Holistic View of ALK TKI Resistance in ALK-Positive Anaplastic Large Cell Lymphoma

Anaplastic lymphoma kinase (ALK) is a receptor tyrosine kinase expressed at early stages of normal development and in various cancers including ALK-positive anaplastic large cell lymphoma (ALK+ ALCL), in which it is the main therapeutic target. ALK tyrosine kinase inhibitors (ALK TKIs) have greatly improved the prognosis of ALK+ALCL patients, but the emergence of drug resistance is inevitable and limits the applicability of these drugs. Although various mechanisms of resistance have been elucidated, the problem persists and there have been relatively few relevant clinical studies. This review describes research progress on ALK+ ALCL including the application and development of new therapies, especially in relation to drug resistance. We also propose potential treatment strategies based on current knowledge to inform the design of future clinical trials.


INTRODUCTION
Anaplastic large cell lymphoma (ALCL) is an aggressive cluster of differentiation (CD)30+ peripheral T-cell lymphoma (PTCL) that accounts for approximately 10%-15% of pediatric and 1%-2% of adult non-Hodgkin lymphoma (NHL) cases (1). Over 90% of children and adolescents with ALCL are ALK positive (ALK+) while the rate among adult patients is 40%-50% (2). The main feature of ALK+ ALCL is the expression of ALK fusion proteins such as nucleophosmin (NPM)-ALK, TNF receptor-associated factor (TRAF)1-ALK, 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase ATIC-ALK, ring finger protein (RNF)213-ALK, and tropomyosin (TPM)3-ALK (3)(4)(5)(6). NPM-ALK is the most prevalent of these fusions and is detected in 75%-80% of adults and 90% of children with ALK+ ALCL (7,8). NPM-ALK arises from the fusion of the ALK gene on chromosome 2p23 with the NPM gene on chromosome 5q35. ALK is normally expressed in cells of the small intestine, testis, and colon but not in lymphocytes (9). In ALK+ ALCL, NPM-ALK is highly expressed as a result of a high copy number of the NPM promoter and constitutively activates NPM-ALK and downstream signaling including signal transducer and activator of transcription (STAT)3, phospholipase (PLC)g, phosphatidylinositol 3-kinase (PI3K)-protein kinase B (AKT), and mitogen-activated protein kinase (MAPK)extracellular signal-regulated kinase (ERK) pathways that are important for cell survival and proliferation, metabolic transformation, and immune evasion via the oligomerization of NPM (10). Therefore, NPM-ALK is a target of therapeutic strategies in ALK+ ALCL. Alectinib was approved in Japan for the treatment of children and adults with relapsed/refractory (R/R) NPM-ALK+ ALCL, while crizotinib has been approved for this indication by the US Food and Drug Administration. However, a subset of patients responds poorly to ALK TKIs due to mutation/amplification of the ALK gene, which reduces sensitivity to these drugs. Tumors also survive by other mechanisms such as autophagy, anti-apoptosis, and ALK bypass substitution, among others.

