Molecular targeted therapy for anticancer treatment

Since the initial clinical approval in the late 1990s and remarkable anticancer effects for certain types of cancer, molecular targeted therapy utilizing small molecule agents or therapeutic monoclonal antibodies acting as signal transduction inhibitors has served as a fundamental backbone in precision medicine for cancer treatment. These approaches are now used clinically as first-line therapy for various types of human cancers. Compared to conventional chemotherapy, targeted therapeutic agents have efficient anticancer effects with fewer side effects. However, the emergence of drug resistance is a major drawback of molecular targeted therapy, and several strategies have been attempted to improve therapeutic efficacy by overcoming such resistance. Herein, we summarize current knowledge regarding several targeted therapeutic agents, including classification, a brief biology of target kinases, mechanisms of action, examples of clinically used targeted therapy, and perspectives for future development.


INTRODUCTION
Cancer is one of the main causes of disease-related death worldwide. According to Global Cancer Observatory (GLOBOCAN) estimates of cancer incidence and mortality, there were approximately 19.3 million new cancer cases and almost 10.0 million cancer deaths in 2020 globally 1 . The cancer-related burden (such as incidence and mortality) is expected to be 28.4 million cases in 2040, which is a 47% increase compared with that in 2020, largely due to increases in risk factors, such as aging, socioeconomic development, overweight status, and smoking 1,2 . Therefore, it is necessary to develop efficacious treatment strategies for patients with cancer.
Several therapeutic modalities, such as surgery, radiation therapy, and systemic anticancer therapy, have been applied clinically for cancer treatment, either alone, in combination, or sequentially, depending on the stage, resectability, biology, comorbidities, and patient's overall functional performance 3,4 . Systemic anticancer therapy, involving a wide range of anticancer drugs for treatment, palliation, symptom alleviation, and quality of life improvement, includes cytotoxic chemotherapy, hormonal agents, targeted therapy, and antitumor immunotherapy 5,6 . Cytotoxic chemotherapy inhibits the survival of actively proliferating cells by disrupting the synthesis of DNA and RNA, blocking mitosis, and/or forming covalent bonds with DNA, RNA, and proteins 7 , and it has been extensively used in adjuvant or neoadjuvant therapy as well as in palliative therapy 7 . Due to the disadvantages of chemotherapy, including side effects and toxicity associated with nonselective action against actively proliferating normal cells 2,8 , there has been innovative development of 'targeted' cancer treatment with increased cancer cell specificity 8 . Targeted therapy may include the following: conventional molecular targeted agents, such as small molecule inhibitors or antibodies that specifically inhibit signal transduction pathways involved in growth, proliferation, and survival 9,10 ; hormonal agents such as estrogen receptor (ER) antagonists and aromatase inhibitors, which have been used for treatment of hormone receptor (HR)-dependent breast cancer and male and female reproductive cancers 11 ; immune checkpoint inhibitors [e.g., antibodies against programmed cell death protein 1 (PD-1), programmed death-ligand 1 (PD-L1), or cytotoxic T-lymphocyteassociated protein 4 (CTLA-4)], which activate host antitumor immunity in a direct or indirect manner 8,12 ; and even targeted cytotoxic therapy that interferes with a specific cellular target (e.g., methotrexate, a dihydrofolate reductase inhibitor) 10 . Despite the anticancer effectiveness of these targeted therapies, these drugs are only applicable for patients harboring targetable driver mutations or aberrations 13,14 . In addition, side effects or toxicity caused by unexpected cross-reactivity with normal cells and emergence of intrinsic or acquired drug resistance hamper their effectiveness 13,14 . Notwithstanding some limitations, targeted therapy has resulted in remarkable survival benefits in some types of cancer and has led to a revolution in the fundamental concept of cancer treatment, providing the fundamental backbone for evolution toward precision or personalized medicine in cancer 13,15 . Herein, we summarize current knowledge with respect to molecular targeted therapy, including the history, types, and mechanism of action, and provide examples of clinically available targeted therapy. In this paper, 'targeted therapy' is confined to conventional molecular targeted therapy (signal transduction inhibitors).
Brief history of molecular targeted therapy Paul Ehrlich first proposed the concept of targeted therapy in the 1890s as a "magic bullet" that would be completely specific for the target and thus safe without any additional toxicity 14,16 . This theory was initially applied to infectious diseases but not to anticancer therapy due to insufficient knowledge of the etiology and biology of cancer 14,16 ; however, this concept has since been expanded to cancer treatment 14,16 . Trastuzumab, an anti-HER2 monoclonal antibody, and imatinib, a small molecule tyrosine kinase inhibitor targeting the BCR-ABL fusion-mediated aberrantly activated ABL kinase, were developed and clinically approved in 1998 and 2001 for treatment of HER2-positive breast cancer and Philadelphia chromosome-positive chronic myelogenous leukemia, respectively 14,[17][18][19] . The success of imatinib in the clinic has served as the paradigm for extensive use of small molecule kinase inhibitors as anticancer therapy 8,17 , and a number of anticancer molecular targeted therapies have been approved for clinical use in cancer patients 8,17 . The timeline for the development of the main molecular targeted therapy is illustrated in Fig. 1.
Types, mechanisms of action and resistance, and adverse effects/toxicity of molecular targeted therapy To date, numerous molecular targeted therapeutic agents have been used clinically for cancer treatment. The classification of molecular targeted therapeutic agents and their targets, mechanism of action, side effects, and toxicity are described below.
