Emerging immunological strategies: recent advances and future directions

Immunotherapy plays a compelling role in cancer treatment and has already made remarkable progress. However, many patients receiving immune checkpoint inhibitors fail to achieve clinical benefits, and the response rates vary among tumor types. New approaches that promote anti-tumor immunity have recently been developed, such as small molecules, bispecific antibodies, chimeric antigen receptor T cell products, and cancer vaccines. Small molecule drugs include agonists and inhibitors that can reach the intracellular or extracellular targets of immune cells participating in innate or adaptive immune pathways. Bispecific antibodies, which bind two different antigens or one antigen with two different epitopes, are of great interest. Chimeric antigen receptor T cell products and cancer vaccines have also been investigated. This review explores the recent progress and challenges of different forms of immunotherapy agents and provides an insight into future immunotherapeutic strategies.


Introduction
Immunotherapy has brought tumor therapy into a new era. From surgery, to radiotherapy, chemotherapy, and targeted therapy, immuno-oncology therapy is almost within reach. However, many obstacles remain for this treatment. Although immune checkpoint inhibitors (ICIs) are now widely studied and have shown promising clinical data, many patients receiving ICIs fail to achieve clinical benefits, show varying response rates among different tumor types [1,2], and suffer from risk of immune-related adverse events (irAEs) [3].
New approaches that promote anti-tumor immunity have recently been developed, such as small molecules, bispecific antibodies (bsAbs), chimeric antigen receptor (CAR) T cell products, and even cancer vaccines. These new drugs can be used alone or in conjunction with existing biological antibodies and traditional therapies (radiotherapy or chemotherapy) to affect various members of the immune system and microenvironment, promote antitumor effectiveness, and benefit many patients.
This review explores the mechanisms and recent advances of small molecule drugs, bsAbs, cancer vaccines, and CAR T cell therapy. Challenges and future directions of these novel immunotherapy strategies are also discussed.

Small molecules in immunotherapy
An overview With deepened understanding of innate immunity and tumor microenvironment (TME), many small molecules and their importance in cancer immunity have been discovered. Small molecule drugs include agonists and inhibitors that can reach the intracellular or extracellular targets of immune cells participating in specific immune pathways, enhancing anti-tumor immunity, or reducing immune suppression. These substances also have potential complementary or synergistic effects with existing immunotherapy. Compared with therapeutic antibodies, small molecule drugs are more permeable to tissues and the TME, and can cross the blood-brain barrier and other physiologic barriers, thus providing new options for the treatment of brain tumors and brain metastases. By adjusting the pharmacokinetic and pharmacodynamic parameters, small molecule drugs may provide the best bioavailability and avoid some of the irAEs associated with long-lasting antibody therapies. These medications also have relatively low production costs and are usually taken orally which enables easy administration.
Although a growing number of small molecules have entered early phase clinical trials, many challenges remain to be solved. Specific issues relate to understanding their mechanisms of action in the immune system and the theoretical basis for further clinical applications, as well as, the need for more safety and efficacy evaluations.

