Evidence for cytoprotective autophagy in response to HER2-targeted monoclonal antibodies

Abstract

The advent of HER2-targeted monoclonal antibodies such as trastuzumab has significantly improved the clinical outcomes for patients with breast cancer overexpressing HER2 and, more recently, also for gastric cancers. However, the development of resistance, as is frequently the case for other antineoplastic modalities, constrains their clinical efficacy. Multiple molecular mechanisms and signaling pathways have been investigated for their potential involvement in the development of resistance to HER2-targeted therapies, among which is autophagy. Autophagy is an inherent cellular mechanism whereby cytoplasmic components are selectively degraded to maintain cellular homeostasis via the generation of energy and metabolic intermediates. Although the cytoprotective form of autophagy is thought to predominate, other forms of autophagy have also been identified in response to chemotherapeutic agents in various tumor models; these include cytotoxic, cytostatic, and nonprotective functional forms of autophagy. In this review, we provide an overview of the autophagic machinery induced in response to HER2-targeted monoclonal antibodies, with a focus on trastuzumab and trastuzumab-emtansine, in an effort to determine whether autophagy targeting or modulation could be translated clinically to increase their effectiveness and/or overcome the development of resistance.

Significance Statement

This manuscript is one in a series of papers that interrogate the role(s) of the autophagy induced in response to antineoplastic agents in various cancer models. This series of papers was developed in an effort to establish whether autophagy targeting or modulation is likely to be an effective adjuvant strategy to increase the efficacy of cancer chemotherapeutic agents. This review explores the relationship between the autophagic machinery and HER2-targeted therapies.

1. Introduction

This manuscript follows a series of papers published by our research group in an effort to evaluate the role(s) of autophagy in response to various antineoplastic modalities. Our previous publications investigated the nature of autophagy in tumor cells in response to radiation (Patel et al, 2020), cisplatin (Xu and Gewirtz, 2022), microtubule poisons (Xu et al, 2022), hormonal therapies in estrogen-positive breast cancer (Finnegan et al, 2022), PARP inhibitors (Elshazly et al, 2022), topoisomerase I poisons (Elshazly et al, 2023b), temozolomide (Elshazly and Gewirtz, 2023c), BET family inhibitors (Elshazly and Gewirtz, 2023b), BRAF-targeted therapies (Elshazly and Gewirtz, 2023a), and, most recently, androgen-targeted therapies (Elshazly and Gewirtz, 2023d). This review addresses the role of the autophagic machinery in response to HER2-targeted monoclonal antibodies; other HER2-targeted therapies, such as HER2 tyrosine kinase inhibitors, will be discussed in a separate article. The primary goal of this series of papers is to determine whether there are particular therapeutic modalities in which the preclinical data and, where available, clinical trial results support the inclusion of autophagy inhibition or modulation as an adjuvant approach.

2. HER2 and autophagy

HER2 (ERBB2) is a transmembrane tyrosine kinase receptor belonging to the ERBB family, which comprises epidermal growth factor receptor 1 (EGFR) (ERBB1/HER1), HER3 (ERBB3), and HER4 (ERBB4) (Garrett and Arteaga, 2011; Elshazly and Gewirtz, 2022). The HER2 receptor consists of 3 primary components: the N-terminal domain and α-helix domain as well as an intracellular domain with tyrosine kinase activity (Garrett and Arteaga, 2011; Elshazly and Gewirtz, 2022). HER2 does not bind directly to any ligand (Citri and Yarden, 2006), whereas the homo- and heterodimeric interactions between the ERBB receptors induce autophosphorylation of the intracellular tyrosine kinase domain (Hsu and Hung, 2016). These phosphorylated residues act as docking sites for various adapter and scaffolding proteins, including Grb2, guanine exchange factor, and SOS, triggering a variety of downstream signaling pathways, including PI3K/AKT, Ras/MEK/ERK, JAK/STAT, and PLCγ/PKC, that regulate cell survival, proliferation, and differentiation (Hsu and Hung, 2016). Despite the crucial role of HER2 in cellular development, HER2 dysregulation, particularly overexpression, may lead to increased signaling pathways activity, particularly the MAPK and PI3K/AKT pathways, ultimately contributing to tumor development (Wong and Hurvitz, 2014). Various HER2-targeted therapies have been developed, including monoclonal antibodies, which are the scope of this review, as well as tyrosine kinase inhibitors. HER2 monoclonal antibodies include trastuzumab and pertuzumab as well as margetuximab (Wong and Hurvitz, 2014; Schlam et al, 2022). Although these agents have markedly improved the prognosis of breast cancer patients bearing tumors when HER2 is overexpressed, resistance development, as the case with other chemotherapeutic agents, often limits their efficacy. Of note, 70% of HER2-overexpressing breast cancer patients either are intrinsically resistant to trastuzumab or develop trastuzumab resistance within 1 year of initiation of treatment (Yang and Klionsky, 2009; Wong and Hurvitz, 2014; Luque-Cabal et al, 2016; Elshazly and Gewirtz, 2022; Schlam et al, 2022). Various molecular mechanisms have been investigated in efforts to restore sensitivity to HER2-targeted therapies (Luque-Cabal et al, 2016; Elshazly and Gewirtz, 2022), among which is that of autophagy.

