Bromodomain/Extra-Terminal Inhibition (BETi) and Ruxolitinib: Immune Profiling Insights from Human Samples in Acute Gvhd

Introduction

Acute-graft versus host disease (GVHD) is a T-cell mediated disorder associated with significant transplant related mortality. Patients non-responsive to front-line steroids and ruxolitinib have poor outcomes, necessitating the need for newer drugs. Pre-clinical studies have shown that inhibition of bromodomain and extra-terminal domain (BET) proteins affects inflammatory gene transcription, reduces T cell proliferation, decreases inflammatory cytokine production, and impairs dendritic maturation both in vivo and in vitro murine models of acute GVHD. Until now, the effects of BET inhibition in acute GVHD human patient samples have never been reported. Herein, we demonstrate the impact of pharmacological inhibition of BET, through OPN-51107, on the immune profile of human samples previously treated with ruxolitinib in vitro.

Methods

We conducted an in-depth review of our database including medical records to identify patients with acute GVHD, adhering to established guidelines and definitions. Medical records were examined by two physicians to categorize patients as either sensitive or refractory to ruxolitinib based on their clinical responses to treatment. Once patients were identified, mononuclear cells (MNC) stored in the OSU GVHD biorepository were obtained. To maintain uniformity, samples were selected from both ruxolitinib sensitive (rux-sen) and refractory (rux-refr) groups, with at least 90-120 days on treatment. A total of 6 samples, 3 in each group, were analyzed in the experiment. The collected samples were activated using Dynabeads ^TM Human T-Activator CD3/CD28 for 48 hours at 37^oC. The samples were plated in RPMI media to test under 3 conditions – unstimulated, stimulated with and without BET inhibitor (BETi) OPN 51107 (500nM) (dose consistent with in vitro models) added simultaneously during stimulation. Samples were stained with antibodies as per the OMIP-042 protocol for flow cytometric analysis of major lymphocyte and myeloid subsets. We separately analyzed rux-sen and rux-refr groups. Supernatants were collected for ELISA to assess Interferon-gamma (INF-g) secretion. All analyses were performed using GraphPad Prism 10.0 and differences between variables were identified using two-sided student’s t-test and ANOVA. With a sample size of 3 per group, we have 80% power to detect a fold change of 2.5 in the mean between two groups, assuming the significance level (alpha) at 0.05.

Results

The flow cytometry analysis demonstrated lower CD3+CD20- T-cell population across all the three conditions in rux-refr vs rux-sen (11.58% vs 64.20%). Subset analysis revealed the proportion of CD4+T cells were higher in the rux-refr group vs rux-sen (mean 63.67% vs 40.59%, p=0.002) while CD8+ T cells were significantly lower (mean 1.4% vs 19.17%, p=0.0002). While CD4+T cells were increased, CD4+CD25+CD127-/lo Tregs were lower in rux-refr group vs rux-sen, possibly due to treatment effect (0.64% vs 10.84%, p=0.006). Percentages of Th1 and Th17 CD4+T helper cells were lower with greater skewing towards Th2 population in the rux-refr group. The total percentage of non-B/T cells (including NK cells, monocytes, dendritic cells and granulocytes) were elevated in the rux-refr group (mean 87.7% vs 32.01%) with granulocytes predominating in both rux-sen and rux-refr.

There was no difference in rux-refr samples treated in vitro with BETi. In rux-sen patient samples, treatment with BETi upon stimulation suppressed total (CD3+) T cells (mean 16.8% vs 3%). CD8+ T cells showed greater suppression compared to CD4+T cells (7.4% vs 5.0%). The Tfh population showed suppression to BETi but wasn’t statistically significant.

ELISA for IFN-g secretion also indicated greater suppression with BETi in the rux-sen group compared to rux-refr (mean 121068 vs 305 pg/ml, p<0.05).

Conclusion

We demonstrate significant T-cell population differences between ruxolitinib sensitive and resistant groups highlighting baseline discrepancies. The data suggests a highly immunosuppressive environment in the resistant group likely due to prior treatments. While cell subset expression was otherwise largely similar, BET inhibition drastically affected IFN-g production. BET inhibition with OPN51107 showed greater efficacy in modulating immune responses in the sensitive group, highlighting its therapeutic potential and need for further studies.