Zoledronic acid
Zoledronic acid, an intravenous nitrogen-containing bisphosphonate
Written by Aaron CuhaReviewed Sep 2026
Also known as: Zoledronate, Reclast, Aclasta, Zometa
An approved once-yearly bone drug carrying the most interesting unexplained mortality signal in geroscience: 9.6 percent died on zoledronic acid against 13.3 percent on placebo in a trial after hip fracture, a secondary finding never satisfactorily explained. A meta-analysis of 38 osteoporosis drug trials then found no mortality benefit.
Overview
This is the most useful page in this batch and the easiest one to overstate, so the numbers come first and the interpretation comes second.
HORIZON Recurrent Fracture Trial, registered as NCT00046254, randomised 2,127 patients with a mean age of 74.5 years to a yearly 5 mg intravenous infusion of zoledronic acid or placebo, starting within 90 days of surgical repair of a hip fracture. Median follow-up was 1.9 years. The primary endpoint was new clinical fracture, and it was met: 8.6 percent against 13.9 percent, a 35 percent risk reduction.
In the safety analysis, 101 of 1,054 patients on zoledronic acid (9.6 percent) and 141 of 1,057 on placebo (13.3 percent) died, a 28 percent reduction in death from any cause, p equals 0.01. Mortality was not the primary endpoint. It emerged from the safety data of a fracture trial.
A 2010 exploratory analysis by the trial's own investigators tried to explain it and largely failed. Only 8 percent of the survival benefit was attributable to preventing subsequent fractures. Adjusting for acute events occurring during follow-up eliminated the death benefit, and patients on zoledronic acid were less likely to die of pneumonia and of arrhythmias than patients on placebo. The authors concluded that further study of zoledronic acid in other acute illnesses might be warranted. In other words, whatever produced the survival difference, it was mostly not the bones.
Then the disagreement. A 2019 meta-analysis in JAMA Internal Medicine pooled 38 randomised placebo-controlled osteoporosis drug trials covering 101,642 participants and found no association between drug treatment and overall mortality, risk ratio 0.98. Restricted to bisphosphonates, 0.95. Restricted to the 6 zoledronate trials, 0.88 with a 95 percent confidence interval from 0.68 to 1.13, not significant, with meaningful heterogeneity between studies. The authors concluded bisphosphonates should be recommended to reduce fracture risk and not for mortality, while explicitly stating that additional trials are needed to clarify whether zoledronate reduces mortality.
A separate large trial does exist in older women. Reid and colleagues randomised 2,000 New Zealand women over 65 with osteopenia to four infusions of zoledronate or saline over six years. It met its fracture endpoint, hazard ratio 0.63. Its non-skeletal adverse event analysis reported fewer myocardial infarctions, fewer total cancers, and a hazard ratio for death of 0.65 with a 95 percent confidence interval from 0.40 to 1.06, which is a trend and not a result. The authors said these apparent effects justify appropriately powered trials with those conditions as primary endpoints. Those trials have not reported.
The senescence connection came later. A Mayo Clinic group showed in 2023 that zoledronic acid killed senescent cells in culture with minimal effect on non-senescent ones, and that eight weeks of treatment in aged mice reduced circulating SASP factors and improved grip strength. That is a plausible mechanism for a non-skeletal effect. It is not proof that the mechanism explains the HORIZON numbers.
Mechanism of action
Zoledronic acid binds avidly to bone mineral and is taken up by osteoclasts, where it inhibits farnesyl pyrophosphate synthase in the mevalonate pathway. That blocks prenylation of small GTPases the osteoclast needs, and the osteoclast stops resorbing bone and dies. This is the established mechanism and it explains the fracture result completely. It does not explain the mortality signal. Proposed non-skeletal mechanisms include a senolytic or senomorphic action, supported by Mayo Clinic work showing selective killing of senescent cells in vitro and reduced circulating SASP factors and improved grip strength in aged mice; effects on pre-osteoclastic cells that carry senescence markers; and general effects of the mevalonate pathway on immune cells. None has been shown to account for the observed survival difference in people.
Human evidence
Zoledronic acid has more randomised human data than anything else on this page, all of it collected to answer a question about bones. The mortality signal is genuine data from a real randomised trial, and it was a secondary or exploratory finding, and the field does not agree on whether it is real.
- HORIZON-RFT, 2,127 randomised after hip fracture: all-cause death 9.6 percent on zoledronic acid against 13.3 percent on placebo, a 28 percent reduction, p equals 0.01. Not the primary endpoint.
- HORIZON-RFT primary endpoint, new clinical fracture: 8.6 percent against 13.9 percent, a 35 percent reduction, p equals 0.001.
