Strategies to overcome drug resistance using SHP2 inhibitors

Abstract

Encoded by PTPN11, the SHP2 (Src homology-2 domain-containing protein tyrosine phosphatase-2) is widely recognized as a carcinogenic phosphatase. As a promising anti-cancer drug target, SHP2 regulates many signaling pathways such as RAS-RAF-ERK, PI3K-AKT and JAK-STAT. Meanwhile, SHP2 plays a significant role in regulating immune cell function in the tumor microenvironment. Heretofore, five SHP2 allosteric inhibitors have been recruited in clinical studies for the treatment of cancer. Most recently, studies have proved the therapeutic potential of SHP2 inhibitor in overcoming drug resistance of kinase inhibitors and programmed cell death-1 (PD-1) blockade. Herein, we review the structure, function and small molecular inhibitors of SHP2, and highlight recent progress in overcoming drug resistance using SHP2 inhibitor. We hope this review would facilitate the future clinical development of SHP2 inhibitors.

Graphical abstract

This review summarizes the structure, function and small molecular inhibitors of SHP2, and highlights recent progress in overcoming drug resistances combining SHP2 inhibitor with its related signaling pathway inhibition.

1. Introduction

Tyrosine phosphorylation is a dynamic and reversible post-translational modification, which plays a vital role in a wide range of cellular functions, including cell proliferation, differentiation, survival or apoptosis, and oncogenic transformation^1,2. This dynamic modification is mediated by protein tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPs). Tyrosine phosphorylation mediates the dynamic and critical regulatory processes of most intracellular signaling pathways, and signal disorders are recognized as the cause of the diseases^3,4. At present, many drugs targeting PTKs have been approved by U.S. Food and Drug Administration (FDA)⁵. Due to the incomplete understanding of PTPs, the unacceptable selectivity of existing inhibitors, and poor pharmacokinetic properties, there are no PTP targeting drugs in clinical⁶.

The PTPs family consists of CX₅R motif, and the structural PTPs consists of a membrane domain (D1) and a proximal domain (D2)⁷. The D1 domain is mainly responsible for catalytic activity, while the D2 domain has almost no activity. According to the structure of the extracellular domain, PTPs can be divided into three subgroups: dual-specificity PTPs (DUSP-PTPs), low molecular weight PTPs (LMW-PTPs), and high molecular weight PTPs (HMW-PTPs)⁸. According to the location, PTPs can be divided into receptor-like transmembrane PTPs and non-receptor cytoplasmic PTPs⁴. The increased tyrosine phosphorylation activity caused by overactivation or overexpression of PTKs are a marker of many cancers⁹, while PTPs are considered to be a negative regulator of signal pathways and tumor suppressor gene products by regulating dephosphorylation. In fact, abnormal PTPs activity can also lead to the occurrence and development of many human disorders such as cancer, metabolic and autoimmune diseases, infectious diseases and neurodegeneration¹⁰.

The non-receptor protein tyrosine phosphatase Src homology-2 domain-containing protein tyrosine phosphatase-2 (SHP2) plays a critical role in many cancer-related signaling pathways, such as RAS-RAF-ERK, PI3K-AKT and JAK-STAT¹¹. Heretofore, five allosteric inhibitors have been recruited for clinical studies, suggesting SHP2 as a promising anti-cancer drug target^6,12, 13, 14. Most recently, a number of studies have shown that SHP2 allosteric inhibitors can be combined with other protein inhibitors to overcome drug resistance. The combined therapies using SHP2 inhibitor has been proved to be more effective than monotherapy. Moreover, SHP2 is also the convergence node of multiple signal pathways in immune cells and cancer cells. In T cells, SHP2 participates in the downstream signal transduction of immunosuppressive receptor PD-1, which is the key immune checkpoint of cancer immunotherapy. Blocking PD-1 or SHP2 can induce T cell help (Th1) immunity, activate T cells and eliminate the immunosuppressive effect of cancer^15,16. It should be mentioned that many oncogenic mutants in SHP2 may cause drug resistance and hamper the clinical development of SHP2 inhibitors. A novel approach based on dual allosteric inhibition may help to improve the inhibition rate of mutants and overcome drug resistance. In addition, it will be a new direction to chemically induce SHP2 degradation using proteolysis-targeting chimeras (PROTACs) technology.

Herein, we introduce the structure and functions of SHP2, and briefly review the development of SHP2 inhibitors. Importantly, we summarized the recent strategies to overcome drug resistance and synergistic tumor immunotherapy using SHP2 inhibitors. We also discuss the future clinical applications of SHP2 inhibitors, hoping to provide a certain reference for the future drug development.

2. Structure and self-inhibition of SHP2

As non-receptor protein tyrosine phosphatase, SHP2 is encoded by the PTPN11 gene and contains 593 amino acid residues. The structure of SHP2 consists of two SH2 domains (N-SH2 and C-SH2), a PTP catalytic domain and a C-terminal with two tyrosine phosphorylation sites (Y542/Y580)^17,18. The C-SH2 domain consists of 112–215 residues responsible for the binding energy but doesn't contribute to the activation of SHP2. On the contrary, the N-SH2 domain contains 2–104 amino acids and acts as a conformational switch in the activation of SHP2^19,20. The PTP catalytic domain is composed of 220–525 residues, of which Cys459 is a highly active cysteine in the conserved characteristic motif of the PTP catalytic domain and has essential catalytic functions^19,21. The C-terminal tail (Tyr542, Tyr580) can be phosphorylated during extracellular stimulation22, 23, 24 (Fig. 1A). In the basal state, SHP2 maintains a self-inhibition state with low catalytic activity through the intramolecular interaction between the N-SH2 domain and the PTP domain¹⁹. Binding of growth factors or cytokine abolish the self-inhibition and activates SHP2^25,26 (Fig. 1B). Therefore, maintaining the self-inhibited conformation would effectively inhibit the activity of SHP2, which provide new mechanism for the design of SHP2 inhibitors²⁷.

Figure 1.

(A) SHP2 functional domains. (B) The self-inhibition state, activation state and inhibitors action sites of SHP2.

At the cellular level, SHP2 located in the cytoplasm and plasma membrane. SHP2 plays a crucial role in different receptor signal pathways, mediating cell growth, cell cycle maintenance, differentiation, migration, adhesion and apoptosis¹⁴. At the same time, SHP2 is also involved in some events in the nucleus and mitochondria²⁸. Mice carrying SHP2 deletion alleles are fatal in embryos²⁹, and some experimental results have shown that SHP2 is necessary for the early development of mice^30,31. Studies on tissue-specific conditional gene knockout mice have revealed the various functions of SHP2⁴.

