Identification of Acetylation, Immune Cell Alterations and Gene Signature As Dynamic Biomarkers of Response to Panobinostat Treatment in Multiple Myeloma Patients

Sridurga Mithraprabhu, Ioanna Savvidou, Malarmathy Ramachandran, Anna Kalff, Tiffany T. Khong, Kate Reed, John Reynolds, Andrew Spencer

Research output: Contribution to journalMeeting AbstractOtherpeer-review

Abstract

Introduction and aims:Panobinostat is a pan-deacetylase inhibitor (DACi) that affects multiple oncogenic and immune-stimulating genes required for tumour cell growth and survival through acetylation of both histone and non-histone proteins. Evaluation of panobinostat as single agent consolidation in patients post-melphalan 200mg/m2 conditioned autologous stem cell transplant (ASCT) was performed in a Phase II, open label, single arm study, wherein newly diagnosed patients (n=35) with less than complete response following ASCT were enrolled. Eighteen patients (12 responders - R, 6 non-responders - NR) were subsequently selected for correlative analysis to identify dynamic biomarkers that could predict for response to treatment. Acetylation is the key downstream effect of treatment with DACi and was therefore assessed to determine its validity as a pharmacodynamic marker of response. The proportions of different immune subsets at specific time-points during treatment were also assessed to identify differences in the immune composition between R and NR. Gene transcription studies were performed to detect differentially regulated genes/pathways.MethodsFor all correlative studies, screening sample (Pre-ASCT), post - ASCT, end of cycle 3 (EOC3) and EOC6 were utilised. Peripheral blood mononuclear cells (PBMC) from post-ASCT, EOC3 and EOC6 were processed to determine acetylation status of lysine residues (K) in histone 2A - H2A (K5), H2B (K5), H3 (K9, K14, K18 and K56), H4 (K5, K8, K12 and K16) and tubulin in CD3+ (T-lymphocytes) utilising flow cytometry. Changes were measured using mean fluorescence intensity (MFI) of the geometric mean. The MFI from a sample was normalized to the respective control sample and expressed as a fold difference. PBMC from all 4 time points were also assessed for proportions of different subsets of T-lymphocytes (T-helper, cytotoxic T-cells and T-regs) and natural killer (NK) cells using flow cytometry. Whole bone marrow (BM) aspirates were collected in PAXgene blood RNA tubes and extracted RNA was subjected to 100 bp paired-end total RNA sequencing.Results and conclusion:Panobinostat-induced acetylation in T-lymphocytes was significantly higher at EOC3 (H2B K5, H3 K9, H3 K14, H3 K56, H4 K8, H4 K12, H4 K16 (p<0.01)) and EOC6 (all histones assessed, p<0.05) in R, with no differences in NR at EOC3 and EOC6 compared to post-ASCT, suggesting that acetylation could be utilised as a pharmacodynamic biomarker of response to panobinostat. Assessment of T-cell proportions in this annotated sample set indicated that the ratio of CD4/CD8 T-cells was significantly higher at EOC3 in R compared to NR (p<0.01), consistent with published literature, which demonstrates that the ratio of CD4/CD8 is significantly lower in patients with poor prognostic tumors. Furthermore, MFI of CD57 in cytotoxic cells was higher in EOC3 and EOC6 compared to post-ASCT in R while it was downregulated in NR (p<0.05), also suggesting an increased cytotoxic activity for T-cells in R. Absolute numbers of NK cells increased significantly in NR at EOC3 and EOC6, while this was significantly reduced in R at EOC3 and EOC6 compared to post-ASCT (p<0.01). This indicates that immunogenic cell death of NK cells appear to occur in the R in response to treatment. Total RNA-sequencing studies indicated that minimal changes were induced between pre-ASCT and EOC3 in NR (6 genes), while comparison of R (Pre-ASCT) to EOC3 and EOC6 demonstrated that 2482 and 2814 genes, respectively, were regulated. A total of 984 genes were differentially regulated in both EOC3 and EOC6 compared to pre-ASCT. Notably, HMGB1, a critical mediator of immunogenic cell death was significantly upregulated in R at EOC3 compared to pre-ASCT sample, consistent with our observations of NK cell proportions. Further analysis revealed that a total of 13 genes were differentially regulated in R at Post-ASCT, EOC3 and EOC6 compared to the pre-ASCT. In conclusion, we present potential pharmacodynamic biomarkers of clinical benefit (acetylation of histones, CD4/CD8 ratio and changes in immune subsets) using PBMC. A subset of genes that can be assessed from whole BM biopsy are also being assessed to determine a genetic signature for response.Disclosures Spencer: Janssen: Honoraria, Research Funding; Amgen: Consultancy, Honoraria, Research Funding.↵* Asterisk with author names denotes non-ASH members.
Original languageEnglish
Article number1823
Number of pages1
JournalBlood
Volume130
Issue numberSuppl 1
Publication statusPublished - 7 Dec 2017

Cite this