Adoptive Cell Therapy Targeting Neoantigens: A Frontier for Cancer Research

Abstract

Adoptive cell therapy (ACT) is a kind of immunotherapy in which T cells are genetically modified to express a chimeric antigen receptor (CAR) or T cell receptor (TCR), and ACT has made a great difference in treating multiple types of tumors. ACT is not perfect, and it can be followed by severe side effects, which hampers the application of ACT in clinical trials. One of the most promising methods to minimize side effects is to endow adoptive T cells with the ability to target neoantigens, which are specific to tumor cells. With the development of antigen screening technologies, more methods can be applied to discover neoantigens in cancer cells, such as whole-exome sequencing combined with mass spectrometry, neoantigen screening through an inventory-shared neoantigen peptide library, and neoantigen discovery via trogocytosis. In this review, we focus on the side effects of existing antigens and their solutions, illustrate the strategies of finding neoantigens in CAR-T and TCR-T therapies through methods reported by other researchers, and summarize the clinical behavior of these neoantigens.

Keywords: CAR-T; TCR-T; adoptive cell therapy; neoantigen; neoantigen screening.

Figure 1

The illustration of the disposition of adoptive cell therapy side effects. (A) An epitope is expressed on CAR-T or TCR-T cells, which can be recognized by epitope-targeted antibodies, thus leading to CAR-T or TCR-T cells being killed through antibody-mediated cytotoxicity. (B) The addition of a dimerizing drug activates iCasp9 signaling and leads to apoptosis. (C) The reduced affinity of TCR/CAR can enhance specificity and reduce off-tumor on-target cytotoxicity. (D) The construction of switchable CAR is an effective method to reduce the side effects of CAR-T therapy; the strategy is to separate the antigen-binding domain from the signal transduction domain through a peptide neoepitope (PNE) that works as a bridge between the antigen-binding domain and the signal transduction domain. (E) On binding one tumor antigen, the synNotch receptor undergoes a conformational change that leads to the release of a transcription factor, which in turn drives the expression of a CAR-T antigen for another inhibitory antigen. (F) Inhibitory CAR (iCAR) dampens the T cell response when a normal antigen is encountered. (G) OR gate CAR is comparable to bispecific CARs.

Figure 2

The workflow of neoantigen screening using whole-exome sequencing (WES) combined with mass spectrometry (MS). WES is conducted to identify the tumor-specific mutations, together with mass spectrometry-based mutated peptide detection, to compare the mutated proteins with those in the transcriptome-generated FASTA database. Mutated proteins will be predicted in silico to narrow down the target mutations. Predicted peptides can be expressed by a patient's APCs, where they are processed and presented in the context of a patient's MHC. The coculture of the patient's autologous T cells with these APCs can be used to identify the mutations processed and presented by APCs. The identification of individual mutations for tumor recognition is applicable because T cells express activation markers such as OX40 or CD137 when they recognize the cognate target antigen.

Figure 3

Neoantigen screening through an inventory-shared neoantigen library. The TCGA and COSMIC databases are used to mine high-frequency mutant genes in nine types of human malignant solid tumors to construct the neoantigen library. Data from the patient who underwent targeted sequencing were compared with those in the neoantigen library to identify the specific neoantigen in the patient's tumor cells. The selected neoantigen will be pulsed and presented by APCs in the context of the patient's MHC. The coculture of the patient's T cells with the APCs can be used to identify the neoantigen presented by the APCs. The detection of immunogenic neoantigens for tumor recognition is performed by ELISPOT-based cytokine detection secreted from activated T cells when they recognize the target neoantigen.

Figure 4

The illustration of neoantigen screening via trogocytosis. The private mutations are identified by exome and RNA sequencing from tumor samples, and the neoepitope ligand can be predicted and presented by an MHC multimer panel to be applied to the patient's autologous T cells. The gene of neoepitope-reactive TCR can be verified and transduced into a T cell line as an effector cell line. Meanwhile, the neoepitope single-chain trimer (SCT) cDNA library can be generated and transduced into K562 or other cancer cell lines to construct the target cell library. The coculture of effector cells with the target cell library will be used to identify the neoepitope because the membrane protein from the T cell line will transfer to the specific target cell whenever the neoantigen matches the T cell with a specific TCR. After two rounds of flow cytometry-based cell sorting, the neoantigen for tumor recognition can be isolated based on the specific membrane protein transferred to neoepitope-transduced target cells.