T cell receptors (TCRs) enable T cells to specifically recognize mutations in cancer cells1–3. Here we developed a clinical-grade approach based on CRISPR–Cas9 non-viral precision genome-editing to simultaneously knockout the two endogenous TCR genes TRAC (which encodes TCRα) and TRBC (which encodes TCRβ). We also inserted into the TRAC locus two chains of a neoantigen-specific TCR (neoTCR) isolated from circulating T cells of patients. The neoTCRs were isolated using a personalized library of soluble predicted neoantigen–HLA capture reagents. Sixteen patients with different refractory solid cancers received up to three distinct neoTCR transgenic cell products. Each product expressed a patient-specific neoTCR and was administered in a cell-dose-escalation, first-in-human phase I clinical trial (NCT03970382). One patient had grade 1 cytokine release syndrome and one patient had grade 3 encephalitis. All participants had the expected side effects from the lymphodepleting chemotherapy. Five patients had stable disease and the other eleven had disease progression as the best response on the therapy. neoTCR transgenic T cells were detected in tumour biopsy samples after infusion at frequencies higher than the native TCRs before infusion. This study demonstrates the feasibility of isolating and cloning multiple TCRs that recognize mutational neoantigens. Moreover, simultaneous knockout of the endogenous TCR and knock-in of neoTCRs using single-step, non-viral precision genome-editing are achieved. The manufacture of neoTCR engineered T cells at clinical grade, the safety of infusing up to three gene-edited neoTCR T cell products and the ability of the transgenic T cells to traffic to the tumours of patients are also demonstrated.

No abstract

Leukocyte-endothelial adhesion is a critical early step in chronic vascular inflammation associated with diabetes, emphysema, and aging. Importantly, these conditions are also marked by abnormal subendothelial matrix crosslinking (stiffness). Yet, whether and how abnormal matrix stiffness contributes to leukocyte-endothelial adhesion remains poorly understood. Using a co-culture of human monocytic cells and human microvascular endothelial cells (ECs) grown on matrices of tunable stiffness, we demonstrate that matrix stiffness exerts biphasic control over monocyte-EC adhesion, with both matrix softening and stiffening eliciting a two-fold increase in this adhesive interaction. This preferential endothelial adhesivity on softer and stiffer matrices was consistent with a significant increase in α-actinin-4-associated endothelial ICAM-1 clustering, a key determinant of monocyte-EC adhesion. Further, the enhanced ICAM-1 clustering on soft and stiff matrices correlated strongly with an increase in Rho activity and ROCK2 expression. Importantly, inhibition of Rho/ROCK activity blocked the effects of abnormal matrix stiffness on ICAM-1 clustering and monocyte-EC adhesion. Thus, these findings implicate matrix stiffness-dependent ICAM-1 clustering as an important regulator of vascular inflammation and provide the rationale for closely examining mechanotransduction pathways as new molecular targets for anti-inflammatory therapy.