MAPK Kinase (MEK)-ERK Pathway
MEK-ERK1/2 signaling stimulates cell proliferation by promoting cell cycle progression via the protein-dependent kinase (CDK)4 and retinoblastoma protein (RB) pathways. CDK4 binds to cyclin D1 to phosphorylate Rb and release the transcription factor E2F, which induces the expression of CDK4 as well as cyclin D1, leading to the entry of cells into S phase; it also promotes cell survival and proliferation via ERK1/2-mTOR signaling (42,43). ERK1/2 phosphorylates JUNB, a member of the activator protein (AP)-1 family and an important transcription factor in ALK+ ALCL. JUNB transcriptional targets have been shown to be involved in cell proliferation, anti-apoptosis, immune evasion, and immune phenotype transition (44,45) (Figure 1). , vincristine sulfate, and prednisone (CHOP) has a favorable clinical outcome, although the rate of recurrence is high. Patients who relapse after first-line chemotherapy usually have a poor prognosis, while some patients are inherently resistant to chemotherapy regimens such as CHOP. ALK TKIs can improve the prognosis of patients with R/R ALK+ ALCL. In a clinical trial of children with R/R ALK+ ALCL treated with crizotinib (NCT00939770), the objective response rate (ORR) was 90% and the complete response (CR) rate was 80% (46). In a phase 2 clinical trial of alectinib for the treatment of R/R ALK+ ALCL(Range, 6-70 years), the ORR after alectinib treatment was 80% and the CR rate was 60%, while the 1-year PFS, event-free survival (EFS), and overall survival (OS) rates were 58.3%, 70.0%, and 70.0% respectively (47). But a new problem has arisen: Among patients who have a high response to crizotinib monotherapy, about 30-40% of patients have developed further resistance to the drug (48). ALK mutation/amplification reduces sensitivity to ALK TKIs, necessitating a switch to another drug. Furthermore, tumors mitigate the cytotoxicity of ALK TKIs through a variety

Activation of Bypass Signaling Pathways
IL-10 signaling bypass leads to crizotinib resistance (59). In ALK+ ALCL, IL-10 promotes the proliferation of ALK+ ALCL and reduces the sensitivity of ALCL cells to ALK TKI. A screen of tumors from ALK+ ALCL patients who progressed within 3 months of crizotinib treatment found that IL-10RA was  In resistant ALK+ ALCL cell lines, autocrine IL-10/IL-10RA signaling activated tyrosine kinase (TYK)2/Janus kinase (JAK)1 signaling instead of the ALK-STAT3 pathway to activate STAT3, which stimulated the expression of IL-10, IL-10RA, and IL-10RB. Thus, the strength of IL-10RA signaling in ALK+ ALCL affects tumor sensitivity to crizotinib. IL-10 is a known immunosuppressive factor that directly inhibits effector T cells, promotes T cell exhaustion, and inhibits T cell activation through myeloid-derived suppressor cells and induction of regulatory T cells (60,61). This could allow ALK+ ALCL to achieve immune evasion and may result in the development of drug resistance, although there have been no studies investigating this possibility. Activation of insulin-like growth factor 1 receptor (IGF-1R) signaling is another mechanism of ALK resistance. Type 1 IGF-1R is commonly expressed in ALK+ ALCL. In a crizotinib resistance model, IGF-1R signaling was shown to be increase with drug concentration and reduced the cytotoxicity of crizotinib; conversely, crizotinib combined with IGF-1R inhibitors partly restored the sensitivity of resistant cells (62). ALK phosphorylates IGF-1R at the C-terminal Y338 residue; in turn, IGF-1R increases the phosphorylation of ALK and its downstream effectors and promotes the survival of ALK+ ALCL cells by increasing the expression of anti-apoptotic proteins (myeloid cell leukemia [Mcl]-1, Bcl-2, Bcl-xl, etc) (63). The combined use of IGF-1R inhibitors and ALK TKIs can suppress ALK and downstream signaling, thereby enhancing the sensitivity of ALK+ ALCL to ALK TKIs and reducing crizotinib resistance caused by activation of the IGF-1R pathway, making combination therapy a potential treatment regimen. (64).
PDGFR expression and activation is a key driver of ALCL proliferation, survival and spread. NPM-ALK promoted the expression of PDGFRB through NPM-ALK/AP-1 (JUN/JUNB) (65). High expression of PDGFRB could also be seen in most ALK+ALCL. PDGFR promoted cell survival via JAK1/STAT3 and AKT(PKB)/mTOR pathways (66,67). The level of PDGFR-mRNA decreased after application of PDGFR inhibitors, indicating that PDGFR is promoting its own expression, ALK-TKI in combination with PDGFR inhibitors reduces lymphoma growth and decreases relapse rates, and is a potential treatment option for patients with drug-resistant lymphoma. (65).
Activation of the MEK pathway also contributes to ALK TKI resistance, mainly through deficiency/low expression of Wiskott-Aldrich syndrome protein (WASP) (68), a tumor suppressor whose loss leads to tumor development and invasion (69). NPM-ALK negatively regulates the expression of WASP by directly phosphorylating Y102, leading to proteasome-mediated protein degradation and preventing protein binding to WASP-interacting protein (WIP) (70). NPM-ALK was also shown to repress WASP protein expression via STAT3-CCAAT/enhancer-binding protein (C/EBP)-b signaling. Downregulation of WASP was shown to enhance MEK pathway activation in ALK+ALCL. NPM-ALK can form a complex with the guanine exchange factor VAV1 to enhance the activity of cell division cycle (CDC)42, which promotes the progression of ALK+ ALCL; meanwhile, WASP inhibits the binding of CDC42 to GTP (71). Protein tyrosine phosphatase (PTPN)1/2 expression was found to be downregulated in ALK+ ALCL, leading to the overactivation of MEK, SHP2, and Janus kinase (JAK)/STAT and resistance to ALK TKIs (72).
Although ALK inhibition undermines the survival and proliferation of ALCL cells, not all cellular changes caused by ALK expression are reversed. For example, IL-10R, WASP, and MEK signaling was not downregulated by application of crizotinib; instead, tumor cells developed resistance to the drug through autocrine IL-10 secretion by ALK+ ALCL cells and activation of IL-10R signaling, which replaced the role of ALK in promoting IL-10R expression, suggesting that cytologic changes caused by ALK led to the emergence of ALK-independent mechanisms of tumor cell survival (59,68). Some oncogenes that were downregulated via epigenetic modifications were not upregulated by crizotinib, but were instead involved in crizotinib resistance, suggesting that the decrease in their expression was not reversed (68, 72) ( Figure 2).

