Although antibodies produced by the hybridoma technique possess attractive advantages, such as good specificity, high purity and large-scale production (25), some defects of murine-derived antibodies remain unavoidable. On the one hand, the low affinity of murine mAbs to the Fc fragment on the immunocyte surface, can cause light antibody-dependent cell-mediated cytotoxicity (ADCC), resulting in a mild killing effect on tumour cells (37). In addition, the killing effect has a short time in which to take effect because of the short half-life of murine mAbs in the blood (38).

On the other hand, murine mAbs cause immunogenicity (39) and can further produce HAMA (27), which means that the repeated use of murine mAbs can lead to decreased efficiency and harm to humans due to allergic reactions (26). Additionally, it is not uncommon for patients treated with mAbs to produce human anti-murine immunoglobulin responses, possibly due to the immune deficiency associated with certain types of tumours (40). Therefore, some early murine mAbs, such as the E5 murine mAb, not only failed to achieve the desired effect in the treatment, but increased the mortality of patients, leading to a period of downturn in the development of mAb preparation and mAb treatment (41).

Rituximab, as the first lymphoma mAb developed in 1982, was shown to alleviate the condition of tumour patients, which raises great hope for the use of mAbs in tumour treatment (42). In addition, the first mAb drug, anti-cd3 mAb OKT3, was approved by the US FDA to enter the market in 1986; this mAb can alleviate the anti-rejection reaction during organ transplantation (23,34). Overall, the mAb drugs produced at that time were murine mAbs and rabbit mAbs, which have some clinical drawbacks (26). Recently, with the advances in technology, human hybridoma technology has been developed as a new mAb preparation technology on the basis of mouse hybridoma technology and rabbit hybridoma technology, which causes the fusion of immunised human B cells and human myeloma cells to produce hybrid cells that can divide indefinitely and secrete antibodies (26). However, the technology has produced a very limited number of multiple myeloma cell lines, has a low cell fusion success rate and easily causes the loss of chromosomes (43).

Compared with murine mAbs prepared by the hybridoma technique which are limited in clinical application due to their ability to cause HAMA reactions in vivo, the chimeric mAbs prepared by DNA recombinant antibody technique can significantly alleviate adverse reactions because of an approximately 70% reduction in immunogenicity of the heterologous antibody, thus improving the curative effect (46). For example, infliximab is capable of preventing and reducing inflammation as the chimeric mAb to tumour necrosis factor (TNF)-α in the treatment of rheumatoid arthritis (RA) and Crohn.s disease (47). In addition, it also has the effector functions of human antibodies because of the existence of a humanised Fc fraction, which makes the chimeric mAb possess more potent complement-dependent cytotoxicity (CDC) and ADCC (48). On the basis of the mechanism, the rituximab chimeric anti-CD20 mAb was developed to treat relapsed indolent lymphoma because the cell-surface antigen CD20 is expressed on more than 90% of B-cell lymphomas and chronic lymphocytic leukaemias (49). Abciximab, a Fab fragment of a chimeric mAb, functions as a GP IIb/IIIa receptor antagonist to significantly decrease the size of coronary artery aneurysms in children with Kawasaki disease by promoting vascular remodelling, and decrease the risk to go through early stent thrombosis in diabetic patients with ST-segment elevation myocardial infarction (50,51). In addition, basiliximab and cetuximab, serving as chimeric mAb, can function to prevent early acute or slow rejection reaction after organ or allogeneic hematopoietic stem cell transplantation so as to treat acute graft-vs. -host disease (52,53), and treat different cancers including recurrent and metastatic head and neck squamous cell carcinoma and metastatic colorectal cancer (54,55), respectively. However, chimeric antibodies can partially solve the problem of heterologous protein rejection, but they may still induce an HAMA reaction, interfere with antibody efficacy and induce a hypersensitivity reaction due to the fact that they also contain the murine V region which limits their clinical application to some extent (56,57).

Phage display techniques can clone the peptide-coding or protein-coding gene fragment into the appropriate position of the phage shell protein structure (58), so that the foreign polypeptide/protein and shell protein are expressed in the form of a fusion of each other and then displayed on the phage surface as the progeny phage reassembles (59,60). The demonstrated polypeptide or protein can maintain a relatively independent spatial structure and biological activity, which is conducive to the recognition and binding of target molecules, providing a way to screen for single-chain antibodies with high specificity and binding ability (61). In theory, as long as enough of this type of peptide is expressed in the library, one or more phage can bind to these targets.

Phage display was pioneered by Greg Winter and his colleagues (59). In 1985, G.P. Smith developed phage display technology (62) based on the research of phage biology and molecular biology, which show unique advantages in virus infection, including HIV infection and tumour diagnosis and treatment. Later, in 1987, Geysen et al proposed that short peptides containing key amino acid residues can mimic the antigenic determinants of proteins and the interaction between proteins is achieved by the interaction between local peptides (63). In 1988, Parmley and Smith proposed the idea that the construction of a random peptide library could provide insight into the antigen-determining cluster epitopes recognised by antibodies (64). Subsequently, Scott and Smith fused random short peptides to the surface protein PIII of filamentous phage and displayed it on the surface of the phage, creating the first phage random peptide library (65). In the same year, McCafferty et al used phage display technology to screen for single-chain antibodies to lysozyme bacteria, propelling phage display technology into an era of widespread application (66). Recently, because of the pioneering work and application in the phage display of peptides and antibodies, Professors George P. Smith and Gregory P. Winter both won a quarter share of the 2018 Nobel Prize in Chemistry (59). Until now, the application scope of phage display technology has been expanding, and the technology has also been constantly improving and developing.