3. SHP2 mutations and diseases

SHP2 mutations lead to dysregulated enzymatic activities, which lead to various diseases such as Noonan syndrome (NS), leopard syndrome (LS), juvenile myelomonocytic leukemia (JMML), acute myelogenous leukemia (AML), myelodysplastic syndrome (MDS), and human malignant tumors³² (Fig. 2). PTPN11 gene mutations occur at a low rate in various types of solid tumors, AML, and neuroblastoma³³. SHP2^E76K is one of the most common SHP2 mutations found in leukemia and solid tumors and exhibited 20-fold increase in the basal phosphatase activity^6,34. NS is an autosomal dominant genetic disease. Germline mutations of PTPN11 are found in about 40%–50% of NS patients. In the PTP domain, the SHP2 related mutant N308D, which accounts for 25% of cases, is 3-fold more active than wild-type SHP2; Asn308 is a mutation hotspot of NS; the other two SHP2 mutants N308S and Q506P show higher catalysis only under specific substrate induction active^4,35. Other mutants T42A, D106A and E139D show relatively basal PTP activity, and low levels of p-Tyr peptide can effectively activate these SHP2 mutants. LS is a rare autosomal dominant genetic disease. Germline mutations in PTPN11 have been found in at least 80% of LS. These mutations weaken the intramolecular interaction between the N-SH2 domain and the PTP domain, resulting in a conformational change in SHP2. Y279C and T468M are the most common mutants that can significantly activate the RAS/ERK signaling pathway^8,36. About 10% of NS also develop into JMML, a relatively rare leukemia that affects approximately one in a million children with a poor prognosis and prone to relapse. The only treatment currently available is bone marrow transplantation^37,38. Studies have shown that leukemia and solid tumor mutations (D61Y and E76K) have higher catalytic activity than NS-related mutations (N308D), which indicate that the PTP activity of leukemia should be higher than that of NS³⁹. Therefore, we speculate that low levels of activated SHP2 cause abnormal diseases, while high levels of activated SHP2 can cause cancer. According to the position and function of mutations, it can be divided into six groups. Group I mutations located in the interaction between N-SH2 and PTP domains, which destroyed the self-inhibition conformation of SHP2. So far, 506 positions mutations in the N-SH2 and PTP domains have been reported and most of them are located at this interface^35,40. For example, SHP2^E76Q and SHP2^S502P mutations resulted in the loss of Glu76-Ser502 hydrogen bond between N-SH2 and PTP domains⁶. Groups II and III of mutations include residues that are exposed to the surface of the PTP domain to stabilize inactive and catalytic conformations. Different from the functions of the first three groups of mutations, group IV mutations have a certain effect on maintaining the entire PTP structure or participating in the interaction of catalytic amino acids. For example, SHP2^F285S destroys the hydrophobic B’C loop of the PTP domain, resulting in a closed conformation opening 2 Å⁴¹. Group V mutations are residues in the phosphor-peptide binding pockets of the two SH2 domains, destroying the effect of binding affinity and specificity. In the last group, the mutated residues are located in the junction region between the N-SH2 and C-SH2 domains, regulating the direction of SH2, and a few mutations are located in this region^32,36.

Figure 2.

Schematic diagram of SHP2 mutations that occur in human diseases (PDB ID: 2SHP).

4. Role of SHP2 in signaling pathways

SHP2 plays an essential role in various signaling pathways⁸. SHP2 regulates physiological and pathological processes through positive (signal enhancement) and/or negative (signal inhibition) of signal transduction pathways in various growth factors, cytokines and extracellular matrix receptors induced signaling pathways^17,42 (Fig. 3). Under the stimulation of cytokines (IL-3, IL-6, GM-CSF, CagA) and growth factors (PDGF, EGF, FGF), the PTP activity (dephosphorylation) of SHP2 is necessary for the complete activation of RAS/RAF/ERK signaling pathway^17,43. As a key GTPase, RAS produces inactive RAS-GDP under the control of GTPase-activating proteins (GAP) (NF1, p120RASGAP), and produces active RAS-GTP under the control of guanine nucleotide exchange factors (GEF)⁴⁴, which cyclically transmits signals from outside the cell to the nucleus⁶. Obviously, GAP has an inhibitory effect on RAS activation. Located at the downstream of receptor tyrosine kinase (RTK), SHP2 dephosphorylate epidermal growth factor receptors (EGFR, EGFR2) and p120RASGAP to indirectly activate RAS^45,46. On the other hand, SHP2 could indirectly induce RAS activation by activating Src family kinases. SHP2 dephosphorylates CSK, a negative regulator of Src or paxillin, which leads to the dissociation of CSK and promotes the activation of Src family kinases47, 48, 49. Moreover, SHP2 can promote dephosphorylation of sprouty, a negative regulator of RAS-MAPK. Dephosphorylated sprouty loss the ability to bind to growth factor receptor-bound protein 2 (GRB2), which promotes the recruitment of GRB2/SOS complexes to fibroblast growth factor receptor substrate (FRS) and finally activate RAS^4,8,12,43,50. Meanwhile, the dephosphorylation of sprouty-related-1 (Spred1) protein by SHP2 weakens the inhibitory effect of Spred1 protein on RAS-ERK pathway^51,52. SHP2 not only can activate the RAS-RAF-ERK signaling pathway with PTP catalytic activity (dephosphorylation), but also act as a scaffolding adaptor that connects upstream and downstream signals to activate the RAS-RAF-ERK signaling pathway³². When stimulated by cytokines or growth factors, SHP2 recruits and binds GRB2 associated binding protein-1/2 (GAB1/2), GRB2, insulin receptor substrate 1 (IRS1), FRS2 and other proteins, resulting in ERK activation. Therefore, the scaffolding adaptor function of SHP2 is significant for the activation of ERK signaling pathway53, 54, 55, 56. Based on current studies, SHP2 plays a positive role in RAS-RAF-ERK signaling pathway.

Figure 3.

Schematic diagram of SHP2 related cytokines and growth factor dependent RAS-RAF-ERK, PI3K-AKT and JAK-STAT signaling pathways. SHP2 dephosphorylates negative regulators in RAS-RAF-ERK pathway through a variety of mechanisms, such as RAS-GAP, paxillin and sprouty, to regulate signal transduction, tumor invasion, cell proliferation, differentiation, apoptosis and survival; SHP2 inhibits the activation of PI3K-AKT pathway and regulates cell proliferation and apoptosis through dephosphorylation of α-catenin, Gab1 and P85 binding sites; SHP2 dually regulates the JAK-STAT signaling pathway, which is essential for regulating DNA damage, cell growth, differentiation, survival and death.