Autophagy and Apoptosis
Autophagy activation and inhibition play different roles according to the tumor type. In ALK+ ALCL, inhibition of autophagy facilitated tumor survival in the presence of ALK TKIs (73). Crizotinib was shown to decrease autophagic flux in ALK+ ALCL, which was important for tumor survival of NPM-ALK+ ALCL; this cytoprotective response reduced the cytotoxicity of crizotinib (56,74). Rapamycin, an mTOR inhibitor that activates autophagy, reduces the survival of ALCL cells (75). The expression level of miR-7-5P was found to be downregulated in ALK+ ALCL by the application of crizotinib; it was later shown that miR-7-5P directly targets the 3′ untranslated region of RAF1 transcript, thereby negatively regulating RAF1 expression and reducing the inhibitory phosphorylation of Unc-51-like autophagy-activating kinase (ULK)1 (S757) to promote autophagy. Conversely, downregulation of miR-7-5P following crizotinib application inhibited autophagy. Potentiating the effect of miR-7-5P using an miRNA mimic enhanced crizotinibinduced autophagic flux and cytotoxicity (76). Interestingly, in ALK+ ALCL stem cells, crizotinib more potently induced autophagy and enhanced tumor cell resistance to the drug, an effect mediated by MYC (77).
Elevated levels of Bcl-2 have been observed upon treatment with crizotinib. Bcl2 is anti-apoptotic protein that inhibits autophagy, which can lead to drug resistance in tumors. Inhibition of Bcl-2 significantly increased crizotinib-induced autophagy (78,79).
Decreased autophagic flux in ALK+ ALCL plays a role in drug resistance, and tumor cells are regulated in various ways that lead to inhibition of autophagy and crizotinib resistance including regulation of miRNAs and Bcl-2 expression. Autophagy inhibition can reverse the crizotinib-induced decrease in cell viability, thus limiting the cytotoxicity of crizotinib ( Figure 2).