Mechanisms of small molecule drugs
Over the past decade, more than 50 small molecule drugs have been produced as single agents or in combination with monoclonal antibodies for tumor immunotherapy [4], and over 100 clinical trials are currently underway (Table 1 and Fig. S1). Small molecule agonists and inhibitors target specific pathways participating in innate or adaptive immunity through different mechanisms (Fig. 1). Understanding the mechanism of small molecule drugs and their current clinical research progress will aid in exploring their role in immunotherapy. occurrence of irAEs must be anticipated and monitored.
The advantages of small molecules are permeabilization, oral delivery, and dose modulation, which promote the development of small molecule inhibitors acting on the PD-1/PD-L1 pathway [5][6][7]. Companies Bristol-Myers Squibb (BMS) and Aurigene are leading the development of small molecule PD-L1 inhibitors, with molecules such as BMS-103, BMS-142, BMS-1166, CA-327, and CA170. Small PD-L1 inhibitors developed by BMS can induce the PD-L1 dimer by filling a deep hydrophobic channel-like pocket between two PD-L1 molecules and then blocking PD-1 binding [8,9]. Oral molecule, CA-327, shows anti-tumor activity in preclinical cancer models by inhibiting the PD-L1 and T cell immunoglobulin domain and mucin domain-3 (TIM-3) [10].
Developed by Aurigene, CA170 is an oral inhibitor that targets PD-L1 and the V-domain Ig suppressor of T cell activation (VISTA), and was reported as the pioneer of oral immunotherapy drugs among small molecule checkpoint inhibitors [11]. A CA-170 phase 2 clinical study is currently ongoing with data obtained from 15 non-small cell lung cancer cases and notable tumor reductions noted in six patients [12]. However, the affinity of small molecules to the target is worse than that of antibodies. Hence, off-targeting may occur and result in reduced efficacy and toxicity. Further mechanism explorations and clinical efficacy evaluations are needed. Although small molecule immune checkpoint inhibitors are mostly in preclinical and early clinical stages, these drugs will open a new avenue for tumor immunotherapy because of their pharmacokinetics and druggability advantages.

Targeting innate immunity
Pattern recognition receptors are key members in innate immunity that can distinguish pathogen-associated molecular patterns and promote T cell effector function [13]. Toll-like receptor (TLR) 7/8 is located in the endosome of cells. By improving the identification of foreign organisms, small molecule TLR agonists activate immune response. Imiquimod, a TLR7 agonist developed as topical cream by the Minnesota Mining & Manufacturing Company (the United States) has been used for superficial basal cell carcinoma [14]. This drug has also shown anti-tumor activity in a phase 2 clinical trial for patients with bladder cancer [15]. Motolimod (VTX-2337), an agonist of TLR8, can mediate the release of IL-18 and activate natural killer (NK) cells [16]. Resiquimod (R848), a TLR7/8 agonist, helps macrophages acquire an anti-tumorigenic phenotype [17]. These TLR7/8 agonists are mostly in phase 1/2 clinical trials ( Table 1).
Stimulator of interferon genes (STING) participate in the innate immune recognition of immunogenic tumors [18]. The activation of the STING pathway contributes to tumor regression in mouse models [19]. STING agonists might also improve the activation of dendritic cells (DCs) and T cells [20]. In June of 2019, Aduro announced the results of a phase 1b clinical trial for a small molecule STING antagonist (ADU-S100) combined with spartalizumab. However, only 6 out of the 83 patients with lymphoma or advanced solid tumors exhibited remarkable responses [21]. In hope of achieving relatively improved results, Aduro is currently preparing to combine ADU-S100 and Keytruda for head and neck cancers in a phase 2 clinical trial. The STING small molecule antagonist MK-1454 is also in a phase 2 clinical trial (NCT04220866).
In addition to antibodies for checkpoint modulation and cell therapy, pattern recognition receptor agonists and STING agonists provide a new approach for small molecules to prompt innate immune members to contribute to anti-tumor immune strategies. Although TLR agonists are promising targets that may exhibit synergistic effects with existing immunotherapy strategies, future research must consider that the TLR pathway is associated with gastric and pancreatic tumorigenesis [22,23]. Additional studies are required to further assess the safety of these small molecule agonists.