PI3K/AKT signal is an important signal pathway that regulates biological and pathophysiological responses such as cell growth, metabolism and survival^57,58. SHP2 can dually regulate PI3K-AKT signal with PTP catalytic activity (PTP dependent) or scaffold function (PTP independent)^59,60. For example, in structurally activated fibroblast growth factor receptor 3 (FGFR3) induced cells, PTP with catalytic activity promoted α-catenin dephosphorylation to activate PI3K/AKT pathway^32,61; SHP2 selectively dephosphorylates platelet-derived growth factor receptor (PDGFR), shortens the binding time of PI3K and RASGAP with receptors and activates PI3K^62,63. These indicate a positive regulation dependent on PTP activity. However, several studies have shown that SHP2 inhibits the activation of PI3K pathway induced by EGF through dephosphorylation of Gab1 and P85 binding sites, indicating a negative PTP dependent regulation^61,64,65. In vascular endothelial growth factor receptor 2 (VEGFR2) mediated ATK signaling, SHP2 forms complexes with Gabs to promote the activation of PI3K pathway, which indicates that SHP2 may play a role as a scaffolding adaptor, leading to PI3K/AKT pathway activation in a manner independent of PTP catalytic activity66, 67, 68, 69.

STAT protein plays an essential role in the physiological functions of cells. The post-translational modification of dephosphorylation involved in SHP2 double-regulates the STAT signaling pathway⁷⁰. SHP2 has a positive function in JAK/STAT pathway. For example, SHP2 indirectly activates STAT5 phosphorylation by activating JAK2‒PrlR complexes and promotes STAT5 activation in mice mammary glands. SHP2 deletion can significantly inhibit STAT5 activity^6,71,72; in SHP2 mutant cells, the JAK2/STAT5 signal stimulated by IL3 was impaired, and wild-type SHP2 could reactivate this signal. In SHP2 inactivated cells, JAK2 activity and STAT5 phosphorylation are decreased³⁶; The Tyr1017 phosphorylation site of JAK forms a complex with SOCS, which prevents JAK from binding with STAT, resulting in the inhibition of JAK-STAT signal. SHP2 can dephosphorylate the tyrosine phosphorylation site of JAK and prevent JAK from binding with SOCS, thus reactivating STAT signaling pathway^8,73. Meanwhile, SHP2 also has a negative regulatory effect on the JAK-STAT signaling pathway. Under IL-3 stimulation, the overexpression of SHP2 increased the dephosphorylation level of STAT5 in BaF3 cells and primary bone marrow hematopoietic progenitor cells, thereby inhibiting STAT5 activity74, 75, 76; STAT3 is a protein that plays a vital role in embryonic stem cell differentiation and hematopoiesis. SHP2 promotes the dephosphorylation of STAT3, thereby negatively regulating the STAT3 signaling pathway⁷⁷; in fibroblasts, SHP2 can dephosphorylate activated STAT1 and down regulate the activity of JAK1–STAT1 signaling pathway induced by IFN^32,78. In conclusion, SHP2 directly or indirectly regulates JAK/STAT mediated signal transduction in a receptor specific or cell specific manners^6,32.

In addition to the mentioned signaling pathways, SHP2 also participates in the regulation of many other signaling pathways through dual regulation. For example, nuclear factor kappa-B (NF-κB), c-Jun N-terminal kinase (JNK), nuclear factor of activated T-cells (NFAT) signal pathways, etc^32,79.

5. Overview of small molecular SHP2 inhibitors

5.1. Catalytic site inhibitors

Since the PTP catalytic sites are positive charge in nature, the catalytic site inhibitors usually possess ionizable functional groups to facilitate interaction with the active-site. In this regard, based on the inhibitors structural characteristics can be divided into the following categories (Fig. 4).

Figure 4.

Structures and inhibitory activities of SHP2 catalytic site inhibitors.

Quinoline hydrazine derivatives NSC-87877 (1). Compound 1 is a SHP2 inhibitor identified for the first time through screening in 2006. It can effectively inhibit SHP2 (IC₅₀ = 0.32 μmol/L) and has higher selectivity for SHP2 than other PTPs (PTP1B, DEP1, HEPTP, LAR, CD45), but no selectivity against SHP1 (IC₅₀ = 0.36 μmol/L) in vitro. Compound 1 inhibits the PTP domain's catalytic activity in SHP2 and effectively blocks EGF-induced RAS/ERK1/2 activation^80,81.

Phenylhydrazonopyrazolone sulfonate derivatives PHPS1 (2a). In 2008, Hellmuth et al.⁸² reported a potential phosphotyrosine inhibitor 2a as a selective SHP2 inhibitor (SHP2 IC₅₀ = 2.1 μmol/L) over SHP1 (SHP1 IC₅₀ = 30 μmol/L) and PTP1B (PTP1B IC₅₀ = 19 μmol/L). Compound 2a is not toxic to normal epithelial cells and can prevent the anchorage-dependent growth of various tumor cells. Mechanism studies have shown that 2a binds to the PTP domain of SHP2 and inhibits the SHP2-dependent RAS-MAPK pathway. Specially, the sulfonic acid group is a p-Tyr mimic and extends to the substrate binding pocket. By introducing different substituents in the 2,4-dihydro-3H-pyrazol-3-one scaffold, Grosskopf et al.⁸³ reported 2b (SHP2 IC₅₀ = 0.37 μmol/L), 2c (SHP2 IC₅₀ = 0.15 μmol/L) with improved activity. GS-493 (2d) exhibited the good SHP2 inhibitory activity (SHP2 IC₅₀ = 0.07 μmol/L) and PTP selective (SHP1 IC₅₀ = 2.08 μmol/L; PTP1B IC₅₀ = 3.17 μmol/L).

Oxindole derivatives NSC-117199 (3a). Lawrence et al.⁸⁴ reported a potential selective oxindole SHP2 inhibitor 3a (SHP2 IC₅₀ = 46.8 μmol/L, SHP1 IC₅₀ = 68 μmol/L, PTP1B K_i = 96.7 μmol/L) through virtual screening. The structure of 3a was further optimized to obtain bis-carboxylic acid derivatives 3b and 3c. The activity of 3b (SHP2 IC₅₀ = 0.8 μmol/L) is higher than that of 3c (SHP2 IC₅₀ = 15.8 μmol/L), and 3b also possess 20-fold selectivity against SHP1. These results indicated the importance of the position of the carboxyl group in the hydrazine aromatic ring. Using 3a as a lead compound, the author identified 3d with higher activity (SHP2 IC₅₀ = 1 μmol/L) and selectivity (SHP1 IC₅₀ = 18.3 μmol/L; PTP1B IC₅₀ = 14.5 μmol/L). Although the introduction of sulfonamide improves the solubility of 3d, it contains negatively charged carboxyl groups, resulting in poor membrane permeability and bioavailability.

Salicylic acid derivatives (4 and 5). Zhang et al.⁸⁵ found that p-Tyr mimics salicylic acid 4 inhibit SHP2 with IC₅₀ value of 212 μmol/L. They developed a series of substituted salicylic acid derivatives through the click reaction. Among them, 5a (SHP2 IC₅₀ = 5.5 μmol/L) has moderate potency and moderate selectivity than other PTPs (SHP1 IC₅₀ = 15.7 μmol/L; PTP1B IC₅₀ = 14.3 μmol/L). In cellular assays, 5a can block the activation of ERK1/2 stimulated by growth factors and inhibit the hyperproliferation of hematopoietic cells induced by the granulocyte-macrophage colony-stimulating factor (GM-GSF) through SHP2 gain-of-function mutants^80,86.

