Activity and mechanism of acquired resistance to tarloxotinib in HER2 mutant lung cancer: an in vitro study.

BACKGROUND: HER2 (ERBB2) activating mutations are present in 2-3% of lung adenocarcinomas; however, no targeted therapy is approved for HER2-altered lung cancers. A novel pan-HER inhibitor, tarloxotinib, is designed to release the active form (tarloxotinib-E) under hypoxic conditions in tumor tissues after being administered as a prodrug. Following the evaluation of the in vitro activity of tarloxotinib-E in HER2-mutant cells, we explored the mechanisms of resistance to tarloxotinib-E in these cells.
METHODS: Growth inhibitory assays were performed with tarloxotinib-E and its prodrug using Ba/F3 cells expressing one of six HER2 mutations or wild-type (WT) HER2, in addition to H1781 cells with HER2 exon 20 insertions. Resistant clones were established from N-ethyl-N-nitrosourea (ENU)-treated HER2-mutant Ba/F3 cells and H1781 cells by chronic exposure to tarloxotinib-E.
RESULTS: Tarloxotinib-E showed potent activity against HER2-mutant Ba/F3 cells and H1781 cells. Furthermore, the half maximal inhibitory concentration (IC50) of tarloxotinib (inactive form) for WT HER2 was 180 times higher than that of tarloxotinib-E, indicating a wide therapeutic window of tarloxotinib. We established 30 resistant clones with secondary mutations of HER2 by ENU mutagenesis, all of which harbored C805S in exon 20. In the analysis of H1781 cells that acquired resistance to tarloxotinib-E, we found that increased HER3 expression was the molecular mechanism of tarloxotinib-E resistance.
CONCLUSIONS: Tarloxotinib-E exhibited potent activity against cell line models with HER2 mutations. We identified a secondary C805S HER2 mutation and HER3 overexpression as the mechanisms of acquired resistance to tarloxotinib-E. 2021 Translational Lung Cancer Research. All rights reserved.

Introduction

HER2 activating mutations, most of which are exon 20 insertion mutations, are present in 2–3% of lung adenocarcinomas (1,2). Despite the rapid development of molecular targeted therapies for non-small cell lung cancers (NSCLCs) with EGFR, ALK, ROS1, BRAF, NTRK or MET alterations, no drug has been approved for the treatment of HER2-mutant NSCLCs. Several clinical trials of HER2-targeted drugs, such as afatinib, neratinib and dacomitinib, for HER2-mutant NSCLC have been conducted; however, the activity of these drugs was limited (3-5).

Recently, we reported potent activity of poziotinib for HER2 exon 20 insertions by comparing nine HER2 inhibitors in an in vitro study (6). Poziotinib has shown potent activity against EGFR and HER2 exon 20 mutant lung cancer in phase II trials and preclinical studies (7,8); however, 67% (8/12) of patients who had received poziotinib experienced dose reduction due to adverse events, and eight patients (67%) had grade 3–4 AEs (8). Therefore, for HER2-mutated lung cancers, a variety of new therapeutic strategies are currently being tested in clinical trials, including a new drug delivery system (tarloxotinib) and an antibody-drug conjugates [trastuzumab deruxtecan (T-DXd)] (9).

Tarloxotinib is a novel pan-HER-targeted drug (10), and a phase II study of NSCLCs harboring HER2-activated mutations or EGFR exon 20 mutations is ongoing (NCT03805841). This drug is designed as a prodrug (tarloxotinib) that releases the activated pan-HER inhibitor [tarloxotinib-effector (tarloxotinib-E)] under pathological hypoxic conditions within the tumor (Figure S1) (11). This activation mechanism is expected to enlarge the therapeutic window and reduce toxicities such as diarrhea, rash and paronychia as a systemically active pan-HER inhibitor.

In this study, we evaluated the activity of tarloxotinib-E using in vitro models of HER2-mutated lung adenocarcinoma. In addition, we also investigated potential acquired resistance mechanisms to tarloxotinib-E. We present the following article in accordance with the MDAR reporting checklist (available at https://dx.doi.org/10.21037/tlcr-21-216).