Targeting amino acid metabolism
The TME contains diverse immunocytes. Tumor-associated macrophages (TAMs) support tumor invasion and metastasis. Treg cells and myeloid-derived suppressor cells (MDSCs) are linked to immunosuppression in the TME. Small molecule drugs navigating metabolic pathways might strengthen the anti-tumor immunity by metabolical reprogramming of tumor and immune cells in the TME [24].
Indoleamine-2,3-dioxygenase-1 (IDO1) participates in the degradation of tryptophan to kynurenine, and selective inhibition of IDO1 enhances NK cell proliferation and reduces conversion to Treg cells [25]. BMS-986205 is one highly-efficient oral IDO1 inhibitor that can shrink bladder tumors when combined with ICIs from a phase 1/2a study [26]. IDO1 inhibitor navoximod has also shown acceptable safety and tolerance in a phase 1 clinical trial of advanced solid tumors, but its combination with atezolizumab was not beneficial [27]. A recent phase 3 trial, ECHO301, tested the efficacy of IDO1 inhibitor epacadostat combined with pembrolizumab in melanoma; however, the reaction was not better than that for pembrolizumab alone [28].
Small molecule arginase 1 (ARG1) or inducible nitric oxide synthase (iNOS) inhibitors targeting MDSCs or TAMs might overcome immunosuppression and aid the restoration of immune function [29]. ARG1 inhibitor CB-1158 promotes the production of inflammatory cytokines and increases CD8 + T cell tumor infiltration [30]. CB-1158 is now under phase 1/2 clinical trials and is also being combined with a small molecule PD-1 blockade (Table 1). Transient treatment with CB-839, an inhibitor of glutaminase 1, also enhances cytotoxic lymphocyte-mediated antitumor responses [31].
Treatments targeting the amino acid metabolism of tumor and/or immune cells in the TME can produce a synergistic effect with existing immunotherapy approaches. However, the unexpected efficacy of IDO1 inhibitor epacadostat combined with pembrolizumab in clinical trial suggests that much efforts are need to further understand the metabolic mechanisms of immune cells to improve the effectiveness of combination therapies.

Targeting adenosine signaling
Ectonucleotidases CD73 and CD39 participate in the dephosphorylation of adenosine triphosphate to produce adenosine, which binds to the Adora2a (A2A) receptor, activates adenosine signaling, and amplifies the immunosuppressive effects of Treg cell [32]. In preclinical studies, the efficacy of ICIs have been enhanced using a combination of A2A receptor antagonists [33]. Preliminary evidence from a phase 1b clinical trial showed that A2A receptor inhibitor CPI-444 combined with atuzumab exhibits disease control in refractory renal cell carcinoma [34]. Other phase 1/2 studies have also assessed the safety of A2A receptor antagonists used alone or combined with ICIs in advanced tumors (Table 1). Given the immunosuppressive role of adenosine signaling in the TME, small molecule antagonists targeting A2A receptor show potential as therapeutics.

Targeting cytokine signaling
Small molecules can regulate the tumor immune response by influencing specific cytokine-mediated pathways. Retinoic acid receptor-related orphan receptor gamma t (RORγt) is a member of the nuclear receptor superfamily of transcription factors and plays an important role in the differentiation of cytokine interleukin-17 expressing immune cells [35]. RORγt agonists enhance anti-tumor immunity by activating Th17 cells and reducing Treg proliferation [36]. RORγt agonist, LYC-55716 in combination with an ICI, is currently undergoing a phase 1 clinical trial (Table 1).
Galunisertib, a transforming growth factor-beta (TGF-β) receptor I inhibitor, suppresses Smad family member 2 phosphorylation and was granted orphan drug designation for the treatment of liver cancer by the European Medical Agency and the FDA in the United States in 2013 [37]. Galunisertib combined with a PD-L1 blockade can enhance the expression of immune-related genes and modulate T cell immunity in colorectal and breast cancer mouse models [38].
Although the relationship between these cytokines and immune regulation has been established, only a few of these drugs are currently undergoing clinical trials, possibly because they mediate complex signaling pathways. Their effects on tumor cells and immune cells in the TME and the risks of combination drugs must be paid attention.

