Myeloma-Specific SRE Hazard Ratios from Pooled Bone-Agent Trials

What the evidence says

The pooled evidence on denosumab versus zoledronic acid in myeloma points one way. A 2025 meta-analysis reports a hazard ratio (HR) of 0.98 (95% CI 0.85–1.13, P = 0.78) for time to first skeletal-related event (SRE) in the myeloma subgroup. For time to first and subsequent SREs, it reports an HR of 1.01 (95% CI 0.89–1.15). Both results show no measured difference. The same analysis reports that solid-tumor subgroups did differ, with a test for subgroup differences at P = 0.04 for the second outcome. Both agents inhibit bone breakdown, but through different routes. The pooled data show no myeloma-specific advantage for either.

The underlying trials agree. In the mixed-malignancy trial, the myeloma subset had an HR of 1.03 (p = 0.89) for time to first SRE, even with greater suppression of bone turnover markers. The myeloma-only non-inferiority trial in newly diagnosed patients reports an HR of 0.98 (95% CI 0.85–1.14), meeting its non-inferiority endpoint. A review of these trials reports the same figure, 0.98 (95% CI 0.85–1.14).

For older comparisons, zoledronic acid decreases skeletal events relative to clodronate and pamidronate. In the Myeloma IX trial, new skeletal lesions were 27% versus 35% (HR 0.74, 95% CI 0.62–0.87). For SRE risk against clodronate, a review reports an HR of 0.77 in patients with bone disease at baseline and 0.53 in patients without it, measured by radiography. Neither of those two HRs has a confidence interval in the sources. A retrospective study of 1,018 U.S. veterans reports an HR of 0.75 (95% CI 0.60–0.94) against pamidronate. That study is observational, so confounding by indication cannot be ruled out.

Treatment timing and duration also matter. In the trials reviewed, zoledronic acid decreases SREs when given at biochemical relapse compared with observation (2 versus 14 events). Giving it for up to four years instead of two decreases SREs in ISS III disease (21% versus 41%). Pamidronate and zoledronic acid in asymptomatic myeloma decrease SREs at progression: 22/56 versus 40/55 for pamidronate, and 55.5% versus 78.3% for zoledronic acid. Starting zoledronic acid within 60 days of diagnosis is associated with a higher SRE-free rate at two years (74.6% versus 56.5%). That last result comes from a non-randomized retrospective study with unbalanced baseline characteristics.

Dosing interval shows no measured effect. Zoledronic acid every 12 weeks and every 4 weeks gave at least one SRE in 28.6% and 29.5% of patients over two years. That trial included mixed cancer populations. A retrospective myeloma comparison of de-escalated versus monthly dosing also found no difference in SRE incidence.

Some outcomes have thin evidence. Three single-arm studies with no comparator report few SREs. Two gave no bisphosphonate after transplant and consolidation, with 1 SRE among 42 and 1 among 59 patients. One used marker-directed dosing, with 12 of 121 patients (9.9%) having an SRE. Without a comparison group, none of these can show what the drug did. Neither the pooled analyses nor the trials here report a myeloma-specific HR for bisphosphonate versus placebo or no treatment.

Where the evidence splits

Denosumab versus zoledronic acid. The pre-specified primary analysis finds an HR of 0.98 for time to first SRE, and the meta-analysis finds no myeloma benefit. A post-hoc analysis reports an HR of 0.66 (95% CI 0.44–0.98, p = 0.039) for SRE occurrence during the study. Two sources describe this analysis. One describes it as using 15 months as a new baseline. The sources state that it was post-hoc, not pre-specified, and that its upper confidence bound sits near 1.0. They do not say why it differs from the primary result beyond that.

Continuing zoledronic acid beyond 2 years. Randomized evidence reports an HR of 0.37 (95% CI 0.15–0.89) in patients with a very good partial response or better, and a progression HR of 0.40 for bone disease. An observational study reports an overall HR of 0.42 but an HR of 1.08 (p = 0.9) in its myeloma subgroup. The sources note that the observational study had no baseline characteristics described and a small myeloma subgroup. They do not show that these differences explain the split. The split is unexplained by the sources on this page.

Thresholds and dose-response

• Start within 60 days of diagnosis: the SRE-free rate at two years is 74.6% with zoledronic acid started within 60 days, versus 56.5% with delayed initiation. The evidence is retrospective.
• Treatment beyond 2 years: the SRE rate is 21% at up to four years versus 41% at two years in ISS III disease. Continuing past two years in patients with a very good partial response or better has an HR of 0.37.
• Dosing interval: a 7% margin defined non-inferiority between 12-week and 4-week dosing. The observed rates were 28.6% and 29.5%.

No threshold is stated for a level at which an effect reverses or changes size. The sources give no dose-response curve for denosumab or zoledronic acid.

Open empirical questions

• What is the myeloma-specific SRE HR for any bone agent against placebo or no treatment in a pooled analysis?
• Is the post-hoc 0.66 HR for denosumab reproduced in a pre-specified analysis?
• What explains the null myeloma subgroup in the observational study of continuing zoledronic acid beyond 2 years, when randomized data show a decrease?
• How do SRE rates compare between bisphosphonate and no bisphosphonate in patients with a deep response after modern consolidation, given that the sources contain only single-arm data?