Sage Journals: Discover world-class research

Abstract

OBJECTIVES:

Pancreatic cancer is one of the most lethal malignancies worldwide. Pancreatic adenosquamous carcinoma (PASC) is a rare histological type of pancreatic carcinoma with a poor prognosis. The median survival time after diagnosis is less than one year. It is believed that the pathogenesis of PASC is different from pancreatic adenocarcinoma. In this study, we tried to reveal the intrinsic gene mutations associated with PASC through whole exome sequencing.

METHODS:

Both cancerous and paracancerous tissues were collected from 12 pathologically diagnosed PASC patients. Their clinical characteristics were collected, and patient survival information was obtained through follow-up. The correlations between the mutations and clinical characteristics were analysed.

RESULTS:

Germline mutations were identified in MAP3K1 (9 cases), PDE4DIP (7), BCR (7), ALK (6), USP6 (5), AR (4), HLA-A (4), SPEN (4), KMT2D (3), NUTM2B (3), ZFHX3 (3), and MN1 (3), while somatic mutations were found in TP53 (5), KRAS (3), HRNR (3), and OBSCN (3). Peripheral tissue invasion was associated with somatic mutations in KRAS ( $P=$ 0.0339). Additionally, there were significant correlations between lymphatic metastasis and germline mutations in USP6 ( $P=$ 0.0228) and somatic mutations in OBSCN and HRNR ( $P=$ 0.0339).

CONCLUSION:

In conclusion, susceptibility genes including MAP3K1, PDE4DIP, and BCR are frequently found to be mutated in the germlines of PASC patients. Somatic mutations in KRAS, OBSCN, and HRNR and germline mutations in USP6 are related to tumour invasion and metastasis, reinforcing the necessity of translating these potential biomarkers into clinical practice.

Keywords

Pancreatic adenosquamous carcinoma germline mutation somatic mutation whole exome sequencing

Abbreviations

PASC	Pancreatic adenosquamous carcinoma
NGS	next-generation sequencing
SNVs	Single Nucleotide Variants
INDELs	insertions and deletions
MAPKs	mitogen-activated protein kinases

1. Introduction

Pancreatic cancer is one of the most lethal malignancies worldwide, with 90,100 estimated new cases and 79,400 deaths in China in 2015 [1]. The majority of patients have regional lymph node (29%) or distant metastases (52%) at initial diagnosis, and their prognosis is dismal with a 5-year survival rate of 11.5% or 2.7%, respectively [2]. Pancreatic adenosquamous carcinoma (PASC) constitutes 1–4% of pancreatic carcinomas, and is a rare, clinically aggressive histological form with a poor prognosis [3, 4, 5]. The median survival time after diagnosis is less than one year [3].

PASC contains both ductal adenocarcinoma and squamous carcinoma within the same tumour. It is widely accepted that PASC is derived from pancreatic tissue, rather than a combination of two sources of malignant tumours. Nevertheless, very little is known about the pathogenesis of PASC due to its rarity, especially the genomic features. Recently, Fang et al. performed whole genome sequencing of 17 PASC tumours [6]. Combined with the 11 PASC cases from the International Cancer Genome Consortium, the most frequently mutated genes included KRAS, TP53, and SMAD4. We have performed whole exome sequencing analysis of 12 PASC patients in order to identify additional cancer-related gene mutations, in hopes of further understanding PASC pathogenesis.

2. Materials and methods

2.1 Patients

Cancerous and paracancerous tissues were collected from 12 patients with pathologically diagnosed PASC at the Department of General Surgery of Changhai Hospital, Shanghai between February 2015 and January 2016. Demographic and clinical baseline characteristics were collected from medical records, and survival data were obtained through follow-up. The present study was approved by the Institution Review Board of the Changhai Hospital, Shanghai. All patients or their families signed informed consent forms before the initiation of the study. All methods were performed in accordance with relevant guidelines and regulations.