ALK Sequencing Analysis
For patients with ALK TKI resistance, tissue biopsy and pathologic examination can reveal the mechanism of resistance and inform clinical decisions, especially drug selection. For example, ALK TKIs can be selected that target specific ALK mutations. On the other hand, the absence of ALK mutation can indicate the development of resistance by the tumor, requiring ALK TKI combination therapy. For highly resistant (especially compound) mutations, ALK amplification may not be detected or the cause of resistance may not be clear, in which case switching ALK TKIs or using combination therapy may not be effective and therapeutic targets other than ALK should be considered. The development of ALK sequencing has aided the precision treatment of ALK+ ALCL and has revealed novel drug resistance mutations; the elucidation of ALK TKI resistance mechanisms will facilitate the development of new treatment strategies.

Intermittent Treatment
For drug resistance caused by NPM-ALK overexpression, crizotinib discontinuation can lead to overactivation of ALK oncogenic signaling and enhanced mitochondrial activity, reactive oxygen species (ROS) production, and activation of the MEK-ERK1/2 pathway, causing DNA damage. This may result in apoptosis of ALK TKI-resistant cells and the resensitization of tumor cells to crizotinib (56,57,(80)(81)(82)(83). NPM-ALK overexpression promotes STAT1 phosphorylation, and high levels of phosphorylated (p)STAT1 antagonize STAT3 and activate tumor suppressor genes, promoting cell death. Enhanced NPM-ALK expression following withdrawal of ALK TKI leads to the upregulation of pSTAT1, which antagonizes STAT3 and induces apoptosis (84,85). Given the damaging effects of excessive tumorigenic signals on DNA, agents that Enhance DNA damage and inhibit DNA repair can be used during drug withdrawal to enhance the effects NPM-ALK ( Figure 3).

Novel ALK TKIs
Second-generation ALK TKIs such as alactinib, ceritinib, brigatinib, lorlatinib, and ZX-29 have more potent activity in the central nervous system than crizotinib and can overcome the effects of most crizotinib resistance mutations (86) ( Table 1). Alectinib is an orally administered drug with greater potency than crizotinib that was shown to be effective in crizotinibresistant tumors (100). including ALK+ non-small cell lung cancer (NSCLC), with neutropenia and elevated levels of creatine kinase as the most serious adverse effects (101,102). Ceritinib is a small molecule ALK TKI that has demonstrated efficacy in ALK-positive tumors including ALCL, inflammatory myofibroblastic tumor, neuroblastoma, and rhabdomyosarcoma; in ALK+ ALCL, the most commonly reported complication was elevated transaminases (103). Brigatinib is an ALK/ROS1 inhibitor with higher selectivity than alectinib or ceritinib that can overcome most crozotinib resistance mutations including G1202R, with an ORR of 74%, median PFS of 14.5 months, and 1-year OS rate of 83% in crizotinib-resistant patients (96,104). Brigatinib has shown superior efficacy to crizotinib in the firstline setting, and an ongoing study is investigating brigatinib in ALCL (105). However, second-generation ALK TKIs are more likely to lead to the development of resistance than crizotinib (54). The third-generation ALK TKI loratinib overcomes nearly all single resistance mutations to second-generation inhibitors and is more potent than brigatinib against the G1202R mutation. However, G1202R was shown to reduce the sensitivity of tumor cells to brigatinib and loratinib, and tumor cells harboring double mutations (D1203N+E1210K and F1174C+D1203N) had lower sensitivity to loratinib, whereas none of the secondgeneration ALK TKIs were effective (54). ZX-29 is a novel ALK TKI that induces apoptosis by stimulating the production of ROS, overcoming the drug resistance conferred by the ALK G1202R mutation and exhibiting greater cytotoxicity than ceritinib (106,107). The Fms-related receptor tyrosine kinase (FLT)3/AXL inhibitor gilteritinib is currently mainly used for R/ R acute myeloid leukemia (108). Gilteritinib inhibits not only wild-type and mutant ALK but also overcomes highly resistant double mutations (I1171N+F1174I, I1171N+L1198H, and others containing I1171N) as well as single mutations by forming hydrogen bonds with the ALK E1197, M1199, and E1210 residues (99,109). Mutations such as L1196M, I1171T/N/S, G1269A, G1202R/del, G1202N, and S1206Y/C that sterically hinder the binding of ALK TKIs to ALK are a common resistance mechanism (47,(49)(50)(51)(52)(53)(54). Gilteritinib inserts into the ATPbinding site of ALK and therefore has an advantage over the first 3 generations of ALK TKI in terms of drug resistance, providing an additional treatment option for patients.  Gilteritinib has also demonstrated efficacy against ALK+ ALCL in preclinical studies (110). ALK kinase structural domain mutations common to NSCLC and ALCL have been identified. Table 1 summarizes mutations known to confer resistance/sensitivity to ALK TKIs that have potential application in the treatment of ALK+ ALCL ( Table 1).

