Antitumor activity of aumolertinib, a third-generation EGFR tyrosine kinase inhibitor, in non-small-cell lung cancer harboring uncommon EGFR mutations

Abstract

Uncommon epidermal growth factor receptor (EGFR) mutations account for 10%–20% of all EGFR mutations in non-small-cell lung cancer (NSCLC). The uncommon EGFR-mutated NSCLC is associated with poor clinical outcomes and generally achieved unsatisfactory effects to the current therapies using standard EGFR-tyrosine kinase inhibitors (TKIs), including afatinib and osimertinib. Therefore, it is necessary to develop more novel EGFR-TKIs to treat uncommon EGFR-mutated NSCLC. Aumolertinib is a third-generation EGFR-TKI approved in China for treating advanced NSCLC with common EGFR mutations. However, it remains unclear whether aumolertinib is effective in uncommon EGFR-mutated NSCLC. In this work, the in vitro anticancer activity of aumolertinib was investigated in engineered Ba/F3 cells and patient-derived cells bearing diverse uncommon EGFR mutations. Aumolertinib was shown to be more potent in inhibiting the viability of various uncommon EGFR-mutated cell lines than those with wild-type EGFR. And in vivo, aumolertinib could also significantly inhibit tumor growth in two mouse allograft models (V769-D770insASV and L861Q mutations) and a patient-derived xenografts model (H773-V774insNPH mutation). Importantly, aumolertinib exerts responses against tumors in advanced NSCLC patients with uncommon EGFR mutations. These results suggest that aumolertinib has the potential as a promising therapeutic candidate for the treatment of uncommon EGFR-mutated NSCLC.

Graphical abstract

Aumolertinib, a novel third-generation EFGR-TKI, exerts promising in vitro and in vivo antitumor activities in uncommon EGFR-mutated NSCLC models by inhibiting EGFR phosphorylation and downstream signal transduction.

1. Introduction

An increasing number of driver genes for non-small-cell lung cancer (NSCLC) have been identified in recent decades. The most common driven mutations occur in the epidermal growth factor receptor (EGFR)¹. These EGFR mutations are usually classified into two categories: common mutations and rare mutations (also known as uncommon mutations)². The common mutations, including exon 19 deletion (Ex19del) mutations and L858R substitution mutation in exon 21, represent over 80% of all EGFR mutations, while the other uncommon EGFR mutations, such as exon 20 insertion (Ex20ins), G719X in exon 18, S768I in exon 20, and L861Q in exon 21, account for the 10%–20% in NSCLC with EGFR mutations (Fig. 1A).

Figure 1.

EGFR-TKIs therapy for NSCLC with uncommon EGFR mutations is still needed. (A) Frequencies of EGFR mutations in NSCLC. Data were obtained from the COSMIC database and filtered to contain only mutations from adenocarcinoma and excluded the two resistance mutations T790M and C797S. (B) Chemical structure of aumolertinib. (C) Structural cartoon model representation of the EGFR catalytic domain with aumolertinib and its interactions with residues nearby. Aumolertinib is displayed in purple, and the key amino acid residues of EGFR were labeled. (D) EGFR signaling pathway and mechanism of action of aumolertinib. NSCLC, non-small-cell lung cancer; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; TKIs, tyrosine kinase inhibitors; TGF-α, transforming growth factor-α; P, phosphorylation; Ex20ins, Exon 20 insertion; Ex19del, Exon 19 deletion.

The identification of EGFR mutations enables the breakthroughs of targeted therapy for the treatment of NSCLC. The tyrosine kinase inhibitors (TKIs) targeting EGFR have been developed for three generations: the first-generation (erlotinib and gefitinib), the second-generation (afatinib and dacomitinib), and the third-generation (osimertinib)^3,4. These drugs have brought remarkable survival benefits to NSCLC patients with common EGFR mutations. However, unlike the NSCLC with common EGFR mutations, the clinical benefits for uncommon EGFR mutations were unsatisfactory^5,6.

For uncommon EGFR mutations, the first-line therapies use standard EGFR-TKIs (including afatinib and osimertinib) or remain conventional cytotoxic therapies similar to the treatment of EGFR wild-type (WT) tumors⁷. Although the FDA approved mobocertinib and amivantamab for Ex20ins mutant NSCLC, the objective response rates of mobocertinib and amivantamab were 28% and 40%, respectively, and the clinical efficacy of these drugs remains imperfect based on current data^8,9. It is also noteworthy that these two drugs are generally associated with undesirable adverse events such as rashes and diarrhea due to the nonspecific inhibition of EGFR WT, which limit their clinical antitumor activity to some extent¹⁰. Furthermore, the affordability and accessibility of mobocertinib and amivantamab were another constraint¹¹. Therefore, it is urgent to develop more novel EGFR-TKIs for treating NSCLC with uncommon EGFR mutations with improved efficacy and safety.

Aumolertinib (Fig. 1B), the second third-generation EGFR-TKI around the world¹², is a novel drug that has been approved in China for locally advanced or metastatic NSCLC with EGFR T790M, Ex19del, or L858R mutations¹³. Compared with osimertinib, the innovative introduction of the cyclopropyl structure into indole nitrogen in aumolertinib increases the metabolic stability, the permeability of aumolertinib through the blood–brain barrier, and the receptor subtype selectivity, which avoids the generation of non-selective metabolites that strongly inhibit EGFR WT during drug metabolism^12,14,15. Moreover, the hydrophobic cyclopropyl group at the 1-position of indole of aumolertinib quite matches the compact hydrophobic pocket formed in EGFR-T790M mutants, which imparts aumolertinib a better antitumor activity due to a higher binding affinity to the target protein (Fig. 1C)¹². As an EGFR-TKI, aumolertinib mainly inhibits the proliferation and growth of tumor cells by blocking the formation of EGFR dimer, inhibiting the phosphorylation of EGFR, and blocking the signal transduction of downstream MAPK and AKT pathways (Fig. 1D). In the clinical treatments of advanced NSCLC with common EGFR mutations, aumolertinib showed robust efficacy and safety^15,16. However, the potential of aumolertinib in NSCLC with uncommon EGFR mutations remains to be fully assessed.