Autophagy is an intrinsic cellular machinery for recycling damaged and misfolded proteins and misfunctional cellular organelles, such as endoplasmic reticulum (ER) and mitochondria, to maintain cellular homeostasis, as well as generate energy and metabolic intermediates necessary for cell survival. Autophagy is induced in response to various stimuli, including starvation, hypoxia, ER stress, damaged organelles, and DNA damage as well as, critically, chemotherapeutic agents (Yang and Klionsky, 2009; Feng et al, 2014; Galati et al, 2019; Xu et al, 2022). The autophagic machinery proceeds through a number of now well established steps, beginning with the generation of the phagophore, followed by formation of the autophagosomes, fusion between autophagosomes and lysosomes, and generation of autolysosomes in which the cellular cargo is degraded (Yang and Klionsky, 2009; Feng et al, 2014; Xu et al, 2022; Elshazly and Gewirtz, 2023d). The detailed mechanism for autophagic machinery has been discussed in detail in our previous publication (Finnegan et al, 2022).

The scientific literature describing autophagy in response to chemotherapeutic modalities as well as radiation generally focuses on the cytoprotective functional form of autophagy. Cytoprotective autophagy is implicated in tumor cell survival and, in certain cases, contributes to resistance development, where its inhibition, pharmacologically and/or genetically, results in sensitization of the tumor cell to the antineoplastic agents. As an example, resistance to tamoxifen has been associated with the induction of autophagy (Liu et al, 2019). The cytoprotective form has been investigated in multiple clinical trials in an attempt to increase the efficacy of distinct chemotherapeutic modalities by using pharmacological autophagy inhibitors such as hydroxychloroquine (HCQ) (Elshazly and Gewirtz, 2023a,c). However, three other functional forms of autophagy have been identified, specifically cytotoxic, cytostatic, and nonprotective autophagy. In the case of cytotoxic autophagy, autophagy induction contributes to cell death and may be associated with apoptosis (Sharma et al, 2014b); conversely, its inhibition results in improved cell survival: an example is the autophagy induced by chlorpromazine in glioblastoma cells (Gewirtz, 2014; Matteoni et al, 2021; Xiao et al, 2021; Xu et al, 2022). The cytostatic form, as the name indicates, is associated with cellular growth arrest and potentially with senescence, although this linkage is not firmly established (Goehe et al, 2012; Sharma et al, 2014a; Elshazly and Gewirtz, 2023b; Elshazly et al, 2023a). As an example, cytostatic autophagy is induced by vitamin D and the vitamin D analog EB 1089 in non–small cell lung cancer cells when combined with radiation (Sharma et al, 2014a). The nonprotective form of autophagy is not implicated in either cell survival or cell death as its inhibition neither results in sensitization nor loss of drug sensitivity. As an example, in the study by Finnegan et al (2023), the combination of fulvestrant plus palbociclib induced autophagy in ER⁺ breast cancer, but autophagy inhibition did not result in either significant sensitization to the drug treatment or a reduction in cell sensitivity; this form of autophagy is less understood than the cytoprotective and cytotoxic forms since it is unclear to what purpose the tumor cell would engage this mechanism. Although these functions of autophagy are quite distinct, currently the only approach for distinguishing one functional form from the other(s) is to evaluate the impact on cell/drug sensitivity upon pharmacologic or genetic inhibition, since no other distinguishing biologic or molecular characteristics have been conclusively identified.

