Promising Molecular Targets and Novel Therapeutic Approaches in Neuroblastoma

This article provides a brief and up-to-date overview of promising molecular targets and novel therapeutic approaches in neuroblastoma (NB). High-risk NB is hard to manage with existing treatment modalities, so more than half of those cases are unable to achieve long-term survival. With a deep understanding of molecular pathogenesis, numerous therapeutic targets have been discovered, offering a wide range of novel strategies to treat high-risk NB. Several molecular targets or pathways of NB are well studied, such as GD2, MYCN, ALK, p53/MDM2, PI3K/Akt/mTOR/, and RAS/MAPK signaling. Novel targeted drugs and combined therapies are being developed and investigated for treating high-risk NB in preclinical and clinical trials. Considering different NB patients respond to molecular-guided therapy and conventional therapy differently, how to design an effective personalized therapy remains a big challenge. Anti-GD2 monoclonal antibodies have been approved to treat high-risk NB. Inhibitors targeting MYCN, ALK, p53/MDM2, RAS/MAPK, and PI3K/Akt/mTOR are being tested in phase I/II clinical trials. However, most research on molecularly targeted therapy stays at the preclinical level. More valuable targets need to be identified, and more efficient therapies need to be developed. Further, exploration of new combinations using inhibitors targeting multiple targets and conventional therapy is still the most important research direction in future, which would advance treatment regimens, improve outcomes, and prolong survival in children with high-risk NB.


Introduction
Neuroblastoma (NB) is an embryonal tumor arising from the adrenal medulla and the sympathetic nervous system. This disorder is the most common extracranial solid tumor of childhood, and 90% of the cases occur in young children before the age of 5 years, particularly in infants (under 1 year old) [1]. The incidence is 10.7 cases per million persons aged 0-14 years, with approximately 700-800 new patients reported annually in the USA [1,2]. The median age at presentation is 22-23 months, and less than 10% of the cases are diagnosed after the age of 5 years [1,2]. Indeed, it rarely presents in adolescence and adulthood, but this age population has a worse prognosis [2]. NB accounts for 8-10% of all pediatric malignancies and is responsible for 15% of all cancer-related mortality in children [3]. It is worth noting that NB is highly heterogeneous in terms of pathogenesis, clinical presentation, and prognosis. NB is considered as a spectrum of diseases other than a single pathological condition, as no solo molecular aberration contributes to all patients [4••]. Based on a number of prognostic factors, such as age at onset, histological category, grade of tumor differentiation, INRG (International Neuroblastoma Risk Group) stage, MYCN (MYCN proto-oncogene, bHLH (basic helix-loop-helix) transcription factor) amplification, 11q aberration, and DNA ploidy, this disease can be divided into four categories: very low-, low-, intermediate-, and high-risk [5]. Remarkably, high-risk patients, accounting for around 50% of all NB cases, have the worst prognosis, with the 5-year event-free survival (EFS) of 51% [6, 7•]. In contrast, the outcome of low-risk or intermediate-risk cases is much better with the 5-year EFS of larger than 85% [7•, 8]. Conventional treatments for NB include surgical tumor resection, systemic chemotherapy, radiation therapy, 13-cis retinoic acid (isotretinoin) differentiation induction therapy, autologous hematopoietic stem cell transplantation, and immunotherapy. However, even with intensive multimodal therapeutics, approximately 60% of high-risk patients eventually relapsed [9]. Up to 20% of children with high-risk NB have suboptimal responses to induction therapy because of drug resistance [10]. Furthermore, the 5-year overall survival (OS) of cases at INSS (International Neuroblastoma Staging System) stage 4 is only about 42% [8]. Thus, exploring potential targets and developing novel treatments are extremely important to improve clinical outcome for highrisk NB.
In recent years, with a deep understanding of molecular etiology of NB, therapeutic markers have been discovered, and targeted drugs have been being examined in clinical trials. Here, we focused on several well-studied molecular targets and relevant targeting therapeutics, ranging from disialoganglioside (GD2), MYCN and anaplastic lymphoma kinase (ALK) to p53/ mouse double minute 2 homolog (MDM2), phosphoinositide 3-kinases (PI3K)/protein kinase B (PKB, also known as Akt)/mammalian target of rapamycin (mTOR), and rat sarcoma virus (RAS)/mitogen-activated protein kinase (MAPK) ( Table 1), providing a brief and upto-date summary of such findings in this review.
GD2 is an exciting, non-protein target for immunotherapy, as it is highly expressed on the surface of NB cells, with a relatively low expression level in normal tissues (e.g., neurons, skin melanocytes, and peripheral nerve fibers), and all NB cases express GD2 virtually despite grading and staging of tumor [4••, 38, 39]. Here, we mainly discuss anti-GD2 monoclonal antibody therapy and anti-GD2 CAR T cell therapy.