The emergence of phage display technology has opened up a simple and fast route for the production of genetically engineered mAbs, which bypasses the technical difficulty of hybridomas (35). It clones and amplifies VH and VL gene fragments in human lymphocyte spectrum by RT-PCR, and randomly combines gene fragments into expression vectors, in order to construct a large-capacity human antibody library (27). In addition, the phage display can simplify the cloning process and acquire a large amount of material to produce peptides or proteins because of the small size of the phage genome and high efficiency of the phage infection (67,68).

The phage display offers the direct physical link between a protein and its genetic material, which helps people to effectively screen the desired cloning again and again, and then amplify it (67). In the process of library screening, specific phage clones are enriched continuously due to their specific affinity for ligands, and relatively rare clones that can bind ligands can be quickly and effectively screened out from a large library (58). Therefore, the biggest advantage of phage display is that once the phage library is established, specific antibodies against the target antigen can be directly screened from the library according to the needs within 23 weeks, which greatly reduces the preparation cycle of mAbs (61). In addition, by specific construction, the filamentous phage may act as a vector, and generate a peptide library of phage display that contains hundreds of millions of unique peptides, which are conducive to their application in antiviral research (69). However, due to the different binding properties of antibodies in bacteria and eukaryotic cells, the applicability of the technology is limited to a certain extent (70). The processes of the phage display refer to bacterial transformation, phage packaging, and even transmembrane secretion processes, which limit the capacity of the phage display library and their molecular diversity. At present, the capacity of the phage display library is usually 10¹¹ (27). Also, limited by the expression system, the antibody library is not large enough to support the acquisition of some rare antibodies and not all sequences are well expressed in phages, because the realisation of some protein functions acquire folding, transportation, membrane insertion and complexation, resulting in the need for additional selection pressure during in vivo screening (45). It is difficult to obtain antibodies that inhibit the growth or function of phages or the expression host, as the phage display system depends on the expression of intracellular genes, which may make the diversity of the library decrease rapidly (71). Also, a phage display library cannot take on the effective mutation and recombination in vitro, which in turn limits the genetic diversity of the molecules in the library (72,73). Nevertheless, these temporary shortcomings cannot obscure the great potential of its applications.

Nowadays, with the establishment of more phage display libraries, the construction of advanced genetic operating system and the development of more efficient phage display systems, phage display technology plays an important role in different fields, especially in protein and antibody-related fields (34). Phage display technology has become an advantageous tool for detecting the protein spatial structure, exploring the binding sites between receptors and ligands, and searching for ligands with high affinity and biological activity (74). It has had a far-reaching impact on research into the mutual recognition of protein molecules, the preparation of phage-functionalised biosensors, the development of new vaccines and tumour therapy (75,76). In addition, Humira, ramucirumab and other mAbs developed by phage display technology have been widely used in clinical practice, especially Humira, which is widely applied in the treatment of rheumatoid arthritis (RA), psoriatic arthritis, Crohn.s disease, ankylosing spondylitis and uveitis (45).

In the past three decades, transgenic animal technology through genetic engineering has been envisaged to improve food quality, animal production and the production of biological products, to reduce or minimise the environmental impact of animal production and to add value to animal products. Recently, with the advance of the ability for targeted genome engineering via genome editing methods such as TALENs, ZFN and the CRISPR/Cas9 system, this technique has been widely used to obtain a series of human mAbs against the interleukin-6 (IL-6) receptor, TNF receptor and epidermal growth factor receptor (EGFR), which play important roles in the treatment of tumours and other diseases (86,87). mAbs against the human IL-6 receptor can show strong antitumour activity in vivo against multiple myeloma cells by inhibiting IL-6 functions (87). H-R3, as a humanised anti-EGFR antibody with antitumour, anti-proliferative, anti-angiogenic and pro-apoptotic properties, can act as an effective EGFR antagonist to inhibit signal transduction, in order to directly or indirectly affect cell proliferation, cell survival and angiogenesis-inducing capacity (88). As an important milestone in validating XenoMouse strains as well as other human immunoglobulin-producing mouse technologies, the first fully human mAb, panitumumab, which was developed from XenoMouse technology and approved by a regulatory agency, has a positive risk-benefit profile in advanced, chemotherapy refractory colorectal tumours and has the potential to increase treatment rates of this disease in earlier lines of therapy (89). Recently, the first transgenic rabbit strain for human antibody production has been created with the discovery that the anti-body diversification mechanism at work in rabbits can act on the fragments of the human transgenic immunoglobulin gene (90), which further expands the application of human mAbs in drug development and promises to lead to new treatments for various diseases.

Ribosome display technology is a powerful tool for protein screening using functional protein interactions in vitro (91). By associating genotypes with protein phenotypes, it can use specific ligands of target proteins to select target proteins and corresponding gene sequences from the protein display library (92). It combines the correctly folded protein and its mRNA on the ribosome at the same time to form mRNA-ribosome-protein trimer, in order to screen some high-affinity proteins with specific binding to target molecules, including antibodies, peptides and enzymes (93,94). The preparative technique involves different key processes, including specific processing and modification of the DNA that encodes proteins, transcription and translation, affinity screening in vitro, the separation of mRNA and molecular orientation evolution in vitro (95).