Mechanistically, HER2 is clearly associated with autophagy induction, as one of the signaling pathways activated by HER2 is mTOR (Janser et al, 2019), which is a negative regulator of autophagic flux (Jung et al, 2010). Furthermore, several studies have reported that autophagy is associated with HER2 resistance and tumor progression (Vazquez-Martin et al, 2009; Zambrano and Yeh, 2016; Hao et al, 2021). Therefore, in an effort to determine whether autophagy targeting or modulation could be an effective strategy to increase the therapeutic benefits of HER2-targeted therapies, we analyzed the nature of the autophagy induced in response to HER2-targeted monoclonal antibodies (Fig. 1).

Fig 1.

Autophagy induced in response to HER2-targeted monoclonal antibodies. Potential outcomes for the autophagy induced by trastuzumab alone and trastuzumab conjugated with emtansine in breast cancer and gastric cancer.

3. Trastuzumab and autophagy

Trastuzumab, a recombinant HER2-targeting monoclonal antibody, was the first biologic drug approved for the adjunctive treatment of HER2-overexpressing breast cancer (Maximiano et al, 2016). Although other available anti-HER2 targeting agents include pertuzumab and lapatinib, trastuzumab remains the gold standard for treatment of this disease subtype (Maximiano et al, 2016). In addition to its utilization for the treatment of patients with HER2-positive early-stage and metastatic breast cancer (Gao et al, 2021), trastuzumab is also used in the treatment of HER2-overexpressing gastric carcinoma (Bang et al, 2010). Trastuzumab targets the HER2 extracellular domain with high affinity. De et al (2013) reported that trastuzumab acts at 3 primary molecular levels, specifically the inhibition of HER2-HER3 hetero-dimerization (Way and Lin, 2005) and prevention of the proteolytic cleavage of the HER2 extracellular domain and the subsequent formation of the active p95^HER2 fragment (Parra-Palau et al, 2014) as well as promoting antibody-dependent cellular cytotoxicity by engaging with Fc receptors on immune effector cells and phagocytosis toward HER2-positive tumors (Petricevic et al, 2013). Furthermore, trastuzumab has been shown to suppress signaling pathways involved in cellular division, including PI3K/AKT and MAPK pathways (De et al, 2013), thereby suppressing cell cycle progression and ultimately halting cell growth and proliferation (Vu and Claret, 2012). Trastuzumab has also been incorporated into an antibody-drug conjugate form, whereby the antibody-drug conjugate is selectively and specifically delivered to the target tumor cells. Composed of a cytotoxic component as well as a linker region, this complex allows for the delivery followed by the internalization of the drug-antibody conjugate and the subsequent release of a highly active cytotoxic agent within cancer cells, ultimately causing cell death (Barok et al, 2014). The first drug-antibody conjugate approved to be used in HER2-overexpressing breast cancer is trastuzumab-emtansine, in which trastuzumab is conjugated with emtansine, a potent microtubule inhibitory and cytotoxic derivative of maytansine (Dhillon, 2014). Another drug-antibody conjugate approved for HER2-positive breast cancer is trastuzumab-deruxtecan, in which trastuzumab is conjugated with deruxtecan, a topoisomerase I inhibitor (Barok et al, 2014; Modi et al, 2020). These agents have improved the overall survival in the second- and third-line settings, with manageable side effect profiles in HER2-positive breast cancer patients (Rassy et al, 2022). Despite the superior clinical activity for trastuzumab either alone or in a conjugation with cytotoxic moieties, resistance development, as is often the case with other chemotherapeutic agents, constrains their efficacy. Various molecular mechanisms contributing to resistance development for HER2-targeted therapies involving trastuzumab have been identified (Elshazly and Gewirtz, 2022), including signaling from other HER receptors, such as HER3 or epidermal growth factor receptor (Garrett and Arteaga, 2011), insulin-like growth factor receptor (Nahta, 2012), activation of PI3K/AKT/mTOR (Saal et al, 2005), overexpression of c-MET (Shattuck et al, 2008), or loss of PTEN (Saal et al, 2005). In addition, autophagy, due to its cytoprotective role in response to various chemotherapeutic agents, may hinder the therapeutic efficacy of these agents (Gewirtz, 2014).

3.1. Breast cancer

Vazquez-Martin et al (2009) studied the potential contribution of autophagy to the development of resistance to trastuzumab. The protein autophagy marker, LC3II, was significantly elevated in the trastuzumab-resistant SKBR3 cells, as compared with the parental SKBR3 cell line. Elevation of LC3II in the resistant cells was confirmed by the immunofluorescence of LC3 puncta. In addition, an increase in the levels of the lysosomal protein, LAMP-1, also shown by immunofluorescence staining, was indicative of increased lysosomal function, consistent with autophagy induction. The induction of autophagy was further confirmed by elevated p62/SQSTM1 degradation in the Tzb-resistant SKBR3 cells as compared with the parental cells.