Anti-GD2 monoclonal antibody therapy is the most widely used immunotherapy for NB in recent years. Importantly, anti-GD2 monoclonal antibodies have different types including murine, chimeric, or humanized antibodies, which are extensively investigated in preclinical and clinical trials. Those novel drugs are currently used in the disease management, such as chemotherapy induction and retinoic acid maintenance. For example, dinutuximab (ch14.18) and dinutuximab beta (ch14.18/CHO), chimeric antibodies containing murine anti-GD2 variable regions of IgG3 and human constant regions of IgG1, have been approved to treat highrisk NB by the US Food and Drug Administration (FDA) in 2015 and by the European Medicines Agency (EMA) in 2017, respectively [11,12]. In addition, a number of humanized monoclonal antibodies with low immunogenicity, including Hu14.18K322A and naxitamab, are being evaluated in clinical trials [14 •, 40]. Notably, naxitamab has been approved to treat relapsed or refractory high-risk NB children with the age older than 1 year by the FDA in November 2020 [13•].
Given the biological roles of GD2 are poorly understood, the mechanism of anti-GD2 monoclonal antibodies against NB is still not clear. However, emerging evidence highlighted that anti-GD2 monoclonal antibodies may regulate immune system as well as survival, apoptosis, and invasion of tumor cells (Fig. 1) [41, 42•, 43, 44]. By way of illustration, the monoclonal antibodies binding to the highly expressed GD2 in cells can trigger dual immune responses that are antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) [41]. Subsequently, ADCC and CDC perform an anti-tumor function [41]. This type of antibodies can also restore immune activation through the block of immune checkpoints [44]. Intriguingly, the anti-GD2 monoclonal antibodies are able to directly regulate survival, invasion, and apoptosis by targeting multiple molecules or pathways, such as PI3K/Akt/mTOR and MYCN/AURKA, which are not associated with immune mechanisms [43].
To improve the efficacy of anti-GD2 monoclonal antibody therapy, the combination of antibodies and cytokines has been explored recently. Data suggested that granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-2 (IL-2) in cooperation with dinutuximab could enhance ADCC with IL-2 promoting NK cells and monocytes and GM-CSF stimulating neutrophils and macrophages [11]. Meanwhile, this regimen may also reduce FDA-approved for high-risk NB [11,12] Dinutuximab beta Human-murine chimeric mAb targeting GD2 EMA-approved for high-risk NB [12] Naxitamab (hu3F8, Danyelza, naxitamab-gqgk) Humanized mAb targeting GD2 FDA-approved for r/r high-risk NB in the bone or bone marrow >  [32] the dose and the toxicity of monoclonal antibody against GD2 [11]. One randomized phase III trial (NCT00026312) demonstrated that the synergy of dinutuximab with IL-2/ GM-CSF improves the survival of high-risk NB, compared to single drug treatments [45•]. Nonetheless, the multicenter phase III trial (NCT01704716) in 2018 showed that subcutaneous IL-2 does not increase the efficacy of dinutuximab beta for treating high-risk NB [46]. Another phase II trial (NCT01334515) uses the hu14.18-IL2 (the humanized 14.18 anti-GD2 monoclonal antibody linked to IL2) in combination with GM-CSF and isotretinoin to treat patients with refractory/relapsed NB, suggesting this combination strategy may be tolerable and suitable for cases with non-bulky disease [16].
Regarding the adverse side effect of anti-GD2 monoclonal antibodies, neuropathic pain is the most significant. This is mainly because the complement is activated when monoclonal antibodies bind to GD2-expressed peripheral nerve fibers. A few novel antibodies against GD2 have been designed to reduce neuropathic pain. For instance, Hu14.18K332A with FC mutations decreases complement fixation and lowers allodynia [47]. Another antibody drug (8B6 monoclonal Fig. 1 The functions of GD2 in NB. GD2 overexpression increases the levels of N-MYC, AKT, mTOR, 4E-BP1, p70-S6 kinase, and AMPKβ subunits, consequently resulting in enhanced survival in NB. Meanwhile, high GD2 expression downregulates the levels of p53, PHLDA1, and caspase 9, leading to apoptotic inhibition. GD2 also modulates PHLDA1, SVIL, RASSF6, JUN, ID1, TLX2, and CDKN1A gene expression, which causes abnormal cell differentiation and RNA metabolic processes in tumor. In addition, GD2 can be released from tumor cells into the circulation or the microenvironment, which suppresses T cell proliferation, induces T cell apoptosis, and blocks CD34 + cells maturation into dendritic cells. Anti-GD2 monoclonal antibodies binding to GD2 triggers the activation of ADCC and CDC, subsequently resulting in apoptosis and cytolysis. NB, neuroblastoma; AKT, protein kinase B; mTOR, mammalian target of rapamycin; 4E-BP1, eukaryotic translation initiation factor 4E (eIF4E)-binding protein 1; AMPK, AMP-activated protein kinase; PHLDA1, pleckstrin homology-like domain family A member 1; ADCC, antibodydependent cell-mediated cytotoxicity; CDC, complement-dependent cytotoxicity antibody) highly selectively binds to O-acetyl GD2, and does not cause neuropathic pain, as O-acetyl GD2, a derivative of GD2, is expressed together with GD2 but not expressed in peripheral nerves [48]. Although advances have been made, more studies are required to evaluate the efficacy, the toxicity, and the side effect of these newly developed drugs in NB.