The ribosomal display technology has undergone a certain period of research from the time it was proposed to the time it was developed and matured. In 1994, Mattheakis et al of the Afflymax Institute in the US put the ideas of their predecessors into practice for the first time and established the prototype of ribosome display technology, which mainly used 'polypeptide-polyribosome-mRNA. complex to construct peptide libraries on polyribosomes, thus screening the polypeptide ligands of immobilized mAbs with an affinity constant of 10⁹ (Nmol level) from a peptide library with a capacity of 10¹² (96). Later, Hanes and Plückthun improved the polyribosome display technology and established a new technology, ribosome display technology, in 1997, for the screening of complete functional proteins such as antibodies in vitro, on the basis of previous research results (93).

Traditional screening techniques have insurmountable drawbacks, mainly related to cell transfection, phage packaging, transmembrane secretion and protein degradation in library construction and screening (97,98). The library capacity and molecular diversity of phage or mRNA display technology are somewhat limited, which reduces the efficiency of library screening (71). In contrast, a ribosome display is a powerful way to screen large libraries and acquire molecular evolution (99,100). It has the advantages of simple library construction, large library capacity, strong molecular diversity, simple screening methods and no need for selection pressure, and can even improve the affinity of target proteins by introducing mutation and recombination technology (93,94). As a system to produce and screen folded proteins entirely in vitro, the ribosome display technique is shown to greatly exploit replicability of mRNA, allowing efficient enrichment of target genes, avoiding the step of bacterial transformation and making the technique unconstrained by the efficiency of cellular transformation (101). On the one hand, the technique greatly increases library capacity and screening throughput, and makes it easy to build a very large volume of antibody library (101). On the other hand, while the expressed proteins have the correct spatial fold conformation, the technique can be combined with some special PCR techniques to improve the protein expression diversity (28,102,103). It also can be used for the screening and research of cytotoxic fractions (94).

However, there are still some technical problems that need to be further advanced. Undoubtedly, maintaining mRNA stability and preventing the degradation of mRNA is the first problem in a ribosome display system (28). Facing the problem, Yamaguchi et al reported a novel screening method-cDNA display, which prevents the degradation of mRNA by promoting the binding of mRNA to linkers and the reverse transcriptional synthesis of cDNA, thus converting mRNA-protein fusions to cDNA-protein fusions and avoiding problems due to the stability of mRNA (104). The use of modified nucleotides as substrates for transcription reactions can also stabilise mRNA (105). In addition, how to construct the more stable 'mRNA-ribosome-protein. complex was one of the problems (106), as it only occurs in cases where the complex is complicated but its stability is poor in practice due to ribosomal display. To solve the problem, Roberts and Szostak developed a simpler and more effective display system-mRNA display system, which allows mRNA to bind to its encoded polypeptide in the presence of puromycin to form a stable mRNA-peptide complex that screens for the target peptide (107). The anti-small stable RNA A (anti-SsrA) oligonucleotides were designed by Muranaka et al to inhibit the function of SsrA and obviously promote the form of the 'mRNA-ribosome-protein. complex (108). In addition, how to improve the display of large molecular protein in the ribosome is also a problem that needs to be solved.

At present, there are numerous reports on the preparation of human mAbs by ribosomal display technology, and the advantages of this technology represent the developmental direction of mAbs (28). On the one hand, it can be applied widely in antibody engineering, proteomics, epitope mapping, and synthetic enzymes (93,103). On the other hand, it also opens up a new way to screen new therapeutic antibodies and new drugs for diagnosis and treatment in tumours, autoimmune diseases, infectious diseases, and inflammatory disorders (94). Ribosome display technology, as a new cloning display technology, will show a more extensive application space in protein interaction research, new drug development and proteomics (95,109).

In recent years, single B-cell antibody preparation techniques have begun to spring up and have gradually become widely used, alongside the development of molecular cellular biology (112). This is because mAbs prepared by single B-cell antibody technology have the characteristics of full human origin, high specificity and uniformity, showing unique advantages and good application prospects in the treatment of pathogenic microbial infections, tumours, autoimmune diseases and organ transplantation (113,114).

Compared with other mAb preparation techniques, monoclonal B-cell technology, is a technique for the cloning and expression of B-cell antibodies with single antigenic specificity in vitro, which preserves the natural pairing in the V region of the light and heavy chain, and has the advantages of good gene diversity, high efficiency, full humanisation and the small number of cells required (115). Studies have shown that human memory B cells can survive in humans for more than 50 years, providing a historical record of the specific antibodies produced during most of the host.s lifetime (116). In contrast, antibodies in the body fluids used in most traditional methods usually decay after macroglobulin clearance, which means that people lose their protective antibody within a few years (117). In addition, it is proposed that memory B cells in the blood of virus-infected patients may store records of early infection with the virus in patients-the genes, which provide a new direction for the research and development of mAbs (118).

