， ，

(Department of Thoracic Oncology，the Second Affiliated Hospital of Zunyi Medical University，Zunyi Guizhou 563099，China)

[Abstract] With the deeper understanding of the pathophysiology and pathogenesis lung cancer treatment has entered a new era.At present,an increasing number of immunotherapeutic agents including monoclonal antibody-targeted drugs have been used in the clinical treatment of malignancy,but it still has many limitations.Chimeric antigen receptor-modified T (chime-ric antigen receptor modified T,CAR-T) cells,a novel adoptive immunotherapy strategy,have not only been used successfully against hematological tumors,but also opens up a new theoretical basis for the immunotherapy of solid tumors including lung cancer.However,targeting lung cancer-specific antigens using engineered CAR-T cells is complicated because of the lack of proper tumor-specific antigens,an immunosuppressive tumor microenvironment,tumor antigen escape,along with off-target effect,etc.This review aims to systematically explain the basic structure and generation characteristics of CAR-T cells,summarize the common tumor-associated antigens in clinical trials of CAR-T cells for the treatment of lung cancer,and point out the current challenges and new strategies,which might provide new ideas and approaches for preclinical and clinical trials of lung cancer.

[Key words] lung cancer; chimeric antigen receptor; T cells; immunotherapy

Adoptive cell therapy (ACT),a specific type of immunotherapy,has progressed greatly in recent years[1-2].In contrast to conventional anti-tumour treatment modalities,such as surgery,chemoradiotherapy,and interventional measures,ACT does not exert direct effects on tumours.Instead,T lymphocytes extracted from tumours or human peripheral blood are modified using genetic engineering,cytokine stimulation,and in vitro culture to amplify the T lymphocytes population with anti-tumour activity.These T lymphocytes are then re-infused into the patient to exert their anti-tumour effects.ACT enables complex treatment with higher precision.

Chimeric antigen receptor (CAR)-T cell immunotherapy has dramatically altered the cancer treatment landscape,and several studies have reported its successful application for the treatment of acute leukaemia and non-Hodgkin lymphoma,without inducing graft-versus-host disease[4-6].The great success of this innovative method for the treatment of haematological malignancies has inspired extensive research and multiple clinical trials on the use of CAR-T cells for solid tumours[7-8].

Lung cancer is a severe threat to human health due to its complex mechanisms and high heterogeneity.Despite tremendous improvements in diagnosis and treatment,the prognosis of lung cancer patients remains poor.In recent years,numerous studies have suggested that CAR-T cells may offer a beacon of hope for patients who respond poorly to standard therapies[10].Consequently,CAR-T cells have become a key research direction in lung cancer treatment.This paper outlines the principles of CAR-T cell therapy for lung cancer and provides a review of the targets,in vivo and in vitro research progress,barriers to treatment,and strategies to improve CAR-T cell therapy for lung cancer.

1 Principles of CAR-T cell therapy

CAR-T cells were first described by Eshhar et al[11].Production of CAR-T cells involves genetic engineering to introduce a CAR gene into T lymphocytes,enabling them to express a specific receptor for binding tumour-related antigens.CARs differ from T cell receptors (TCRs) in that they contain a synthetic extracellular antigen recognition domain,transmembrane domain,and cytoplasmic signalling domain.Notably,CARs recognise antigen-targeted cells in a major histocompatibility complex (MHC)-independent manner[12-13],which is one advantage of this approach.CAR-T cells also differ from TCRs in that they can bypass MHC-restricted T cell recognition,including downregulation of human leukocyte antigen (HLA) expression and proteasomal antigen processing,two mechanisms that contribute to TCR-mediated tumour evasion[14-16].