Diterpenoid quinone derivatives (6). Liu et al.⁸⁷ screened a natural product database and identified cryptotanshinone 6, which possess moderate SHP2 inhibitory activity (SHP2 IC₅₀ = 22.5 μmol/L) and low selectivity (SHP1 IC₅₀ = 39.5 μmol/L; PTP1B IC₅₀ = 33.5 μmol/L).

Other inhibitors (7 and 8). Wu et al.⁸¹ identified 7 as dual SHP1/2 inhibitor (SHP1 IC₅₀ = 2.3 μmol/L; SHP2 IC₅₀ = 2.1 μmol/L). Zhou et al.⁸⁸ reported a SHP2 inhibitor 8 (IC₅₀ = 2.11 μmol/L) with weak selectivity against SHP1 (IC₅₀ = 4.28 μmol/L) and good selectivity against PTP1B (IC₅₀ = 50.2 μmol/L).

5.2. Allosteric inhibitors

Because of the highly conserved sequence of PTP catalytic domain, developing high selective SHP2 catalytic site inhibitors is still very difficult, which is one of the major challenges in future clinical development. In addition, due to the positive charge environment of PTP catalytic sites, the catalytic site inhibitors are required to possess multiple negative ionizable functional groups. These functional groups usually have low membrane permeability and oral bioavailability, which are factors that hinder the possibility of such inhibitors to become approved drugs²¹.

SHP2 allosteric inhibitors are essential components of tumor therapeutic molecules with high therapeutic potential^89,90. At present, four different allosteric binding sites have been reported in SHP2 protein, including tunnel-like site formed by N-SH2, C-SH2 and PTP domains⁹¹, latch-like and groove-like sites located between the N-SH2 and PTP domains⁹², non-conserved cysteine residue 333 (Cys333) site located in the PTP domain (Fig. 5). It should be mentioned that SHP2 allosteric inhibitors targeting the Groove-like site have not yet been reported. According to different binding sites, current SHP2 allosteric inhibitors can be divided into three categories (Fig. 6).

Figure 5.

Allosteric pockets of SHP2 self-inhibitory conformation. PTP domain is colored in orange, N-SH2 in green, C-SH2 in marine, allosteric site in red. (A) Tunnel-like allosteric pocket with allosteric inhibitor SHP099 (10) (PDB ID: 5HER). (B) Latch-like allosteric pocket with allosteric inhibitor SHP244 (16) (PDB ID: 6MBR). (C) Groove-like allosteric pocket (PDB ID: 6MBR). (D) Non-conserved Cys333 allosteric site (PDB ID: 3B7O).

Figure 6.

Structures and IC₅₀ values of representative SHP2 allosteric inhibitors.

Through high-throughput screening, Novartis discovered a novel allosteric inhibitor SHP836 (9) (SHP2 IC₅₀ = 12 μmol/L; SHP^PTP IC₅₀ > 100 μmol/L) based on the aminopyrimidine scaffold. Crystal structure revealed that compound 9 binds to the tunnel-like region formed between the C-SH2, N-SH2 and PTP domains. Structure–activity relationship studies show that the chlorine in the benzene ring is essential for the activity against SHP2^81,93. In 2016, Novartis^6,91 announced a novel SHP2 allosteric inhibitor SHP099 (10) (SHP2 IC₅₀ = 0.07 μmol/L; SHP^PTP IC₅₀ > 100 μmol/L; SHP1 IC₅₀ > 100 μmol/L; PTP1B IC₅₀ > 100 μmol/L) with in vivo activity, highly selectivity, high orally bioavailability, illuminating to the fact that allosteric inhibition can serve as a promising direction for the development of SHP2 inhibitors. Novartis also reported another series of pyrazolopyrimidinone derivatives, of which 11a is a very effective SHP2 inhibitor (IC₅₀ = 6 nmol/L). Unfortunately, the development of 11a was terminated due to high human ether-a-go-go-related gene (hERG) inhibition (hERG IC₅₀ = 4 nmol/L). Finally, through structural optimization, SHP389 (11b) with similar potency and acceptable hERG inhibition (IC₅₀ = 17,000 nmol/L) was produced⁹⁴. In 2019, Novartis⁹⁵ reported the aminopyrimidinone derivative SHP394 (12) which showed higher SHP2 activity (IC₅₀ = 23 nmol/L), better hERG selectivity (hERG IC₅₀ > 30 μmol/L) and pharmacokinetic properties in the Detroit-562 xenograft model, and also resulted in a dose-dependent decrease in tumor growth. The 13 (SHP2 IC₅₀ = 3 nmol/L) identified by Novartis⁹⁶ showed stronger inhibitory activity in vitro, inhibited the growth of ALK rearranged non-small cell lung cancer (NSCLC) cells, and inhibited tumor growth of MGH049 and MGH045-2A xenograft models in vivo. Nichols et al.⁹⁷ reported that RMC-4550 (14) is a more potent small molecule SHP2 allosteric inhibitor (SHP2 IC₅₀ = 1.5 nmol/L; SHP1 IC₅₀ > 10 μmol/L; PTP1B IC₅₀ > 10 μmol/L), which stabilizes the self-inhibitory conformation of SHP2, but has no inhibitory activity on the mutant proteins (SHP2^E76K, SHP2 ^T253M, SHP2^Q257L). Specially, compound 14 can prevent the excessive activation of RAS-ERK signal and inhibit tumor growth by inhibiting the activity of RAS protein.

The interference of the N-SH2/PTP interface leads to the instability of the self-inhibition conformation, which is the primary mechanism of resistance to SHP2 allosteric inhibitors⁹⁸. Moreover, some mutations destroy the integrity of the self-inhibition interaction, and current SHP2 allosteric inhibitors exhibited low activity against specific oncogenic SHP2 mutant proteins, such as SHP2^E76A, SHP2^G60V, SHP2^S502P, etc.⁹⁹. Also, the SHP2^{E76K/T253M/Q257L} and SHP2^T253M/Q257L mutations reduce the inhibitory activity of 10 against SHP2^100,101. These mutations lead to the inherent instability of the self-inhibitory conformation, whereas the binding of 10 and similar type of allosteric inhibitors required a stable self-inhibitory conformation of SHP2⁹⁹. Through structure-based drug design, discovered an effective allosteric inhibitor 15 for the mutant protein SHP2^E76A (IC₅₀ = 0.71 μmol/L) by targeting the tunnel-like site.