2.2 Sample acquisition and preparation

DNA was purified from the macro-dissected frozen cancer tissues and paracancerous tissues. Sections from the frozen tissues ( $\sim$ 7 $\mu$ M thick) were stained with haematoxylin. DNA was extracted using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). The DNA concentration was measured by a Qubit dsDNA assay (Life Technologies, Carlsbad, CA, USA). Only samples with an OD260/OD280 of 1.8–2.0 were used for analysis. Fragmentation of DNA was conducted by sonication (Covaris M220), followed by end repair and adaptor ligation. Fragments of 150–200 bp were purified with Agencourt AMPure XP Beads (Beckman Coulter Brea, CA, USA; Gat No. A63881) and amplified by PCR using a SureSelectXT Reagent kit (Agilent Technologies, Santa Clara, CA; Gat No. G9611A). A Bioanalyzer 2100 with DNA 1000 LabChip kit (Agilent Technologies Deutschland GmbH) was then used to assess quality and size range.

2.3 Whole exome sequencing

Indexed samples were pooled for loading onto flow cells for sequencing on an Illumina sequencing platform (HiSeq ${}^{\rm TM}$ 4000) with 150 bp paired-end reads. Raw data (raw reads) were processed using NGS QC Toolkit [7]. Reads containing poly-N and low quality reads were not used for analysis.

Table 1
Demographic and clinical characteristics of the 12 patients with pancreatic adenosquamous carcinoma

ID	Gender	Tumor size, cm	TNM stage	Lymph atic metastasis	Peripheral tissue invasion	CA199 $>$ 37 U/ml	p53	Postoperative therapy	Outcome
1	Male	3.5	T3N0M0	No	Duodenum	Yes	Positive	Tegafur	Death
2	Female	3	T3N1bM0	Yes	No	No	Positive	None	Alive
3	Male	5	T3N0M0	No	Left adrenal gland	Yes	Positive	Interventional therapy for liver metastasis	Death
4	Male	5	T3N0M0	No	Duodenum	No	Positive	Gamma-knife radiosurgery lymphatic metastasis in peritoneum	Alive
5	Male	4	Unknown	No	Stomach	Yes	Positive	None	Death
6	Female	2	T3N1M0	Yes	CBD	Yes	Positive	None	Alive
7	Female	12	T3N1M0	Yes	No	Yes	Positive	Tegafur	Alive
8	Male	6	T3N0M0	Yes	Spleen, stomach, colon	No	Positive	Four lines of chemotherapy	Death
9	Male	5.5	T3N1M0	Yes	No	Yes	Negative	Three lines of chemotherapy	Death
10	Male	3.5	T3N1M0	Yes	CBD	No	Negative	Tegafur	Alive
11	Male	4.5	T3N1bM0	Yes	No	No	Positive	None	Death
12	Male	5	T3N1M0	Yes	CBD	No	Negative	None	Death

CBD: Common bile duct. Data were collected from clinical medical records. Immunohistochemistry was used to assess the p53 expression using mouse anti-human p53 monoclonal antibody (Maixin-Bio Co., Fuzhou, China). p53 was interpreted as positive if $\geqslant$ 20% tumors cell show strong positive staining.

2.4 NGS data analysis and filtering

The Illumina next-generation sequencing (NGS) platform was used for whole exome sequencing, and the results were saved in the FASTQ file format. All the raw sequencing reads in FASTQ format were quality controlled using FastQC v0.11.3 [8]. The trimmomatic V0.32 (trimmomatic-0.36.jar PE-threads N R1 R2 R1.paired.fq.gz R1.unpaired.fq.gz R2.paired.fq.gz R2.unpaired.fq.gz ILLUMINACLIP: adapter. fa:2:30: 10 LEADING:3 TRAILING:3 SLIDINGWINDOW: 4:15 MINLEN:36) was used to trim adapters and also bases with Phred quality scores under 30 [9]. Trimmed reads longer than 36 bp were mapped to human geno-me sequence hg19 (NCBI Build 37, http://hgdownload. cse.ucsc.edu/goldenPath/hg19/bigZipss) using the Burrows-Wheeler Alignment tool (BWA-MEM version 0.7.12) [10] with default settings. To perform the second round of quality control before variant calling, PicardTools-1.119 (http://broadinstitute.github.io/picard) was used to remove duplicates, and GATK [11] was applied for local INDEL realignment, duplicate removal, and base quality score recalibration.