In this study, the antitumor activity of aumolertinib was systematically evaluated in various cell and animal models bearing diverse uncommon EGFR mutations. Aumolertinib demonstrated promising potential for treating NSCLC with uncommon EGFR mutations and was worthy of further clinical investigation.

2. Materials and methods

2.1. Chemicals and reagents

Aumolertinib was provided by Hansoh Pharmaceutical Group Co., Ltd. (Shanghai, China). Afatinib and osimertinib were purchased from SelleckChem. Mobocertinib was purchased from MedChem Express. Kinase of EGFR WT and all mutants, including EGFR T790M, EGFR L858R, EGFR delE746-A750, EGFR T790M/L858R, and EGFR delE746-A750/T790M, were purchased from Cama Bioscience. Substrates and detection reagents, including ULight-polyGT, Europium-anti-phosphotyrosine (PT66), detection buffer, and opaque white OptiPlate™-384, were purchased from PerkinElmer. ATP solution, HEPES, EDTA, and MgCl₂ were purchased from Life Technologies. EGTA and Tween-20 were purchased from Sigma–Aldrich. DMEM medium and F12 medium were purchased from Gibco. RPMI-1640 medium was purchased from Hyclone. CellTiter-Glo™ Luminescent Cell Viability Assay Kit and Caspase-Glo™ 3/7 Assay Kit were purchased from Promega. BCA (Bicinchoninic Acid) Protein Assay Kit was purchased from Thermo.

2.2. Cell lines

The Ba/F3 EGFR D770-N771insSVD (SVD), Ba/F3 EGFR V769-D770insASV (ASV), Ba/F3 EGFR H773-V774insNPH (NPH), Ba/F3 EGFR A763-Y764insFQEA (FQEA), and Ba/F3 EGFR L861Q cell lines were obtained from PRECEDO (Hefei, China). Ba/F3 EGFR G719S, Ba/F3 EGFR G719S/T263P, Ba/F3 EGFR T790M/L861Q, Ba/F3 EGFR S768I, Ba/F3 EGFR D770-N771insG (G), Ba/F3 EGFR D770-N771insNPG (NPG), Ba/F3 EGFR H773-V774insH (H), Ba/F3 EGFR E709K/G719A, and Ba/F3 EGFR L747S/G719A cell lines were obtained from KYinno Biotechnology (Beijing, China). The LU0387 cell line was provided by Crown Bioscience (Taicang, China). Ba/F3 cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum. LU0387 cells were cultured in DMEM/F12 medium containing 10% FBS. All cell lines were cultured at 37 °C with 5% CO₂.

2.3. Measurement of cell-free EGFR kinase activity

Cell-free EGFR kinase activity was performed by LANCE® Ultra time-resolved fluorescence resonance energy transfer technology. Briefly, different EGFR-TKI solutions (including aumolertinib, afatinib, and osimertinib) were prepared by serial dilution and mixed with each enzyme (WT or mutant EGFR), the U-Light-PolyGT/ATP mixture (substrate and ATP in a kinase buffer) into 384-well plate. Kinase reactions were incubated at room temperature for one hour and then quenched by EDTA. The specific LANCE® Eu-W1024 anti-phosphotyrosine (PT66) was added to the reaction mixture in the LANCE® detection buffer, followed by incubation for 1 h. The LANCE® signal was recorded on the microplate reader (BioTek) at the emission wavelength of 665 nm using the time-resolved fluorescence program. The half-maximal inhibitory concentration (IC₅₀) value was determined by GraphPad Prism 9.3 software.

2.4. Cell proliferation assay

Ba/F3 cells were seeded in 96-well plates with RPMI-1640 media containing a ten-point, threefold dilution series of EGFR-TKI for 72 h. LU 0387 cells were seeded into 96-well plates, dosed with a nine-point, triple dilution series of EGFR-TKIs, and incubated for seven days. Cell viability was measured using the CellTiter-Glo assay and calculated by Eq. (1):

Cell viability (%) = T/C × 100 (1)

where, T is the luminescence readout of the experimental group treated with drugs, and C is the average luminescence readout of the control group treated with the solvent. GraphPad Prism 9.3 software was used to plot the S-type curve of the inhibition rate vs. the dosage and calculate the IC₅₀ value using the nonlinear regression model.

2.5. Immunoblot analysis

Cells in the exponential growth phase were collected and counted with the Vi-Cell XR cell counter. The cell suspension concentration was adjusted with the corresponding media, and the cells were added into the 6-well plate and cultured overnight in a 5% CO₂ incubator at 37 °C. After drug treatment, cell samples were collected, lysed, and followed centrifugation for protein extraction. The protein concentration was determined by the BCA method. Protein samples were mixed with SDS-PAGE loading buffer and boiled for 5 min. For each sample, 20 μg of protein was loaded for electrophoresis and transferred to the PVDF membrane by the wet-transfer method. Membranes were incubated with primary antibodies overnight at 4 °C. Membranes were washed with TBST solution at room temperature for 5 min each time and three times in total. After washing, membranes were incubated with HRP-labeled secondary antibodies in the dark for 1 h at room temperature. After three times of washing, the membranes were either scanned by the Odyssey imaging system at 700 or 800 nm wavelength or visualized by the Tanon 5200 chemiluminescence imaging system. Grey values of the blots were quantified by the Image J software.

2.6. Colony-forming assay

Cells (10,000) in the exponential growth phase were seeded in each well of the 6-well plate and cultured overnight. The medium was replaced with a fresh medium containing aumolertinib with different concentrations (30, 100, 300, and 1000 nmol/L). All samples were tested in triplicate, and the medium was changed twice a week. After 21 days, the culture medium was discarded, and the cells were gently washed twice with PBS. Cells were fixed in 4% paraformaldehyde and stained with 0.5% crystal violet solution (Beyotime) for 30 min at room temperature. The staining solution was discarded, and the cells were gently washed with distilled water. Plates were air dried at room temperature, colony pictures were captured, and numbers were counted.

