Direct Evidence for a Chronic Antigen Driven T Cell Response In CLL Lymph Nodes

Abstract 915

In this study we put forward a novel approach to investigate the T cell compartment of chronic lymphocytic leukaemia (CLL). The role of T cells in the pathobiology of CLL has become the subject of much research due to compelling evidence that CLL is derived from antigen experienced B cells that are subject to ongoing activation. The malignant proliferation occurs particularly within the pseudofollicles of lymph nodes where the CLL cells are in close proximity with T cells and stromal cells of the microenvironment. Our group has previously shown, by multi-parameter confocal microscopy of paraffin embedded CLL lymph node biopsy sections, that almost all proliferating CLL cells are in close contact with CD4+ CD25+ FOXP3- T cells. In order to gain access to this T cell compartment and gain further insight into the phenotype and role of these T cells we have utilized the technique of lymph node fine needle aspiration (LN-FNA). Patients with a diagnosis of CLL and who have lymphadenopathy accessible to LN-FNA were invited to participate in the study. A total of 21 LN-FNAs were collected from 18 patients. The median cell yield obtained was 4.1 x10⁶ (range 1–13×10⁶). Multi-colour flow cytometry using antibody panels of up to 7 flurochrome combinations was performed using a BD FACSCanto II (BD Biosciences) and analyzed with FlowJo (Treestar) and GraphPad Prism (GraphPad Software Inc.). Uniquely we have been able to compare the phenotype of the T cells derived from the LN-FNA with cells from the peripheral blood (PB) of the same patient taken at the same time. A panel consisting of CD3 (APC-Cy7), CD4 (PerCp-Cy5.5), CD8 (Pacific Blue), CD25 (FITC), CD40L (PE), CTLA-4 (Pe-Cy7), PD-1 (APC) was designed to investigate T cell subsets and activation markers. We found a statistically significant higher number of CD3+ T cells in the lymphocyte gate of CLL LN-FNAs (20.75% +/−13.80) compared to PB (6.25% +/− 4.06); p=0.0152. Additionally, of the CD3+ cells, there were more CD4+ cells in the LN-FNA (72.25% +/− 8.10) compared to PB (59.38% +/− 15.73); p=0.005, and conversely less CD8+ cells in the LN-FNA (17.43% +/− 6.50) compared to PB (27.43% +/− 14.18); p=0.03. The percentage of CD4+ and CD8+ cells expressing CD25, CD40L and CTLA-4 was not found to be significantly different. A further panel consisting of CD4 (PerCP-Cy5.5), CD8 (Pacific Blue), CD45RA (Pe-Cy7), CD62L (APC-Cy7), CCR7 (PE), CD28 (APC) and CD25 (FITC) was used to differentiate between effector memory, central memory and naive T cells. In the LN-FNA CD4+ cells there was a trend towards an increased effector memory phenotype compared to the peripheral blood (60%+/− 20 vs. 29%+/−13). One of the most striking observations in this study was the significantly increased mean fluorescence intensity of the exhaustion marker PD-1 on the CD4+ LN-FNA cells (1041+/− 1130) compared to PB (415.9+/− 487); p=0.0002. This was also mirrored in the CD8+ cells, LN-FNA (1342+/− 1219) compared to PB (465.8+/− 658.7); p=0.0015. Additionally significantly less PD-1 expression was observed in both CD4+ and CD8+ T cells from age matched normal controls confirming the relevance of these findings. Importantly, we have reproduced these observations using multi-parameter immunofluorescence microscopy on paraffin embedded CLL and normal lymph node biopsy samples. Lymph node sections were antigen retrieved and fluorescently labelled with combinations of CD4, CD23, Ki67 as well as PD-1 and then visualised using a Zeiss Axiovert LSM 510 microscope. Our results support the hypothesis that T cells in CLL are chronically stimulated by antigen within the lymph node. In addition, the results from the PB imply that either PD-1 expression is down-regulated on these cells or that the T cell pools derived from LN-FNA and PB are distinct from one another. To investigate this further we are currently examining the T cell receptor repertoires from these two compartments. In addition, it is known that expression levels of PD-L1, a ligand of PD-1, correlates with poor prognosis in many solid tumours and we are therefore investigating the role of this ligand pairing in CLL. In order to understand how this disease persists and progresses it is necessary to unravel the complex network of signals that mediate the survival and proliferation of the CLL cells. Our data suggest that T cells are part of this complex cellular crosstalk and blocking receptor ligand interactions, such as PD-1 and PD-L1, may offer potential future therapeutic targets.