Engineered chimeric antigen receptor (CAR) T cell therapy is another potent targeted GD2 therapy, which integrates the specificity of an antibody with the cytolytic capacity of T cells [49]. The anti-GD2 CAR T cells have advantages, such as increasing potency and crossing the blood-brain barrier, in comparison with monoclonal antibody therapy [50••]. However, anti-GD2 CAR T cell therapy has the same adverse side effect with antibodies, which is neurotoxicity [49]. Recently, the phase I clinical trial (NCT02761915) showed that anti-GD2 CAR T cell therapy is well tolerated without on-target off-tumor toxicity but does not achieve objective clinical response in children with relapsed/refractory NB [17•, 51••]. Therefore, more efforts should be made on the modification of GD2 CAR T cells to enhance antitumor activity and promote CAR T cell longevity.

Targeting MYCN
The MYCN gene belongs to the MYC proto-oncogene family, encoding a basic helix-loop-helix transcription factor N-MYC. N-MYC, a master regulator of cell fate, interacts with other transcription factors to regulate gene expression (Fig. 2). In normal condition, MYCN is predominantly expressed in hematopoietic stem cells and cells within the developing nervous system. By contrast, gene amplification of MYCN has been observed in 20-30% of NB patients with a dismal prognosis [9]. Regardless of age and INSS stage, MYCN amplification is thought to be an independent poor prognostic factor [52••]. The biological functions of N-MYC have been implicated in various cellular processes including proliferation, differentiation, apoptosis, metabolism, and maintenance of stem cell (Fig. 2) [53,54,55]. N-MYC overexpression results in uncontrolled cell proliferation, apoptosis inhibition, and differentiation arrest in NB. MYCN gene amplification can modulate tumor microenvironment and influence immune response by regulating cytokines [53,56]. Additionally, MYCN amplification or N-MYC overexpression is one of the earliest molecular abnormalities identified in NB, which is closely associated with stage, aggressiveness, and prognosis of tumor [53]. Considering its important roles in NB pathogenesis, N-MYC has become an attractive target for curing this disorder.
Currently, there is no specific small molecular drug available for directly targeting N-MYC, since it acts as a general transcription factor in normal and tumor cells and its protein structure lacks enzymatic pocket for drug binding. Thus, approaches to target N-MYC mostly consist of two strategies: (1) downregulation of N-MYC expression by targeting its transcription, translation, and degradation; (2) impairment of N-MYC functions by targeting N-MYC/MYC-associated factor x (Max) interactions or other molecules. Here, a few simplified examples are presented for those strategies.
Exploring the inhibitors of bromodomain and extra-terminal domain (BET) and cyclin-dependent kinases (CDKs) to treat NB is an example of targeting MYCN transcription. The BET family recruits transcriptional regulatory complexes to acetylate chromatin, thereby regulating the transcription of MYCN [57]. BET inhibitors JQ1 and GSK1324726A (I-BET726) were firstly found to have the therapeutic effects on NB in xenograft mouse models [57,58]. MZ1, a novel BET inhibitor, utilizes proteolytic-targeting chimera (PRO-TAC) technology to facilitate BET proteasomal degradation, resulting in a rapid and effective inhibition of MYCNamplified NB in vitro and in vivo [59]. Another two BET inhibitors, BMS-986158 and BMS-986378, have entered a phase I clinical trial (NCT03936465) for treating NB with no data reported yet [21]. Moreover, CDKs are oncogenic drivers through regulating MYCN transcription [60 •, 61]. The CDK2/9 inhibitor fadraciclib has been shown to repress growth and promote apoptosis in MYCN-amplified NB through MYCN inhibition in a preclinical study [61]. Importantly, the combination of CDK7 inhibitor YKL-5-124 and BET inhibitor JQ1 showed a synergistic effect in xenograft mice with NB [62]. Current studies about BET inhibitors and CDKs inhibitors for curing NB maintain in the preclinical stage, so further drug testing on animal models and clinical trials is required.
