pubmed.ncbi.nlm.nih.gov

Multi-omics-driven discovery of invasive patterns and treatment strategies in CA19-9 positive intrahepatic cholangiocarcinoma - PubMed

  • ️Mon Jan 01 2024

. 2024 Nov 15;22(1):1031.

doi: 10.1186/s12967-024-05854-9.

Affiliations

Multi-omics-driven discovery of invasive patterns and treatment strategies in CA19-9 positive intrahepatic cholangiocarcinoma

Delin Ma et al. J Transl Med. 2024.

Abstract

Background: Intrahepatic cholangiocarcinoma (ICC) is a malignant tumor with a poor prognosis, predominantly CA19-9 positive. High CA19-9 levels correlate with increased aggressiveness and worse outcomes. This study employs multi-omics analysis to reveal molecular features and identify therapeutic targets of CA19-9 positive ICC, aiming to support individualized treatment.

Methods: Data from seven clinical cohorts, two whole-exome sequencing cohorts, six RNA sequencing/microarray cohorts, one proteomic cohort, 20 single-cell RNA sequencing samples, and one spatial transcriptome sample were analyzed. Key findings were validated on tissue microarrays from 52 ICC samples.

Results: CA19-9 positive ICC exhibited poorer OS (median 24.1 v.s. 51.5 months) and RFS (median 11.7 v.s. 28.2 months) compared to negative group (all P < 0.05). Genomic analysis revealed a higher KRAS mutation frequency in the positive group and a greater prevalence of IDH1/2 mutations in the negative group (all P < 0.05). Transcriptomic analysis indicated upregulated glycolysis pathways in CA19-9 positive ICC. Single-cell analysis identified specific glycolysis-related cell subclusters associated with poor prognosis, including Epi_SLC2A1, CAF_VEGFA, and Mph_SPP1. Higher hypoxia in the CA19-9 positive group led to metabolic reprogramming and promoted these cells' formation. These cells formed interactive communities promoting epithelial-mesenchymal transition (EMT) and angiogenesis. Drug sensitivity analysis identified six potential therapeutic drugs.

Conclusions: This study systematically elucidated the clinical, genomic, transcriptomic, and immune features of CA19-9 positive ICC. It reveals glycolysis-associated cellular communities and their cancer-promoting mechanisms, enhancing our understanding of ICC and laying the groundwork for individualized therapeutic strategies.

Keywords: CA19-9; Glycolysis; Hypoxia; Intrahepatic cholangiocarcinoma; Multi-omics.

© 2024. The Author(s).

PubMed Disclaimer

Conflict of interest statement

Declarations Ethics approval and consent to participate This study was approved by the Peking University People’s Hospital Ethics Committee. All procedures involving human participants were performed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study. Consent for publication Consent for publication was obtained from all individual participants included in the study. Competing interests The authors declare that they have no competing interests.

Figures

Fig. 1
Fig. 1

Survival analysis and aggressive characteristics of CA19-9 positive ICC patients in combined cohort. A, B The KM curves show that patients in the CA19-9 positive group exhibit poorer OS (A) and RFS (B) in the combined ICC cohort. C–F The bar plots show the proportion differences of perineural invasion (C), vascular invasion (D), lymph node metastasis (E), and distal metastasis (F) between CA19-9 positive and negative groups in the combined ICC cohort. G, H The forest plots display the multivariate Cox regression result for OS (G) and RFS (H) in the combined ICC cohort. P-values were calculated using the Log-rank test in (A, B), the Chi-square test in (C–F), and the multivariate COX regression analysis in (G, H)

Fig. 2
Fig. 2

Gene mutation pattern and tumor mutation burden of CA19-9 positive patients in the ZS-ICC cohort. A The waterfall plot shows the top 50 mutated genes in the ZS-ICC cohort. The genes marked in red are common to both cohorts. B The bar plot shows the mutation frequency differences of the ten shared high-ranked genes between CA19-9 positive and negative groups in the ZS-ICC cohort. ns: no significance. *P < 0.05. C The bar plot shows the genes with significant mutation frequency among all genes between CA19-9 positive and negative groups in the ZS-ICC cohort. Red represents a higher frequency in the positive group, while blue represents a higher frequency in the negative group. D, E The KM curves of OS between patients with and without KRAS (D) and IDH1/2 (E) mutations in the ZS-ICC cohort. F The violin plot shows the difference in tumor mutation burden between the CA19-9 positive and negative groups in the ZS-ICC cohort. P-values were calculated Using the Chi-square test in (B, C), the Log-rank test in (D, E), and the Student’s t-test in (F)

