Skip to main content
Download PDF

Single-cell RNA sequencing reveals cellular and molecular heterogeneity in extensive-stage small cell lung cancer with different chemotherapy responses

Cancer Cell International volume 25, Article number: 157 (2025) Cite this article

Abstract

Despite its rapid growth and early metastasis, small cell lung cancer (SCLC) is more chemosensitive than other lung cancers. However, some patients with extensive-stage SCLC (ES-SCLC) do not respond to first-line chemotherapy, resulting in poorer prognoses due to inter- and intratumoral heterogeneity. In this study, we conducted single-cell RNA sequencing of 9 treatment-naive ES-SCLC samples. Based on comprehensive imaging evidence collected before and after two cycles of first-line chemotherapy and sample types, the 9 samples were categorized into three groups: progressive disease with the pleural effusion sample (PD_PE group, n = 1), progressive disease with the primary tumor samples (PD_TU group, n = 2), and partial response with the primary tumor samples (PR_TU group, n = 6). Based on transcriptomic landscape and cell type composition, the PD samples represent a multicellular ecosystem distinct from PR samples. The immune response, along with the elevated expression of immune-related genes such as LTF, SLPI, SPARC and IGLV1-51, might correlate with a poor first-line chemotherapy response in ES-SCLC. We also observed that T cells, particularly effector T cells, were more abundant in PD_TU group, with TNFA signaling via NFκB being significantly enriched. The PD_TU group was strongly enriched with macrophages and tumor-associated macrophages (TAMs), and angiogenesis in TAMs was highly enriched. Immunomodulatory fibroblasts were highly abundant in PD_TU group, and the pathways of epithelial-mesenchymal transition and angiogenesis were upregulated. This study offers the first comprehensive insights into the cellular and molecular heterogeneity in treatment-naive patients with ES-SCLC with different chemotherapy responses.

Introduction

Small cell lung cancer (SCLC) is an extremely aggressive neuroendocrine carcinoma known for its exceptionally high proliferation and early metastasis [1, 2], and accounts for approximately 15% of all lung cancer cases [3]. The Veterans Administration Lung Study Group staging system classifies SCLC into limited-stage and extensive-stage based on whether it has spread beyond one hemithorax [4]. Unfortunately, about two-thirds of SCLC patients are diagnosed with extensive-stage SCLC (ES-SCLC) at the time of initial diagnosis due to the cancer’s rapid progression [5]. With a 2-year survival rate of only 7%, ES-SCLC is among the deadliest malignancies [6, 7].

For the past 40 years, the platinum combined with etoposide chemotherapy regimen has been the unshakable standard first-line treatment for ES-SCLC. In recent years, the immunotherapy has demonstrated some effectiveness in improving treatment efficiency and prolonging survival in ES-SCLC patients. However, the clinical benefits remain limited [8, 9]. Previous studies have shown that ES-SCLC patients are highly sensitive to first-line chemotherapy, with response rates of 60–65% [10]. Nevertheless, some ES-SCLC patients do not respond to first-line chemotherapy and progress more rapidly, and the underlying mechanisms for this lack of response remain unclear. The tumor microenvironment (TME) plays a crucial role in tumor formation and development, as well as shaping tumor behavior [11, 12]. Accumulating evidence suggests that intratumor heterogeneity between malignant and nonmalignant cells, and their interactions within the TME are vital to chemotherapy responses [13,14,15]. Therefore, it is crucial to clarify the cellular and molecular heterogeneity within the TME that are associated with responses to first-line chemotherapy in treatment-naive ES-SCLC patients, as this information could guide personalized treatment.

Single-cell RNA sequencing (scRNA-seq) technologies allow for the profiling of cell components within the TME and reveal tumor heterogeneity [16]. In this study, we performed scRNA-seq of 9 untreated ES-SCLC samples. Based on computed tomography (CT) scan results of the lungs before and after first-line chemotherapy, the 9 samples were categorized into the progressive disease (PD) group and the partial response (PR) group. We compared the scRNA-seq profiles of samples from PD and PR patients. Additionally, we comprehensively characterized the transcriptomic features of malignant cells, immune cells and stromal cells, enhancing our understanding of the biological basis for chemotherapy responses in ES-SCLC and identifying factors associated with treatment efficacy. This research provides new insights into the biological basis of chemotherapy responses in ES-SCLC, helping to identify factors that contribute to better or worse treatment efficacy.

Methods and materials

Patients and sample collection

A total of 9 previously untreated patients diagnosed with ES-SCLC were prospectively enrolled in our study at Shanghai Pulmonary Hospital, China, between September and December 2022. All patients were diagnosed with ES-SCLC through histologic diagnosis by pathologists and adequate imaging evidence by radiologists. The patients all received two cycles (three weeks per cycle) of first-line chemotherapy with etoposide plus carboplatin (EC) and had not undergone immunotherapy, radiotherapy, or other anti-tumor therapies during this period.

