Abstract 122 The development of cancer is driven by the accumulation of scores of alterations affecting the structure and function of the genome. Equally important in this process are genetic alterations and epigenetic changes. Whereas the former disrupt normal patterns of gene expression, sometimes leading to the expression of abnormal, constitutively active proteins, the latter deregulate the mechanisms such as transcriptional control leading to the inappropriate silencing or activation of cancer-associated genes. In this report, the effect of differential epigenetic alterations on the clinical outcome in multiple myeloma (MM) patients is presented. Peripheral blood mononuclear cells (PBMC) from twelve human healthy volunteers (7M/5F; median age 41 years, range 28 to 52 years) and 32 MM patients (14M/18F; median age 59 years, range 23 to 71 years) who underwent high dose melphalan with autologous stem cell support (ASCT) as part of their first line therapy. Twenty-three patients achieved an objective response post-ASCT while 9 patients did not respond according to IMWG criteria as well as human primary fibroblasts with different repair backgrounds (VH25, GGR+/TCR+; CS1AN, GGR+/TCR-; XP21RO, GGR-/TCR+; XP25RO, GGR-/TCR-) were exposed to melphalan. Chromatin condensation (using micrococcal nuclease digestion), transcription activity (steady-state levels using RNA slot-blots and rates of transcription using run-off assay), as well as in vitro melphalan-induced damage formation/repair (monoadducts and interstrand cross-links using Southern blot) were measured in three transcriptionally active genes (the housekeeping beta-actin gene, the tumor suppressor p53 and the protooncogene N-ras) as well as in the non-transcribed delta-globin gene. In repair-deficient human primary fibroblasts, melphalan adducts accumulated to similar levels in all four genes studied, indicating that the state of transcription and the local chromatin condensation did not affect adduct formation. In all MM patients examined, a close association was observed between the locus-specific repair efficiency, the transcriptional activity and the chromatin condensation. Strikingly, the order of variation of DNA repair efficiency, transcriptional activity and local chromatin structure between different genes was beta-actin>p53>N-ras>delta-globin in all healthy volunteers and in 95% of responders to chemotherapy, while a perturbation of this order was found in 80% of non-responders. In agreement with our previous reports (Dimopoulos et al, J Clin Oncol 2005;23:4381-9; Dimopoulos et al, Haematologica 2007;92:1505-12) in the p53, N-ras and delta-globin gene, responders to HDM showed higher melphalan-induced damage and lower rates of repair compared to non-responders. These differences were statistically significant for all types of melphalan adducts. No difference in the repair activity between responders and non-responders was observed in the beta-actin gene. Also, in all genes examined, DNA damage formation and repair induced by in vitro melphalan treatment of PBMC taken from MM patients prior to therapy, correlates with the respective data obtained in vivo, i.e. after therapeutic administration of HDM. Furthermore, the kinetics of melphalan adduct formation/repair along the transcribed N-ras gene showed a 5' to 3'-end gradient of repair efficiency, with faster repair at the 5'-end. Interestingly, the difference in the repair activity between responders and non-responders becomes maximal at the 5'-end of the N-ras gene, decrescent towards the 3'-end of the gene. As for the delta-globin gene, an induction of the transcription activity, a higher “looseness” of the local chromatin structure and an increase in the repair efficiency was observed specifically in MM patients who did not respond to melphalan therapy. No difference between responders and non-responders was found in regions on both sides of the genes as well as at the overall genome repair, suggesting that in non-responders the repair system active in the removal of melphalan-induced adducts was not grossly affected. In conclusion, our study suggests that the DNA repair efficiency, the transcriptional activity and the local chromatin structure in different gene loci are tightly connected, and can be used as molecular markers to select those patients with MM who are more likely to benefit from HDM therapy. Disclosures: No relevant conflicts of interest to declare.