Deletion of the protein tyrosine phosphatase PTPN22 for adoptive T cell therapy facilitates CTL effector function but promotes T cell exhaustion

Background: Adoptive cell therapy (ACT) is a promising strategy for treating cancer, yet it faces several challenges such as lack of long-term protection due to T cell exhaustion induced by chronic TCR stimulation in the tumor microenvironment. One benefit of ACT, however, is that it allows for cellular manipulations, such as deletion of the phosphotyrosine phosphatase non-receptor type 22 (PTPN22), which improves CD8+ T cell antitumor efficacy in ACT. We tested whether Ptpn22KO cytolytic T cells (CTLs) were also more effective than Ptpn22WT CTL in controlling tumors in scenarios that favor T cell exhaustion.

Methods: Tumor control by Ptpn22WT and Ptpn22KO CTL was assessed following adoptive transfer of low numbers of CTL to mice with subcutaneously implanted MC38 tumors. Tumor infiltrating lymphocytes were isolated for analysis of effector functions. An in vitro assay was established to compare CTL function in response to acute and chronic restimulation with antigen-pulsed tumor cells. The expression of effector and exhaustion-associated proteins by Ptpn22WT and Ptpn22KO T cells was followed over time in vitro and in vivo using the ID8 tumor model. Finally, the effect of PD-1 and TIM-3 blockade on Ptpn22KO CTL tumor control was assessed using monoclonal antibodies and CRISPR/Cas9-mediated knockout.

Results: Despite having improved effector function at the time of transfer, Ptpn22KO CTL became more exhausted than Ptpn22WT CTL, characterized by more rapid loss of effector functions, and earlier and higher expression of inhibitory receptors (IRs), particularly the terminal exhaustion marker TIM-3. TIM-3 expression, under the control of the transcription factor NFIL3, was induced by IL-2 signaling which was enhanced in Ptpn22KO cells. Antitumor responses of Ptpn22KO CTL were improved following PD-1 blockade in vivo, yet knockout or antibody-mediated blockade of TIM-3 did not improve but further impaired tumor control, indicating TIM-3 signaling itself did not drive the diminished function seen in Ptpn22KO CTL.

Conclusions: This study questions whether TIM-3 plays a role as an IR and highlights that genetic manipulation of T cells for ACT needs to balance short-term augmented effector function against the risk of T cell exhaustion in order to achieve longer-term protection.