Methods

Cell lines

We used the immortalized murine pro-B cell line Ba/F3, which was obtained from the RIKEN Bio Resource Center (Tsukuba, Japan). NCI-H1781 (RRID:CVCL_1494) cells, which are a lung adenocarcinoma cell line harboring the HER2 exon 20 G776delins VC mutation (VC), were purchased from the American Type Culture Collection (Virginia, USA) and authenticated by short-tandem repeat profiling (Takara, Kusatsu, Japan). These cells were cultured in RPMI 1640 (Wako, Osaka, Japan) with 10% FBS (Sigma-Aldrich, St. Louis, MO, USA) and penicillin-streptomycin (P/S, Wako) at 37 °C with 5% CO2. We routinely checked for mycoplasma contamination using the TaKaRa PCR Mycoplasma Detection Set (Takara).

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). All the in vitro experiments were performed in compliance with the institutional and national guidelines.

Establishment of Ba/F3 cells with HER2 mutations

Ba/F3 cells harboring the A775_G776insYVMA (YVMA), VC, or P780_Y781insGSP (GSP) HER2 mutation and Ba/F3 cells with wild-type (WT) HER2 were generated in our previous study (6,12). Ba/F3 cells with one of four HER2 point mutations (S310F, V659E, L755A, or L755P) were established using retroviral transfection as described previously (6,12). The list of designed primers is summarized in Table S1. Briefly, a pBABE retrovirus vector subcloned cDNA of the human WT HER2 gene was purchased from Addgene (Cambridge, MA, USA). We introduced each HER2 mutation into the pBABE vector of WT HER2 using a Prime STAR Mutagenesis Basal Kit (Takara) and designed primers (Table S1). The pBABE constructs were transfected with a pVSV-G vector [Takara (Clontech, Fremont, CA, USA), RRID:Addgene_138479] into gp-IRES 293 cells using FuGENE6 transfection reagent (Promega, Madison, WI, USA). Forty-eight hours after transfection, we collected the culture medium, and the viral particles were concentrated using a Retrovirus Concentration kit [Takara (Clontech)] and stored at −80 °C. Ba/F3 cells (3×103/well) were seeded onto 24-well plates and cultured for 24 hours. A retroviral suspension was added to each well. After a few days of incubation, the infected Ba/F3 cells were selected using puromycin (1.0–1.5 μg/mL) in the presence of interleukin-3 (IL-3). Then, we cultured puromycin-selected cells without IL-3 for approximately one week.

Reagents

Tarloxotinib and its activated drug, tarloxotinib-E, were provided by Rain Therapeutics Inc. We also used afatinib, poziotinib [second-generation (2G) EGFR TKIs], osimertinib [third-generation (3G) EGFR TKIs], and pyrotinib, which were purchased from Selleck Chemicals (Houston, TX, USA). Each drug was dissolved to a concentration of 10 mmol/L in dimethyl sulfoxide (DMSO, Sigma-Aldrich) and stored at −80 °C until use.

Cell proliferation assay

We seeded 3×104 transfected Ba/F3 cells into six-well plates. As a control, nontransfected Ba/F3 cells were also cultured in the presence and absence of IL-3. We counted the number of cells in each well using OneCell Counter (Biomedical Medical Science, Tokyo, Japan) in triplicate every 24 hours for 96 hours.

Cell viability assay

Cell viability assays were performed using Cell Counting Kit-8 (Dojindo Laboratories, Kumamoto, Japan). A total of 3×103 cells were plated onto 96-well plates in technical triplicates and incubated for 24 hours. The cells were treated with each TKI at ten different drug concentrations. After 72 hours, 10 μL of tetrazolium salt WST-8 was added to each well, and plates were incubated for 2–4 hours. To determine bioluminescence, the formazan dye generated from tetrazolium in the reduction reaction with NADH, which reflects cell viability, was measured by reading the absorbance at 450 nm using a multiplate reader (Tecan, Mannedorf, Switzerland). Cell viability assay was performed at least twice.

