Immune Crosstalk Profiling of Non-Hodgkin Lymphoma (NHL) Tumor Immune Microenvironment Identifies Two Clusters Based on T Cell Exhaustion, Correlating with Different Clinical Outcomes

• Published in: Blood 2024-11, Vol.144, p.6251-6251
• Main Authors: Medina, Daniel, Palomo, Laura, Navarro, Víctor, García, Cristina, Nieto, Juan Camilo, Carabia, Júlia, Pagès Geli, Carlota, Acha, Pamela, Pujadas, Gemma, Palacio, Carlos, Gallur, Laura, Castellvi, Josep, Abrisqueta, Pau, Bosch, Francesc, Crespo, Marta
• Format: Article
• Language: English
• ISSN: 0006-4971
• DOI: 10.1182/blood-2024-205108

Introduction: Tumor cells use different mechanisms to evade immune surveillance, leading to the remodeling of the tumor immune microenvironment (TIM). TIM is considered to play an important role in the pathogenesis and response to treatment in non-Hodgkin lymphoma (NHL). Herein, we hypothesized that different NHL subtypes share common TIM patterns that can predict patient outcomes. Therefore, our objective was to characterize the crosstalk between cancer cells and T cells across different NHL subtypes and its clinical relevance. Methods: Biopsies from 64 NHL patients and 16 reactive/non-malignant lymphoid tissues (rLT) were analyzed by flow cytometry. Immune profiling included major lymphocyte subpopulations, as well as the expression of inhibitory (PD-L1, CD112, CD155) and co-stimulatory ligands (CD58, 4-1BBL, ICOSL, OX40L) and HLA-I in B cells, the expression of inhibitory (PD-1, TIGIT, TIM3, LAG3, CD244, CD160) and co-stimulatory receptors (4-1BB, ICOS, OX40) on T cells, T cell differentiation status, Tregs and T-helper cells. Overall survival (OS) and progression-free survival (PFS) was available for 59 out of 64 patients. Results: The characterized tumors included follicular lymphoma (FL) (n=35), diffuse large B cell lymphoma (DLBCL) (n=14), marginal zone lymphoma (MZL) (n=7), mantle cell lymphoma (MCL) (n=4), Burkitt's lymphoma (BL) (n=3) and high-grade B cell lymphoma (HGBCL) (n=1). The immune profiling showed a higher percentage of CD58+ and HLA-I negative NHL cells compared to healthy B cells in rLT. Moreover, we observed a significantly higher number of patients with NHL cells expressing PD-L1, CD112 and ICOSL compared to rLT. These results confirmed the loss of surface HLA-I and the expression of PD-1 and TIGIT ligands as common immune evasion mechanisms in NHL. CD4+ T cells were reduced in NHL compared to rLT, while CD8+ T cell levels were not distinct. Both CD4+ and CD8+ T cells exhibited depletion of the naïve pool and elevated expression of PD-1, TIGIT, LAG3 and TIM3. The percentage of CD4+ and CD8+ T cells co-expressing PD-1 and TIGIT was also significantly higher in NHL. An increased expression of the co-stimulatory receptor 4-1BB was observed in CD4+ and CD8+ T cells from NHL tumors. Additionally, Treg frequency was higher in NHL while there were no significant differences in T-helper subpopulations. Overall, our data showed a skewing of T cells towards an effector profile with increased expression of exhaustion markers in NHL. Unsupervis