2.7. Assessment of caspase-mediated apoptosis using caspase-3/7 assay

The caspase 3/7 activities were measured by the Caspase-Glo 3/7 assay (Promega) according to the manufacturer's instructions. Briefly, cells were seeded in a 96-well plate at the optimal density, cultured overnight, and then treated with compounds. After exposure for 24 and 48 h, 100 μL of Caspase-Glo 3/7 reagent was added to each well, followed by a gentle mixture using a plate shaker at 300 rpm for 30 s. Incubate at room temperature for 1 h; the luminescence of each sample was then measured by Envision 2104 Multilabel Reader (PerkinElmer, USA).

2.8. Computational atomistic modeling and docking simulation

2.8.1. Aumolertinib with EGFR complexes

The preparations were performed using Schrӧdinger's Maestro (version 13.0.135, Schrödinger, LLC, New York, NY, USA), including covalent docking, mutating, and homology modeling. The structure for EGFR WT in complex with aumolertinib was generated by covalent docking (PDB code: 6JXT¹⁷). EGFR L861Q or EGFR E709K/G719A in complex with aumolertinib was generated by mutating and covalent docking in the 6JXT X-ray structure. Other EGFR mutation types in complex with aumolertinib, osimertinib, or afatinib were generated by homology modeling and covalent docking in the 6JXT X-ray structure.

2.8.2. Molecular dynamics and molecular mechanics Poisson–Boltzmann surface area (MM/PBSA)

The parameters for the protein and ligand model structures were obtained from the AMBER ff14SB protein force field¹⁸ and the general AMBER force field¹⁹, respectively. Molecular dynamics simulations (MDs) were performed on the model system using GROMACS molecular dynamics software package (version 2021.4). MDs were carried out in the absence of aumolertinib covalently bound to cysteine (C797 in EGFR WT, EGFR L861Q, and EGFR E709K/G719A; C798 in EGFR H and C800 in EGFR NPH.). A 10 Å TIP3P water molecule octahedron box was set to solvate the complex system along with Na⁺ and Cl⁻ counter-ions to neutralize the system. Simulations were performed under periodic boundary conditions at a constant temperature of 300 K and pressure of 1 atm using the NVT ensemble for 100 picoseconds (ps). Next, 100 ps of NPT equilibration was conducted. Finally, the production run was 2 ns, with trajectories generated every 2 fs and saved every 2 ps. All the preliminary analyses such as root mean square deviation and conformation extraction of protein and ligand were carried out by GROMACS analysis programs. MM/PBSA was performed using AMBER Tools21 module²⁰.

2.9. Animal models

The genetically engineered Ba/F3 EGFR ASV and L861Q cell lines (1 × 10⁶ cells in 0.2 mL PBS) were inoculated into the right flank of 6–8-week-old BALB/c nude mice (GemPharmatech Co., Ltd.). After reaching optimal tumor volume (100–150 mm³), the mice were randomized into groups (six mice per group) and orally administered the drugs or vehicle daily.

For the LU0387 (Crown Bioscience) patient-derived xenografts (PDX) model, tumor fragments of human lung cancer LU0387 (2–3 mm in diameter) which expresses EGFR NPH were inoculated subcutaneously into the right abdomen of 6–7-week-old BALB/c nude mice (GemPharmatech Co., Ltd.). When the tumor size reached 150 mm³, animals were randomly grouped (five mice per group). Mice in each group were treated daily with the vehicle and 40 mg/kg aumolertinib, respectively, by intragastric administration for 26 days. The mice were euthanized 6 h after the last administration, and blood samples and tumor tissues were collected. Complete blood count and blood biochemical analysis were performed on the blood samples of nude mice using an automated hematology analyzer (Sysmex, XN-1000V) and an automated chemistry analyzer (Sysmex, BX-3010), respectively. Tumor tissue samples were added to the RIPA lysis buffer. The homogenized lysate was placed on ice for 30 min, centrifuged at 4 °C for 30 min, and collected the supernatant. Proteins were quantified by the BCA method. phospho-EGFR (pEGFR) (#3777; Cell Signaling Technology), EGFR (#54359; Cell Signaling Technology), phospho-AKT (pAKT) (#4060; Cell Signaling Technology), AKT (#9272; Cell Signaling Technology), phospho-ERK (pERK) (#4376; Cell Signaling Technology), ERK (#4695; Cell Signaling Technology), proliferating cell nuclear antigen (PCNA, 13110; Cell Signaling Technology), cleaved PARP (#9548; Cell Signaling Technology), PARP (ab227244; Abcam), cleaved caspase-3 (AF7022; affbiotech), and caspase-3 (AF6311; affbiotech) were assayed by Western blotting.

The tumor volume and body weight of mice were measured twice a week for all models. Eq. (2) used to calculate the tumor volume:

Tumor volume (mm³) = [Length (mm) × Width (mm) × Width (mm)] × 0.5 (2)

The percentage of tumor growth inhibition (TGI) was calculated according to Eq. (3):

TGI (%) = [1−((T_final−T_initial)/(C_final−C_initial))] × 100 (3)

where T represents the tumor volume in EGFR TKI-treated groups, and C represents the tumor volume in the vehicle control group.

All animal protocols were approved by the Institutional Animal Care and Use Committee of PRECEDO or Crown Bioscience.

2.10. Histology and immunohistochemistry

Hematoxylin and eosin (HE) staining was conducted according to routine protocols. Briefly, the deparaffinized and rehydrated tumor and tissue sections were stained with hematoxylin solution for 5 min and then rinsed with tap water. Then sections were treated with hematoxylin differentiation solution, rinsed with tap water, treated with hematoxylin Scott tap bluing, and rinsed with tap water. Then the sections were dehydrated in 85% and 95% gradient alcohol for 5 min each, stained with eosin solution for 5 min, and followed by dehydration with graded alcohol and clearing in xylene. The mounted slides were then examined and photographed by microscopy (Nikon Eclipse E100).