Both germline and somatic mutations associated with PASC were identified. Normal tissue was used to call the germline Single Nucleotide Variants (SNVs) and short insertions and deletions (INDELs) using GATK HaplotypeCaller. Raw germline variants were filtered using the standard hard filtering parameters according to GATK Best Practice recommendations [12, 13]. Comparing normal and cancer tissues, somatic mutations (SNPs and INDELs) were identified using GATK Mutect2. Both germline and somatic variants which passed the filtering criteria were annotated with ANNOVAR [14]. Variants were considered as potential candidates when three of five prediction tools called them damaging (SIFT, Polyphen2_HDIV, Polyphen2_HVAR, LRT, and MutationTaster) [15, 16, 17, 18].

2.5 Statistical analysis

All statistical analyses were carried out using SPSS (SPSS Inc, Woking, Surrey, UK) version 22. Correlations between germline or somatic mutations and clinical characteristics or outcomes were analysed using the Chi-square test or $t$ -test. $P<$ 0.05 was considered statistically significant.

3. Results

3.1 Subjects characteristics

We conducted whole exome sequencing of 12 PASC patients including three females and nine males. Their demographic and clinical characteristics are shown in Table 1. The primary tumour size ranged from 3 to 12 cm. Eight out of 12 patients had lymphatic metastasis. Common bile duct invasion occurred in three cases, duodenum invasion in two cases, stomach invasion in two, left adrenal gland invasion in one, spleen invasion in one, colon invasion in one, while four patients did not experience peripheral tissue invasion. At the end of our data collection, seven patients had died from local recurrence, liver metastasis and/or pulmonary metastasis, while the remaining five cases had achieved good clinical outcomes.

3.2 Identified germline and somatic mutations

A total of 122 germline mutated genes were identified, harbouring mostly missense mutations (Supplemental Table). There are 21 genes which are often mutated in pancreatic cancer, of which four were also mutated in the PASC samples in this study, including TP53 ( $n=$ 2), EP300 ( $n=$ 2), SMARCA4 ( $n=$ 1), and CREBBP ( $n=$ 1). Figure 1 shows the germline mutated genes that were identified in at least two samples. The majority of mutations (54.3%; 57/105) were missense mutations. The three most common susceptibility genes were MAP3K1, PDE4DIP, and BCR, with germline mutations found in nine, seven, and seven patients, respectively. MAP3K1 had 6 inframe deletions and 3 missense mutations, PDE4DIP had 6 missense mutations and 1 inframe insertion, and all germline mutations in BCR were frameshift insertions. Among the genes that were mutated in the germlines of two or more patients, pathogenic variants were found in MAP3K1, ALK, AR, ZFHX3, TP53, RAD51B, PARP1, LPP, KMT2C, FAT4, FANCD2, and CDH1.

Figure 1.

Germline mutation spectrum in 12 PASC patients with 36 genes that were identified in at least two samples.

Moreover, multiple somatic mutations were found in PASC cancer tissues that did not occur in paracancerous tissues. A total of 488 somatically mutated genes were identified (Supplemental Table). Figure 2 shows the genes with somatic mutations found in at least two samples. The majority of the mutations (74.3%; 52/70) were missense mutations. TP53, as expected, was the most frequently mutated gene in tumours, followed by KRAS, OBSCN, and HRNR. TP53 had 3 missense mutations, 1 nonsense mutation, and 1 inframe deletion; all mutations in KRAS, OBSCN, and HRNR were missense mutations. Among the genes with somatic mutations in two or more samples, pathogenic variants were found in TP53, KRAS, TGFBR2, KLF17, and CASP9.

