Combined therapy with ALK inhibitors and a novel EGFR-targeting antibody‒drug conjugate as a strategy for ALK-rearranged non–small cell lung cancers

Abstract

Non-small cell lung cancer (NSCLC) remains the leading cause of cancer-related death. Targeted monotherapies against the actionable oncogenic drivers, including anaplastic lymphoma kinase (ALK) rearrangement and epidermal growth factor receptor (EGFR) dysregulation, are well-established for molecularly selected NSCLC patients. However, therapeutic resistance remains a major clinical challenge, and combinatorial strategies targeting distinct signaling pathways may overcome the limitations of single-agent targeted therapies. LR004-VC-MMAE, a novel EGFR-targeting antibody‒drug conjugate (ADC), shows potent antitumor efficacy in multiple EGFR-positive xenografts, yet its therapeutic potential in ALK-rearranged NSCLC remains largely unexplored. Here, we evaluated a novel combinatorial strategy using ALK inhibitors (crizotinib, ceritinib, alectinib, and lorlatinib) in combination with LR004-VC-MMAE in ALK-rearranged NSCLC models. We found that both ALK and cell-surface EGFR were highly expressed in ALK-rearranged NCI-H3122 and NCI-H2228 cell lines but were expressed at low levels in the ALK-wild-type NCI-H460 cell line. All ALK inhibitor-ADC regimens synergistically suppressed proliferation in ALK-rearranged NSCLC cells, while exerting little synergy in ALK wild-type and EGFR-low counterparts. Moreover, crizotinib or ceritinib combined with LR004-VC-MMAE induced significantly greater apoptosis in NCI-H3122 cells compared with either single agent. Mechanistically, the crizotinib-LR004-VC-MMAE combination markedly downregulated ALK and EGFR expression and triggered apoptosis via mitochondrial depolarization-mediated caspase cascade activation and endoplasmic reticulum stress-induced unfolded protein response. In NCI-H3122 xenografts, this combination achieved synergistic tumor inhibition without increased toxicity compared with either treatment alone. Collectively, our findings demonstrate the synergistic antitumor activity of ALK inhibitor plus LR004-VC-MMAE against ALK-rearranged, EGFR-high NSCLC, providing a mechanistic rationale for further clinical development of this combination strategy.

Introduction

Lung cancer, a very aggressive and highly prevalent disease worldwide, can be categorized into two types: small-cell lung cancer and non-small cell lung cancer (NSCLC), with almost 2.5 million new cases and an estimated 1.8 million deaths reported in 2022 [1, 2]. NSCLC, accounting for 85–90% of all lung cancers, is the most common histological subtype and is characterized by high morbidity and mortality [3]. Early-stage NSCLC patients frequently miss optimal treatment windows because of the diseaseˊs asymptomatic progression [4]. Nearly seventy-five percent of NSCLC patients are diagnosed at an advanced stage, resulting in a 5-year survival rate of 4–17% [5]. Standard NSCLC treatments include surgery, radiotherapy, chemotherapy, immunotherapy, and targeted therapy [6]. Among them, targeted therapy has made significant advances at an astonishing rate of development, bringing new hopes for patients with advanced NSCLC [7, 8]. To date, the US Food and Drug Administration (FDA) has approved multiple targeted therapies for NSCLC directed against well-characterized oncogenic drivers, including anaplastic lymphoma kinase (ALK) rearrangement and epidermal growth factor receptor (EGFR) dysregulation [8–10]. However, substantial unmet needs remain for NSCLC patients worldwide, as targeted monotherapies often face critical challenges in inadequate efficacy, dose-limiting toxicity, and inevitable drug resistance. Therefore, multimodal combination therapies targeting distinct signaling pathways are increasingly recognized in oncology for their multifunctional benefits to overcome the challenges of single-agent targeted therapies [11].

ALK rearrangement and EGFR dysregulation represent well-characterized oncogenic drivers in lung cancer [9, 12]. ALK is a transmembrane receptor tyrosine kinase within the insulin receptor superfamily that mediates proliferative signaling pathways through ligand-dependent activation (via ALK and LTK ligand 1/2 binding), including Ras-extracellular signal-regulated kinase pathway, Janus kinase 3-signal transducer and activator of transcription 3 pathway, and phosphatidylinositol 3-kinase pathway [13, 14]. Approximately 3–5% of NSCLC patients harbor ALK rearrangements, which often cause various malignancies, with echinoderm microtubule-associated protein-like 4 (EML4) as the most common fusion partner [14–16]. Unlike ALK, EML4-ALK facilitates ligand-independent dimerization of ALK kinase domains via the coiled-coil domain within EML4, thereby promoting oncogenic cell proliferation and ultimately leading to tumor progression through sustained downstream signaling activation [17]. Several ALK inhibitors, including crizotinib, ceritinib, alectinib, ensartinib, brigatinib and lorlatinib, have become the main treatments for advanced ALK-rearranged NSCLC [18]. EGFR, a proto-oncogene in the erythroblastosis oncogene B receptor tyrosine kinase family, represents the first validated therapeutic target in NSCLC [19]. EGFR dysregulation manifests in more than 60% of NSCLC cases through either a kinase-activating mutation in EGFR or overexpression of wild-type EGFR, driving tumor progression via constitutive activation of proliferation, survival, and metastatic pathways [20, 21]. Current therapeutic strategies for patients with EGFR dysregulation involve two primary mechanistic categories: small molecule tyrosine kinase inhibitors (TKIs, e.g., gefitinib, erlotinib, afatinib, dacomitinib, and osimertinib) and anti-EGFR antibody‒drug conjugates (ADCs, e.g., becotatug vedotin) [22, 23]. Taken together, significant advances have been achieved in ALK- or EGFR-targeted therapies. However, the treatment efficacy of these single-agent targeted therapies remains substantially limited by both intrinsic and acquired resistance mechanisms, with durable clinical responses observed in fewer than 25% of NSCLC patients receiving current targeted therapies [24, 25].

ADCs are an emerging class of novel therapeutics for NSCLC [26]. Developing rational ADC-based combinations represents a key objective in optimizing ADC-based treatment regimens due to the emergence of drug resistance following ADC monotherapy [27]. LR004-VC-MMAE, a novel EGFR-targeting ADC developed by our group, is composed of the humanized IgG₁ monoclonal antibody (mAb) LR004, enzyme-cleavable peptide valine‒citrulline (VC) linkers, and the microtubule inhibitor monomethyl auristatin E (MMAE) [28]. Our in vitro studies have demonstrated nanomolar-potency cytotoxicity of LR004-VC-MMAE against multiple EGFR-expressing malignancy cell lines derived from esophageal squamous carcinoma (ESCC, KYSE520 and KYSE150), epidermoid carcinoma (A431), ALK wild-type NSCLC (HCC827, A549 and NCI-H1975), breast cancer (MDA-MB-468, BT-549, MDA-MB-231 and MCF-7), and pancreatic cancer (AsPC-1) [28, 29]. In vivo studies have also demonstrated potent tumor growth suppression by LR004-VC-MMAE across multiple EGFR-positive xenograft models, including the KYSE 520, A431 and triple-negative breast cancer (TNBC) xenografts [28, 29], suggesting that this ADC is a promising therapeutic for EGFR expression-driven tumors. Nevertheless, critical knowledge gaps persist concerning the therapeutic potential of this ADC for ALK-rearranged NSCLC models, especially when it is combined with ALK inhibitors.