Statistical analysis

The growth inhibitory curves were generated using GraphPad Prism 9 (GraphPad Software, San Diego, CA). Calculation of the standard deviation and estimation of the IC50 values were also performed using GraphPad Prism 9.

Generation of tarloxotinib-E resistant clones

We established tarloxotinib-E-resistant (TR) clones using HER2-mutant Ba/F3 cells through the N-ethyl-N-nitrosourea (ENU, Sigma-Aldrich) mutagenesis technique, as previously reported (13). HER2 mutant Ba/F3 cells were exposed to 100 μg/mL ENU for 24 hours. After removal of ENU, 1×105 cells were plated onto 96-well plates. These cells were incubated with 200 nM tarloxotinib-E for 14-21 days (the medium was changed every a few days), and TR clones were selected.

In the chronic exposure assay, H1781 cells were cultured with an increasing concentration of tarloxotinib-E starting at 2 nM. Acquired resistance was defined by the 50-fold higher IC50 of the TR cells than that of the parental H1781 cells.

HER2 mutation analyses

Total RNA from the transfected cells and resistant clones was extracted using a mirVana miRNA Isolation Kit (Qiagen, Hilden, Germany) and transcribed to cDNA using ReverTra Ace (TOYOBO, Osaka, Japan). Each transduced region and HER2 tyrosine kinase domains (TKD, exons 18 to 24) were amplified using designed primers. Sanger sequencing was conducted by a Genetic Analyzer 3130 or 3500XL (Applied Biosystems, Waltham, MA).

Western blot analysis

The cells were treated with the indicated concentrations of tarloxotinib-E or DMSO for 6 hours and washed with PBS. Total cell pellets were dissolved in lysis buffer, and a total of 15–30 μg of protein was loaded into SuperSep TM Ace (Wako) and blotted. Protein was transferred to the membrane [Trans-Blot® Turbo™ Mini PVDF Transfer Pack (Bio-Rad, Hercules, CA)], and membranes were probed using one of following antibodies: anti-HER2 (Cell Signaling Technology Cat# 2242, RRID:AB_331015), phosphorylated (p)-HER2 (Tyr1221/1222, Cell Signaling Technology Cat# 2243, RRID:AB_490899), anti-HER3 (Cell Signaling Technology Cat# 4754, RRID:AB_10691324), p-HER3 (Tyr1289, Cell Signaling Technology Cat# 4791, RRID:AB_2099709), anti-Akt (Cell Signaling Technology Cat# 9272, RRID:AB_329827), p-Akt (Ser473, Cell Signaling Technology Cat# 9271, RRID:AB_329825), anti-Erk1/2 (p44/42 MAPK, Cell Signaling Technology Cat# 9102, RRID:AB_330744), p-Erk1/2 (p44/42 MAPK, Thr202/Tyr204, Cell Signaling Technology Cat# 9100, RRID:AB_330741), and β-actin (Cell Signaling Technology Cat# 4967, RRID:AB_330288) overnight at 4 °C. Horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (Cell Signaling Technology Cat# 7074, RRID:AB_2099233) was used as the secondary antibody. All antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). In the chemiluminescence assay, the membranes were reacted with enhanced chemiluminescence (ECL) solution (GE Healthcare, Chicago, Illinois) and scanned by an Amersham imager 680 (GE Healthcare).

p-receptor tyrosine kinase (RTK) array analysis

To detect the changes in the phosphorylation status of 49 different RTKs, we used a Human Phospho-RTK Array (R&D, Minneapolis, MS, USA). Parental H1781 and TR cells were treated with DMSO or 100 nM tarloxotinib-E for 24 hours, and then, cell lysates were prepared using lysis buffer according to the protocol. After blocking the array sheets, 200 µg of prepared sample was added to the array sheets and incubated overnight at 4 °C. After washing with wash buffer, the diluted anti-p-tyrosine-HRP detection antibody was pipetted into each array sheet. The array sheets were exposed to Chemi-Reagent Mix and scanned by an Amersham imager 680 (GE Healthcare).