Types of molecular targeted therapy. The two major types of molecular targeted therapy are monoclonal antibodies (mAbs) and small molecule kinase inhibitors (SMKIs) 8,14 . mAbs target extracellular ligands (e.g., bevacizumab targets vascular endothelial growth factor [VEGF]), membrane receptors (e.g., trastuzumab targets HER2 and cetuximab; panitumumab targets EGFR), and membrane-bound proteins (e.g., rituximab targets CD20), acting through ligand-binding blockade, ligand-receptor interaction neutralization, or target molecule internalization/degradation 14,20 . Except for inhibitors targeting nonkinase cellular proteins (e.g., mutated KRAS and proteasome) or epigenetic modulators (e.g., histone deacetylases), most SMKIs suppress protein kinases involved in the transformation, growth, proliferation, and survival of cancer cells. As deregulation of protein kinases (e.g., activation by gain-of-function genetic mutation, gene amplification, autonomous activation, and chromosomal rearrangement) has been associated with cancer development and progression [21][22][23][24] , protein kinases have been regarded as important targets for developing molecular targeted therapies. Protein kinases are classified into receptor tyrosine kinases, nonreceptor (cytoplasmic) tyrosine kinases, serine/threonine kinases, and lipid kinases based on their subcellular localization, substrate type, and hallmark roles in cancer 21 (Fig. 2). A detailed explanation of the signal transduction by receptor tyrosine kinase is described in previous studies 24, 25 .
SMKIs block the enzymatic activity of the aforementioned kinases via several modes of action 26 . Type I kinase inhibitors bind to the ATP-binding pocket of the active conformation of the enzyme [DFG (Asp-Phe-Gly)-in and αC-helix-in] 26 , whereas type I 1/2 or type II inhibitors bind the enzyme in an inactive conformation (type I 1/2 : DFG-Asp in; type II: DFG-Asp out) 21,26 . Type III and type IV inhibitors allosterically suppress kinase activity by binding either to a site next to the ATP-binding pocket or one remote from the ATP-binding pocket located in the kinase substrate-binding site 21,26,27 . Type V inhibitors act as bivalent inhibitors binding to two different portions of the kinase lobe 21,26 . Type VI inhibitors covalently bind an enzyme to inhibit kinase activity 26,28 . A recent paper describes the detailed mode of action of each type of kinase inhibitor 26 , and some examples are listed in Table 1.
Mechanisms of the anticancer effects of molecular targeted therapy. Molecular targeted therapies achieve anticancer effects through various mechanisms, such as inhibition of cell proliferation, metastasis, and angiogenesis, induction of apoptosis, and reversal of multidrug resistance 2 (Fig. 2a). Several molecular targeted therapeutic agents also facilitate host antitumor immunity by potentiating CD8 + T-cell recruitment and natural killer cell cytotoxicity, downregulating immunosuppressive myeloid cells, and inducing immunogenic cell death, either alone or in combination with chemotherapeutic agents 29 . Therapeutic mAbs create a bridge between tumor cells and immune cells via Fab region-mediated binding to a target protein of tumor cells and recognition of immune cells through the Fc region of antibodies 30 , resulting in opsonization and antibody-dependent cellular cytotoxicity (ADCC) toward tumor cells 30 (Fig. 2b). A recent study demonstrated that neutrophils mediate trogoptosis (Fig. 2c), the phenomenon of transferring surface molecules of interacting cells onto immune cells 31,32 , which causes lytic/necrotic death of antibody-opsonized cancer cells 33 . mAbs and SMIs also exert immune cell-induced cytotoxic effects on cancer cells by activating complement and complement-dependent cytotoxicity 30,34 , facilitating antigen processing by increasing expression of major histocompatibility complex molecules 30,35,36 and regulating cytokine/chemokine expression 30,37 .
Mechanisms underlying resistance to molecular targeted therapy. The emergence of drug resistance is a major hurdle of efficacious anticancer treatment. Primary (intrinsic) resistance is defined as a refractory status to initial therapy due to intrinsic cellular, genetic, and/or epigenetic alterations. Hyperactivation of compensatory signaling pathways [e.g., truncated HER2 expression (p95HER2) for resistance to anti-HER2 mAbs 38 ; KRAS mutation or MET amplification for resistance to anti-EGFR therapy 38,39 ], mutations in kinase domains (e.g., EGFR exon 20 insertion for resistance to anti-EGFR therapy 38 ), isoform switching (e.g., BRAF/CRAF switching for resistance to anti-BRAF therapy 40 ), and metabolic reprogramming 40 during disease development are involved in primary resistance to molecular targeted therapy.
Human cancers often exhibit substantial intratumor heterogeneity, which is a main driver for emerging acquired therapy resistance as a result of expansion of rare preexisting refractory populations during treatment in initial responders 39,41,42 47 , reactivation of developmental processes, such as the epithelial-mesenchymal transition (EMT), acquisition of cancer stem cell (CSC)-associated phenotypes, and transdifferentiation to small-cell lung cancer 43,48 ] have also been shown to induce acquired therapy resistance. The mechanisms Fig. 1 Timeline for the approval of selected molecular targeted therapeutic agents. The first FDA-approved targeted therapeutic agent for each cellular target (denoted in blankets) is indicated in the timeline.
of resistance to each molecular targeted therapy are summarized in Tables 2-6.