Targeting protein stability is another method to decrease N-MYC expression. In normal cells, N-MYC is precisely controlled by the ubiquitin-proteasome system (UPS) with a short half-life (less than 30 min) [63]. AURKA is considered as one of the negative regulators to control N-MYC degradation in NB, which means N-MYC degradation can be increased by AURKA inhibition [63]. A recent study demonstrated that one AURKA inhibitor MLN8237 (alisertib) induces cell senescence, blocks cell cycle arrest at G2/M and accelerates N-MYC degradation in NB cell line IMR32 [64]. Further, the tolerability and the anti-tumor activity of alisertib combined with irinotecan and temozolomide were evaluated by phases 1 and 2 clinical trials (NCT01601535), highlighting the potency of this regimen for managing NB [19]. Another AURKA inhibitor LY3295668 (erbumine) is being tested for treating relapsed/refractory NB, with the primary aim of assessing safety and tolerability and anti-tumor activity (phase I clinical trial NCT04106219) [20••]. Notably, PHA-680626, a novel small molecule inhibitor targeting interactions between AURKA and N-MYC, is designed to selectively decrease N-MYC protein level, and not to impair other functions of AURKA (e.g., cell cycle and mitosis). This inhibitor is currently being trialed at preclinical level [65••].
Targeting N-MYC/MYC associated factor X (Max) interplays is one of the first strategies to suppress NB development and progression. N-MYC forms heterodimers with MAX, which is essential for N-MYC to properly function as a transcription factor [66••]. For example, 10,058-F4 and 10,074-G5, small molecular drugs, are known to inhibit the N-MYC/MAX interaction, consequently inducing differentiation and apoptosis in MYCN-amplified NB cells [67].
Recently, the small molecule MYCMI-6 with a high affinity for MYC is reported to selectively inhibit the tumor growth in MYCN-amplified cell and animal models, and this compound has no cytotoxic effect on normal cells [68]. Remarkably, MYCI975 has multiple roles against NB rather than disrupting N-MYC heterodimerization with MAX, including enhancement of N-MYC phosphorylation, promotion of N-MYC degradation, and sensitization of anti-programmed cell death protein 1 (PD1) immunotherapy [69]. It is worth noting that omomyc, a dominant-negative form of MYC, binds to MAX to inhibit MYC-mediated gene transcription, because MYC/MAX heterodimers are replaced by omomyc/ MAX heterodimers [70]. However, the anti-tumor activity of omomyc in NB has not been assessed [71]. Impairment of N-MYC function by targeting other molecules is also employed to treat NB. By way of explanation, the biological function of N-MYC depends on polyamine homeostasis [72]. Hence, targeting ornithine decarboxylase 1 (ODC1) is another method to disrupt N-MYC functions, as this enzyme is a rate-limiting enzyme in polyamine synthesis. Difluoromethylornithine (DFMO), an inhibitor of ODC1, was initially reported to impede cell proliferation, block tumor initiation, and enhance chemotherapeutic efficacy in NB cells [73]. The data from a phase II clinical study (NCT02395666) showed DFMO maintenance after completion of standard therapy has a positive association with improved outcomes in high-risk NB, with 5-year EFS and 5-year OS being 85.2% and 95.1%, respectively [22•]. Remarkably, the integration of DFMO and AMXT 1501 (a polyamine transport inhibitor) significantly reduces tumor progression and prolongs survival in vitro and in vivo, suggesting incorporation of polyamine synthesis and uptake inhibitors with standard chemotherapy could be a new therapeutic approach [74].
Taken together, numerous drugs that impede MYCN directly and indirectly are being intensively investigated at preclinical and clinical levels, from monotherapy to combination therapy. These indicate that MYCN-targeted strategies are very promising to be incorporated into the clinical management of NB in future.

Targeting ALK
ALK, a tyrosine kinase receptor, has an important role in the initiation and progression of malignancies including NB. ALK is the second most common oncogenic driver with its mutations being correlated with familial NB, which accounts for 1-2% of all NB cases [75]. In sporadic NB, approximately 9-14% of patients carry ALK mutations or amplifications [76]. ALK mutations within the tyrosine kinase domain often occur in NB, among which the most common variants are Arg1275Gln, Phe1174Leu, and Phe1245Cys. The mutated ALK leads to ligand-independent autophosphorylation and activation of ALK, which consequently activates RAS/MAPK and PI3K/Akt/mTOR pathways (Fig. 3

) [77••].
It is noted that hyperactivated ALK in NB contributes to proliferation, migration, and recurrence [78]. More importantly, ALK and MYCN have a close relationship (Figs. 2 and 3), since they are adjacently located on chromosome 2, at 2p23 and 2p24, respectively [79]. Further, ALK and MYCN are generally co-amplified [80], and ALK mutations are more frequently observed in MYCN-amplification positive NB [81]. Both ALK and MYCN have a correlation with a poor prognosis and interact with each other in a synergic manner. MYCN can induce ALK activation, which in turn enhances MYCN transcription and stability; however, the development of NB may require both activated proteins [82]. MYCN expression is also positively linked with ALK expression in MYCN non-amplified NB patients [83•]. Nevertheless, the elevated ALK expression may be found in cases without ALK mutations as well, which indicates a grim prognosis [84]. Briefly, ALK is a crucial therapeutic target, particularly for high-risk NB with MYCN amplification.