The role of autophagy in this experimental system was investigated using the pharmacological autophagy inhibitors 3-MA, LY294002, and bafilomycin A1. 3-MA, an inhibitor of early stages of autophagy as shown by Wu et al (2010), reduced the viability of Tzb-resistant SKBR3 cells without affecting that of the parental cells. Similar results were obtained with LY294002, in which the trastuzumab-resistant cell line proved to be exquisitely sensitive to this agent. LY294002 blocks phosphatidylinositol 3-kinase activity and prevents autophagic sequestration, as demonstrated by Xing et al (2008), suggesting a cytoprotective role of autophagy in the resistant cell line. These results were further validated with bafilomycin A1, which prevents the fusion between autophagosomes and lysosomes (Klionsky et al, 2008), demonstrating markedly greater cytotoxicity in Tzb-resistant SKBR3 cells than the SKBR3 cells. The cytoprotective function of autophagy in this model was supported by a genetic approach using ATG8/LC3-targeted small interfering RNA. Genetic knockdown of ATG8 sensitized Tzb-resistant SKBR3 cells to trastuzumab, confirming the role of autophagy in the development of resistance to trastuzumab.

Vazquez-Martin et al (2009) also showed that trastuzumab treatment promoted autophagy in HER2-dependent BT474 breast carcinoma cells without affecting autophagy levels in HER2-negative MCF-7 cells as assessed by LC3 expression, lysosomal function, and p62/SQSTM1 expression. These data reinforce the premise that HER2 expression is required for autophagy induction following exposure to the anti-HER2 monoclonal antibody trastuzumab.

Rodríguez et al (2015) studied the effect of trastuzumab and its relationship with autophagy using two-dimensional and three-dimensional cultures of HER2-overexpressing BT474 cells and HER2-low-expressing MCF-7 cells. These investigators showed that trastuzumab treatment mildly increased the percentage of the BT474 cells undergoing G0/G1 cell cycle arrest, with no apoptosis induction, whereas MCF-7 cells were largely unaffected by trastuzumab treatment (as would be expected in cells that are HER2 low or negative). Trastuzumab treatment increased the ratio of LC3 II/I expression in both two-dimensional and three-dimensional cultures of the BT474 cells, suggesting an active autophagic flux. The latter was confirmed by autophagosome formation detected by immunofluorescence. Further studies of autophagy inhibition in trastuzumab-resistant BT474 cells as well as in the parental cells demonstrated that the combined treatment of trastuzumab together with the autophagy inhibitor, 3-MA, reduced the viability of both cell lines as compared with each drug alone, accompanied by the promotion of apoptosis, consistent with a cytoprotective role for the autophagic flux induced by trastuzumab.

Cufí et al (2013) investigated the potential of autophagy inhibition for restoring sensitivity to trastuzumab in HER2-positive breast cancer cells using trastuzumab-refractory JIMT-1 cells as well as trastuzumab-sensitive SKBR3 cells. Initially, immunoblotting indicated that JIMT-1 cells have higher levels of LC3II than in the SKBR3 cell line, suggesting a higher level of basal autophagic flux in the refractory cells; this observation was confirmed using confocal imaging indicating intense punctate LC3 fluorescence in the refractory cells. The active autophagic flux in the refractory cells was further validated by the utilization of a GFP-LC3 reporter, indicating higher GFP-LC3 puncta, together with a greater degree of p62/SQSTM1 degradation in the refractory cell line. Trastuzumab treatment in the JIMT-1 cells further increased GFP-LC3 puncta, suggesting that trastuzumab promoted autophagic flux. Of note, the dynamics/induction of the autophagic machinery in the refractory JIMT-1 cells was comparable to that observed in SKBR3 cells made resistant to trastuzumab by continued exposure for 10 months, modeling the primary resistance to trastuzumab.