By screening against the SHP2^T253M/Q257L double mutant, which is a defective mutant for SHP099 (10), Fodor et al.⁹² identified a novel triazole-quinazolinone molecule SHP244 (16) (SHP2 IC₅₀ = 60 μmol/L; SHP2^T253M/Q257L IC₅₀ = 68 μmol/L; SHP^PTP IC₅₀ > 100 μmol/L; aqueous solubility = 0.047 mmol/L) targeting the latch-like allosteric pocket (Fig. 6B). In order to further improve the activity and aqueous solubility of 16, the authors performed structural optimization and obtained new allosteric inhibitors SHP844 (17) (SHP2 IC₅₀ = 18.9 μmol/L; SHP^PTP IC₅₀ > 100 μmol/L; aqueous solubility = 0.895 mmol/L) and SHP504 (18) (SHP2 IC₅₀ = 21 μmol/L; SHP^PTP IC₅₀ > 100 μmol/L; aqueous solubility = 0.535 mmol/L) (Fig. 6). These inhibitors bound with SHP2 in a similar manner to 16, and both show higher activity to SHP2 and possess improved aqueous solubility. Through X-ray structure analysis, it can be observed that the tunnel-like allosteric pocket is not interfered by the binding of 17 or 18 to the latch-like allosteric pocket, indicating that both allosteric sites may be double-occupied. Then, the authors determined the crystal structure of SHP2 complexed with both tunnel-like site binder (10) and latch-like site binders (16–18), confirmed the simultaneous binding hypothesis. Moreover, a dose-dependent decrease in SHP2 activity and a modest enhancement of IC₅₀ for 10 was observed with increasing concentrations of 18, indicating possible cooperativity between the two binding modes⁹². Furthermore, in KYSE-520 cells, 10, 17 and 18 significantly reduced DUSP6 level (a downstream marker of MAPK pathway)⁹². Meanwhile, the combination treatment of KYSE-520 cells with 18 (30 μmol/L) and 10 (0.2 μmol/L) improved DUSP6 downregulation compared to either of the single agents. The results showed that combining two different but compatible SHP2 inhibitors (dual allosteric inhibitors) improved the inhibition rate of SHP2 and may overcome drug resistance⁹².

5.3. Covalent inhibitor

The non-conserved Cys333 allosteric site on the PTP catalytic domain can also be used as a target for selective SHP2 inhibition¹⁰². The covalent binding of Cys333 with 19 could significantly inhibit the activity of wild-type SHP2 (IC₅₀ = 35 μmol/L)¹⁰² (Fig. 7). 19 showed weak but still significant time-dependent inhibition of the mutant protein SHP2^C333A103. Therefore, targeting non-conserved Cys333 with covalent inhibitors may be a meaningful new way for developing more effective allosteric inhibitors to block SHP2 activity or overcome drug resistance^36,102.

Figure 7.

(A) Structure and IC₅₀ value of 19. (B) X-ray crystallography structure of 19 with SHP2 (PDB ID: 3B7O). 19 is shown in green sticks and Cys333 is depicted as red sticks.

6. Combine use of SHP2 inhibitors to overcome drug resistance

The rapid emergence of drug resistance by tyrosine kinase inhibitors largely limits the efficacy of targeted tumor therapy. Studies have demonstrated that patients who are resistant to most kinase inhibitors have identified as point mutations in the kinase domain of the corresponding target kinase. Gene amplification, overexpression and changes in protein expression levels are the other two main mechanisms for oncogenic activation or signal pathway modification to produce drug resistance. In contrast to secondary drug resistance, primary drug resistance can be caused by multiple mechanisms that prevent or reduce kinase inhibitors and their kinase targets in the cytoplasm. The intracellular drug concentration depends on the expression of transporters that mediate the influx of kinase inhibitors into or out of plasma. In tumor cell lines, multidrug resistance is usually associated with decreased accumulation of ATP-dependent cellular drugs, which is attributed to the overexpression of certain ATP-binding cassette (ABC) transporters. In general, the overexpression of the ABC transporter protects tumor cells from kinase inhibitor inhibition, that is, the chem-immune system appears to recognize kinase-targeting drugs as xenobiotics at the membrane and tissue barriers. In the case of active efflux, it protects intracellular targets from kinase inhibitors103, 104, 105. Overexpression of drug transporters may confer more potential drug resistance. In addition, cancer stem cells (CSCs) are responsible for tumor initiation and possess hyperproliferative potential and are insensitive to periodic chemotherapy drugs, which become a major obstacle to cancer therapy. In the treatment of leukemia, kinase inhibitors can effectively target proliferating mature cells, but failed to eliminate leukemia stem cells. In breast CSCs, cells with high CD44 and low CD24 expression (CD44^highCD24^low cells) has been shown to be enriched. Knockout of SHP2 reduces the CD44^highCD24^low cell population in MCF10A-HER2/3 and SUM159 breast cancer cells, indicating SHP2 is a potential biomarker and reasonable therapeutic target for breast cancer stem cells. Also, immune checkpoint inhibitors have been used to treat advanced NSCLC, but about 80% of patients are resistant to immunosuppressants alone^106,107.

The combined therapy of SHP2 inhibitors with existing kinase targeting drugs or immune checkpoint inhibitors to improve efficacy and/or combat drug resistance can be hot area of research.

6.1. Combination of RTK inhibitors with SHP2 inhibitors

Several compelling evidences have shown that cancer cells acquire drug resistance through a series of signal pathways activated by RTK. For example, EGFR is a kind of RTK whose gene mutation (especially the secondary mutation T790M) and protein overexpression abnormally activate downstream pathways to induce drug resistance. At the same time, the simultaneous activation of redundant kinases can induce drug resistance by activating by-pass pathways¹⁰³. In this sense, inhibiting RTK activity can play a vital role in tumor treatment¹⁰⁸. As an important part of RTK signal, SHP2 is the downstream effector of many RTK activation signal cascades. Studies have shown that SHP2 is significantly up-regulated when RTK is activated to acquire adaptive resistance¹⁰⁸.

In v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutant tumor cells, the combination of RTK inhibitor and SHP2 inhibitor 10 can enhance the inhibitory effect on KRAS mutant tumor cells¹⁰⁹. This indicates that cancer with KRAS mutation depends on the upstream signal of RTK and SHP2 and provides a new direction for RTK inhibitors and SHP2 inhibitors to treat cancer with KRAS mutation^93,109. Inhibition of SHP2 by 10, a selective allosteric inhibitor, has revealed the therapeutic prospect of RTK dependent cancers. Experimental evidence suggests that inhibition of SHP2 and RTK is useful in treating various KRAS mutant cancers that depend on upstream growth factor signaling, including KRAS^G13D and KRAS^Q61H mutations^93,97,109.

Several genetic and biochemical evidences show that SHP2 is an integral part of RTK signals, including FGFR, VEGFR, PDGFR, and EGFR signals, leading to the complete activation of ERK and AKT pathways^47,110. Sorafenib is a multi-kinase inhibitor, including RTK, which explicitly targets multiple growth factor pathways to block tumor cell proliferation and resist angiogenesis. Under the treatment of sorafenib, hepatoma cells obtain adaptive drug resistance by reactivation of RAS-MEK-ERK and AKT pathways. Combined use of SHP2 inhibitor (10) with sorafenib can block the reactivation of MEK/ERK and AKT signaling pathways, thus overcoming the adaptive resistance to sorafenib (Fig. 8). The combination of 10 and sorafenib can maximize tumor growth inhibition and significantly improve the survival rate, which may be a new treatment strategy against hepatocellular carcinoma (HCC)¹⁰⁸.

