CoChat

IDH Mutation Subgroup Status Associates with Intratumor Heterogeneity and the Tumor Microenvironment in Intrahepatic Cholangiocarcinoma

Open access

The DOI matches its Crossref record; 56 of the 57 references it cites were found in a registry or an index.

Abstract

Intrahepatic cholangiocarcinoma (ICC) is highly heterogeneous. Here, the authors perform exome sequencing and bulk RNA sequencing on 73 tumor regions from 14 ICC patients to portray the multi‐faceted intratumor heterogeneity (ITH) landscape of ICC. The authors show that ITH is highly concordant across genomic, transcriptomic, and immune levels. Comparison of these data to 8 published datasets reveals significantly higher degrees of ITH in ICC than hepatocellular carcinoma. Remarkably, the authors find that high‐ITH tumors highly overlap with the IDH (isocitrate dehydrogenase)‐mutant subgroup ( IDH ‐SG), comprising of IDH ‐mutated tumors and IDH ‐like tumors, that is, those IDH ‐wildtype tumors that exhibit similar molecular profiles to the IDH ‐mutated ones. Furthermore, IDH ‐SG exhibits less T cell infiltration and lower T cell cytotoxicity, indicating a colder tumor microenvironment (TME). The higher ITH and colder TME of IDH ‐SG are successfully validated by single‐cell RNA sequencing on 17 503 cells from 4 patients. Collectively, the study shows that IDH mutant subgroup status, rather than IDH mutation alone, is associated with ITH and the TME of ICC tumors. The results highlight that IDH ‐like patients may also benefit from IDH targeted therapies and provide important implications for the diagnosis and treatment of ICC.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1002/advs.202101230