To date, there are two types of drugs for targeting ALK that are small molecular inhibitors and monoclonal antibodies. ALK small molecular inhibitors have been well developed to treat a range of cancers over the past years, while monoclonal antibodies against ALK are still largely unknown. The first-generation ALK inhibitor crizotinib competitively binds to the receptor tyrosine kinases. Treatment of ALK-positive cancers with crizotinib achieved a prolonged survival, in comparison with chemotherapy [85]. This agent has been approved to treat ALK or ROS1-positive non-small cell lung cancer (NSCLC), relapsed or refractory systemic anaplastic large cell lymphoma (ALCL), and ALKpositive inflammatory myofibroblastic by the FDA in 2016, 2021, and 2022, respectively [85,86,87]. In 2021, the thirdgeneration ALK inhibitor lorlatinib has been approved to be the first-line therapy for metastatic ALK-positive NSCLC [85]. Regarding the efficiency of ALK inhibitors in NB, the response rate is low in clinical trials, although the inhibitors exert a good anti-tumor activity in tumor cell lines and xenograft mice [24 •, 25, 88]. For example, a phase II clinical study (NCT00939770) found only three of twenty ALK-aberrant patients with relapsed/refractory NB effectively respond to crizotinib [24•]. Ceritinib is a secondgeneration ALK inhibitor and 20% of NB cases (n = 30) respond to this drug in a clinical study (NCT01742286) [25]. Lorlatinib (PF-06463922), a third-generation ALK inhibitor, has been demonstrated to be more efficient for treating ALK-driven NB than crizotinib [26,89]. In a preclinical study, lorlatinib (ALK/ROS1 inhibitor) shows ALK inhibition and induces tumor regression in crizotinib-resistant and crizotinib-sensitive xenograft mice as well as in patient-derived xenografts carrying the crizotinib-resistant Phe1174Leu and Phe1245Cys mutations [89]. Another case study showed that one patient with relapsed refractory, ALK Phe1174Leu-mutated NB had a complete response to lorlatinib, but the disease recurred after 13 months of treatment [26]. Typically, ALK inhibitor efficacy partially depends on ALK-mutant sites, as evidenced by the fact that ALK Arg-1275Gln mutation is susceptible to crizotinib in NB [24•]. Data about monoclonal antibodies against ALK are rather limited, with an ALK antibody identified to suppress the tumor cell growth and trigger ADCC in human NB-derived cell lines [90].
In summary, there is currently no ALK inhibitors or monoclonal antibodies approved for the treatment of NB by the FDA or the EMA. New findings pointed out that combining chemotherapy with ALK inhibitors could be a potential method to improve the efficacy and the sensitivity to ALKtargeted drugs. More studies should focus on how to improve ALK inhibitors efficacy and avoid resistance and adverse side effects. More anti-ALK monoclonal antibodies should also be further designed, developed, and tested in cell and animal models and clinical trials.

Targeting Intracellular Signaling Transduction Pathways
Several intracellular signaling pathways are involved in the development and the progression of NB, including p53/ MDM2, PI3K/Akt/mTOR, and RAS/MAPK (Fig. 3). Nowadays, inhibition of multiple signaling pathways is being extensively explored as a novel and promising therapeutic strategy.