Fig. 3
Fig. 3

The analysis of the transcriptome and proteome between CA19-9 positive and negative patients. A–C The volcano plots show the significantly differentially expressed genes (logFC > 0.5 & P < 0.05) between CA19-9 positive and negative groups in ZS-ICC (A), TCGA-ICC (B) and GSE107943-ICC (C) cohorts. D–F The bar plots show the differences of GSEA on hallmark pathways between CA19-9 positive and negative groups in ZS-ICC (D), TCGA-ICC (E), and GSE107943-ICC (F) cohorts. G The Venn diagram shows upregulated hallmark pathways overlapping across different ICC cohorts. H–J The KM curves show that patients in the CA19-9 positive group exhibit poorer OS in ZS-ICC (H), TCGA-ICC (I), and GSE107943-ICC (J) cohorts. K–M The scatter plots show the correlation between the serum CA19-9 levels and hypoxia glycolysis scores in ZS-ICC (K), TCGA-ICC (L) and GSE107943-ICC (M) cohorts. N The volcano plots show the significantly differentially expressed proteins (logFC > 0.5 & P < 0.05) between CA19-9 positive and negative groups in the ZS-ICC cohort. O The bubble plot shows the enrichment results on glycolysis-related pathways of the upregulated proteins (logFC > 0 & P < 0.05) in the ZS-ICC cohort. P-values were calculated using the Log-rank test in (H–J) and the Spearman correlation test in (K–M)

Fig. 4
Fig. 4

Single-cell analysis for all cells and epithelial cells. A The UMAP plot shows all cells colored by major cell types. B The dot plot shows the expression of the canonical marker genes across the eight major cell types. C The bar plot shows the numbers of each major cell type. D The UMAP feature plot shows the hallmark glycolysis scores in all cells. E The violin plot shows the difference in hallmark glycolysis scores in eight major cell types. F The heatmap shows eight major cell types’ interaction number differences between CA19-9 positive and negative groups. G The UMAP plot shows epithelial cells colored by seven epithelial subclusters. H The dot plot shows the expression of the marker genes across the seven epithelial subclusters. I, J The bar plot (I) and heatmap (J) show the proportion differences and Ro/e values of seven epithelial subclusters between CA19-9 positive and negative groups. *: Ro/e > 1. K The UMAP feature plots show the hallmark glycolysis scores (left) and glycolysis/gluconeogenesis scores (right) in epithelial cells. L The violin plots compare hallmark glycolysis scores (left) and glycolysis/gluconeogenesis scores (right) between Epi_SLC2A1 and other epithelial cells. M The GSEA result of hallmark glycolysis pathway between Epi_SLC2A1 and other epithelial cells. N The dot plot shows the expression of the LDHA, HK2, and PKM across the seven epithelial subclusters. O The violin plot shows the differences in GRCC scores between the CA19-9 positive and negative groups in the ZS-ICC cohort (up left panel). The KM curves compare the high and low GRCC score groups on OS in the ZS-ICC cohort (upright panel). The scatter plots show the correlation between GRCC scores, hallmark glycolysis scores (bottom left panel), and serum CA19-9 levels (bottom left panel) in the ZS-ICC cohort. P-values were calculated using the Student’s t-test in (L) and (O) (up left panel), the Log-rank test in (O) (upright panel), and the Spearman correlation test in (O) (bottom panel)

Fig. 5
Fig. 5

Glycolysis-related cell subclusters were verified using mIF and spatial transcriptomics analysis. A–C The representative mIF images of Epi_SLC2A1 (A, blue: DAPI, green: GLUT1, red: CK19), CAF_VEGFA (B, blue: DAPI, green: VEGFA, red: αSMA), and Mph_SPP1 (C, blue: DAPI, green: SPP1, red: CD68) in CA19-9 positive (bottom panel) and negative (up panel) groups. Scale bar = 50um. D The box plots show the differences in counts/unit area in Epi_SLC2A1 (left panel), CAF_VEGFA (mid panel), and Mph_SPP1 (right panel) between CA19-9 positive and negative groups (positive: n = 14, negative: n = 12). E The UMAP (up left panel) and spatial (up right panel) plots show all spots colored by tissue types. The spatial feature plots show the expression of KRT19 (bottom left panel) and ALB in all spots (bottom left panel). F The spatial feature plots show the Epi_SLC2A1 (left panel panel), CAF_VEGFA (mid panel panel), and Mph_SPP1 (right panel panel) signature scores in tumor region spots. G The correlation analysis between the Epi_SLC2A1 and CAF_VEGFA signature scores, between the Epi_SLC2A1 and Mph_SPP1 signature scores, and between the CAF_VEGFA and Mph_SPP1 signature scores in tumor region spots. H The violin plots show that the CA19-9 positive group has higher GRCC scores in the ZS-ICC cohort (up left panel). The KM curves show that the high GRCC scores group has a poorer OS in ZS-ICC cohorts (upright panel). The scatter plots show that the GRCC scores significantly correlate with hallmark glycolysis scores (bottom left panel) and serum CA19-9 levels (bottom right panel) in the ZS-ICC cohort. I The forest plot shows the multivariate Cox regression result for OS in the ZS-ICC cohort. P-values were calculated using the Mann-Whitney U test in (D), the Pearson correlation test in (G), the Student’s t-test in (H) (up left panel), the Log-rank test in (H) (upright panel), and the Spearman correlation test in (H) (bottom panel).