Currently, the preparation of memory B-cell antibodies has become a popular method used to prepare the humanised antibody, which also promotes immunological research including antibody affinity maturation, the defence mechanism against vaccine immunity, vaccine development, and the treatment of tumours and autoimmune diseases at the same time (1,119). With the maturity and improvement of B-cell sorting technology, subsequent PCR gene amplification methods and the high-throughput analysis and identification of antibody genes, memory B-cell antibody preparation technology will play an unprecedented important role in the diagnosis, pharmacodynamic and clinical application in the future, leading to a new era of therapeutic antibody research (120-122). As this approach has also been successful in widely isolating neutralizing antibodies against viruses including SARS and H5N1 influenza, it can provide not only neutralizing antibodies for passive serum therapy, but also information for vaccine design, and is expected to accelerate the development of therapeutics in the field of infectious diseases (119,123). Additionally, Wrammert et al produced anti-H1N1 antibodies in 2009 by isolating plasma cells from peripheral blood, to analyse the characteristics of antibodies in detail generated from plasma blasts induced by pandemic H1N1 infection (124).

Traditionally, mAbs produce cytotoxic effects in tumour cells though ADCC, CDC, changing signal transduction, elimination of the cell-surface antigen, and targeted conveying payloads (125).

In general, ADCC is achieved by the specific binding of mAbs and the targeted antigen of tumour cells (126). Namely, the Fc fraction of mAbs can bind to the receptor of immune effector cells (NK cells, macrophages, neutrophils, granulocytes), and achieve activation of intracellular signals in the next moment, resulting in ADCC (126,127). NK cells activated by antibodies can release cellular cytotoxic granules (perforin and granzyme) to achieve cell apoptosis on the one hand, while they can release cytokines and chemokines to inhibit cell proliferation and angiogenesis on the other (128).

CDC refers to the cytotoxic effect involving complement; that is, after the binding of specific mAbs and the corresponding antigens on the surface of the cell membrane, the complex activates the classical pathway of complement and forms an attack membrane complex to induce a lysis effect on target cells (127,129). It is worth noting that, although CDC does not directly preside over the antitumour effects of most mAbs, it produces a variety of factors that enhance ADCC (130,131).

Almost every clinically effective unconjugated mAb, directly or indirectly, interferes with the signal transduction that influences the proliferation and survival of targeted cell populations (132). Growth factor receptors are some of the most commonly targeted tumour-associated antigens whose activation under normal conditions induces mitotic reactions and promotes cell survival, are overexpressed in numerous malignancies, leading to promotion of tumour cell growth and insensitivity to chemotherapy drugs (133-135). Therefore, the use of mAbs is likely to normalise the cell growth rate and restore the sensitivity of cells to cytotoxic drugs by reducing the signal that passes through these receptors (136). For example, the pertuzumab block shows receptor heterodimerisation (the dimerisation of HER2 with HER3 and other HER family receptors) that is required for signal transduction to play an antitumour role (130,137).

Targeted antitumour drugs provide a new concept concerning tumour therapy, referring to several tumour-associated signaling pathways and targets. For instance, the most common target antigens in solid tumours refer to epithelial cell adhesion molecule (Ep-CAM; also known as epithelial glycoprotein-2, EGP-2/GA 733-2), carcinoembryonic antigen (CEA), EGFR family including EGFR (also known as c-erbB-1), HER2/neu (c-erbB-2), HER3 (c-erbB-3) and HER4 (c-erbB-4) (135,138,139). Compared with them, the mAbs applied in lymphoma usually target CD52, CD20, CD30, CD22, CD37 and CD79 (135,136), with the easier achievement of a better effect because it is simpler to manage tumour penetration. In contrast to the above, tumour stroma and tumour vasculature offer some unique targets for antibody-based interaction because the new generation of tissue and vasculature show some components that differ from those in the normal situation, leading to the situation whereby fibroblast activation protein (FAP) (140) and tenascin-C (TNC) (141) are regarded as targets in the tumour stroma, Fibronectin ED-B (142) and prostate-specific membrane antigen (PSMA) (143) are regarded as targets in the tumour vasculature. In addition, ligands including vascular endothelial growth factor (VEGF) are thought to target cell-surface receptors expressed on tumour cells or their supporting tissue (140,144).

The fight between the immune system and tumour cells is a long-term dynamic process, which has both positive and mutual influences (145). In the process whereby a healthy individual.s immune cells detect and kill tumour cells via the antitumour immune response, activated T cells cause upregulated expression of several surface receptors, which can bind with relevant ligands expressed highly on the surface of tumour cells, resulting in inhibition of the immune response and downregulation of potent immune response (146). These surface receptors, namely the suppressive regulatory molecules, are essential for maintaining self-tolerance, preventing autoimmune response and minimizing tissue damage by regulating the time and intensity of the immune response; this is called the immune checkpoint (146,147). The immune checkpoint results in the inhibition of cellular function, meaning that the body cannot produce an effective antitumour immune response, ultimately leading to immune escape of the tumour (148).

On the theoretical basis of the immune checkpoint, some mAbs have been developed as immune checkpoint inhibitors to block the interaction between tumour cells expressing immune checkpoints and immune cells, in order to block the inhibitory effect of tumour cells on immune cells (148,149). During the occurrence and development of tumours, immune checkpoint inhibitors can enhance the immune function of the body and restore the recognition ability of T cells, in order to eliminate tumours or slow down the development of tumours (150,151) (Fig. 3). Recently, tumour-related immune checkpoint molecules mainly include programmed death-1 (PD-1), cytotoxic T lymphocyte-associated antigen-4 (CTLA-4), T-cell immunoglobulin mucin 3 (TIM3) and lymphocyte activation gene-3 (LAG3) (145,146,150).

