T-Cell Profiling Shows Marked Differences in Gene Expression Between Tumour Infiltrating (TIL) and Peripheral Blood (PB) Antibody-Bead Selected T-Cells From Fresh Lymph Node and PB Samples in Follicular Non-Hodgkin's Lymphoma (FL): Possible implications for Immunomodulatory Therapy

FL is one of the commonest B-cell lymphomas accounting for 25–30% of all newcases of non-Hodgkin's lymphoma. Median survival ranges from 6–10 years with a constant annual rate of relapse and death. The cellular immune system through T-cells and macrophages/monocytes (MMs) has an important role in the response to therapy and long term remissions. Evidence for this stemmed from increased incidence and the poor response rates and survival in patients with inherited or acquired T-cell defects, and the graft-versus-leukaemia effect of allogeneic stem cell transplants. The number of tumour infiltrating lymphocytes (TILs) and MMs appear to predict clinical response and outcome in FL. A recent National Cancer Institute study looked at the prediction of survival in FL based on the molecular features of TILs with positive findings. However, this study used frozen material which could have changed the signatures, T-cells were not positively selected and would have been mixed with other non B-cells, negatively selected cells were stimulated with CD3 which could have changed the gene expression profile, the study looked only at the predictive value in relation to prognosis and did not look at PB T-cell (PBT)profile.

In the first stage reported here, we looked at positively selected TILs, PBT and MMs gene expression profile signatures in fresh lymph node samples (LNs) of histologically confirmed grades 1–3 FL in 14 patients and compared the profile results with similar cells positively selected from PB samples taken at the same time. The analysis was controlled with 4 histologically proven reactive hyperplasia LNs. Tissue samples were digested with a cocktail of enzymes, ficol separated and then positively selected for CD2, CD19, CD14 in that order using Dynabeads (Dynall) initially then Microbeads (Miltenyl Biotech) subsequently. CD2 was used to avoid T-cell receptor stimulation and change of expression with CD3. Some isolated T-cells were initially expanded in culture for purity assessment only by flow cytometry. T-cells and MMs were liquid nitrogen frozen in trisol whilst B-cells were stored in DMSO. RNA was extracted at the same time and analysed using Affymetrix Microarray Chips. Statistical analysis used 3 parameters to define significant change in expression: fold change of >1.5, p value of > 1×10⁻³and FDR of >0.5. Samples were analysed as paired and pooled as some cases did not yield adequate paired samples.

T-cell paired (10) samples showed 97 over (41) or under (56) expressed genes in TILs compared to PBT whilst pooled samples (14) showed 778 over (380) or under (398) expressed genes. The top over expressed genes (10 fold plus) using both methods of analysis were CXCL13 a B-cell chemotactic gene, IGJ which helps in assembling IgG and IgM, CTLA4 a regulatory molecule, CD200 which delivers an inhibitory signal for the macrophage lineage and several heat shock protein (HSP) genes. The top under-expressed genes (10 fold plus) include genes that regulate T-cell adhesion, migration and direct and indirect cytotoxicity such as a number of FC receptor genes needed for antibody dependant cell cytotoxicity, killer cell lectin-like receptor genes, fractalkine genes involved in the adhesion and migration of leukocytes, and granzyme, granulosin, lysosyme and other genes involved in T-cell cytotoxocity. MMs: Only pooled samples were analysed as the extracted cell numbers were too small to have paired analysis. Many more (3239) genes were over (1494) or under (1745) expressed in LN MMs compared to PB MMs. The top over expressed genes include genes that support B cell growth such as IGJ, BLNK and C4orf7, the B cell chemotactic gene CXCL13, signal transduction inhibitors and HSPs. The top underexpressed genes include those that regulate cell adhesion and survival, cell proliferation and migration and extracellular matrix assembly. Others include FC receptor, IL-1 and compliment genes necessary for cell cytotoxicity, and thrombomodulin genes.

We believe this is novel data suggesting that TILs and tissue MMs appear to support B-cell growth in FL whilst inheriting several defects in direct and indirect cell cytotoxicity. Downregulating the significantly overexpressed and upregulating the under expressed genes through targeted therapy may provide the basis of non-chemotherapy immunomodulatory treatment of B-cell lymphomas.

Rassam:BMS: Honoraria; Johnson and Johnson: Unrestricted research grant Other; ROCHE: Honoraria. Mufti:Celgene: Consultancy, Research Funding.