Fig. 6
Fig. 6

The analysis of evolutionary trajectories and cellular interaction network. A-C. Trajectory plots of epithelial cells (A), stromal cells (B), and macrophages (C) colored by hallmark OXPHOS (left panel) and glycolysis (right panel) scores. D–F The correlation analysis between pseudotime and hallmark OXPHOS (left panel) and glycolysis (right panel) scores in epithelial cells (D), lineage 3 of stromal cells (E), and lineage 2 of macrophages (F). G The correlation analysis between pseudotime and hallmark hypoxia scores in epithelial cells (left panel), lineage 3 of stromal cells (mid panel), and lineage 2 of macrophages (right panel). H The scatter plots show the correlation between the hallmark hypoxia scores of each sample and the proportion of Epi_SLC2A1 (left panel), CAF_VEGFA (mid panel), and Mph_SPP1 (right panel). I The representative IHC images of the HIF1αstaining in CA19-9 positive (up panel) and negative group (bottom panel). Scale bar = 100um. J The box plot shows the differences in the fraction of HIF1αstaining between CA19-9 positive and negative groups (positive: n = 31, negative: n = 22). K The heatmaps of Nichenet analysis show regulatory patterns between CAF_VEGFA to Epi_SLC2A1 (up panel) and Mph_ SPP1 to Epi_SLC2A1 (bottom panel). L The bar plots show the GO and KEGG pathways enrichment results of the predicted target genes in Epi_SLC2A1. P-values were calculated using the Student’s t-test in (J) and the Spearman correlation test in (D–H)

Fig. 7
Fig. 7

The effects of glycolysis-related cellular communities on EMT, angiogenesis, and drug screening strategy. A The GSEA result of hallmark EMT pathway between Epi_SLC2A1 and other epithelial cells. B The UMAP feature (up panel) and violin (bottom panel) plots show the Epi_SLC2A1 has higher hallmark EMT scores compared to other epithelial cells. C The trajectory plot (up panel) shows the hallmark EMT scores and the correlation analysis (bottom panel) between pseudotime and hallmark EMT scores in epithelial cells. D The circle plots show the differences in the number (up panel) and intensity (bottom panel) of interactions within the three glycolysis-related cell subclusters between CA19-9 positive and negative groups. Red represents increasing, and blue represents decreasing in the positive group. E The dot plot shows the differences in TGFB1 signaling interaction of three glycolysis-related cell subclusters to Epi_SLC2A1 between CA19-9 positive and negative groups. F The dot plots show the expression of the VEGFA across the subclusters of stromal cells (left panel), epithelial cells (mid panel), and macrophages (right panel). G The dot plot shows the differences in VEGFA signaling interaction of three glycolysis-related cell types to endothelial cells between CA19-9 positive and negative groups. H The UMAP plot shows endothelial cells colored by seven endothelial subclusters. I The dot plot shows the differences in VEGFA signaling interaction between the three glycolysis-related cell subclusters and seven endothelial subclusters between CA19-9 positive and negative groups. J The overall workflow for screening candidate drugs targeting GRCC and CA19-9 positive ICC patients. P-values were calculated using the Student’s t-test in (B) (bottom panel) and the Pearson correlation test in C (bottom panel)

Similar articles

References

    1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229–63. - PubMed
    1. Ali I, Wani WA, Haque A, Saleem K. Glutamic acid and its derivatives: candidates for rational design of anticancer drugs. Future Med Chem. 2013;5(8):961–78. - PubMed
    1. Ali I, Wani WA, Saleem K, Wesselinova D. Syntheses, DNA binding and anticancer profiles of l-glutamic acid ligand and its copper(II) and ruthenium(III) complexes. Med Chem. 2013;9(1):11–21. - PubMed
    1. Ali I, Wani WA, Saleem K, Hsieh M-F. Anticancer metallodrugs of glutamic acid sulphonamides: in silico, DNA binding, hemolysis and anticancer studies. RSC Adv. 2014;4(56):29629–41.
    1. Ilyas SI, Affo S, Goyal L, Lamarca A, Sapisochin G, Yang JD, et al. Cholangiocarcinoma—novel biological insights and therapeutic strategies. Nat Rev Clin Oncol. 2023;20(7):470–86. - PMC - PubMed

MeSH terms

Substances

LinkOut - more resources