Targeting p53/MDM2 signaling pathway has a great potentiality for curing NB, as this pathway is central for protecting cells against genome instability and malignant transformation. The p53 tumor suppressor, known as the guardian of the genome, is a critical multi-function player that regulates DNA damage and cell cycle, represses cell division, and promotes apoptosis [91••]. TP53 mutations occur in fewer than 2% of the relapsed or refractory NB cases, which is usually associated with multi-drug resistance [91••, 92•]. Despite TP53 mutations rarely seen in NB, p53 accumulation with increased stability and MDM2 overexpression induced p53 inhibition have been commonly identified [91••]. MDM2 is an E3 ubiquitin ligase involved Fig. 3 The functions of ALK and PI3K/AKT/mTOR and RAS/ MAPK pathways in NB. ALK activates PI3K/AKT/mTOR and RAS/ RAF/MEK/ERK pathways, which regulates growth, proliferation, survival, apoptosis, invasion, metastasis, and angiogenesis in NB. PI3K signaling can also be triggered by TrkB and RTKs, while EGFR can stimulate RAS/MAPK pathway. Of note, a range of modulators can control different sites of these pathways, including eEF1A2, PIM and SHP2. N-MYC or MYCN that is regulated by PI3K/AKT/mTOR pathway, can interact with MDM2, thus causing aberrant VEGFmediated angiogenesis and p53-induced apoptosis. (Fig. 3 was generated by Figdraw with permission ID being TYUIWc48a4). ALK, anaplastic lymphoma kinase; PI3K, phosphatidylinositol-3-kinase; AKT, protein kinase B; mTOR, mammalian target of rapamycin; RAS, rat sarcoma; MAPK, mitogen-activated protein kinase; RAF, serine/ threonine kinase; MEK, mitogen-activated protein; ERK, extracellular signal-regulated kinase; NB, neuroblastoma; TrkB, tyrosine kinase receptor tropomyosin-related kinase B; RTK, receptor tyrosine kinase; EGFR, epidermal growth factor receptor; eEF1A2, eukaryotic translation elongation factor-1, alpha-2; PIM, proviral insertion site in Moloney murine leukemia virus; SHP2, Src homology-2 domaincontaining protein tyrosine phosphatase-2; MDM2, murine double minute 2; VEGF, vascular endothelial growth factor; BDNF, brainderived neurotrophic factor; Wip1, wild-type p53-induced phosphatase 1, Ub, ubiquitin; WIP1, wild-type p53-induced phosphatase 1 in p53 degradation, and high expression of MDM2 is linked to metastasis, high staging, chemotherapy resistance, and poor prognosis of NB [91••]. Increasing data proved that the involvement of MDM2 in tumor initiation and progression is partially because of MDM2-mediated-p53 reduction [91••]. MDM2 also performs p53-indenpent activities, such as elevation of gene stability and gene transcription [55]. Given the functional wild-type p53 and the crosstalk of MDM2-p53 with MYCN in NB, it is prudent to develop inhibitors for directly targeting MDM2-p53 interaction, thus enhancing p53 function [93]. For instance, nutlin-3a is the first-generation inhibitor to block the interaction between MDM2 and p53 and reactivate the p53 activity in tumor cells and xenograft mice [94]. Derived from the structure modification of nutlin-3, the new inhibitor RG7112 is developed with an increased affinity and a decreased toxicity, but it is still not being examined in clinical trials [95]. Another second-generation MDM2 antagonist idasanutlin (RG7388) shows elevated potency, selectivity and bioavailability in preclinical studies and is being investigated in a phase I/II clinical trial (NCT04029688) for the treatment of relapsed or refractory NB [28]. Beyond small molecular antagonists directly disrupting MDM2-p53 interactions, targeting protein modulators to inhibit MDM2 or restore p53 is gaining more and more attention. For example, the serine/threonine phosphatase WIP1 (wild-type p53-induced phosphatase 1), encoded by protein phosphatase, Mg2 + /Mn2 + dependent 1D (PPM1D) gene at 17q23, can reduce p53 activation as well as control cell cycle, apoptosis, and DNA repair in NB [96]. WIP1 overexpression links to poor prognosis in NB patients, while inhibition WIP1 decreases survival and growth of tumor cells [96]. Moreover, the WIP1 inhibitor SL-176 has been reported to have similar or stronger effects on tumor growth than nutlin-3 in NB xenograft models [97]. Although disruption of p53/MDM2 is an appealing strategy, treatment-related toxicity and drug resistance have been discovered. Thus, better inhibitors should be designed and developed.
Targeting PI3K/Akt/mTOR pathway has been identified as a promising target because this pathway has a significant role in many cellular processes [29 •, 98]. PI3K/Akt/mTOR pathway, the key pro-survival signaling, is activated in most cases of NB, which results in deregulated growth, angiogenesis, invasion, and metastasis [51••, 99]. The aberrant activation of PI3K/AKT/mTOR pathway likely predicts poor prognosis and drug resistance in NB [100,101]. Further, PI3K/Akt/mTOR pathway has been found to affect MYCN amplification and stabilization to enhance the NB phenotype. For example, high Akt phosphorylation correlates with MYCN amplification, and in turn, inhibition of Akt causes a decreased N-MYC protein level [100]. Hence, inhibitors directly or indirectly targeting PI3K, Akt or mTOR could be employed to manage NB. For instance, a phase I trial (NCT00776867) found that Akt inhibitor perifosine retards disease progression with no apparent toxicity in 33% of patients with refractory/relapsed NB (n = 27) [29•]. PP242, a mTOR inhibitor, reverses hypoxia-induced gene expression profiles that contribute to an unfavorable prognosis in NB [101]. Another mTOR inhibitor temsirolimus has been approved for the treatment of renal cell carcinoma in 2007 and is currently undergoing a phase II trial (NCT01767194) for patients with relapsed or refractory NB [30, 51••]. AZD8055, a dual mTORC1/mTORC2 inhibitor, suppresses growth and induces apoptosis in NB cells and mice [102]. VS-5584, a dual PI3K/mTOR inhibitor, exhibits anti-tumor effects on NB in vitro and in vivo [103]. Targeting modulators of this pathway is also a potential therapeutic approach. For instance, the eukaryotic translation elongation factor-1, alpha-2 (eEF1A2) is considered as an activator of PI3K/ AKT/mTOR pathway, and its knockdown leads to an inhibition of Akt/mTOR phosphorylation in NB cells [104]. This finding may highlight a role of eEF1A2 as a new molecular target for NB therapy [104]. Interestingly, the serine/threonine proviral insertion site in murine leukemia virus (PIM) kinases has a correlation with MYC overexpression, PI3K inhibitor resistance, and an unfavorable outcome, and its inhibition significantly prolongs the survival of NB mice with wild-type neurofibromatosis type 1 (NF1) [105,106]. It is of interest that the triple PIM/PI3K/mTOR inhibitor IBL-302 is more effective than the single target inhibition, which causes enhanced apoptosis and differentiation and decreased N-MYC in NB cells and patient-derived xenografts [106]. It appears that simultaneously combining standard therapy with multiple targeted therapy could be more powerful. How to design and develop inhibitors with multiple targets is extremely important.
