975P - Gene methylation in myelodysplastic syndrome - a comparative study between bone marrow and peripheral blood

Date 28 September 2014
Event ESMO 2014
Session Poster Display session
Topics Plasma Cell Dyscrasias
Pathology/Molecular Biology
Presenter Joana Jorge
Citation Annals of Oncology (2014) 25 (suppl_4): iv327-iv339. 10.1093/annonc/mdu339
Authors J.M. Jorge1, E. Cortesão2, A.C. Gonçalves1, A. Pires1, R. Alves1, C. Moucho2, L. Rito2, E. Magalhães2, J. Carda2, C. Geraldes2, A.I. Espadana2, M.A. Pereira3, M.D.A.R..F. Dourado4, L. Ribeiro2, J.M. Nascimento-Costa5, A.B. Sarmento-Ribeiro1
  • 1Faculty Of Medicine, Coimbra University, Applied Molecular Biology and University Clinic of Hematology, 3000 - Coimbra/PT
  • 2Clinical Hematology Service, Centro Hospitalar e Universitário de Coimbra, 3000 - Coimbra/PT
  • 3Medicine Service, Hospital Distrital da Figueira da Foz, 3094 - Figueira da Foz/PT
  • 4General Pathology, Faculty of Medicine of University of Coimbra, PT-3004-504 - Coimbra/PT
  • 5Oncology And Paliative Care, Faculty of Medicine of University of Coimbra, Coimbra/PT



Blood-based specimens may be a potential source of non-invasive DNA methylation cancer biomarkers. Peripheral blood leukocytes from patients with solid tumors exhibit complex and distinct cancer-associated DNA methylation patterns, which might be seen as epigenetic biomarkers with significant clinical potential. However, peripheral blood cell methylation profiles are largely unknown in hematopoietic cancers. Our aim was to compare DNA methylation status in bone marrow (BM) aspirate and peripheral blood (PB) of Myelodysplastic Syndrome (MDS) patients at diagnosis.


We compare DNA methylation status of the tumor suppressor genes, p15, p16, p53, DAPK and MGMT, and of TRAIL (TNF-Related Apoptotic Inducing Ligand) receptor genes, TRAIL-DcR1, -DcR2, -DR4 and -DR5), in 45 Myelodysplastic Syndrome (MDS) patients at diagnosis, in genomic DNA obtained from BM aspirate and PB samples, after informed consent. Genomic DNA was isolated by standard protocols and modified by sodium bissulphite. The MS-PCR for each gene was performed using two sets of primers, one for methylated DNA and other for unmethylated DNA. χ2 Test was used to analyses association between groups and Kappa statistics to evaluate concordance, results were considered statistically significant when p < 0.05.


We observed a good concordant results between BM and PB samples in 77,8% of patients for p16 (p = 0,017), 75,6% for p15 (p = 0,028), 62,2% for DAPK (p = 0,031), 84,4% for TRAIL-DcR1, 68,9% for TRAIL-DcR2, 66,7% for TRAIL-DR4 and a discordant results for TRAIL-DR5 gene (p = 0,012), since only 37,8% of the tested samples were concordant. In some cases discrepancies were also bidirectional, with cases presenting demethylated PB and methylated BM aspirate and vice versa. No patient presented p53 and MGMT gene methylated.


Our results show a correlation between gene methylation patterns in PB and BM aspirate in MDS patients. Although DNA methylation patterns measured in PB may have great potential as informative biomarkers of cancer risk and prognosis, large systematic and prospective studies will be needed.


All authors have declared no conflicts of interest.