The Spine Blog

Tuesday, October 8, 2013

Does BMP-2 cause cancer? Results from the Medicare Database

The use of BMP-2 in spine surgery remains controversial, with some authors suggesting it increases the risk of complications and two recent independent reviews of the original industry-sponsored trial data (the Yale Open Data Access project, or YODA) concluding that BMP-2 did not improve clinical outcomes compared to iliac crest bone graft (ICBG).1-3  While Carragee et al. concluded that BMP-2 may be associated with complications such increased post-operative pain, wound healing problems, and retrograde ejaculation, most concerning was a possible association with cancer.4 The YODA papers found a 2-3 fold increased risk of cancer in the BMP patients compared to the ICBG patients, and this difference was found to be statistically significant in one of the two reviews.2,3 More recently, Carragee et al. published an analysis of the AMPLIFY study data and concluded that the rate of new cancer events in the BMP-2 patients was 6 times as high as in the ICBG patients.1 AMPLIFY is a higher dose preparation of BMP-2, and the study protocol used a 40 mg dose for a one level posterolateral fusion. Inspection of the data table in this paper demonstrates that 7 of the 12 new cancer events reported in the first two years in the BMP-2 group were skin cancers occurring in three patients. In comparing non-skin cancer events reported in the first 24 months, 5 were observed in the BMP-2 group and 2 in the ICBG group. Given the low numbers of cancers reported in the clinical trials, it is hard to draw strong conclusions based on these data.


It is on this background that Drs. Cooper and Kou, in a study that was sponsored by Medtronic, decided to query the Medicare database in order to study a large cohort of patients and determine if there was an association between BMP-2 and cancer. They evaluated all Medicare patients undergoing lumbar fusion from 2003-2008, excluding those with a prior or current diagnosis of cancer. They included over 146,000 patients, with BMP-2 having been used in about 15%. They subsequently followed these patients into the future for an average of 4.7 years and compared the rates of cancer among those who received BMP and those who did not. In order to bolster their results, the authors ran the analyses using 3 different definitions of cancer, ranging from a highly sensitive definition requiring just one code for cancer to a more stringent definition requiring two separate diagnosis codes for the same cancer as well as a code for cancer treatment. They found that BMP was not associated with overall cancer risk or risks of specific types of cancer. In the analysis using the most stringent definition of cancer, the BMP patients actually had a slightly lower risk of all cancer types and specifically brain cancer.


This analysis based on Medicare data comes to a markedly different conclusion than the other recent papers on this topic. It seems likely that the cause of the discrepant results is the different methodologies employed and the different patient populations in each study. While there were significantly more cancer events in the BMP-2 patients than in the ICBG patients in the AMPLIFY study, a single patient experienced 6 of the 20 new cancer events observed over five years, and five of these were skin cancers. Another patient in the BMP-2 group experienced three skin cancers. These two patients accounted for nearly 50% of the new cancer events, and 50% of the new cancer events were basal cell carcinoma, squamous cell carcinoma or melanoma. While it is possible that BMP-2 causes skin and other cancers, it is also possible that two patients prone to skin cancer were randomly assigned to the BMP-2 group and would have developed these cancers without exposure to BMP-2. Another high dose BMP trial did not demonstrate an increased risk of cancer, though this study remains unpublished and was reported only in the YODA reviews.2,3 Bridwell et al. recently published a paper comparing outcomes and complications in about 30 adult deformity patients treated with very high dose BMP-2 (140 mg on average) and a control group treated with ICBG.5 They reported one patient with an acoustic neuroma in the BMP -2 group and one patient with skin cancer and another with uterine fibroids in the ICBG group. While the current Medicare study does have all of the limitations of a large database study, it seems likely that coding for the use of BMP-2 and cancer is fairly reliable. With the large numbers involved, the chance of a Type II error is very low, and it seems reasonable to conclude that BMP-2 is not associated with cancer in the Medicare population. It should be noted that squamous cell carcinoma and basal cell carcinoma were not included in the database, so this study does not provide any information about the association with BMP-2 and these types of cancer. Additionally, it should be noted that Medtronic funded the study. Given the conflicting results on this topic, it is likely that the only method to get a more definitive answer would be the creation of a registry that could prospectively track complications in all patients who receive BMP-2.


Please read Dr. Cooper’s article on this topic in the October 1 issue. Does this study change the way you consider the association between BMP-2 and cancer? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor



1.            Carragee EJ, Chu G, Rohatgi R, et al. Cancer risk after use of recombinant bone morphogenetic protein-2 for spinal arthrodesis. The Journal of bone and joint surgery American volume 2013;95:1537-45.

2.            Fu R, Selph S, McDonagh M, et al. Effectiveness and harms of recombinant human bone morphogenetic protein-2 in spine fusion: a systematic review and meta-analysis. Annals of internal medicine 2013;158:890-902.

3.            Simmonds MC, Brown JV, Heirs MK, et al. Safety and effectiveness of recombinant human bone morphogenetic protein-2 for spinal fusion: a meta-analysis of individual-participant data. Annals of internal medicine 2013;158:877-89.

4.            Carragee EJ, Hurwitz EL, Weiner BK. A critical review of recombinant human bone morphogenetic protein-2 trials in spinal surgery: emerging safety concerns and lessons learned. The spine journal : official journal of the North American Spine Society 2011;11:471-91.

5.            Kim HJ, Buchowski JM, Zebala LP, Dickson DD, Koester L, Bridwell KH. RhBMP-2 is superior to iliac crest bone graft for long fusions to the sacrum in adult spinal deformity: 4- to 14-year follow-up. Spine 2013;38:1209-15.