CAR-modified T cells consist of three main structural components:(1) An MHC-independent extracellular domain containing a single-chain variable fragment (scFv) that recognises and binds specific tumour antigens.CAR function can be predicted based on the affinity of the scFv.Although almost all scFVs used in clinical practice are antigen-binding domains derived from heavy and light chain variable regions[17-18],the extracellular targeting domain of CAR is not limited to scFvs.For example,IL13Rα2-specific CARs with an extracellular domain consisting of a modified IL13 molecule have been successfully applied in clinical practice[19-21].Other studies have reported the addition of a ζ signalling domain to the cytoplasmic tail of NKG2D or a CD70 (CD27) receptor for the respective construction of NKG2D ligand-and CD70-specific CARs[22-24].Peptide ligands have also been used as extracellular domains for CARs,which showed strong anti-tumour activity in preclinical studies using animal models[25].(2) The second component is a homodimeric transmembrane domain typically derived from CD3 or CD8 connected to the extracellular domain by a hinge region.Differences in the length and flexibility of the hinge profoundly affect the function of CAR-T cells.The most used hinge structures are the IgG1 hinge and IgG1/4 CH2CH3 domains.Current research works have shown that CAR-T cells with short hinge regions possess superior anti-tumour activity.Other studies further showed that certain hinge regions impair CAR-T cell activity.Hombach et al.[26]demonstrated that a CH2-CH3 hinge significantly decreased epitope recognition of a first generation CD30-specific CAR.The transmembrane domain,located between the hinge region and intracellular signalling domain,is commonly derived from CD3,CD4,CD8,or CD28.Its structure significantly influences the expression,location,and effector function of CAR-T cells.For instance,first generation CD19 CAR-T cells containing a CD3-ζ transmembrane domain are less stable than second generation CD19 CAR-T cells containing a CD28 transmembrane domain[27].(3) The third component is an intracellular signalling domain consisting of a signalling adapter (usually CD3ζ or FcεRIγ) and a costimulatory receptor (4-1BB,CD28,or OX40)[28-31].After antigen recognition,the signalling domain of the CAR transmits a costimulatory signal to T cells that activates its tumour-killing effects.Due to the complex interactions among the scFV,hinge region,transmembrane domain,and intracellular signalling domain,there is no single optimum configuration.CAR optimisation is largely empirical and requires extensive testing in a range of preclinical models.

2 Effective targets for lung cancer treatment

Since their development,CARs have advanced through four generations,but their efficacy for the treatment of solid tumours remains low compared with that for haematological tumours.The greatest challenge to the successful application of CAR-T cells in solid tumours is the identification of suitable target antigens.

Lung cancer is divided into two main pathological types,non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC),and immunotherapy provides benefits for patients with both pathological types32-33].Accumulating studies have successively demonstrated the potential for the broad application of CAR-T cells in the treatment of solid tumours.Effective CARs possess the following characteristics:(1) Cytotoxic T lymphocyte (CTL) specificity,(2) absence of MHC restriction,(3) large scale preparation,(4) in vivo amplification,(5) generation of immunological memory,and (6) efficacy for the treatment of refractory tumours.Several effective targets for lung cancer are described below.

2.1 CD56 CD56,also known as NCAM1,is a glycoprotein expressed on the surface of SCLC cells and most neuroendocrine tumours.Given its key roles in cell development,neural differentiation,and immune surveillance,CD56 is a candidate target antigen[34].SCLC is characterized by low HLA expression and relatively poor T cell infiltration.Multiple phase I clinical trials demonstrated the efficacy of CD56-targeted antibody-drug conjugates for the treatment of SCLC[35-36].Crossland et al[37]reported that CD56R-CAR-T cells effectively inhibited the growth of SCLC and significantly reduced tumour burden in xenograft models.They also showed that optimisation of the CD28 costimulatory molecule may increase the persistence of anti-tumour activity.However,the study had significant shortcomings; only 35% of the CAR-T cells expressed CD8,and CD8+T cells are considered effective anti-tumour cells.

2.2 Epidermal growth factor receptor (EGFR) Epidermal growth factor receptor (EGFR),also known as human epidermal growth factor receptor 1 (HER-1),is a transmembrane glycoprotein expressed in epithelial cells.Overexpression of EGFR can cause gene amplification and mutation,thereby affecting cell proliferation,cell migration,and angiogenesis[38].The extracellular domain of EGFR has ideal tumour specificity and immunogenic epitopes,which make it a suitable target for adoptive cellular immunotherapy[39].Using the non-viral piggyBac transposon system,Li et al.generated EGFR-CAR-T cells containing the EGFR scFv,transmembrane domain,and intracellular 4-1BB-CD3ζ signalling domain.These cells had a higher CD3+CD8+CTL population,and the efficacy of EGFR-CAR-T cell therapy for NSCLC was shown in basic experiments.A phase I clinical study investigating the use of second generation lentivirus-transduced EGFR-CAR-T cells in 11 patients with NSCLC reported tolerable and manageable target-related toxicities[10].Examination of EGFR-CAR-T cell levels in blood and tumour biopsies showed that the cells effectively migrated to tumours.However,the study also suggested that the levels of EGFR-CAR-T cells in peripheral blood were inconsistent with tumour lesions or clinical outcomes,indicating that EGFR-CAR-T cells in peripheral blood cannot be used as a marker for monitoring CAR-T cell function.One difficulty with current CAR-T cell therapy is the lack of an accurate indicator of CAR-T cell function in the body.In another phase I clinical trial of EGFR-CAR-T cells for the treatment of NSCLC,piggyBac-generated EGFR-CAR-T cells were well-tolerated in all nine patients in the study[40].The most common adverse events were grade 1-3 fever,and no patients experienced grade 4 adverse events or serious cytokine release syndrome.After treatment,eight of the nine patients showed detectable EGFR-CAR-T cells in their peripheral blood.One patient showed a partial response that lasted for more than 13 months,six had stable disease,and two experienced disease progression.The progression-free survival of these nine patients was 7.13 months (95%CI:2.71-17.10 months),and the median overall survival was 15.63 months (95%CI:8.82-22.03 months).These results suggest that non-viral piggyBac transposon system-engineered EGFR-CAR-T cell therapy is feasible and safe for the treatment of patients with EGFR-positive advanced relapsed/refractory NSCLC.Additional research is required to further assess this cell therapy technique in more patients or at higher doses.