Proteolysis-Targeting Chimera (PROTAC)
PROTACs are used to promote proteasome-mediated protein degradation (111). PROTAC is composed of 2 ligands that connect the target protein and E3 ubiquitin ligase, forming a ternary complex. The ubiquitin-binding enzyme E2 binds to the E2 binding site on E3 ligase and transfers ubiquitin to the target protein, leading to its degradation (111)(112)(113). ALK TKIs (alectinib, brigatinib, ceritinib, etc.) are often used as PROTAC ligands to promote the degradation of ALK protein and thereby inhibit tumor growth driven by ALK (114). This allows a small dose of PROTAC to achieve a strong inhibitory effect, which is especially advantageous for overcoming drug resistance due to ALK amplification. PROTAC directly targets and degrades proteins and is unaffected by ALK point mutations; this explains how PROTAC designed with alectinib/brigatinib can degrade the ALK G1202R mutant, which is resistant to the drugs themselves. PROTAC designed with alectinib as the ligand was more effective than alectinib in ALK+ ALCL patients and enhanced ALK degradation (115)(116)(117). ARV-110, the first PROTAC drug targeting the androgen receptor, has been used in patients with metastatic debulking-resistant prostate cancer with promising results (118). A second PROTAC drug, ARV-471 (targeting the estrogen receptor [ER]), is being evaluated in clinical trials for the treatment of patients with locally advanced or metastatic ER+/human epidermal growth factor receptor (HER)2− breast cancer. Interim results have shown that ARV-471 reduced ER expression levels by 62% and up to 90% and was effective against both wild-type and mutant ER (119). PROTACs targeting ALK are not yet in clinical use but several ALK-RROTACs are being developed such as SIAIS001, SIAIS117, and SIAIS164018 that can potentially overcome resistance to ALK TKIs (115, 116, 120) ( Figure 4A).

