Study design: Retrospective cohort study among Medicare beneficiaries with lumbar spinal fusion surgery.
Objective: To determine the risk of subsequent cancer among patients who received recombinant human bone morphogenic protein (rhBMP) at surgery compared with those who did not.
Summary of background data: rhBMP is commonly used to promote bone union after spinal surgery. BMP receptors are present on multiple cancer types, but the risk of cancer after receiving rhBMP has not been well studied.
Methods: We identified 146,278 subjects aged 67 years and older who underwent surgery in 2003 to 2008 and were followed through 2010 for a new diagnosis of 1 of 26 cancers. Proportional hazards models were used to determine cancer risk associated with rhBMP use.
Results: rhBMP was administered in 15.1% of the cohort. After an overall average follow-up of 4.7 years, 15.4% of rhBMP-treated and 17.0% of untreated patients had a new cancer diagnosis, with most commonly recorded types as prostate, breast, lung, and colorectal. In a multivariate proportional hazards model, there was no association of rhBMP with cancer risk (hazard ratio: 0.99, 95% confidence interval: 0.95-1.02). There was also no association of rhBMP with the risk of any individual cancer types. The results were consistent in analyses using 2 secondary definitions of incident cancer.
Conclusion: In this large population-based analysis of Medicare beneficiaries, we found no evidence that administration of rhBMP at the time of lumbar fusion surgery was associated with cancer risk.
Level of evidence: 4.