In this study, we developed a novel combination strategy via pairing ALK inhibitors (crizotinib, ceritinib, alectinib and lorlatinib) with a novel EGFR-targeting LR004-VC-MMAE, assessed its synergistic antitumor effects, and further elucidated its dual pro-apoptotic mechanisms by activating both mitochondrial/caspase signaling and endoplasmic reticulum (ER) stress-mediated unfolded protein response (UPR) in ALK-rearranged NSCLC models.

Materials and methods

Cell lines and reagents

The ALK-rearranged NSCLC cell lines (NCI-H3122 and NCI-H2228) were commercially sourced from Jinyuan Biology (Shanghai, China). The ALK-wild-type NSCLC cell line (NCI-H460) was maintained in our laboratory. All NSCLC cell models were cultured in RPMI 1640 medium (BasalMedia, Shanghai, China, cat# L210KJ) supplemented with 10% fetal bovine serum (FBS, Sorfa, cat# SX1101) and 1% penicillin/streptomycin (Solarbio, Beijing, China, cat# P1400), and maintained at 37 °C in a humidified incubator (HERA Cell 150i, Thermo Fisher, Waltham, MA, USA) containing 5% CO₂ and 95% air. ALK inhibitors crizotinib (cat# 1019151) and ceritinib (cat# A13238) were purchased from J&K Scientific (Beijing, China) and AdooQ Bioscience (Nanjing, China), respectively. Additional ALK inhibitors, including alectinib (cat# S5232) and lorlatinib (cat# S7536), were obtained from Selleck Chemicals (Beijing, China). MMAE (cat# RSP03215) was acquired from Shanghai Resuperpharmtech Co., Ltd. LR004-VC-MMAE was freshly prepared by our research group.

Western blot

Western blot analysis was performed to determine protein expression profiles, with methodological adjustments to our established procedures [30]. Cells were sonicated and lysed on ice for 30 minutes (min) in radioimmunoprecipitation assay lysis buffer (Solarbio, cat# R0010) containing 1× proteinase/phosphatase inhibitor cocktail (NCM Biotech, cat# P002). Lysates were centrifuged at 14000 revolutions per minute for 30 min, and the supernatants were collected for protein quantification using a Pierce BCA kit (Thermo Scientific, cat# 23225). Total proteins (20 μg/lane) were resolved by sodium dodecyl sulfate‒polyacrylamide gel electrophoresis and transferred onto PVDF membranes (Millipore, cat# IPVH00010). Membranes were blocked with 5% nonfat milk in tris-buffered saline with 0.05% Tween 20 (TBST) and separately probed overnight at 4 °C with primary antibodies (1:1000 dilution) directed against phospho-ALK (Y1507) (CST, cat# 14678), ALK (CST, cat# 3633), phospho-EGFR (CST, cat# 3777), EGFR (Proteintech, cat# 66455-1-Ig), Bcl-2 (CST, cat# 15071), Bax (Proteintech, cat# 66455-1-Ig), cytochrome c (CST, cat# 4280), Caspase-3 (CST, cat# 14420), Caspase-7 (CST, cat# 12827), PARP (CST, cat# 9542), inositol-requiring enzyme 1 (IRE1, Affinity, cat# DF7709), phospho-IRE1 (S724) (Affinity, cat# AF7150), PKR-like endoplasmic reticulum kinase (PERK, Selleck, cat# F0027), phospho-PERK (T982) (Affinity, cat# DF7576), activating transcription factor 6 α (ATF6, Affinity, cat# DF6009) and phospho-ATF6 (T166) (Affinity, cat# AF4507). β-tubulin (CST, cat# 2146,) or β-actin (Proteintech, cat# 66009-1-Ig) served as a loading control. After three TBST washes, membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit (ZSGB-BIO, Beijing, China, cat# ZB-2301) or anti-mouse (ZSGB-BIO, Beijing, China, cat# ZB-2305) IgG at a 1:2000 dilution for 2 hours (h) at room temperature (RT). Protein bands were visualized using Tanon™ High-sig ECL Western Blotting Substrate (Shanghai Tanon, cat# 180-5001) and imaged with a Tanon 5200 system (cat# LAB-0002-0007-SHTN). Band intensities were quantified via ImageJ software (version 1.53e).

Cell viability assay

Cell viability was measured via a cell counting kit-8 (CCK-8, Dojindo Laboratories, cat# CK04), a water-soluble tetrazolium salt-based assay for sensitive detection of cell proliferation and cytotoxicity. NSCLC cells were seeded in 96-well plates and cultured overnight at 37 °C for adherence. The cells were then treated with serial dilutions of ALK inhibitors (crizotinib, ceritinib, alectinib and lorlatinib), LR004-VC-MMAE/MMAE, or their combinations, with blank (cell-free) and untreated control wells included for normalization. Following 72 h of drug exposure, the supernatant from each well was removed and replaced with 100 μL of serum-free RPMI 1640 medium containing 10 μL of CCK-8 solution. After 30–120 min of incubation at 37 °C to allow for formazan formation by cellular dehydrogenases, the optical density (OD) at 450 nm was read via a SpectraMax i3x microplate reader (Molecular Devices, San Jose, CA, USA). Cell viability was calculated according to the following formula, and the half-maximal inhibitory concentration (IC₅₀) values for the main test agents were determined using GraphPad Prism software (version 10.4.1).

$$\mathrm{Cell}\mathrm{viability}\left(\%\right)=\frac{{OD}_{Treated}-{OD}_{Blank}}{{OD}_{Control}-{OD}_{Blank}}\times 100$$

Drug combination effects were quantitatively assessed via two different approaches. For the multi-dose experiments, the combination index (CI) was calculated via Compusyn software (Combosyn, Inc., Paramus, New Jersey, USA), which implements the Chou‒Talalay CI method on the basis of median-effect principles, where CI > 1.1 indicates antagonism, 0.9 ≤ CI ≤ 1.1 represents an additive effect, 0.3 ≤ CI < 0.9 denotes synergism, and CI < 0.3 indicates strong synergism. In single-dose experiments, drug interactions were evaluated via the coefficient of drug interaction (CDI), where a CDI > 1 indicates an antagonistic effect, a CDI = 1 represents an additive effect, and a CDI < 1 indicates a synergistic effect [31].

Apoptosis assay

Apoptosis was quantified via an Annexin V-FITC/PI Apoptosis Detection Kit (MultiSciences, cat# AP101). Briefly, NCI-H3122 cells (4 × 10⁵ cells/well in 6-well plates) were cultured overnight and then treated with ALK inhibitors (crizotinib and ceritinib), LR004-VC-MMAE, or their combinations for 72 h. All cells were harvested via centrifugation or trypsinization and then immediately resuspended in 1×Annexin V binding buffer. The samples were simultaneously stained in the dark with the working solutions containing Annexin V-FITC and PI for 15 min at RT. Analysis was performed via an Agilent NovoCyte flow cytometer equipped with NovoCyte software (version 1.5.6).

In vivo experiment

Female NOD/ShiLtJGpt-Prkdc^em26Cd52Il2rg^em26Cd22/Gpt (NCG, #T001475) mice aged 6–8 weeks were obtained from GemPharmatech Co., Ltd. (Beijing, China). The animal experimental protocol was approved by the Ethics Committee of the Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences & Peking Union Medical College (Approval No. IMB-202307011D603). All procedures strictly adhered to the ARRIVE guidelines for preclinical study reporting, ensuring humane treatment and standardized experimental practices throughout the investigation [32]. To establish NCI-H3122 xenografts, 5 × 10⁶ cells suspended in 200 μL of a 15% Matrigel matrix (BD, NJ, USA, cat# 354262)/PBS solution were subcutaneously injected into the right flank of each NCG mouse. When the average tumor volume reached approximately 80 mm³, the tumor-bearing mice were randomized into four treatment groups (n = 5): (1) control (sterile PBS), (2) crizotinib monotherapy, (3) LR004-VC-MMAE monotherapy, and (4) combination therapy with crizotinib plus LR004-VC-MMAE. Blinding was not performed in this animal study. The mice in the LR004-VC-MMAE group were intravenously dosed at 5 mg/kg once every 5 days for a total of 5 injections, whereas those in the crizotinib group were gavaged daily at a dose of 15 mg/kg on a 5-day-on, 2-day-off schedule for a total of 21 doses. The control mice received equivalent volumes of sterile PBS on the same schedule. Body weights were measured every 3 days, and tumor growth was concurrently monitored via caliper measurements of tumor length (a) and width (b). The tumor volume (V) was calculated via the following standard formula: V = $\frac{1}{2}$×a×b².

Statistical analysis

Data are expressed as the mean ± standard deviation (SD) from at least 3 independent experiments. Statistical analyses were performed using GraphPad Prism Software (version 10.4.1). Comparisons between two groups were conducted using a two-tailed unpaired Mann Whitney test. Statistically significant differences among multiple groups were assessed via one-way analysis of variance (ANOVA) or two-way ANOVA, followed by Tukey′s multiple comparison test. A P value of less than 0.05 was considered statistically significant.

Results

Expression levels of ALK and EGFR in lung adenocarcinoma patient samples and NSCLC cell lines

We analyzed the transcript levels of ALK and EGFR in lung adenocarcinoma patient specimens from the publicly accessible The Cancer Genome Atlas (TCGA) database. No significant correlation was observed across the entire cohort. Notably, five ALK-rearranged patient samples were found, all of which exhibited high EGFR expression together with ALK overexpression (Supplementary Fig. 1A). We also analyzed transcript-per-million (TPM) data of NSCLC cell lines from the Cancer Cell Line Encyclopedia (CCLE) database. Using a cutoff of log₂(TPM + 1) ≥ 4 for high expression, 77.5% of 142 NSCLC cell lines exhibited high EGFR transcript levels. Among these EGFR-high cell lines, five were ALK-rearranged (NCI-H3122, SNU2535, SNU2292, HCC366 and NCI-H2228), with NCI-H3122 showing the lowest EGFR expression (Supplementary Figs. 1B, C). Subsequent investigations into cell-surface EGFR expression by flow cytometry revealed significantly greater EGFR expression on the cell surface of both ALK-rearranged cell lines (NCI-H3122 and NCI-H2228) than did the ALK-wild-type NCI-H460 cells (Supplementary Figs. 1D–G). We further examined ALK expression in the aforementioned NSCLC cell lines by western blot analysis. Distinct expression patterns were observed between ALK-rearranged and ALK-wild-type cells: ALK-rearranged cell lines showed prominent overexpression of both total ALK and phospho-ALK (Y1507), whereas NCI-H460 cells (ALK-wild-type) displayed minimal ALK expression (Supplementary Figs. 1H–K). This dual overexpression pattern of ALK and cell-surface EGFR in ALK-rearranged NSCLC cellular models suggests that they may act as complementary molecular targets, and their synergistic targeting is expected to provide a rationale therapeutic strategy for ALK-rearranged NSCLC with EGFR-overexpression.

The combination of ALK inhibitors with LR004-VC-MMAE exhibited synergistic antiproliferative activity in ALK-rearranged and EGFR-overexpressing NCI-H3122 and NCI-H2228 cells

To assess the therapeutic potential of LR004-VC-MMAE and ALK inhibitors (crizotinib and ceritinib) in NSCLC cells with different ALK statuses, a dose‒response analysis was first performed via CCK-8 assays after 72 h of drug exposure. All three single agents exhibited dose-dependent cytotoxicity across ALK-wild-type NCI-H460 and ALK-rearranged NCI-H3122 and NCI-H2228 cells, with notably increased potency in ALK-rearranged cells (Fig. 1A; Supplementary Tab. 1). Assessments of combined treatments with ALK inhibitors (crizotinib, ceritinib) and LR004-VC-MMAE revealed distinct synergistic antiproliferative effects: in most combinations, no synergy (additive or antagonistic effects) occurred in NCI-H460 cells, whereas pronounced synergy (CI ≤ 0.8) emerged in both ALK-rearranged cell lines (Figs. 1B–G). Specifically, in NCI-H3122 cells, the cell viability decreased from 52.1 ± 4.2% (80 nM crizotinib alone) and 67.9 ± 11.7% (5 nM LR004-MMAE alone) to 7.8 ± 0.2% (CI = 0.035) with the combination treatment and from 32.8 ± 1.8% (30 nM ceritinib alone) and 58.9 ± 4.5% (5 nM LR004-VC-MMAE alone) to 8.0 ± 0.6% (CI = 0.246) with the cotreatment. In NCI-H2228 cells, the cell viability decreased from 20.0 ± 0.5% (4000 nM crizotinib alone) and 52.0 ± 6.5% (5 nM LR004-VC-MMAE alone) to 12.6 ± 0.2% (CI = 0.658) in combination and from 43.4 ± 0.6% (1000 nM ceritinib alone) and 64.7 ± 4.7% (5 nM LR004-VC-MMAE alone) to 21.3 ± 0.8% (CI = 0.102) with the cotreatment. Similarly, in ALK-wild-type NCI-H460 cells, only a few combinations of ALK inhibitors (alectinib, lorlatinib) with LR004-VC-MMAE exhibited weak synergistic effects, with CI values near the 0.9 synergism boundary, and most other combinations yielded CI values above 0.9; In contrast, in two ALK-rearranged NSCLC cell lines (NCI-H3122 and NCI-H2228), most CI values obtained from the combination treatments were less than 0.9, demonstrating robust synergistic antiproliferative effects specifically in ALK-rearranged rather than ALK-wild-type cells (Supplementary Fig. 2). To investigate whether the observed synergy stems from the MMAE payload, we performed co-treatments of crizotinib/lorlatinib with free MMAE. Compared with the synergistic effects observed for ALK inhibitor combined with LR004-VC-MMAE (Figs. 1D, F, and Supplementary Figs. 2E, G), attenuated or absent synergy was detected in ALK-rearranged NCI-H3122 and/or NCI-H2228 cells (Supplementary Fig. 3), suggesting that the synergy between crizotinib or lorlatinib in combination with LR004-VC-MMAE was partially driven by free MMAE under defined experimental conditions. We next knocked down EGFR expression in ALK-rearranged NCI-H3122 and NCI-H2228 cells using EGFR-targeting shRNA lentivirus (Supplementary Figs. 4A, B). CCK-8 assays (Figs. 1D, F; Supplementary Figs. 4C, D) showed that the synergistic antiproliferative effect of crizotinib combined with LR004-VC-MMAE was attenuated in EGFR-knockdown NCI-H3122 and NCI-H2228 cells relative to their parental counterparts. These results suggest that EGFR upregulation contributes to the synergistic antiproliferative activity of crizotinib combined with LR004-VC-MMAE in ALK-rearranged NSCLC cells.

Fig. 1. Cytotoxicity of both ALK inhibitors (crizotinib, ceritinib) and EGFR-targeting LR004-VC-MMAE toward various NSCLC cell lines.

A Antiproliferative activity of single agents. NCI-H460, NCI-H3122 and NCI-H2228 cells were treated with serial dilutions of single agents for 72 h, followed by a CCK-8 assay. The cell viability (normalized to that of the untreated controls) and IC₅₀ values were calculated via GraphPad Prism 10.4, with the data presented as the mean ± SD (n = 3). B, C Nonsynergistic cytotoxicity of ALK inhibitors (crizotinib and ceritinib) plus LR004-VC-MMAE in ALK-wild-type NCI-H460 cells, as determined by a CCK-8 assay across most dose combinations. The cell viability and CI values were calculated according to the descriptions in the text. D–G Synergistic inhibitory effects of ALK inhibitors (crizotinib and ceritinib) plus LR004-VC-MMAE on ALK-rearranged NCI-H3122 and NCI-H2228 cells. CI > 1.1 indicates antagonism, 0.9 ≤ CI ≤ 1.1 represents an additive effect, 0.3 ≤ CI < 0.9 denotes synergism, and CI < 0.3 indicates strong synergism. All of these experiments were performed in triplicate. CRI crizotinib, CER ceritinib, LR-ADC LR004-VC-MMAE.

ALK inhibitors (crizotinib and ceritinib) plus LR004-VC-MMAE enhanced apoptosis of ALK-rearranged NCI-H3122 cells

To examine whether the synergistic antiproliferative responses induced by the combinations of ALK inhibitors (crizotinib, ceritinib) with LR004-VC-MMAE were due to enhanced apoptosis, flow cytometry analysis was conducted using the Annexin V-FITC/PI Apoptosis Detection Kit in ALK-rearranged NCI-H3122 cells. As shown in Fig. 2 and Supplementary Fig. 5, single-agent treatments induced dose-dependent apoptotic responses: crizotinib (40–80 nM) induced 16.6–34.4% apoptosis, LR004-VC-MMAE (0.5–1 nM) triggered 38.4–48.9% apoptosis, and ceritinib (10–30 nM) caused 20.2–44.8% apoptosis. Notably, combination therapies yielded substantially higher apoptotic rates than either monotherapy did. Crizotinib (40–80 nM) combined with 0.5 nM LR004-VC-MMAE resulted in 55.4–73.8% apoptosis, whereas the crizotinib-LR004-VC-MMAE (1 nM) combinations resulted in 66.9–79.7% apoptotic cells. The most potent apoptotic response was observed following combination treatment with ceritinib (30 nM) and 0.5 or 1 nM LR004-VC-MMAE, which induced 76.3% and 85.1% apoptosis, respectively, significantly surpassing the single-agent apoptotic response. Collectively, these findings demonstrate that combining ALK inhibitors (crizotinib, ceritinib) with LR004-VC-MMAE enhances apoptosis of ALK-rearranged NCI-H3122 cells through synergistic interactions, with the combined efficacy showing a marked concentration dependence for both therapeutic components.

Fig. 2. Induction of apoptosis by crizotinib combined with LR004-VC-MMAE in ALK-rearranged NCI-H3122 cells.

A Representative dot plots of apoptosis. NCI-H3122 cells were treated with crizotinib (0, 40, 60 or 80 nM), LR004-VC-MMAE (0, 0.5 or 1 nM) or their combinations for 72 h. Untreated cells served as the control. Following the treatments, the cells were subjected to FITC-labeled Annexin-V and PI, which detect early and late apoptosis, and analyzed by flow cytometry. B Statistical analysis of cell apoptotic rates. The data revealed a ‌dose-dependent increase ‌in the percentage of apoptotic NCI-H3122 cells following ‌cotreatment with crizotinib and LR004-VC-MMAE. Error bars represent mean ± SD of three independent experiments. **P < 0.01, ***P < 0.001, and ****P < 0.0001 versus the control group; ^####P < 0.0001 versus the 40 nM crizotinib group; ^{^^^^}P < 0.0001 versus the 60 nM crizotinib group; ^δδδδP < 0.0001 versus the 80 nM crizotinib group; ^θθθP < 0.001, ^θθθθP < 0.0001 versus the 0.5 nM LR-ADC group; ^λλP < 0.01, and ^λλλP < 0.001 versus the 1 nM LR-ADC group (one-way ANOVA followed by Tukey′s multiple comparison test). CRI crizotinib, LR-ADC LR004-VC-MMAE.

The combination of crizotinib and LR004-VC-MMAE synergistically downregulated total ALK and EGFR expression and activated the mitochondrial membrane potential (MMP)-mediated, caspase-dependent apoptotic pathway in ALK-rearranged NCI-H3122 cells

MMP reduction serves as an early biomarker of mitochondria-mediated apoptosis [33]. To investigate this mechanism, MMP changes were detected using the JC-1 dye with an Enhanced MMP Assay Kit. JC-1 is a dual-emission fluorogenic probe that exhibits membrane potential-dependent spectral characteristics: under high-MMP conditions, JC-1 accumulates in mitochondrial cristae as red-fluorescent aggregates, whereas dissipation of JC-1 aggregates triggers disaggregation into green-fluorescent monomers under low-MMP conditions, indicating impairment of mitochondrial function and initiation of the apoptotic cell death process [34]. As shown in Fig. 3A, control cells exhibited characteristic polarized mitochondrial status, as evidenced by the dominant red fluorescence and minimal green signal. Compared with single agents, the crizotinib-LR004-VC-MMAE combination induced pronounced mitochondrial depolarization in ALK-rearranged NCI-H3122 cells, as evidenced by a dramatic shift from red JC-1 aggregates to green monomers, which is indicative of MMP collapse. Quantitative flow cytometry analysis (Figs. 3B–E) revealed that the combination significantly reduced the high-MMP cell population to 28.5% compared with single agents (69.4% for crizotinib; 51.2% for LR004-VC-MMAE) and dramatically expanded the low-MMP population from the baseline 25.6% to 71.5%, accompanied by a decreased JC-1 red/green ratio (combination 0.22 versus control 0.91) and a corresponding increased JC-1 green/red ratio (combination 5.22 versus control 1.11), suggesting that MMP reduction induced early apoptosis in ALK-rearranged NCI-H3122 cells. Western blot analysis showed that cotreatment with crizotinib and LR004-VC-MMAE synergistically downregulated the expression of total ALK, total EGFR, and phospho-EGFR, whereas crizotinib-mediated ALK inhibition alone did not upregulate total or phospho-EGFR expression (Fig. 3F; Supplementary Figs. 6, 7). Additionally, the combined treatment with crizotinib and LR004-VC-MMAE also increased the Bax/Bcl-2 ratio, promoted cytochrome c release, and enhanced the cleavage of caspase-3, caspase-7 and PARP, as compared with either agent alone (Supplementary Fig. 6). These findings conclusively suggest that simultaneous downregulation of ALK and EGFR expression induced by crizotinib combined with LR004-VC-MMAE contributes to the synergistic induction of mitochondrial apoptosis through MMP reduction and caspase cascade initiation in ALK-rearranged NCI-H3122 cells.

Fig. 3. Effects of crizotinib and/or LR004-VC-MMAE on MMP and the expression of ALK, EGFR and mitochondrial apoptosis-related proteins in ALK-rearranged NCI-H3122 cells.

A Representative flow cytometry density plots of JC-1-stained NCI-H3122 cells treated with 80 nM CRI and 5 nM LR004-VC-MMAE for 72 h. Mitochondrial JC-1 aggregates (red fluorescence) and cytosolic JC-1 monomers (green fluorescence) were quantified to assess the MMP or early apoptosis. B, C Proportions of cells with high or low MMP. D, E Statistical analysis of the JC-1 red/green ratio (reflecting the MMP) and green/red ratio (indicating early apoptosis). The data represent the mean ± SD of triplicate independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 (one-way ANOVA followed by Tukey′s multiple comparison test). F Representative western blot images of phsopho-ALK, ALK, phospho-EGFR, EGFR and mitochondrial apoptotic proteins in NCI-H3122 cells treated with 80 nM crizotinib and/or 1 nM LR004-VC-MMAE for 72 h. The combined therapy markedly suppressed the expression of ALK, EGFR, and phospho-EGFR; increased the Bax/Bcl-2 ratio; triggered cytochrome c release; and enhanced the cleavage of caspase 3, caspase 7 and PARP. CTRL control, MMP mitochondrial membrane potential, CRI crizotinib, LR-ADC LR004-VC-MMAE.

Crizotinib combined with LR004-VC-MMAE triggered ER stress-mediated UPR and induced cell cycle arrest at both S and G2/M phases in ALK-rearranged NCI-H3122 cells

To further investigate the molecular mechanisms underlying the enhanced apoptosis induced by the combination of crizotinib and LR004-VC-MMAE, transcriptome sequencing analysis was conducted in ALK-rearranged NCI-H3122 cells exposed to drugs for 72 h. The number of differentially expressed genes (DEGs) between any two groups is presented in Supplementary Fig. 8A, with a cutoff value of q < 0.05. Comparative transcriptomic analysis revealed distinct DEG profiles: crizotinib monotherapy modulated 5048 genes, LR004-VC-MMAE alone affected 3181 genes, whereas combination therapy regulated 6070 genes relative to the untreated control. Venn diagram analysis revealed 1920 coregulated DEGs across all the treatments, with hierarchical clustering revealing two distinct transcriptional signatures: one cluster showing augmented downregulation and another exhibiting enhanced upregulation in the combination group compared with either monotherapy (Supplementary Figs. 8B, C). Functional annotation through Gene Ontology (GO) analysis demonstrated that these shared DEGs were enriched in the cell cycle, IRE1-mediated unfolded protein response, response to endoplasmic reticulum stress, unfolded protein binding, enzyme activities and other biological process and molecular functions (Supplementary Figs. 8D, E). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment further prioritized “protein processing in endoplasmic reticulum” and “cell cycle” as the top two pathways (Supplementary Fig. 8F). These findings collectively suggest that crizotinib may synergize with LR004-VC-MMAE through activating the ER stress-mediated UPR and inducing cell cycle arrest.

Based on RNA-seq profiling, we first verified the modulation of UPR sensor proteins by western blot analysis. Compared with either monotherapy, the combination treatment with crizotinib and LR004-VC-MMAE significantly modulated UPR sensor proteins by: (1) suppressing total IRE1 expression while elevating the phospho-IRE1(S724)/IRE1 ratio, (2) increasing phospho-PERK(T982) levels and the phospho-PERK(T982)/PERK ratio, and (3) enhancing phospho-ATF6(T166) abundance and the phospho-ATF6(T166)/ATF6 ratio without affecting total ATF6 expression (Figs. 4A, B), implying that crizotinib and LR004-VC-MMAE cooperatively activate the ER stress-mediated UPR pathways, thereby potentiating apoptotic responses. We also investigated cell cycle progression through flow cytometry analysis. Our findings demonstrated combination-induced cell cycle arrest at both S and G2/M phases compared with the untreated control (Figs. 4C, D), with an obvious reduction in the G1 population compared with either monotherapy. However, no statistically significant differences in either S or G2/M arrest were observed between the combination treatment and LR004-VC-MMAE alone, suggesting that cell cycle modulation may be ancillary to the primary apoptotic synergy mediated by the mitochondrial caspase cascade and ER-stress-induced UPR pathways.

Fig. 4. Impact of crizotinib and LR004-VC-MMAE on UPR proteins and cell cycle in ALK-rearranged NCI-H3122 cells.

A Representative western blot images of UPR proteins, including IRE1, pho-IRE1 (S724), PERK, pho-PERK (T982), ATF6 and pho-ATF6 (T166), in NCI-H3122 cells exposed to 80 nM crizotinib and 5 nM LR004-VC-MMAE for 72 h. B Densitometric quantification of UPR proteins via ImageJ software (version 1.53e). The combination of crizotinib and LR004-VC-MMAE significantly increased the phospho-IRE1(S724)/IRE1, phospho-PREK(T982)/PERK, and phospho-ATF6(T166)/ATF6 ratios, indicating the induction of ER stress-mediated apoptosis. Densitometric band analyses are shown from at least three independent experiments. C Representative flow cytometry histograms of the cell cycle distribution (DNA content analysis). D Statistical analysis of the cell cycle phases in each group. The data are shown as the mean ± SD of triplicate independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 (one-way ANOVA followed by Tukey′s multiple comparison test). CTRL control, CRI crizotinib, LR-ADC LR004-VC-MMAE.

Coadministration of crizotinib with LR004-VC-MMAE synergistically suppressed NCI-H3122 xenograft tumor growth in vivo

To evaluate the in vivo antitumor efficacy of crizotinib and LR004-VC-MMAE, we established NCI-H3122 cell-derived xenograft models in NCG mice. Upon reaching a mean tumor volume of ~80 mm³, the mice were randomized into four groups and treated with sterile PBS (control), crizotinib (15 mg/kg), LR004-VC-MMAE (5 mg/kg) or their combination. No obvious body weight loss was observed in any of the treatment groups compared with the control group (Fig. 5A), indicating favorable treatment tolerability. On day 33, compared with the control, monotherapy with either crizotinib or LR004-VC-MMAE had limited antitumor efficacy, with tumor growth inhibition rates of 24.6% (P < 0.001) and 30.1% (P < 0.0001), respectively. In contrast, the combination therapy demonstrated synergistic antitumor efficacy, significantly increasing the tumor growth inhibition rate to 54.2% (P < 0.0001 versus crizotinib and P < 0.001 versus LR004-VC-MMAE), with a CDI of 0.869, thus confirming pharmacological synergy in vivo (Fig. 5B). More importantly, histopathological evaluation through H&E staining revealed no notable toxico-pathological alterations in the heart, liver, spleen, lung, kidney, intestine, or bone marrow across all groups, suggesting that the combination regimen does not elicit increased toxicity relative to single-agent treatments or the control group (Fig. 5C).

Fig. 5. Therapeutic efficacy of crizotinib and LR004-VC-MMAE against ALK-rearranged NCI-H3122 tumor xenograft models in vivo.

A Body weight changes of NCG mice bearing subcutaneous NCI-H3122 tumors treated with crizotinib (15 mg/kg orally via gavage, administered continuously for 5 days followed by a 2-day break, totaling 21 administrations), LR004-VC-MMAE (5 mg/kg intravenously once every 5 days, totaling 5 injections), and their combination. The data are expressed as the mean ± SD (n = 5). B Growth curves of NCI-H3122 cell-derived xenograft tumors in NCG mice as described in A. Data are shown as the mean ± SD (n = 5). Significant differences among treatment groups were evaluated via two-way ANOVA followed by Tukey′s multiple comparison test. For the CRI group, ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 versus the control group; for the LR-ADC group, ^#P < 0.05 and ^####P < 0.0001 versus the control group; for the CRI + LR-ADC group, ^{^^^}P < 0.001 and ^{^^^^}P < 0.0001 versus the control group^{; δδδδ}P < 0.0001 versus the CRI group; ^θP < 0.05 and ^θθθP < 0.001 versus the LR-ADC group. C Histopathological examination of harvested organs (heart, liver, spleen, lung, kidney, intestine and bone) from the tumor-bearing NCG mice was performed via H&E staining. Pictures were captured and analyzed via SlideViewer software (version 2.5.0.143918). CTRL control, CRI crizotinib, LR-ADC LR004-VC-MMAE.

Discussion

Non-small cell lung cancer (NSCLC) remains a leading cause of cancer-related mortality worldwide [3]. Targeted single-agent therapies have been extensively developed to treat NSCLC and have markedly improved clinical outcomes [6]. However, limitations persist, including intrinsic/acquired resistance and suboptimal treatment efficacy in advanced disease. Consequently, multimodal combination therapies targeting distinct signaling pathways are increasingly prioritized in oncology for their synergistic antitumor effects, reduced dosage, mitigated toxicity, and potential to overcome resistance [11]. Among them, the integration of small-molecule inhibitors with ADCs has become a promising strategy to increase the therapeutic efficacy of individual agents in ADC-based therapies [27]. For example, an ALK inhibitor enhances anti-CD30-LDM ADC-induced antitumor efficacy in nucleophosmin-ALK-positive anaplastic large-cell lymphoma [35]. Combination treatment with an enhancer of zeste homolog 2 inhibitor and an anti-disialoganglioside ADC significantly enhances therapeutic efficacy against osteosarcoma [36]. The combination of a cyclin-dependent kinase inhibitor and a death receptor 5-targeting ADC synergistically inhibits the growth of patient-derived xenograft organoids from both advanced microsatellite stable and microsatellite instability-high colorectal cancer [37]. The monocarboxylate transporter 1 inhibitor-programmed death ligand 1 (PD-L1)-ADC combination can effectively treat the programmed death 1/PD-L1 blockade-resistant tumors [38]. A P-glycoprotein inhibitor combined with an anti-nectin-4 ADC elicits rapid regression of ADC-resistant breast tumors [39]. PARP inhibitors synergize with anti-trophoblast surface antigen 2 ADC in mice bearing TNBC tumors [40]. Collectively, these findings validate the small molecule inhibitor-ADC combination as a transformative paradigm for precision oncology.

LR004-VC-MMAE‌, a novel EGFR-targeting ADC, has demonstrated potent efficacy in EGFR-positive KYSE520, A431, and TNBC xenograft models [28, 29]. However, this ADC remains poorly characterized in ALK-rearranged NSCLC models. To investigate the antitumor activity and the underlying mechanisms of LR004-VC-MMAE in ALK-rearranged NSCLC models, we evaluated the combinations of ALK inhibitors with LR004-VC-MMAE because ‌ALK rearrangements‌ and EGFR dysregulation‌ are clinically actionable oncogenic drivers, and acquired resistance to EGFR- or ALK-targeted agents remains a major clinical challenge [41].