CTLA-4 (also known as CD152) is a transmembrane protein expressed on the surface of activated CD4⁺ T cells, activated CD8⁺ T cells, and regulatory T cells (Treg cells), which can bind to the CD86(B7-2) and CD80(B7-1) ligand to negatively regulate T-cell activation (152). In addition, there is an intracellular pool of CTLA-4 in recycling endosomes of Treg and memory T cells which can be rapidly cycled to the cell surface upon activation (153). Therefore, the anti-CTLA-4 mAb can decrease Treg cells and activate T cell immune response by terminating the activity of CTLA-4 (145). Ipilimumab is an mAb that blocks CTLA-4, which was approved by the FDA to treat melanoma in 2011 (150,154). The blocking effect of ipilimumab on CTLA-4 is, that Ipilimumab binds to CTLA-4 to further impede the interaction between CTLA-4 receptor and the B7 ligand, thus increasing the activation and proliferation of T cells (145). This means that the antitumour effect of ipilimumab on melanoma is indirect and the action mechanism may be to help the body.s immune system recognize, target and attack melanoma cells (155). Hence, it is possible for ipilimumab to be used as a carrier of imaging agent with good targeting properties (156). In addition, it is reported that patients with human CTLA-4 insufficiency develop immune dysregulation, lymphadenopathy and hepatosplenomegaly (153,157,158). Therefore, anti-CTLA-4 mAbs should be able to inhibit the inhibitory signalling pathway and maintain the killing function of T cells, thereby killing tumour cells (146).

Although some anti-CTLA-4 mAbs do not cause an adverse reaction similar to cytotoxic drugs, including myelosuppression and alopecia, due to the different actions of cytotoxic drugs, they cause pathological damage to the body while producing an antitumour reaction, namely immune-related adverse events (IRAEs) (146,149). Different from others, the toxicity profile of ipilimumab mainly manifests as symptoms associated with infusion, including rigor, pruritus, fatigue, nausea, dizziness, colitis, less frequently hypophysitis, hepatitis pneumonitis, and even hypotension, angioneurotic oedema, and dyspnoea (159-161). When facing the problems that occur in anti-CTLA-4 immunotherapy, combination therapy is put forward which may thereby provide greater antitumour activity than either agent alone by enhancing antitumour immune responses, and presenting a miraculous, unprecedented therapeutic effect (159). For example, in the treatment of progressive melanoma, the combined blocking of PD-1/PD-L1 and CTLA-4 can further improve efficacy in patients compared with the single blocking of PD-1/PD-L1 or CTLA-4 (162). Based on this, a trial involving patients with advanced melanoma was carried out and revealed that nivolumab plus ipilimumab provides the longer progression-free and overall survival, and better health-related quality of life than ipilimumab alone (159,163). From another perspective, CTLA-4 inhibition can synergize with local chemotherapy, improving applicability and sensitivity to immune-checkpoint inhibition (164).

At present, not only ipilimumab, but also some other anti-CTLA-4 mAbs, have been optimised and developed, and have now attracted more and more attention from the public (148). Another anti-CTLA-4 mAb, tremelimumab, which was exploited by Pfizer, is being investigated in clinical trials (165). However, the research into tremelimumab has made little progress, meaning that ipilimumab is still regarded as the most promising anti-CTLA-4 mAb applied in tumour treatment (148).

PD-1 (also known as CD279), as a member of the immunoglobulin superfamily which is mainly expressed on the surface of activated T cells, can be used as an immunosuppressive molecule to regulate the immune system and promote self-tolerance by downregulating the response of the immune system to human cells and suppressing the inflammatory activity of T cells, which may prevent the immune system from killing tumour cells (166,167). PD-1 has at least two ligands, PD-L1 (also known as CD274 or B7-H1) and PD-L2 (CD273 or B7-DC) (146). In general, as one of the means by which human tissue protects them, PD-1 can bind with specific ligands on the immunocyte surface to prevent immune cells from activating and killing normal cells (166,168). Additionally, the binding of PD-1 with ligands promotes the programmed death of T cells and reduces the apoptosis of regulatory T cells by suppressing the T cell activation signal primed by the interaction between MHC and TCR, in order to make the tumour cells acquire immune escape (146). Certain types of malignant tumours express a mass of PD-1 on the cell surface; therefore, they evade the attack of immune cells by powerfully suppressing the activation of immune cells (129). PD-1 is also expressed on the surface of activated B cells and macrophages, indicating that PD-1 negatively regulates the immune response more widely than CTLA-4 (145). Therefore, immune regulation targeting of PD-1 plays an important role in antitumour, anti-infection, anti-autoimmune diseases and organ transplantation survival (169). Facing the situation, anti-PD-1 mAbs are manufactured as PD-1 inhibitors to activate the immune system to attack tumours and treat some types (145).

Compared with anti-CTLA-4 mAb therapy, immunotherapy with anti-PD-1 mAbs has a broader antitumour effect and fewer overall side effects (170,171). The biggest difference between immune drugs and others is that it shows more persistent efficacy which could lead to long-term survival or even a clinical cure for patients with advanced disease (146). Currently, there are various drugs approved by the FDA, including nivolumab (also known as Opdivo) and pembrolizumab (also known as Keytruda) (172). The indications of nivolumab include melanoma, non-small cell lung cancer (NSCLC), renal cell carcinoma (RCC), classical Hodgkin.s lymphoma (CHL), squamous cell carcinoma of the head and neck (SCCHN), and urothelial carcinoma (148). In contrast, pembrolizumab is mainly applied in melanoma, NSCLC, and SCCHN (167,172).

