A Population of Suppressive Monocytes Inhibiting T Cell Proliferation and Dendritic Cell Differentiation in Relapsed Non-Hodgkin's Lymphoma

Abstract

Despite advances in treatments for patients with non-Hodgkin's lymphoma (NHL), the relapse rates remain high and 40% of diffuse large B-cell NHL (DLBCL) patients die of disease. New therapies to augment the host anti-tumor immune response are needed. Reports of graft-versus-lymphoma responses in patients who have received allogeneic hematopoietic cell transplant indicate a role for cellular immunotherapy. However, these patients have variable levels of immunodeficiency which may impact the efficacy of cellular therapy. To study this we first evaluated the cellular immune status of patients with relapsed NHL. Proliferation of peripheral blood mononuclear cells (PBMNC) stimulated with anti-CD3/CD28 beads was reduced by more than 3.5 folds for patients with DLBCL (n = 3) compared to that of age-matched healthy donors (n = 5; p = 0.02). Removal of monocytes from PBMNC by use of anti-CD14 immunomagnetic beads restored proliferation to that of healthy donors. Further, monocytes from these patients were deficient in stimulating allogeneic T cell proliferation by 3 folds compared to monocytes from healthy donors (n = 3 NHL; n = 8 normal; p < 0.01). Peripheral blood from 12 NHL patients (9 DLBCL; 1 grade 3 follicular lymphoma; 2 composite) and 12 age-matched healthy donors were characterized by flow cytometry to determine the phenotype of these suppressive monocytes. There was no difference in the % monocytes in the blood between NHL patients and healthy donors; however, NHL patients had elevated % monocytes with a suppressive phenotype (CD14+HLA-DRneg) compared to normals (NHL 38.9 ± 4.93%; normal 8.3 ± 2.15; p < 0.0001). This phenotype is distinct from other myeloid suppressors (Lin-CD33+HLA-DR-) or non-classical monocytes (CD16+), neither of which was different in numbers between NHL and normal donors. This suggests that the CD14+HLA-DRneg monocytes are responsible for the observed T cell suppression. To further characterize the function of these cells, we cultured purified CD14+ monocytes from NHL and compared their differentiation capacity with those from normal donors. The percentage of CD14+HLA-DRneg monocytes in initial ex vivo culture was inversely correlated with the percentage of pure, mature dendritic cells (mDC) generated with TNF-a and PGE2 as maturation factors (CD80+CD83+; n = 9; p = 0.015). As CpG oligonucleotides are also capable of immune stimulation and have some anti-tumor activity in clinical trials, we investigated the effect of CpG on mDC differentiation in NHL. In healthy donors, maturation of monocytes with CpG yielded highly pure mDC (90.7 ± 2.15%, n = 3). However, preliminary results of mDC yield from monocytes of NHL patients matured with CpG was only 22.6 ± 11.2% (n = 2). This data suggests alternative signaling pathways of these suppressive monocytes. Preliminary analysis of a proteome array for 46 kinase phosphorylation sites from 37 proteins in 2 NHL and 3 healthy donors suggest changes in phosphorylation of 17 protein kinases for CD14+HLA-DRneg compared to CD14+HLA-DR+ cells. Functional correlation of these protein kinase phosphorylation changes is needed to definitively target the pathway characteristic of CD14+HLA-DRneg monocytes. Finally, we have identified a serum-free culture method that can consistently generate highly pure mDC from monocytes of NHL patients (93.0 ± 3.6%, n = 9) and is readily adaptable to good manufacturing practice for clinical use in immunotherapy. Taken together we have described for the first time a population of CD14+HLA-DRneg monocytes that is a significant source of immunosuppression in NHL patients and are beginning to target methods of overcoming this suppression.