Targets for Combination Therapy
ALK-independent drug resistance is normally treated by a combination of drugs that inhibit ALK along with other pathways involved in cell survival, thereby enhancing the cytotoxicity of ALK TKIs. The use of ALK TKI combinations can delay the development of ALK resistance. Crizotinib combined with everolimus, CHOP, decitabine, and trametinib prolonged the emergence of resistance, possibly because of a reduction in the dose of individual drugs and the effects of dual targeting (48).
Elevated expression of IL-10RA not only confers resistance to crizotinib but also to second-or third-generation inhibitors. Accordingly, ALK TKI combined with IL-10 pathway inhibitors (e.g., STAT3/pan-JAK/TYK2 inhibitors) may be effective (48) in patients who are resistant to ALK TKIs and have high IL-10RA expression. TYK2 acts upstream of STAT3; it was reported that the survival of ALK+ ALCL cells depended on TYK2/STAT1/ MCL1. Regardless of the presence of ALK fusion protein, TYK2 inhibitors can induce tumor cell apoptosis. Inhibiting TY2 blocks the activation of STAT3 by IL-10 and other pathways via T2, suppressing tumor cell survival via a bypass mechanism (121). The combination of IGF-1R and NPM-ALK inhibited tumor growth in a mouse model of ALCL cell lymphoma. Inhibition of IGF-1R promoted cell apoptosis and blocked phosphorylation of NPM-ALK and its downstream effectors (122). Thus, IGF-1R  inhibition combined with crizotinib can have a synergistic effect, allowing dose reduction of crizotinib and delaying the emergence of drug resistance (123). The expression level of platelet-derived growth factor receptor (PDGFR)B in ALK+ ALCL was found to be positively correlated with the cytotoxicity of PDGFR inhibitors, which may be effective in the treatment of PDGFRB+ ALK+ ALCL. Findings from in vitro experiments and mouse models have demonstrated that imatinib treatment suppresses the proliferation of tumor cells and promotes apoptosis. Notably, a rapid and durable antitumor response was achieved with imatinib in a patient with advanced refractory ALCL (65). Activation of the MEK pathway was observed in WASP− ALK+ ALCL and was associated with reduced efficacy of ALK TKIs. The combination of the MEK inhibitor trametinib with crizotinib has achieved better clinical outcomes than crizotinib by delaying the emergence of drug resistance (65); thus, targeting MEK can reduce the survival of lymphoma with low or absent WASP expression via the MAPK pathway (68). Additionally, g-secretase inhibitors suppressed the proliferation of ALK+ ALCL cells with crizotinib resistance, suggesting NOTCH1 as a therapeutic target in crizotinibresistant ALK+ ALCL (124,125). Crizotinib combined with everolimus more potently induced cell cycle arrest, DNA damage, and apoptosis than monotherapy in the Karpas299 transplantation model (126,127). Although everolimus can inhibit mTOR, it can cause the activation of AKT and RAS-ERK, an effect that is blocked by the addition of crizotinib. The combination treatment also prevented the occurrence of selective drug resistance after long-term use of monotherapy (126). In neuroblastoma, crizotinib combined with mTOR inhibitor was shown to overcome crizotinib resistance and promote tumor cell apoptosis (128). In ALK+ ALCL cells, the mTOR inhibitor rapamycin combined with crizotinib increased autophagic flux and promoted cell death (75) (Figure 2).
Given the reversibility and importance of epigenetic regulation of gene expression in the development of ALK+ ALCL, drugs targeting epigenetic modifications and allowing "re-expression of tumor suppressor proteins" or "reduced expression of oncogenic proteins", such as DNA methylation, HDAC and SIN3A inhibitors, are a promising therapeutic approach. (20,22,23,30,31,33). In a multicenter clinical study of cidapenem in R/R PTCL, patients with ALK+ ALCL treated with cidapenem had a higher ORR (66.67%) and disease control rate (83.33%) and better prognosis compared to those with other PTCL subtypes (129). The use of decitabine (a DNA methyltransferase inhibitor) not only enhanced the efficacy of ALK TKIs but also prolonged the time to emergence of drug resistance by more than 3 fold compared to monotherapy (48) ( Figure 4B).
ALK promotes cell cycle progression through a variety of mechanisms (eg, activation of MEK/ERK and STAT3 signaling, induction of cell cycle-related gene expression, and downregulation of P27 and P53) (12,42,130,131). NPM-ALK also promotes cell cycle progression via regulation of miRNAs. In ALK+ ALCL, the combination therapy of cell cycle inhibitors has shown the potential of resistance to ALK (132). In a mouse xenograft model of neuroblastoma with ALK F1174L and F1245C mutations, the CDK4/6 inhibitor ribociclib blocks the binding of CDK4/6 to CyclinD1, thereby inhibiting the operation of the cell cycle. In combination with Ceritinib exerts synergistic cytotoxicity, inhibits tumor growth, enhances cycle arrest, and promotes cell death (133) (Figure 4C).

Targets Other Than ALK
ALK+ ALCL expresses CD30, PD-L1, and B7-H3, which are potential therapeutic targets for promoting tumor cell death through pathways independent of ALK, thereby overcoming ALK TKI resistance.