Small interfering RNA

TR clones generated from H1781 cells were reverse-transfected with scrambled or HER3-targeting small interfering RNA (siRNA) purchased from Sigma-Aldrich (scramble siRNA, siHER3). Briefly, siRNA dissolved in OptiMEN (Thermo Fisher, Waltham, MA, USA) was spread to 60 mm dishes or 96-well plates, and H1781 TR cells were seeded in the plates. After 24 hours of incubation, H1781 TR cells were treated with tarloxotinib-E or DMSO. We collected H1781 TR cells after 12 hours of treatment and used them for western blotting.

Copy number analysis

Genomic DNA was extracted using a DNA extraction kit (Qiagen). Quantitative real-time PCR (qPCR) was performed using CYBR Green Real-time PCR Master Mix (Applied Biosystems) and the designed primers described in Table S2 with the StepOnePlus System (Applied Biosystems, Foster City, CA). The copy number of the LINE-1 gene was utilized as the internal control as described previously (14).

mRNA expression analysis

We extracted the total mRNA from H1781 and TR cells and then reverse-transcribed it to cDNA as described above. qPCR was performed using TaqMan probes targeting HER3 and TaqMan Fast Advanced Master Mix (Applied Biosystems). We quantitated the mRNA expression of HER3 by the delta-delta CT method using the expression of 18S ribosomal RNA as the internal control as described previously (14).

Results

HER2 mutations in non-small cell lung cancer

We searched for the frequency of HER2 mutations in NSCLC using the cBioPortal database (https://www.cbioportal.org) on March 1st, 2019 (). To avoid overlapping data, we referred to eight studies to identify the common HER2 mutation in NSCLC (2,15-21). As previously reported (6,12), HER2 exon 20 insertions, YVMA, VC and GSP, were the most common HER2 mutations in NSCLC. In addition to these insertions, four single base substitutions, namely, S310F, V659E, L755A and L755P, were identified as common HER2 gene alternations in NSCLC.

Figure 1

Mutation spectrum of HER2 in NSCLC detected in the cBioPortal database. In the cBioPortal database, we identified three HER2 exon 20 insertions and four HER2 point mutations that are common in NSCLCs from eight studies that performed comprehensive genetic screening (2,15-21). NSCLC, non-small cell lung cancer.

In our previous study (6), we observed that Ba/F3 cells harboring YVMA, VC, GSP and WT HER2 proliferated well independent of IL-3. In this study, we found that Ba/F3 cells harboring one of the three point mutations (V659E, L755A and L755P) were also viable in the absence of IL-3 but not Ba/F3 cells with the S310F mutation (Figure S2).

Activity of tarloxotinib-E in HER2 mutated in vitro models

In the growth inhibitory assay, tarloxotinib-E showed potent efficacy against all Ba/F3 cells harboring HER2 exon 20 insertions as well as those harboring HER2 point mutations (). The IC50 values of tarloxotinib-E for HER2-mutant Ba/F3 cells and H1781 cells were less than 5 nM, which were lower than afatinib, osimertinib and pyrotinib and equivalent to that of poziotinib. The IC50 values of tarloxotinib (prodrug) for WT HER2 were much higher than those of other TKIs; for example, tarloxotinib had an 800-fold higher IC50 value than poziotinib. Western blotting showed that tarloxotinib-E inhibited the phosphorylation of HER2 exon 20 insertions and point mutations at 10 nM (Figure S3). On the other hand, tarloxotinib had no effect on the phosphorylation of HER2, even at a concentration of 100 nM (Figure S3).