Adverse effects and toxicity of molecular targeted therapy. Despite improved specificity for cancer cells, epidemiological studies have indicated that cancer patients who receive targeted therapy may experience various side effects and toxicity. The side effects of targeted therapy include asthenia, anorexia, dyspnea, diarrhea, nausea, vomiting, mucositis, skin rash, fever, hand-foot syndrome, fatigue, cardiotoxicity, hypertension, and bleeding 49,50 .   Specifically, acneiform rash, a skin rash with an acne-like appearance, is a common side effect of anti-EGFR therapy 50,51 , and hypertension is a common side effect of bevacizumab and anti-VEGF receptor (VEGFR) therapy 52 . These common side effects are related to therapy response 52 . Severe toxicities, such as colitis, digestive perforation, toxic cardiomyopathy, pneumonitis/interstitial lung disease, acute respiratory distress syndrome, posterior reversible encephalopathy syndrome, necrotizing fasciitis, acute renal failure, and hypersensitivity, have been observed in patients receiving molecular targeted therapy, such as antiangiogenic agents, anti-EGFR therapy, and anti-HER2 therapy 53 . The side effects and toxicity of each molecular targeted therapy are summarized in Tables 2-6.
SMKIs and mAbs in targeted cancer therapy By focusing on U.S. Food and Drug Administration (FDA)-approved kinase inhibitors, target kinases and examples of clinically used inhibitors are briefly introduced below.  [57][58][59] . These genetic changes cause abnormal EGFR activation in a ligand-independent fashion 60 . Exon 19 microdeletions and L858R point mutations are commonly found in patients with non-small cell lung cancer (NSCLC), particularly in nonsmoking east Asian females 59,61 , and EGFRvIII is frequently observed in glioblastoma [57][58][59] . Additional EGFR mutations, including E884K, D761Y, T854A, and exon 20 insertion, have been detected in NSCLC and found to confer EGFR TKI resistance 62 . Several EGFR TKIs have been developed over the past decades and are clinically used for treatment of patients with NSCLC harboring kinase-activating mutations ( Table 2). Gefitinib and erlotinib are first-generation EGFR-TKIs 8,63,64 that interact with the ATP-binding pocket of EGFR in either the active or inactive conformation 26 . Second-generation EGFR TKIs, such as afatinib and dacomitinib, are irreversible EGFR inhibitors that covalently bind to the ATP-binding pocket of EGFR 8,65,66 . Despite the great efficacy of firstand second-generation EGFR-TKIs in patients with kinase-activating mutations in EGFR 64,67 , the EGFR T790M mutation in exon 20 68 is associated with acquired resistance to these firstand second-generation EGFR-TKIs 67,69 (e.g., approximately half of NSCLC patients acquire resistance to first-generation EGFR-TKIs 69 ). EGFR T790M provides advantages for the growth and survival of cancer cells 69 and limits the therapeutic efficacy of EGFR TKIs through both steric hindrance and potentiated ATP binding 62,69 . Accordingly, EGFR TKIs targeting the T790M mutation have been developed and clinically utilized. Osimertinib, a thirdgeneration EGFR TKI, inhibits EGFR kinase activity by forming a covalent bond with the cysteine-797 residue in the ATP-binding pocket and shows an approximately 200 times greater inhibitory effect on mutant EGFR [L858R or exon 19 deletion mutations additionally harboring T790M (L858R/T790M or exon19del/ T790M)] than on wild-type EGFR 70,71 . Another third-generation EGFR TKI, lazertinib, is an orally available, CNS-penetrable, and irreversible EGFR TKI that inhibits EGFR T790M and kinaseactivating mutations 72 . Despite the approval of these agents for  86 .

Receptor tyrosine kinase inhibitors
In addition to SMKIs, mAbs targeting EGFR and HER2 have been used in the clinic ( Table 2). EGFR mAbs, including cetuximab and panitumumab, have been clinically used for treatment of patients with metastatic colorectal cancer [87][88][89] . HER2-targeting mAbs, such as trastuzumab and pertuzumab, are approved for clinical use in patients with HER2-positive breast cancer 90 . Recently, the HER2bispecific antibody zanidatamab was approved for patients with HER2-expressing biliary tract cancers 91 , and several clinical trials for recently developed HER2-targeting monoclonal antibodies are ongoing 86 .
ALK inhibitors: ALK is an receptor tyrosine kinase (RTK) with structural homology to leukocyte tyrosine kinase (LTK), which belongs to the insulin receptor superfamily 92 . In normal tissues, ALK expression is predominant in the nervous system and is known to play an important role in physiological regulation of nervous system development and function 92,93 . Chromosomal rearrangement of the ALK gene and consequent generation of a fusion protein with a number of partner proteins, including echinoderm microtubule-associated protein-like 4 (EML4), nucleophosmin (NPM), tropomyosin 3 (TPM3), and tropomyosin 4 (TPM4), ALK gene amplification, or ALK mutations lead to overexpression of a constitutively activated ALK protein 92 . ALK alterations have been found in several types of cancer, such as anaplastic lymphoma, neuroblastoma, and NSCLC 92 . Approximately 3-7% of patients with NSCLC, especially for those with the adenocarcinoma subtype, have been reported to harbor ALK rearrangements; ALK mutations are mutually exclusive with KRAS and EGFR mutations 94,95 .
