What HORIZON-PFT Actually Changes in Clinical Practice

By
Published Updated Last reviewed
Clinical medical image for trials horizon pft: What HORIZON-PFT Actually Changes in Clinical Practice

At a glance

| Detail | Value | |---|---| | N | 7,765 postmenopausal women | | Intervention | Zoledronic acid 5 mg IV once yearly x 3 years | | Comparator | Placebo IV infusion | | Duration | 3 years | | Primary endpoint | New morphometric vertebral fracture | | Key result | 70% relative risk reduction (3.3% vs. 10.9%; p <0.001) | | Registration | NCT00049829 |

Why the Trial Was Needed

Before 2007, oral bisphosphonates (alendronate, risedronate) dominated osteoporosis prescribing. They worked, but real-world adherence was poor. Published pharmacy-claims analyses showed that roughly half of patients stopped oral bisphosphonate therapy within the first year. Weekly or monthly dosing, fasting requirements, and esophageal irritation drove discontinuation. Clinicians needed a regimen that removed those barriers without sacrificing fracture protection. HORIZON-PFT was designed to test whether a once-yearly IV zoledronic acid infusion could fill that gap.

Trial Design: What the Abstract Leaves Out

HORIZON-PFT randomized 7,765 postmenopausal women (ages 65 to 89) across 240 centers in 27 countries. Eligibility required either a femoral-neck T-score of <−2.5 with or without existing vertebral fracture, or a T-score of <−1.5 with radiologic evidence of at least two mild or one moderate vertebral fracture.

Several design choices matter for translating results to practice:

Calcium and vitamin D loading. All participants received calcium 1,000 to 1 to 500 mg/day plus vitamin D 400 to 1 to 200 IU/day. Patients with 25-hydroxyvitamin D levels below 15 ng/mL at screening received a loading dose of 50,000 to 125 to 000 IU of vitamin D2 or D3 before randomization. This pre-treatment vitamin D repletion is critical. Acute-phase reactions (fever, myalgia) after IV zoledronic acid are more common in vitamin D-deficient patients, and hypocalcemia risk rises without adequate supplementation. Clinicians who skip this step in practice are not replicating the trial protocol.

Concomitant therapy allowed. Patients already on hormone therapy, raloxifene, calcitonin, or tibolone could enroll as long as they had been on a stable dose for at least one year. Prior oral bisphosphonate use was permitted if it had been stopped at least six months before randomization. This makes the trial population more heterogeneous than a typical "treatment-naive" efficacy study.

Morphometric vertebral fracture as primary endpoint. Vertebral fractures were assessed by digitized lateral spine radiographs at baseline, 12, 24, and 36 months using the Genant semi-quantitative method. This approach captures fractures that are clinically silent (roughly two-thirds of vertebral fractures never present with acute symptoms). The clinical relevance is that morphometric vertebral fractures predict future hip and clinical fractures.

Results Beyond the Headline

The HealthRX Fracture-Reduction Breakdown

The 70% vertebral fracture reduction is the number most sources repeat. The full picture across fracture types tells a more useful story for prescribing decisions.

| Outcome | Zoledronic Acid | Placebo | Relative Risk Reduction | NNT (3 yr) | |---|---|---|---|---| | Morphometric vertebral fracture | 3.3% | 10.9% | 70% (p <0.001) | 13 | | Clinical vertebral fracture | 0.5% | 2.6% | 77% (p <0.001) | 48 | | Hip fracture | 1.4% | 2.5% | 41% (p = 0.002) | 91 | | Non-vertebral fracture | 8.0% | 10.7% | 25% (p = 0.001) | 37 | | Any clinical fracture | 8.4% | 12.8% | 33% (p <0.001) | 23 |

Source: calculated from data in Black et al., NEJM 2007.

The hip fracture NNT of 91 over three years is comparable to what oral alendronate showed in the FIT trial (NNT ~91 for hip fracture in the vertebral fracture arm). Zoledronic acid did not outperform oral bisphosphonates on fracture endpoints. It matched them while solving the adherence problem.

BMD Changes

Bone mineral density gains were consistent and dose-dependent over each infusion year:

  • Lumbar spine: +6.7% at 3 years vs. +0.7% in placebo
  • Total hip: +6.0% at 3 years vs. +1.0% in placebo
  • Femoral neck: +5.0% at 3 years vs. +0.3% in placebo

These BMD gains exceed what most oral bisphosphonate trials reported over comparable intervals. Whether larger BMD gains translate to proportionally greater fracture reduction remains debated. The FLEX extension study of alendronate and the subsequent HORIZON extension data both suggest a plateau effect, where BMD gains above a threshold provide diminishing fracture protection.

Bone Turnover Markers

Serum C-telopeptide (CTX) and bone-specific alkaline phosphatase (BSAP) dropped sharply within days of the first infusion and remained suppressed throughout. CTX reached its nadir at 9 to 11 days post-infusion and stayed below baseline until the next annual dose. This rapid, sustained suppression confirms that a single yearly infusion produces antiresorptive potency comparable to daily or weekly oral regimens.

Safety Profile: What Prescribers Need to Know

Acute-phase reaction. The most common adverse event was a flu-like syndrome after the first infusion: fever, myalgia, headache, and arthralgia affecting roughly 32% of zoledronic acid patients vs. 6.7% on placebo. By the second infusion, incidence dropped to 7%, and by the third, to 3%. Pre-treatment with acetaminophen reduces severity. This reaction is the leading reason patients cite for not wanting a repeat infusion, so counseling before the first dose matters.