Cites

56 of 56 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Research strategy and the genomic landscape. A) Research strategy. B) The genomic landscape. Top: number of somatic mutations, viral hepatitis, and ITH group. Middle: mutations of 19 driver genes across all the tumor regions. Driver genes were arranged by signaling pathways. Types of mutations and CNAs are indicated. Bottom: CNAs of 8 recurrent regions: dark red for amplifications, light red for gains, dark blue for deletions, and light blue for losses. Altered frequency based on patient and region is shown on the left.
Figure 1 Research strategy and the genomic landscape. A) Research strategy. B) The genomic landscape. Top: number of somatic mutations, viral hepatitis, and ITH group. Middle: mutations of 19 driver genes across all the tumor regions. Driver genes were arranged by signaling pathways. Types of mutations and CNAs are indicated. Bottom: CNAs of 8 recurrent regions: dark red for amplifications, light red for gains, dark blue for deletions, and light blue for losses. Altered frequency based on patient and region is shown on the left.CC BY · PubMed Central
Figure 2 Extensive ITH in ICC observed on the DNA level. A) Heatmaps show the regional distribution of all mutations in six selected patients. Clonal and subclonal mutations are marked in blue and light blue, respectively. The columns next to each heatmap show four categories of mutations and their percentages: trunk clonal mutations (purple); trunk subclonal mutations (sky blue); branch mutations (pale green); and region‐specific mutations (orange). Phylogenetic trees were constructed using a maximum parsimony algorithm based on mutations identified in each patient. The length of each line is proportional to the number of mutations. Mutations in potential driver genes are indicated, including CGC genes (black) and actionable genes (red). Sorafenib‐targeted amplifications are annotated in blue. Patient IDs and region names are labeled in each tree. MRCA, most recent ancestor, IM, intrahepatic metastasis, MO, multiple occurrences. B) Comparison of clonal ITH and mutation‐ITH. ** P < 0.01, paired Student's t test. C) Comparison of mutation‐ITH between ICC and HCC studies. In the box plots, the lines in the box indicate the median, the boxes indicate the first and third quartiles. * P < 0.05, ** P < 0.01, Student's t test. D) Comparison of the pooled mutation‐ITH of ICC and HCC studies. *** P < 0.001, Student's t test.
Figure 2 Extensive ITH in ICC observed on the DNA level. A) Heatmaps show the regional distribution of all mutations in six selected patients. Clonal and subclonal mutations are marked in blue and light blue, respectively. The columns next to each heatmap show four categories of mutations and their percentages: trunk clonal mutations (purple); trunk subclonal mutations (sky blue); branch mutations (pale green); and region‐specific mutations (orange). Phylogenetic trees were constructed using a maximum parsimony algorithm based on mutations identified in each patient. The length of each line is proportional to the number of mutations. Mutations in potential driver genes are indicated, including CGC genes (black) and actionable genes (red). Sorafenib‐targeted amplifications are annotated in blue. Patient IDs and region names are labeled in each tree. MRCA, most recent ancestor, IM, intrahepatic metastasis, MO, multiple occurrences. B) Comparison of clonal ITH and mutation‐ITH. ** P < 0.01, paired Student's t test. C) Comparison of mutation‐ITH between ICC and HCC studies. In the box plots, the lines in the box indicate the median, the boxes indicate the first and third quartiles. * P < 0.05, ** P < 0.01, Student's t test. D) Comparison of the pooled mutation‐ITH of ICC and HCC studies. *** P < 0.001, Student's t test.CC BY · PubMed Central
Figure 3 Extensive ITH in ICC observed on the RNA and TME levels. A) Hierarchical clustering of highly variable genes across ICC tumor samples. ITH subgroup, IDH mutation, and patient ID are annotated above the heatmap. The Spearman correlation coefficient was used as the distance metric for clustering. B) RNA heterogeneity quadrants for the ICC samples. RNA intratumor (y–axis) and intertumor heterogeneity (x–axis) are plotted on the axes as density curves. The plot is divided into quadrants by the mean intratumor (dashed horizontal line) and mean intertumor (dashed vertical line) heterogeneity scores. RNA heterogeneity quadrant is indicated for each cancer type as non‐significant (gray), significantly enriched (red; odds ratio > 1), or significantly depleted (blue; odd ratio < 1). Odds ratios are plotted on a natural log scale. Statistical significance was tested with a two‐sided Fisher's exact test. C) Survival association of RNA heterogeneity quadrants across 33 cancer types from the PRECOG database. Boxplots represent median values and 25th and 75th percentiles. The vertical bars span the fifth to 95th percentiles, Student's t test. D) Boxplot comparing the RNA‐ITH scores of cases from ICC_this study and HCC_Losic. et al. ** P < 0.01, Wilcoxon rank‐sum test. E) IHC of PD‐1 in 5 tumor regions from P10. Scale bar, 100 µm. F) Box plot showing the MD values of immune cells from ICC_this study and HCC_Losic. et al. Red dashed line denotes the MD value of 5. A narrow MD denotes a 0–5 range of MD values for all regions within a tumor. G) Bar plot comparing the proportion of cases with a narrow MD among ICC_this study, HCC_Losic. et al. and HCC_Shen. et al. *** P < 0.001, Chi‐square test. H) Box plots comparing the mutation‐ITH, RNA‐ITH, and immune‐ITH of high‐ITH and low‐ITH patients of our cohort. *: P < 0.05, **: P < 0.01, Student's t test. I) Scatter plots showing the Pearson correlations between clonal‐ITH and mutation‐ITH, between mutation‐ITH and RNA‐ITH, and between RNA‐ITH and immune‐ITH.