The immunohistochemistry (IHC) staining was conducted with primary antibodies of Ki-67 (Servicebio), EGFR (Servicebio), and pEGFR (CST) at 4 °C overnight, then covered with secondary antibodies (HRP-linked) of Ki-67 (Servicebio), EGFR (Servicebio) and p-EGFR (Servicebio) and incubated at room temperature for 50 min. Diaminobenzidine (DAB, Servicebio) was used to visualize the antigen–antibody reaction. The slides were photographed by microscopy (Nikon E100). ImageJ was used for the quantifications of all IHC staining.

2.11. Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay

After deparaffinization and rehydration, tumor sections were permeabilized with proteinase K for 20 min, washed with PBS (pH 7.4), and treated with the TUNEL assay kit (Servicebio). After the sections were developed by DAB chromogenic reagent (Servicebio), counterstained with hematoxylin, dehydrated, and mounted, the apoptosis of tumor tissues was detected by microscopy (Nikon E100).

2.12. Patients

The clinical cases of this work were from a retrospective study (ChiCTR2100049155), which has been approved by the Medical Ethics Committee of Union Hospital Affiliated to Tongji Medical College, Huazhong University of Science and Technology. All patients have provided written informed consent. The data was collected between January 2020 and August 2022. Computed tomography, pathologic examination, and gene detection were used to determine eligibility for this study. Inclusion criteria were as follows: age over 18 years, diagnosis of NSCLC, harboring uncommon EGFR mutations and having received administration of aumolertinib. Tumor response was assessed using computed tomography scans according to the Response Evaluation Criteria in Solid Tumor Criteria Version 1.1. The disease control rate was defined as the proportion of patients who achieved the best overall response of complete response, partial response, or stable disease was observed. The objective response rate was defined as the percentage of patients in whom the best response is determined as complete response or partial response.

2.13. Statistical analysis

GraphPad Prism 9.3 software was used for statistical analysis and plotting of the data. If not specifically noted, data are presented as the mean ± SD. Differences were analyzed using one-way ANOVA and Student's t-test, and P < 0.05 was considered statistically significant.

3. Results

3.1. Aumolertinib selectively inhibits the cell proliferation of NSCLC with uncommon EGFR mutations

As one of the third-generation of EGFR-TKIs, the in vitro kinase activity of aumolertinib was first examined by time-resolved fluorescence resonance energy transfer technology. Similar to standard osimertinib and afatinib, aumolertinib also offers IC₅₀ values ranging from 0.84 to 82.80 nmol/L, indicating the potent inhibition effect against six different EGFR kinases (Fig. 2A). Especially in L861Q, D761Y, and L747S three mutations, aumolertinib showed a ten times greater inhibition than the other mutations. Importantly, we found that the inhibition effect of aumolertinib on EGFR WT was lower than afatinib, indicating a higher selectivity of aumolertinib to use as the targeted therapy (Fig. 2A; Supporting Information Table S1).

Figure 2.

Aumolertinib selectively inhibits NSCLC with EGFR uncommon mutations in vitro. (A) Inhibitory activity of aumolertinib against different kinases with uncommon EGFR mutations. (B) Inhibitory activity of aumolertinib against A431 cells with EGFR WT and various Ba/F3 cells with uncommon EGFR mutations. Cells were exposed to aumolertinib for 72 h, and the cell viability was measured by the CellTiter-Glo assay. (C) Uncommon EGFR mutation selectivity of aumolertinib, osimertinib, and afatinib. The WT/mt ratio was determined using the IC₅₀ value shown in Supporting Information Table S2. (D) Western blot assay of Ba/F3 cells with uncommon EGFR mutations. Cells were exposed to EGFR-TKIs with indicated concentrations for 4 h. (A, B) Data from at least two independent experiments are presented as the mean ± SD. ∗P < 0.05 vs. WT. EGFR, epidermal growth factor receptor; NSCLC, non-small-cell lung cancer; TKIs, tyrosine kinase inhibitors; WT, wild type; mt, mutant; H, H773-V774insH; G, D770-N771insG; NPG, D770-N771insNPG; NPH, H773-V774insNPH; ASV, V769-D770insASV; SVD, D770-N771insSVD; FQEA, A763-Y764insFQEA; AUM, aumolertinib; OSI, osimertinib; AFA, afatinib.

Ba/F3 cell line has been a popular system to generate models with specific EGFR mutations for drug screening. Next, the anticancer effects of aumolertinib were investigated on engineered Ba/F3 cells with uncommon EGFR mutations. As a result, aumolertinib significantly reduced the viability of Ba/F3 cells with uncommon EGFR mutations, with the average IC₅₀ values ranging from 10.68 to 453.47 nmol/L, which is close to those treated with osimertinib (Supporting Information Table S2 and Fig. S1). In some specific mutations, such as FQEA, ASV, SVD, NPH, S768I, L861Q, T790M/L861Q, and G719S, the inhibitory effects of aumolertinib on cell proliferation were the most potent (Fig. 2B). Similar with the inhibition on in vitro kinase activity, aumolertinib showed a lower activity against EGFR WT cells compared with osimertinib and afatinib. It is indicated that aumolertinib is less able to inhibit EGFR in normal tissues, resulting in lower dose-limiting toxicity.

The selectivity of aumolertinib was further characterized by the ratio of wild type IC₅₀ value to uncommon-mutated type IC₅₀ value. As shown in Fig. 2C, aumolertinib showed excellent selectivity for uncommon-mutated EGFR cells, especially for the mutations such as FQEA, ASV, SVD, G, NPG, H, and NPH mutations. Therefore, in the aspect of targeted selectivity, aumolertinib was better than osimertinib and afatinib to some extent.