Figure 8.

The combination of 10 and sorafenib blocks the reactivation of MEK/ERK and AKT signaling pathways.

6.2. Combination of MEK inhibitors with SHP2 inhibitors

RAS protein is a crucial driver of cancer, and RAS gene is easily mutated in human malignant tumors. In tumor cells, KRAS, a member of the Ras family, is the most prone to mutation¹¹¹. MEK inhibitors have been widely used to treat cancers with RAS mutations. However, mutations in MEK lead to excessive activation of MEK or prevent the inhibitor from binding to MEK, resulting in drug resistance. Most cancers that are resistant to MEK inhibitors reactivate multiple RTKs upstream of the MAPK pathway, thereby initiating a signal cascade and eventually leading to excessive cell proliferation¹¹². Therefore, the emergence of adaptive drug resistance limited future clinical use of MEK inhibitors¹¹¹. Interestingly, recent study revealed that SHP2 inhibitors can prevent adaptive drug resistance of MEK inhibitors. Therefore, the combination of MEK and SHP2 inhibitors could be a new strategy for treating RAS-driven cancer⁹⁰. In KRAS mutant lung cancer cell lines, SHP2 inhibitors alone have almost no effect on cell proliferation. In contrast, the combination of SHP2 and MEK inhibitors show a strong synergistic anti-proliferation effect¹¹³. However, inhibition of SHP2 in the KRAS mutant NSCLC in vivo can induce a senescence response, which is aggravated by MEK inhibition¹¹³. The combination of MEK inhibitors and SHP2 inhibitors also overcome the adaptive drug resistance of wild-type RAS tumor cells that are difficult to treat, such as gastric cancer, triple-negative breast cancer (TNBC) and high-grade serous ovarian cancer^111,114. In some tumors, the deletion or inhibition of SHP2 can delay tumor growth, but the effects are not enough to achieve tumor regression¹¹⁵. The combination of low-dose SHP2 inhibitor RMC-4630 and MEK inhibitor cobimetinib showed synergistic effects in xenograft model of KRAS^G12C NCI-H358 cells and prevented tumor growth⁶. In KE39 and CAT12 tumor cells, the combination of SHP2 inhibitor 10 and MEK inhibitor trametinib (GSK1120212) can significantly inhibit tumor growth and induce regression¹¹⁴. When MEK inhibitor selumetinib (AZD6244) is used in combined therapy with SHP2 inhibitors, pancreatic cancer cell lines and colon cancer cell lines show higher sensitivity to selumetinib¹¹³ (Fig. 9). In general, the combined inhibition of SHP2 and MEK activity has strong synergistic effects in KRAS mutant tumors, especially in NSCLC113, 114, 115.

Figure 9.

Structures of MEK inhibitors used in combination with SHP2 inhibitors.

6.3. Combination of ERK signal suppression and SHP2 inhibitors

ERK activation is one of the main signals of SHP2 functional gain mutation¹¹⁶. In a variety of ERK dependent tumor environments, the remission of negative feedback of RAF or MEK inhibitors promotes the upregulation of various RTK, while in turn RTK activates RAS, which leads to the rebound of ERK activity and tumor adaptive resistance to inhibitors¹¹⁷. Simultaneously manipulating ERK signaling and SHP2 activity can effectively overcome the adaptive resistance of specific ERK-dependent tumors to RAF and MEK inhibitors. In order to evaluate the efficacy of combined inhibition of ERK signal and SHP2 in vivo, the combination of dabrafenib, Trametinib and SHP2 allosteric inhibitor 10 significantly suppressed p(Y542) SHP2 and ERK signals in mice carrying RKO xenografts while showed no significant effect on body weight. It is more effective than the combination of dabrafenib and trametinib to suppress ERK signal (Fig. 10). However, the use of dabrafenib or trametinib or 10 alone had little effect on the tumor growth or ERK signal transduction of xenografts. It is further proved that ERK signal and SHP2 suppression may be effective methods for treating BRAF (V600E) colorectal tumors. The current findings establish a combination of ERK signal transduction and SHP2 inhibition, which can effectively overcome the adaptive resistance of ERK-dependent tumors to RAF and MEK inhibitors¹¹⁷.

Figure 10.

The combination of ERK signal suppression and SHP2 inhibitor overcomes the adaptive resistance of tumors to inhibitors.

6.4. Combination of ALK inhibitors with 10

Under normal circumstances, anaplastic lymphoma kinase (ALK) activates cell growth after ligand binding. However, when EML4 on DNA is fused with ALK, the ALK kinase region is abnormally activated, which has carcinogenic activity. ALK inhibitors initially restrain most NSCLC with ALK rearrangement, but SHP2 provides a parallel survival input downstream of multiple tyrosine kinases that promote resistance to ALK inhibitors. Recently, it has been found that SHP2 inhibitor 10 has little effect on cell proliferation of several tumor cells, however, when used in combination with ALK tyrosine kinase inhibitor ceritinib, it can prohibit the growth of drug resistant patient-derived cells by preventing the reactivation of RAS and ERK1/2 (Fig. 11). These findings suggest that the combined inhibition of ALK and SHP2 might provide a promising strategy for drug-resistant cancer therapy. Moreover, short-term or long-term use of 10 alone doesn't reduce the activity of RAS in any patient-derived tumor cells, while short-term combined use of ceritinib and 10 could reduce the level of GTP-RAS in all models¹¹⁸.

Figure 11.

The combination of ALK inhibitor and SHP2 inhibitor prevents the growth of resistant cells.

Similarly, treatment of ceritinib in MGH049-1A and MGH073-2B xenografts produced mild and transient responses, while MGH045-2A xenografts were completely resistant. However, the combination of SHP2 inhibitor 10 and ceritinib resulted in a deep regression of MGH049-1A and MGH073-2B xenografts and moderately inhibited the growth of MGH045-2A tumors, consistent with the significantly reduced DUSP6 mRNA level. In the MGH073-2B and MGH049-1A models, when the combination of 10 and ceritinib stopped, the tumor cells grow again, and these tumor cells are still sensitive when the drugs are used once more. The combination of 10 and ceritinib show mild toxicity at the initial stage and are alleviated during treatment. In conclusion, inhibition of ALK and SHP2 activity may provide a broad therapeutic strategy for overcoming the ALK-independent mechanism of acquired drug resistance in NSCLC patients¹¹⁸.