These investigators then proceeded to determine whether the autophagy induced in the refractory cells played a cytoprotective role via genetic inhibition of autophagy using ATG8-targeted short hairpin RNA (shRNA). Genetic autophagy inhibition in combination with trastuzumab resulted in a small but significant reduction in the viability of JIMT-1 cells as compared with the controls, suggesting a possible cytoprotective role for the autophagic machinery induced in this system. Of note, autophagy inhibition increased the efficacy of other HER2-targeted therapies, including gefitinib, erlotinib, and lapatinib (Cufí et al, 2013). Pharmacologic inhibition of autophagy using chloroquine (CQ) in combination with trastuzumab synergistically reduced the viability of JIMT-1 cells, as shown by an MTT assay together with apoptosis induction validated by annexin V staining. These findings were further confirmed by determination of clonogenicity, in which the combination of CQ and trastuzumab markedly suppressed the frequency of surviving cells, whereas each drug alone did not affect the clonogenic capacity of JIMT-1 cells. Genetic inhibition of autophagy using shRNA directed to ATG8, ATG5, or ATG 12 or pharmacological inhibition of autophagy using CQ in combination with trastuzumab resulted in upregulation of cleaved caspase-3 levels as compared with either therapy alone, indicative of apoptosis induction. These data rigorously support the premise that the autophagic machinery induced in this model plays a cytoprotective role, participating in the development of resistance to HER2-targeted therapies, including trastuzumab, and indicating that autophagy targeting could be a viable strategy for tumor cell sensitization. Cufí et al (2013) further validated their results in vivo using JIMT-1 xenograft animal models. The combination of CQ and trastuzumab substantially reduced the tumor volume as compared with each drug alone, confirming the cytoprotective role played in this system by autophagy. Furthermore, the combination increased the expression of the proapoptotic protein BAX, along with reducing that of the antiapoptotic protein BCL2, suggesting that inhibition of the autophagic machinery induced apoptotic cell death in this experimental model.

Cufí et al (2012) also investigated which specific autophagy-related genes might be contributing to the development of resistance to trastuzumab in HER2 breast cancer. Initially, they showed that trastuzumab-refractory JIMT1 cells have significantly upregulated expression of ATG-12 by quantitative reverse transcription polymerase chain reaction (qRT-PCR). Furthermore, upon addition of trastuzumab to JIMT1 cells, autophagy-related genes, including PRKAA2, CDKN2A/p16, DAPK1, and ATG12, were further upregulated, consistent with the involvement of autophagy in the response to trastuzumab. To further confirm these results, these investigators analyzed the ATG-12 gene prolife across the Adai (GSE1090) gene expression dataset (which includes 56 breast cancer cell lines) as well as across the Neve’s gene expression data set (which includes 54 widely used breast cancer cell lines), where they classified the cell lines according to whether they were trastuzumab sensitive or trastuzumab refractory. Interestingly, ATG-12 mRNA expression was statistically elevated in the refractory cell lines as compared with the sensitive ones in both data sets. Upon performing ATG-12 knockdown using shRNA in JIMT1 cells, the sensitivity of JIMT1 to the cytotoxic effects of trastuzumab was clearly enhanced, suggesting a cytoprotective role for the autophagy induced by trastuzumab. Similarly, ATG-12 inhibition sensitized the JIMT11 cells to various HER1/2-targeted therapies, including gefitinib, erlotinib, and lapatinib. Cufí et al (2012) further investigated the effect of ATG-12 knockdown in vivo using the JIMT1 xenograft animal model. Here, as in the cell culture work, trastuzumab in combination with ATG-12 depletion drastically reduced the tumor volume as compared with each treatment alone, confirming the cytoprotective role of ATG-12 and, indeed, the autophagic machinery in this system.

Recently, Franco et al (2023) confirmed that combining CQ with trastuzumab resulted in a synergistic interaction in reducing the viability of trastuzumab-resistant HER2-positive JIMT-1 breast cancer cells.

In patient-related studies, Koukourakis et al (2014) investigated changes in various biomarkers in 28 patients diagnosed with locally advanced primary breast cancer who had received 1 cycle of trastuzumab, followed by a new biopsy on day 21, then treated with taxol/trastuzumab chemotherapy for 4 cycles prior to surgery. Trastuzumab administration markedly reduced clinical and positron emission tomography–detected tumor dimensions, proliferation index, and GLUT-1 as well as HER2 expression. Although the percentage of the LC3A-expressing cells as well as LC3A/HER2-expressing cells was upregulated, no change was found in LC3B- or LC3B/HER2-expressing cells. These results emphasize the need for further patient-related studies; however, the difficulties associated with monitoring autophagy and modifications to autophagy in patients make it extremely difficult to identify biomarkers that could be used to determine whether autophagy inhibition might be useful as a therapeutic strategy (Gewirtz, 2014).