We collected primary tumor tissues prior to first-line chemotherapy using percutaneous pulmonary biopsy, bronchoscopy biopsy, or endobronchial ultrasound (EBUS) biopsy. Based on chest CT scans and sufficient imaging evidence before and after two cycles of first-line chemotherapy, three patients were assigned to the PD group and six patients were assigned to the PR group. PD is defined as the appearance of new lesions or a 20% or greater increase in the size of existing lesions, while PR is defined as a 30% or greater reduction in lesion size. Notably, one PD patient’s sample was derived from pleural effusion collected via closed thoracic drainage. The response evaluation was based on an increase in pleural effusion assessed by ultrasound, along with the appearance of brain metastasis. Considering the different sample types, the 9 treatment-naive ES-SCLC samples were finally classified into three groups: PD with the pleural effusion sample (PD_PE group, n = 1), PD with the primary tumor samples (PD_TU group, n = 2), and PR with the primary tumor samples (PR_TU group, n = 6). The clinical information of the 9 patients is presented in Supplementary Table 1. The CT scan of the lungs of the 9 patients before and after chemotherapy is shown in Supplementary Fig. 1. This study was approved by the Ethics Committee of Shanghai Pulmonary Hospital affiliated to Tongji University (K24-530), and written informed consent was obtained from all patients.

Preparation of single-cell suspensions

Fresh samples were quickly isolated and sent to the GENECHEM laboratory. The samples were stored in MACS Tissue Storage Solution (Miltenyi Biotec, Bergisch Gladbach, Germany) until processing. First, the samples were washed with phosphate-buffered saline (PBS) and then minced into small pieces (approximately 1 mm3) on ice. They were enzymatically digested for 90 min at 37 ℃ with agitation using 1 mg/mL collagenase I (Gibco, Waltham, MA), 1 mg/mL collagenase II, 60 U/mL Hyaluronidase (Sigma, Taufkirchen, Germany), 10 U/mL Liberase (Roche, Basel, Switzerland), and 0.02 mg/mL DNase I (Roche, Basel, Switzerland). After digestion, the samples were filtered through 100 and 40 µm cell strainers and then centrifuged at 300×g for 5 min. After washing with Dulbecco’s phosphate-buffered saline (DPBS) containing 0.5% bovine serum albumin (BSA), the cell pellets were re-suspended, stained, and counted in DPBS with 0.5% BSA.

scRNA-seq

The Chromium instrument and the Single Cell 3′ Reagent Kit V3.1 (Dual Index) were used to prepare individually barcoded scRNA-Seq libraries according to the manufacturer’s protocol (10× Genomics, Pleasanton, CA). Briefly, sample partitioning and molecular barcoding were performed using the Chromium Controller (10× Genomics, Pleasanton, CA). Cellular suspensions were loaded with the Single-cell 3′ Gel Beads onto a Single-cell 3′ chip, where gel beads in emulsion (GEM) were generated. The RNA from the barcoded cells was reverse-transcribed, and sequencing libraries were then constructed using reagents from the Chromium Single Cell Reagent Kit V3.1 (10× Genomics, Pleasanton, CA). Sequencing was performed using Illumina (HiSeq 2000 or NovaSeq, depending on the project) according to the manufacturer’s instructions.

scRNA-seq data processing

We utilized fastp to conduct basic quality statistics on the raw reads. The raw reads were demultiplexed and mapped to the reference genome GRCh38. This was done using the 10× Genomics Cell Ranger pipeline (https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines) with default parameters. All downstream single-cell analyses were performed using Cell Ranger and Seurat [17, 18]. In brief, for each gene and each cell barcode (filtered by Cell Ranger), unique molecule identifiers were counted to construct digital expression matrices. Seurat’s secondary filtration process considered a gene expressed if it was detected in more than three cells. Additionally, each cell was required to have at least 200 expressed genes, leading to the exclusion of certain foreign cells. The Seurat package was utilized for data normalization, dimensionality reduction, clustering, and differential expression analysis. The batch effect was removed by using the Harmony R package [19]. For clustering, highly variable genes were selected, and the principal components based on those genes were used to build a graph, which was segmented with a resolution of 0.6. Based on the filtered gene expression matrix by Seurat, differential expression analysis of samples was carried out using the edgeR package to obtain zone-specific marker genes. We clustered cells using the FindNeighbors and FindClusters functions and performed nonlinear dimensional reduction with the RunUMAP function with default settings.

Cell type annotation

After using the graph-based uniform manifold approximation and projection (UMAP) for nonlinear dimensional reduction, the cells grouped together based on shared features. The FindAllMarkers function in Seurat was used to identify markers for each cluster. Next, we classified and annotated the clusters based on the expression of canonical markers specific to certain cell types. Clusters without canonical cell type markers were considered low-quality, while those with two or more canonical markers were identified as doublet cells. We removed both low-quality clusters and doublets to ensure the accuracy and reliability of the analysis.

Identification of differentially expressed genes

The FindAllMarkers function in Seurat, which calculates differentially expressed genes (DEGs) among various cell subpopulations and groups, was used in this study. P values are determined using the Wilcoxon rank-sum test. The adjusted P value after Bonferroni correction was <0.05. The cut-off for log (fold change) of the average expression between the two conditions (avg_logFC) being compared was set to be >0.5.