Figure 3 Extensive ITH in ICC observed on the RNA and TME levels. A) Hierarchical clustering of highly variable genes across ICC tumor samples. ITH subgroup, IDH mutation, and patient ID are annotated above the heatmap. The Spearman correlation coefficient was used as the distance metric for clustering. B) RNA heterogeneity quadrants for the ICC samples. RNA intratumor (y–axis) and intertumor heterogeneity (x–axis) are plotted on the axes as density curves. The plot is divided into quadrants by the mean intratumor (dashed horizontal line) and mean intertumor (dashed vertical line) heterogeneity scores. RNA heterogeneity quadrant is indicated for each cancer type as non‐significant (gray), significantly enriched (red; odds ratio > 1), or significantly depleted (blue; odd ratio < 1). Odds ratios are plotted on a natural log scale. Statistical significance was tested with a two‐sided Fisher's exact test. C) Survival association of RNA heterogeneity quadrants across 33 cancer types from the PRECOG database. Boxplots represent median values and 25th and 75th percentiles. The vertical bars span the fifth to 95th percentiles, Student's t test. D) Boxplot comparing the RNA‐ITH scores of cases from ICC_this study and HCC_Losic. et al. ** P < 0.01, Wilcoxon rank‐sum test. E) IHC of PD‐1 in 5 tumor regions from P10. Scale bar, 100 µm. F) Box plot showing the MD values of immune cells from ICC_this study and HCC_Losic. et al. Red dashed line denotes the MD value of 5. A narrow MD denotes a 0–5 range of MD values for all regions within a tumor. G) Bar plot comparing the proportion of cases with a narrow MD among ICC_this study, HCC_Losic. et al. and HCC_Shen. et al. *** P < 0.001, Chi‐square test. H) Box plots comparing the mutation‐ITH, RNA‐ITH, and immune‐ITH of high‐ITH and low‐ITH patients of our cohort. *: P < 0.05, **: P < 0.01, Student's t test. I) Scatter plots showing the Pearson correlations between clonal‐ITH and mutation‐ITH, between mutation‐ITH and RNA‐ITH, and between RNA‐ITH and immune‐ITH.CC BY · PubMed Central
Figure 4 High degree of ITH exhibited by the IDH mutation subgroup A) Unsupervised hierarchical clustering of genes related to chromatin modifier and metabolism across ICC samples. ITH status, IDH mutation, IDH‐SG, and patient ID are annotated. B) 33 cases from ICC_TCGA. ITH status, IDH mutation, IDH‐SG, and patient ID are annotated. C) Bar plot comparing the number of IDH‐SG patients in the groups of high‐ITH and low‐ITH patients in ICC_this study (left) and ICC_TCGA (right). *: P < 0.05, ***: P < 0.001, Chi‐square test. D) Box plot comparing the clonal‐ITH of IDH‐SG and IDH‐NO patients in ICC_this study (left) and ICC_TCGA (right). *: P < 0.05, ***: P < 0.001, Chi‐square test.
Figure 4 High degree of ITH exhibited by the IDH mutation subgroup A) Unsupervised hierarchical clustering of genes related to chromatin modifier and metabolism across ICC samples. ITH status, IDH mutation, IDH‐SG, and patient ID are annotated. B) 33 cases from ICC_TCGA. ITH status, IDH mutation, IDH‐SG, and patient ID are annotated. C) Bar plot comparing the number of IDH‐SG patients in the groups of high‐ITH and low‐ITH patients in ICC_this study (left) and ICC_TCGA (right). *: P < 0.05, ***: P < 0.001, Chi‐square test. D) Box plot comparing the clonal‐ITH of IDH‐SG and IDH‐NO patients in ICC_this study (left) and ICC_TCGA (right). *: P < 0.05, ***: P < 0.001, Chi‐square test.CC BY · PubMed Central
Figure 5 IDH mutation subgroup exhibited a distinct tumor microenvironment. A) Unsupervised hierarchical clustering of CD8+ T cell‐related markers in our study. B) Heatmap of CD8+ T cell‐related markers in the ICC_TCGA study. Boxplots comparing the expression of CD8+ T cell‐related markers between IDH‐SG and IDH‐NO patients C) in our study and D) the ICC_TCGA study. *: P < 0.05, **: P < 0.05, ***: P < 0.001, Student's t test. E) IHC of CD8 for six patients. Scale bar, 200 µm.
Figure 5 IDH mutation subgroup exhibited a distinct tumor microenvironment. A) Unsupervised hierarchical clustering of CD8+ T cell‐related markers in our study. B) Heatmap of CD8+ T cell‐related markers in the ICC_TCGA study. Boxplots comparing the expression of CD8+ T cell‐related markers between IDH‐SG and IDH‐NO patients C) in our study and D) the ICC_TCGA study. *: P < 0.05, **: P < 0.05, ***: P < 0.001, Student's t test. E) IHC of CD8 for six patients. Scale bar, 200 µm.CC BY · PubMed Central
Figure 6 Single‐cell analysis of IDH‐SG and IDH‐NO tumors. A) Schematic diagram of the scRNA‐seq analysis workflow. B) t‐SNE plots for cell type identification of 15 037 single cells from 4 ICC tumors. C) Large‐scale CNAs of single cells (rows) of 4 ICC tumors. Red, amplifications; blue, deletions. D) Epithelial scores of malignant and non‐malignant cells. In the boxplots, the central rectangles span the first quartile to the third quartile, with the segments inside the rectangle corresponding to the median. The whiskers extend 1.5 times the interquartile range. E) Heatmap showing gene expression programs deciphered from a representative tumor. F) Pearson correlation clustering of 40 intra‐tumor expression programs. The dot size is proportional to the absolute value of the correlation. G) GSVA analysis of malignant cells in IDH‐SG and IDH‐NO tumors based on the Hallmark Signature from Molecular Signatures Database (MSigDB). The normalized enrichment score (NES) was used to indicate enrichment of the related pathways. H) EMT scores of single cells of IDH‐SG and IDH‐NO tumors. Violin plots of EMT‐related markers (NNMT, VCAN, CDH2, FOXC2, VIM, and FN1) from IDH‐SG and IDH‐NO tumors. The width of a violin plot indicates the kernel density of the expression values. All P < 0.001, Student's t test. I) Stemness scores of single cells of IDH‐SG and IDH‐NO tumors. Violin plots of stemness‐related markers (CD44, SOX4, SOX6, ICAM1, CD47, and NES) from IDH‐SG and IDH‐NO tumors. All P < 0.001, Student's t test. J) Dot plot comparing the RNA‐ITH scores of IDH‐SG and IDH‐NO tumors.