2.3 Human epidermal growth factor receptor-2 (HER2) Human epidermal growth factor receptor-2 (HER2; erbB-2/neu) is a member of the erbB family of receptor tyrosine kinases.The ERBB2 gene that encodes HER2 is a dominant driver of cell proliferation[41-42].Ahmed et al.demonstrated the effectiveness of HER2-CAR-T cells in solid tumours; the median overall survival of the 19 study patients was 10.3 months (range,5.1-29.1 months)[43].Another clinical study of HER-targeted CAR-T cell therapy for the treatment of lung cancer (NCT02713984) is currently underway in China.

2.4 Carcinoembryonic antigen (CEA) Carcinoembryonic antigen (CEA) is present on the surfaces of cancer cells differentiated from endodermal cells and functions as a structural protein in the cell membrane.Abnormal CEA levels are found in many cancers,including colorectal,lung,pancreatic,stomach,and breast cancers,making it a valuable tumour marker.A clinical study of CEA-targeted CAR-T cell therapy for the treatment of NSCLC (NCT02349724) is currently in progress.The study includes patients with colorectal,stomach,breast,and pancreatic cancers.A clinical study of solid tumours with liver metastases included lung adenocarcinoma patients with liver metastases who were treated with the maximum CAR-T dose (1010 cells).One patient was alive at 23 months after CAR-T therapy,whereas five patients died of progressive disease[44].

2.5 Mesothelin (MSLN) Mesothelin (MSLN) is a surface glycoprotein expressed at low levels in normal mesothelial cells and the cells of the pulmonary pleura,pericardium,peritoneum,and capsules.It is abnormally expressed in most malignant pleural mesothelioma and lung,pancreatic,and ovarian cancers,which makes it a candidate antigen target for CAR-T cell therapy[45-46].Several phase I clinical trials of MSLN-targeted CAR-T cell therapy (NCT02414269,NCT01583686,NCT02580747,NCT02159716,and NCT01355965) are currently underway,and lung cancer patients are included in certain trials.Adusumilli et al[47].investigated intra-pleural administration of a single dose of CD28-costimulated MSLN-targeted CAR-T cells.The study recruited a total of 20 patients; 18 had malignant pleural mesothelioma,one had lung cancer,and one had breast cancer; 35% of the patients received ≥3 lines of therapy.A key feature of the study was the adoption of a PD1-CD28 chimeric switch receptor to reduce immunosuppression.

2.6 Mucin 1 (MUC1) Mucin 1 (MUC1),a glycoprotein that is abnormally expressed in multiple tumours,is closely associated with tumour progression and poor prognosis.Therefore,it is a candidate target antigen for CAR-T cell therapy[48].Recruitment of patients with MUC1-positive advanced refractory solid tumours for participation in a phase I/II clinical trial in China (NCT02587689) is currently underway.Eligible cancer types are NCSCLC,hepatocellular carcinoma,pancreatic carcinoma,and triple-negative basal-like breast carcinoma.

2.7 Receptor tyrosine kinase-like orphan receptor 1 (ROR1) Receptor tyrosine kinase-like orphan receptor 1 (ROR1),a member of the receptor tyrosine kinase (RTK) family,is highly expressed in multiple malignancies and tissues.ROR1 is a distinctive tumour-specific marker and is thus regarded as an attractive,potential treatment target[49].A clinical trial initiated by the National Cancer Institute (NCI) (NCT02706392) is investigating the safety and maximum tolerated dose (MTD) of ROR1-CAR-T cells for the treatment of advanced metastatic NSCLC and triple-negative breast cancer.The study involves the infusion of ROR1-CAR-T cells in a dose-escalation,from 3.3×105/kg to 1×107/kg cells.At present,treatment has commenced for 6 of the 30 recruited patients,although dose-limiting toxicities (DLTs) have not yet been observed.