As a classic EGFR-TKI, the main mechanism of action of aumolertinib involves inhibiting EGFR phosphorylation. We next selected nine cell lines with the strongest inhibition of cell proliferation (as shown in Fig. 2B) to examine whether aumolertinib inhibited EGFR phosphorylation by immunoblotting. As expected, after aumolertinib treatment for 4 h, the expression levels of pEGFR in all cell lines with uncommon EGFR mutations were significantly reduced in a dose-dependent manner compared to that of the control group (Fig. 2D). Besides, aumolertinib significantly suppressed EGFR downstream signaling, including pAKT and pERK in EGFR L861Q, EGFR ASV, and EGFR NPH cells.

3.2. Predicted binding modes of aumolertinib with uncommon EGFR mutations

As the binding process of covalent inhibitors proceeds through multiple steps, including noncovalent binding and kinetics of the irreversible covalent binding, the full computational description of this process is challenging²¹. In view of the same kinetic values of irreversible covalent binding of aumolertinib with the different EGFR mutants. Therefore, we used GROMACS to simulate the dynamic behavior of aumolertinib with four representative mutants (EGFR L861Q, EGFR E709K/G719A, EGFR NPH, and EGFR H), on which aumolertinib showed distinct inhibitory activity in the cellular assay. Their non-covalent interactions were then calculated using MM/PBSA.

The binding free energies of aumolertinib to EGFR WT, EGFR L861Q, EGFR E709K/G719A, EGFR NPH, and EGFR H were predicted to be −0.3, −1.9, −0.2, −1.3 and −1.2 kcal/mol, respectively (Fig. 3A; Supporting Information Table S2). The computational results have a good correlation with their activities in the cellular assay. Especially, aumolertinib showed better inhibitory activity (IC₅₀ = 10.68–453.47 nmol/L) in EGFR mutants than EGFR WT (IC₅₀ = 596.60 nmol/L), which is consistent with the predicted binding free energy. In addition, the binding free energy of aumolertinib to EGFR L861Q and EGFR E709K/G719A were also in good agreement with their cell proliferation activity. Compared with EGFR L861Q, the alanine 719 of EGFR E709K/G719A was close to aumolertinib, so the indole-ring of aumolertinib was flipped to an uncomfortable conformation in the protein (Fig. 3B and C), resulting in decreased inhibition of aumolertinib against EGFR E709K/G719A. On the other hand, we found that the ligand binding pocket of EGFR NPH was more compact than that of EGFR H through MDs, which could lead to an improvement on the binding affinity of aumolertinib with EGFR NPH and the inhibition of cells proliferation activity (Fig. 3D).

Figure 3.

Predicted binding modes of aumolertinib with mutant EGFR. (A) Experimental inhibition (IC₅₀) and the binding free energy (ΔGcalc) values of aumolertinib in EGFR mutants. (B, C) Binding mode of aumolertinib (salmon) with EGFR L861Q (B) and EGFR E709K/G719A (C). (D) Binding mode of aumolertinib with EGFR H773-V774insNPH (green) and EGFR H773-V774insH (yellow). EGFR, epidermal growth factor receptor.

3.3. Aumolertinib inhibits cell growth of patient-derived tumor cell lines harboring uncommon EGFR mutations in vitro

As established cell lines have undergone unspecified long-term passaging in vitro, they may no longer faithfully represent the molecular heterogeneity of primary patient tumors. We next evaluated the anticancer activity of aumolertinib in patient-derived cancer cell lines with uncommon EGFR mutations. These patient-derived cell lines would better retain the molecular characteristics of patient tumors and be a better alternative to study cancer biology and drug sensitivity. In this study, the LU0387 cell line was derived from a 64-year-old female patient with NSCLC, carrying the EGFR NPH mutation. As shown in Fig. 4A, aumolertinib showed significant inhibition on the proliferation of the LU0387 cell line with an IC₅₀ value of 312 nmol/L. Consistent with the results from engineered Ba/F3 cells with uncommon EGFR mutations, aumolertinib also demonstrated significant inhibition of EGFR phosphorylation and its downstream signal markers (pERK) in the time- and dose-dependent manners (Fig. 4B and C).

Figure 4.

Aumolertinib inhibits cell growth and EGFR signal transduction in tumor cell lines derived from the NSCLC patient harboring uncommon EGFR mutation. (A) Inhibitory activity of aumolertinib on LU0387 cells. Cells were incubated with aumolertinib for seven days, and the cell viability was measured by the CellTiter-Glo assay. (B, C) Effect of EGFR-TKIs on EGFR signal transduction in the LU0387 cell line. The cells were exposed to different concentrations of EGFR-TKIs for 6 and 24 h, and the phosphorylation of EGFR, AKT, and ERK was assayed by Western blot. (D) Colony formation test of LU0387 cells. Cells were seeded and exposed to aumolertinib for 21 days. Three independent experiments were performed to determine the number of cell colonies relative to that of the DMSO control group. (E) Aumolertinib activates caspase-3/7 in LU0387 cells. The cells were treated with aumolertinib for 24 h and 48 h, and caspase-3/7 activation was determined by the Caspase-Glo 3/7 kit. DMSO was used as the control. Data are presented as the mean ± SD, n = 3; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. Abbreviation: EGFR, epidermal growth factor receptor; NSCLC, non-small-cell lung cancer; TKIs, tyrosine kinase inhibitors; NPH, H773-V774insNPH; AUM, aumolertinib; OSI, osimertinib; AFA, afatinib.

Moreover, the inhibitory effect of aumolertinib on the proliferation of LU0387 cells was assessed by colony formation assay. The results showed that 100 nmol/L of aumolertinib significantly inhibited the proliferation of LU0387 cells (Fig. 4D). In addition, whether aumolertinib induces LU0387 cell apoptosis was investigated by monitoring the activity of caspase 3/7. It was found that aumolertinib could increase the caspase-3/7 activity in LU0387 cells in a concentration-dependent manner, indicating the inhibition of the proliferation of NSCLC with uncommon EGFR mutations by aumolertinib might be associated with the induction of apoptosis (Fig. 4E; Supporting Information Fig. S2).