6.5. Combination of PD-1 blockade with SHP2 inhibitors

Two groups of costimulatory receptors are expressed on the surface of T cells: costimulatory receptors and co-inhibitory receptors. PD-1 is a co-suppressor receptor expressed on T cells¹¹⁹. It is highly expressed in tumor-infiltrating lymphocytes (TIL) and inhibits T cell activation. PD-1 deficient mice showed immunoglobulin production disorder under the background of C57BL/6, and autoimmune cardiomyopathy occurred under the background of BALB/c. This might provide a solid evidence of the vital role of PD-1 as a negative regulator of T cell activation^120,121. PD-1 is a crucial immune checkpoint in cancer immunotherapy¹²². When PD-1 binds to programmed death ligand-1 (PD-L1), T cell receptor (TCR) targeting gene and Th1 cytokines significantly inhibited and transmit inhibitory signals¹⁶. Inhibition of the interaction between PD-1 and PD-L1 can enhance T cell response and mediate preclinical antitumor activity¹²³. Although anti-PD-1/PDL-1 treatment has achieved great success, many patients with solid tumors still exhibited primary and acquired drug resistance. Under treatment of PD-1/PDL-1, tumors can form tumor microenvironment (TME) to block the anti-tumor effect of T cells. This may be due to insufficient antigen immunogenicity, disfunction of antigen presentation, irreversible T cell exhaustion, resistance to IFN-γ signaling and immunosuppression. Some patients will eventually develop resistance or relapse after PD-1/PDL-1 treatment. In the presence of PD-1/PDL-1 inhibitors, through tumor immune editing, tumor cells that escape anti-tumor immunity gradually dominate. In addition, activation of PD-1/PDL-1 independent inhibitory pathways and re-depletion of activated T cells can once again cause the loss of T cell function¹²⁴.

Hui et al.¹¹⁹ demonstrated that the co-receptor CD28 is more suitable as the target of SHP2 phosphatase dephosphorylation recruited by PD-1 in relative to TCR. Dephosphorylated and inactivated CD28 inhibits T cell function by restraining the activation of the PI3K-AKT signaling pathway and reducing the activation of transcription factors, indicating that SHP2 mediates PD-1 inhibition of T cell function by inactivating CD28 signal^15,125 (Fig. 12). Therefore, SHP2 is considered to be the key mediator of PD-1 signal inhibition^16,126,127. Although SHP2 is commonly expressed in T cells, the level of SHP-2 in TIL is significantly higher compared with peripheral blood lymphocytes (PBL). The expression level of SHP2 in TIL of head and neck squamous cell carcinoma (HNSCC) patients was positively correlated with the expression of PD-1¹⁶. Blockade of PD-1 or SHP2 is sufficient to restore strong Th1 immunity and T cell activation, thus reversing immunosuppression in tumor microenvironment¹⁶. SHP2 can bind to a variety of immunosuppressive receptors to inhibit the activation of immune cells, which explains the powerful tumor-killing effect of SHP2 inhibitors. PD-1 antagonists, as immunotherapeutic agents, are being actively explored in clinical trials and have shown clinical efficacy in several solid tumors. Sun et al.¹²⁸ reported that the combination of 10 and anti-PD-1 antibody showed a higher efficacy than monotherapy in inhibiting tumor growth. Therefore, the development of a specific SHP2 inhibitor combined with PD-1 antagonists will be a promising strategy for tumor immunotherapy in the future¹⁶.

Figure 12.

SHP2-mediated direct inhibition of the downstream PI3K-AKT signaling pathway to prevent T cell activation.

6.6. Combination of other inhibitors with SHP2 inhibitors

Preclinical studies have shown that, compared with a single drug, combination therapy is more effective, overcomes drug resistance⁶, and solves the over activation of signaling pathway caused by a single drug. The combination of SHP2 inhibitors and other inhibitors has attracted more attention. For example, in the treatment of liver cancer, multiple myeloma and chondrosarcoma, the use of SHP2 inhibitors can activate an essential cancer-promoting factor STAT3, thus SHP2 inhibitors should be used carefully in these tumors. It also suggests that close attention should be paid to the phosphorylation level of STAT3 when using SHP2 inhibitors to treat related tumors. The combination of SHP2 inhibitors and STAT3 inhibitors may be a new treatment strategy¹⁵. Besides, SHP2 inhibitor TNO-155 is combined with PD-1 antibody spartalizumab or CDK inhibitor ribociclib in the treatment of advanced solid tumors and combined with KRAS^G12C inhibitor MRTX849 in the treatment of KRAS^G12C solid tumors. TNO-155 can also be combined with BRAF inhibitor dabrafenib, ERK inhibitor LTT462, MEK inhibitor trametinib and RAF inhibitor LXH254 in the treatment of advanced/metastatic BRAF V600 colorectal cancer⁶ (Fig. 13).

Figure 13.

Structures of SHP2 inhibitor and other inhibitors.

7. Future clinical development of SHP2 inhibitor

At present, there are five SHP2 allosteric inhibitors in clinical research, as summarized in Table 1^6,80. In January 2018, JAB-3068, a small molecule oral anticancer drug independently designed and developed by Jacobio with global intellectual property rights, was officially approved by FDA to enter clinical trials, mainly for the treatment of adult advanced solid tumors. JAB-3068 is currently in phase II clinical research stage. Studies have shown that JAB-3068 alone can promote the antitumor function of CD8⁺ T cells and can also be combined with PD1/PD-L1 antibody. Therefore, JAB-3068 can be used in the treatment of non-responsive tumors with PD-1/PD-L1 antibody. In August 2019, JAB-3312, the second original antineoplastic drug independently developed by Jacobio, launched phase I clinical trials in HealthONE Clinic Oncology Center, USA. JAB-3312 can block the PD-1 pathway of T cells and the KRAS pathway of tumor cells by inhibiting SHP2, thereby having dual roles of tumor immunity and tumor targeting. JAB-3312 is used to treat solid tumors such as NSCLC, colorectal cancer, and pancreatic cancer. It can also relieve the tumor immunosuppressive microenvironment and enhance the efficacy of existing tumor immunotherapy. Novartis' TNO-155 is a practical, selective, and orally active SHP2 allosteric inhibitor (IC₅₀ = 0.011 μmol/L). TNO-155 has potential in treating RTK-dependent malignant tumors, especially advanced solid tumors, and is currently in phase I clinical trials. RLY-1971 developed by Relay Therapeutics is used for the treatment of solid tumors. A phase I dose escalation study is being conducted in patients with advanced or metastatic solid tumors to evaluate the safety and tolerability of RLY-1971. Revolution Medicines’ RMC-4630 is a practical, sensitive, oral SHP2 allosteric inhibitor, which can inhibit SHP2-dependent RAS signal mutations (such as KRAS^G12C, NF1, BRAF, KRAS amplification, etc.). The results of the phase I clinical study clarified that RMC-4630 showed reasonable tolerability and preliminary clinical activity in KRAS mutant NSCLC patients, especially KRAS^G12C mutant patients. Also, the tolerability of intermittent administration was improved compared with daily administration. For patients with other mutations in the RAS pathway and patients with disease progression after receiving KRAS^G12C inhibitors, a study of the use of RMC-4630 in combination with the MEK inhibitor cobimetinib is ongoing (NCT03989115).