Several ALK inhibitors are currently available in the clinical setting (Table 2), and these drugs are approved for the treatment of NSCLC patients. Crizotinib, a first-generation ALK inhibitor, is an orally available ATP-competitive inhibitor that was clinically approved in 2011 95,96 . Crizotinib was initially developed as a MET inhibitor; however, based on the inhibitory effect of crizotinib on ALK at pharmacologically relevant concentrations and the structural homology of the ATP-binding site between ALK and ROS1, the clinical efficacy of crizotinib has been evaluated in patients carrying alterations in these genes 95,97 . Consequently, crizotinib has been used as a firstor second-line therapy in patients with NSCLC harboring ALK, ROS1, or MET alterations 96,98,99 . However, due to the rapid emergence of resistance to crizotinib and its weak ability to penetrate the central nervous system (CNS) 95,96,100 , additional ALK inhibitors have been developed. The second-generation ATP-competitive ALK/ROS1 inhibitor ceritinib and the ATP-competitive ALK inhibitor alectinib have been approved for treatment of patients with crizotinib resistance 96 . In contrast to crizotinib and ceritinib, alectinib can penetrate the CNS, curing NSCLC patients with brain metastasis and preventing progression of CNS metastasis 8,96 . Additional blood-brain barrier (BBB)-permeable ATP-competitive ALK TKIs have been developed, including brigatinib, which is effective against FMS-like tyrosine kinase 3 (FLT3), insulin-like growth factor receptor (IGF-1R), EGFR, and several ALK mutations associated with resistance to crizotinib, ceritinib, and alectinib 101,102 , and lorlatinib, with inhibitory effects against all recognized ALK mutations except the L1198F mutation 8,92 .
MET inhibitors: MET is an RTK activated by hepatocyte growth factor (HGF) and mediates several physiological processes, such as embryogenesis and tissue repair; aberrant activation of MET by genetic alterations plays an important role in the proliferation, invasion, and metastasis of tumor cells 103 . Alterations in the MET gene, such as amplification, mutation, and alternative splicing (MET exon 14 skipping), have been detected in NSCLC and other solid tumors 8,99 . MET overexpression is associated with poor prognosis and resistance to chemotherapeutic agents, including EGFR targeted therapy 8,104 . In addition, MET gene exon 14 skipping leads to constitutive activation of the MET signaling pathway and confers sensitivity to MET inhibitors 105 . MET inhibitors, such as orally available ATP-competitive smallmolecule TKIs and monoclonal antibodies, have been developed and evaluated in preclinical and clinical trials 99 . Among them, capmatinib and tepotinib are approved for clinical use in treatment of patients with metastatic NSCLC harboring MET exon 14 skipping 99 (Table 2).
TRK and FLT3 inhibitors: Neurotrophic tyrosine receptor kinases (NTRKs) are oncogenes that encode tropomyosin receptor kinase (TRK) proteins, including TRKA, TRKB, and TRKC 106 . TRKs are activated by binding of intrinsic neurotrophin ligands, such as nerve growth factor (NGF) for TRKA, brain-derived neurotrophic factor (BDNF) and neurotrophin 4 (NT-4) for TRKB, and neurotrophin 3 (NT-3) for TRKC 106,107 . NTRK gene fusion caused by chromosomal rearrangements of NTRK genes with various fusion partners drives ligand-independent, constitutive activation of TRKs, which has been found in a wide range of cancer types, including mammary analog secretory carcinoma, secretory breast carcinoma, and infantile fibrosarcoma 106,107 . FLT3 (CD135), a class III RTK, is exclusively expressed in hematopoietic stem and progenitor cell populations 108 . Constitutive activation of FLT3 kinase through internal tandem duplications (FLT3-ITD) or missense mutations in the FLT3 tyrosine kinase domain 109 has been observed in approximately 30% of patients with acute myeloid leukemia (AML) and a normal karyotype 109,110 . Several TKIs targeting TRKs (e.g., larotrectinib and entrectinib) or FLT3-ITD (e.g., midostaurin, sorafenib, and gilteritinib) have been developed and approved for clinical use. Examples are listed in Table 2.
Inhibitors targeting PDGFR, VEGFR, or FGFR family receptors and Ret: Tumor angiogenesis is a hallmark of cancer. Several growth factors and their receptors, such as platelet-derived growth factor (PDGF)/PDGFR, vascular endothelial growth factor (VEGF)/VEGFR, fibroblast growth factor (FGF)/FGFR, stem cell factor (SCF)/c-Kit, glial cell line-derived neurotrophic factor (GDNF)-family ligands/ rearranged during transfection (RET), and angiopoietin/Tie 22,111 , regulate the growth, differentiation and migration of cancer cells and angiogenic activities of vascular endothelial cells 22,111 . PDGFs are members of the 'cysteine knot' growth factor superfamily, the members of which contain at least three disulfide bridges and forms homo-or heterodimers 112 . Five types of PDGF dimers (PDGF-AA, PDGF-AB, PDGF-BB, PDGF-CC, and PDGF-DD) have been identified, and these PDGFs transduce signals by binding to two isotypes of PDGFRs (PDGFR-α and PDGFR-β) 113 . PDGF-AA and PDGF-CC, ligands that bind to these PDGFRs with different affinities, have a high affinity for PDGFR-α, whereas PDGF-BB and PDGF-DD transduce signaling through PDGFR-β 113 . PDGFR-α plays both general and specific roles in the development of mesenchymal and fibroblastic cell compartments; PDGFR-β plays an important role in the formation of vascular mural cells, including vascular smooth muscle cells and pericytes 113 . Alterations in PDGFR-α and PDGFR-β are associated with vascular diseases and mesenchymal cell/fibroblast-driven pathological conditions, respectively 113 . Alterations in PDGFR-α, such as point mutations and amplification, exist in approximately 5% of patients with gastrointestinal stromal tumors (GISTs) and 5-10% of patients with glioblastoma multiforme 114 .