As reported by previous studies, both ALK and EGFR are transmembrane receptor proteins with conserved topological features: their extracellular N-terminal domains cross the plasma membrane via a single α-helical transmembrane segment, whereas their intracellular C-terminal domains mediate downstream signaling [14, 42]. Specifically, wild-type ALK retains its membrane localization, whereas oncogenic EML4-ALK fusion proteins accumulate in the cytoplasm because of the loss of extracellular and transmembrane domains caused by chromosomal rearrangement [43]. Unlike EML4-ALK, EGFR signaling dynamics are tightly regulated via receptor internalization, endosomal sorting and trafficking, and lysosomal degradation [44], highlighting the critical role of cell-surface EGFR in determining the efficacy of EGFR-targeting antibody or ADC therapies, beyond its gene expression level. Consequently, our preliminary screening for NSCLC cell lines employed complementary protein quantification approaches: western blot for ALK expression and flow cytometry for cell-surface EGFR levels. This dual approach confirmed elevated expression levels of ALK and cell-surface EGFR in ALK-rearranged NCI-H3122 and NCI-H2228 cells in contrast to ALK-wild-type, EGFR-low NCI-H460 cells, providing a rational cellular basis for subsequent combination experiments.

Building on this foundation, we evaluated the antiproliferative activity of ALK inhibitors (crizotinib, ceritinib, alectinib and lorlatinib) combined with LR004-VC-MMAE across multiple NSCLC cellular models with varying ALK and EGFR expression profiles. Surprisingly, most ALK inhibitors combined with LR004-VC-MMAE produced prominent synergistic antiproliferative effects in ALK-rearranged, EGFR-overexpressing cell lines (NCI-H3122, NCI-H2228), while only a few combination regimens exhibited weak synergistic effects approaching additive effects in ALK-wild-type, EGFR-low NCI-H460 cells. Furthermore, we demonstrated that both free MMAE and EGFR expression contributed to the synergistic effects observed in ALK-rearranged NCI-H3122 and/or NCI-H2228 cells under defined experimental conditions. Validation via NCI-H3122 cell-derived xenograft models also revealed that, compared with either monotherapy group, the crizotinib-LR004-VC-MMAE combination group had a significantly reduced tumor volume to 45.8% of the control level, with a CDI of 0.869, implying synergistic enhancement of antitumor efficacy. Collectively, both in vitro and in vivo studies demonstrate that ALK inhibitors (crizotinib, ceritinib, alectinib and lorlatinib) serve as optimal combination partners with LR004-VC-MMAE for the treatment of ALK-rearranged NSCLC characterized by high EGFR expression.

Mechanistically, our previous studies have demonstrated that LR004-VC-MMAE induces apoptosis through two mechanisms in EGFR-positive ESCC and TNBC cells: (1) release of MMAE following EGFR-mediated endocytosis and subsequent intracellular trafficking into lysosomes and (2) inactivation of EGFR [28, 29]. Apoptosis, a canonical form of programmed cell death, is typically characterized by elevated apoptotic cell ratio, decreased MMP, cytochrome c release, activation of the caspase cascade, and cleavage of the caspase-specific substrate PARP [45, 46]. To validate whether enhanced apoptosis occurs and contributes to the observed inhibition of cell proliferation and tumor growth in the combination treatment group, we further assessed apoptotic cell death and MMP by flow cytometry, and examined the expression of phospho-ALK(Y1507), total ALK, phospho-EGFR (Y1068), total EGFR, and key apoptosis-related proteins by western blot analysis. Our results demonstrated that ALK inhibitors combined with LR004-VC-MMAE obviously enhanced apoptosis in ALK-rearranged NCI-H3122 cells compared with either agent alone. The crizotinib/LR004-VC-MMAE-triggered apoptotic effect was further characterized by the loss of mitochondrial MMP, an elevated Bax/Bcl-2 ratio, cytochrome c release, and cleavages of caspase 3/7 and PARP, demonstrating that the combination treatment induces apoptosis via an MMP downregulation-mediated, caspase-dependent apoptotic signaling pathway Meanwhile, we also observed that combined treatment with crizotinib and LR004-VC-MMAE exerted a more pronounced inhibitory effect on total ALK, total EGFR, and phospho-EGFR(Y1068) relative to either monotherapy, indicating effective suppression of ALK and EGFR signaling (Fig. 6). Notably, previous studies have reported that lorlatinib-mediated ALK inhibition can activate EGFR, conferring resistance to ALK-TKIs by binding to the kinase domain of ALK [47, 48]. Unexpectedly, EGFR activation was not detected in ALK-rearranged NCI-H3122 cells treated with crizotinib for 72 h or even extended to 120 h. We speculate that the crizotinib concentration employed in this study may be insufficient to trigger EGFR bypass signaling in ALK-rearranged NCI-H3122 cells, and that increasing crizotinib concentrations may activate EGFR signaling, thereby conferring resistance to crizotinib. Furthermore, given the marked downregulation of ALK and EGFR following combined treatment with crizotinib plus LR004-VC-MMAE, this regimen may effectively counteract ALK-TKI–induced EGFR bypass signaling upon the development of ALK inhibitor resistance.

Fig. 6. Potential antitumor mechanisms of the crizotinib-LR004-VC-MMAE combination on ALK-rearranged, EGFR-overexpressing NSCLC through activating mitochondrial and ER apoptosis.

The figure illustrates the synergistic antitumor mechanisms of the combination of crizotinib and LR004-VC-MMAE for the treatment of NSCLC with ALK rearrangement and EGFR overexpression. The single agent crizotinib or LR004-VC-MMAE potently inhibits ALK expression, with combination therapy markedly increasing the suppression of both ALK and EGFR. Compared with single agent treatments, the crizotinib-LR004-VC-MMAE combination has superior apoptotic effects through coordinated activation of two key pathways: (1) mitochondrial apoptosis via both MMP downregulation and caspase cascade activation and (2) the ER stress-triggered apoptotic signaling pathway, thereby increasing the degree of cytotoxic effects on ALK-rearranged, EGFR-overexpressing NSCLC. MMP mitochondrial membrane potential, ER endoplasmic reticulum.