RAS/MAPK pathway is not only involved in the early development, but also emphasized in the recurrence, with nearly 80% of relapsed NB patients harboring mutations in this pathway (e.g., NF1, BRAF, protein tyrosine phosphatase non-receptor type 11 (PTPN11), fibroblast growth factor receptor 1 (FGFR1), KRAS, NRAS, HRAS, and ALK) [107]. Current treatment options are less effective in relapsed NB, so targeting RAS/MAPK may offer an approach to treat disease recurrence. In NB, RAS/MAPK pathway is often activated by receptor tyrosine kinase, in particular ALK. Mutations in ALK or RAS gene family members usually cause the constant activated RAS/MAPK signaling in NB. By way of illustration, RAS mutations lock RAS into an active form and enhance GTP-binding affinity and effector-binding affinity, thus leading to persistent activation of RAS/MAPK [108 •]. Sotorasib (AMG510), an inhibitor of KRAS-G12C mutant protein, has been successfully approved by the FDA for treating NSCLC in 2021 [108•]. Nevertheless, there is no report about the effects of this inhibitor on NB. Among RAS inhibitors, tipifarnib, a general inhibitor of RAS proteins, is the only one being tested in early-stage clinical trials for NB. Tipifarnib prevents RAS from binding to the cellular membrane, thereby inhibiting signaling transduction [109, 110••]. Strikingly, tipifarnib, an FDA-approved inhibitor, sensitizes NB tumors to dinutuximab, providing a promising treatment option for high-risk NB [109]. MEK 1/2 inhibitors, such as binimetinib and trametinib, also show potent cytotoxicity against RAS mutated NB in vitro and in vivo [107, 110••, 111]. Evidence from several studies proved that NB cells or mice xenografts with RAS/BRAF mutations are sensitive to MEK inhibition [107, 110••, 111, 112, 113]. For instance, binimetinib and trametinib inhibit the growth in several NB cell lines and xenograft models harboring RAS mutations [107, 110••, 111, 112 113]. A phase I/IIa clinical trial (NCT02124772) is ongoing for the investigation of safety and activity of trametinib monotherapy in NB, and the combined therapy of dabrafenib with trametinib in cancers harboring BRAF V600E mutations [31•, 51••]. To overcome the acquired resistance to RAS/MAPK inhibitors, concurrent inhibition of this pathway at distinct sites has been suggested. PTPN11, one of the most important mutations in relapsed NB, encodes tyrosine phosphatase SHP2. SHP2, an activator of RAS, promoting its dephosphorylation to enhance RAS binding to RAF and consequently activates RAS/MAPK signaling [114]. RAS mutations (e.g., NRAS Q61K) in NB confer resistance to SHP2 inhibitors; however, the combination of SHP2 inhibitors and RAS downstream component inhibitors (RAF, MEK, or ERK) shows synergistic effects in RAS mutated NB cells [114]. In brief, targeting RAS/MAPK could be an effective strategy for relapsed NB; nonetheless, it should be aware that single inhibition of this way may be insufficient to manage NB. Combination therapies could be vital to achieve effective treatments and avoid resistance development.

Combination Therapies
It is well recognized that rational combination therapies will be essential to further progress therapies for high-risk NB, such as increasing the response to standard therapies and developing new therapies for refractory/relapsed disease. In general, there are two types of combined therapies: one is conventional therapies synergized with molecular targeted therapies; the other is combination targeted therapy using inhibitors that target many different molecules or pathways. For maximum therapeutic benefit, molecularly targeted therapies are often combined with chemotherapeutic agents. For example, dinutuximab in combination with irinotecan and temozolomide has shown a significant anti-tumor activity [115••]. Another study implied that integrating conventional chemotherapy with PIM/PI3K/mTOR inhibition could improve outcomes and avoid adverse effects in children with high-risk NB [106]. Regorafenib (BAY73-4506), an inhibitor of multiple kinases, has been approved by the FDA for the treatment of metastatic colorectal cancer, advanced gastrointestinal stromal tumors (GIST), and progressive hepatocellular carcinoma [116]. In NB, regorafenib simultaneously inhibits RAS/MAPK, PI3K/AKT/mTOR, and Fos/Jun pathways, resulting in growth suppression and apoptosis induction [32,116]. This inhibitor combined with standard therapy is being used to treat refractory/relapsed NB in a phase I clinical trial (NCT02085148). More molecular-guided therapy in combination with standard therapy is being currently explored for high-risk NB managements in preclinical and clinical trials.