Collectively, these results highlight the cytoprotective role of the autophagic machinery in response to trastuzumab in HER2-positive breast cancer as well as the involvement of autophagy in the development of resistance to HER2-targeted therapies, particularly in the case of monoclonal antibody therapy.

3.2. Gastric cancer

Ye et al (2018) studied the relationship between the autophagic machinery and trastuzumab using HER2-positive NCI-N87 gastric cancer cells, trastuzumab-resistant NCI-N87 cells, and HER2-negative SGC 7901 cells. Here, trastuzumab treatment reduced the viability of NCI-N87 parental cells, but not SGC 7901 cells, based on MTT and CCK-8 assays, consistent with trastuzumab acting specifically at the level of the HER2 receptor. Trastuzumab was also shown to inhibit the colony-forming abilities of NCI-N87 parental cells. In contrast to the studies in breast tumor cells, trastuzumab reduced the expression of LC3 I/II, while at the same time promoting p62/SQSTM1 accumulation in NCI-N87 parental cells, indicative of autophagy inhibition. Trastuzumab-resistant NCI-N87 cells exhibited an elevation in p62/SQSTM1 levels as compared with the parental cells, suggesting autophagic flux inhibition. The latter was further confirmed using transmission electron microscopy (TEM), where the resistant cell line exhibited smaller autolysosomes as compared with the parental cells, with no change in the autophagic flux upon trastuzumab addition. Interestingly, the autophagy inhibitors 3-MA, BAF A1, and HCQ reduced the viability of the parental cell line, whereas only HCQ reduced the viability of the resistant cell line. In combination with trastuzumab, neither BAF A1 nor 3-MA influenced the viability of the resistant cells, whereas HCQ reduced the viability of the resistant cells but to a lower extent than the parental cell line. Mechanistically, the resistant cell line exhibited an elevation in the phosphorylation levels of Akt and mTOR as compared with the parental cell line, suggesting that the Akt/mTOR pathway may play a role in resistance development, as activation of mTOR interferes with autophagy. In further studies, these investigators used everolimus, an autophagy inducer, which alone significantly reduced the viability of both parental and refractory gastric cancer cells to a greater extent than the autophagy inhibitors, as shown by MTT and CCK-8 assays. Furthermore, the levels of both p-mTOR and p-Akt were reduced upon combining trastuzumab with everolimus as compared with trastuzumab alone in the resistant cell line. These results suggested that autophagy inhibition (rather than autophagy induction, as is generally the case) via Akt and mTOR phosphorylation may mediate the resistant phenotype in the trastuzumab-resistant NCI-N87 gastric cancer cells. Although trastuzumab clearly inhibited the autophagic machinery, no clear conclusion can be established as to the nature of the autophagy in this system, especially in the parental cells, given the absence of genetic knockdown studies for autophagy inhibition (Klionsky et al, 2021).

It is important to emphasize the apparent differences in the roles of trastuzumab-associated autophagy in breast and gastric carcinoma, which is consistent with the premise that the nature of autophagy is likely to be dependent both on the chemical compound being used as well as the cell line/tumor being investigated (Gewirtz, 2014; Elshazly and Gewirtz, 2023d).

4. Trastuzumab-emtansine and autophagy

4.1. Breast cancer

In addition to trastuzumab in its monoclonal antibody form, the antibody-drug conjugate trastuzumab-emtansine and its relationship with autophagy is worthy of consideration. Liu et al (2020) studied the role of the autophagic machinery in the cytotoxicity of trastuzumab-emtansine using the HER2-positive human breast cancer cell lines BT-474 and SK-BR-3. Trastuzumab-emtansine induced apoptosis in both cell lines based on annexin V/PI flow cytometric measurements; these outcomes were confirmed based upon the upregulation of caspase-3/7 activity. Trastuzumab-emtansine treatment also induced the autophagic machinery, evidenced by the formation of autophagosomes as detected by TEM and LC3 II accumulation as well as autophagic fluorescent accumulation of Cyto-ID autophagy green dye. Autophagy inhibition using CQ reversed the apoptosis induced in response to trastuzumab-emtansine treatment, indicating that in these studies, autophagy had taken on a cytotoxic function. These results were further confirmed using another autophagy inhibitor, LY294002, in which trastuzumab-emtansine cytotoxicity was partially suppressed upon autophagy inhibition in both cell lines. Mechanistically, these investigators demonstrated that trastuzumab-emtansine induced the autophagic machinery via the suppression of mTOR phosphorylation. These results were further validated via evidence for the dephosphorylation of both 4E-binding protein 1 (4EBP1) and Protein S6 kinase (p70s6K), downstream targets of mTOR signaling (Jung et al, 2010; Yang et al, 2013). Additionally, the phosphorylation level of Akt was also significantly reduced upon trastuzumab-emtansine treatment. These findings of cytotoxic autophagy with the antibody-drug conjugate likely reflect the activity of the payload and clearly differ from the outcomes where autophagy function was interrogated with trastuzumab alone. At this point, there are relatively limited data relating to autophagy associated with the antibody-drug conjugate.