3.3 Correlations between germline or somatic mutations and clinical characteristics and outcomes

Associations between gene mutations and clinical characteristics and outcomes were analysed (Fig. 3 and Table 2). The results show that germline mutations in USP6 are negatively correlated with lymphatic metastasis (4/4 patients without lymphatic metastasis vs. 1/8 patients with lymphatic metastasis; $P=$ 0.0228). No relationship was observed between the identified germline mutations and peripheral tissue invasion, elevated CA199, p53 positivity, or death. In terms of somatic mutations, mutations in HRNR and OBSCN were related to a lack of lymphatic metastasis (HRNR: 3/4 patients without lymphatic metastasis vs. 0/8 patients with lymphatic metastasis, $P=$ 0.0339; OBSCN: 3/4 patients without lymphatic metastasis vs. 0/8 patients with lymphatic metastasis, $P=$ 0.0339), whereas KRAS mutations were significantly associated with a lack of peripheral tissue invasion (3/4 patients without peripheral tissue invasion vs. 0/8 patients with peripheral tissue invasion, $P=$ 0.0339). There was no relationship between the identified somatic mutations and elevated CA199, p53 positivity, or death.

Table 2
Association between gene mutations and clinical characteristics and outcome

Mutation		Tumour size	Lymphatic metastasis	Peripheral tissue invasion	Elevated CA199	P53 positive	Outcome
Germline mutation	MAP3K1	0.8520	0.4795	0.4795	1.0000	1.0000	1.0000
	PDE4DIP	0.4970	1.0000	0.8360	1.0000	0.7353	1.0000
	BCR	0.9860	1.0000	1.0000	1.0000	1.0000	0.0600
	ALK	0.7470	1.0000	1.0000	1.0000	1.0000	0.2416
	USP6	0.4280	0.0228	0.1473	1.0000	1.0000	1.0000
	AR	0.6200	0.8286	1.0000	1.0000	0.4795	0.8360
	HLA-A	0.1370	1.0000	1.0000	1.0000	1.0000	0.3006
	SPEN	0.7900	0.2790	1.0000	1.0000	0.4795	0.8360
	KMT2D	0.7560	1.0000	0.4795	1.0000	1.0000	1.0000
	NUTM2B	0.4830	1.0000	0.4795	1.0000	0.7003	0.7353
	ZFHX3	0.3420	0.4795	1.0000	1.0000	0.7003	0.7353
	MN1	0.6630	1.0000	0.4795	1.0000	0.7003	1.0000
Somatic mutation	TP53	0.2690	0.8360	1.0000	1.0000	0.3105	1.0000
	KRAS	0.5290	0.4795	0.0339	1.0000	0.7003	0.7353
	HRNR	0.8520	0.0339	0.4795	1.0000	0.7003	1.0000
	OBSCN	0.8520	0.0339	0.4795	1.0000	0.7003	1.0000

The values in this table were the $P$ values of Chi-square test or $t$ -test.

Figure 2.

Somatic mutation spectrum in 12 PASC samples with 32 genes that were identified in at least two samples.

Figure 3.

Integrative clustering of key mutated genes in 12 pancreatic adenosquamous carcinoma patients. Clinical features: gender, outcomes, lymphatic metastasis, and peripheral tissue invasion.

4. Discussion

As a rare tumour type, limited studies have been performed regarding the genomic features of PASC [6, 19]. In the present study, we performed a comprehensive analysis of germline and somatic alterations in 12 pathologically diagnosed PASC patients by whole exome sequencing. To our knowledge, this is only the second study of PASC in a relatively large population.