At present, the anti-PD-1 mAbs are mostly used in combination therapies, because traditional therapy using anti-PD-1 mAbs alone can lead to various IRAEs, including fatigue, skin rash, colitis, hypophysitis, pneumonitis, myasthenia gravis and interstitial nephritis (171). For tumours that are highly dependent on the immunosuppressive mechanism of PD-1/PD-L1, such as malignant melanoma, Hodgkin.s lymphoma, certain types of lung tumour, and colon tumours, it has shown remarkable efficacy (147), but for the vast majority of unselected solid tumours, the efficacy of PD-1 inhibitor alone is not high. In comparison, combination therapy improves the treatment effect and transforms patients who are not suitable for PD-1 inhibitor treatment into those who can benefit from it (173). First, the anti-PD-1 mAb can combine with another immunotherapy drug (174). A combination of the PD-1 and CTLA-4 mAbs has been shown to be more effective in treating several tumours, including malignant melanoma, than either antibody alone (162). In addition, PD-1 can combine with anti-CTLA-4 or VEGF tyrosine kinase inhibitors (TKIs) to adjust first-line therapy for metastatic kidney carcinoma (175). Secondly, PD-1 in combination with chemotherapy is considered a promising treatment strategy (176). Chemotherapy has a profound impact on the antitumour immune by directly regulating immune cellular subsets or indirectly stimulating the immune system through the induction of immunogenic cell death, revealing the existence of synergy between cytotoxic chemotherapy and immune checkpoint inhibition (176). This combination therapy is also approved for the first-line treatment of advanced NSCLC (176). Nevertheless, there are some patients with chemotherapy-refractory metastatic solid tumours; for those patients, PD-1 inhibitor combined with radiotherapy is regarded as a salvage treatment (177,178). This means that the anti-PD-1 may combine with radiotherapy to improve the overall survival (179). In addition, anti-PD-1 can combine with targeted drugs including acitinib, and levatinib (180,181). Recent studies have shown that the association of acitinib with pembrolizumab provides improved clinical benefit in patients with previously untreated advanced renal cell carcinoma, which is well tolerated (182). In addition, the strategy of using anti-PD-1 mAbs in combination with oncolytic virus (OV) to enhance antitumour immunity and therapy has been developed (183). To validate this, research into the effect of IL-15-armed OV in combination with PD-1 inhibitors in mice with colon or ovarian carcinoma processes, has shown some results including tumour regression and the prolongation of overall survival (184). Moreover, a personalised mutanome vaccine can be used in combination therapy with anti-PD-1 mAb as it enhances the persistence of anti-PD-1-mediated effect and extends anti-PD-1 therapies to patients with no preexisting T cell response (185). Finally, the anti-PD-1 mAb can combine with novel tumour-specific immune cells, such as the chimeric antigen receptor T-cell (CAR-T), to produce better therapeutic effects (186). Some research has found that PD-1 blocks CAR-T cell therapy within solid tumours, therefore the anti-PD-1 mAb which prevents the PD-1-related inhibition of CAR-T cell response can increase the levels of cytolysis and cytokine secretion and enhance the in vivo antitumour function of CAR-T cells (187,188).

PD-L1, expressed on the surface of tumour cells, can bind with PD-1 on the surface of activated T cells and B cells to conduct inhibitory signals and reduce T cell proliferation (146,189). As promising new agents, there are some anti-PD-L1 mAbs approved by the FDA in clinic, including atezolizumab (also known as Tecentriq) which is used to treat locally advanced or metastatic urothelial carcinoma, durvalumab (also known as Imfinzi) which is used to treat locally advanced or metastatic urothelial carcinoma and NSCLC, and avelumab (also known as Bavencio) which is used to treat meningioma, metastatic Merkel cell carcinoma and carcinoma of the urinary bladder (128,167,190). In theory, compared with anti-PD-1 mAbs which bind to PD-L2, anti-PD-L1 mAbs have specific effects and demonstrate a certain superiority (146). The anti-PD-L1 mAbs can block the co-suppression of B7-1 and PD-1, which is conducive to fully activate the function of T cells and produce cytokines (191,192). Therefore, anti-PD-L1 mAbs may more fully activate the immune system to kill tumours (191). Furthermore, it has been shown that durvalumab as a third-line or later treatment can significantly benefit advanced NSCLC patients with EGFR mutations or ALK rearrangements (EGFR⁺/ALK⁺) with ≥25% of tumour cells expressing PD-L1, although it is unsuitable for patients with EGFR⁺/ALK⁺ to use the immune checkpoint inhibitor because of the low curative effect and subsequent severe adverse reactions (193). In summary, anti-PD-1 mAbs and anti-PD-L1 mAbs each have their own indications and application scope, and the combined utilisation can achieve mutual complementarity in the interest of our common development (146).

In addition, there are various other useful checkpoints, such as the lymphocyte activation gene-3 (LAG-3), B7-H3, B7-H4, T cell immunoglobulin-3 (TIM-3), T cell immunoglobulin and ITIM domain protein (TIGIT), and V-domain immunoglobulin-containing suppressor of T cell activation (VISTA) (194-197), which are either entering the clinic or under active development.

