AI-generated research brief — verify at source
ICI Therapy Shows No Significant Association with Fracture Risk in Solid Malignancy Trials
A pooled analysis and meta-analysis of 35 randomized controlled trials found no statistically significant increase in fracture risk among patients receiving immune checkpoint inhibitors compared to controls, with odds ratios of approximately 1.10 for any fracture (OR 1.10, 95% CI 0.80–1.50), 1.14 for osteoporotic fracture (OR 1.14, 95% CI 0.78–1.64), and 0.84 for pathologic fracture (OR 0.84, 95% CI 0.26–2.73). These neutral findings, however, are accompanied by wide confidence intervals and a recognized gap in fracture reporting across ICI trials, which limits definitive conclusions.
What Was Studied
This study examined whether immune checkpoint inhibitor therapy is associated with an elevated risk of fracture — including any fracture, osteoporotic fracture, and pathologic fracture — compared to non-ICI treatment in patients with solid malignancies. The question arose because ICIs may promote osteoclast activation through off-target immune effects, potentially increasing skeletal vulnerability, yet fractures are not formally classified as immune-related adverse events in current frameworks.
How It Was Studied
Researchers conducted a systematic literature review of PubMed and Embase from database inception through September 25, 2024, identifying phase II and III randomized controlled trials of ICIs in solid malignancy populations that reported fracture outcomes. Supplementary materials from all eligible studies were reviewed to capture fracture data that may not have appeared in primary publications. A total of 35 RCTs encompassing 23,404 patients were included. Pooled analysis, random-effects meta-analysis, and multivariate meta-regression were all employed, along with two pre-specified subgroup analyses focusing on trials without an active comparator and on those with at least six months of ICI exposure.
What Was Observed
- Across all fracture types, ICI-treated patients showed no meaningfully elevated risk compared to controls — a roughly 10% higher odds that was not statistically significant (OR 1.10, 95% CI 0.80–1.50, p = 0.54). In absolute terms, 91 fractures occurred in ICI groups versus 70 in control groups across the 35 trials.
- Osteoporotic fractures, which represented the largest fracture subcategory (66 in ICI groups vs. 49 in controls), similarly showed no significant association with ICI use — a roughly 14% higher odds that did not reach statistical significance (OR 1.14, 95% CI 0.78–1.64, p = 0.48), with a wide confidence interval reflecting considerable uncertainty.
- Pathologic fractures were rare in both groups (5 each) and showed no significant difference, with the odds ratio actually slightly favoring ICI users in point estimate — though the estimate is highly imprecise (OR 0.84, 95% CI 0.26–2.73, p = 0.79).
- Statistical heterogeneity across all analyses was negligible (I² = 0.0%), and meta-regression identified no variables predictive of increased fracture risk. Subgroup analyses did not alter the overall neutral findings.
Why This Matters
ICIs are approved for numerous solid malignancies, and understanding their full adverse event profile — including skeletal effects — is essential for patient management. Because fractures are not currently categorized as immune-related adverse events, they may be systematically underreported or overlooked in ICI trial designs. This analysis represents a first attempt at pooling fracture data specifically from ICI randomized trials, and its findings highlight that current trial infrastructure is insufficiently equipped to capture skeletal outcomes, underscoring the need for standardized fracture reporting as a mandated adverse event category in future ICI studies.
How to Read This Result
These neutral findings should be interpreted cautiously: the wide confidence intervals across all fracture outcomes, the short observation periods in most included trials, and the infrequent fracture reporting mean that a true long-term skeletal signal from ICI-mediated osteoclast activation cannot be excluded on the basis of currently available data.
Limitations
Only a minority of eligible trials reported fracture outcomes, creating the potential for underascertainment bias. Observation periods across included studies were generally short, making it difficult to detect fracture risks that may accumulate over time. The small absolute number of fractures — particularly pathologic fractures — limits statistical power for subgroup comparisons. The authors call for longer-term, comprehensive fracture adverse-event reporting in future ICI randomized controlled trials to adequately evaluate skeletal safety in this growing patient population.