Compared to the research by Fang et al. [6], the present study identified more PASC susceptibility genes. There are approximately 21 well-recognized pancreatic cancer susceptibility genes, including KRAS, TP53, SMAD4, MLL3, TGFBR2, ARID1A, SMARCA4, PBRM1, SF3B1, PCDH18, MAP2K4, EPC1, ATM, ARID2, CREBBP, AHANK, ANK3, ACVR1B, EP300, PCSK6, and ARHGAP3, among which three genes including KRAS, TP53 and SMAD4 were also identified as the most frequently mutated genes by Fang et al. [6]. Germline mutations in TP53, EP300, SMARCA4, and CREBBP were also detected in PASC patients at a low frequency, suggesting that these genes are also PASC susceptibility genes. We found an additional 12 germline mutated genes occurring in three or more PASC patients. The top three frequently mutated genes were MAP3K1, PDE4DIP, and BCR. All three genes had novel germline variants related to PASC.

MAPKs (mitogen-activated protein kinases) are key mediators of signalling pathways and are involved in multiple aspects of cell physiology [20, 21]. Dysregulation of MAPK-mediated signalling is a common event in tumorigenesis. MAP3Ks (mitogen-activated protein kinase kinase kinases) catalyse the first phosphorylation in a cascade of three sequential phosphorylation steps. MAP3K1 participates in the ERK and JNK pathways and the NF- $\kappa$ B pathway, that are all associated with tumorigenesis [22, 23, 24]. Somatic missense or nonsense mutations in MAP3K1 have been identified in multiple types of cancer, especially in hormone receptor positive breast cancers [25]. In Chinese breast cancer patients, deleterious somatic mutations in the MAP3K1 gene, including c.1280delC (exon6), c.1461_1465delATGTC (exon8), and c.1604_1605delAA (exon9), were identified by microfluidic PCR-based target enrichment and next-generation sequencing technologies [26]. However, MAP3K1 has not been reported as a cancer susceptibility gene yet. In this study, we found a novel germline variant in MAP3K1 at a high frequency of 75.0% (9 out of 12 PASC patients). It is worth mentioning that MAP3K4, another member of the MAP3K family, was found to harbour germline truncation variants in patients with ovarian cancer [27]. These findings suggest that variants in different members of the MAP3K family may cause susceptibility to different cancer types.

PDE4DIP encodes phosphodiesterase 4D interacting protein, a regulatory binding protein associated with calcium-mediated cAMP signalling (a pathway which is known to be dysregulated in cancer). Somatic mutations in PDE4DIP have been reported in some types of cancers, such as neuroendocrine Merkel cell carcinoma [28], ovarian cancer [29], malignant peritoneal mesothelioma [30] and pancreatic adenocarcinoma [31]. The malignant disease driven by the BCR-ABL oncogene is chronic myeloid leukaemia [32, 33]. Neither PDE4DIP nor BCR germline mutations have been reported in PASC patients. In fact, to our knowledge, mutations in the PDE4DIP and BCR genes are extremely rare in East Asian populations. Variants in these genes have not been reported in the dbSNP database yet. However, germline mutations in PDE4DIP and BCR were significantly enriched in the patients in this study, indicating that the susceptibility mutations might be unique to the Chinese population. We will focus on these genes in our future studies.

In addition to identifying germline and somatic alterations in patients with PASC, we also investigated the relationship between these mutations and clinical characteristics and outcomes. The results show that somatic mutations in KRAS are associated with diminished peripheral tissue invasion, and germline mutations in USP6 and somatic mutations in OBSCN and HRNR are associated with a decreased incidence of lymphatic metastasis. None of the identified mutations correlate with the outcomes in PASC patients; however, the small sample size, and varied efficacy of different therapies might have affected this analysis.

KRAS is a proto-oncogene encoding monomeric membrane-localised guanine nucleotide (GTP/GDP)-binding protein. Somatic mutations in KRAS have been reported in pancreatic cancer [25]. Other studies have shown that most pancreatic ductal adenocarcinoma patients and PASC patients carry KRAS activating mutations [34, 35, 36]. Consistently, we also identified somatic mutations in KRAS in our cohort. These findings suggest that mutations in KRAS are a common molecular feature of pancreatic cancer. Rachagani et al. reported that knockdown of the activated Kras ${}^{\rm G12D}$ allele in pancreatic cancer cells resulted in notable decreases in cell growth, motility, and invasion [37]. Correspondingly, the expression of key compounds, such as phoshpo-ERK-1/2, NF- $\kappa$ B, and MMP-9, in the related signalling pathways were decreased. Disruption of E-cadherin mediated cell-cell adhesion is an important step in epithelial mesenchymal transition, which promotes tumour cell migration and invasion [38]. Expression of E-cadherin was increased by Kras knockdown, which suppressed the invasion capability of pancreatic cancer cells. This mechanism might also explain the relationship between somatic mutations in KRAS and peripheral tissue invasion in PASC patients; however, this requires further confirmation.