Targeting oncogenic phosphatases and kinases
Phosphatases and kinases that regulate signal transduction are potential targets for small molecule drugs. Src homology-2-containing protein tyrosine phosphatase 2 (SHP2) is involved in the downstream signaling of PD-1, which suppresses T cell function [39]. Owing to its crucial role in T cell activation, SHP2 has emerged as a treatment strategy. In colon cancer xenograft models, SHP2 inhibitor SHP099 combined with an anti-PD-1 antibody showed better reducing ability for tumor load than monotherapy [40]. The SHP2 inhibitor RMC-4630s is currently under phase 1/2 clinical trials, and its pharmacokinetic profile and safety are also being evaluated (Table 1).
Colony stimulating factor 1 receptor (CSF1R) is activated by phosphorylation; pexidartinib, an oral CSF1R inhibitor, decreases TAMs and increases CD8 + T cells when used in combination with a dendritic cell cancer vaccine in mesothelioma mouse models [41]. Two clinical trials of pexidartinib monotherapy and two clinical trials of pexidartinib combined with monoclonal antibodies in advanced tumors are currently ongoing.
These small molecule drugs targeting phosphorylases and kinases usually affect tumor cell signal transduction. Additional research is needed to clarify their overall influence on tumor and immune cells prior to clinical trials.

Targeting chemokine receptors
The chemokine superfamily consists of a large number of ligands and receptors that participate in the homing, retention, circulation, and activation of immune cells [46]. C-C chemokine receptor (CCR) 2 inhibitor PF-04136309 depletes macrophages and inflammatory monocytes from the primary lesion and premetastatic liver, thereby enhancing antitumor immunity, depressing tumor growth, and reducing metastasis [47]. Inhibiting C-X-C chemokine receptor (CXCR) 4 may also reduce the accumulation of macrophages in the TME [48]. Plerixafor, a CXCR4 antagonist, has achieved good results as a chemosensitizer in phase 1/2 leukemia clinical trials [49]. Other ongoing clinical trials of small molecule drugs targeting chemokine receptors have focused on CCR2/5 antagonist BMS-813160, CCR4 inhibitor FLX475, and CXCR2 antagonist AZD5069 ( Table 1).
The small molecule targeting of chemokine receptors is often used in combination with ICIs and chemotherapeutics in clinical trials. Given the important role of chemokines in the TME, the combinational strategies may provide meaningful clinical benefits. At present, numerous small molecule drugs have been developed to target the extracellular or intracellular pathways in adaptive or innate immunity; however, most of them are in the early stage of clinical trials. Additional basic experiments and clinical trials are urgently required to clarify their mechanism, clinical efficacy, and pharmacokinetics.
BsAbs are utilized in various ways, including receptoractivation, receptor-blocking, receptor-internalization, receptor-clustering, or retargeting of cytotoxic effector cells [55]. Cancer is a complicated and polyfactorial disease. Compared with monospecific monoclonal antibodies, bsAbs can synchronously bind two individual epitopes or antigens for greater impact and better treatment effects. Multi-combined regions in one antibody could help regulate diverse functional pathways in cancer, thus avoiding drug resistance and decreasing the side effects on intravital tissues [56][57][58][59].
With the rapid development of gene engineering antibodies and immunology, the construction, technology platform, product research, and development of bsAbs are continuously being innovated at high speed. BsAbs are expected to be the next generation of biological therapeutics for tumors, autoimmune illness, contagious diseases, diabetes, Alzheimer's disease, and osteoporosis [51,60]. However, several challenges have been encountered during their development, namely, how to prevent poisoning and immunogenicity due to neo-antigenic determinants, how to meet the threshold for sensitizing diverse molecular mechanisms, and how to ensure the manufacturing quality [61].

Preparation method of bsAbs
BsAbs contain two different antigen binding domains that cannot be found in nature and can only be prepared artificially. Chemical coupling [62], two-hybrid method [63], and genetic engineering [64] are the most common preparation techniques for bsAbs. The most attractive application is the realization of new biological functions and therapeutic mechanism of action (MOA). However, new MOAs pose undiscovered risks that cannot be estimated in preclinical research. The indeterminacy over their safety is the major hurdle in the exploration of bsAbs. Molecular imaging studies could be used to create predictive models for the pharmacokinetic parts of bispecific constructs and to develop optimal dosing strategies [65].