4.2. Gastric cancer

In contrast to the findings with breast cancer, Zhang et al (2021) reported that targeting autophagy in gastric carcinoma improves the therapeutic response to trastuzumab-emtansine using HER2-positive NCI-N87 cells. These investigators showed that trastuzumab-emtansine reduced the viability of NCI-N87 cells in a dose-dependent manner, together with the promotion of apoptosis, as shown by the annexin V/PI assay as well as the upregulation of caspases 3/9 and PARP levels by immunoblotting. Trastuzumab-emtansine also induced the autophagic machinery, as evidenced by the upregulation of LC3 II and p62/SQSTM1 degradation. The latter results were further confirmed via increased punctate autophagosome-like fluorescence by confocal microscopy using Cyto-ID and LysoTracker double staining as well as the formation and aggregation of autophagosomes visualized by TEM, confirming an active autophagic flux. Pharmacological autophagy inhibition using 3-MA and LY294002 markedly enhanced the cytotoxicity of trastuzumab-emtansine as compared with that for each drug alone, suggesting that the autophagy induced by trastuzumab-emtansine has a cytoprotective role. These investigators further validated their results in vivo using nude mice subcutaneously injected with NCI-N87 cells, in which the combination of trastuzumab-emtansine and LY294002 enhanced antitumor activity compared with that for each drug alone. Furthermore, autophagy blockade by LC3II reduction and p62/SQSTM1 accumulation, as well as apoptosis induction via cleaved PARP and caspase 9/3 upregulation, confirmed the cytoprotective role for the autophagic flux induced in this model by trastuzumab-emtansine. Mechanistically, they showed that trastuzumab-emtansine treatment reduced the phosphorylation of S2448 of mTOR and the phosphorylation of S473 on Akt, an upstream activator of mTOR (Memmott and Dennis, 2009), as well as inhibiting the downstream inducers of Akt/mTOR, 4E-BP1, and p70S6K (Jung et al, 2010; Yang et al, 2013). These data in gastric carcinoma are diametrically opposite to the effect of trastuzumab-emtansine in breast cancer, in which the cytotoxic role predominates, emphasizing the need for further investigation to clearly define whether autophagy targeting in gastric carcinoma could be a beneficial strategy to increase the effectiveness of trastuzumab-emtansine cleaved.

5. Conclusions

HER2-targeted therapies, including monoclonal antibodies, have improved overall survival in patients with breast and gastric cancer when HER2 is overexpressed (Yang et al, 2023). However, as is the case with other antineoplastic agents, resistance development often limits and compromises their therapeutic activity in the clinic (Elshazly and Gewirtz, 2022). Therefore, overcoming molecular mechanisms that contribute to resistance, such as autophagy, could serve to improve their clinical utility. Autophagy is fundamentally considered to be both a homeostatic and survival mechanism (Xu et al, 2022; Elshazly and Gewirtz, 2023a), whereby cells respond to stress, including starvation and nutrient deprivation (Xu et al, 2022). In the past decade, many clinical trials have been initiated to investigate the possible targeting of autophagy to overcome the development of resistance as well as increasing the efficacy of distinct antineoplastic agents (Chude and Amaravadi, 2017; Elshazly and Gewirtz, 2023a,c). Four different roles of the autophagic machinery have been identified in different tumor models, specifically cytostatic, nonprotective, and cytotoxic as well as cytoprotective autophagy, ostensibly the resistance-associated form (Gewirtz, 2014). Of note, it is generally accepted in the literature that the nature of the autophagy induced is dependent on the chemical nature of the inducing compound and the cell line/tumor type being used (Gewirtz, 2014; Elshazly and Gewirtz, 2023d).