Analysis of differentially expressed genes

To explore the biological functions of differentially expressed genes, we used clusterProfiler for functional enrichment analysis of the gene sets. This study presents Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses. Multiple testing was managed using the false discovery rate, with a threshold of <0.05 considered significant [20]. We performed pathway analyses on the 50 hallmark pathways listed in the molecular signatures database. We performed gene set variation analysis (GSVA) on each cell type to estimate pathway activity and identify enriched pathways [21]. Pathway activity per cell was evaluated using the GSVA R package with default settings. Differential expression analysis of pathways was performed using the Limma R package [22]. Significantly disturbed pathways were identified with Benjamini-Hochberg-corrected P value of ≤0.01.

Copy number variations (CNVs) inferred with scRNA-seq data

To identify malignant cells from epithelia, we used the InferCNV R package to estimate the CNVs [23]. A raw counts matrix, annotation file and gene or chromosome position file were prepared according to data requirements. The immune cells were used as normal references, and the default parameters were applied. The sum of calculated CNV for each gene per cell was defined as the CNV score of the cell.

Cell–cell communication analysis

To explore the complex network of intercellular communication signals, we conducted an in-depth analysis using the CellChat R package [24]. On the basis of the default database CellChatDB, we identified signaling pathways, calculated communication probabilities, and constructed communication networks. Ultimately, we determined the quantity and weight relationships of ligand-receptor pairs between different cell types, as well as the communication probabilities of ligand-receptor interactions among different cell types. Those with P value <0.05 were visualized using dot plots.

Pseudotime trajectory analysis

The Monocle2 was used to model lineage development of cell subtypes and construct pseudotime trajectory based on the gene expression profiles [25]. After dimensionality reduction and cell ordering, the differentiation trajectories were inferred with the default parameters of Monocle.

Results

Transcriptomic landscape and cell type composition in ES-SCLC with different chemotherapy responses

We performed scRNA-seq profiling of 9 samples from 9 treatment-naive ES-SCLC patients, of which 2 samples were primary tumors from 2 PD patients, 6 samples were primary tumors from 6 PR patients, and 1 sample was pleural effusion from 1 PD patient (Fig. 1A). After initial quality controls and filtering low-quality cells, the batch effect was removed to avoid confounding technical biases. A total of 74,422 single cells were obtained. Among these cells, 19,380 were from the PD_TU group, 45,032 were from the PR_TU group, and 10,010 were from the PD_PE group. After data normalization and principal components analysis, 41 high-confidence cell clusters were identified using UMAP. Based on the expression of canonical gene markers, we identified major cell types known to exist in lung tissue, including epithelial cells, T cells, natural killer (NK) cells, myeloid cells, fibroblasts, endothelial cells, B cells, plasma cells and mast cells among others (Fig. 1B, C).

Fig. 1
figure 1

ScRNA-seq profiling of cellular heterogeneity in ES-SCLC with different chemotherapy responses. A Design of scRNA-seq experiments on treatment-naive ES-SCLC samples. B UMAP projection of 74,422 single cells from 9 treatment-naive ES-SCLC samples. Each dot corresponds to a single cell, colored by cell type and group. C Canonical markers were used to label clusters by cell identity as represented in the UMAP plot. D The proportion of cell types among three groups

Full size image

We observed variations in the proportions of these cell types in the 9 samples among three groups (Fig. 1D, Supplementary Fig. 2A, B and C). From the results, the average proportion of cell types in pleural effusion varied substantially from that in tumor tissues. In the PD_PE group, immune cells were highly abundant, including 5271T cells (52.7%), 2169 B cells (21.7%), and 871 NK cells (8.7%). The relative abundance of NK cells in the PD_TU group was comparable to that in the PR_TU group. In the PD_TU group, the relative abundance of endothelial cells, fibroblasts, mast cells, and T cells increased compared to the PR_TU group, while the relative abundance of B cells, epithelial cells, monocytes, neutrophils, and plasma cells decreased. Overall, PD samples represent a multicellular ecosystem distinct from PR samples, reflecting intertumoral heterogeneity.

Analyses of differentially expressed genes between the PD group and PR group

The comparison of differentially expressed genes between the PD_TU and PR_TU groups showed 145 genes upregulated and 98 genes downregulated (Fig. 2A). Enrichment analysis of GO and KEGG pathways indicated that the differentially expressed genes in the PD_TU group were associated with immune response, adaptive immune response, and immune system process, as well as relaxin signaling pathway and PI3K-Akt signaling pathway (Fig. 2B, C). The immune response may be related to the chemotherapy responses in SCLC [26]. The higher expression of several genes associated with immune response, such as lactotransferrin (LTF), secretory leukocyte peptidase inhibitor (SLPI), secreted protein acidic and rich in cysteine (SPARC) and immunoglobulin lambda variable 1–51 (IGLV1-51), might correlate with a poor first-line chemotherapy response in ES-SCLC (Fig. 2D).