Combining multiple molecular targeted therapies concentrated on genetic alterations and deregulated pathways represents a novel approach for NB, which could improve efficacy, overcome resistance, avoid relapse, and reduce toxicity. For example, several inhibitors have been discovered to improve the efficacy and reduce the adverse side effects of GD2 monoclonal antibodies, such as PD-1 inhibitors, RAS inhibitors as well as ODC1 inhibitors [109,117,118,119]. A combined treatment with a PD-1 inhibitor and dinutuximab beta showed a synergistic anti-NB immune response in the mouse model, and two patients with refractory NB have a good response to the combination of nivolumab (PD-1/PD-L1 monoclonal antibody) and dinutuximab beta [120]. A phase I trial (NCT02914405) is going to investigate safety and efficacy of nivolumab in combination with dinutuximab beta for relapsed/refractory NB with no results available. In addition to PD-1 inhibitors, the effectiveness of dinutuximab can be elevated by tipifarnib (RAS inhibitor) [109]. In contrast, the combination of DFMO (ODC1 inhibitor) and dinutuximab can relieve anti-GD2-induced allodynia in rat models, and the potency of this regimen is being evaluated in a phase II clinical trial (NCT03794349) [119]. Additionally, the benefit of DFMO in synergy with ceritinib (ALK inhibitor) or bortezomib (proteasome inhibitor) is being studied in phase II clinical trials (NCT02559778, NCT02139397). Trametinib (MEK inhibitor) combined with dabrafenib (BRAF inhibitor) has been reported to resolve vasoactive intestinal peptide-induced diarrhea in two NB cases with BRAF V600E mutation [31 •]. The combination of MEK inhibitors and ribociclib (CDK4/6 inhibitor) has also shown a synergistic effect [111], and a phase I clinical trials (NCT02780128) for investigation of the synergy of trametinib and ribociclib is ongoing. Strikingly, preclinical studies found that the combination of trametinib and a YAP (yes-associated protein 1) inhibitor (CA3) is capable to overcome trametinib resistance [113], whilst trametinib combined with GD2 immunotherapy can increase the killing efficacy of GD2-CAR T cell therapy [121]. More drug combinations among distinct inhibitors with different targets are being tested intensively in NB preclinical studies, such as ALK inhibitors with ATR inhibitors (BAY1895344) [122], MDM2 inhibitors with targeted radiation therapy ( 177 Lu-DOTATATE) [123], MDM2 inhibitors (CGM097) with BET inhibitors (OTX015) [124], CDK7 inhibitors (YKL-5-124) with BET inhibitors (JQ1) [62], mTOR inhibitors (temsirolimus) with BET inhibitors (JQ1 or OTX-015) [125], and FGFR1 inhibitors (AZD4547) with PI3K inhibitors (GDC0941) [126]. More combination options using inhibitors to target many various pathways would be further inspected at preclinical and clinical levels.

Conclusion
Despite significant discoveries in the field of NB molecularly targeted therapy, the translation of such findings to clinical disease managements remains low in past years. To date, only anti-GD2 monoclonal antibodies have been approved for the treatment of pediatrics with relapsed or refractory high-risk NB. There is a remarkable gap between preclinical and clinical trials. On one hand, the biological and the pathological roles of molecular targets are not fully elucidated in NB. On the other hand, new drug testing is currently conducted on the traditional NB cell and mouse models. New disease models should be evolved using advanced technology, such as 3D tissueengineered system and patient-derived xenograft model system. Owning to the heterogeneity of NB, different patients respond differently to therapies. How should an efficient personalized therapy be designed using molecularly targeted drugs and standard therapy? How should an effective therapy be progressed to prolong survival for high-risk NB or for refractory/relapsed NB? Overall, new drugs with more efficacy and less toxicity will be eagerly expected; novel combination options, employing inhibitors targeting multiple molecular targets, and conventional therapy, will be desperately needed.
Acknowledgements The figures were created by the BioRender and Figdraw.
Author contribution XY drafted the manuscript. JL and JY provided supervision and guidance, and revised the manuscript. All authors approved the submitted version.

Conflict of Interest
The authors declare competing interests.

Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
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 licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence 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. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.