Figure 2

Tarloxotinib-E demonstrated high inhibition activity against HER2 mutations, and tarloxotinib showed low activity against WT HER2. (A) Growth inhibition curves of H1781 and Ba/F3 cells harboring WT HER2, HER2 with exon 20 insertions or HER2 with point mutations for each HER2 TKI. (B) A mosaic table that summarizes the measured IC50 values of (A) shows comparable efficacies of tarloxotinib-E and poziotinib and the minor activity of tarloxotinib. Each value indicates the IC50 calculated from the growth inhibition curve. Green indicates high sensitivity, corresponding to when the IC50 is less than 5 nM. Yellow indicates moderate sensitivity, corresponding to when the IC50 is higher than 5 nM but lower than 50 nM. Red indicates resistance, corresponding to when the IC50 is higher than 50 nM. (C) The plot table of SI of HER2 TKIs revealed the high SI of tarloxotinib/tarloxotinib-E. Each plot was calculated by [IC50 of Ba/F3 cells expressing WT HER2]/[IC50 of Ba/F3 cells expressing each HER2 mutant]. Regarding the plots of tarloxotinib-E, the SI was calculated by [IC50 of WT HER2 treated with tarloxotinib]/[IC50 of each HER2 mutant treated with tarloxotinib-E]. YVMA, A775_G776insYVMA; VC, G776_delinsVC; GSP, P780_Y781insGSP; WT, wild type; TKI, tyrosine kinase inhibitor; IC50, half maximal (50%) inhibitory concentration; SI, sensitivity index.

To compare the therapeutic window, we calculated the selectivity index (SI) of these drugs according to a previous report (22). The SI was calculated as the ratio of the IC50 of each TKI for WT HER2 and for each HER2 mutation (). Because tarloxotinib circulates as a prodrug and is activated under hypoxic intratumor conditions, the SI was indicated as [IC50 of tarloxotinib for WT HER2 cells]/[IC50 of tarloxotinib-E for HER2 mutant cells]. Many of the SIs of afatinib, poziotinib, osimertinib and pyrotinib were <1, meaning that these TKIs have inhibitory activity against WT HER2 compared with HER2 mutants. In contrast, the SI of tarloxotinib/tarloxotinib-E was >10 for all HER2-mutant Ba/F3 cells, indicating that tarloxotinib will have a wide therapeutic window in the clinical setting.

C805S mutation as an acquired resistance mechanism to tarloxotinib-E

To investigate the mechanisms of acquired resistance to tarloxotinib-E, we performed ENU mutagenesis using Ba/F3 cells with HER2 mutations. In this assay, we obtained a total of 50 clones that acquired tarloxotinib-E resistance from HER2-mutant Ba/F3 cells (). Thirty clones among these 50 developed HER2 secondary mutations, all of which were C805S substitutions, which are homologous to C797S of EGFR (). No secondary mutation was found in the HER2 TKD of the remaining 20 clones.

Figure 3

The secondary C805S mutation was identified in TR clones of Ba/F3 cells harboring a HER2 mutation through ENU mutagenesis. (A) Number of TR clones established from ENU mutagenesis. The only secondary mutation identified in the HER2 TKD was C805S. (B) Chromatogram of the secondary C805S mutation identified in TR clones. (C) Ba/F3 cells harboring either HER2 exon 20 insertions or HER2 point mutations with the secondary C805S mutation acquired resistance to tarloxotinib-E and poziotinib. The killing curves of parental cells with HER2 exon 20 insertions or point mutations were extracted from as a control. (D) A mosaic table of the measured IC50 values of the acquired C805S clones shows that the secondary C805S mutation causes significant resistance to all HER2 TKIs. Yellow indicates moderate sensitivity, corresponding to when the IC50 is higher than 5 nM but lower than 50 nM. Red indicates resistance, corresponding to when the IC50 is higher than 50 nM. YVMA, A775_G776insYVMA; VC, G776_delinsVC; GSP, P780_Y781insGSP; TKD, tyrosine kinase domain; CS, C805S; TR, tarloxotinib-E-resistant; ENU, N-ethyl-N-nitrosourea; IC50, half maximal (50%) inhibitory concentration.

The IC50 values of tarloxotinib-E and poziotinib for Ba/F3 cells that acquired C805S were 50–200 times higher than those of parental cells (). We also observed that Ba/F3 cells with secondary C805S mutations were resistant to all currently available HER2-TKIs ( and Figure S4A). As shown in the immunoblots, the phosphorylation of HER2 in resistant cells treated with C805S was not reduced by tarloxotinib-E, even at a concentration of 100 nM (Figure S4B).