Taken together, these findings suggest that aumolertinib inhibits the proliferation of NSCLC with uncommon EGFR mutations by inhibiting EGFR phosphorylation and its downstream signal transduction, activating apoptotic proteases and inducing apoptosis.

3.4. Aumolertinib induces tumor growth inhibition in uncommon EGFR-mutant lung cancer allografts models in vivo

Next, to explore the in vivo antitumor activity of aumolertinib, we firstly examined the potency of aumolertinib against two mouse tumor models implanted with Ba/F3 cells genetically engineered to express human EGFR with the ASV and L861Q mutations. In the Ba/F3 EGFR ASV (Ex20ins) allografts model, after oral administration daily, aumolertinib showed significant tumor-growth inhibition within a very short period after the first administration at a dose of 40 mg/kg/day (P < 0.01 on Day 4, Supporting Information Fig. S3A). At the end of the experiment, the tumor volumes in the aumolertinib group, as well as the tumor weight, exhibited significant reductions compared to the vehicle group (P < 0.01, Fig. S3A–S3C). The overall TGI ratios of aumolertinib were 41% and 64% in the 20 mg/kg group and 40 mg/kg group, respectively. Importantly, we found that the treatment of 40 mg/kg aumolertinib could rival mobocertinib and osimertinib, achieving similar TGI ratios at around 65% with no significant differences. While in the Ba/F3 EGFR L861Q (non-Ex20ins) allograft model, aumolertinib demonstrated more dramatic effects on tumor growth inhibition (Fig. 5A–C). After the daily treatment of 40 mg/kg aumolertinib, tumor regression was achieved with a TGI ratio of 103%, which was also close to the efficacies in positive control groups using afatinib and osimertinib with no significant differences (Fig. 5B; Fig. S3A–S3C).

Figure 5.

Aumolertinib inhibits tumor growth in Ba/F3 EGFR L861Q mutant mouse allograft model in vivo. Tumor weight (A), tumor volume (B), the photograph of tumors (C), representative immunochemistry image and quantification of EGFR, p-EGFR, Ki-67, and TUNEL (D, E), and representative HE staining image of tumor tissue and main organ sections (F) in mice bearing Ba/F3 EGFR L861Q allografts. Mice were treated for 13 days with aumolertinib (20 and 40 mg/kg), afatinib (7.5 mg/kg), osimertinib (20 mg/kg), or vehicle once a day after the tumor volume reached 100 mm³. Scale bar, 25 μm (D, F). Orange arrows in (F), a number of tubular epithelial cells degenerated at the junction between the cortex and medulla with loose and pale cytoplasm. The data of (A) and (B) are presented as the mean ± SEM (n = 6/group) and compared with the vehicle (ns, no significant difference; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001). EGFR, epidermal growth factor receptor.

Aumolertinib was well tolerated with no clinical signs or drug-related mortality, and little body weight loss (less than 10% of starting body weight) was observed compared with predose starting body weight (Figs. S3D and S4). Histopathologic examination of tumor sections obtained following aumolertinib treatment using HE staining, IHC staining for EGFR, p-EGFR, and Ki-67, and TUNEL staining, further confirmed receptor, cell proliferation inhibition and engagement of apoptotic machinery in uncommon EGFR-driven tumors in vivo (Fig. 5D–F; Fig. S3E and S3F). We also carried out the histological examination of the kidney, liver, and lung of mice. HE staining results show that aumolertinib-treated mice, even the high-dose group, had no noticeable damage in these tissues compared to the vehicle group (Fig. 5F; Fig. S3F). However, some lung lesions, such as alveolar injury and minimal hemorrhage, were present in mobocertinib and osimertinib-treated mice (Fig. S3F). Besides, more renal tubular epithelial cell degeneration was observed in osimertinib and afatinib-treated mice compared with the vehicle group (Fig. 5F; Fig. S3F).

3.5. Aumolertinib demonstrates potent antitumor activity in the uncommon EGFR-mutated PDX model in vivo

Although the convincing results obtained from the above investigations on in vitro and in vivo antitumor activities of aumolertinib, these data mainly represent non-endogenous engineered systems. Therefore, we next expected to evaluate and validate its antitumor activity using a more directly relevant PDX model. In this work, LU0387 NSCLC PDX mice carrying the NPH mutation of EGFR were orally administered with aumolertinib at 40 mg/kg once daily for 26 days. Compared with the vehicle group, aumolertinib showed a significant reduction in tumor volume and weight after the treatment of 26 days, with a 90.29% TGI ratio (P < 0.05; Fig. 6A and B). Besides, aumolertinib dosed orally at 40 mg/kg daily was well tolerated with no substantial changes in weight loss compared to the weight at baseline (Fig. 6C), and no significant abnormalities in the complete blood count and the blood biochemical tests were found (Supporting Information Tables S3 and S4). Moreover, EGFR signal suppression, cell proliferation inhibition, and apoptosis induction by aumolertinib were confirmed by assessing the expression of phospho-proteins, proliferation markers, and apoptosis markers by immunoblotting. As shown in Fig. 6E and F, aumolertinib treatment significantly inhibited the phosphorylation of EGFR and concurrently decreased the expression of pERK and pAKT, indicating a reduction in EGFR signal transduction. Accordingly, aumolertinib led to decreased levels of PCNA, which has a critical role in cell proliferation²². Besides, aumolertinib treatment significantly increased the pro-apoptosis effect, as indicated by the notably higher expression of cleaved caspase-3 and cleaved PARP.

Figure 6.