Structure types
The basic structure of bsAbs consists of two pairs of heavy-light polypeptide chains connected by interchain disulfide and noncovalent bonds resembling a "Y" shape compound, including antigen binding fragments (Fabs) and a fragment crystallizable region (Fc). BsAbs could help immune cells target tumor cells by binding to one surface antigen expressed on cancer cells and to a second antigen expressed on immune cells, such as NK cells or effector T cells. The fusing of the antitumor binding domain with the Fc receptor (FcR) or the anti-CD3 binding domain may help produce bsAbs that can recruit immune cells. FcR is the terminal area of the antibody that interplays with the neonatal receptor, which results in lethal immune-mediated effects [66,67].
BsAbs can be divided into two categories according to their structure: one contains the Fc region, and the other lacks the Fc region. These types can be further classified into asymmetric IgG-like bsAbs, symmetric IgG-like bsAbs, and non-IgG-like bsAbs [68]. IgG-like bsAbs can achieve effector functions, and non-IgG-like binding antibody (bAbs) are diminutive, which can improve penetration. IgG-like bAbs contain three arms/binding sites: two Fab arms and an Fc arm. The IgG-like bsAb structure promotes Fc domain-mediated effects and defends the physical properties endowed by the FcR [69,70]. A unique kind is asymmetric IgG-like bsAbs that possess an integrated Fc and a couple of distinguishing arms combining different antigens; some examples include M802, M701 [51], KN026 [71], MBS301 [72], IBI318 [73], IBI315 [74], and KN046 [75].

Mechanisms of bsAbs
BsAbs have manifold targets and special MOA.

T cell redirection
BsAbs characteristically target the antigen connected to T cells. By bonding to T cells and cancer cells, they can redirect the toxicity of effector T cells and obliterate cancer cells [85,86].
Co-localized blockage SHR-1701 can simultaneously block the PD-L1 immune checkpoint and TGF-β on cancer cells. The aforementioned combination therapy could also increase the antineoplastic effect compared with mono-treatment in cancer cell pathways [95,96].