Figure 6 Single‐cell analysis of IDH‐SG and IDH‐NO tumors. A) Schematic diagram of the scRNA‐seq analysis workflow. B) t‐SNE plots for cell type identification of 15 037 single cells from 4 ICC tumors. C) Large‐scale CNAs of single cells (rows) of 4 ICC tumors. Red, amplifications; blue, deletions. D) Epithelial scores of malignant and non‐malignant cells. In the boxplots, the central rectangles span the first quartile to the third quartile, with the segments inside the rectangle corresponding to the median. The whiskers extend 1.5 times the interquartile range. E) Heatmap showing gene expression programs deciphered from a representative tumor. F) Pearson correlation clustering of 40 intra‐tumor expression programs. The dot size is proportional to the absolute value of the correlation. G) GSVA analysis of malignant cells in IDH‐SG and IDH‐NO tumors based on the Hallmark Signature from Molecular Signatures Database (MSigDB). The normalized enrichment score (NES) was used to indicate enrichment of the related pathways. H) EMT scores of single cells of IDH‐SG and IDH‐NO tumors. Violin plots of EMT‐related markers (NNMT, VCAN, CDH2, FOXC2, VIM, and FN1) from IDH‐SG and IDH‐NO tumors. The width of a violin plot indicates the kernel density of the expression values. All P < 0.001, Student's t test. I) Stemness scores of single cells of IDH‐SG and IDH‐NO tumors. Violin plots of stemness‐related markers (CD44, SOX4, SOX6, ICAM1, CD47, and NES) from IDH‐SG and IDH‐NO tumors. All P < 0.001, Student's t test. J) Dot plot comparing the RNA‐ITH scores of IDH‐SG and IDH‐NO tumors.CC BY · PubMed Central
Figure 7 Comparison of non‐malignant cells between IDH‐SG and IDH‐NO. A) t‐SNE plot of non‐malignant cells from 4 tumors. Cells were annotated based on known lineage‐specific marker genes. 4 subclusters of T cells were highlighted below, including CD4+ T cells, CD8+ T cells, Tregs, and proliferating T cells. B) Average expression of cell type markers across different clusters. The dot size is proportional to the relative expression level of each gene. C) Expression of T cell markers across different clusters. D) Composition of non‐malignant cells from 4 tumors. E) Bar plot comparing the proportion of T cells and CAFs between IDH‐SG and IDH‐NO patients. *: P < 0.05, ***: P < 0.001, Student's t test. F) Bar plot comparing the proportion of subclusters of T cells between IDH‐SG and IDH‐NO patients. *: P < 0.05, ***: P < 0.001, Student's t test. G) Scatterplot showing DEGs in CD8+ T cells derived from IDH‐SG tumors in comparison with those from IDH‐NO tumors. H) The developmental trajectory of CD8+ T cells inferred by Monocle2. I) Heatmap showing scaled expression of dynamic genes along the pseudotime. Rows of the heatmap represent genes that show dynamic changes along the pseudotime, and these genes were clustered into two groups according to their expression pattern along the pseudotime. The color scheme represents the z‐score distribution from blue to red. 2D density plot of the cytotoxicity and exhaustion states of CD8+ T cells in J) IDH‐SG tumors and K) IDH‐NO tumors. Cells are partitioned into “high cytotoxicity & high exhaustion” (CyhighExhigh), “high cytotoxicity & low exhaustion” (CyhighExlow), “low cytotoxicity & high exhaustion” (CylowExhigh) and “low cytotoxicity & low exhaustion” (CylowExlow) groups. L) Cytotoxicity and exhaustion scores of IDH‐SG and IDH‐NO tumors. *: P < 0.05, ***: P < 0.001, Student's t test. M) Interaction analysis showing enriched receptor‐ligand pairs in T cell subclusters and malignant cells in IDH‐SG tumors (Top) and IDH‐NO tumors (down). N) Immunofluorescence for CD8, GZMB, and DAPI for four patients (800×).
Figure 7 Comparison of non‐malignant cells between IDH‐SG and IDH‐NO. A) t‐SNE plot of non‐malignant cells from 4 tumors. Cells were annotated based on known lineage‐specific marker genes. 4 subclusters of T cells were highlighted below, including CD4+ T cells, CD8+ T cells, Tregs, and proliferating T cells. B) Average expression of cell type markers across different clusters. The dot size is proportional to the relative expression level of each gene. C) Expression of T cell markers across different clusters. D) Composition of non‐malignant cells from 4 tumors. E) Bar plot comparing the proportion of T cells and CAFs between IDH‐SG and IDH‐NO patients. *: P < 0.05, ***: P < 0.001, Student's t test. F) Bar plot comparing the proportion of subclusters of T cells between IDH‐SG and IDH‐NO patients. *: P < 0.05, ***: P < 0.001, Student's t test. G) Scatterplot showing DEGs in CD8+ T cells derived from IDH‐SG tumors in comparison with those from IDH‐NO tumors. H) The developmental trajectory of CD8+ T cells inferred by Monocle2. I) Heatmap showing scaled expression of dynamic genes along the pseudotime. Rows of the heatmap represent genes that show dynamic changes along the pseudotime, and these genes were clustered into two groups according to their expression pattern along the pseudotime. The color scheme represents the z‐score distribution from blue to red. 2D density plot of the cytotoxicity and exhaustion states of CD8+ T cells in J) IDH‐SG tumors and K) IDH‐NO tumors. Cells are partitioned into “high cytotoxicity & high exhaustion” (CyhighExhigh), “high cytotoxicity & low exhaustion” (CyhighExlow), “low cytotoxicity & high exhaustion” (CylowExhigh) and “low cytotoxicity & low exhaustion” (CylowExlow) groups. L) Cytotoxicity and exhaustion scores of IDH‐SG and IDH‐NO tumors. *: P < 0.05, ***: P < 0.001, Student's t test. M) Interaction analysis showing enriched receptor‐ligand pairs in T cell subclusters and malignant cells in IDH‐SG tumors (Top) and IDH‐NO tumors (down). N) Immunofluorescence for CD8, GZMB, and DAPI for four patients (800×).CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 1

