Introduction

Pancreatic cancer is highly aggressive and lethal malignancy. It is seventh most common cause of cancer-related death worldwide [1] and third most in the United States of America (USA) [2]. The 5 year survival rate is 7.2% and estimated mortality is 40,560 in 2015 in USA [2]. The two main types of tumor are pancreatic ductal adenocarcinoma (PDA) and endocrine tumors. PDA constitutes 90% of all cases, and is characterized by highly aggressive and malignant behavior, delayed diagnosis, resistant to treatment, and high mortality [3, 4]. Generally, at the time of diagnosis, PDA already had local or distant metastasis which make them ineligible for surgical resection, and palliative care is their only option with 5 year survival rate in single digits [5]. Despite recent advances in therapeutic options like FOLFIRINOX (Folinic acid, 5 Fluorouracil, Irinotecan, Oxaliplatin), median overall survival still remains poor [6]. Therefore, novel therapeutic options need to be devised to deal with this highly lethal and dreadful tumor. Immunotherapy recently showed promising results in various tumors and pancreatic cancer can be a potential target for them. Although initial studies of immune checkpoint inhibitors as single agents failed to show therapeutic response in PDA with advancing field and better understanding of immune system in relation to pancreatic cancer, there are still promising chances to have effective immunotherapy for advanced PDA [7].

Chimeric antigen receptor (CAR) T cell therapy is a type of Adoptive T cell therapy (ACT), where genetically engineered T cells are used for lysis and degradation of cancer cells. It’s a novel therapeutic modality which has shown outstanding results in hematological malignancies and currently research is going on in solid tumors; PDA can be a potential target for CAR T cell therapy. Immune system generally detects damaged or genetically modified tumor-potential cells and kill them. Cancer cells have the potential to evade these immune mechanisms and can modify immune system to help cancer cells in their proliferation, invasion, and metastasis, known as immune escape. There are various mechanism behind it; some of them are defect in antigen processing and presentation [8, 9], loss of immunogenic cancer antigens, secretion of inhibitory cytokines like IL-10 and TGF-β, genetic modification in tumor oncogenes and tumor suppressor genes, overexpression of CTLA-4, and PD-1 immune checkpoints. Therefore, if we can reverse these immune escape phenomena of cancer cells and activate immune cells against cancer antigens, it can be promising therapy for cancer. Immunotherapy is based on this principle. Conventional activation of T cells is a complex process and is dependent on antigen processing, antigen presentation, and major histocompatibility complex(MHC)- mediated activation to peptide antigens. But CAR T cells uses light and heavy chains of immunoglobulins for antigen binding ectodomain, and therefore, it doesn’t require antigen processing, MHC dependence [10, 11], and can have immunogenic response to non-peptide antigens also [12]. Therefore, CAR T cells can evade various immune escape mechanisms of cancer cells and can have cancer specific immune reactions. There are four main components of CAR. Ectodomain which is extracellular cancer antigen binding domain, consists of single chain variable fragment (scFv) derived from heavy and light chains of antibody. Spacer domain, also known as hinge domain, is formed from IgG4 and CD8 molecule and connects extracellular domain and transmembrane domain. Intracellular domain also known as signaling domain consists of CD3ζ and it initiates and triggers antigen specific immune response. CAR T cell therapy is autologous customized therapy, where T cells are collected from patients’ own peripheral blood with leukapheresis; these T cells are transduced with genetic material for CAR with the help of lentiviruses and retroviruses. Transduced T cells are genetically amplified with in vitro culture systems and each culture system is specific for a particular type of cell line like effector T cell, memory, or naïve T cell. Therefore, we can control the phenotype and can generate required subset of T cells [13]. CAR T cells are then transfused back into patients and they can recognize and bind to tumor specific antigens (TAA). This start a cascade of reaction and lead to phosphorylation of immune receptor tyrosine-based activation motifs (ITAMs) and, therefore, activation of T cells [14]. Cytotoxicity of CAR T cells is mediated by activation of death receptor with binding of fas/fas-ligand and by release of perforin enzyme which induces lysis of tumor cells [15, 16]. On the basis of intracellular domain, CAR T cells are divided into 4 generations. In first generation, CD3ζ is intracellular signaling domain, in second and third generation, there are additions of co-stimulatory molecules like 4-1BB (CD137), CD 28, OX40 which enhances the T cell response; in second generation, there is one co-stimulatory molecule and in third generation, there are two. Fourth generation of CAR T cells have addition of cytokine releasing genes (IL-12 and IL-15) which improves the survival of CAR T cells in immunosuppressive tumor microenvironment [17, 18].