The VEGF family is composed of five glycoproteins, including VEGFA (VEGF), VEGFB, VEGFC, VEGFD (c-Fos-induced growth factor, FIGF), and placental growth factor (PIGF or PGF) 115 . VEGF is expressed as multiple alternative splicing isoforms, with pro-or antiangiogenic effects; among them, VEGF 165 is the predominant proangiogenic isoform that is overexpressed in various solid tumors 115,116 . VEGF activates signal transduction by binding to VEGFR family receptors, VEGFR1 (FLT1), VEGFR2 (KDR), and VEGFR3 (FLT4) 25,115 . VEGFR2 is primarily expressed in vascular endothelial cells, and VEGF/VEGFR2 signaling plays a crucial role in angiogenesis by controlling vascular permeability, proliferation, migration, and survival of vascular endothelial cells 115,117 . VEGF also stimulates vasculogenesis in tumors by recruiting bone marrow-derived hematopoietic progenitor cells and endothelial progenitor cells 115 . VEGFC and VEGFD bind VEGFR3 and regulate lymphangiogenesis, contributing to metastatic spread through the lymphatic system 118 . In addition to these angiogenic effects on vascular endothelial cells, VEGF exerts several tumorpromoting effects, such as increased cancer cell proliferation, migration, invasion, stemness [119][120][121] , immune suppression 115 , and premetastatic niche formation 122 .
The FGF family growth factors, comprising 18 members that are categorized into six subfamilies, activate signal transduction by binding to FGFRs 123 . Five FGFRs (FGFR1-FGFR5) are known 123,124 . FGFR1-FGFR4 possess tyrosine kinase activity; in contrast, FGFR5 lacks the intracellular tyrosine kinase domain but acts as a coreceptor of FGFR1 and modulates ligand-mediated signaling [123][124][125] . Heparan sulfate glycosaminoglycan (HSGAG) binds to both FGF and FGFR, protecting FGFs from degradation, stabilizing the interaction between ligand and receptor, and facilitating dimerization of FGF-bound FGFR 123 . In cancer cells, aberrant activation of FGF/FGFR signaling caused by FGFR amplification, activating FGFR mutations, FGFR single-nucleotide polymorphisms, FGFR fusion protein formation with various binding partners, and deregulation of phospholipase Cγ1 (PLCγ1, FRS1) and FGFR substrate 2 (FRS2) all promote cell survival, cell proliferation, angiogenesis, acquisition of an epithelial-mesenchymal transition (EMT) phenotype, invasion, and metastasis in cancer cells 124,126 .
Nonreceptor tyrosine kinase inhibitors BCR-ABL and SFK inhibitors: Abelson (ABL) family kinases (ABL1 and ABL2) are nonreceptor tyrosine kinases that commonly contain a specific domain cassette consisting of the Src homology 3 (SH3) domain (a protein module that binds to proline-rich sequences), the SH2 domain (a protein module that binds to tyrosine phosphorylated sites), the tyrosine kinase domain (SH1 domain), the PXXP motif mediating interaction with SH3 domaincontaining proteins, and the C-terminal F-actin binding domain 135,136 . ABL1, but not ABL2, additionally includes a DNAbinding domain, nuclear localization signal motifs, and nuclear export signal motif and mediates DNA damage repair 135,136 . ABL2 is mainly found at actin-rich sites, including focal adhesion and invadopodia in the cytoplasm, through its F-actin and microtubule-binding domains and mediates cytoskeletal remodeling 135,136 . Activation of ABL kinases is tightly regulated through autoinhibitory intramolecular interactions, intermolecular interactions with other proteins to disrupt or maintain autoinhibitory conformation, and posttranslational modifications such as transor Src-mediated tyrosine phosphorylation (e.g., activation of ABL1 by phosphorylation at Y245 and Y412), serine/threonine phosphorylation, acetylation, myristoylation, and polyubiquitination 135,136 . Oncogenic alterations in ABLs, including fusion protein formation caused by chromosome translocations in leukemia [e.g., BCR-ABL1 in Philadelphia chromosome-positive (Ph + ) chronic myeloid leukemia (CML)] and amplification and somatic mutations in solid tumors, constitutively activate ABLmediated signaling pathways and promote survival, proliferation, dedifferentiation, migration, and invasion in cancer cells 135 .