- Investigator analysis of the mortality effect: only 8 percent was explained by preventing further fractures, and adjusting for acute events during follow-up eliminated the effect entirely. Fewer deaths from pneumonia and from arrhythmias on drug.
- The 2019 JAMA Internal Medicine meta-analysis of 38 trials and 101,642 participants found no mortality benefit for osteoporosis drugs overall (0.98), for bisphosphonates (0.95), or for the six zoledronate trials pooled (0.88, CI 0.68 to 1.13, not significant).
- Reid and colleagues, 2,000 osteopenic New Zealand women over 65 across six years: fragility fracture hazard ratio 0.63, the trial's primary endpoint.
- In the same 2,000-woman trial, as prespecified and post hoc adverse event analyses: fewer myocardial infarctions (rate ratio 0.58), fewer total cancers (hazard ratio 0.67), and a death hazard ratio of 0.65 with a confidence interval crossing 1.
- No trial anywhere has been run with mortality or any ageing endpoint as its primary outcome.
What this does not tell you: Every mortality number here is secondary, exploratory or from an adverse event analysis. That is not a technicality: secondary findings in trials powered for something else are exactly the findings that most often fail to replicate, and the 2019 meta-analysis is what failing to replicate looks like. The HORIZON population had just broken a hip and had a high near-term death rate, which is where a survival difference is easiest to detect and hardest to generalise from. The Reid trial's cancer and cardiac findings were not its primary endpoint either, and its own authors called for properly powered trials rather than claiming the result. Nobody has shown that a healthy person without osteoporosis gains anything from this drug.
Reading the research record
This page deliberately does not settle the disagreement, because the field has not.
The case that the mortality effect is real: it came from a large double-blind placebo-controlled randomised trial, not an observational cohort, so confounding by indication cannot explain it. It was large in absolute terms, 40 fewer deaths. A second independent randomised trial in a different country and a healthier population found effects pointing the same way for cancer, myocardial infarction and death. And a plausible non-skeletal mechanism has since been demonstrated in cells and aged mice, with zoledronic acid killing senescent cells selectively and reducing circulating SASP factors.
The case that it is not: it was not the primary endpoint, and the investigators' own mediation analysis could not find a route for it, with adjustment for acute events during follow-up abolishing the effect entirely. When 38 randomised trials covering 101,642 people were pooled, the mortality benefit disappeared, and the zoledronate-only subset remained non-significant with real heterogeneity between studies. The Reid trial's death hazard ratio crossed 1. No trial has ever been run with mortality as the question it was designed to answer.
What both camps agree on is the remedy, and it is the same sentence in the 2019 meta-analysis and in the 2020 Reid analysis: an appropriately powered trial with these endpoints as the primary outcome is needed. That trial has not reported. Until it does, zoledronic acid is a well-evidenced fracture drug with an unexplained mortality signal attached, which is a genuinely interesting thing to be and is not the same thing as a proven anti-ageing drug.
One conflict of interest worth stating: the senior author of the meta-analysis that found no mortality benefit discloses grants and personal fees from Amgen, which markets a competing osteoporosis drug. This is disclosed in the paper. It is noted here because readers should weigh it themselves, not because it invalidates a meta-analysis of 38 randomised trials.
The evidence, charted
Fig. 1 · evidence composition
4of 6 citations (67%) are in people
Counted from the citation list on this page. The Human count is the same number shown in the badge at the top. Both understate any literature larger than the sources we cite.
Fig. 2 · evidence over time
Evidence spans 6 distinct years, 2007 to 2023, counted from the citation list on this page.
Fig. 3 · legal status at a glance
US
Approved
UK
Approved
AU
Approved
CA
Approved
Approved in all four jurisdictions shown. A jurisdiction's classification is a regulatory fact, not a verdict on the science; see Legal status below for the full text and any notes.
Fig. 4 · dose response
No human dose response curve exists
We draw this figure where the data supports it. For this compound in humans it does not, so the panel stays empty rather than borrowing an animal curve and implying it transfers.
Key studies & citations
- Human2007
Zoledronic acid and clinical fractures and mortality after hip fracture
HORIZON Recurrent Fracture Trial, NCT00046254. 1,065 assigned to yearly 5 mg intravenous zoledronic acid and 1,062 to placebo within 90 days of hip fracture repair, mean age 74.5, median follow-up 1.9 years. Primary endpoint new clinical fracture: 8.6 percent against 13.9 percent, a 35 percent reduction. In the safety analysis 101 of 1,054 (9.6 percent) on drug and 141 of 1,057 (13.3 percent) on placebo died, a 28 percent reduction in all-cause death, p equals 0.01. Mortality was not the primary endpoint. Commonest adverse events were pyrexia, myalgia and bone and musculoskeletal pain; no osteonecrosis of the jaw was reported.
