Genomic and molecular characterization of esophageal squamous cell carcinoma

Authors

De-Chen Lin

Jia-Jie Hao

Yasunobu Nagata

Liang Xu

Li Shang

Xuan Meng

Yusuke Sato

Yusuke Okuno

Ana Maria Varela

Ling-Wen Ding

Manoj Garg

Li-Zhen Liu

Henry Yang

Dong Yin

Zhi-Zhou Shi

Yan-Yi Jiang

Wen-Yue Gu

Ting Gong

Yu Zhang

Xin Xu

Ori Kalid

Sharon Shacham

Seishi Ogawa

Ming-Rong Wang

H. Phillip Koeffler

Doi

10.1038/ng.2935

PMID: 24686850 · DOI: 10.1038/ng.2935 · Journal: Nature Genetics (2014)

TL;DR

Lin et al. performed whole-exome or targeted deep sequencing of 139 paired esophageal squamous cell carcinoma (ESCC) tumor/germline samples plus somatic copy number analysis of over 184 ESCC cases. They identified novel significantly mutated genes (FAT1, FAT2, ZNF750, KMT2D) alongside previously reported ESCC drivers (TP53, PIK3CA, NOTCH1), implicated APOBEC3B as a major mutagen in ESCC, found recurrent dysregulation of RTK-MAPK-PI3K, cell-cycle, and epigenetic pathways, and proposed XPO1 as a therapeutic target — demonstrating that the XPO1 inhibitor KPT-330 (selinexor) induced apoptosis and growth arrest in ESCC cell lines.

Cohort & data

  • Discovery Cohort: 20 paired ESCC germline/tumor samples, whole-exome sequencing (mean coverage 79×); RNA-seq on 4 of these 20 tumors.
  • Frequency Cohort: 119 additional matched ESCC germline/tumor pairs plus 10 ESCC cell lines, targeted deep sequencing (mean coverage 111×) of all coding exons of Discovery-Cohort mutated genes plus 277 cancer-relevant genes.
  • Total mutational cohort: 139 paired ESCC cases.
  • SCNV cohort: 22 tumors via SNP-array (Affymetrix 250K, 18 from Discovery Cohort), 59 samples via array-CGH (44K Agilent), plus three additional public SNP-array ESCC datasets — total 184 analyzable primary ESCC samples.
  • FISH validation cohort: Additional 53 ESCC tumors for FGFR1 amplification.
  • IHC validation cohorts: TMAs containing 50 primary ESCCs (and additional 40-case cohort for FBXW7, 60-case for FGFR1).
  • Samples: Collected from Cancer Institute/Hospital, Chinese Academy of Medical Sciences (CAMS) and Linxian Cancer Hospital; only tumors with >70% malignant cells used.
  • Cancer type: ESCC (esophageal squamous cell carcinoma).
  • cBioPortal study: escc_ucla_2014.
  • Reference public expression datasets: GSE20347, GSE23400 (cited for APOBEC3B, ZNF750, XPO1 mRNA comparisons); CCLE for ZNF750 cell-line expression.
  • Sequence Read Archive accession: SRP033394.

Key findings

  • 1,847 non-silent somatic mutations identified across the combined cohort with an average of 15 mutations per case; Sanger validation true-positive rate 96.2% PMID:24686850.
  • 13 genes significantly mutated by MutSigCV (FDR q < 0.2); many had not previously been implicated in ESCC.
  • Trinucleotide signature analysis implicated APOBEC3B as a major driver of ESCC mutagenesis; APOBEC3B mRNA was up-regulated in ESCC tumors (datasets GSE20347, GSE23400).
  • 14 recurrent focal SCNVs identified; the most frequent amplification peak spanned CCND1 on 11q13.2. Additional amplification peaks involved EGFR, MYC, KRAS, and CDKN2A deletion.
  • FGFR1 was newly identified as recurrently amplified in ESCC: FISH validation showed amplification in 11/53 additional ESCCs; IHC showed FGFR1 protein up-regulation in 17.3% of ESCC tumors.
  • Significantly altered pathways: MAPK (P = 0.0005), PI3K (P = 0.0004), JAK-STAT3 (P = 0.0006), G1-S cell cycle (P = 1.63E-05), and epigenetic modification (P = 0.0013).
  • ZNF750 significantly mutated (q = 1.24E-06), predominantly truncating mutations; focally deleted in 3.4% of ESCC tumors; mRNA significantly under-expressed in tumor vs. normal; functional knockdown promoted ESCC proliferation while ectopic expression suppressed it and induced epithelial differentiation genes — ZNF750 proposed as a novel tumor suppressor.
  • FAT1, FAT2, FAT3 harbored frequent, mutually-exclusive truncating mutations. FAT1 showed loss of heterozygosity in 2/2 mutated tumors examined by SNP-array and homozygous deletions in 3.4% of ESCCs. FAT1 siRNA knockdown promoted proliferation; FAT1 ectopic expression inhibited proliferation and soft-agar colony formation; FAT2 shRNA knockdown promoted ESCC xenograft growth in NOD/SCID mice — FAT1 and FAT2 proposed as tumor suppressors.
  • XPO1 carried a recurrent D624G missense mutation (identical to a CLL mutation [PMID:23415222]); structural modeling predicts loss of a salt bridge with Snurportin Lys144, presumably enhancing nuclear export activity. XPO1 mRNA and protein were frequently overexpressed in ESCC tumors and overexpression correlated with larger tumor size (P = 0.016).
  • XPO1 functional/therapeutic data: shRNA knockdown of XPO1 induced apoptosis (cleaved PARP) and inhibited proliferation; submicromolar KPT-330 (selinexor) treatment of ESCC cells induced significant apoptosis, inhibited proliferation, and altered XPO1 cargo proteins (p53, cyclin D1, c-Myc, PUMA, BIM).
  • FBXW7 mutations confirmed at frequencies described by Agrawal et al. (PMID:22877736); IHC showed mutation-associated loss of FBXW7 protein, with down-regulation in 33% of tumors in an additional 40-case cohort.
  • 31 genes with potentially actionable alterations identified, including ERBB, HDAC, PI3K family, XPO1, FGFR1, TP53, JAK-STAT3 and MTOR-RPS6K signaling.

Genes & alterations

  • FAT1 — recurrent truncating mutations, LOH, and 3.4% homozygous deletion; functional tumor suppressor in ESCC.
  • FAT2 — recurrent truncating mutations (mutually exclusive with FAT1/FAT3); shRNA knockdown promotes xenograft growth.
  • FAT3 — recurrent truncating mutations; mutually exclusive with FAT1 and FAT2.
  • ZNF750 — significantly mutated (q = 1.24E-06), predominantly truncating; focal deletion in 3.4% of ESCC; mRNA and protein down-regulated; functional novel tumor suppressor regulating squamous-cell differentiation.
  • KMT2D — significantly mutated histone methyltransferase; part of epigenetic dysregulation pathway.
  • KMT2C — mutated histone methyltransferase contributing to epigenetic dysregulation.
  • KDM6A — mutated histone demethylase contributing to epigenetic dysregulation.
  • ARID2 — mutated SWI/SNF complex member.
  • PBRM1 — mutated SWI/SNF complex member.
  • XPO1 — recurrent D624G missense mutation; protein and mRNA overexpressed; therapeutic target via KPT-330 (selinexor).
  • FBXW7 — frequent inactivating mutations leading to protein loss (33% of additional cohort).
  • TP53 — mutated as part of G1-S cell-cycle dysregulation.
  • PIK3CA — activating mutations and amplification contributing to PI3K pathway activation.
  • NOTCH1 — mutated (confirms prior reports).
  • PTEN — inactivating mutations in PI3K pathway.
  • ERBB4 — activating mutations in RTK signaling.
  • MAP3K13 — inactivating mutations in MAPK pathway.
  • MAP3K15 — inactivating mutations in MAPK pathway.
  • JAK1 — mutations activating JAK-STAT3 signaling.
  • IL7R — amplification activating JAK-STAT3 signaling.
  • SOX2 — amplified lineage-survival oncogene cooperating with STAT3 activation.
  • FGFR1 — focal amplification in 11/53 ESCC validated by FISH; protein over-expression in 17.3% of tumors; proposed druggable target.
  • CCND1 — most frequent focal amplification peak (11q13.2); drives cell-cycle dysregulation.
  • CDKN2A — focal deletion and mutation; cell-cycle dysregulation.
  • EGFR — focal amplification.
  • MYC — focal amplification.
  • KRAS — focal amplification.
  • APOBEC3B — up-regulated in ESCC tumors; implicated as the cytidine deaminase responsible for the trinucleotide mutational signature observed.

