.6% (P<0.001)

Headache: 7.1% vs. 2.3% (P<0.001)

Arthralgia: 6.3% vs. 2.0% (P<0.001)

Renal

Creatinine increase >0.5 mg/dL: 1.2% vs. 0.4% (P=0.001)

The authors note that these increases were transient and there were no differences in renal function at 3 years.

Urinary protein >2+: 0.5% vs. 0.2% (P=0.06)

CrCl <30 mL/min: 4.1% vs. 3.9% (P=0.69)

CVD-related

Any AF: 2.4% vs. 1.9% (P=0.12)

Serious AF-related event: 1.3% vs. 0.5% (P<0.001)

Stroke: 2.3% vs. 2.3% (P=0.94)

Stroke death: 0.5% vs. 0.3% (P=0.15)

MI: 1.0% vs. 1.2% (P=0.44)

CVD mortality: 1.0% vs. 0.9% (P=0.55)

Criticisms

• Manuscript authored in part by representatives from the sponsor.

Funding

• Supported by Novartis.