T cell-based immunotherapy (TCBI) pioneers a new stage of tumor treatment in recent years [1,2]. However, the clinical response rates from the TCBI remain unsatisfactory [3], which were stymied by stumbling blocks from the insufficient immunogenicity, the T cell dysfunction from T cell exhaustion and the immune evasion from programmed death-1/programmed death-ligand 1 (PD-1/PD-L1) pathway. Up to date, the vast majority of researches mainly focus on modulating tumor microenvironment for improving the efficacy of TCBI [1,[4], [5], [6]], ignoring the straightforward approach via pharmacologically modulating T cells. In view of the molecular and cellular biology of T cells, the pharmacological modulation of T cells might be a powerful approach to enhance the efficacy of TCBI.

For TCBI, naive T cells are first activated by antigen presentation from dendritic cells (DCs) [7]. In this stage, antigen is essential for triggering immunogenicity. A subset of immunogenic neoantigens such as classical melanoma-associated antigen 3 (MAGE-A3) and gp100 were previously utilized for designing therapeutic tumor vaccines, but these predefined antigenic epitopes also restrict the target of immune responses [8]. In contrast, whole tumor lysates (TL) comprise a variety of known or unknown tumor-associated antigens (TAA) that are able to trigger a broader range of tumor-specific T cells, enabling to avoid the possibility of off-target immune response [9]. Particularly, the TL is also capable of recruiting DCs for further amplifying the immunogenicity [10]. Therefore, the TL provides a powerful antigen tool for designing advanced immunotherapeutic platforms with strong induced immunity. Particularly, the local delivery via hydrogel enables to significantly augment the drug bioavailability in solid tumors while minimizing systemic exposure as well as reducing side effects in the normal tissue. Unfortunately, up to date, there is no report on the design of TL as the gelator to realize an injectable hydrogel. Therefore, it is urgent to develop a TL-based gelator to construct an injectable hydrogel for inducing robust immunogenicity.

Upon triggering immunogenicity, T cells proliferate and differentiate into effector T cells and memory T cells [11]. Tumor-infiltrating effector T cells are the principal immunological cells that kill tumor cells. However, their high expression of programmed cell death protein 1 (PD-1) immune checkpoint activates PD-1/programmed death-ligand 1 (PD-L1) pathways for inducing immune evasion, thus significantly compromising the killing capability of T cells [2]. Until now, most of the research was focused on utilizing the specific protein antibodies to disrupt PD-1/PD-L1 pathway, where poor permeability in tumor tissues and unfavorable immune-related defects were identified [12]. Recently, glycogen synthase kinase 3 (GSK-3) was identified as the key upstream kinase regulating PD-1 expression in the CD8+ T cells [13], and pharmacological suppression of GSK-3 could activate the T-bet-modulated downregulation of PD-1 expression. This genetic modulation of PD-1 expression thus offers a new approach to reverse PD-1/PD-L1-based immune evasion.

Nevertheless, T cell differentiation has a “parallel” program—exhaustion [14]. The exhausted T cells can be divided into progenitor exhausted T cells and terminally exhausted T cells according to different epigenetic states [15]. Progenitor exhausted T cell subpopulations can differentiate into T cells with effector capacity after PD-1 checkpoint blockade [16,17]. However, the terminally exhausted T cell subpopulation formed at the end of differentiation, highly expresses multiple inhibitory receptors and loses effector functions, even after PD-1 checkpoint blockade. A recent study indicated that the accumulation of depolarized mitochondria in CD8+ T cells facilitates the phenotypic and epigenetic reprogramming into terminally exhausted T cells [18]. Consequently, the removal of depolarized mitochondria in T cells and the enhancement of mitochondrial fitness are expected to be a promising strategy to inhibit the differentiation into terminally exhausted T cells.

Based on the above, an injectable tumor lysates-based hydrogel is reported to potentiate TCBI via pharmacologically modulating the body's own T cells. The oxidized sodium alginate (OSA)-modified tumor lysates (O-TL) were, for the first time, designed as the gelator. After the mix solution of O-TL, nicotinamide adenine dinucleotide (NAD) precursor nicotinamide riboside (NR), and GSK-3 inhibitor SB415286 (SB) was intratumorally injected, the [email protected](N + S) hydrogel formed quickly owing to the chelation of OSA of O-TL with Ca2+ in the tumor environment (Scheme 1). The hydrogel was constructed by the chemically modified tumor lysates, enabling the hydrogel as a robust antigen reservoir to induce strong immunogenicity. The SB encapsulated in hydrogel enabled potent suppression of PD-1 expression for regulating immune evasion. Furthermore, the hydrogel-encapsulated NR stimulated powerful mitophagy of T cells for reversing their differentiation into terminally exhausted T cells. Therefore, our injectable hydrogel creates a robust immune niche within the tumor to address the insufficient immunogenicity, the T cell dysfunction from T cell exhaustion and the immune evasion from PD-1/PD-L1 pathway, which otherwise severely compromises the efficacy of TCBI clinically. Our strategy via pharmacologically modulating T cells offers a conceptually new approach and opens a new avenue for TCBI.