Research status
Many multinational pharmaceutical companies and biotechnology companies have committed to developing bsAb-related drugs. Many Chinese companies are also involved in the research and development of bsAbs, some of which have entered the clinical or clinical application stage.
More than 100 bsAb constructions and 200 clinical trials and over 30 technology research platforms, including CrossMab (Roche), CRIBTM (China), ItabTMv (China), and FIT-IgTM (China), have been conducted over the past decade [61]. Despite starting later than other countries, Chinese bsAb development has rapidly progressed. By using the aforementioned bsAb technology research platforms, China has created 18 bsAb structures and initiated 25 clinical trials ( Table 2). As of August 2020, PD-L1 and CTLA-4 are the most commonly studied targets in China [106]. In particular, 10 bsAb and 41 clinical trials were noted for both China and other countries (  (Table 4).
AK104 (PD-1/CTLA-4) is under a recent phase 2 multicenter study on advanced gastric adenocarcinoma. The common targeted cancer bsAb simultaneously blocks the PD-1 and CTLA-4 immune regulatory checkpoints, resulting in the potential suppression of double checkpoints and antineoplastic activity [107]. Developed by Alphamab, KN046 (PD-L1/CTLA-4) is currently in a phase 3 trial. Some studies have recently reported treatment-related toxic side effects of anti-CTLA-4 antibody [108,109]. Compared with each parental mAbs, KN046 can improve the safety and efficacy [110].
Developed by Biokin, SI-B001 is an anti-HER3 Â anti-EGFR bsAb that is currently in a phase 1 trial and could firsthand activate the downstream paths and inhabit tumorigenesis [100,101].
BsAbs have great clinical potential because of their unique characteristics that cannot be found in monoclonal antibodies. Most bsAbs are in clinical or preclinical research. Adverse reactions, such as cytokine storms, neurotoxicity, and production processing, are the main problems for this therapy. Designing a reasonable antibody structure according to different effect mechanisms is the focus of bsAb research and development. The continued development of clinical studies and advances in upstream and downstream technology will hopefully help to solve these bsAb-related problems.   linked to an intracellular domain without any costimulatory domain. However, no promising antitumor response was observed largely due to the lack of adequate activation [111]. As a solution, second-and thirdgeneration CARs are being developed by adding one or two co-stimulatory domains, respectively, to enhance their activity [112,113]. Second-generation autologous (patient-derived) CAR T cell therapy has changed the treatment of hematologic malignancies; four CD19-targeting CARs have achieved   [114][115][116][117]. Clinical trials are also ongoing, and CAR T cells specific for CD30 (CD30-targeting CARs) have shown potential to treat Hodgkin's lymphoma (HL) in two phase 1/2 clinical trials (NCT02690545, NCT02917083) [118]. A clinical trial of anti-CD7 universal CAR-T (U-CAR-T) cells indicated that patients with T cell lymphoma displayed robust CAR-T cell expansion (NCT04264078) [119]. In a phase 1/2 clinical study (NCT01869166), anti-EGFR CAR-T cells were found to be a feasible therapeutic strategy for EGFRpositive patients with NSCLC [120]. However, the success of CAR T cell therapy is yet to be applied clinically. Several impediments have been encountered, namely, poor availability of tumor specific antigens, immunosuppressive characteristics of the TME, and variability in manufacturing quality and high processing costs [121][122][123]. The use of "off-the-shelf" allogeneic CAR T cells from healthy donors could potentially overcome these issues. Allogeneic T cells are primarily derived from peripheral blood mononuclear cells, embryonic stem cells, and induced pluripotent stem cells. Allogeneic CAR T products can markedly decrease the costs owing to industrialized and scaled-up production, thereby rendering CAR T treatment immediately accessible to a large number of patients due to the batch manufacturing of cryopreserved T cells. The use of allogeneic cells would also provide a high-quality product based on donor selection and allow for standardized dosing and re-dosing and a combination of CAR targets [122,123]. Other major issues must be addressed, including toxicities such as graft versus host disease (GVHD) and limited anti-tumor efficacy against solid tumors. Various safeguarding strategies, such as applying non-αβ T cells including γδ T cells [124], gene editing with αβ T cell receptor (TCR) deletion [125], and using virus-specific T cells [126] or donor-derived allogeneic T cells [127], are needed to improve the clinical safety of CAR T cell therapy. All these techniques have been designed to specifically reduce GVHD toxicity.

Chimeric antigen receptor (CAR) T cell therapy
Although CAR T cell therapies have shown unsatisfactory efficacy in solid tumors, many promising methods can be applied for optimization. Improving CAR T structures [128] and combining with different treatment strategies such as chemotherapy [129], local therapy [130], checkpoint blockades [131], bsAbs [132], epigenetic modulators [133], vaccines [134], and oncolytic viruses [135] have all been explored to enhance the persistence and antitumor activity of CAR T cell therapy.
Despite the bumpy road ahead, the future of CAR T cell therapy looks promising because of the continuous evolution of advanced gene editing techniques and novel solutions. These innovations will help "off-the-shelf"   allogeneic CAR T cell therapy to be effective, safe, and perhaps even revolutionize cancer treatment.