As summarized in Table 1, the predominant role of the autophagy induced in response to trastuzumab in breast cancer is the cytoprotective form, highlighting the possible translation of autophagy targeting as an adjuvant therapy to increase the effectiveness of trastuzumab or to overcome resistance development in HER2-overexpressing breast cancer. Conversely, the preclinical data, although quite limited, suggest that trastuzumab may interfere with autophagy in HER2-expressing gastric carcinoma. Trastuzumab, as an antibody-drug conjugate, demonstrated essentially two different roles in breast and gastric cancer, cytotoxic and cytoprotective autophagy, respectively. The effects of trastuzumab and trastuzumab in the antibody-drug conjugate form on autophagy are likely to differ given that the drug emtansine, a microtubule poison, is the payload being delivered to the tumor cell. Consequently, instead of the sole interaction of trastuzumab with the HER2 receptor, whatever effects are observed with respect to autophagy will reflect some potentially complex but as yet undefined interaction with both the membrane receptor (trastuzumab) and microtubule function (emtansine).

Table 1.

Different forms of autophagy induced in response to HER2-targeted therapies

Agent	Tumor or Cell Type	Autophagy Induction	Role of Autophagy	References
Trastuzumab	Breast cancer; Tzb-resistant SKBR3, SKBR3, HER2-dependent BT474 and HER2-negative MCF-7 cells	Autophagy induced except in MCF-7 cells	Cytoprotective	Vazquez-Martin et al, 2009
Trastuzumab	Breast cancer; HER2-overexpressing BT474 cells, HER2-low expressing MCF-7 and resistant BT474 cell line	Autophagy induced	Cytoprotective	Rodríguez et al, 2015
Trastuzumab	Breast cancer; trastuzumab-refractory JIMT-1 cells as well as trastuzumab-sensitive SKBR3 cells In vivo using a JIMT-1 xenograft animal model	Autophagy induced, mainly in the resistant cells	Cytoprotective	Cufí et al, 2013
Trastuzumab	Breast cancer; Trastuzumab refractory JIMT1 In vivo using JIMT1 xenograft animal models	Autophagy induced	Cytoprotective	Cufí et al, 2012
Trastuzumab	Breast cancer; trastuzumab refractory JIMT1		Cytoprotective	Franco et al, 2023
Trastuzumab	Breast cancer; patients with locally advanced primary breast cancer–related studies		The role of autophagy is not established	Koukourakis et al, 2014
Trastuzumab	Gastric cancer; HER2-positive NCI-N87 cells, trastuzumab-resistant NCI-N87 cells, and HER2-negative SGC 7901 cells	Autophagy inhibited	NA	Ye et al, 2018
Trastuzumab-emtansine	Breast cancer; HER2-positive human breast cancer cell lines; BT-474 and SK-BR-3 cells	Autophagy induced	Cytotoxic	Liu et al, 2020
Trastuzumab-emtansine	Gastric cancer; HER2-positive NCI-N87 cells In vivo using nude mice subcutaneously injected with NCI-N87 cells	Autophagy induced	Cytoprotective	Zhang et al, 2021

NA, not applicable.

Additional studies confirming the role of autophagy in vivo using different animal models would likely clarify these differences (Klionsky et al, 2021). It is critical to emphasize the major difficulty in extrapolating from preclinical to clinical settings in that no accurate, reliable, noninvasive methods are currently available to determine the induction/inhibition of autophagic flux in patients (Bristol et al, 2013; Gewirtz, 2016), thereby limiting efforts to the measurement of autophagy markers in peripheral blood mononuclear cells (Wu et al, 2018; Bensalem et al, 2021). Regarding the autophagy inhibitors, only HCQ is currently recommended for in vivo studies as it is the only autophagy inhibitor that is also approved for clinical trials. However, reports by Kimura et al (Kimura et al, 2013) and Gies et al (Gies et al, 2020) have argued that the effects of chloroquine and hydroxychloroquine may not be limited to autophagy inhibition. There are other potential autophagy inhibitors being investigated in preclinical studies, for example, the ULK inhibitors described by Karmacharya and Jung (2023) and Martin et al (2018), in addition to others further discussed by Chude and Amaravadi (Chude and Amaravadi, 2017). With regard to the other available monoclonal antibodies, pertuzumab and margetuximab, as well as trastuzumab-deruxtecan, only limited data are available regarding the involvement of autophagy in the tumor cell response in either breast cancer or gastric cancer.

Conflict of interest

The authors declare no conflicts of interest.