CD30
ALK promotes CD30 expression through the MEK-ERK-AP1-JUNB pathway (134,135). Brentuximab vedotin (BV) is an antibody-drug conjugate (ADC) that targets CD30 in which anti-CD30 antibody is linked to the microtubule destroyer monomethyl auristatin E (MMAE) via a protease-cleavable linker. After binding to CD30, BV forms phagosomes through receptor-mediated endocytosis before it is hydrolyzed by lysosomal proteases; this releases MMAE, leading to cell cycle arrest and cell apoptosis (136)(137)(138)(139). At the 5-year follow-up of a clinical trial of BV in R/R ALCL, 66% of patients continued to show a CR (140). In another study of BV in the treatment of R/R ALCL, 86% of patients achieved ORR including 57% with CR and 29% with partial remission; the duration of response was 12.6 months and the duration of CR was 13.2 months (141). In a clinical trial of BV combined with chemotherapy in the treatment of pediatric patients with first-onset ALK+ ALCL (NCT01979536), the 2-year EFS rate was 79.1% and 2-year OS rate was 97.0% (142). Only one patient (1.5%) relapsed during treatment. These results demonstrate that BV is superior to conventional chemotherapy in preventing recurrence.
In a clinical trial of chimeric antigen receptor (CAR) T-cell (CAR-T) therapy targeting CD30 in ALCL patients (NCT01316146), the duration of CR was up to 9 months after 4 infusions. Importantly, CD30 CAR-T was still detectable after 6 weeks of treatment, indicating that CAR-T had sustained antitumor effects. ADCs targeting CD30 are associated with a number of adverse effects including gastrointestinal reactions that reduce the tolerability of these drugs and can lead to treatment discontinuation. CD30 CAR-T therapy circumvents this problem and under standardized care and execution, is a safe and effective treatment for CD30+ lymphoma (143).

HSP90
HSP90 is highly expressed in tumors including ALK+ ALCL. HSP90 inhibitors not only block the binding of HSP90 to ATP but also promote proteasome-mediated degradation of HSP90 target proteins (144). Inhibition of HSP90 resulted in the downregulation of NPM-ALK. The combined use of the HSP90 inhibitor onalespib and ALK TKI in ALK+ NSCLC delayed the emergence of ALK resistance and preserved sensitivity to onalespib (145). A dual-target inhibitor designed based on the active structure of HSP90 and ALK TKIs (resorcinol and 2,4-diaminopyrimidine motifs) has been proposed that is expected to overcome ALK TKI resistance (146) (Figure 4D).

Immune Checkpoint
In ALK+ ALCL, whether NPM-ALK promotes PD-L1 expression through downstream signals or in view of the role of PD-1/PD-L1 in TME, PD1/PD-L1 is expected to become a potential therapeutic target for ALK+ ALCL. Recently, studies have found that the expression of PD-L1 and the number of tumor-infiltrating T cells are related to the prognosis of ALK+ ALCL (147). Geptanolimab (GB226) is a PD-1 monoclonal antibody. In an open study (NCT03502629), it was found that the higher the level of PD-L1 in R/R PTCL, the effect of PD-1 blockers is better. In R/R PTCL, the ORR was 40.4% and the 12month Duration of Response (DOR) was 48.5%. In PTCL with PD-L1 expression>50%, the ORR (53.3%) and median PFS (6.2 months) were higher, especially in ENKTL, ALK -ALCL, ALK + ALCL, Geptanolimab has a better curative effect (148). Similar to this, in an ALK-resistant ALK+ ALCL patient whose tumor tissue highly expresses PD-L1, the tumor tissue completely disappeared after 5 months of Navumab treatment and the complete remission was maintained for up to 18 months (149). In addition, when B7-H3 CAR-T is used to treat ALK+ ALCL, it shows strong cytokine secretion and cytotoxicity both in in vivo and in vitro experiments. It has obvious proliferative activity and memory phenotype after receiving B7-H3 stimulation. Under the premise of strict control of CRS, B7-H3 CAR-T is expected to become another important therapeutic target for ALK+ ALCL after ALK and CD30 (150).