Radiation can act directly on DNA molecules and cause their damage, by ionising water molecules in living organisms to produce free radicals which break macromolecules and lead to cell damage (199). Based on this theory, mAbs with specific affinity for tumours can be utilised as a carrier of highly active radiopharmaceutical agents, thus forming radioimmunoconjugates that target tumour tissue to kill tumour cells or inhibit their growth while reducing radiation damage to normal tissue, using the ionising radiation effects of radioisotopes (11).

The cytotoxic agent can conjugate with antitumour mAbs to form chemo-immunoconjugate, which can bind to the surface of antigen-positive tumour cells through the guidance of mAbs, inducing the internalisation of the conjugates (200). After that, these chemical drugs play their cytotoxic effects by binding to DNA molecules, thus killing tumour cells by inhibiting cell DNA and protein synthesis, interfering with cell nucleic acid or protein function, and inhibiting mitosis (201,202). Common conjugates include cisplatin, cyclophosphamide, etoposide, adriamycin, paclitaxel, methotrexate, and vinblastine (203,204).

Immunotoxin has a specific affinity for tumour cell surface antigens and can release bacterial or plant protein toxins to tumour cells without harming normal cells (205). Once the toxin enters the cell, it kills the tumour cell by inhibiting protein synthesis and altering signalling transmission (206). The main toxins currently used in reagents are diphtheria toxin, abrin, ricin, gelonin, and Pseudomonas aeruginosa endotoxin (198,207,208).

Antibody-based immunotherapy has been a major and rational therapeutic strategy in the clinical management of oncology (209,210). In clinical practice, therapeutic mAbs have a limited effectiveness in the treatment of solid tumours due to their large molecular weight, but a high degree of targeting (198). With a few exceptions such as mAbs to HER2, EGFR and CD20, most mAbs can bind with effector molecules by using specific linkers to produce antibody-drug conjugates (ADCs), which expand the scope of medical treatment while possessing highly targeted selection, achieving the complementary advantages of the two therapeutic drugs, which have little antitumour effects after binding the target antigen (15,198). In contrast, small molecule chemicals are highly effective against tumour cells, in spite of the fact that they are less selective and may cause serious side effects, accidentally injuring normal cells due to off-target toxicity (211). Therefore, mAbs can bind with effector molecules by using specific linkers to produce ADCs, which expand the scope of medical treatment while owning highly targeted selection, achieving the complementary advantages of the two therapeutic drugs (17,135). The mAbs can bind with effector molecules by using specific linkers to produce ADCs, which expand the scope of medical treatment while possessing highly targeted selection, achieving the complementary advantages of the two therapeutic drugs that can usually target tumour-associated antigens or specific receptors on the surface of tumour cells and show a selective directing effect on tumour cells (135). The effector molecules that act as payloads which produce a killing effect on tumours include radiopharmaceutical agents, cytotoxic agents and bacterial or plant protein toxins, conjugating respectively with mAbs to form radioimmunoconjugates, chemo-immunoconjugates and immunotoxins used in tumour-guided therapy (198). Currently, ADCs that have been approved by the FDA include brentuximab vedotin, trastuzumab emtansine, gemtuzumab ozogamicin, Inotuzumab ozogamicin, and polatuzumab vedotin (211,212).

Generally, ADCs are injected intravenously into the blood system to prevent the hydrolysis of mAbs by gastric acid and protease and are distributed into the tumour tissue by exosmosis of the endothelial pores and the endocytosis of endothelial cells (198). After the mAbs specifically direct drugs to the surface of tumour cells expressing tumour-specific antigens, ADCs come into tumour cells by internalisation (211,213). As a general rule, there are three distinguished pathways to internalise, including clathrin-mediated endocytosis, caveolae-mediated endocytosis and pinocytosis (211). Later, with the influence of the acidic environment of the cytoplasm, some ADCs with cleavable linkers release effector molecules which can damage tumour cells, while other ADCs undergo the enzymatic fracture of linkers or mAb degradation with the influence of lysosomal protease (214). Finally, payload and degradation products are released into the cytoplasm of tumour cells, disturbing their cellular action mechanism, affecting the tumour microenvironment and inducing the death of cells (213). As ADCs are formed from maytansinoid drugs which are derivatives of maytansine and huC242, humanised mAbs which bind to the CanAg antigen expressed on colorectal tumours, pancreatic tumours and certain NSCLCs, huC242-maytansinoid conjugates can be disintegrated with the influence of lysosomal acidic conditions in order to release the maytansinoid, which contributes greatly to the antitumour effect of conjugates and the bystander effect (214). If the released payload is permeable, a bystander effect is produced, which means that the internalised payload enters and kills adjacent tumour cells, showing the effect of apoptotic tumour cells on bystander tumour cells (215). Although ADCs cannot directly kill the adjacent antigen-negative tumour cells, they can kill the antigen-positive tumour cells in order to indirectly kill the adjacent tumour cells by the bystander effect (216). Not only do ADCs damage the growing tumour, they can also disrupt the structure supporting tumour growth, such as tumour stromal cells and tumour vessels, in order to enhance the antitumour effect (217). More importantly, it was proposed that the activation of bystander effects by apoptotic tumour cells may be crucial to achieving the permanent eradication of tumours (215).

