BET inhibition targets pathogenic interstitial macrophages to limit antibody mediated fibrosis in bronchiolitis obliterans chronic gvhd

Abstract

Abstract Introduction: Chronic graft-versus-host disease (cGVHD) is a major cause of late mortality in allogeneic-hematopoietic stem cell transplantation with bronchiolitis obliterans syndrome (BOS) among the most lethal treatment-resistant complications. Despite approval of ibrutinib, ruxolitinib, belumosidil, and axatilimab, clinical response in moderate and severe BOS patients remains poor, highlighting the need for new therapies. Bromodomain and Extra terminal (BET) proteins are epigenetic readers driving oncogenic and inflammatory transcriptional programs in multiple cell types. We reported that BET inhibition reduced acute GVHD pathogenicity via suppression of the IL23/T helper 17 axis, without impairing the graft-versus-leukemia effect. Thus, we hypothesized that BET inhibition would be a feasible therapeutic option by restraining the direct inflammatory T cells as well as macrophage (mac) polarization to profibrotic (M2) phenotype known to be critical to cGVHD pathophysiology. In a BOS model, pulmonary function tests showed mice treated with BETi improved lung function and decreased collagen deposition. Here, we present our results identifying novel mechanisms driving cGVHD pathogenesis that are targeted by BET inhibition that ameliorates disease. Methods: We used an established BOS cGVHD model wherein B.10.BR (H2k) recipients conditioned with chemotherapy (120mg/kg x 2 cytoxan) and total-body irradiation (800cGy) received 7.5e6 T cell depleted bone marrow (TCD-BM) cells from allogeneic donor CD45.2+ C57BL/6 (B6, H2b) mice +/- 7.5e5 CD45.1+ B6 splenocytes (Day 0). TCD-BM only recipients that do not develop disease served as controls. Recipients with established cGVHD/BOS were treated days 28-~55 (study end) with vehicle or the BET inhibitor (BETi, OPN-51107, 10mg/kg 3x/weekly), that also has anti-tumor effects on malignant B cells. Spectral flow cytometry, histopathology, immunofluorescence (IF) and single-cell RNA sequencing (scRNA-seq) of lung tissue were performed to evaluate effector cells associated with BOS pathology. Results: BETi decreased circulating CXCR5+ B cells and CD4+ T helper cells (p<0.001) as well as plasma levels of the B cell chemoattractant CXCL13 that binds CXCR5 (p<0.001). BETi treated mice displayed reduced splenic activated CD44+CD62L- T follicular helper (Tfh) cells (p<0.0001) and restored IL10+ T follicular regulatory (Tfr):Tfh ratios (p<0.05). BETi did not impact germinal center B cell numbers (p=0.56) but did reduce alveolar plasma B cells (p<0.05). Plasma ELISA showed mice with cGVHD had increased circulating pathogenic IgG1/IgM, but not IgG2c, which was attenuated under BETi (p<0.01). Lung IF staining confirmed reduced pathogenic IgG1/IgM deposition in BETi mice compared to vehicle controls (p<0.05). Antibodies (Abs) in cGVHD can be both allo- and auto-antigen (Ag) reactive. To test this, we harvested lung tissue sections from B6 Rag2-null mice (lacking Abs) for staining with serum from TCD-BM, cGVHD and BETi mice. cGVHD serum had increased Abs reactive to auto (H-2b)-Ags, that was lower in the BETi cohort (p<0.001). scRNA-seq revealed 22 distinct cell types in the BOS lung. In cGVHD mice, GSEA analysis revealed upregulation of pathways for migration and cell adhesion pathways in alveolar macrophages and those related to bacterial defense response in interstitial macrophages (IM), all downregulated in BETi mice (p<0.05). cGVHD mice had enriched anti-inflammatory Arginase1 and profibrotic Tgfb1 expression (p<0.001) in tissue-resident alveolar macrophages (TRAM) and IMs, which was significantly reduced in BETi treated mice. Spectral flow cytometry confirmed that BETi targeted lung-infiltrating M2 macs, through selective diminishment of CD206+ M2 IMs (p<0.01) but not TRAMs (p=0.66). We found that FcgR expressing M2 IMs were increased in cGVHD, with reduced levels in BETi treated mice (p<0.01). Reduced FcgR-Ab interactions in M2 macs could mechanistically contribute to the observed reduction in TGF-b expression and crosstalk to collagen depositing fibroblasts. Conclusion: We identify BET proteins as a key regulator of inflammation in BOS. BETi suppresses Tfh cells, thereby reducing Ab production and deposition. Within the lung, BETi selectively targets M2 IM macs to reduce FcgR and TGF-b expression, thereby regulating collagen deposition and fibrosis. These findings reveal a previously unrecognized mechanistic axis driving cGVHD and highlight BET inhibition as a promising therapeutic strategy.