DISCUSSION
ALK TKI resistance is a major challenge in the treatment of ALK+ ALCL; therefore, therapeutic strategies to overcome this resistance are a key research direction for this malignancy. Many aspects of the resistance mechanisms remain to be elucidated including the upregulation of Bcl-2, ALK amplification after the application of crizotinib, bypass signaling, autophagy, and apoptosis. There have been few studies of single drugs/combination therapies that can improve/reverse ALK resistance, which may be related to the low prevalence of ALK+ ALCL and do not provide sufficient impetus for clinical trials. There are also considerable disparities in the treatment of ALK TKI resistance, especially in terms of therapeutic options for patients. Strategies that target resistance mutations are expected to greatly improve the clinical outcome of ALK+ ALCL.
Selection of appropriate ALK TKIs by sequencing ALK mutation sites can alleviate ALK TKI resistance, although highly resistant mutants and multiple mutations are problematic. The mechanisms by which mutations lead to ALK resistance have been systematically investigated in studies of ALK protein structure and binding to TKI and ATP, and they have served as the basis for the evaluation of candidate drugs such as ZX-29 and ginitinib. The latter in particular-whose binding to ALK is largely unaffected by ALK mutations-can overcome the effects of most ALK single mutations as well as double mutations conferring loratanib resistance (99,106,107).
ALK amplification also underlies resistance to ALK TKIs (57) although it may not benefit tumor cells as it can trigger oncogenic stress and induce DNA damage especially upon discontinuation of an ALK TKI, resulting in tumor cell apoptosis and restoration of sensitivity to the inhibitor (85). PROTAC technology is a new treatment strategy that targets ALK amplification and is effective against drug resistance mutations as it induces mutant ALK proteins to undergo ubiquitin-mediated degradation (114).
ALK-independent drug resistance usually necessitates treatment with a combination of drugs, and many studies have shown that although ALK inhibition decreases tumor cell survival and proliferation, many cellular changes caused by ALK overexpression are not reversed such as IL-10RA and IGF-1R dysregulation (59,(62)(63)(64)68). When the inhibition of ALK fusion proteins is alleviated, these changes allow tumor cells to survive and proliferate, leading to the development of ALK TKI resistance. Pro-survival pathways other than ALK can be blocked by a combination of drugs that synergistically enhance the cytotoxicity of ALK TKIs. Additionally, drug combinations can also delay the emergence of resistance through dose reduction of single agents and dual targeting. Although there have been only a few clinical studies examining ALK TKI combinations in ALK+ ALCL, clinical trials are currently underway ( Table 2). ALK TKIs also cause adaptive changes in tumor cells that favor their survival such as the upregulation of Bcl-2 and inhibition of autophagy; therapeutic strategies that target these changes may be effective in the treatment of ALK+ ALCL (56,77,79).
In addition to ALK fusion proteins and their associated signaling pathways, activation of ALK and its downstream effectors in ALK+ ALCL results in the expression of CD30, PD-L1, and B7-H3, which are potential drug targets in patients who are highly resistant to ALK TKIs. However, it is unclear whether such targeted therapies can restore tumor cell sensitivity to the inhibitors.

CONCLUSION
In this review, we summarized research progress on ALK resistance to provide a reference for the design of clinical studies and development of new drugs for the treatment of ALK+ ALCL. ALK is an important therapeutic target in ALK+ ALCL. ALK TKIs have broadened the therapeutic options for ALK+ALCL patients who are resistant to or relapse on chemotherapy, but the emergence of drug resistance is an outstanding problem. Considerable progress has been made in the elucidation of ALK resistance mechanisms including those associated with and independent of ALK, and novel TKIs are being developed that can bring lasting remission to ALK TKIresistant patients.

ACKNOWLEDGMENTS
We sincerely appreciate all members. We thank Charlesworth Author Services (www.cwauthors.com) for its linguistic assistance during the preparation of this manuscript.