As an important part of ADCs, mAbs used in tumour-guided therapy must have some special characteristics. First, the ideal mAb must effectively bind to antigens on target cells so that the cytotoxic drugs are concentrated at the site of the tumour (212). Secondly, the mAbs should bind selectively to tumour cells and have little cross-reactivity with normal cells (212). If the antibody selectivity is poor or the selected antigen is present in normal tissues, cytotoxic drugs will be delivered to normal cells, resulting in targeted toxicity including allergic reactions, rashes and alopecia (218,219). In addition, the Fc fractions of some mAbs should have the affinity to bind with the Fc receptor of immune cells, thereby activating the killer effect of immunocytes (211). Thirdly, mAbs must own the ability to induce the internalisation of tumour cells, resulting in the release of the payload in the cytoplasm (220). In this way, the cytotoxic molecule can be released to extend the extent of damage to tumour cells, while mAbs play a certain antitumour role (212). Fourth, mAbs should be optimised to significantly reduce the non-specific binding of ADC drugs and prolong the half-life of ADCs in the blood (221). The immune interaction of the constant Fc fragment of an ADC is one of the major determinants of its cyclic half-life (222). As a consequence, humanised mAbs and fully human mAbs should be selected and the Fc fraction should be modified to decrease a part of immunogenicity and immunotoxicity and increase the cyclic half-life (223). Fifth, the molecular weight of mAbs should be appropriate. If the molecular weight is too large, ADCs will have difficulty penetrating the capillary endodermis and extracellular spaces (217). If the molecular weight is too small, the half-life will be influenced (224). Finally, the mAbs of ADCs should have some of the function of mAbs, including ADCC and CDC (217), which means that the mAb alone can be seen as an effective drug.

Nowadays, all ADCs in clinical trials use IgG because the biomolecule not only contains multiple natural sites for conjugation, but can also be modified to produce other conjugate sites (225). Due to their high affinity to target antigens and long circulating half-life in the blood, IgG can accumulate in the tumour region (226). Also, compared with others, IgG1 is most often chosen as the antibody part of ADCs (211). Generally, different IgG subtypes have different immune functions including ADCC and CDC (217). Compared with IgG4 and IgG2 subtypes, human IgG1 and IgG3 have stronger ADCC and CDC (217,227). Furthermore, IgG3 antibodies have a short half-life and rapid clearance compared to IgG1, IgG2 and IgG4, making them impossible to use in ADC synthesis (223). As a result, IgG1 is being used more selectively for ADC development.

In the research of tumour treatments, the development of fully human mAbs is very important, as murine mAbs and chimeric mAbs induce the immunogenicity of allogenic antibodies which can cause allergic reactions in humans (212). In early studies, the use of murine mAbs often triggered a severe immune response in humans, and patients produced human anti-murine antibodies which greatly reduced the therapeutic effect (228). Therefore, it is necessary to develop an mAb preparation technique, in order to make it possible to use better humanised or fully human mAbs as an essential component in an ADC in the future.

Traditional tumour treatments, such as surgery, chemotherapy, radiation therapy and photodynamic therapy (PDT), often damage the function of normal cells while killing diseased tissue; this can break the delicate balance between the pathogen tissue and the surrounding healthy cells (229). For example, after the photosensitiser is administered and enriched in the tumour, the PDT uses certain wavelengths of visible light to activate the photosensitiser, generating singlet oxygen to kill the tumour cells, in order to achieve a therapeutic effect (230). In the process, it is inevitable to damage normal tissues or organs because of the accumulation of photosensitisers in normal cells (231). In contrast, NIR-PIT, as a molecularly targeted phototherapy with the selective killing of diseased tissue developed on the basis of photodynamic therapy and immunotherapy, uses mAbs to direct the near-infrared, water-soluble, silicon-phthalocyanine derivative, IRdye700DX(IR700), to tumour sites, solving the problem of the low selectivity of photodynamic therapy (229,232).

As early as the beginning of the 1980s, Mew et al started studying PIT in vitro and in vivo, indicating that photo-immunoconjugates show higher selectivity to tumour tissues than photosensitisers or mAbs alone (233,234). In subsequent years, although various photosensitisers, cross-linking methods and mAbs were developed and applied to photoimmunotherapy (PIT), the application of photo-immunoconjugate was still limited in vivo due to the hydrophobic nature of the photosensitisers (235). Later, a new type of PIT was developed in 2011 by Mitsunaga et al, NIR-PIT, which uses a target-specific photosensitiser based on NIR phthalocyanine dye, IR700, in combination with mAb targeting EGFR (236). In this treatment, mAb-IR700 conjugate binds to tumour cells that overexpress antibody targets (235). When irradiated with near-infrared light, IR700 in the conjugate is activated and rapidly destroys the hydrophobic tumour cell membrane, resulting in the death of the cancer cells (229,237). In addition, the conjugate in vivo can indirectly activate cytotoxic T cells to kill tumours and inhibit tumour metastasis and recurrence by targeting CD44, CD133 and other tumour stem cell biomarkers (229,232).

In short, mAbs, as a specialised tool for identifying specific proteins on the surface of cancer cells, can act as a carrier to selectively deliver the photosensitisers which have a poor targeting ability to the tumour site, helping photosensitisers to locate and attach to cancer cells (238). More importantly, photoimmunotherapy could be applied to a range of cancers simply by altering mAbs in photo-immunoconjugates which have different targets, such as EGFR, HER2, PSMA, CD25, CEA, mesothelin, GPC3, CD47, CD20 and PD-L1 (139,229,238). Therefore, it is necessary to design and manufacture mAbs with better properties so that the PIT has a greater prospect.