Antitumor activity of aumolertinib in the PDX model harboring EGFR H773-V774insNPH mutation. Tumor volume (A), tumor weight (B), average body weights of mice at treatment start (Day 0) and study end (Day 26) (C), and the photograph of tumors (D) in a PDX LU0387 (EGFR H773_N774insNPH) NSCLC tumor model in mice over 26-day treatment with aumolertinib. The vehicle was used as the control. (E, F) Expression of pEGFR, EGFR, pAKT, AKT, pERK, ERK, PCNA, cleaved PARP, PARP, cleaved caspase-3, and caspase-3 in tumor tissue after administration of EGFR-TKIs. Tumors were obtained and lysed at 6 h after the last treatment. Data are presented as the ratio relative to the vehicle group. The data of (A), (B), (C), and (F) are presented as the mean ± SEM (n = 6/group) and compared with the vehicle (∗P < 0.05; ∗∗P < 0.01). PDX, patient-derived xenografts; EGFR, epidermal growth factor receptor; TKIs, tyrosine kinase inhibitors; NPH, H773-V774insNPH.

Taken together, these data demonstrate that aumolertinib is safe, well tolerated, and has a proven inhibitory effect on the growth of tumors harboring uncommon EGFR mutation by suppressing EGFR signal transduction, resulting in cell proliferation inhibition and apoptosis induction.

3.6. Aumolertinib showed potent antitumor activity in uncommon EGFR-mutated NSCLC patients

At last, we performed an observational case series study on the treatment of uncommon EGFR-mutated advanced NSCLC with aumolertinib, in which 22 patients were included. The baseline sociodemographic characteristics are shown in Table 1. The median age of the patients was 63.5 years (range 49–75), 63.6% were male, 95.5% had ECOG ≤ 1, 100% were lung adenocarcinoma, 86.4% were stage Ⅳ, and 40.9% had brain metastasis. Uncommon EGFR mutation types include Ex20ins (27.3%), G719X (13.6%), L861Q (27.3%), S768I (4.5%), and double mutations (27.3%). Aumolertinib was administered in 14 patients as first-line therapy and in 4 patients as second-line. Nine patients (40.9%) received aumolertinib as monotherapy, among whom a patient was treated with 1.5 times the regular dose of 165 mg. Other 13 patients (59.1%) received combination drug therapy with chemotherapy, bevacizumab, or anlotinib.

Table 1.

Characteristics of patients.

Characteristic	Total (n = 22)
Gender
Male	14 (63.6)
Female	8 (36.4)
Age (years)
Median (range)	63.5 (49–75)
ECOG performance status
0–1	21 (95.5)
2	1 (4.5)
Histology
Adenocarcinoma	22 (100.0)
Brain metastasis
Yes	9 (40.9)
No	13 (59.1)
Stage prior therapy
Ⅲ	3 (13.6)
Ⅳ	19 (86.4)
EGFR mutation types
Ex20ins	6 (27.3)
G719X	3 (13.6)
L861Q	6 (27.3)
S768I	1 (4.5)
Double mutations	6 (27.3)
G719X + S768I	3 (13.6)
G719X + L861Q	1 (4.5)
S768I + L858R	1 (4.5)
S768I + V774M	1 (4.5)
Aumolertinib as line of treatment
First	14 (63.6)
Second	4 (18.2)
Beyond second	4 (18.2)
Aumolertinib dosage per day
110 mg (regular dose)	21 (95.5)
165 mg	1 (4.5)
Drug combination
No	9 (40.9)
Yes	13 (59.1)
Chemotherapy	9 (40.9)
Bevacizumab	2 (9.1)
Chemotherapy + Bevacizumab	1 (4.5)
Anlotinib	1 (4.5)

Data are n (%) unless otherwise indicated. Abbreviation: ECOG, Eastern Cooperative Oncology Group; EGFR, epidermal growth factor receptor; Ex20ins, Exon 20 insertion.

Of the evaluated patients, 13 (59.1%) achieved partial response, 6 (27.3%) stable disease, and 3 (13.6%) progressive disease (Table 2; Supporting Information Table S5). The objective response rate and the disease control rate were 59.1% and 86.4%, respectively (Table 2). Representative imaging responses are shown in Fig. 7. According to the available data, the maximum progression-free survival was up to 15 months. No patients suffered unmanageable adverse events during treatment, and no treatment-related dose reductions or discontinuations, even at the high dose of 165 mg daily.

Table 2.

Antitumor activity of aumolertinib in patients with uncommon EGFR-mutated NSCLC.

Response	Total (n = 22)
Best response, n (%)
Partial response	13 (59.1)
Stable disease	6 (27.3)
Progressive disease	3 (13.6)
ORR, % (95% CI)	59.1 (38.5 to 79.6)
DCR, % (95% CI)	86.4 (72.0 to 100.7)

Abbreviation: EGFR, epidermal growth factor receptor; NSCLC, non-small-cell lung cancer; ORR, objective response rate; DCR, disease control rate.

Figure 7.

Aumolertinib showed effective antitumor activity in NSCLC patients with uncommon EGFR mutations. (A) Radiological responses of a 62-year-old female patient with EGFR L861Q mutant stage IV lung adenocarcinoma after 16-week treatment of aumolertinib 110 mg orally once daily combined with chemotherapy (700 mg pemetrexed disodium, 500 mg carboplatin). (B) Radiological responses of a 74-year-old male patient with EGFR D770-N771insSVD (Ex20ins) mutant stage IV lung adenocarcinoma after 16-week treatment of aumolertinib 110 mg orally once daily. The Blue arrow indicates the tumor lesion. EGFR, epidermal growth factor receptor; NSCLC, non-small-cell lung cancer; Ex20ins, Exon 20 insertion.

4. Discussion

Compared with NSCLC patients with common EGFR mutations, patients harboring uncommon EGFR mutations usually have a poor prognosis with limited treatment options, except mobocertinib and amivantamab have been approved by the FDA for the Ex20ins subgroup²³. Most EGFR uncommon mutation types are insensitive to the first-generation and second-generation EGFR-TKIs⁵. Although afatinib has been approved for the first-line treatment for NSCLC with uncommon EGFR mutations, afatinib only shows a certain level of activity against L861Q, G719X, and S768I mutations²⁴. Nevertheless, afatinib is highly toxic, and the incidence of treatment-related adverse events is significantly higher than that of the other EGFR-TKIs²⁵. Despite their outstanding performance, mobocertinib and amivantamab are of limited use due to their cost and lack of availability. Therefore, treatment for uncommon EGFR-mutated NSCLC is still a significant clinical challenge, and more work is required to develop more effective, economical, and accessible EGFR-targeting therapies.