Immunosuppression by PDA

PDA is highly aggressive, malignant, and treatment resistance cancer with extremely high mortality likely due to evasion of immune surveillance by PDA cancer cells. Although PDA is a T cell rich tumor, it induces local or systemic immune dysfunction to prevent cancer cells’ death from cytotoxic immune cells [19, 20]. PDA downregulate the MHC class 1 molecule and inhibit antigen insertion into MHC groove, therefore, inhibit antigen cross-presentation [21]. PDA cells has non-functional Fas receptor to avoid Fas-mediated apoptosis and along with that they express Fas ligand which can induce apoptosis in immune effector cells [22]. PDA secrete IL-10 and transforming growth factor-beta (TGF-β) which are immunosuppressive molecules and inhibit the function of cytotoxic T cells [23, 24]. PDA also secrete Indoleamine 2,3-dioxygenase (IDO) enzyme which catalyzes the breakdown of tryptophan to kynurenine, which is an important nutrient for the survival of lymphocytes, and thereby, suppress antitumor T cell response, along with that, IDO also increase the T regulatory cells (Tregs) in lymph nodes which aggravates immunosuppressive environment [25, 26]. PDA increases the expression of PD-L1 on their surface, which are immune checkpoints; therefore, decrease the activation of T cells [27]. The stromal matrix of PDA comprises of cellular and acellular components like myofibroblasts, pancreatic stellate cells, fibroblasts, blood vessels, extracellular matrix, immune cells, cytokines, and growth factors, and all of them contribute to tumor proliferation, invasion, and metastasis [28]. Along with excessive desmoplasia, PDA also lack vasculature which leads to impaired perfusion and limited drug delivery. The tumor microenvironment of PDA consists of dense fibrotic stroma with variety of inflammatory cells, infiltration with immunosuppressive cells such as myeloid derived suppressive cells (MDSCs), tumor-associated macrophages (TAMs), Tregs, and various other mechanism to avoid tumor cell detection makes tumor microenvironment highly immunosuppressive [29]. As consequence of all the above mentioned mechanisms. As there are multiple steps of immune escape which can be potential targets for immunotherapy also, this makes PDA an immune privilege tumor. Multiple immune escape steps should be targeted simultaneously to have a sustained therapeutic response and single agent immunotherapy should be avoided, as it will lead to failure of immune response due to other immunosuppressive factors, as we have seen in immune checkpoint inhibitors in PDA. Therefore, PDA is a promising candidate for immunotherapy like CAR T cell therapy and these immunosuppressive factors are two edged sword as they can act as obstacles, but with further advances in therapeutic options, they can be excellent targets for different forms of immunotherapy also.

Antigenic targets for CAR T cell therapy

Mesothelin (MSLN)

MSLN is a normal cell surface glycoprotein which is expressed on the surface of mesothelial cells, lining the peritoneum, pericardium, and pleura. It is formed as a precursor protein which is cleaved by the action of endoprotease enzyme to release N-terminal known as megakaryocyte potentiating factor (MPF) and mature MSLN remain behind attached to membrane [30]. As per the distribution of MSLN, it likely functions as a differentiating factor for mesothelial cells but the exact biological function is unknown. MSLN is overexpressed in various cancer cells and its overexpression has been correlated with poor prognosis in various tumors like ovarian cancer [31], PDA [32, 33], cholangiocarcinoma [34, 35], lung adenocarcinoma [36, 37], and triple negative breast cancer [38]. As most of the cancer specific antigens which are used as targets for solid tumors are also present on normal cells, there is nonspecific toxicity. But MSLN is only expressed on dispensable tissue, and therefore, the chances of nonspecific toxicity are less.

