731P - The predictive value of bone morphogenetic protein 9 (BMP-9) in the serum of HCC patients treated with sorafenib

Date 29 September 2014
Event ESMO 2014
Session Poster Display session
Topics Anticancer Agents
Hepatobiliary Cancers
Translational Research
Basic Principles in the Management and Treatment (of cancer)
Biological Therapy
Presenter Leonidas Chelis
Citation Annals of Oncology (2014) 25 (suppl_4): iv210-iv253. 10.1093/annonc/mdu334
Authors L. Chelis1, K. Anagnostopoulos2, S. Vradelis3, N. Xenidis1, K. Amarantidis1, D. Pitsiava4, P. Michailidis1, E. Christakidis1, A. Chalvatzis1, D. Matthaios1, I. Biziota1, A. Kortsaris2, S. Kakolyris1
  • 1Medical Oncology, University Hospital of Alexandroupoli, 68100 - Alexandroupolis/GR
  • 2Biochemistry, Democritus University of Thrace, Medical School, 68100 - Alexandroupolis/GR
  • 3Internal Medicine, Democritus University of Thrace, Medical School, 68100 - Alexandroupolis/GR
  • 4Internal Medicine, University Hospital of Alexandroupoli, 68100 - Alexandroupolis/GR



BMP-9 is produced in the liver and is the main ligand of ALK1 ( activin like-receptor kinase1) . BMP-9 is considered a vascular quiescence factor that acts in the maturation phase of angiogenesis as an angiogenesis and lymphangiogenesis inhibitor. Sorafenib is an anti-angiogenic agent that represents the only treatment for advanced hepatocellular carcinoma (HCC). Therefore we conducted a prospective clinical trial in order to investigate the potential role of BMP-9 as a predictive factor in sorafenib treated HCC patients (pts).


Our study population was 70 HCC pts who received 1st line sorafenib and 30 healthy controls. Serum samples were collected at baseline and after 4 weeks on treatment. Using commercially available ELISA kits we measured the serum levels of BMP-9 and correlate them with overall survival (OS) and progression free survival (PFS). To the best of our knowledge this is the first time that blood levels of BMP-9 is assessed in cancer patients.


Sixty seven (n = 67) pts were evaluable for efficacy. The OS for all pts was 9,6 months and the PFS was 4,2 months. At baseline the median values of BMP-9 for pts and controls were about the same : 248,5 pg/ml and 273 pg/ml respectively. After 4 weeks of treatment, BMP-9 levels were decreased to 180,4 pg/ml, median change (-27,4%) and this reduction was highly significant (p = 0.006). The OS for pts with higher or lower BMP-9 levels at baseline was 9,7 and 11,6 months respectively (p = NS). After 4 weeks of sorafenib, pts who demonstrated higher reduction of BMP-9 were experienced shorter OS : 5,2 months vs 13,4 months for pts with lower reduction of BMP-9, p = 0,038. This was also true for PFS, 2,8 months vs 6,4 months respectively, p = 0.021.


Measuring serum levels of BMP-9 in HCC pts is feasible. A significant decrease of BMP-9 levels was observed and this is a treatment effect of sorafenib. This reduction of BMP-9 had a negative predictive value since pts with higher decrease of BMP-9 experienced shorter OS and PFS . Our hypothesis generating trial suggests that the reduction of BMP-9 levels represents a novel mechanism of tumour escape from anti-angiogenic treatments. Whether BMP-9 could also serve as a predictive marker needs further exploration in bigger trials.


All authors have declared no conflicts of interest.