Aumolertinib is a novel third-generation EGFR-TKI with an innovative molecular structure. It has been approved in China for common EGFR mutations, such as Ex19del and L858R, as well as the drug-resistant mutation T790M^15,16. For patients with EGFR-mutant advanced NSCLC, aumolertinib represents an effective treatment alternative, particularly in low-to middle-income nations; even in high-income countries, it plays a vital role as a potential cost disruptor¹¹. In this study, we intend to further evaluate the activity of aumolertinib in NSCLC with uncommon EGFR mutations to expand its application in lung cancer treatment to benefit more patients.

In this study, multiple preclinical models were employed to examine the antitumor activity of aumolertinib in NSCLC with uncommon EGFR mutations. The results showed that aumolertinib inhibited the proliferation of uncommon EGFR-mutated cells by inhibiting phosphorylation of EGFR and its downstream molecules, including AKT and ERK (Fig. 2). Molecular dynamics simulations showed that the ligand binding pocket of EGFR NPH was more compact, which improved the binding affinity of aumolertinib to NPH and enhanced the inhibition of cell proliferation (Fig. 3D). In the LU0387 cell line harboring the EGFR NPH mutation, we also observed that aumolertinib inhibited EGFR phosphorylation, downstream signal transduction, cell proliferation, and induced apoptosis (Fig. 4). A consistent mechanism was also observed in the in vivo model of LU0387 (Fig. 6E and F). According to the Cosmic database, among all uncommon EGFR mutations, S768I, L861Q, G719X, FQEA, ASV, SVD, G, and NPH mutations accounted for more than 50%. In our preclinical NSCLC models, aumolertinib effectively inhibited all these mutations, as well as compound mutations, including G719S/T263P. At last, a case series study was performed to investigate the potential efficacy of aumolertinib in patients with advanced uncommon EGFR mutant NSCLC. In terms of efficacy, 19 (86.4%) of 22 patients achieved partial response or stable disease after the administration of aumolertinib-based therapy.

Considering that aumolertinib belongs to one of the EGFR-TKIs, afatinib and osimertinib were used as controls for cell experiments. In the cell proliferation assays, aumolertinib showed a higher selectivity than afatinib or osimertinib, especially for the Ex20ins mutants (Fig. 2C). The free energy of binding calculated by the MM/PBSA method was consistent with the results of the cell experiments, indicating a better inhibitory activity of aumolertinib in EGFR mutants than that in EGFR WT (Fig. 3). Similarly, in the murine animal models, aumolertinib could also suppress tumor growth and rival those positive control drugs without any adverse reactions such as weight loss, abnormal changes in blood and biochemical indexes or tissue injury occurred in the high- and low-dose groups during treatment (Figure 5, Figure 6; Fig. S3). In contrast, certain toxic reactions were observed in afatinib, osimertinib, or mobocertinib groups during treatment, such as renal and lung lesions in the osimertinib group, and lung lesions in the mobocertinib group. In addition, problems such as diarrhea and renal lesions occurred in the afatinib group. Thus, compared with drugs with low selectivity, aumolertinib may have lower dose-limiting toxicity related to EGFR inhibition in normal tissues (e.g., skin, lung, and gastrointestinal tract).

With the innovative structure design, aumolertinib shows better metabolic stability to reduce the production of non-selective metabolites, which generally induced nonspecific inhibition on EGFR WT to cause undesirable toxicity^14,15. In terms of the in vitro and in vivo results in this study, the selectivity effect on EGFR WT of aumolertinib was superior to the other EGFR-TKIs, and fewer toxic reactions were observed in aumolertinib than those in afatinib, osimertinib, or mobocertinib on the murine animal models. In the observational case series study, aumolertinib also showed an acceptable safety profile and tolerance, even up to 1.5 times the regular dose, which is consistent with the mice data. In the context of similar efficacy, a better safety profile would be crucial for clinicians to select the optimal treatment agent for NSCLC patients with uncommon mutations. Therefore, results in our study suggest that aumolertinib is a promising alternative third-generation EGFR-TKI for NSCLC patients harboring uncommon mutations. While our observational study showed promising efficacy and tolerance of aumolertinib, more rigorous clinical trials need to be conducted to validate its clinical efficacy and safety profile in uncommon EGFR mutant NSCLC, with a particular focus on drug combinations to determine the best treatment regimen. It should be noted that a multicenter, open-label, phase 3 clinical trial of aumolertinib in patients with uncommon EGFR-mutated advanced NSCLC (NCT04951648) is in progress.

5. Conclusions

In conclusion, our study shows that aumolertinib significantly inhibits tumor cell proliferation and induces cell apoptosis both in vitro and in vivo by inhibiting EGFR phosphorylation and its downstream signaling pathways in various NSCLC preclinical models with uncommon EGFR mutations. Notably, the favorable efficacy of aumolertinib in the clinical case series investigation demonstrates a good clinical relevance to the results obtained from pre-clinical models. Our findings support further clinical evaluation of aumolertinib as a promising effective drug candidate for patients harboring uncommon EGFR mutations, an area of high unmet medical need.

Author contributions

Chen Shi: conceptualization, writing-original draft, writing-review & editing, project administration. Cong Zhang: methodology, investigation, formal analysis, writing-original draft. Zhiwen Fu: methodology, formal analysis, data curation. Jinmei Liu: methodology. Yuanfeng Zhou and Bao Cheng: resources. Cong Wang and Shijun Li: investigation, visualization. Yu Zhang: conceptualization, writing-review & editing, supervision. All authors have read and approved the final manuscript.

Conflicts of interest

Yuanfeng Zhou and Bao Cheng are employees of Shanghai Hansoh Biomedical Co., Ltd. No potential conflicts of interest were disclosed by the other authors.

Appendix A. Supplementary data

The following is the supplementary data to this article: