63. Bone morphogenetic protein and cancer in spinal fusion: a propensity matched analysis

Abstract

<h3>BACKGROUND CONTEXT</h3> Since its FDA approval in 2002, recombinant human bone morphogenetic protein-2 (rhBMP-2) has been increasingly used in spinal surgery to promote arthrodesis and avoid the morbidity of autologous iliac crest bone graft harvest. Recent data estimate its utilization in 25% of spinal fusions. Because growth factors such as rhBMP-2 stimulate cellular proliferation, the association of rhBMP-2 with tumorigenesis in spinal fusion is an area of concern. Previous research has generated conflicting conclusions on the risk of cancer in patients receiving rhBMP-2 for spinal fusion. <h3>PURPOSE</h3> The purpose of this study is to compare the incidence of new solid organ and hematopoietic malignancies in patients undergoing spinal arthrodesis with or without BMP in a large administrative database. <h3>STUDY DESIGN/SETTING</h3> Retrospective propensity-matched analysis. <h3>METHODS</h3> Patients undergoing thoracolumbar fusion with or without BMP utilization between 2015-2020 were identified from the PearlDiver Mariner Patient Claims Database (PearlDiver Technologies, Colorado Springs, CO, USA) using ICD-10 codes. Patients with pre-existing malignancy diagnosis were excluded. Data were analyzed for incidence of solid organ malignancy (lung, breast, kidney, prostate, thyroid) and hematopoietic malignancy (lymphoma, myeloma, leukemia) diagnosed after spinal surgery. Propensity score matching was performed between patients who did and did not receive BMP. Propensity matched variables were selected a priori and included age, sex, and year of surgery. <h3>RESULTS</h3> Among patients without a history of solid organ malignancy undergoing thoracolumbar fusion, BMP was used in 9,593 patients and was not used in 193,713. In the propensity matched group, 1.5% of the BMP group developed solid organ malignancy following surgery compared to 1.9% of the non-BMP group. The odds ratio of developing solid organ malignancy in the BMP group compared to the control group was 0.80 (95% CI 0.64-1.00); P=0.056. Among patients without a history of hematopoietic malignancy undergoing thoracolumbar fusion, BMP was used in 9,913 patients and was not used in 191,006 patients. In the propensity matched group, 0.3% of the BMP group developed hematopoietic malignancy compared to 0.5% of the non-BMP group. The odds ratio of developing hematopoietic malignancy in the BMP group compared to the control group was 0.69 (95% CI 0.44-1.07); P=0.12. <h3>CONCLUSIONS</h3> BMP use in a large patient population undergoing thoracolumbar fusion was not associated with increased risk of new solid organ or hematopoietic malignancy. Limitations of this study include its reliance on accurate documentation and coding, including documentation of BMP use which may not be as robust as CPT coding since BMP coding may not be reimbursable. However, for the subset of patients with clear documentation of BMP use, this study further supports emerging data on the lack of a causal relationship between BMP use and malignancy. <h3>FDA DEVICE/DRUG STATUS</h3> Bone morphogenetic protein: Approved for certain types of spinal fusion.