Therapeutic cancer vaccines
Cancer vaccines trigger immune responses against tumor cells by amplifying and broadening antigen-specific T cells [136]. Tumor antigens, immune adjuvants, delivery vehicles, and formulations are the four key components of therapeutic vaccines and are vital for efficacy. Tumor antigens can be delivered in the form of genetic vaccines (DNA/RNA/viral), protein/peptide vaccines, and cell vaccines. Delivery method is also a major factor influencing vaccine efficacy [136]. Antigens for tumor vaccines include TAAs and tumorspecific antigens (TSAs). Early cancer vaccines focused on TAAs, self-antigens that have elevated levels on tumor cells but may also be expressed on normal cells. However, TAAs lacking tumor specificity increase the risk of autoimmune toxicities and have been unsuccessful in generating effective antitumor immune responses due to immune tolerance [136,137].
TSAs comprise antigens expressed by neoantigens or oncoviruses and are found exclusively in cancer cells. Neoantigen-based cancer vaccines are tumor-specific, can enhance a tumor-specific T cell response, and prevent toxicities caused by "off-target" damage. Recent development on bioinformatics technologies has enabled the systematic identification of tumor neoantigens; several promising studies have explored neoantigen cancer vaccines [138]. In a phase 1 clinical trial, Ott et al. reported a neoantigen vaccine that was formulated with up to 20 personized HLA-A/B-restricted peptides and has expanded neoantigen-specific T cells in patients with melanoma (NCT01970358) [139]. After a 4-year median follow-up of neoantigen vaccine therapy, a persistent T cell response was observed in patients with melanoma [140]. Neoantigen-specific T cells from peripheral blood also show the potential to migrate into intracranial tumors in glioblastoma after surgical resection cases in a phase 1b clinical trial (NCT02287428) [141]. These initial studies suggest that neoantigen-specific cancer vaccines are safe in patients with melanoma and glioblastoma. For further understanding on their therapeutic efficacy, in-depth studies must be conducted on the function of vaccineinduced T cells and the persistence of neoantigen-specific memory T cells.
Most therapeutic cancer vaccines are in ongoing trials, and their development can possibly enhance the efficacy of immunotherapy. In a phase 1b study of a neoantigen-based peptide vaccine NEO-PV-01 in combination with a PD-1 inhibitor, epitope spreading was detected post-vaccination and correlated with improved progression free survival in patients (NCT02897765) [142]. Compared with sunitinib monotherapy, sunitinib in combination with ilixadencel, a cell-based allogeneic off-the-shelf product, exhibited a higher overall response rate in patients with synchronous metastatic renal cell carcinoma [143]. In a clinical trial of personalized tumor lysate-pulsed DCs for patients with recurrent ovarian cancer, a vaccine plus therapy seemed to improve the overall survival compared with a low-dose cyclophosphamide and bevacizumab combination therapy [144].
Although the above preliminary findings are encouraging, numerous challenges remain to be addressed. First, further discovery of personalized neoantigen targets is required to maximize their effects. Second, delivery strategies are an important factor affecting vaccine efficacy; the effectiveness of different delivery methods varies among tumor types. Finally, when a vaccine is being combined with existing treatment approaches, the timing, sequence, and dose of combination therapy must be further explored.

Challenges and future direction
New immunotherapeutic approaches provide opportunities for further drug development and bring benefits to patients. However, challenges persist during their development. Therefore, further basic and clinical research is needed.

Assessment of combination therapy
Given that anti-tumor immunity involves various steps, rational combinations to modulate different biological steps might strengthen anti-tumor responses. Effective transformation from basic discovery to clinical application could be achieved by exploring the molecular mechanisms and optimizing the strategies and timing of combination therapy to maximize its effects. The combination of four components (anti-PD1 therapy, tumor antigen-targeting antibody, interleukin-2, and a T cell vaccine) that engage in innate and adaptive immune responses was reported to eliminate large tumors in mouse model [145]. However, most drugs are in the early stages of clinical trials with complicated combinations and pose various challenges, specifically how to maximize their synergistic effects and how to avoid combinational toxicities. For a partial solution, MORPHEUS and FRACTION platforms were designed to evaluate the safety and effectiveness of combination immunotherapies in multiple phase 1b/2 trials [146,147]. A novel Quick efficacy seeking trial (QuEST1) was also designed to assess different immunotherapy combinations in patients with prostate cancer [148]. The rational selection of the combination and dosage based on known molecular mechanisms to maximize their synergistic effects is yet to be elucidated.