Fig. 2
figure 2

Analyses of differentially expressed genes between PD group and PR group. A Heatmap of genes differentially expressed between PD_TU group and PR_TU group. B, C GO and KEGG pathway enrichment of genes differentially expressed between PD_TU group and PR_TU group. D Violin plots with dots showing differential expression of genes associated with immune response between PD_TU group and PR_TU group. Each dot corresponds to a single cell

Full size image

Characterization of epithelial cells and malignant cells between the PD group and PR group

In our subsequent analyses, we focused on the transcriptomic characteristics of each major cell type. All the epithelial cells were obtained from the three groups and further clustered into neuroendocrine cells, basal airway epithelial cells, ciliated airway epithelial cells, tuff cells, alveolar epithelial type 2 cells and others (Fig. 3A, B). The variations in the proportions of these epithelial cell subtypes in the 9 samples among three groups were presented (Supplementary Fig. 3A, B, C and D). The distribution of epithelial cell subtypes in pleural effusion was significantly different from that in tumor tissues. The number and proportion of neuroendocrine cells in the PD_PE group were limited. In the PD_TU group, the relative abundance of alveolar epithelial type 2 cells increased compared to the PR_TU group.

Fig. 3
figure 3

Characterization of epithelial cells and malignant cells between PD group and PR group. A UMAP projection of subclustered epithelial cells from 9 treatment-naive ES-SCLC samples. Each dot corresponds to a single cell, colored by cell type and group. B Heatmap of marker gene expression in epithelial cell subtypes. C Heatmap of large-scale CNVs in epithelial cells using immune cells as references. D Heatmap of genes in malignant cells differentially expressed between PD_TU group and PR_TU group. E, F GO and KEGG pathway enrichment of genes in malignant cells differentially expressed between PD_TU group and PR_TU group. G Violin plots with dots showing differential expression of genes associated with ribosome between PD_TU group and PR_TU group. Each dot corresponds to a single cell. H Number of cellular interactions with malignant cells in two groups. I, J Ligand-receptor bubble diagram of malignant cells acting on different cells in PD_TU group and PR_TU group

Full size image

Malignant cells were identified by using immune cells as normal references, and InferCNV was employed to analyze and score chromosomal CNVs (Fig. 3C, Supplementary Fig. 3E). A total of 34,946 malignant cells were identified, with 9625 (accouting for 49.66%) from the PD_TU group, 24,540 (accouting for 54.49%) from the PR_TU group and 781 (accouting for 7.80%) from the PD_PE group. Subsequently, we compared the differentially expressed genes in malignant cells between the PD_TU group and PR_TU group, discovering that 778 genes were upregulated and 1022 genes were downregulated (Fig. 3D). GO and KEGG enrichment analyses linked differential gene expression in malignant cells to ribosome, cytoplasmic translation and others (Fig. 3E, F). Compared with the PR_TU group, the ribosome-associated genes including ribosomal protein S4 Y-linked 1 (RPS4Y1) and FosB proto-oncogene (FOSB), were upregulated in the PD_TU group (Fig. 3G).

To characterize intercellular interactions within ES-SCLC with different chemotherapy responses, we revealed the quantity of cellular interactions with malignant cells as source cells in the PD_TU group and PR_TU group (Fig. 3H). The malignant cells exhibited the highest communication patterns with fibroblasts and endothelial cells in the PD_TU group and PR_TU group, respectively. Furthermore, we delved into the ligand-receptor interactions of malignant cells acting on different cells in both groups (Fig. 3I, J). It is demonstrated that the malignant cells engaged with other cells through the MIF-(CD74 + CXCR4), MIF-(CD74 + CD44), and APP-CD74 receptor-ligand pairs in both groups.

Effector T cells are more abundant in the PD_TU group

We subclustered 17067 T cells and NK cells from three groups into 8 main subtypes (Fig. 4A, B), with 3686 from the PD_TU group, 7239 from the PR_TU group and 6142 from the PD_PE group. The variations in the proportions of these T/NK cell subtypes in the 9 samples among three groups were revealed (Fig. 4C, Supplementary Fig. 4A, B and C). The distribution of T/NK cell subtypes in pleural effusion varied substantially from that in tumor tissues and the naive CD4+ T cells exceeded half in the PD_PE group. For NK cells, the distribution in the PD_TU group was similar to that in the PR_TU group.

Fig. 4
figure 4

Characterization of T/NK cells between PD group and PR group. A UMAP projection of subclustered T/NK cells from 9 treatment-naive ES-SCLC samples. Each dot corresponds to a single cell, colored by cell type and group. B Heatmap of marker gene expression in T/NK cell subtypes. C The proportion of T/NK cell subtypes among three groups. D Differentially expressed pathways are scored per cell by GSVA among T/NK cell subtypes. E Potential developmental trajectory of CD4+ T cells inferred by Monocle. Each dot corresponds to a single cell, colored by pseudotime value, cell type and group

Full size image

For CD4+ T cells, the relative percentages of exhausted CD4+ T cells and Treg cells in the PD_TU group were reduced compared to the PR_TU group, while the relative percentage of memory or effector CD4+ T cells was higher than that in the PR_TU group. GSVA enrichment analysis by differentially expressed pathways scored per cell showed that memory or effector CD4+ T cells exhibited notable enrichment in pathways related to tumor necrosis factor-α (TNFA) signaling via NFκB (Fig. 4D). Pseudotime trajectory analysis to model changes in the transcriptome over time indicated that the PD_TU group had a broader distribution throughout the differentiation process compared to the PR_TU group (Fig. 4E).

For CD8+ T cells, the relative percentage of exhausted CD8+ T cells in the PD_TU group was reduced compared to the PR_TU group, while memory or effector CD8+ T cells were more abundant in the PD_TU group. There were no significant differences in the potential developmental trajectory of CD8+ T cells between the two groups (Supplementary Fig. 4D).

The PD_TU group is strongly enriched with macrophages and TAMs

A total of 8144 myeloid cells (1846 from the PD_TU group, 5739 from the PR_TU group and 559 from the PD_PE group) were obtained from three groups and further clustered as neutrophils, monocytes, macrophages, myeloid dendritic cells (mDCs), plasmacytoid dendritic cells (pDC), tumor-associated macrophages (TAMs) and others (Fig. 5A, B). The variations in the proportions of these myeloid cell subtypes in the 9 samples among three groups were presented (Fig. 5C, Supplementary Fig. 5A, B and C). The PD_PE group was enriched with mDCs and neutrophils. In the PD_TU group, the relative abundance of neutrophils decreased compared to the PR_TU group, while that of macrophages and TAMs increased.

Fig. 5
figure 5

Characterization of myeloid cells between PD group and PR group. A UMAP projection of subclustered myeloid cells from 9 treatment-naive ES-SCLC samples. Each dot corresponds to a single cell, colored by cell type and group. B Heatmap of marker gene expression in myeloid cell subtypes. C The proportion of myeloid cell subtypes among three groups. D Differentially expressed pathways are scored per cell by GSVA among myeloid cell subtypes. E Potential developmental trajectory of myeloid cells inferred by Monocle. Each dot corresponds to a single cell, colored by pseudotime value, cell type and group

Full size image

GSVA linked differentially expressed pathways in macrophages to the reactive oxygen species pathway, Notch signaling, and interferon gamma response, while angiogenesis was highly enriched in TAMs (Fig. 5D). Based on Monocle, the potential developmental trajectory of myeloid cells in the PD_TU group was comparable to that in the PR_TU group (Fig. 5E).

Immunomodulatory fibroblasts are highly abundant in the PD_TU group

We subclustered 2531 fibroblasts (1664 from the PD_TU group, 771 from the PR_TU group and 96 from the PD_PE group) from three groups as cancer-associated fibroblasts (CAFs), myofibroblasts, pericytes, immunomodulatory fibroblasts, lipofibroblasts and others (Fig. 6A, B). Notably, the pericytes were categorized as subtypes within fibroblasts as reported previously [12, 14]. The variations in the proportions of these fibroblasts subtypes in the 9 samples among three groups were revealed (Fig. 6C, Supplementary Fig. 6A, B and C), and there were no fibroblasts present in one PR sample. The distribution of fibroblasts subtypes in pleural effusion significantly differed from that in tumor tissues. For CAFs, myofibroblasts, pericytes and lipofibroblasts, the PD_TU group had a similar distribution as the PR_TU group. The immunomodulatory fibroblasts were highly abundant in the PD_TU group compared to the PR_TU group.

Fig. 6
figure 6

Characterization of fibroblasts between PD group and PR group. A UMAP projection of subclustered fibroblasts from 9 treatment-naive ES-SCLC samples. Each dot corresponds to a single cell, colored by cell type and group. B Heatmap of marker gene expression in fibroblasts subtypes. C The proportion of fibroblasts subtypes among three groups. D Differentially expressed pathways are scored per cell by GSVA among fibroblasts subtypes. E Potential developmental trajectory of fibroblasts inferred by Monocle. Each dot corresponds to a single cell, colored by pseudotime value, cell type and group

Full size image

Next, GSVA enrichment analysis linked differentially expressed hallmark pathways in immunomodulatory fibroblasts to epithelial-mesenchymal transition (EMT) and angiogenesis (Fig. 6D). The inferred developmental trajectory of fibroblasts exhibited a branched structure, and the PD_TU group was abundant at the initiation stage (Fig. 6E).

Discussion

Despite its rapid growth and early metastasis, SCLC is more chemosensitive than other lung cancers. However, there is still a proportion of ES-SCLC patients who do not respond to chemotherapy exhibiting a poorer prognosis due to inter- and intratumoral heterogeneity [14, 27]. Since there has few reports focusing on this field, we have comprehensively characterized the heterogeneity of malignant cells, immune cells, and stromal cells within the TME in treatment-naive ES-SCLC patients with different first-line chemotherapy responses in this study.

According to transcriptomic landscape and cell type composition, the PD samples represent a multicellular ecosystem distinct from the PR samples. Based on DEG analyses and enrichment analyses, immune response and higher expression of immune response-associated genes LTF, SLPI, SPARC and IGLV1-51, as well as ribosome-associated genes RPS4Y1 and FOSB, might correlate with a poor first-line chemotherapy response in ES-SCLC. The immune system is a critical determinant of chemotherapy responses to cancers [28], and immune response within the TME is clinically relevant [29]. So far, no biomarkers have been identified for predicting chemotherapy responses in ES-SCLC. Therefore, we selected several highly expressed immune response-associated genes as potential new markers. For example, suppressing SLPI may inhibit the growth of human colorectal cancer cells and has been identified as a novel resistance factor to cisplatin [30]. Overexpression of the gene SPARC has been reported as one of the major causes of chemotherapy resistance to pancreatic adenocarcinoma [31]. Hence, these genes might be further explored as potential therapeutic targets against ES-SCLC. Additionally, we discovered the malignant cells exhibited the highest communication patterns with fibroblasts and endothelial cells in the PD_TU group and PR_TU group, respectively, which indicated the cell–cell interaction mechanisms might be involved in chemosensitivity in ES-SCLC.

We also found that T cells and effector T cell subtypes were more abundant in the PD_TU group, with a significant enrichment in TNFA signaling via NFκB. The PD_TU group showed a significant increase in macrophages and TAMs, as well as the angiogenesis pathway in TAMs, which might correlate with a poor response to first-line chemotherapy in ES-SCLC. This TAM subtype was similar to a previously reported TAM in colon cancer that promoted tumor angiogenesis [32], and in lung adenocarcinoma with lymph node metastasis that expressed higher vascular endothelial growth factor A (VEGFA) [12]. TAMs are pivotal in cancer progression, influencing tumor growth, angiogenesis, and immune evasion, as well as chemoresistance [33,34,35]. Accordingly, combining chemotherapy with anti-angiogenic therapy may be more effective against ES-SCLC than monotherapies. Immunomodulatory fibroblasts were highly abundant in the PD_TU group, and the pathways of EMT and angiogenesis were upregulated. The EMT is a potential driver of invasion and metastasis by human epithelial tumors [36], and may mediate chemosensitivity and chemoresistance in SCLC [37, 38]. A recent proteogenomic characterization study of SCLC identified a subtype with the highest EMT score, which was associated with SCLC chemoresistance [39]. Thus, inhibition of the EMT pathway may enhance chemosensitivity to ES-SCLC.

In addition, we characterized the transcriptomic landscape and cell type composition of a pleural effusion sample with a poor first-line chemotherapy response in ES-SCLC in this study. We found the average proportion of cell types and subtypes in pleural effusion of ES-SCLC, which was strongly enriched with immune cells, varied substantially from that in tumor tissues of ES-SCLC.

There are several limitations in our study. First, we were unable to compare samples before and after chemotherapy, so it is impossible to comprehensively decode the tumor heterogeneity associated with chemotherapy responses. Second, our observations are based on bioinformatic analyses, so the results need to be validated in experiments with patient tissues or animal models. Third, we analyzed a pleural effusion PD sample and lacked a PR sample as the control.

Conclusion

In summary, our study provides the first comprehensive understanding of the cellular and molecular heterogeneity in treatment-naive patients with ES-SCLC with different chemotherapy responses by scRNA-seq. The variations of transcriptomic features of malignant cells, immune cells and stromal cells were observed. The immune response, EMT and angiogenesis might correlate with a poor first-line chemotherapy response. The findings provide new insights into the biological basis of chemotherapy responses in ES-SCLC, helping to identify factors that contribute to better or worse treatment efficacy.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

CAFs:

Cancer-associated fibroblasts

CT:

CNVs  Copy number variations

CT:

Computed tomography

EBUS:

Endobronchial ultrasound

EMT:

Epithelial-mesenchymal transition

ES-SCLC:

Extensive-stage small cell lung cancer

GO:

Gene Ontology

GSVA:

Gene set variation analysis

IGLV1-51:

Immunoglobulin lambda variable 1–51

KEGG:

Kyoto Encyclopedia of Genes and Genomes

LTF:

Lactotransferrin

NK:

Natural killer

PD:

Progressive disease

PR:

Partial response

RPS4Y1:

Ribosomal protein S4 Y-linked 1

SCLC:

Small cell lung cancer

scRNA-seq:

Single-cell RNA sequencing

SLPI:

Secretory leukocyte peptidase inhibitor

SPARC:

Secreted protein acidic and rich in cysteine

TAMs:

  Tumor-associated macrophages

TME:

Tumor microenvironment

TNFA:

Tumor necrosis factor-α

UMAP:

Uniform manifold approximation and projection

References

  1. Wang Q, Gümüş ZH, Colarossi C, Memeo L, Wang X, Kong CY, et al. SCLC: epidemiology, risk factors, genetic susceptibility, molecular pathology, screening, and early detection. J Thorac Oncol. 2023;18(1):31–46.

    Article  CAS  PubMed  Google Scholar 

  2. Remon J, Aldea M, Besse B, Planchard D, Reck M, Giaccone G, et al. Small cell lung cancer: a slightly less orphan disease after immunotherapy. Ann Oncol. 2021;32(6):698–709.

    Article  CAS  PubMed  Google Scholar 

  3. Rudin CM, Brambilla E, Faivre-Finn C, Sage J. Small-cell lung cancer. Nat Rev Dis Primers. 2021;7(1):3.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Detterbeck FC, Nicholson AG, Franklin WA, Marom EM, Travis WD, Girard N, et al. The IASLC lung cancer staging project: summary of proposals for revisions of the classification of lung cancers with multiple pulmonary sites of involvement in the forthcoming eighth edition of the TNM classification. J Thorac Oncol. 2016;11(5):639–50.

    Article  PubMed  Google Scholar 

  5. Oronsky B, Reid TR, Oronsky A, Carter CA. What’s new in SCLC? A review. Neoplasia. 2017;19(10):842–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Gay CM, Stewart CA, Park EM, Diao L, Groves SM, Heeke S, et al. Patterns of transcription factor programs and immune pathway activation define four major subtypes of SCLC with distinct therapeutic vulnerabilities. Cancer Cell. 2021;39(3):346–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Li X, Tong L, Wang S, Yu J, Lu B, Wang Q, et al. Integration of clinical and blood parameters in risk prognostication for patients receiving immunochemotherapy for extensive stage small cell lung cancer: real-world data from two centers. BMC Med. 2024;22(1):381.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Zhu L, Qin J. Predictive biomarkers for immunotherapy response in extensive-stage SCLC. J Cancer Res Clin Oncol. 2024;150(1):22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Saida Y, Watanabe S, Kikuchi T. Extensive-stage small-cell lung cancer: current landscape and future prospects. Onco Targets Ther. 2023;16:657–71.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Horn L, Mansfield AS, Szczęsna A, Havel L, Krzakowski M, Hochmair MJ, et al. First-line atezolizumab plus chemotherapy in extensive-stage small-cell lung cancer. N Engl J Med. 2018;379(23):2220–9.

    Article  CAS  PubMed  Google Scholar 

  11. Altorki NK, Markowitz GJ, Gao D, Port JL, Saxena A, Stiles B, et al. The lung microenvironment: an important regulator of tumour growth and metastasis. Nat Rev Cancer. 2019;19(1):9–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Xing X, Yang F, Huang Q, Guo H, Li J, Qiu M, et al. Decoding the multicellular ecosystem of lung adenocarcinoma manifested as pulmonary subsolid nodules by single-cell RNA sequencing. Sci Adv. 2021;7(5):eabd9738.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Tian Y, Li Q, Yang Z, Zhang S, Xu J, Wang Z, et al. Single-cell transcriptomic profiling reveals the tumor heterogeneity of small-cell lung cancer. Signal Transduct Target Ther. 2022;7(1):346.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Cheng C, Zhu G, Li Y, Wang H, Wang S, Li C, et al. Single-cell transcriptome analysis reveals cellular and molecular alterations in small cell lung cancer tumors following chemotherapy. Int J Cancer. 2023;153(6):1273–86.

    Article  CAS  PubMed  Google Scholar 

  15. Xu R, Shao J, Liu J, Qu B, Liu J. Single-cell analysis of chemotherapy-resistant microenvironment identifies a chemo-response biomarker for pancreatic cancer. J Gastrointest Oncol. 2024;15(4):1836–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Lim B, Lin Y, Navin N. Advancing cancer research and medicine with single-cell genomics. Cancer Cell. 2020;37(4):456–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Macosko EZ, Basu A, Satija R, Nemesh J, Shekhar K, Goldman M, et al. Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. Cell. 2015;161(5):1202–14.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Satija R, Farrell JA, Gennert D, Schier AF, Regev A. Spatial reconstruction of single-cell gene expression data. Nat Biotechnol. 2015;33(5):495–502.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Korsunsky I, Millard N, Fan J, Slowikowski K, Zhang F, Wei K, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods. 2019;16(12):1289–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Reiner A, Yekutieli D, Benjamini Y. Identifying differentially expressed genes using false discovery rate controlling procedures. Bioinformatics. 2003;19(3):368–75.