New England Journal of Medicine - Human2010
Potential mediators of the mortality reduction with zoledronic acid after hip fracture
Exploratory retrospective analysis of the same 2,111 randomised patients, by the trial investigators. Adjusted for baseline risk factors, zoledronic acid reduced death by 25 percent. Only 8 percent of the death benefit was attributable to preventing subsequent fractures. Adjusting for acute events during follow-up eliminated the death benefit, and treated patients were less likely to die from pneumonia and from arrhythmias. The authors concluded the mechanism is not known.
Journal of Bone and Mineral Research - Review2019
Association Between Drug Treatments for Patients With Osteoporosis and Overall Mortality Rates: A Meta-analysis
38 randomised placebo-controlled trials, 101,642 participants. No association between osteoporosis drug treatment and overall mortality, risk ratio 0.98 (95 percent CI 0.91 to 1.05). Bisphosphonates 0.95 (0.86 to 1.04). Six zoledronate trials: 0.88 (0.68 to 1.13), not significant, with heterogeneity. The authors concluded bisphosphonates should be recommended to reduce fracture risk rather than mortality, and that further trials are needed to clarify whether zoledronate reduces mortality. The senior author discloses grants and personal fees from Amgen.
JAMA Internal Medicine - Human2018
Fracture Prevention with Zoledronate in Older Women with Osteopenia
2,000 New Zealand women aged 65 or older with osteopenia, randomised to four infusions of 5 mg zoledronate or saline at 18-month intervals over six years. Fragility fracture occurred in 190 on placebo and 122 on zoledronate, hazard ratio 0.63 (95 percent CI 0.50 to 0.79). Number needed to treat 15. This is the large trial in older women, and its primary endpoint was fracture, not mortality.
New England Journal of Medicine - Human2020
Effects of Zoledronate on Cancer, Cardiac Events, and Mortality in Osteopenic Older Women
Adverse event analysis of that same 2,000-woman trial. Myocardial infarction in 39 women on placebo against 24 on zoledronate (hazard ratio 0.60, 95 percent CI 0.36 to 1.00). Total cancers reduced, hazard ratio 0.67 (0.51 to 0.89). Hazard ratio for death 0.65 (0.40 to 1.06, p equals 0.08), and 0.51 (0.30 to 0.87) in women without an incident fragility fracture. The authors state these apparent effects justify further appropriately powered trials with these nonskeletal conditions as primary endpoints.
Journal of Bone and Mineral Research - Animal2023
In vitro and in vivo effects of zoledronic acid on senescence and senescence-associated secretory phenotype markers
Mayo Clinic. In human lung fibroblasts and DNA-repair-deficient mouse embryonic fibroblasts, zoledronic acid killed senescent cells with minimal effect on non-senescent cells. In aged mice treated for eight weeks it significantly reduced circulating SASP factors including CCL7, IL-1 beta, TNFRSF1A and TGF beta 1, and improved grip strength. Cell and mouse work proposing a senolytic or senomorphic mechanism for the non-skeletal effects; it does not test the mortality question in people.
Aging (Albany NY)
Frequently asked questions
Does zoledronic acid make you live longer?
Unproven. In a randomised trial of 2,127 people after hip fracture, 9.6 percent on the drug died against 13.3 percent on placebo. That was a secondary finding in a fracture trial, the investigators could not explain it, and a later meta-analysis of 38 trials found no mortality benefit. No trial has tested the question directly.
Is it a senolytic?
There is evidence for it in a dish and in mice. A Mayo Clinic group showed it kills senescent cells selectively in culture and that eight weeks of treatment in aged mice lowered circulating SASP factors and improved grip strength. Whether that explains anything seen in people is unknown.
Should a healthy person take it for longevity?
There is no evidence supporting that. Every human trial enrolled people with fractures, osteoporosis, osteopenia or cancer. It is an intravenous infusion given in a clinical setting, with acute phase reactions (fever, aches) common after the first dose, and rare risks including osteonecrosis of the jaw and atypical femoral fracture with long-term use.
What did the big trial in older women show?
Two thousand New Zealand women over 65 with osteopenia, six years, four infusions: fragility fractures fell, hazard ratio 0.63, which was the primary endpoint. Cancers and heart attacks were also lower and deaths trended lower, but those were adverse event analyses and the death result crossed statistical significance. Its own authors called for properly powered trials rather than claiming the finding.
Why do sources disagree about the mortality effect?
Because the evidence genuinely disagrees. One randomised trial found a clear reduction as a secondary outcome, the mediation analysis could not identify a mechanism, and pooling 38 trials made the effect vanish. Anyone telling you it is settled in either direction is going beyond the data.
How is it given?
As an intravenous infusion, usually once a year for osteoporosis, administered by a clinician. It is not an oral supplement and not something to source privately.