As there is overexpression of MSLN in pancreatic cancer, different therapeutic approaches have been evaluated to target MSLN like MSLN vaccine, antibody drug conjugate, chimeric antibody, and CAR T cell therapy. Recent data suggest that new immunotoxins like LMB-100 (old name RG 7787) might have synergistic effects with standard therapy in management of PDA [39]. Initial clinical trials of CRS-207 (Listeria monocytogene-mesothelin vaccine) also have shown promising outcome as compared to standard chemotherapy [40]. Therefore, MSLN can be a potential target for CAR T cell therapy for PDA patients. Other modalities of targeting MSLN are SS1P (Recombinant anti-MSLN immunogenic toxin) [41], BAY 94-9343 (antibody drug targeting MSLN) [40], DMOT4039A (humanized IgG1 anti-MSLN mAb) [42], and BMS-986148 (Anti-MSLN MAb-cytotoxic drug conjugate) [43]. In a clinical trial of MSLN-targeted CAR T cell therapy, a patient with PDA, after receiving 3 weeks of intravenous CAR T cell therapy had a stable disease. After finishing the therapy, FDG PET/CT imaging also showed decrease in uptake value at all sites of disease transiently. Ascitic fluid analysis on day 3 and 15 of CAR T cell therapy have shown decrease in cancerous cell expressing MSLN and c-Met [44]. The transient immune reaction is a promising result and MSLN CAR T cell therapy can be a potential therapeutic tool for management of advanced pancreatic cancer.

CD24

One of the major reasons for highly aggressive and invasive nature of PDA is presence of self-renewable, chemoresistant, and multipotent cells in tumor, known as cancer stem cells (CSC). CSC are known to be responsible for tumor proliferation, invasion, metastasis, and recurrence [45,46,47,48]. Initially, Li et al. found highly aggressive subpopulation of cancer cells in PDA, and transfer of these cells in immunocompromised mice lead to initiation, rapid proliferation, and metastasis of PDA [49]. It has been found that these cells have surface markers like CD44, CD24, and epithelial specific antigen. These surface markers are present only in 0.2–0.8% of cancer cells, but they can increase the tumorigenic potential more than 100 folds as compared to other cancer cells [49, 50]. Therefore, CD24 can be a potential target to control the tumorigenic potential of PDA cancer cells. Maliar et al. [51] successfully eliminated the CD24 + tumor cells with second generation CD24 CAR T cell therapy in human orthotopic pancreatic cancer xenograft model. In this study, total 107 CAR T cells were injected intravenously or intratumorally on 3 or 4 alternating days and then mice were given IL-2 intraperitoneally, two times a day for 10 days. This study showed that as CD24 CAR T cell therapy was more efficacious as compared to HER2 CAR T cell therapy in curing mice, although CD24 expression is markedly less on PDA cancer cells in comparison to HER2. Therefore, this study showed that CD24 can be a potential target for CAR T cell therapy of PDA.

Carcinoembryonic antigen (CEA)

CEA is a 180-kDa GPI-linked glycoprotein which is normally expressed on the surface of colon. However, this oncofetal protein expression increases exponentially during carcinogenesis. Along with that, CEA can be shed into blood stream and can be measured as tumor marker. On a variety of gastrointestinal tumors, expression of CEA is markedly elevated including PDA [52,53,54,55]. Chmielewski et al. [56] did a study regarding efficacy of CEA CAR T cell therapy in mice having orthotopic pancreatic cancer cells. In all, mice size of tumor was reduced to below the limit of detection and long term tumor eradication was seen in 67% of mice. No lymphodepletion was required and there was no destruction of CEA + healthy tissue. CEA targeting CAR T cells therapy trials have not been reported yet, but phase 1 dose escalation trial of CEA CAR T cell therapy with preconditioning chemotherapy and intravenous IL2 in patients with advanced CEA positive tumors have reported that after 6 weeks of treatment, 7 out of 13 patients had stable disease and rest 6 patients progressed [57]. At the second dose escalation level, this trial was discontinued due to pulmonary toxicity in 4 patients [57, 58]. Therefore, CEA can be a promising target for advanced PDA but further research is needed.

MUC1

It is a membrane bound glycoprotein and is commonly overexpressed in various tumors like breast, ovarian, colon, and lung cancer [59]. MUC1 is also overexpressed in around 90% of PDA patients and considered as potential diagnostic, prognostic, and therapeutic marker [60]. Strength of expression is associated with poor prognosis. In various clinical and preclinical trials, MUC1-targeted therapy have been tested [61]. Posey et al [62] developed a CAR that targets MUC1 and it showed target specific cytotoxicity and was able to control growth of tumor cells in xenograft model of pancreatic cancer and T cell leukemia. In this study, MUC-1 CAR T cell therapy led to 100% (6/6) survival of mice as compared to 40% (2/5) survival with untransduced T cells at 16 weeks of therapy, therefore, establishing the role of MUC1 CAR T cell therapy for PDA.

HER2

HER2 is overexpressed in various types of cancers like breast cancer, lung cancer, and gastric cancer [63, 64]. It has been shown in many studies that HER2 can increase the proliferation, invasion, and metastasis of cancer cells, and it is an important biomarker and target of therapy for breast cancer. Presence of HER2 on PDA is controversial [65] but some of the studies have shown its presence on 20–60% of PDA [66, 67]. In the same study as mentioned above for CD24, HER2 CAR T cell therapy has been tested against human orthotopic pancreatic cancer in immunodeficient mice. HER2 CAR T cell therapy was able to reduce the tumor burden even in advanced cases of PDA within 1 week. Sequential therapy with CD24 CAR T cell therapy after 2 months HER2 CAR T cell therapy led 90% survival at 14 weeks [51]. Therefore, HER2 expression was little lower on PDA but it can be a promising therapeutic modality especially in combination.

Toxicity of CAR T cell therapy

CAR T cell therapy have demonstrated positive clinical outcome in hematological malignancy and modest outcome has been seen in various solid tumors. However, CAR T cell therapy also have toxicities like cytokine release syndrome (CRS), on target/off tumor effects, neurological toxicity, and anaphylaxis. The most prevalent and dreadful toxicity of CAR T cell therapy is over immune activation leading to CRS [68]. CRS can present as high grade fever, weakness, malaise, nausea, cardiac dysfunction, renal impairment, tachycardia/hypotension, and disseminated intravascular coagulation with elevation of interferon gamma, IL-10, IL-6, and GM-CSF (granulocyte macrophage colony stimulating factor) [68,69,70]. Generally, for diagnosis of CRS, monitoring of cytokines is technically difficult; therefore, CRP (C-reactive protein) is used as marker for onset and severity of CRS, as it is produced by liver cells in response to IL-6 [69, 70]. The treatment for CRS generally include high dose steroid therapy, ventilator support, vasopressors, and anti-IL-6R antibody (Tocilizumab) which has shown clinical benefit in some of the patients [71]. In various patients receiving CD19 CAR T cell therapy, neurological toxicities has been reported including confusion, aphasia, delirium, myoclonus, and seizure. Cause for neurological toxicity is not identified, likely it is related to elevated cytokine levels [69, 70]. Direct toxic effect of CAR T cells on neurological tissue is another possibility. The ideal target for CAR T cell therapy should be present only on cancer cells, but generally these targets are shared by normal tissues also. Therefore, action of CAR T cells on nonpathogenic tissue lead to on-target/off-tumor toxicity [72]. Targeting of HER2/neu lead to rapid respiratory failure, multi-organ dysfunction, and death in a patient due to reactivity against pulmonary tissue [73]. As there is wide range of expected and unexpected toxicities, integration of suicide gene or elimination gene may be an important component for improvement of this therapy [74,75,76].

