Abstract:Chimeric antigen receptor (CAR) T cells have shown promising results in patients (pts) with B cell malignancies, yet approximately 60% of pts with diffuse large B cell lymphoma (DLBCL) will relapse. Therefore, future efforts are needed to improve the outcomes of these pts. A total of 18 pts with relapsed/refractory B cell malignancies, DLBCL (n=17) and ALL (n=1), were enrolled on a phase 1b/2 study (NCT02772198) of locally produced CD19 CAR T cells. The median age was 40.5 years (range, 23-70). All pts received a lymphodepleting preparative regimen with cyclophosphamide and fludarabine, followed by intravenous infusion of autologous CD19 CAR T cells with a CD28 costimulatory domain. Dosing of CAR T cells was 1-1.5 million CAR+ cells per kg. Clinical response was determined at 28 days following cell administration. Blood samples obtained prior to the lymphodepleting conditioning and at days 7, 14, 21, 30 and 60 after CAR T administration were collected. Cell phenotype was assessed on peripheral blood mononuclear cells (PBMCs) using multiparametric flow cytometry. The manufactured CAR T products (n=9) were also subjected to immunophenotypic analysis. Clinically, 11 of 18 pts (61%) responded to CAR T therapy, 6 (33%) with complete response (CR), and 5 (28%) with partial response (PR). Analysis of manufactured CAR T products (n=9) revealed high CD3+ purity (99%), composed with CD4+ (28%) and CD8+ (72%) T cells. Phenotypic characterization demonstrated marked heterogeneity in the percentages of naïve, central memory, effector memory and effector cells within the CD4+ and CD8+ subsets in the CAR T product. In order to assess the homing potential of CAR T cells, expression of chemokine receptors was evaluated in the product cells. High CXCR3 expression was detected (77% and 96% positive within CD4+ and CD8+ subsets, respectively), indicating high migratory capacity of CAR T cells toward inflamed tissues with high levels of CXCL9 and CXCL10 ligands. Furthermore, co-expression of CXCR4 (56% and 54% positive within CD4+ and CD8+ cells) suggests increased homing ability of the manufactured CAR T toward CXCL12-rich bone marrow microenvironment and lymph nodes. Interestingly, higher CCR7 expression (32% vs 8.5%) and lower CCR6 levels (15% vs 28%) were detected on CD8+ CAR T cells from responding pts who achieved CR and PR (n=6) in comparison to non-responders (n=3), suggesting that less differentiated phenotype together with increased trafficking of CAR T to lymphoid tissue corresponds with improved clinical outcome. Additionally, we assessed the immunoregulatory and senescent/exhausted phenotype in CAR T products. Low percentage of CD4+CD25+CD127- Treg cells (13.5%) was detected, with no correlation to clinical response. However, significantly higher frequency of exhausted CD57+CD39+CD28- cytotoxic CD8+ cells stand out as signature population in CAR T products of non-responders in comparison to CR pts (37% vs 9.5%, p<0.02). It is known that immunosuppressive environment affects CAR T cell activation, dampening anti-tumor responses. Therefore, we next evaluated the frequency and kinetics of regulatory T cells and myeloid suppressor cells in the peripheral blood of the CAR T treated pts. Notably, responding and non-responding pts presented distinct Treg patterns. Pts achieving CR demonstrated modest and delayed increase in Treg cells, reaching maximal frequency of 23% Treg out of CD4+ cells at day 21 post CAR T infusion, declining to basal low levels (12.5%) at day 30. In contrast, non-responders possessed rapidly increasing percentage of Treg cells (35%) at day 14 post-infusion. In line with this finding, notable increase in proportion of immunosuppressive CD11b+CD14+ myeloid cells expressing CD163, CD206 and MERTK M2 markers was detected in blood of non-responders, while pts achieving CR experienced transient increase in myeloid suppressor cells at day 7 that went back to normal levels at day 14. Overall, these results elucidate in part the mechanisms of CAR T traffic, immunosuppressive responses as well as induction of T cell senescence/exhaustion that most probably downregulate CAR T effectiveness as observed in non-responding pts. It is conceivable that deeper understanding of these processes will help not just establishing surrogate markers predicting clinical responses but may lead to new strategies for restoration of CAR T activation, thereby improving their efficacy and patient's prognosis. Disclosures No relevant conflicts of interest to declare.

Abstract B033: Excessive Chimeric Antigen Receptor signaling correlates with poor clinical outcomes

Abstract A042: Quantitative multiplex immunoprecipitation reveals distinct protein interaction networks responsible for discrepant in vivo activity exhibited by two second generation CD22-targeted chimeric antigen receptors

Abstract 4017: CAR T cell-mediated cytotoxicity and cytokine release in response to varying levels of antigen expression on target cells

Abstract PO1-26-05: Animal Study Compares CAR T cell Exhaustion & Ability to Kill Low Antigen expressing breast cancer cells among three CAR constructs including one with 1XX mutations

Upregulation of Senescent/Exhausted Phenotype of CAR T Cells and Induction of Both Treg and Myeloid Suppressive Cells Correlate with Reduced Response to CAR T Cell Therapy in Relapsed/Refractory B Cell Malignancies

Abstract 882: Harnessing single cell multi-omics data to identify predictors of clinical outcome in CD19 CAR T cell therapy

Abstract 3609: Chimeric cytokine receptors increase memory phenotypes and improve expansion and efficacy of CAR-T cell therapies

CD28 and 41BB Costimulatory Domains Alone or in Combination Differentially Influence Cell Surface Dynamics and Organization of Chimeric Antigen Receptors and Early Activation of CAR T Cells

Abstract 44: A novel, codon-based method to modulate off-tumor toxicity of chimeric antigen receptor (CAR) T cells

INSPIRED Symposium Part 1: Clinical Variables Associated with Improved Outcomes for Children and Young Adults treated with Chimeric Antigen Receptor T cells for B cell Acute Lymphoblastic Leukemia

Abstract 3995: A novel chimeric Fas signal redirect receptor enhances the durability of anti-tumor activity and serial killing potential of CAR T cells

Abstract 6332: Evaluating the relationship of affinity, functional avidity, and in vivo potency in KIR-CAR T cells

Determinants of response and resistance to CD19 chimeric antigen receptor (CAR) T cell therapy of chronic lymphocytic leukemia

Abstract LB070: In vitro modeling of antigen-driven exhaustion in human CAR T cells

Abstract B043: CD3ζ ITAM diversity determines chimeric antigen receptor signaling and function

Abstract 4002: Investigating CAR-T cell efficacy and activation in the disseminated NALM6-luc human B-cell acute lymphoblastic leukemia model

Abstract 3867: Identification of novel biomarkers of immune toxicity from CAR T-cell therapy using ultrasensitive NULISATM proteome technology

Abstract 6324: Antagonism-enforced braking system to enhance CAR T cell therapeutic specificity

Abstract 4010: Assessing impacts of tumor burden and CAR-T cell dosage on the toxicity and efficacy profile of CD19 CAR-T therapy in a PBMC-humanized mouse model

CAR19 monitoring by peripheral blood immunophenotyping reveals histology-specific expansion and toxicity.

Abstract 1148: 1XX mutations slow CAR T cell signaling and increase in vivo persistence