Table 1.

Clinical trials of SHP2 allosteric inhibitors^6,80.

Drug	Company	Phase	Indication	NCT identifier
JAB-3068	Jacobio	Phase I/II	Advanced solid tumors	NCT03518554
				NCT03565003
JAB-3312	Jacobio	Phase I/II	Advanced solid tumors	NCT04121286
				NCT04045496
TNO-155	Novartis	Phase I/Ib	Advanced solid tumors	NCT03114319
				NCT04000529
RLY-1971	Relay Therapeutics	Phase I	Advanced or metastatic solid tumors	NCT03989115
RMC-4630	Revolution Medicines	Phase I/II	Relapsed or refractory solid tumors	NCT03634982
				NCT03989115

Preclinical studies have demonstrated that combination therapy is more effective than monotherapy and is an effective way to overcome resistance to a single drug⁶. SHP2 inhibitors combined with other kinase inhibitors are more effective than single therapy and are less likely to develop drug resistance. At the same time, development of multi-target inhibitors is also worth studying in the future. In addition, immunochemotherapy can not only inhibit the proliferation of tumor cells, but also activate the immune response of T cells to tumor cells, which is an important research direction. The current research data provides a strong theoretical basis for the clinical combination strategy of SHP2 inhibitors and drugs that directly target the immune system¹²⁹. Fodor et al.⁹² reported a rare case of dual, simultaneously occupying both tunnel-like and latch-like allosteric pockets of SHP2 protein. Studies have shown that dual inhibition prevents the emergence of resistance to each drug in preclinical animal models¹³⁰. Therefore, exploring dual allosteric inhibitors may help improve the inhibitory activity of SHP2 and overcome drug resistance caused by mutations in SHP2. The discovery of irreversible inhibitors for PTP non-conservative Cys333 also provides a new direction for the development of SHP2 allosteric inhibitors. The inhibitors screened based on Cys333 may be a new tool for PTP targeted drug discovery¹⁰². Several in vivo and in vitro studies on different target proteins have shown that the combination of allosteric and orthomorphic inhibitors can keep the protein conformation stable, thus delaying the emergence of drug-resistant mutations of target proteins⁹⁸. This suggests that the combined use of SHP2 PTP catalytic site inhibitors and allosteric inhibitors may be a helpful direction for clinical development. Future research should focus on deciphering the new molecular mechanism targeting SHP2 and accelerating the development of selective SHP2 inhibitors³².

In the past few years, the proteolysis-targeting chimeras (PROTACs) technology has become another hot spot in drug discovery. This strategy recruits the target protein to E3 ligase system and induces the degradation of targeted protein through ubiquitin proteasome system^36,131 (Fig. 14A). PROTACs strategy to achieve endogenous degradation of target proteins has been increasingly reported¹³². Most recently, the PROTACs technology has been successfully applied to the design of small molecule SHP2 degraders, opening a new field for targeting SHP2 degradation into medicines^6,133. Small molecular degrader SHP2-D26 was synthesized by linking the known SHP2 inhibitor and VHL-1 ligand and exhibited low DC₅₀ values in KYSE520 and MV4; 11 cells (Fig. 14B). Importantly, SHP2-D26 is significantly better than classical SHP2 allosteric inhibitors in inhibiting ERK phosphorylation as well as proliferation in KYSE520 and MV4; 11 cells. Thus, this study proved for the first time that targeted degradation of SHP2 is a very effective strategy to inhibit SHP2 activity¹³³.

Figure 14.

(A) The mechanism of PROTAC. (B) Structure of SHP2 degrader SHP2-D26.

8. Conclusions and outlook

In the past two decades, we have gained a great understanding of the molecular structure, functional characteristics, and signal regulation of SHP2. SHP2 gene abnormalities, including mutations (GOF and LOF) and abnormal expression (upregulated and down-regulated), are closely related to leukemia and solid tumors^36,40. The function of SHP2 protein is different in distinct environments. For example, overexpression of SHP2 is associated with breast cancer, gastric cancer and lung cancer. Knockout or inhibition of SHP2 can significantly prevent the growth of cancer cells and exert anticancer activity. However, SHP2 may has an inhibitory effect in other cancers such as liver cancer and osteosarcoma³⁶. With the progress in studying the mechanism of action, SHP2 become an important biomarker and a promising therapeutic target for tumor therapy. It is of great significance to clarify these problems for understanding the relationship between SHP2 function and related diseases⁴⁸.

Due to the high conservation and positive charged PTP catalytic domain, the development of catalytic site SHP2 inhibitors meets lot of difficulties such as low selectivity and cell permeability, which severely limits the clinical use of SHP2 catalytic site inhibitors⁸¹. This once put the research and development of SHP2 inhibitors in a dilemma. Until 2016, Novartis developed a novel allosteric inhibitor 10, which can stabilize SHP2 in the self-inhibitory conformation by acting as a “molecular glue”. Hereafter, the development of new SHP2 allosteric inhibitors has aroused great enthusiasm. With the discovery of several other allosteric pockets, a number of new allosteric inhibitors have been reported. These SHP2 allosteric inhibitors exhibited high selectivity, water solubility, cell permeability, oral availability, and in vivo activity. Currently, the clinical trials of five SHP2 allosteric inhibitors in various solid tumors are also in progress, which further promote an in-depth study of SHP2 in tumor diseases.

The development of SHP2 inhibitors have been systematically reviewed^6,36,80,81 elsewhere. In this review, we focused on the recent development of combined therapy using SHP2 inhibitors, which showed a significant advantage than single-drug therapy and have the potential to overcome drug resistance. Drug resistance is one of the major challenges in current cancer treatment. The mechanism of drug resistance determines the proliferation and metastasis of tumors and ultimately leads to the death of patients. Therefore, developing new strategies to overcome drug resistance is an important direction for future cancer treatment. Recent studies have already revealed that the signal pathways involved in the regulation of SHP2 play an important role in the drug resistance mechanism of kinase inhibitors. Therefore, the development of efficient, highly selective SHP2 inhibitor and the combination of SHP2 inhibitor with kinase inhibitors will become the main direction of SHP2 inhibitor research. Meanwhile, the development of immunochemotherapy, dual target inhibitors, dual allosteric inhibitors, covalent inhibitors, and the combined use of allosteric inhibitors and orthosteric inhibitors, PROTACs technology to degrade mutant proteins provide new ideas for overcoming drug resistance using SHP2 inhibitor. We hope this review would be helpful for the future development of SHP2 inhibitors.

Author contributions

Prof. Hao Fang and Prof. Xuben Hou provided the writing ideas and guided the revision of manuscript content. Meng Liu summarized the literature and wrote the manuscript. Shan Gao provided ideas for figures and revised the manuscript. Reham M. Elhassan edited the language of the manuscript. All authors gave approved to submit the final manuscript.

Conflicts of interest

The authors have no conflicts of interest to declare.