FOXA1 drives chemoresistance through activating Wnt pathway in small cell lung cancer

Abstract

Background

Small cell lung cancer (SCLC) is an aggressive neuroendocrine tumor. While chemotherapy has been the cornerstone of first-line treatment, most SCLC patients develop chemoresistance shortly after an initial response, and the molecular basis of this resistance remains poorly understood. This study aims to identify novel drivers of SCLC chemoresistance.

Methods

FOXA1 copy number alterations and expression patterns were analyzed using public databases and validated by immunoblotting and immunohistochemistry. The functional role of FOXA1 in chemoresistance was determined through in vitro and in vivo gain- and loss-of-function assays. RNA sequencing and chromatin immunoprecipitation sequencing were performed to investigate underlying mechanisms. The functional significance of FOXA1-WNT5A-Wnt axis was tested using genetic perturbations and pharmacological inhibition.

Results

FOXA1 was a frequently amplified and upregulated gene in SCLC. FOXA1 overexpression promoted resistance to chemotherapy. Mechanistically, WNT5A was identified as a direct downstream target of FOXA1, mediating chemoresistance through activation of the Wnt/β-catenin pathway. FOXA1 also induced epithelial-mesenchymal transition (EMT) in SCLC cells. Consistently, human SCLC tumors with elevated FOXA1 expression exhibited enriched EMT and Wnt/β-catenin pathway signatures. Importantly, pharmacological inhibition of the Wnt pathway selectively suppressed the growth of FOXA1-overexpressing SCLC cells.

Conclusions

This study identifies FOXA1 as a key driver of chemoresistance in SCLC and highlights Wnt pathway inhibition as a promising therapeutic strategy to overcome FOXA1-driven chemoresistance.

Highlight box.

Key findings

• FOXA1 is frequently amplified and overexpressed in small cell lung cancer (SCLC), particularly in the ASCL1-high subtype.

• FOXA1 promotes chemoresistance in SCLC by directly activating WNT5A, leading to stimulation of Wnt/β-catenin signaling and induction of epithelial-mesenchymal transition (EMT).

• Inhibition of the Wnt/β-catenin pathway selectively suppresses tumor growth of FOXA1-overexpressing SCLC.

What is known and what is new?

• Chemoresistance is a major cause of treatment failure in SCLC, and activation of Wnt/β-catenin signaling has been implicated in relapsed disease. FOXA1 is recognized as a lineage-associated pioneer transcription factor in neuroendocrine SCLC, but its functional role in chemotherapy resistance has remained unclear.

• This study identifies FOXA1 as a novel driver of chemoresistance in SCLC and uncovers a previously unrecognized FOXA1-WNT5A-Wnt/β-catenin signaling axis involved in chemoresistance.

What is the implication, and what should change now?

• FOXA1 may serve as a clinically relevant biomarker to identify chemoresistant SCLC, enabling improved patient stratification.

• Targeting the Wnt/β-catenin pathway represents a rational therapeutic strategy for FOXA1-high SCLC, supporting biomarker-guided approaches to overcome chemotherapy resistance.

Introduction

Small cell lung cancer (SCLC) is an aggressive neuroendocrine (NE) malignancy with a pronounced tendency for rapid chemoresistance (1). For decades, platinum-based chemotherapy has been the cornerstone of SCLC treatment (2). Despite this, disease relapse occurs in approximately 80% of limited-stage patients and nearly all extensive-stage patients, ultimately leading to therapeutic failure. The molecular drivers of this chemoresistance, however, are still not fully elucidated (3). Although immune checkpoint blockade has been approved for first-line treatment, only a small subset of patients with inflamed tumors can benefit (4). Molecular subtyping of SCLC based on expression of distinct transcription factors has revealed distinct biological subsets with unique therapeutic vulnerabilities (5). In parallel, DLL3-targeted therapies, including chimeric antigen receptor (CAR) T-cell therapy, antibody-drug conjugates (ADCs), and T-cell engagers (TCEs), have been developed for SCLC treatment, with the recent U.S. Food and Drug Administration (FDA) approval of tarlatamab for patients with previously treated SCLC (6-8). However, results from the phase III DeLLphi-304 trial showed that tarlatamab achieved only a modest improvement in median progression-free survival (4.2 months) compared with chemotherapy (3.2 months) (8). Despite these advances, effective targeted therapeutic strategies remain limited for most patients with SCLC, and overall outcomes remain poor, with a 5-year survival rate of less than 7% (9). Consequently, elucidating the mechanisms driving chemoresistance in SCLC is essential for guiding the development of precise therapeutic strategies.

SCLC is notable for its high tumor mutation burden (TMB) (10). Recurrent copy number alterations, including loss of RB1 and TP53 and amplification of MYC and SOX2 genes, have been identified in SCLC (11-14). Notably, amplification of MYC paralogs has been shown to promote acquired cross-resistance to chemotherapy in SCLC (15). Recently, amplification of FOXA1 gene was reported in 4 to 8% of human SCLCs (16,17). FOXA1 is a pioneer transcription factor that binds condensed chromatin to facilitate transcriptional activation and regulate tissue-specific gene expression (18,19). FOXA1 dysregulation has been implicated in the progression and therapeutic resistance of multiple cancers (19-24). For example, FOXA1 is upregulated in docetaxel-resistant non-SCLC (NSCLC) cells and promotes chemoresistance by upregulating SOX5 (24). Furthermore, reprogramming of the FOXA1 cistrome has been shown to be essential for the development of NE prostate cancer following androgen signaling inhibition in prostate adenocarcinoma (25). Recently, FOXA1 has been identified as a multifaceted oncogene driving prostate tumorigenesis and therapy-resistant cellular plasticity (26). Notably, FOXA1 is highly expressed in human and mouse SCLC tumors and is associated with super-enhancers in ASCL1+ SCLC cells (27). FOXA1 has also been identified as a candidate driver of SCLC metastasis (28). Despite these observations, the functional roles of FOXA1 in SCLC remain poorly understood.

In this study, we show that FOXA1 is a frequently amplified and upregulated gene in SCLC. We demonstrate that FOXA1 promotes chemoresistance in SCLC by activating a WNT5A-mediated Wnt signaling axis. Importantly, our findings provide a rationale to target the Wnt pathway for FOXA1-overexpressing SCLC. We present this article in accordance with the MDAR and ARRIVE reporting checklists (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0010/rc).