Several kinase inhibitors targeting the BCR-ABL fusion protein have been developed and used clinically (Table 4). Imatinib is an orally active first-generation BCR-ABL inhibitor. Imatinib is an ATPcompetitive type II TKI that binds to the inactive conformation of the ABL kinase (DFG-out conformation 137 ) 135,137 . Mutation in the ATP-interacting gatekeeper residue of the ATP-binding pocket (T315I) leads to maintenance of the active conformation of ABL and resistance to imatinib and related TKIs 137 . The amide substitution in the central aminophenyl ring of imatinib is crucial for tyrosine kinase inhibition, and the 6-methyl residue in the aminophenyl ring increases selectivity for BCR-ABL 137 . Due to the structural similarity among ABL, c-Kit, and PDGFRs (class III RTK) 25 , imatinib also inhibits PDGFR and c-Kit 8,135,137,138 . Secondgeneration BCR-ABL inhibitors have been developed and clinically utilized to overcome imatinib resistance caused by ABL kinase point mutations. Nilotinib is an ATP-competitive and orally active type II kinase inhibitor with greatly improved potency compared to imatinib 137,138 . Similar to imatinib, nilotinib inhibits the inactivated conformation of the ABL kinase, and resistance in the presence of BCR-ABL harboring the T315I mutation has been reported; however, nilotinib suppresses most imatinib-resistant BCR-ABL mutants and is not a substrate of drug influx/efflux transporters [137][138][139] . In addition, nilotinib displays inhibitory effects regarding activation of multiple kinases, such as c-Kit, PDGFR, the ABL-related kinase ARG, DDR1 kinase, oxidoreductase NQO2, and  ephrin receptor EPHB4 138 . Bosutinib is an orally active and ATPcompetitive dual SFK/ABL inhibitor 135,137,138 showing similar inhibitory effects against mutated or amplified BCR-ABL associated with imatinib resistance 137,140 and BCR-ABL harboring the T315I mutation 137,140 . Accordingly, bosutinib has been used for treatment of patients with Ph + CML who are resistant to or intolerant of imatinib 141 . Other agents approved in the clinic include radotinib, an orally active second-generation BCR-ABL inhibitor that exhibits inhibitory effects on wild-type and some imatinibresistant mutant forms of BCR-ABL and PDGFR 8,142 , and asciminib, an allosteric inhibitor that binds to the myristate pocket of BCR-ABL and is effective against T315I-mutant BCR-ABL 143,144 . Ponatinib, an orally available third-generation inhibitor against both wild-type and T315I-mutant BCR-ABL 135,137,145 , also displays inhibitory effects on the activity of multiple kinases, including FLT3, c-Kit, VEGFR, PDGFR, and Src 145 . Since 2012, ponatinib has been used for treatment of patients with T315I-positive CML (including accelerated phase, chronic phase, or blast phase) or those with T315I-positive Ph + ALL 145 . Additional BCR-ABLtargeting inhibitors have been developed and evaluated preclinically and clinically 137 . Src-family kinases (SFK: Blk, Fgr, Frk, Fyn, Hck, Lck, Lyn, Src, Yes, and Yrk) contain a conserved domain organization consisting of a myristoylated N-terminal segment (SH4 domain), followed by SH3, SH2, linker, and tyrosine kinase domains and a short C-terminal tail 146,147 . Among SFKs, Src, Fyn, and Yes are ubiquitously expressed; Hck, Lck, Lyn, Blk, Yrk, and Fgr are primarily expressed in hematopoietic cells and Frk-related kinases in epithelial-derived tissues. Similar to ABL, SFKs adopt an inactive conformation through autoinhibitory intramolecular interactions involving phosphorylation at Y527/Y530 146 . Dephosphorylation of Y527/ Y530 causes destabilization of intramolecular interactions, leading to SFK activation by interaction with RTKs, G protein-coupled receptors, and focal adhesion kinase via its SH2 or SH3 domains and subsequent autophosphorylation at Y416/Y419 146,147 . Activated SFKs play a crucial role in cell proliferation, adhesion, migration, invasion, metastasis, angiogenesis, and therapeutic resistance in cancer and act as key nodes of multiple oncogenic signal transduction pathways 147,148 , indicating the potential of SFK targeting for efficacious anticancer therapeutic regimens. ABL in the active conformation is structurally similar to SFKs 138,149 , and dasatinib, which preferentially interacts with the active conformation of the ABL kinase domain 135,137,138 , shows inhibitory effects on SFKs 149 . Dasatinib targets multiple kinases, including c-Kit, PDGFR, and SFK (Src, Fgr, Fyn, Hck, Lck, Lyn, and Yes) 138,146,149 , but not BCR-ABL harboring the T315I mutation 137 . Currently, there are no clinically approved anticancer regimens targeting SFKs, and some clinical trials evaluating the effectiveness of SFK inhibitors are still ongoing 150 .
BTK and JAK inhibitors: BTK is a nonreceptor tyrosine kinase that plays an essential role in the development and function of B cells 151 . BTK contains five typical domains, including from the N-terminus to the C-terminus the pleckstrin homology (PH) domain required for binding to phosphatidylinositol lipids, the proline-rich TEC homology (TH) domain, a zinc-finger motif for optimal activity and stability of the protein, SH3 and SH2 domains, and the catalytic domain 151,152 . Antigen engagement by the B-cell receptor causes activation of BTK through phosphorylation at Y551 in the kinase domain by spleen tyrosine kinase (Syk), Lyn, or Src 152 , which leads to subsequent autophosphorylation at Y223 in the SH3 domain and activation of downstream signaling pathways, including phospholipase Cγ, mitogen-activated protein kinase (MAPK), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-кB), and Akt, leading to regulation of B cell survival, proliferation, differentiation, and antibody secretion 151,152 Overexpression and hyperactivation of BTK have been observed in a number of non-Hodgkin B-cell malignancies, including chronic Table 4.  151,152 . The Janus kinase (JAK) family comprises the nonreceptor tyrosine kinases JAK1, JAK2, JAK3, and TYK2 153,154 . Cytokine binding to receptors leads to receptor dimerization and recruitment, trans-autophosphorylation, and activation of JAK, resulting in phosphorylation and activation of downstream signaling cascades such as phosphatidylinositol-3kinase (PI3K)/Akt, MAPK, and signal transducer and activator of transcription (STAT) transcription factors [153][154][155] . Deregulation of JAK through hyperactivation and activating mutations (e.g., JAK2 V617F) has been reported in myeloproliferative neoplasms, including myelofibrosis 154 . Examples of clinically approved inhibitors targeting BTK (e.g., ibrutinib, acalabrutinib, and zanubrutinib) or JAK (e.g., ruxolitinib and fedratinib) are listed in Table 4.