In infectious disease, polyclonal T cell responses against immunodominant epitopes drive successful immune responses. In cancer, neoepitopes (neoE) derived from non-synonymous mutations, similarly to the immunodominant epitopes in viral infections, are potentially highly immunogenic because the T cells recognizing these antigens are not subjected to the mechanisms of tolerance. Indeed, early studies support that neoE derived from non-synonymous mutations are the primary target of T cell responses induced by immune checkpoint blockade therapy and have been successfully targeted by adoptively transferred T cell therapies (ACT) in multiple cancer histologies. However, there is limited knowledge on the immunodominance and evolution of neoE's, or the clonality of the T cell responses against these neoE. Furthermore, little is known regarding the correlation between the presence and expansion of neoE-specific T cells and the clinical response to immunotherapy in patients. To characterize the neoE-specific T cell responses induced after immunotherapy, we collected peripheral blood mononuclear cells (PBMCs) over time (longitudinally) and established expanded tumor infiltrating lymphocyte cultures (TILs) and autologous tumor cell lines from the patient's tumor biopsies. We performed whole exome and RNA sequencing of the tumor and normal tissue controls for the computational prediction and ranking of patient-specific neoEs. We then generated a library of capture reagents consisting of the patient HLA class I molecules loaded with predicted neoE (Peng et al. AACR 2019) and isolated neoE-specific T cells from the patients' PBMC or TIL samples. Once isolated, the paired neoE-specific TCR alpha and beta chains (neoTCR) were obtained by single cell sequencing. For functional characterization of the neoTCRs, healthy donor primary human T cells were modified to express the neoTCR using CRISPR-based, non-viral precision genome engineering by replacing the endogenous TCR with the respective neoTCR (Jacoby et al., AACR 2019, Sennino et al., AACR 2019). These gene-edited T cells were then used in co-culture experiments with the patient autologous cell lines. We analyzed T cell responses in three patients (PT1, PT2, and PT3) with metastatic melanoma receiving immunotherapy. PT1 had a fast and durable anti-tumor response to anti-PD-1 therapy. Sequencing identified 2556 somatic coding mutations. A library of 243 neoE-specific pMHC capture reagents across 3 HLA types, HLA-A*03:01, A*24:01, and C*12:03 was generated and used for screening of PBMCs or TILs derived from multiple longitudinal time points. Several hundred neoE-specific T cells were isolated. Importantly, this neoE-specific T cell response was comprised of 17 different neoE-specific T cells clones targeting only 5 different HLA-neoE complexes supporting the immunodominance hypothesis. On the other hand, PT2 and PT3 showed marginal responses to immunotherapy. Patient two progressed after being treated with anti-PD1. This patient had 24 somatic coding mutations. Seventeen neoE-HLA reagents across 3 HLAs, B*35:03, C*12:03, and C*08:01 were generated and used to capture neoE-specific T cells from TILs and PBMCs. While 14 different TCRs targeting 7 HLA-neoE complexes were identified from expanded TILs, no neoE-reactive T cells were captured from the peripheral blood. PT3 presented with progressive disease after being treated with local TVEC. This patient had 61 somatic coding mutations; 78 neoE-specific pHLA capture reagents covering HLA-A*02:01, A*03:01, B*07:02, C*05:01, and C*07:02 were generated and used to screen for neoE-specific T cells in the patient's TIL and PBMCs. In contrast to PT2, 2 different neoTCRs targeting the same HLA-neoE complexes were isolated from PBMCs, but none from TILs. To further characterize the T cell responses from patients that responded or did not respond to immunotherapy, we generated 18 separate T cell products, each expressing a different neoTCR isolated from PT1, PT2 and PT3. For PT1, we characterized 14 different neoTCRs specific for neoE's in the mutated IL8, PUM1 and TPP2 genes. All 14 T cell products displayed specific cytotoxicity against the matched autologous melanoma cell line established from a biopsy of patient one (50-75% tumor growth inhibition compared to melanoma cell line growth in co-culture with a mismatched control TCR, 96 hour assay using a product to target ratio (P:T) of 1:1, p < 0.000001 for each comparison). No cytotoxic effect against an unmatched human melanoma cell line was observed. Furthermore, neoE TCR T cells upregulated 4-1BB and OX-40, secreted IFNγ, IL-2, TNFα, and IL6, and induced T cell proliferation and degranulation. Again, no unspecific T cell activation was observed when T cells were co-cultured with unmatched targets. Interestingly, precision genome engineered T cell products expressing neoTCRs identified from patients that did not respond to therapy (PT2 and PT3), also potently killed autologous tumor cells. Four neoTCRs were studied (2 TCR for PT2 and 2 TCRs for PT3), and three of them showed specific cytotoxicity against the matched autologous melanoma cell line (50-100% tumor growth inhibition compared to melanoma cell line growth in co-culture with a mismatched control TCR, 96 hour assay using P:T 5:1, p < 0.05 for each comparison). Additionally, upon co-culture with the matched melanoma cell line, but not against an unmatched melanoma cell line control, neoE TCR T cells upregulated 4-1BB and OX-40, secreted IFNγ, IL-2, TNFα, and IL6, and induced T cell proliferation and degranulation. These data demonstrate that even patients that did not respond to immunotherapy harbor neoTCRs that, when expressed in ‘fresh' T cells, are able to kill the autologous tumor cell lines. Using newly developed techniques to isolate and capture neoE-specific single T cells, as well as non-viral gene editing, we isolated and characterized neoE-specific T cells that can recognize the cancer cells and induce an anti-tumor response. We also studied the neoE immunodominance and TCR clonality over time of the natural T cell repertoire that induce anti-tumor responses to ICB therapy. Our results show that in a patient with a good response to anti-PD-1, there is a polyclonal response that targets a limited number of neoE-HLA complexes (2% of the neoE tested in the case of patient one) highlighting the immunodominance of these epitopes. Interestingly, different T cell clonotypes targeting the same mutations evolve over time, suggesting functional differences amongst the TCRs. In addition, our results demonstrate that even patients that did not respond to these therapies harbor neoE-specific T cells, as we were able to isolate neoE-specific T cells that recognized and killed patient-derived cancer cells. This suggests that even in patients that do not respond to immunotherapy, neoE-specific TCRs can be isolated and could be potentially used for personalized ACT. Finally, our results also show how non-viral precision genome engineering can successfully redirect T cells to neoE-expressing tumors, enabling the personalized ACT. Citation Format: Cristina Puig-Saus, Barbara Sennino, Bhamini Purandare, Duo An, Boi Quach, Songming Peng, Huiming Xia, Sidi Zhao, Zheng Pan, Yan Ma, Justin Saco, Sameeha Jilani, Christine Shieh, Katharine Heeringa, Olivier Dalmas, Robert Moot, Diana Nguyen, William Lu, Kyle Jacoby, Andrew Conroy, Jasreet Hundal, Malachi Griffith, Stefanie Mandl, Alex Franzusoff, Antoni Ribas. Landscape analysis of neoepitope-specific T-cell responses to immunotherapy [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr NG11.

No abstract