Activation of HER3 as an acquired resistance mechanism to tarloxotinib-E

We also established H1781 cells with acquired resistance to tarloxotinib-E (H1781 TR cells) via a chronic exposure assay at increasing concentrations of tarloxotinib-E from 2 nM to 100 nM. This clone showed approximately 100 times higher IC50 values than the parental H1781 cells (). There was no morphologic change that suggests the acquisition of epithelial to mesenchymal transition in H1781 TR cells (). HER2 phosphorylation was suppressed in H1781 TR cells, while Erk1/2 and Akt phosphorylation remained after 100 nM tarloxotinib-E treatment (). We detected no secondary mutations in the HER2 TKD, including C805S, in H1781 TR cells.

Figure 4

p-HER3 activation was identified as the acquired resistance mechanism to tarloxotinib-E in H1781 cells through chronic exposure. (A) Growth inhibition curves of the H1781 TR clone and its parental H1781 cells. (B) Images of parental H1781 and TR cells through the chronic exposure assay (EVOS XL Core, 4×). (C) Western blotting revealed that downstream signals were not inhibited by tarloxotinib-E in TR cells. (D) Compared with parental H1781 cells treated with DMSO (upper-left), the phosphorylation level of HER3 in H1781 cells treated with 100 nM tarloxotinib-E (upper-right) was decreased in the human phospho-RTK array. On the other hand, the phosphorylation level of the H1781 TR clone remained the same after exposure to 100 nM tarloxotinib-E (lower right). (E) Western blotting of p-HER3 and total HER3 in parental and H1781 TR cells. TR, tarloxotinib-E-resistant; tar-E, tarloxotinib-E; p-HER3, phosphor-HER3; p-Erk1/2, phosphor-Erk1/2; t-Erk1/2, total-Erk1/2; p-Akt, phosphor Akt; t-Akt, total-Akt.

To explore the resistance mechanism of H1781 TR cells, we used the Human Phospho-RTK Array (). In this immunoassay, the phosphorylation of HER3 was increased in H1781 TR cells treated with 100 nM tarloxotinib-E compared with H1781 parental cells treated with 100 nM tarloxotinib-E. Using western blotting, we confirmed that p-HER3 was higher in H1781 TR cells than in parental cells in the presence of tarloxotinib-E (). We identified that the mRNA expression of HER3 was increased in TR cells compared with parental H1781 cells (). In addition, we found that H1781 TR cells acquired a gene copy number gain of HER3 by 1.5 times compared with parental cells ().

Figure 5

HER3 copy number increased in TR cells, and interfering with HER3 expression restored sensitivity to tarloxotinib-E. (A) HER3 expression levels are higher in TR cells than in parental H1781 cells, regardless of tarloxotinib-E exposure. (B) The HER3 DNA copy number of TR cells was approximately 1.5 times higher than that of parental H1781 cells. (C) TR cells were resensitized to tarloxotinib-E by combined treatment with siHER3. (D) The phosphorylation of HER3 and its downstream signal, p-Akt, were suppressed by the combination of tarloxotinib-E and siHER3. On the other hand, the decrease in p-Erk1/2 was limited. tar-E, tarloxotinib-E; TR, tarloxotinib-E-resistant; siHER3, small interfering HER3; p-Erk1/2, phosphor-Erk1/2; t-Erk1/2, total-Erk1/2; p-Akt, phosphor Akt; t-Akt, total-Akt.

To analyze the role of HER3 in H1781 TR cells, we silenced HER3 using siRNA in this cell line. We observed that the combination of tarloxotinib-E and siRNA-mediated HER3 knockdown effectively killed H1781 TR cells (). Western blotting revealed that pAkt was significantly suppressed by the combination of tarloxotinib-E and siRNA-mediated HER3 knockdown ().