    Article  CAS  PubMed  Google Scholar 

  21. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013;14:7.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7): e47.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM, Wakimoto H, et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science. 2014;344(6190):1396–401.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Jin S, Guerrero-Juarez CF, Zhang L, Chang I, Ramos R, Kuan CH, et al. Inference and analysis of cell-cell communication using Cell Chat. Nat Commun. 2021;12(1):1088.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Qiu X, Mao Q, Tang Y, Wang L, Chawla R, Pliner HA, et al. Reversed graph embedding resolves complex single-cell trajectories. Nat Methods. 2017;14(10):979–82.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Murayama T, Mahadevan NR, Meador CB, Ivanova EV, Pan Y, Knelson EH, et al. Targeting TREX1 induces innate immune response in drug-resistant small-cell lung cancer. Cancer Res Commun. 2024;4(9):2399–414.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Chan JM, Quintanal-Villalonga Á, Gao VR, Xie Y, Allaj V, Chaudhary O, et al. Signatures of plasticity, metastasis, and immunosuppression in an atlas of human small cell lung cancer. Cancer Cell. 2021;39(11):1479–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Tilsed CM, Fisher SA, Nowak AK, Lake RA, Lesterhuis WJ. Cancer chemotherapy: insights into cellular and tumor microenvironmental mechanisms of action. Front Oncol. 2022;12: 960317.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Zhang X, Yang J, Du L, Zhou Y, Li K. The prognostic value of immunoscore in patients with cancer: a pooled analysis of 10,328 patients. Int J Biol Markers. 2020;35(3):3–13.

    Article  CAS  PubMed  Google Scholar 

  30. Wei Z, Liu G, Jia R, Zhang W, Li L, Zhang Y, et al. Inhibition of secretory leukocyte protease inhibitor (SLPI) promotes the PUMA-mediated apoptosis and chemosensitivity to cisplatin in colorectal cancer cells. Discov Oncol. 2023;14(1):1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Thota R, Pauff JM, Berlin JD. Treatment of metastatic pancreatic adenocarcinoma: a review. Oncology (Williston Park). 2014;28(1):70–4.

    PubMed  Google Scholar 

  32. Zhang L, Li Z, Skrzypczynska KM, Fang Q, Zhang W, O’Brien SA, et al. Single-cell analyses inform mechanisms of myeloid-targeted therapies in colon cancer. Cell. 2020;181(2):442–59.

    Article  CAS  PubMed  Google Scholar 

  33. Chu X, Tian Y, Lv C. Decoding the spatiotemporal heterogeneity of tumor-associated macrophages. Mol Cancer. 2024;23(1):150.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Shao S, Miao H, Ma W. Unraveling the enigma of tumor-associated macrophages: challenges, innovations, and the path to therapeutic breakthroughs. Front Immunol. 2023;14:1295684.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. McWhorter R, Bonavida B. The role of TAMs in the regulation of tumor cell resistance to chemotherapy. Crit Rev Oncog. 2024;29(4):97–125.

    Article  PubMed  Google Scholar 

  36. Pastushenko I, Blanpain C. EMT transition states during tumor progression and metastasis. Trends Cell Biol. 2019;29(3):212–26.

    Article  CAS  PubMed  Google Scholar 

  37. Stewart CA, Gay CM, Xi Y, Sivajothi S, Sivakamasundari V, Fujimoto J, et al. Single-cell analyses reveal increased intratumoral heterogeneity after the onset of therapy resistance in small-cell lung cancer. Nat Cancer. 2020;1:423–36.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Cañadas I, Rojo F, Taus Á, Arpí O, Arumí-Uría M, Pijuan L, et al. Targeting epithelial-to-mesenchymal transition with Met inhibitors reverts chemoresistance in small cell lung cancer. Clin Cancer Res. 2014;20(4):938–50.

    Article  PubMed  Google Scholar 

  39. Liu Q, Zhang J, Guo C, Wang M, Wang C, Yan Y, et al. Proteogenomic characterization of small cell lung cancer identifies biological insights and subtype-specific therapeutic strategies. Cell. 2024;187(1):184-203.e28.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We thank all the partners and staff who helped us in the process of this study.

Funding

The study was supported by the Science and Technology Commission of Shanghai Municipality Program (No. 19401930800, No. 21Y11921900), the Shanghai Pulmonary Hospital Program (No. fkgg1807, No. fkzr2119), and the National Natural Science Foundation of China (No. 82004115).

Author information

Authors and Affiliations

  1. Department of Integrative Medicine, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Shanghai, China

    Zhan Gu, Yongqing Heng, Rui Fan & Lixia Ju

  2. Department of Medical Oncology, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Shanghai, China

    Jie Luo

Authors
  1. Zhan Gu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Yongqing Heng
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Rui Fan
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Jie Luo
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Lixia Ju
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

ZG and LXJ designed the study; ZG, YQH and RF collected the data and completed the experiments; ZG and JL analyzed the data; ZG wrote the manuscript; LXJ revised the manuscript content. All authors have read and approved the final submitted manuscript.

Corresponding author

Correspondence to Lixia Ju.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Shanghai Pulmonary Hospital affiliated to Tongji University (K24-530), and written informed consent was obtained from all patients.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gu, Z., Heng, Y., Fan, R. et al. Single-cell RNA sequencing reveals cellular and molecular heterogeneity in extensive-stage small cell lung cancer with different chemotherapy responses. Cancer Cell Int 25, 157 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12935-025-03785-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12935-025-03785-z

Keywords

Cancer Cell International

ISSN: 1475-2867

Contact us