  1. 10.1016/j.cell.2017.05.046 · DOI · Paper page

Cited by

See all 71

Metadata

Authors
Xiang, X., Liu, Z., Zhang, C., Li, Z., Gao, J., Zhang, C., Cao, Q., Cheng, J., Liu, H., Chen, D., Cheng, Q., Zhang, N., Xue, R., Bai, F., & Zhu, J.
Affiliations
  • Department of Hepatobiliary Surgery, Peking University People's Hospital Beijing Key Surgical Basic Research Laboratory of Liver Cirrhosis and Liver Cancer Beijing 100044 China
  • Peking University (CN)
  • Peking University People's Hospital (CN)
  • Peking University First Hospital (CN)
Corresponding author
  • Xue, Ruidong
  • Bai, Fan
  • Zhu, Jiye
Venue
Advanced Science
Publisher
Wiley
Published
2021-07-11
Volume
8
Issue
17
Pages
e2101230-e2101230
ISSN
2198-3844 (ISSN-L)
OpenAlex
W3182403227
Cited by
71
References
57
Citations by year
2026: 13 · 2025: 12 · 2024: 21 · 2023: 12 · 2022: 10 · 2021: 3
FWCI
10.81
Type
journal article
Language
en
Open access
Locations
  • Advanced Science (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Greater South Information System
Funders
  • National Science and Technology Major Project (2018ZX10302205, 2019YFC1315702, 2018ZX10723204)
  • National Natural Science Foundation of China (81802813, 81988101, 81972656, 81570590, 81872508)
Topics
  • Health Sciences › Medicine › Surgery › Cholangiocarcinoma and Gallbladder Cancer Studies
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Cancer Genomics and Diagnostics
  • Health Sciences › Medicine › Surgery › Pediatric Hepatobiliary Diseases and Treatments
Keywords
intrahepatic cholangiocarcinoma, intratumoral heterogeneity, IDH mutation, tumor microenvironment, T cell infiltration, T cell cytotoxicity, tumor immune microenvironment, single-cell RNA sequencing
MeSH terms
Bile Duct Neoplasms, Humans, Isocitrate Dehydrogenase, Mutation, Cholangiocarcinoma, Tumor Microenvironment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.