Discussion

In this in vitro study, tarloxotinib-E demonstrated potent activity against Ba/F3 models with HER2 exon 20 insertions or with HER2 point mutations that have been detected in NSCLCs. Recently, some novel agents have shown promising activity against HER2-altered NSCLCs in early-phase clinical trials. Socinski et al. (23) reported a 27.8% overall response rate (ORR) and 5.6-month progression-free survival (PFS) in twelve patients in a phase II study of poziotinib. T-DXd, a HER2-targeted antibody-drug conjugate, achieved a 61.9% ORR in eleven NSCLC patients with HER2 mutations in a dose-escalation phase I study (9). Most recently, pyrotinib showed efficacy against HER2-mutant lung adenocarcinomas with an ORR of 30.0% and a PFS of 6.9 months in a phase II study (24). However, it is also true that these novel HER2 inhibitors have high toxicity, such as diarrhea, skin rash, and drug-induced interstitial lung disease. In the early-phase trials mentioned above, dose reduction was needed in 67% of patients who received poziotinib and in 24% of patients who received T-DXd due to adverse events (8,25).

In this study, we demonstrated that tarloxotinib (prodrug) was not toxic against Ba/F3 cells with WT HER2 in contrast to other HER2 TKIs (afatinib, poziotinib, pyrotinib, and osimertinib). Because tarloxotinib-E (active form) inhibited the growth of Ba/F3 cells with HER2 mutations at a similar range as poziotinib, we expect high efficacy and lower toxicity of tarloxotinib in patients with HER2 mutations. In fact, in the first report of an ongoing phase II study of tarloxotinib in NSCLC patients, tumor response by RECIST was observed in 44% (4/9) of a cohort of NSCLC patients with a HER2-activating mutation, while dose reduction was required only in 21.7% (5/23) of patients in a whole cohort (NSCLCs harboring EGFR exon 20 insertion or HER2 mutations and solid tumors with NRG1/ERBB gene fusions) (26).

In this study, we also explored acquired resistance mechanisms to tarloxotinib-E in vitro. In the Ba/F3 models, we identified a secondary HER2 C805S mutation. HER2 C805 is homologous to C797 in the EGFR gene, in which irreversible TKIs form covalent bonds; therefore, it is reasonable that the C805S secondary mutation conferred resistance to all irreversible HER2 inhibitors, including tarloxotinib-E.

In addition to the secondary C805S mutation, we identified an acquired resistance mechanism by increased HER3 expression in H1781 cells. To the best of our knowledge, this is the first report of HER3 activation as an acquired resistance mechanism to HER2 inhibitors in lung cancer cells with HER2 mutations, although HER3 has been reportedly involved in the acquisition of resistance to gefitinib [an EGFR-TKI (27)] and alectinib [an ALK-TKI (24)]. Western blot analysis revealed an increased pAkt level in H1781 TR cells with HER3 activation (), which is reasonable because HER3 contains six Tyr-Xaa-Xaa-Met (YXXM) consensus binding sites for the SH2 domains of the p85 regulatory subunit of PI3K (28). We found that siRNA-mediated HER3 knockdown resensitized H1781 TR cells to tarloxotinib-E, confirming the role of HER3 in the acquisition of resistance to tarloxotinib-E in H1781 cells. Therapeutically, we expect that anti-HER3 monoclonal antibodies, such as patritumab or lumretuzumab, or bispecific HER2/HER3 antibodies, such as istiratumab or duligotumab, may be able to overcome tarloxotinib-E resistance with HER3 activation (29). In this study, we explored acquired resistance mechanisms mediated by secondary HER2 mutations using Ba/F3 models and those mediated by bypass signaling using H1781 cells. However, it is not clear which mechanism, secondary mutations vs. bypass pathway, is more likely to occur in the actual treatment of HER2-mutated lung cancer patients. In addition, it is possible that other types of resistances, such as histological transformation or modification of tumor microenvironment, may arise. Therefore, future analyses using clinical specimens obtained from tarloxotinib-E refractory patients are necessary to confirm our in vitro findings.

In conclusion, tarloxotinib-E showed potent activity against cell lines expressing HER2 mutations, both exon 20 insertions and HER2 point mutations. In addition, we identified a secondary C805S HER2 mutation as well as increased HER3 expression as molecular mechanisms of acquired resistance to tarloxotinib-E in our in vitro models.

The article’s supplementary files as