3 Limitations and strategies of CAR-T cell therapy in lung cancer

In general,research on CAR-T cell therapy for lung cancer is still in the early exploratory stage.A large number of clinical trials are progressing slowly,the efficacy is very limited,and there are still some technical bottlenecks to be solved.Different from hematological malignancies,the main adverse reactions of CAR-T cell therapy for lung cancer are cytokine release syndrome and nervous system toxicity.How to overcome these challenges is currently a hot field of CAR-T cell therapy for lung cancer.

3.1 Cytokine release syndrome Cytokine release syndrome (CRS) is a systemic inflammatory response triggered by T-cell activation,usually manifested by fever,chills,muscle pain,generalized weakness,and systemic organ failure[50].Clinical studies have shown that CRS was mainly caused by the activated CAR-T cell resulted in a significant increase in the secretion of proinflammatory factors (e.g.,IL-6,IFN-γ,and TNF-α) by immune cells (T cells,B cells,natural killer cells,and monocytes/macrophages) that disrupt the balance between proinflammatory and anti-inflammatory responses[51].Some studies have demonstrated that regulating the in vivo lifespan and kinetics of CAR-T cells by optimizing CAR gene transfection[52]and using nanoparticles[53]can reduce and avoid CRS.Therefore,avoiding CRS damage after CAR-T cell immunotherapy will be a key issue to address in the treatment of lung cancer in the future.

3.2 Neurological toxicity CAR-T therapy may also cause central nervous system (CNS) toxicity[54].Some patients develop reversible neurological complications such as confusion and epilepsy-like symptoms.There may also be progressive confusion,aphasia,and eventually dementia.Some cases even require air protection,such as intubation and mechanical ventilation[55].Clinical studies have shown that endothelial cell activation promotes disruption of the blood-brain barrier,allowing immune effector cells and inflammatory mediators to infiltrate the central nervous system,leading to neurotoxicity[56-57].CRES is largely reversible and completely resolved after treatment with tocilizumab and dexamethasone,although recovery from neurotoxicity is slower with tocilizumab treatment with dexamethasone treatment with endothelial cell activation[56].Taken together,a better understanding of pathophysiology will be an important factor in reducing systemic CRES in CAR-T cell immunotherapy of solid tumors.

4 Future perspectives for CAR-T cell therapy in lung cancer

The success of CAR-T cells in the treatment of hematological malignancies has brought new hope for the clinical treatment of lung cancer and has entered a stage of rapid development.However,due to the heterogeneity of malignant solid tumors and the limitations of preclinical experiments,the clinical application of CAR-T cells should take a more cautious approach,and future research on CAR-T cells may include:(1) finding more stably expressed and specific target antigens; (2) modifying the structure of CAR to enhance the efficacy,specificity,and survival time of CAR-T cells;(3) reducing the toxicity of CAR-T cells; (4) optimizing CAR-T cells that target the TME of lung cancer; (5) exploring combination therapies,such as combining immune checkpoint inhibitors,PD-1 inhibitors,or dual-CAR-T; and(6) establishing natural ligand-receptor-based CAR-T cells.These modified CARs are being studied in animal models and clinical trials in an attempt to alleviate the heterogeneity of tumor antigens and may eventually form the next generation of CAR-T cells.In conclusion,the above efforts will provide safer and more effective clinical application of CAR-T cell immunotherapy for solid tumors such as lung cancer.

5 Conclusion

At present,although the research of CAR-T cell immunotherapy for solid tumors is still in its infancy,the beneficial results of the preliminary trials provide a theoretical basis for its application in the subsequent clinical treatment of solid tumors.With the continuous improvement of CAR-T technology for lung cancer treatment,such as insights into TME,screening of novel target antigens,and advances in molecular biotechnology,it provides a lot of hope,but there are also many challenges.We are still faced with many problems,especially killing effect,before it can be used as an alternative or as an adjuvant therapy for lung cancer in the future.In addition,combining CAR-T cells with chemoradiotherapy may enhance their antitumor activity.To sum up,with the continuous innovation of CAR-T design concept and treatment plan,CAR-T cell immunotherapy is expected to become an important direction of lung cancer treatment with broad prospects.