In 2013, the University of Pennsylvania reported the first case of a child with acute lymphoblastic leukemia who achieved complete remission after CAR-T therapy
[1]. Subsequently, CD19-targeted CAR-T has been widely used in the treatment of hematologic tumors. In long-term follow-up, patients with ALL and NHL showed significant remission after CD19-CAR-T treatment
[2][3][4]. For patients with multiple myeloma, anti-BCMA CAR-T had the highest effective rate. One study showed that the median follow-up time was 417 days, the overall response rate (ORR) was 88.2%, and the 1-year overall survival (OS) was 82.3%
[5]. On the basis of these encouraging results, in 2017, the U.S. Food and Drug Administration approved two CAR-T immunotherapies, Kymriah and Yescarta, primarily for the treatment of ALL and NHL
[6][7]. In 2020, with the results of the KTE-X19 CAR-T (Tecartus) therapy, the FDA approved a third CAR-T therapy for adult patients with MCL
[8]. Overall, CAR-T therapy has shown ideal clinical efficacy in patients with hematologic malignancies. There are some adverse events, such as cytokine release syndrome, recurrence of disease due to immune escape of tumor antigens, reduction of blood cells, adverse reaction of central nervous system, infection, and ineffective platelet transfusion. However, these side effects are usually manageable
[9].
CAR-T therapy has limited efficacy in solid tumors. Hou et al.
[10] reported that the overall effectiveness of CAR-T cells in solid tumors was only 9%, with an overall response rate of 11% in hepatobiliary and pancreatic tumors, 12% in neurologic tumors, and 12% in other tumors, hence far less effective than in hematologic tumors.
There are major impediments to using CAR-T for solid tumors. Target specificity is the most important; however, no tumor-specific antigen has been found in solid tumors. Most target antigens are tumor-associated antigens that express at low levels in normal tissues, leading to the risk of off-target effects and even death
[11]. Second, in the treatment of solid tumors, the primary killing effect of CAR-T cells is achieved only when the CAR-T cells migrate from the peripheral blood to the tumor site, especially to the interior of the primary tumor and other metastatic lesions
[12][13][14]. However, solid tumors often have abnormal vascular beds and high levels of interstitial fibrosis that inhibit delivery of CAR-T cells or drugs to deep tumors
[15][16]. More importantly, there is a complex immunosuppressive microenvironment in solid tumors. In detail, depleted T cells often express inhibitory receptors, including programmed death receptor 1 (PD-1), cytotoxic T lymphocyte-associated antigen 4 (CTLA-4), T cell immunoglobulin mucin-3 (Tim-3), and lymphocyte activation gene 3 (LAG-3). These inhibitory receptors bind their corresponding ligands to induce apoptosis of T cells by different mechanisms, thereby down-regulating the immune response
[17][18][19]. Immunosuppressive factors secreted by tumor cells, Tumor-associated macrophages (TAMS) and regulatory T cell (Treg) in the tumor microenvironment, such as IL-10, IL35, and TGF-β, are key factors in T cell failure. TGF-β not only contributes to the activation of Tregs and tumor angiogenesis cytokines, but also TGF-β induces upregulation of CTLA-4 expression by Treg
[20][21]. In addition, tumors consume a large amount of glucose and essential amino acids, and tumors produce many metabolites such as fatty acids and lactic acid that cause a hypoxic acidic microenvironment, which may reduce the cellular function of CTL
[22].
Several methods have been used to improve CAR-T cell function in the tumor microenvironment. Leonid showed that depleted CAR-T was reactivated by application of PD-1 antibody
[23]. John et al. found increased efficacy of CAR-T by using PD-1 blocking antibodies combined with HER2-CAR-T in HER2+ sarcoma cells
[24]. Researchers also engineered CAR-T to secrete PD-1 scFv locally to block the PD-1/PD-L1 pathway, which not only prolonged the survival time of mice, but also avoided adverse events caused by systemic administration of PD-1 antibodies
[25]. A more promising approach is the use of gene-editing technology, which enables modification of single or multiple genes. Ren et al.
[26] used CRISPR-Cas9 technology to create PD-1 and CTLA-4 double knockout CAR-T cells, which improved the activity of the CAR-T cells. Later, they knocked out the TCR β2 microglobulin to achieve a still better antitumor effect
[27]. However, other evidence showed that proliferation of CAR-T cells was inhibited after PD-1 receptor silencing or knock out
[28]. Therefore, more experiments are needed to assess the feasibility of this scheme.
2. Targets of CAR-T Therapy in Prostate Cancer
CAR-T studies in prostate cancer are currently focused on preclinical studies, with a small number of phase I trials conducted to assess safety. Finding specific targets for prostate cancer is the first step in the development of effective CAR-T therapy. The ideal tumor target should be expressed exclusively in cancer cells, and CAR-T can generate specific immune responses in tumor tissues without damaging normal tissues. Prostate-specific antigen (PSA) is the most commonly used marker for the diagnosis of prostate cancer. PSA is secreted by prostate vesicles and epithelial duct cells, and can be detected in serum normally with a concentration of less than 4 μg/mL
[29]. When prostate tissue is destroyed, PSA is released into the blood through capillaries
[30], However, these secreted target antigens are not suitable as CAR-T targets because they cannot be localized to target cells, so it is critical to find highly specific membrane surface antigens. Currently, three main targets of CAR-T therapy for prostate cancer research are prostate specific membrane antigen (PSMA), prostate stem cell antigen (PSCA), and epithelial cell adhesion molecule (EpCAM).
Prostate specific membrane antigen (PSMA) is one of the most common targets. The gene is located on chromosomes 11p11-12 and expressed as a 750-amino acid II type intrinsic membrane protein. PSMA is over-expressed on the membrane of prostate cancer and endothelial cells of tumor neovasculature
[31]. It was also found in other normal tissues, such as salivary gland, brain, small intestine, renal tubular epithelium and breast epithelium
[32]. In a mouse model of prostate cancer constructed from PC3 cells, Ma et al.
[33] compared the efficacy of first-generation and second-generation CAR-T cells targeted to PSMA and normal T cells. They found that 75% (6/8) of the second-generation CAR-T group achieved complete remission, significantly superior to the first-generation CAR-T (1/8) and normal T cells (0/8). Zuccolotto et al.
[34] introduced CD28 as a costimulatory molecule in PSMA-CAR to construct the second generation of CAR-T cells. In the treatment group, tumor volume gradually decreased after 1 week and tumors almost disappeared after 3 weeks. The survival time of the mice was also prolonged (SCID mice: 54 d in the control group, 74 d in the experimental group; NOD-SCID mice: 60 d in the control group, >150 d the experimental group). In addition, in order to weaken immunosuppressive factors, researchers added anti TGF-β to PSMA targeted CAR-T and found that the tumor killing ability of CAR-T was significantly improved
[35]. In 2016, the results of a phase I clinical trial showed that two of the five patients with prostate cancer achieved partial remission with reduced serum PSA, and no toxicity caused by PSMA CAR-T cells was observed
[36]. In 2018, Kloss et al.
[37] built an inhibitor of TGF-β receptor expression PSMA-CAR-T cells, which improved the CAR-T effect; a phase I trial (ClinicalTrials.gov Identifier: NCT03089203) was then launched to evaluate CAR-T in mCRPC patients. Currently, some phase I/II clinical studies are ongoing, but no published data are available (ClinicalTrials.gov Identifier: NCT04249947, NCT04633148, NCT04429451).
Prostate stem cell antigen (PSCA) is a tumor-related antigen discovered by Reiter et al.
[38] in a study of prostate cancer gene expression. The protein was named as a prostate stem cell antigen due to its 30% homology with stem cell antigen. PSCA has some functions of stem cells, such as cell self-renewal, proliferation and adhesion, and is involved in tumor genesis and development
[39]. The expression rate of PSCA in normal prostate tissues is about 60–70%, and more than 90% in prostate cancer tissues
[38]. Further studies have shown that PSCA expression gradually increased from normal prostate cancer, prostate intraepithelial tumor, hormone dependent, hormone independent prostate cancer, and bone metastases of prostate cancer
[40][41]. Therefore, PSCA is an ideal target in advanced or metastatic diseases. Hillerdal et al.
[41] constructed a third generation CAR-T cell targeting PSCA for the treatment of prostate cancer in mice. The vitro experiments showed that when CAR-T cells specifically bound to target cell PSCA, they secreted a large amount of IL-2 and interferon γ, which promoted CTL proliferation and killed tumors effectively
[42]. Priceman et al.
[42] confirmed the advantage of PSCA-CAR-T in the model of bone metastasis of prostate cancer, and they found from a selection of different costimulatory molecules that 4-1 BB enabled PSCA-CARs to have higher disease control ability and to exhibit better T cell persistence compared with CD28 as a costimulatory molecule
[43]. Currently, two phase I/II clinical trials are underway to evaluate the efficacy and safety of CAR-T targeting PSCA in patients with advanced prostate cancer (ClinicalTrials.gov Identifier: NCT03873805, NCT02744287).
The third effective target is the epithelial cell adhesion molecule (EpCAM), also known as CD326, which belongs to the adhesion molecule family. The EpCAM gene is located on chromosome 2p21 and encodes a 40 kDa type I transmembrane glycoprotein. EpCAM functions as an epidermal cell adhesion molecule, participating in signal transduction and cell proliferation
[44][45]. EpCAM is associated with oncogenesis and is strongly expressed in various types of human epithelial carcinoma, such as lung, breast, prostate, ovarian, cervical, and colorectal cancer (CRC), and the expression of EpCAM is related to the degree of disease, suggesting that it may be a promising target for cancer diagnosis and treatment
[46]. Some studies suggest that EpCAM can be used as a predictor of prostate cancer; it has an important activity in CaP proliferation, invasion, metastasis, and chemo-/radio-resistance associated with the activation of the PI3K/Akt/mTOR signaling pathway
[47]. Using EpCAM as TAA, researchers built EpCAM-specific chimeric antigen receptors. Although the EpCAM on PC3 cells was expressed at a low level, in the transfer model, EpCAM CAR-T still inhibited the growth of tumors and increased the survival time of mice. Thus, EpCAM may be better for high proliferation and metastasis of cancer cells
[48]. However, the expression of EPCAM in prostate cancer is inconsistent. Some studies have shown that EpCAM expression has no significant correlation with Gleason score and progression after radical treatment in prostate cancer
[49]. Another study confirmed that the overexpression of EpCAM was significantly associated with high Gleason grade by tissue microarray method. Therefore, the merits and demerits of EPCAM as a target for CAR-T therapy in prostate cancer need further confirmation. One clinical trial has begun to evaluate the safety and efficacy of CAR-T cells that target EpCAM in patients with EpCAM-positive cancer (ClinicalTrials.gov Identifier: NCT03013712).