Clinical implications

  • FGFR1 amplification in ~6–17% of ESCC (FISH/IHC) suggests potential sensitivity to selective FGFR inhibitors (cited NVP-BGJ398 precedent from PMID:23002168) — novel candidate biomarker for FGFR-targeted therapy in ESCC.
  • XPO1 protein overexpression positively correlated with larger tumor size (P = 0.016); selinexor (KPT-330) induced apoptosis and growth arrest in ESCC cell lines at submicromolar concentrations — proposed therapeutic target for XPO1-overexpressing ESCC.
  • 31 candidate actionable alterations identified across ERBB, HDAC, PI3K family, XPO1, FGFR1, TP53, JAK-STAT3, and MTOR-RPS6K signaling — most not previously considered actionable in ESCC.
  • Authors propose multiple novel therapeutic entry points for what they describe as a typically chemo-resistant cancer.

Limitations & open questions

  • No survival/outcome analysis correlating individual mutations or amplifications with patient prognosis is reported.
  • Frequency Cohort uses targeted deep sequencing (not WES) so non-coding events and mutations outside the targeted gene set are not captured.
  • Cell-line cohort lacks paired germline controls, requiring more stringent variant-calling thresholds and reducing confidence in cell-line-specific somatic calls.
  • XPO1 inhibitor KPT-330 efficacy demonstrated only in ESCC cell lines; in vivo / clinical evidence in ESCC remains to be established.
  • Functional roles of FAT3 truncating mutations are inferred from mutual exclusivity with FAT1/FAT2 but not directly validated by knockdown experiments.
  • Whether ZNF750 inactivation cooperates with specific co-mutations or copy-number events is not dissected.
  • Cohort is exclusively from high-incidence Chinese centers — generalizability to ESCC in other geographic populations remains open.

Citations from this paper used in the wiki

  • “We identified novel significantly mutated genes such as FAT1, FAT2, ZNF750 and KMT2D, in addition to previously discovered ones (TP53, PIK3CA and NOTCH1).” (Abstract)
  • “RTK-MAPK-PI3K pathways, cell cycle and epigenetic regulation are frequently dysregulated by multiple molecular mechanisms in this cancer.” (Abstract)
  • “trinucleotide signature analysis suggested that DNA cytidine deaminase APOBEC3B is responsible for ESCC mutagenesis.” (p.2)
  • “FGFR1 was shown to be frequently amplified, which has not been reported before in ESCC … FGFR1 was amplified in 11 samples … FGFR1 protein was up-regulated in 17.3% ESCC tumors.” (p.3)
  • “ZNF750 was significantly mutated in ESCC (q = 1.24E-06) … ZNF750 might function as a novel tumor suppressor in ESCC through regulating squamous cell differentiation.” (p.4)
  • “ESCC harbored very frequent, mutually-exclusive truncating mutations affecting FAT1, FAT2 and FAT3 … homozygous deletions of FAT1 occurred in 3.4% ESCCs.” (p.4)
  • “We found one missense mutation D624G affecting XPO1. Importantly, this mutation is identical to the one discovered in chronic lymphocytic leukemia.” (p.5)
  • “Submicromolar concentrations of KPT-330 inhibited ESCC cell proliferation and induced significant apoptosis.” (p.5)
  • “we identified 31 genes with potentially actionable alterations in ESCC. Recurrent candidate druggable targets included ERBB, HDAC and PI3K family, XPO1, FGFR1, TP53, JAK-STAT3 and MTOR-RPS6K signaling.” (p.5)

This page was processed by crosslinker on 2026-05-11.