Inhibitors targeting downstream signaling pathways: RAS inhibitor and serine/threonine kinase inhibitors. Activated tyrosine kinases trigger phosphorylation and activation of downstream signaling mediators that are mostly serine/threonine kinases. The main relevant downstream signaling pathways are the PI3K/Akt/ mammalian target of rapamycin (mTOR) and RAS/RAF/MEK/ERK pathways. Alterations in several components of these pathways (e.g., RAS, RAF, MEK, and PI3K) have been found in various types of cancer and thus considered druggable targets [156][157][158] . Cyclins are also downstream effector molecules of these signaling cascades and play an important role in regulating cell cycle progression and various cellular processes, such as gene transcription, DNA damage repair, and metabolism, by associating with cyclin-dependent kinases (CDKs) 159 . Alterations in cyclins and CDKs have been observed in various cancer types, and several CDK inhibitors have been developed and approved for clinical use 160 . Examples of these targeted therapeutic drugs are described below.
RAS/RAF/MEK inhibitors: RAS is a guanine nucleotide-binding protein that plays an important role in cell proliferation and differentiation, and farnesylation of RAS by RAS farnesyltransferase (FTase) is crucial for RAS to associate with membranes and its transforming activity 161 . Mutations in RAS result in constitutive activation 161 . Among the three RAS isoforms (KRAS, HRAS, and NRAS), KRAS is the most frequently mutated isoform, and five mutations (G12D, G12V, G12C, G13D, and Q61R) are the most prominent RAS mutations observed in cancer patients 156 . Based on the important role of RAS FTase in the regulation of RAS transforming activity, several FTase inhibitors have been developed and evaluated, yet none of them have been clinically used because of limited efficacy 162 . Recently, a small molecule inhibitor targeting mutated KRAS (KRAS G12C ) was developed and approved for clinical use. Sotorasib is an orally available inhibitor that binds to inactive guanosine diphosphate (GDP)-bound KRAS via a covalent bond between the C12 residue and the acrylamide warhead and noncovalent bonds between the isopropylpyridine substituent and a cryptic pocket comprising H95, Y96, and Q99 residues; this results in inhibition of KRAS G12C without affecting wild-type KRAS 163,164 . Another KRAS G12C inhibitor, adagrasib (MRTX849), is under clinical trial evaluation 165 . Activated RAS in the GTP-bound state leads to association of RAF proteins, causing formation of RAF homo-or heterodimers, RAF phosphorylation, and consequent activation of the downstream signaling mediators MEKs and ERKs 157,166 . Among the three isoforms of RAF (ARAF, BRAF, and CRAF), mutations in BRAF, especially at the V600 residue (e.g., V600E) in the activation loop, are frequently observed in several types of cancer, including melanoma, papillary thyroid cancer, and colorectal cancer 157,166,167 . Indeed, the V600E mutation, which causes RASindependent activation of BRAF, accounts for more than 90% of BRAF mutation cases in cancer 157,166,167 . Thus far, three RAF inhibitors and three MEK inhibitors have been used for anticancer treatment. Currently available RAF inhibitors target monomeric V600E-mutant BRAF; thus, for dimeric RAF, inhibition of one protomer by the drug paradoxically leads to transactivation of the other protomer and downstream signaling 157 . Therefore, a combination of MEK inhibitors (e.g., vemurafenib plus cobimetinib, dabrafenib plus trametinib, and encorafenib plus binimetinib) has been clinically utilized 168 . Examples of clinically approved BRAF and MEK inhibitors are listed in Table 5. PI3K/mTOR inhibitors: The PI3K/Akt/mTOR pathway plays a central role in regulating cell proliferation, survival, growth, and metabolism 158,169 . Deregulation of the PI3K/Akt/mTOR pathway through mutation or amplification of PIK3CA (encoding the p110α subunit of PI3K), loss or inactivation of phosphatase and tensin homolog (PTEN), and hyperactivation of mTOR have been commonly found in various cancer types 158,169 and related anticancer drug resistance 158,170 . Hence, inhibitors targeting PI3K, Akt, and mTOR have been evaluated in preclinical studies and clinical trials, and some inhibitors have been used clinically for cancer treatment.
Because of the specific expression of PI3K, p110γ, and p110δ subunits in the hematopoietic system, the association of the PI3K pathway with regulating B-cell receptor (BCR) signaling, and the undesirable toxicity of pan-PI3K or dual PI3K/mTOR inhibitors 171,172 , PI3K inhibitors that specifically target PI3Kδ or PI3Kγ have been employed for treatment of patients with lymphoma. Some mTOR inhibitors, especially rapamycin analogs (rapalogs) that form a complex with FK506-binding protein 12 (FKBP12) and inhibit mTORC1 (but not mTORC2) activity, have been approved for clinical use 8 . Additionally, ATP-competitive mTOR inhibitors have been developed and are under preclinical and clinical evaluation 8 . Examples of clinically utilized PI3K (e.g., idelalisib, duvelisib, copanlisib and alpelisib) and mTOR inhibitors (e.g., sirolimus, temsirolimus, and everolimus) are listed in Table 5. CDK inhibitors: Among more than 20 members of CDK family proteins 159 , CDK4 and CDK6 (in complex with cyclin D) play a crucial role in promoting cell cycle progression by sequestering CDK inhibitors and inducing various proteins involved in cell cycle progression from G1 to S phase, DNA replication, chromatin structure, chromosome segregation, and the spindle assembly checkpoint through phosphorylation of various targets, including retinoblastoma protein (RB), and activating E2F-mediated transcription 160 . Hence, CDK4/6 has been considered attractive] for targeted anticancer therapy. Three CDK4/6 inhibitors have been used clinically for treatment of patients with HR-positive advanced breast cancer (Table 5). Palbociclib, ribociclib, and abemaciclib are orally available, reversible, and selective CDK4/6 inhibitors that have been used clinically in combination with an aromatase inhibitor for treatment of postmenopausal women with ER-positive and HER2-negative advanced or metastatic breast cancer 8,173,174 .
