The First-in-Class Alkylating Histone-Deacetylase Inhibitor (HDACi) Fusion Molecule Edo-S101 Exerts Potent Preclinical Activity Against Tumor Cells of Hodgkin Lymphoma (HL) Including Bendamustine-Resistant Clones

Background:The highly unfavorable outcome of patients with recurrent HL, who progress after stem cell transplantation or are ineligible for such procedure make the development of new active agents an impellent medical need in this clinical setting. EDO-S101 is fusion molecule combining the DNA damaging effects of bendamustine (BDM) with the pan-histone deacetylase (HDAC) inhibitor, vorinostat. Given that BDM and HDAC inhibitors are active agents in recurrent HL we investigated the preclinical activity of EDO-S101 in this malignancy.

Methods:We assessed the patterns of  EDO-S101 cytotoxicity (0.39 to 50 µmol/L) in a panel of HL-derived cell lines (L1236, L428, KMH2, HDLM2, L540) and its regulatory effects on genes involved in DNA-damage/repair response, apoptosis and cell cycle checkpoints.  As a further model we exploited an L1236 cell clone (R100) selected for resistance (R) to BDM through continuous exposure to increasing concentrations of the agent. R100 cells display a growth pattern indistinguishable from parental L1236 cells when cultured in the presence of  BDM (100 µmol/L). Clonal identity of R100 cells with parental L1236 was confirmed by sequencing of V3-21 (FR2/FR3) and JH3-JH4 Ig DNA regions.

Results: EDO-S101 induced a significant time- and dose-dependent inhibition of growth and survival in all HL cell lines. L1236 cells displayed the highest sensitivity to the agent with an IC50, at  48 hrs, of 1.88 µmol/L, as opposed to KMH2, L428, L540 and HDLM2 cells with IC50 of 2.06, 2.53, 2.26 and 16.2 µmol/L, respectively. These values were about 10-fold  lower than the IC50 of BDM in the same cell lines. While exposure of L1236 cells to EDO-S101 caused cell accumulation in S-phase, qRT-PCR disclosed that cell death was mainly dependent on triggering of apoptosis, as shown by the early (24 hrs) and sustained (48 hrs) upregulation of NOXA, p21 and p27 genes. Data were confirmed by the significant increase (>150%) of Annexin V-expressing L1236 cells. In contrast, expression levels of PLK1, AKA and cyclin B1 genes remained unchanged or were increased. This excluded induction of the mitotic catastrophe (MC) as a major determinant of cytotoxic activity for EDO-S101 in L1236 cells. Exposure to EDO-S101 induced a strong DNA stress/repair response as shown by the activation of  pATR/pATM and increase of the downstream DNA damage checkpoint proteins  pCHK1-/-2 and CCNB1, along with the upregulation of the EXO1 gene. Most intriguingly, BDM-resistant L1236 cells (R100) were highly sensitive to EDO-S101, with an IC50 of only 4.56 µmol/L, but less responsive to vorinostat (IC50: 6.17 µmol/L) than  parental L1236 cells (IC50: 0.58 µmol/L). Differently from native cells, EDO-S101 induced a late downregulation of transcripts for PLK1 and AKA genes and of cyclin B1 gene and protein in R100 cells, along with the early induction of  NOXA and p21, but not p27 genes. In both L1236 and R100 cells, expression of MC-genes was unaffected by exposure to vorinostat. This suggests a more complex mechanism for EDO-S101 in BDM-resistant HL cells involving activation of the both apoptotic and MC pathways. Notably, we documented that EDO-S101 corrected the constitutive  ATM/ATR unbalance of R100 cells by triggering the early (24 hrs) upregulation of ATR and a late (48 hrs) downregulation of ATM transcripts and proteins, along with increased levels of EXO1 and MGMT at 24 hrs. Vorinostat induced a similar effect. Finally, while baseline expression levels of HDAC isoforms were comparable among HL cell lines, EDO-S101 caused a significant (>40%) late downregulation of transcripts for all HDAC isoforms (HDAC-1 to -8) in R100 cells but only of HDAC-6 in native L1236. This pattern diverged from results obtained in both L1236, i.e. increase of all HDAC isoform transcripts except HDAC-6, and R100 cells, i.e.upregulation of all isoforms and reduction of HDAC-6, with vorinostat and BDM as single agents.

Conclusions: We have described for the first time that EDO-S101 is effective in preclinical models of  HL including cells resistant to BDM. The combined functions of in one molecule of a bifunctional alkylator and panHDAC inhibitor confer this agent unique antitumor property different from both of its single drug components. Following a strong DNA damage response, triggering of apoptosis and/or MC  may take place in HL cells according to their sensitivity status to BDM. A phase 1/2 study in recurrent HL, including patients pretreated with BDM, is next to be launched.