Current limitations and future perspectives of CAR T cell therapy

After outstanding success in hematological malignancies, now CAR T cell therapy is one of the most potential therapies for advanced solid tumors like pancreatic cancer. However, solid tumors have additional challenges as compared to hematological malignancies in success of CAR T cell therapy. The main three basic pathophysiological challenges are lack of antigen, hostile tumor microenvironment, and poor trafficking of engineered T cells. These challenges were minimal in liquid cancers as they have one standard antigen, i.e., CD19 which is present on all B cells. Treatment with CD19 CAR T therapy may lead to B cell aplasia but this can be managed by intravenous immunoglobulins and close monitoring for infection. Trafficking of T cells and immunosuppressive microenvironment are also not issue in hematological tumors as CAR T cells are infused directly into blood.

Most common limiting factor for CAR T cell therapy in PDA is hostile tumor microenvironment. As microenvironment of PDA is highly immunosuppressive, it makes survival of engineered T cells difficult, and therefore reduces their efficacy. To deal with this issue, combinational therapies should be tried, like combination of Immune checkpoint inhibitors with CAR T cell therapy [77]. Both are immunotherapies and target cancer cells in different ways, and hence, they can have synergistic effect and improve outcome in PDA. Higher bulk of tumor is associated with poor outcome with CAR T cell therapy has been shown in studies on leukemia patients. Similar response is also expected in PDA; therefore, we should reduce the bulk of tumor before providing CAR T cell therapy with cryoablation, radiation, radiofrequency ablation, or even surgical debulking. Along with reducing bulk, radiation therapy also potentiates immune response which can further potentiate efficacy of CAR T cells [78, 79]. Preconditioning with traditional chemotherapy like cyclophosphamide and fludarabine can also be used as it can also reduce the bulk of tumor along with that it inhibits Tregs, which improves tumor microenvironment for proliferation of CAR T cells [80]. In a research, combination of gemcitabine with rosiglitazone in PDA patients has reduced tumor progression, metastasis, and improved survival also. Rosiglitazone altered the tumor suppressive mediators and made the microenvironment more immunogenic [81]. Similarly, metformin also increased CD8 + TIL in tumor microenvironment [82]. Therefore, if CAR T cell therapy can be combined with rosiglitazone or metformin, immune suppressive nature of tumor microenvironment will reduce and efficacy of CAR T cells will improve.

CAR T cells also need to be modified to improve their efficacy and specificity, like addition of chemokine receptors. Chemokines are molecules released by cancer cells for proper trafficking to tumor site. Receptors of CAR T cells should be complimentary to these molecules for improvement in migration of T cells. Therefore, if CAR T cells have overexpression of these chemokines like CCR2b, CCR4 (CCL17 receptor), and CXCR2 (CXCL1 receptor), migration will improve [83, 84]. In one study on mice migration of MSLN, CAR T cells improved 12.5 times with overexpression of CCR2b in malignant pleural mesothelioma [85]. It can also be tried in PDA. For penetration of extracellular matrix, T cells secrete Heparanase (HPSE) enzyme but [86], while preparing CAR T cells, they become unable to secrete HPSE enzyme due to functional changes. Therefore, researchers developed CAR T cells which can express HPSE and these cells have shown better anticancer response [87]. This strategy can be useful in PDA also.

Conclusion

CAR T cell therapy is promising emerging treatment in cancer care. After its success in hematological malignancies, it is currently tried in Solid tumors. PDA is highly aggressive malignant tumor with high mortality and extremely limited therapeutic option. CAR T cell therapy can be a promising treatment modality for advanced PDA but it has to travel a long distance to come into clinical practice. Currently, there are various limitations and toxicities which are obstructing its way to reality like immunosuppressive tumor microenvironment, nutrient deprivation, hypoxia, dense stoma, poor trafficking of T cells, CRS, on target, off tumor effects, and many more. Active research is going on to deal with these limitations and we are optimistic that in near future, CAR T cell therapy alone or in combination will improve the outcome and survival of PDA patients.