We further explored the synergistic antitumor mechanisms of the crizotinib-LR004-VC-MMAE combination in ALK-rearranged NCI-H3122 cells using transcriptome analysis. Bioinformatics analysis revealed that ER protein processing and cell cycle were the top two factors contributing to the synergistic antitumor activity of the combined therapy. Reportedly, the ER plays critical roles in maintaining calcium homeostasis and mediating protein synthesis and processing, and excessive ER stress induces apoptosis through the activation of UPR sensors, including IRE1, PERK and ATF6 [49, 50]. Thus, we evaluated UPR sensor protein levels and cell cycle distributions to verify the results deduced from transcriptome analysis. Surprisingly, the combination of crizotinib and LR004-VC-MMAE significantly increased the phosphorylation ratios of IRE1(S724)/IRE1, PERK(T982)/PERK, and ATF6(T166)/ATF6 relative to either monotherapy, indicating potent activation of ER stress. This finding strongly aligns with previous reports that both the (S)-enantiomer of crizotinib and free MMAE induce severe ER stress in cancer cells [51, 52]. Cell cycle analysis revealed that the crizotinib-LR004-VC-MMAE combination significantly induced dual arrest at both the S and G2/M phases in the ALK-rearranged NCI-H3122 cells relative to the control group. However, no obvious difference was observed between the combination group and the LR004-VC-MMAE monotherapy group, implying that S and G2/M phase arrest is unlikely to represent the primary mechanism underlying the observed synergistic antitumor effects. Collectively, our findings demonstrate that the combination of crizotinib and LR004-VC-MMAE exerts synergistic antitumor activity through dual activation of the mitochondria/caspase-dependent apoptosis and ER stress-induced UPR (Fig. 6).

To date, considerable progress has been made in the development of EGFR-targeting ADCs, with numerous candidates undergoing clinical investigation—including becotatug vedotin, BL-B01D1 [53], CPO301, ALX2004, FH006, TQB6411, SIM0610, and QLS5316. Notably, becotatug vedotin is the first and only conditionally approved cytotoxic EGFR-targeting ADC for recurrent or metastatic nasopharyngeal carcinoma [23]. Despite this progress, cytotoxicity remains a major limitation for EGFR-targeting ADCs. As previously reported, the clinical toxicities of these agents are largely attributed to on-target off-tumor effects arising from physiological EGFR expression in normal epithelial tissues, systemic payload-related toxicity, and target-independent bystander effects [54]. This challenge is underscored by the fact that several EGFR-directed ADCs (e.g., losatuxizumab vedotin, depatuxizumab mafodotin) were discontinued due to poor therapeutic index and unmanageable adverse toxicities [55, 56]. In contrast, the approval of becotatug vedotin hinges on its favorable therapeutic index and manageable, reversible adverse events [57], supporting further clinical development of EGFR-targeting ADCs in EGFR-overexpressing malignancies. Nevertheless, the toxicity profiles of LR004-VC-MMAE, alone or combined with ALK inhibitors, remain incompletely characterized and warrant further preclinical and clinical investigation.

Second- and third-generation ALK inhibitors have demonstrated excellent and durable efficacy and are now standard first-line treatment for ALK-rearranged NSCLC [58]. Therefore, we do not recommend the ALK inhibitor plus LR004-VC-MMAE regimen as a routine first-line therapy for unselected ALK-rearranged NSCLC patients. However, given that acquired resistance to ALK-TKIs is inevitable, and EGFR upregulation serves as a key bypass mechanism limiting the efficacy of single-agent ALK inhibitors [47], along with our preclinical data supporting that LR004-VC-MMAE monotherapy inhibits EGFR activation [29], and when combined with an ALK inhibitor, it synergistically suppresses tumor cell proliferation, induces apoptosis, and reduces phospho-EGFR, total EGFR, and total ALK levels, we envisioned this combination strategy primarily for patients with acquired resistance to ALK-TKIs, especially those with concomitant high EGFR expression or activation.

Although ADC-based combination therapies have shown promising efficacy and manageable safety in NSCLC patients [59, 60], several key limitations and translational challenges should be noted. First, optimal patient selection criteria remain unclear. Reliable biomarkers (e.g., validated EGFR expression thresholds, EGFR amplification status, and pathway activation signatures) are urgently needed to identify potential responders, while patients with low or absent EGFR expression may gain little benefit. Second, overlapping and unique toxicities warrant careful clinical monitoring. Combination treatment may worsen common adverse events such as dermatological reactions, gastrointestinal toxicity, and hematological abnormalities. EGFR-targeting ADCs may also carry inherent risks of interstitial lung disease and infusion-related hypersensitivity, necessitating strict safety evaluation and individualized dose adjustment. Third, polyclonal resistance remains a major clinical challenge, including secondary ALK kinase domain mutations, enhanced EGFR downstream cascades, and activation of parallel bypass pathways such as MET and HER2. Histological transformation from NSCLC to small cell lung cancer also represents an unaddressed mechanism of treatment failure. In addition, our preclinical models cannot fully recapitulate intra-tumoral heterogeneity, clonal evolution, metastatic tropism (particularly central nervous system recurrence), or long-term resistance dynamics observed clinically, which may restrict the direct extrapolation of our findings to real-world patient cohorts.

In conclusion, this study developed a novel and synergistic combinatorial strategy composed of ALK inhibitors and a new EGFR-targeting ADC (LR004-VC-MMAE) for the treatment of ALK-rearranged and EGFR-overexpressing NSCLC. Mechanistic interrogation revealed that the combined therapy exerted its pharmacological synergism through coordinated activation of the mitochondria-mediated caspase cascade and the ER stress-triggered UPR, culminating in programmed cell death via the intrinsic apoptotic signaling axis following ALK and EGFR downregulation (Fig. 6). Although further mechanistic studies are needed, our findings at least partly elucidate the underlying basis for the synergistic antitumor efficacy of this combination regimen in ALK-rearranged and EGFR-overexpressing NSCLC models. This investigation also provides a mechanistically driven framework for developing precision therapeutic strategies for ALK-rearranged and EGFR-overexpressing NSCLC, offering new perspectives for refining clinical management paradigms in translational oncology research. Additional biomarker-driven clinical investigations and rigorous validation of advanced translational models are indispensable to further confirm the antitumor efficacy of this novel combination therapy.

Author contributions

MQF, GRJ and CAL conceptualized the framework for this research, guided the work and revised the manuscript. ZLW and GRJ performed experiments, analyzed data and wrote the manuscript. WXF prepared LR004-VC-MMAE. GJF was responsible for a subset of the western blot experiments and CCK-8 viability assays, in addition to constructing NSCLC cell lines with stable EGFR knockdown. ZDD, XJ and ZSY provided support for the animal experiments. LZR reviewed the manuscript. All authors read and approved the final version of the manuscript.

Funding

This work was supported in part by the National Natural Science Foundation of China (project approval No. 82173211), the CAMS Innovation Fund for Medical Sciences (Grant No. 2021-I2M-1-026) and the Beijing Natural Science Foundation (7242201).

Data availability

The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2025) in National Genomics Data Center (Nucleic Acids Res 2025), China National Center for Bioinformation / Beijing Institute of Genomics, Chinese Academy of Sciences (GSA-Human: HRA016321) that are publicly accessible at https://ngdc.cncb.ac.cn/gsa-human.

Competing interests

The authors declare no competing interests.

Ethics approval

This study was approved by the Ethics Committee of the Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences & Peking Union Medical College.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41389-026-00644-y.