Background: We and others have published that antigen-stress induced functional exhaustion of chimeric antigen receptor (CAR) T cells can be rescued by addition of anti-PD1 agents. To avoid the need for multiple administrations of anti-PD1 agent, we developed CAR T-cell intrinsic PD1 dominant negative receptor (PD1 DNR). Herein, we investigated the anti-tumor efficacy of cell-extrinsic versus intrinsic anti-PD1 strategies. Methods: Human T cells transduced with CD28 costimulated mesothelin-targeted CAR T cells (M28z) with or without anti-PD1 agent, and M28zPD1DNR CAR T cells were investigated against mesothelin-expressing cancer cells with inducible or constitutive PDL1 over expression. In vitro, cytotoxicity upon single and multiple antigen stimulation and proliferation, inhibitory receptor expression (PD1, TIM3 LAG3) were tested. In vivo, tumor burden regression kinetics and median survival were determined in an orthotopic model of pleural mesothelioma. Results: Following single antigen stimulation, both CAR T cells exhibited equivalent cytotoxicity. Following multiple antigen stimulations (antigen stress test), M28z PD1DNR CARs showed sustained and relatively higher cytotoxicity and proliferation compared to M28z CAR T cells, even when combined with anti-PD1 agent. PD1, TIM3 and LAG3 expression upregulation was noticed in both CAR T cells. CAR T cell therapy combined with extrinsic PD1 blockade or intrinsic PD1 DNR demonstrated enhanced and sustained tumor regression and prolonged survival (median survival 66 and 60 days) compared to CAR T cell therapy alone (27.5 days). In mouse model with constitutive PDL1 overexpressing cancer, similar trend is observed (74, 69 versus 28 days). Conclusions: A single dose of M28z CAR T cells with cell-intrinsic PD1DNR demonstrated equal anti-tumor efficacy compared to multiple doses of extrinsic anti-PD1 agent administration in models with inducible and constitutive high-PDL1 expressing cancer with no on-target, off-tumor toxicity with either strategy. This data provides the rationale for our ongoing phase II trial of combination therapy with M28z CAR T cells and anti-PD1 agent, and upcoming M28zPD1DNR CAR T cell trial in 2020.