Other targeted anticancer agents. In addition to PARP inhibitors, other types of clinically used or recently approved targeted therapies, including epigenetic modulators (e.g., DNA methyltransferase inhibitors, histone deacetylase inhibitors, EZH2 inhibitors, and isocitrate dehydrogenase inhibitors), proteasome inhibitors, Bcl-2 inhibitors, and smoothened inhibitors, are summarized in Table 6.
PARP inhibitors: The PARP family plays a crucial role in regulating DNA repair processes upon the DNA damage response (DDR) and chromatin modulation 175,176 . PARP family proteins, especially PARP1 and PARP2, bind to DNA lesions and mediate poly-ADP ribosylation (PARylation) of chromatin and DNA damage response components, resulting in DNA repair by recruiting DNA repair effectors such as XRCC1 175,176 . After autoPARylation, PARP  dissociates from DNA, and the DNA repair process is completed by recruitment of DNA repair proteins 176 . BReast CAncer gene 1 (BRCA1) and BRCA2 (BRCA1/2) are tumor-suppressor genes that play a key role in repair of double-strand DNA breaks via the conservative homologous recombination repair (HRR) process 175,177 . Mutations in BRCA1/2 genes have been found in some cancer types, including breast, ovarian, pancreatic, and prostate cancers 177 . Defects in BRCA function due to BRCA1/2 gene mutations cause loss of the HRR process and mediate the DNA repair process in a nonconservative manner, such as nonhomologous end joining, leading to DNA alteration 175  Summary and future perspectives in the development of molecular targeted therapy Owing to advances in molecular diagnosis, genome-wide analysis, and in-depth understanding of cancer biology, numerous tyrosine kinase inhibitors have recently been developed, tested preclinically and clinically, and utilized for cancer treatment in the clinic. Nevertheless, poor efficacy, toxicity, and tumor relapse due to drug resistance are major obstacles for targeted therapy-based efficacious anticancer treatment. Therefore, further investigation is required to develop efficacious personalized targeted therapies that overcome drug resistance and reduce side effects and toxicity.
To this end, a fundamental template for drug discovery by identifying additional druggable targets through in-depth biochemical, genomic, and molecular studies and structural investigations is needed. Drug discovery with different chemical entities or modes of action is also necessary for the development of molecular targeted therapy. In addition to direct or allosteric modulation of cellular targets, strategies for indirect manipulation of cellular targets [e.g., posttranslational modification 179 or targeted protein degradation using proteolysis-targeting chimera (PROTAC) 180 ] based on biological and functional studies for cancer-specific modulation would be applicable. Furthermore, the development of small molecule inhibitors that concurrently block signaling pathways associated with cancer cell proliferation and drug resistance and design of optimized combinatorial therapeutic strategies using molecular targeted therapy, either alone or in combination with other types of anticancer therapy (e.g., chemotherapy and immune checkpoint inhibitors), would be of importance for increased efficacy, limited toxicity, and minimal drug resistance.
Because the side effects and toxicity of targeted therapy are mediated by nonspecific inhibition of the same target in normal cells 10 , strategies for cancer cell-specific targeting are also important. A relevant example is the recent development of KRAS G12C inhibitors. Since the clinical failure of farnesyltransferase inhibitors, KRAS has been considered an undruggable target 181 . In a recent study utilizing the high reactivity of cysteine, compounds that covalently bind to KRAS via the mutated cysteine residue and allosterically inhibit GTP binding to KRAS were designed 182 ; this approach can inhibit KRAS without occupying the GTP/GDPbinding pocket on the surface and achieve specificity for mutant   KRAS beyond wild-type KRAS, thus avoiding the unfavorable effects caused by inhibition of wild-type KRAS 182,183 . Based on this innovative study and a better understanding of the crystal structure of mutant KRAS, several potent KRAS G12C inhibitors have been developed and approved for clinical use 183,184 ; agents targeting other types of mutant KRAS, such as KRAS G12D , have also been developed and evaluated in preclinical settings 185,186 . Studies on molecular diagnosis and discovery of predictive biomarkers are necessary to properly select eligible populations for better efficacy and reduced toxicity 183 . Several newly developed approaches, such as next-generation sequencing technology 187 , whole-genome sequencing 188 , and machine learning 189 , can be applied to this end. In fact, artificial intelligence (AI)-based strategies 190 are expected to be extensively utilized for the design of the structure and chemical synthetic procedures, identification of potential hits, prediction of pharmacokinetic profiles, assessment of side effects and toxicity, and drug repurposing. Finally, emerging evidence has shown the role of the host microbiome in cancer development and progression, drug responsiveness, and therapy-induced side effects 191,192 . For example, the gut microbiome promotes the function of mutant p53 toward oncogenicity 193 and modulates responsiveness to antitumor therapy such as anti-PD-1 immunotherapy 194 . A number of investigations into the influence of the gut microbiome on chemotherapy and anticancer immunotherapy are ongoing; however, the effect of the host microbiome on molecular targeted therapy remains elusive. Further studies are necessary to investigate the role of the host microbiome in the efficacy and toxicity of molecular targeted therapy and to identify key factors to develop safer and more efficacious therapeutic strategies based on microbiome-targeted therapy.
In summary, the present paper briefly reviews the current status of molecular targeted therapy and discusses future directions, providing novel therapeutic strategies with better efficacy and safety to improve the prognosis of cancer patients.