Renal safety. Transient serum creatinine elevations (>0.5 mg/dL above baseline) occurred in 1.3% of zoledronic acid patients vs. 0.4% on placebo. All cases resolved within 30 days. The Reclast FDA label now contraindicates the drug in patients with creatinine clearance <35 mL/min and requires pre-infusion hydration. Post-marketing reports of acute renal failure, some requiring dialysis, led to an infusion-time minimum of 15 minutes (not the 5-minute pushes some early adopters attempted).

Atrial fibrillation. Serious atrial fibrillation occurred in 1.3% of zoledronic acid patients vs. 0.5% on placebo (p = 0.003). This signal prompted FDA review but was not replicated in the companion HORIZON-RFT trial or in pooled analyses of other bisphosphonate data. Current consensus from the American Association of Clinical Endocrinology (AACE) treats this as a possible but unconfirmed risk. Patients with pre-existing arrhythmias should be informed.

ONJ and atypical fractures. No confirmed cases of osteonecrosis of the jaw (ONJ) or atypical femoral fractures (AFF) occurred during the three-year trial period. Both complications emerged in longer-duration bisphosphonate exposure and are now recognized in the Reclast label. The three-year timeframe of HORIZON-PFT was likely too short and the population too small to detect these rare events at rates of roughly 1 per 10,000 to 100,000 patient-years.

What Changed in Guidelines After HORIZON-PFT

HORIZON-PFT, published in May 2007, directly influenced multiple society guideline updates.

AACE/ACE (2020 update). Zoledronic acid is listed as a first-line option for postmenopausal osteoporosis in patients at high fracture risk. The AACE guideline specifically recommends it when oral bisphosphonate adherence is a concern or when gastrointestinal contraindications exist.

ACP (2023). The American College of Physicians clinical practice guideline recommends bisphosphonates (including IV zoledronic acid) as initial pharmacologic treatment for primary osteoporosis in postmenopausal females and older males at high fracture risk.

NOGG (UK). The National Osteoporosis Guideline Group incorporated IV zoledronic acid as an alternative to oral agents after HORIZON-PFT, with specific mention of adherence advantages.

In practice, IV zoledronic acid occupies a clear niche: patients who have failed oral bisphosphonates due to GI intolerance, those with demonstrated non-adherence, and those who prefer annual dosing. It is also used as "bridge therapy" after discontinuing denosumab (Prolia), where a rapid-onset antiresorptive is needed to prevent rebound vertebral fractures.

Limitations the Authors Acknowledged

The original publication listed several caveats that are often omitted in secondary coverage:

  1. Population homogeneity. Over 99% of participants were white postmenopausal women. Efficacy and safety in men, premenopausal women, and non-white populations cannot be directly inferred.
  2. Concomitant calcium/vitamin D. All patients were supplemented. The results assume adequate mineral intake, a condition not guaranteed in real-world prescribing.
  3. Three-year duration. Optimal treatment duration and the safety profile beyond three years were unknown at publication. Subsequent extension data (HORIZON extension, published 2012) suggested that six years of annual infusion maintained fracture protection, but the conversation about bisphosphonate holidays had already begun.
  4. Atrial fibrillation signal. The authors flagged this finding as unexpected and called for further investigation.

Who Benefits Most (and Who Does Not Match This Trial)

The HORIZON-PFT population was narrowly defined: postmenopausal women aged 65 to 89 with established osteoporosis. The following patient groups fall outside the trial's direct evidence base:

  • Men with osteoporosis. Zoledronic acid is FDA-approved for male osteoporosis based on smaller studies, but the fracture reduction magnitude seen in HORIZON-PFT should not be directly extrapolated.
  • Glucocorticoid-induced osteoporosis. The companion HORIZON trial (Black et al., 2007, separate publication) addressed this, but with different endpoints and a shorter duration.
  • CKD stage 3b or worse. Creatinine clearance <35 mL/min is a contraindication. Patients with borderline renal function need pre-infusion hydration and post-infusion monitoring.
  • Patients on denosumab switching to zoledronic acid. This clinical scenario, now common in practice, was not part of the trial design. Observational data suggest a single zoledronic acid infusion after denosumab discontinuation may not fully prevent bone loss, and two infusions spaced six months apart may be needed.

The Adherence Argument in Numbers

The strongest practical argument for IV zoledronic acid is adherence. Oral bisphosphonate adherence studies consistently show 12-month medication possession ratios below 50%. A 2012 meta-analysis found that 57% of oral bisphosphonate users were non-adherent within one year. HORIZON-PFT reported 88.8% of zoledronic acid patients completed all three annual infusions. The gap between 50% adherence and 89% adherence explains why real-world fracture reduction with IV zoledronic acid often exceeds what oral agents achieve in practice, even though efficacy in controlled trials is comparable.

Bottom Line for Prescribers

HORIZON-PFT did not prove that zoledronic acid is superior to oral bisphosphonates in fracture reduction. It proved equivalence in efficacy with a dosing schedule that dramatically improves adherence. For a disease where the biggest barrier to fracture prevention is patients stopping therapy, that distinction matters more than any head-to-head endpoint.

Frequently asked questions

›
›
›
›
›
›
›
›
›
›

References

  1. Black DM, Delmas PD, Eastell R, et al. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med. 2007;356(18):1809-1822. PubMed
  2. Reclast (zoledronic acid) prescribing information. Novartis. Revised 2022. FDA Label
  3. Camacho PM, Petak SM, Binkley N, et al. American Association of Clinical Endocrinologists/American College of Endocrinology clinical practice guidelines for the diagnosis and treatment of postmenopausal osteoporosis, 2020 update. Endocr Pract. 2020;26(Suppl 1):1-46. PubMed
  4. Prolia (denosumab) prescribing information. Amgen. Revised 2022. FDA Label
  5. Black DM, Reid IR, Boonen S, et al. The effect of 3 versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-Key Fracture Trial (PFT). J Bone Miner Res. 2012;27(2):243-254. PubMed