Methods

Public databases data mining

Data from the AACR Project GENIE BPC 16.0-public release were analyzed using cBioPortal (www.cbioportal.org/genie/) (29). Genomic alteration frequencies of FOXA1, MYC, MYCL, and MYCN were assessed in 1,081 human SCLC tumors. The CellMiner-SCLC platform (https://discover.nci.nih.gov/SclcCellMinerCDB/) was used to evaluate the associations between FOXA1 mRNA expression, copy number, and promoter methylation in SCLC NCI-DTP cell line datasets (30). FOXA1 mRNA and protein expression levels in SCLC tumors were analyzed using the Tongji University SCLC (TU-SCLC) cohort (11). Additionally, transcriptomic profiles alongside clinical data from the IMpower133 cohort (5) were used to investigate the clinical relevance of FOXA1 expression in SCLC.

Patient specimens

Tissue samples, including both SCLC tumors and matched adjacent non-tumor lung tissues, were collected from patients who underwent surgery at Shanghai General Hospital (Table S1). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Shanghai General Hospital (Approval No. 2020SQ103). Written informed consent was obtained from all participants.

Cell lines and cell culture

Human SCLC cell lines, including DMS53 (RRID: CVCL_1177), H1092 (RRID: CVCL_1454), H146 (RRID: CVCL_1473), H1417 (RRID: CVCL_1469), H82 (RRID: CVCL_1525), H446 (RRID: CVCL_1562), DMS114 (RRID: CVCL_1174), H1048 (RRID: CVCL_1453) and H211 (RRID: CVCL_1529) were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). The H146 and H82 cell lines were cultured in RPMI 1640 medium, while the H1092 and other human SCLC cell lines were maintained in DMEM/F12 and DMEM, respectively. All media were supplemented with 10% FBS and 1% penicillin/streptomycin, and cells were incubated in a 37 ℃ incubator. Cells were regularly examined for mycoplasma contamination. For drug treatment experiments, tumor cells were treated with DMSO (#20688, Sigma-Aldrich, St. Louis, MO, USA) or MSAB (0.25 µM, HY-120697, MCE, Monmouth Junction, NJ, USA) for 72 hours. After treatment, representative images were captured using a Leica DMi8 microscope and cell pellets were collected for immunoblotting analysis.

Lentiviral vector construction and infection

The lentiviral pLenti-puro (#39481, Addgene, Watertown, MA, USA) vector was a gift from Ie-Ming Shih. The pLenti-CMV-tetR Blast (#17492, Addgene) was a gift from Eric Campeau and Paul Kaufman. The pLKO.1-TRC vector (#10878, Addgene) was a gift from David Root. The pMD2.G (#12259, Addgene) and psPAX2 (#12260, Addgene) vectors were gifts from Didier Trono. The open reading frame (ORF) sequences of human FOXA1 and WNT5A genes were amplified from cDNA derived from DMS53 cells, and subcloned into the pLenti-puro vector using a standard molecular cloning protocol. The ORF sequence of GFP was subcloned into the pLenti-puro vector and used as a negative control. Lentiviral shRNA vectors targeting human FOXA1 gene and a non-silencing control were constructed using the pLKO.1-TRC backbone following the protocol provided by the Broad Institute (https://portals.broadinstitute.org/gpp/public/). The shRNA oligos used in this study include non-silencing control-1 (shNS1): 5-CCTAAGGTTAAGTCGCCCTCG-3; non-silencing control-2 (shNS2): 5-ACCTCCACCCTCACTCTGCCAT-3; shFOXA1-2: 5-ATACGAACAGGCACTGCAATA-3.

Lentiviruses were packaged using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA, USA, #11668019) in accordance with the manufacturer’s instructions. The produced lentiviruses were collected and used to transduce human SCLC cell lines following established protocols. For constitutive overexpression, tumor cells were infected with pLenti-GFP, pLenti-FOXA1 or pLenti-WNT5A viruses, and stable transfects were selected using puromycin (A1113803, Thermo Fisher Scientific). For inducible overexpression, tumor cells were first infected with pLenti-CMV-tetR Blast virus, and stable transfectants were selected using blasticidin (A1113903, Thermo Fisher Scientific). The pLenti-FOXA1 or pLenti-GFP viruses were subsequently used to infect tetR-expressing tumor cells and stable transfectants were selected using puromycin (A1113803, Thermo Fisher Scientific). FOXA1 expression was induced by adding 0.5 µg/mL doxycycline (D3447, Sigma-Aldrich) every two days.

Cell viability assay

To assess cell viability, cells were plated in 96-well plates at 3,000 cells per well. After incubation, 10 µL of Cell Counting Kit-8 (CCK-8; Dojindo, Kumamoto, Japan) solution was added per well, and after a further 1.5-hour incubation, absorbance was measured at 450 nm using a BioTek Synergy H1 microplate reader (BioTek, Winooski, VT, USA). For in vitro drug treatment assays, 3,000 cells per well were seeded into 96-well plates and cultured for 24 h. Cisplatin (HY-17394, MCE), diluted with sterile 0.9% saline, and etoposide (HY-13629, MCE) or MSAB (HY-120697, MCE), dissolved in DMSO, were added to each well. After 72 h of incubation, the CCK-8 assay was performed as described above. Phase-contrast images of tumor cells after 72 hours of drug treatment were obtained using a Leica DMi8 microscope (Leica, Wetzlar, Germany). All experiments were performed in triplicate and independently repeated.

Animal experiments

All animal experiments were performed under a project license (No. 2022AWS0012) granted by the Institutional Animal Care and Use Committee of Shanghai General Hospital, in compliance with institutional guidelines for the care and use of animals. To evaluate tumor growth in vivo, male NOD SCID mice (Cat. No. SM-019, purchased from Shanghai Model Organisms Center, Inc., Shanghai, China) at 5–6 weeks of age and housed under specific pathogen-free (SPF) conditions received a subcutaneous injection of 5.0×10⁶ cells suspended in 100 µL of phosphate-buffered saline (PBS) into the right flank. Tumor development was monitored daily, and sizes were measured using calipers. Tumor volume was calculated using the formula: volume = 0.5 × length × width². For in vivo drug treatment studies, 5.0×10⁶ DMS53-tetR-FOXA1 cells in 100 µL PBS were subcutaneously injected into male NOD SCID mice. When tumors reached approximately 200 mm³, mice were randomly assigned to two groups. One group received doxycycline-containing chow, while the control group received regular chow. Cisplatin (6 mg/kg, HY-17394, MCE) and MSAB (15 mg/kg, HY-120697, MCE) were administered intraperitoneally every other day for 8 days and 6 days, respectively. Mice were monitored daily or every other day, and tumor volumes were measured as described above.

Immunoblotting

Immunoblotting was performed following standard protocols. SCLC tumor and lung tissues were homogenized and lysed using T-PER (78510, Thermo Fisher Scientific). For SCLC cells, direct lysis was performed using T-PER. The total protein extracts were separated by SDS-PAGE and transferred onto nitrocellulose membranes. The following antibodies were used: FOXA1 (#53528, Cell Signaling Technology, Danvers, MA, USA, RRID: AB_2799438), E-cadherin (#3195, Cell Signaling Technology, RRID: AB_2291471), β-catenin (#8480, Cell Signaling Technology, RRID: AB_11127855), Tau (#46687, Cell Signaling Technology, RRID: AB_2783844), Vimentin (#5741, Cell Signaling Technology, RRID: AB_10695459), YAP1 (#14074, Cell Signaling Technology, RRID: AB_2650491), ZO-1 (#8193, Cell Signaling Technology, RRID: AB_10898025), β-tubulin (#2128, Cell Signaling Technology, RRID: AB_823664), Lamin A/C (#4777, Cell Signaling Technology, RRID: AB_10545756), WNT5A (#2530, Cell Signaling Technology, RRID: AB_2215595), phospho-LRP6 Ser1490 (#2568, Cell Signaling Technology, RRID: AB_2139327), LRP6 (#3395, Cell Signaling Technology, RRID: AB_1950408), DVL2 (#3224, Cell Signaling Technology, RRID: AB_2093336), ASCL1 (ab213151, Abcam, Cambridge, UK), NEUROD1 (ab213725, Abcam, RRID: AB_2801303), and β-actin (A3854, Sigma-Aldrich, RRID: AB_262011). Signals were detected using the SuperSignal West Pico PLUS chemiluminescent substrate (34580, Thermo Fisher Scientific) and imaged with the FluorChem E system (ProteinSimple, San Jose, CA, USA). Band intensities were quantified using ImageJ software and normalized to β-actin.

Immunohistochemical staining

Immunohistochemistry (IHC) was conducted following standard protocols. Briefly, tumor tissues were fixed in neutral buffered formalin (NBF) for 24 hours, followed by storage in 70% ethanol. After paraffin embedding, 4 µm sections were prepared for subsequent IHC staining. The following primary antibodies were used: FOXA1 (#53528, Cell Signaling Technology, RRID: AB_2799438), β-catenin (#8480, Cell Signaling Technology, RRID: AB_11127855), cleaved caspase-3 (CC3) (#9661, Cell Signaling Technology, RRID: AB_2341188). The secondary antibody used in this study was Biotinylated Goat anti-Rabbit IgG antibody (BA-1000, Vector, Newark, CA, USA, RRID: AB_2313606). The stained sections were examined under a microscope, and representative images were taken using a Leica DM6B microscope. H-score was calculated using the formula: H-score = staining intensity (0–3) × percentage of positive cells (0–100%).

Immunofluorescence (IF) staining

For IF analysis, 2,000 cells per well were seeded into glass chamber slides (PEZGG0816, Millipore, Burlington, MA, USA) and cultured for 24 hours. Subsequently, cells were washed with PBS and fixed using 4% paraformaldehyde, in accordance with standard protocols. After washing and permeabilization, cells were treated with primary antibodies and incubated overnight at 4 ℃. The primary antibodies included Tau (#46687, Cell Signaling Technology, RRID: AB_2783844) and β-catenin (#8480, Cell Signaling Technology, RRID: AB_11127855). Nuclei were stained with DAPI (P36931, Thermo Fisher Scientific), and target proteins were visualized using an Alexa Fluor 488-conjugated goat anti-rabbit secondary antibody (A11034, Thermo Fisher Scientific). Fluorescent images were captured with a Leica DM6B microscope or Leica DMi8 microscope. Fluorescence intensity and the proportion of β-catenin-positive nuclei were quantified using ImageJ software.

Reverse transcription quantitative PCR (RT-qPCR) analysis

Total RNA was isolated from tumor cells with TRIzol reagent (15596026CN, Thermo Fisher Scientific) and subsequently reverse transcribed into cDNA using the HiScript III All-in-one RT SuperMix kit (R333-01, Vazyme, Nanjing, China). Finally, RT-qPCR was performed in triplicate on a QuantStudio 7 Flex System (Applied Biosystems, Waltham, MA, USA) with ChamQ Universal SYBR qPCR Master Mix (Q711-03, Vazyme). Primer sequences are listed in Table S2.

RNA-seq and data analysis

Total RNA was isolated from tumor cells using TRIzol reagent (15596026CN, Thermo Fisher Scientific). Subsequently, sequencing libraries were constructed with the NEBNext Ultra RNA Library Prep Kit for Illumina (#7530S, New England BioLabs, Ipswich, MA, USA) and sequenced on an Illumina HiSeq platform (Illumina, San Diego, CA, USA) to generate PE150 reads. Following alignment to the human GRCh38 genome with HISAT2, raw counts were quantified by featureCounts, and the fragments per kilobase of transcript per million mapped reads (FPKM) expression values were estimated using StringTie. Differential expression analysis was conducted with the limma package (|log₂FC| ≥1, P<0.05). Finally, functional enrichment analysis of the identified genes was performed via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. Gene Set Enrichment Analysis (GSEA) was conducted using FPKM values and the Hallmark gene set collection.

Chromatin immunoprecipitation (ChIP)-seq, ChIP-PCR, and data analysis

ChIP was performed using the SimpleChIP Enzymatic Chromatin IP Kit (9004, Cell Signaling Technology) according to the manufacturer’s instructions. An anti-FOXA1 antibody (#53528, Cell Signaling Technology, RRID: AB_2799438) was used for immunoprecipitation. Sequencing libraries were prepared from 10 ng of ChIP DNA using the NEBNext Ultra Kit (E7103, New England BioLabs) and sequenced on an Illumina HiSeq platform (paired end 150 bp). Following sequencing, adaptors were trimmed with Skewer (v0.2.2) and read quality was assessed via FastQC (v0.11.5). Clean reads were aligned to the GRCh38 genome with BWA (v0.7.10), and peaks were called using MACS2 (v2.2.6). For visualization, BigWig files were generated by deepTools (v3.1.3) and viewed in IGV (v2.18.2). ChIP-qPCR validation was performed using ChamQ SYBR qPCR Master Mix (Q711-03, Vazyme) with primers listed in Table S2.

Public RNA-seq data analysis

Processed transcriptomic data and subtype assignments for patients from the IMpower133 cohort (n=271) were obtained from the European Genome-Phenome Archive (EGAD00001006926 and EGAD00001006927). GSEA was performed using log₂(TPM + 1) values and the Hallmark gene set collection. The ranked gene lists used for GSEA analysis are provided in Tables S3,S4.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.0 (GraphPad, Boston, MA, USA). For in vitro cell viability assays and in vivo tumor growth experiments, comparisons were performed using two-way analysis of variance (ANOVA), and data are presented as mean ± standard error of the mean (SEM). Other comparisons were performed using paired or unpaired two-tailed Student’s t-tests, and data are presented as mean ± standard deviation (SD). A P value <0.05 was considered statistically significant. The sample size was not predetermined using statistical methods.

Results

FOXA1 copy number amplification and upregulation in SCLC

Copy number alterations are known to drive chemoresistance in various cancers, including both SCLC and NSCLC (31-35). To identify novel factors contributing to chemoresistance, we examined genes with frequent copy number alterations in human SCLC. By leveraging data from the AACR Project GENIE BPC database (29), we found that FOXA1 is amplified in approximately 3.0% (20 of 675) of human SCLC samples, a frequency comparable to MYC family gene amplification (Figure 1A). Consistent with this, both FOXA1 mRNA and protein expression levels were significantly upregulated in human SCLC tumors compared with normal lung tissues (Figure 1B,1C). We further confirmed FOXA1 upregulation in human SCLC tissues by immunoblotting and IHC (Figure 1D,1E).

Figure 1.

FOXA1 copy number amplification and upregulation in SCLC. (A) Bar plot showing the frequencies of the indicated altered genes in human SCLC samples from the AACR Project GENIE BPC database. (B) Scatter diagram showing FOXA1 mRNA expression in 106 paired human SCLC and normal lung tissues from TU-SCLC cohort (11). Paired Student’s t-test, two-tailed. (C) Scatter diagram showing FOXA1 protein expression in 112 paired human SCLC and normal lung tissues from TU-SCLC cohort (11). Paired Student’s t-test, two-tailed. (D) Representative immunoblotting results showing FOXA1 and ASCL1 protein levels in four paired human SCLC tumors and normal lung tissues. β-actin was used as loading control. (E) Representative IHC images showing FOXA1 protein levels in human SCLC tumors and adjacent normal lung tissues. Scale bar, 100 µm. (F) FOXA1 expression across SCLC molecular subtypes based on RNA-seq data from the IMpower133 cohort (n=271). Unpaired Student’s t-test, two-tailed. (G) Pearson correlation analysis of FOXA1 and ASCL1 mRNA expression in human SCLC samples based on RNA-seq data from the IMpower133 cohort (n=271). ***, P<0.001; ****, P<0.0001; ns, not significant. IHC, immunohistochemistry; mRNA, messenger RNA; N, normal; SCLC, small cell lung cancer; SNV, single nucleotide variation; T, tumor; TPM, transcripts per million; TU-SCLC, Tongji University SCLC cohort.

Notably, through analysis of CellMiner-SCLC NCI-DTP datasets (30), there was no significant correlation between FOXA1 mRNA expression and copy number in human SCLC cell lines (Figure S1A). Instead, we found a significant negative correlation between FOXA1 mRNA expression and promoter DNA methylation in human SCLC cell lines (Figure S1B), suggesting that promoter hypomethylation, rather than copy number alone, may contribute to FOXA1 upregulation. Together, these results indicate that increased FOXA1 expression in SCLC is driven by both copy number amplification and epigenetic regulation.

We next examined FOXA1 expression across SCLC molecular subtypes. Notably, NE SCLC subtypes (SCLC-A and SCLC-N) displayed relatively high FOXA1 expression compared to non-NE SCLC subtypes (SCLC-P and SCLC-I) (Figure 1F). Moreover, a significant positive correlation was observed between FOXA1 and ASCL1 expression levels in human SCLC tumors (Figure 1G and Figure S1C-S1E). Supporting this finding, immunoblotting analysis showed that high FOXA1 expression is predominantly present in ASCL1-high SCLC cell lines (Figure S1F). Consistently, assessment of SCLC lineage transcription factor expression revealed that all four SCLC samples were ASCL1-positive, with high FOXA1 expression observed in three of the four cases (Figure 1D). These findings indicate that FOXA1 is primarily upregulated within the ASCL1-high SCLC subtype. This is consistent with previous reports identifying FOXA1 as a super-enhancer-associated factor that cooperates with ASCL1 in SCLC (27). Together, these findings establish FOXA1 as an amplified and upregulated gene in SCLC.

FOXA1 overexpression promotes chemoresistance in SCLC

Next, we investigated the functional roles of FOXA1 copy number amplification and upregulation in SCLC. Based on endogenous FOXA1 expression in human SCLC cell lines (Figure S1F), we selected DMS53 as the primary model for functional studies, as it represents a well-characterized ASCL1-high SCLC cell line with moderate endogenous FOXA1 expression, thereby providing a suitable system for evaluating both overexpression and knockdown effects. To further examine the role of FOXA1 across distinct SCLC subtypes, we additionally employed the YAP1-high H1048 cell line for FOXA1 overexpression and the NEUROD1-high H446 cell line for FOXA1 depletion. Enforced FOXA1 expression suppressed proliferation in both DMS53 and H1048 cells (Figure 2A,2B). This is consistent with prior findings showing that FOXA1 inhibited tumor cell proliferation in breast cancer, gastric cancer, and hepatocellular carcinoma (36-38). Similar paradoxical effects have also been observed for other oncogenes, such as MYC (39). Mechanistically, overexpression of oncogenes can suppress tumor proliferation by activating intrinsic cellular safety mechanisms, such as apoptosis or senescence (39). Despite this growth-inhibitory effect, FOXA1 overexpression significantly promoted resistance to cisplatin, but not etoposide, in both cell lines (Figure 2C and S2A-S2C). Consistently, knockdown of FOXA1 promoted proliferation in both DMS53 and H446 cells while sensitizing them to cisplatin treatment (Figure 2D-2F and Figure S2D). In xenograft mouse models, cisplatin treatment significantly suppressed tumor growth in control DMS53 xenografts, whereas tumors derived from FOXA1-overexpressing DMS53 cells exhibited significantly increased resistance to cisplatin compared with control tumors (Figure 2G). Furthermore, a significant reduction in CC3-positive cells was observed in xenograft tumors derived from FOXA1-overexpressing cells (Figure 2H), indicating reduced apoptosis.

Figure 2.

FOXA1 overexpression promotes chemoresistance in SCLC. (A) Immunoblotting results showing FOXA1 overexpression in DMS53 and H1048 cells. β-actin was used as a loading control. (B) Cell viability assays showing the effects of FOXA1 overexpression on proliferation of DMS53 and H1048 cells. Two-way ANOVA. Data are presented as mean ± SEM from three independent experiments. (C) Representative dose-response curves showing the effects of FOXA1 overexpression on cell viability of DMS53 and H1048 cells following cisplatin treatment. (D) Immunoblotting results confirming FOXA1 knockdown in DMS53 and H446 cells. β-actin was used as a loading control. (E) Cell viability assays showing the effects of FOXA1 knockdown on proliferation of DMS53 and H446 cells, respectively. Two-way ANOVA. Data are presented as mean ± SEM from three independent experiments. (F) Representative dose-response curves showing the effects of FOXA1 knockdown on cell viability of DMS53 and H446 cells upon cisplatin treatment, respectively. (G) Effects of FOXA1 overexpression on in vivo growth of DMS53 cells in NOD SCID mice treated with vehicle or cisplatin. n=5 mice per group. Two-way ANOVA. Data are presented as mean ± SEM. (H) Representative IHC images showing staining of CC3 in xenograft tumors derived from DMS53 cells with or without FOXA1 overexpression. Scale bar, 100 µm. Quantification of CC3 expression levels is shown. Each dot represents one tumor region (three regions per tumor; n=4 tumors per group). Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001; ns, not significant. ANOVA, analysis of variance; IC₅₀, half-maximal inhibitory concentration; IHC, immunohistochemistry; OD, optical density; SCLC, small cell lung cancer; SD, standard deviation; SEM, standard error of the mean.

FOXA1 overexpression induces partial EMT in SCLC

To elucidate the molecular mechanisms underlying FOXA1-induced chemoresistance, we conducted RNA sequencing (RNA-seq) on DMS53 cells with or without FOXA1 overexpression. This analysis identified 3,231 differentially expressed genes (DEGs) (Figure 3A and Table S5). KEGG pathway analysis showed considerable enrichment in pathways such as neuroactive ligand-receptor interaction, ECM-receptor interaction, focal adhesion, and axon guidance (Figure 3B and Table S6). GSEA further demonstrated that FOXA1 overexpression led to significant enrichment of pathways associated with interferon-α response, interferon-γ response, inflammatory response, and EMT (Figure 3C).

Figure 3.

FOXA1 overexpression induces partial EMT in SCLC. (A) Volcano plot showing DEGs in DMS53 cells upon FOXA1 overexpression, with representative genes highlighted. Red indicates upregulated genes; blue indicates downregulated genes. (B) KEGG analysis of genes regulated by FOXA1 in DMS53 cells. (C) GSEA results showing enrichment of signaling pathways in DMS53 cells upon FOXA1 overexpression. (D) Representative phase-contrast images showing cell morphology of DMS53 cells with or without FOXA1 overexpression. Scale bar, 100 µm. (E) Representative phase-contrast images showing cell morphology of H446 cells with or without FOXA1 knockdown. Scale bar, 100 µm. Red arrows in (D) and (E) indicate cells with elongated, axon-like protrusions. (F) Immunoblotting results showing Tau protein levels in DMS53 and H446 cells with FOXA1 overexpression or knockdown, respectively. β-actin was used as a loading control. (G) Representative immunofluorescence images showing Tau protein staining in DMS53 cells with or without FOXA1 overexpression. Scale bar, 100 µm. (H) Representative immunofluorescence images showing Tau protein staining in H446 cells with or without FOXA1 knockdown. Scale bar, 100 µm. (I) Immunoblotting analysis showing the protein levels of ZO-1, E-cadherin, Vimentin, β-catenin and FOXA1 in DMS53 cells upon FOXA1 overexpression. β-actin was used as a loading control. (J) Immunoblotting analysis showing the protein levels of ZEB1, ZO-1, E-cadherin, Vimentin, β-catenin and FOXA1 in H446 cells following FOXA1 knockdown. β-actin was used as a loading control. DEGs, differentially expressed genes; EMT, epithelial-mesenchymal transition; GSEA, gene set enrichment analysis; KEGG, Kyoto Encyclopedia of Genes and Genomes; NES, normalized enrichment score; SCLC, small cell lung cancer.

Given the established roles of EMT in metastasis and chemoresistance (40) and its emerging relevance in SCLC (41), we hypothesized that FOXA1 may promote chemoresistance by inducing an EMT program. Indeed, DMS53-FOXA1 cells exhibited EMT-like morphological changes, including elongated, axon-like protrusions (Figure 3D). Conversely, FOXA1 knockdown in H446 cells impaired axon-like protrusion formation (Figure 3E). Tau, a microtubule-associated protein that facilitates cytoskeletal remodeling and membrane protrusion formation, has been implicated in EMT-associated phenotypic changes and chemoresistance in cancer (42-45). Therefore, we further examined Tau expression in these cells. FOXA1 overexpression increased Tau protein levels in DMS53 cells (Figure 3F,3G and Figure S3A), whereas FOXA1 knockdown reduced Tau expression in H446 cells (Figure 3F,3H and Figure S3A).

Moreover, FOXA1 overexpression decreased epithelial markers (ZO-1 and E-cadherin) and increased mesenchymal markers (Vimentin and β-catenin) (Figure 3I). Conversely, FOXA1 knockdown in H446 cells increased epithelial markers and decreased mesenchymal markers (Figure 3J). Importantly, GSEA revealed that an EMT signature is enriched in human SCLC tumors with high FOXA1 expression (Figure S3B). We also assessed representative EMT markers across SCLC cell lines, but given the marked heterogeneity of SCLC, no obvious correlation was observed between FOXA1 expression and EMT markers in these SCLC cell lines (Figure S3C). Collectively, our findings suggest that FOXA1 drives chemoresistance in part through inducing a partial EMT program.

FOXA1 activates WNT5A expression and Wnt pathway in SCLC

To further determine the mechanisms by which FOXA1 promotes chemoresistance and induces EMT, we analyzed transcriptomic changes in FOXA1-overexpressing DMS53 cells. We found a significant upregulation of multiple EMT-related genes, including TGM2, COL11A1, FN1, NTM, and WNT5A (Figure 3A and Table S5). Among them, WNT5A is of particular interest, as it has been shown to induce EMT in several tumors, including NSCLC (46,47). Moreover, WNT5A has been reported to promote chemoresistance in pancreatic cancer and NSCLC (48,49). WNT5A has also been implicated in activating non-canonical and canonical Wnt pathways (50-53). However, the biological role of WNT5A in SCLC remains unknown. We therefore validated WNT5A upregulation in FOXA1-overexpressing SCLC cells using RT-qPCR and immunoblotting (Figure 4A,4B). Conversely, knockdown of FOXA1 in H446 cells reduced WNT5A expression at both the mRNA and protein levels (Figure 4C,4D). Consistently, a significant positive correlation between FOXA1 and WNT5A expression was observed in two independent human SCLC cohorts (Figure S3D,S3E).

Figure 4.

FOXA1 activates WNT5A expression and Wnt pathway in SCLC. (A) RT-qPCR analysis showing upregulation of WNT5A in DMS53 cells upon FOXA1 overexpression. Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD from three independent experiments. (B) Immunoblotting analysis showing increased WNT5A protein levels in DMS53 cells upon FOXA1 overexpression. β-actin was used as a loading control. (C) RT-qPCR analysis showing downregulation of WNT5A in H446 cells following FOXA1 knockdown. Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD from three independent experiments. (D) Immunoblotting analysis showing downregulation of WNT5A protein levels in H446 cells following FOXA1 knockdown. β-actin was used as a loading control. (E) Immunoblotting results showing increased β-catenin levels in both nuclear and cytoplasmic compartments of DMS53 cells upon FOXA1 overexpression. Lamin A/C and β-tubulin were used as loading controls for the nuclear and cytoplasmic fractions, respectively. (F) Representative immunofluorescence images showing β-catenin localization in DMS53 cells with or without FOXA1 overexpression. Scale bar, 100 µm. White arrows indicate cells with nuclear β-catenin accumulation. Quantification of nuclear β-catenin-positive cells is shown on the right. Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD from three independent experiments. (G) Immunoblotting results showing phospho-LRP6, LRP6, DVL2, β-catenin and WNT5A protein levels in DMS53 cells upon FOXA1 overexpression. β-actin was used as a loading control. (H) Immunoblotting results showing phospho-LRP6, LRP6, DVL2, β-catenin and WNT5A protein levels in H446 cells following FOXA1 knockdown. β-actin was used as a loading control. (I) GSEA analysis showing enrichment of Wnt/β-catenin signaling in human SCLC tumors (IMpower133 cohort) with high FOXA1 expression (Top 25%) versus low FOXA1 expression (Bottom 25%). ***, P<0.001. FDR, false discovery rate; GSEA, gene set enrichment analysis; NES, normalized enrichment score; RT-qPCR, reverse transcription quantitative PCR; SCLC, small cell lung cancer; SD, standard deviation.

Given the established roles of WNT5A in regulating Wnt pathway, we next examined whether FOXA1 activates the Wnt pathway in SCLC. FOXA1 overexpression increased total β-catenin protein levels and promoted its nuclear accumulation, as demonstrated by immunoblotting (Figure 4E). We further confirmed the increased β-catenin nuclear accumulation by immunofluorescence staining (Figure 4F). In addition, FOXA1 overexpression increased the levels of DVL2 and phospho-LRP6, two key mediators of Wnt pathway activation (Figure 4G). Conversely, knockdown of FOXA1 in H446 cells resulted in decreased levels of β-catenin, DVL2 and phospho-LRP6 (Figure 4H). We also examined basal Wnt pathway activity across SCLC cell lines. However, no obvious correlation was observed between endogenous FOXA1 levels and Wnt pathway markers (Figure S3F), likely reflecting the molecular heterogeneity of SCLC and the involvement of additional regulatory factors beyond FOXA1 alone. Consistent with our experimental findings, GSEA revealed significant enrichment of the Wnt pathway signature in human SCLC tumors with high FOXA1 expression (Figure 4I). Notably, Wnt pathway activation has been previously reported in relapsed SCLC following chemotherapy (54). Together, these findings indicate that FOXA1 may promote chemoresistance in SCLC by activating the WNT5A/Wnt signaling axis.

As a pioneer transcription factor, FOXA1 regulates gene transcription through direct promoter binding. To determine how FOXA1 upregulates WNT5A expression in SCLC, we performed ChIP-seq analysis. We observed no significant changes in FOXA1 binding signals across the genome between DMS53-GFP and DMS53-FOXA1 cells (Figure 5A). However, FOXA1 binding was detected at the WNT5A promoter region in both DMS53-GFP and DMS53-FOXA1 cells, with increased occupancy observed in DMS53-FOXA1 cells (Figure 5B). ChIP-PCR analysis further confirmed significantly enhanced FOXA1 binding at the WNT5A promoter in DMS53-FOXA1 cells compared to control cells (Figure 5C). In line with our findings, published ChIP-seq datasets (55) from breast and prostate cancer cells also demonstrate FOXA1 occupancy at the WNT5A promoter (Figure 5D). Collectively, these findings indicate that FOXA1 directly activates the transcription of WNT5A, thereby promoting activation of the Wnt pathway in SCLC.

Figure 5.

FOXA1 directly binds WNT5A promoter in SCLC. (A) Heatmaps illustrating FOXA1 binding signal intensities at TSS loci ± 3 kb surrounding regions in constitutive GFP-overexpressing (DMS53-GFP) and FOXA1-overexpressing (DMS53-FOXA1) cells. (B) IGV plots showing FOXA1 binding intensities at the promoter region of WNT5A gene in DMS53-FOXA1 and DMS53-GFP cells by ChIP-seq analysis. (C) ChIP-PCR results showing increased binding of FOXA1 at the WNT5A promoter in DMS53-tetR-FOXA1 cells following doxycycline induction. Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD from three independent experiments. (D) Representative ChIP-seq tracks showing FOXA1 binding at the WNT5A promoter region in breast and prostate cancer cells. FOXA1 ChIP-seq signals were re-analyzed and visualized by the authors from publicly available datasets obtained through the ChIP-Atlas database (55) (accession numbers: SRX11375722, SRX6878605, SRX6712578, SRX5493758). *, P<0.05; ns, not significant. ChIP, chromatin immunoprecipitation; ChIP-seq, chromatin immunoprecipitation sequencing; Dox, doxycycline; GFP, green fluorescent protein; IgG, immunoglobulin G; IGV, Integrative Genomics Viewer; PDX, patient-derived xenograft; SCLC, small cell lung cancer; SD, standard deviation; TSS, transcription start site.

WNT5A mediates FOXA1-driven chemoresistance in SCLC

We next investigated whether FOXA1 promotes SCLC chemoresistance through WNT5A upregulation and subsequent activation of the Wnt pathway. To test this hypothesis, we overexpressed WNT5A in DMS53 cells (Figure 6A). WNT5A overexpression did not significantly affect cell proliferation but markedly increased resistance of SCLC cells to both cisplatin and etoposide (Figure 6B-6D). These findings indicate that WNT5A partially phenocopies the effect of FOXA1 overexpression in promoting chemoresistance in SCLC. Consistently, GSEA revealed that EMT and Wnt/β-catenin signaling signatures were significantly enriched in human SCLC tumors with high WNT5A expression (Figure 6E). To further determine whether FOXA1-driven EMT and chemoresistance depend on activation of the Wnt pathway, we treated cells with MSAB, a selective Wnt/β-catenin inhibitor. MSAB treatment reduced β-catenin levels and attenuated EMT-associated features induced by FOXA1 overexpression in SCLC cells (Figure 6F) and partially impaired the formation of axon-like protrusions (Figure 6G). Together, these findings suggest that FOXA1 drives chemoresistance in SCLC, at least in part, by activating the WNT5A/Wnt/β-catenin signaling axis.

Figure 6.

FOXA1 promotes chemoresistance through activating Wnt pathway in SCLC. (A) Immunoblotting analysis showing the WNT5A protein levels in DMS53 cells upon WNT5A overexpression. β-actin was used as a loading control. (B) Cell viability assays showing the effects of WNT5A overexpression on proliferation of DMS53 cells. Two-way ANOVA. Data are presented as mean ± SEM from three independent experiments. (C,D) Representative dose-response curves showing the effects of WNT5A overexpression on DMS53 cell viability following cisplatin (C) or etoposide (D) treatment. Quantification of IC₅₀ values is shown on the right. Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD from three independent experiments. (E) GSEA showing enrichment of EMT and Wnt/β-catenin signaling signatures in human SCLC tumors (IMpower133 cohort) with high WNT5A expression (top 50%) compared with low WNT5A expression (Bottom 50%). (F) Immunoblotting analysis showing the effect of MSAB treatment (0.25 µM) on ZO-1, E-cadherin, Vimentin, β-catenin and FOXA1 protein levels in DMS53 cells with or without FOXA1 overexpression. (G) Representative phase-contrast images showing the morphology of DMS53 cells with or without FOXA1 overexpression after 72 hours of DMSO or MSAB treatment (0.25 µM). Scale bar, 100 µm. *, P<0.05; **, P<0.01; ns, not significant. ANOVA, analysis of variance; EMT, epithelial-mesenchymal transition; FDR, false discovery rate; GSEA, gene set enrichment analysis; IC₅₀, half-maximal inhibitory concentration; NES, normalized enrichment score; OD, optical density; SCLC, small cell lung cancer; SD, standard deviation; SEM, standard error of the mean.

Wnt pathway inhibition suppresses growth of FOXA1-overexpressing SCLC

Based on the above findings, we hypothesized that FOXA1-overexpressing SCLC cells may be more sensitive to inhibition of WNT/β-catenin signaling. Notably, MSAB treatment inhibited the proliferation of DMS53 cells with or without FOXA1 induction (Figure 7A). However, FOXA1-overexpressing DMS53 cells exhibited higher sensitivity to MSAB treatment (Figure 7B). Furthermore, we demonstrated that doxycycline-induced FOXA1-overexpressing DMS53 xenografts were more sensitive to MSAB treatment in mouse models (Figure 7C and Figure S3G). IHC analysis of subcutaneous tumors revealed that β-catenin levels were significantly upregulated in xenografts derived from FOXA1-overexpressing cells (Figure 7D), consistent with enhanced Wnt pathway activation in these tumors. Together, these data indicate that FOXA1-expressing SCLC tumors exhibit increased vulnerability to Wnt pathway inhibition, suggesting that targeting Wnt pathway may offer a promising strategy to overcome FOXA1-driven chemoresistance.

Figure 7.

Wnt pathway inhibition suppresses growth of FOXA1-overexpressing SCLC. (A) In vitro proliferation of DMS53 cells with or without FOXA1 overexpression upon MSAB treatment (0.25 µM). Two-way ANOVA. Data are presented as mean ± SEM from three independent experiments. (B) Representative dose-response curves showing the effect of MSAB treatment on cell viability of DMS53 cells with or without FOXA1 overexpression (left). Quantification of IC₅₀ values is shown on the right. Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD from three independent experiments. (C) In vivo growth curves showing tumor growth of MSAB-treated DMS53-tetR-FOXA1 xenografts with or without doxycycline induction in NOD SCID mice. MSAB treatment was initiated when tumor volume measurement began and continued for 6 days. Two-way ANOVA. Data are presented as mean ± SEM (n=6 mice per group). (D) Representative IHC images showing FOXA1 and β-catenin staining in MSAB-treated DMS53-tetR-FOXA1 xenografts with or without doxycycline induction. Scale bar, 100 µm. Quantification of FOXA1 and β-catenin expression levels is shown. Each dot represents one tumor region (three regions per tumor). Unpaired Student’s t-test, two-tailed. Data are presented as mean ± SD (n=4 tumors per group). **, P<0.01; ****, P<0.0001. ANOVA, analysis of variance; IC₅₀, half-maximal inhibitory concentration; IHC, immunohistochemistry; OD, optical density; SCLC, small cell lung cancer; SD, standard deviation; SEM, standard error of the mean.

Discussion

Chemoresistance has been a barrier in the clinical management of SCLC (56). Although most SCLC patients are sensitive to initial chemotherapy, nearly all eventually relapse and develop chemoresistance (57). Previous studies have reported that amplification of MYC family members and loss of KEAP1, among other alterations, can promote chemoresistance in SCLC (15,58-60). However, the mechanisms underlying SCLC chemoresistance remain largely unexplored. In the current study, we identify FOXA1 as a novel driver of chemoresistance in SCLC, contributing to chemoresistance in a subset of tumors. In line with our findings, FOXA1 has been shown to confer drug resistance in multiple cancer types. For example, FOXA1 was shown to be overexpressed in metastatic castration-resistant prostate cancer, whereas FOXA1 overexpression was able to promote resistance to immunotherapy and chemotherapy in prostate cancer (61,62). We also demonstrated that FOXA1 expression is upregulated in human SCLC tissues, especially in ASCL1-high SCLC subtype. Consistently, FOXA1 is considered one of the regulators of NE and NEv2 SCLC subtypes (63). To our knowledge, this study provides the first functional characterization of FOXA1 in SCLC and establishes its role in promoting chemoresistance.

Rapidly proliferating tumor cells are generally more sensitive to cytotoxic chemotherapy, as conventional chemotherapeutic agents preferentially target actively dividing cells by disrupting DNA replication and mitosis (64,65). Consistent with this principle, our findings show that FOXA1 overexpression suppresses cell proliferation while promoting chemoresistance in SCLC cells. Thus, the reduced proliferation induced by FOXA1 may partially contribute to decreased cisplatin sensitivity. However, IC₅₀-based analyses indicate that the observed shift in cisplatin response cannot be fully explained by slower proliferation alone, suggesting the involvement of additional resistance mechanisms. Mechanistically, our findings indicate that FOXA1 overexpression induces a partial EMT program in SCLC cells, which has been linked to therapeutic resistance. Furthermore, FOXA1 activates the WNT5A/Wnt/β-catenin signaling axis, which further contributes to enhanced cisplatin resistance. Together, these results suggest that although FOXA1 suppresses cell proliferation, its overexpression promotes chemoresistance in SCLC through multiple mechanisms.

EMT plays important roles in tumor development and progression, especially in promoting metastasis and drug resistance (66). For example, EMT has been uncovered to be a common mechanism for acquired resistance to EGFR inhibitors, whereas restoring CDH1 expression rescues sensitivity to EGFR inhibitors in NSCLC (67,68). Moreover, EMT has been revealed to contribute to chemoresistance, but not lung metastasis, in breast cancer using an EMT lineage tracing system (69). Although classical histopathological evidence of EMT is rarely observed in SCLC patient specimens, accumulating evidence supports the presence of EMT-like transcriptional and functional programs that contribute to chemoresistance in SCLC (41,70-75). For example, Chen et al. recently used in vivo genetic lineage tracing to demonstrate a causal role of EMT in driving chemoresistance in SCLC (70). Moreover, our previous work revealed that CREBBP inactivation promotes SCLC development through inducing a partial EMT program (76). Additionally, targeting EMT with a Met inhibitor reversed chemoresistance in SCLC (75). In the present study, we demonstrate that FOXA1 induces EMT-like changes in SCLC cells, which may contribute to FOXA1-driven chemoresistance. Interestingly, FOXA1 has been shown to inhibit EMT in prostate and pancreatic cancers (77,78). These findings indicate that the role of FOXA1 in regulating EMT is context dependent.

Notably, the recently described inflamed SCLC (SCLC-I) subtype, characterized by EMT features, has been linked to increased chemoresistance (5). In our study, FOXA1 was linked to EMT-like phenotypes and chemoresistance, yet FOXA1 upregulation was predominantly observed in the ASCL1-positive SCLC subtype. Although this observation may appear inconsistent with the SCLC-I subtype, increasing evidence highlights substantial intertumoral heterogeneity and lineage plasticity in SCLC, whereby tumor cells can adopt transitional or mixed transcriptional states (79). In this context, FOXA1-driven EMT-like reprogramming in ASCL1-high cells may represent an alternative route to phenotypic plasticity and chemoresistance that partially converges with features of the SCLC-I subtype. Together, these findings suggest that EMT-associated chemoresistance in SCLC is not confined to a single molecular SCLC subtype but may arise through multiple transcriptional mechanisms in a context-dependent manner.

Activation of the Wnt/β-catenin pathway has recently been implicated in relapsed SCLC and promoting chemoresistance (54). However, the mechanisms underlying Wnt/β-catenin pathway activation in SCLC remain elusive. Here, we demonstrate that FOXA1 overexpression activates the Wnt/β-catenin pathway by transcriptionally upregulating WNT5A expression and promoting β-catenin accumulation. Furthermore, we show that WNT5A overexpression promotes chemoresistance in SCLC, partially phenocopying the effects of FOXA1 overexpression. Together, these findings reveal a previously unrecognized FOXA1-WNT5A-Wnt signaling axis in driving SCLC chemoresistance. Nevertheless, we acknowledge that the direct regulation of additional Wnt-associated genes by FOXA1 requires further validation through more comprehensive chromatin-based analyses, as well as genetic suppression of key components of the Wnt signaling pathway. Importantly, we demonstrate that pharmacological inhibition of the Wnt pathway significantly suppresses the growth of FOXA1-overexpressing SCLC cells, providing a potential therapeutic rationale for targeting Wnt pathway in SCLC with high FOXA1 expression. Notably, MSAB treatment also partially reversed EMT-associated features in control cells (Figure 6F), indicative of the presence of basal Wnt/β-catenin signaling in DMS53 cells (Figure S3F). This observation is consistent with the detection of endogenous FOXA1 protein in parental DMS53 cells (Figure S1F), which may sustain low-level Wnt pathway activity even in the absence of exogenous FOXA1 overexpression. These findings suggest that MSAB suppresses both basal and FOXA1-enhanced Wnt signaling, with more pronounced functional effects observed in the context of elevated FOXA1 expression. Finally, targeting the Wnt/β-catenin pathway has been recognized as a promising strategy in cancer therapy (80,81). For example, multiple Wnt pathway inhibitors are already being tested in clinical trials (NCT02675946, NCT01351103, NCT06399757).

This study has several limitations. First, FOXA1 expression was primarily evaluated using the TU-SCLC cohort, which includes 112 paired tumor and adjacent normal lung tissues. Although this dataset allowed us to assess FOXA1 expression in SCLC, direct comparisons between primary chemo-sensitive and relapsed chemo-resistant tumors were not available and warrant further investigation in future studies. Second, the RNA-seq analyses were performed by comparing doxycycline-induced FOXA1-overexpressing cells with uninduced controls. Although doxycycline at 0.5 µg/mL is widely used in inducible expression systems and is generally considered to exert minimal transcriptional off-target effects, the absence of doxycycline-treated empty vector controls represents a limitation. Future studies incorporating this control will help more definitively attribute the observed transcriptional and phenotypic changes to FOXA1 overexpression rather than potential effects of doxycycline exposure. Third, the involvement of WNT/β-catenin signaling in FOXA1-driven chemoresistance was primarily supported by pharmacological inhibition using MSAB. Additional studies employing complementary genetic approaches or the use of independent pharmacological inhibitors will be necessary to further validate the role of Wnt/β-catenin signaling in this process. Finally, FOXA1 function was assessed in only three SCLC cell lines, and the generalizability of these findings remains to be validated in additional cell lines or patient-derived xenograft models.

Conclusions

In summary, our study reveals a novel mechanism of chemoresistance in SCLC, in which FOXA1 drives WNT5A expression and subsequent activation of Wnt/β-catenin signaling. These findings not only deepen our understanding of FOXA1-driven tumor adaptation to chemotherapy but also provide a mechanistic rationale for therapeutically targeting the Wnt pathway in FOXA1-high SCLC.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Shanghai General Hospital (Approval No. 2020SQ103). Written informed consent was obtained from all participants. All animal experiments were performed under a project license (No. 2022AWS0012) granted by the Institutional Animal Care and Use Committee of Shanghai General Hospital, in compliance with institutional guidelines for the care and use of animals.

Data Sharing Statement

Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0010/dss

tlcr-15-07-202-dss.pdf ^{(70.9KB, pdf)}

DOI: 10.21037/tlcr-2026-1-0010