Validated biomarkers
Over 3000 interventional clinical trials of immunological drugs either alone or in combination are being conducted globally [149]. Nevertheless, the clinical benefits of many novel immunotherapies cannot be determined at this stage. Strategies for the identification of valid biomarkers are essential in identifying patients who will benefit the most. Many current clinical trials include the detection of serial sampling of peripheral blood or tumor specimens for the analyses of corresponding biomarkers (such as NCT01928576 and NCT03220477). In a phase 2 study of immunotherapy combinations of motolimod and doxorubicin for ovarian cancer, statistically significant differences in the overall survival of motolimod-treated patients were observed in a subgroup of patients who experienced injection site reactions; this investigation may provide biomarkers to evaluate the efficacy of combinational immunotherapies [150]. Owing to the complex interactions required for effective treatment, the development of actionable information and identifying feasible markers that can accurately classify patients is imperative.

Autoimmune toxicities
The mechanisms of immune-related toxicities must be understood to produce the best personalized treatment approach. Small changes in the molecular structures of small molecules may lead to tremendous variations in efficacy and toxicity. Diverse challenges have emerged during the exploration of bsAbs, such as reducing toxicity and immunogenicity induced by neo-epitopes, satisfying thresholds for sensitizing various molecular pathways, and assuring the quantity and quality of bsAbs.
The application of CAR T is also not without concerns. This treatment can lead to adverse effects, such as cytokine release syndrome and on-target off-tumor toxicity. Early recognition of cytokine release syndrome and aggressive steroid administration in CAR T treatment are important [151]. Moreover, drug-drug interactions must be considered for the toxicities of combination treatments. In a phase 2 study, the combination of pembrolizumab plus oral azacitidine CC-486 was associated with an increase in treatment-related adverse events compared with the pembrolizumab plus placebo group. This phenomenon can be attributed to the intestinal and hematological toxicities noted for the oral formulation of azacitidine [152].

Improving manufacturing practices
The production of bsAbs, CAR T cells, and neoantigenbased vaccines is expensive and time consuming. In the development of biological products, optimizing the structures and workflows according to the biological mechanisms requires special attention. Designing a reasonable antibody structure according to different effect mechanisms is the focus of current bsAb research and development.
Complete CAR T cell therapy is complex compared with autologous products; however, allogeneic CAR T products offer the advantages of industrialized production and low costs [122]. Manufacturing some biological products is also time consuming, and the production of personalized vaccines is more expensive than off-the-shelf therapeutic agents. Vaccine preparation usually takes 3-5 months at best [139]. Technological developments, such as automated flow peptide production, might help promote peptide manufacturing and decrease the production time of personalized vaccines [153]. For these emerging immunotherapy drugs, their research, development, and production time and costs must be considered.

Conclusions
Immune checkpoint therapies, such as PD-1, PD-L1, and CTLA-4 antibodies, have made considerable headway in tumor treatments for the past decade. However, only a small number of cases respond to immunotherapy and are often accompanied by adverse reactions. Therefore, new treatment options are essential to enhance immunotherapy efficacy, overcome immunosuppression, and reduce toxicity. Understanding novel immuno-oncology therapeutic strategies allow us to provide additional opportunities for patients with advanced cancer. Small molecule drugs, bsAbs, CAR-T treatment, and cancer vaccines provide appealing avenues for immunotherapy. Related preliminary preclinical and clinical studies are already underway. Cost of treatment, lack of biomarker responses, and combination therapies targeting different immune mechanisms remain as challenges to be overcome. Nevertheless, these emerging strategies can bring about new opportunities for patients with cancer.

Compliance with ethics guidelines
Hongyun Zhao, Fan Luo, Jinhui Xue, Su Li, and Rui-Hua Xu declare that they have no conflicts of interest. This manuscript is a review article and does not involve a research protocol requiring approval from relevant institutional review board or ethics committee.

Electronic Supplementary Material
Supplementary material is available in the online version of this article at https://doi.org/ 10.1007/s11684-021-0886-x and is accessible for authorized users.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.