The OBSCN gene encodes obscurin, which belongs to a family of giant cytoskeletal proteins. Mutations in OBSCN are common in cancers [39, 40, 41], but not in PASC. We found somatic mutations in OBSCN in PASC patients, and these were negatively correlated with lymphatic metastasis. Altered expression of obscurin was detected in breast cancer tumour tissue rather than in matched normal tissue. Furthermore, loss of obscurin causes breast cancer cells to fail to form adhesion junctions, thus promoting tumour metastasis [41]. Whether the somatic mutations in OBSCN lead to a similar outcome in PASC should be cautiously investigated.

However, it should be noted that there were several limitations in this study. Firstly, the number of samples is a little small. PASC is a rare, clinically aggressive histological tumour, which constitutes only 1–4% of pancreatic carcinomas, and therefore, it is very difficult to collect PASC samples. There were only 12 samples included in this study. Considering the small sample size, our results should be further confirmed in a validation cohort or by experimental methods. Although the limited samples may bias results, we believe that this study has taken an important step towards revealing the underlying pathogenesis of PASC, and may aid in the determination of novel therapeutic targets. Secondly, the relationship between the identified mutations and PASC development/progression is only basing on data analysis with the absent of functional study. Whether these additional mutations are biologically meaningful should be further validated in vitro or in vivo models. We hope that these useful clues will help other researchers to carry out relevant research.

5. Conclusions

Three PASC susceptibility genes were identified: MAP3K1, PDE4DIP, and BCR. Additionally, we found negative associations between tumour metastasis and germline mutations in USP6 and somatic mutations in OBSCN and HRNR, as well as a negative correlation between peripheral tissue invasion and somatic KRAS mutations.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-190236.

sj-xlsx-1-cbm-10.3233_CBM-190236.xlsx - Supplemental material

Supplemental material, sj-xlsx-1-cbm-10.3233_CBM-190236.xlsx

Footnotes

Acknowledgments

This work was funded by the National Natural Science Foundation of China (Grant NO. 81672830, 81502054, 81871992 and 61905279), PLA Youth Talent Training Program Foundation (Grant NO. 16QNP095), Scientific Research Project of Shanghai Municipal Commission of Health and Family Planning (Grant NO. 201840054), Scientific Research and Innovation Project of Shanghai Municipal Education Commission (Grant NO. 2017-01-07-00-07-E00012) and China’s National Key Research and Development Project (Grant NO. SQ2019YFC130043).

Conflict of interest

The authors declare no conflict of interest.

References

Chen

Zheng

Baade

P.D.

Zhang

Zeng

Bray

Jemal

X.Q.

and He

, Cancer statistics in China, 2015, CA Cancer J Clin 66 (2016), 115–32.

Howlader

Noone

A.M.

Krapcho

et al., SEER Cancer Statistics Review, 1975–2014, National Cancer Institute. Bethesda, MD, https://seer.cancer.gov/csr/1975_2014/, based on November 2016 SEER data submission, posted to the SEER web site, April 2017.

Simone

C.G.

Zuluaga Toro

Chan

Feely

M.M.

Trevino

J.G.

and George

T.J.

, Jr., Characteristics and outcomes of adenosquamous carcinoma of the pancreas, Gastrointest Cancer Res 6 (2013), 75–9.

Boyd

C.A.

Benarroch-Gampel

Sheffield

K.M.

Cooksley

C.D.

and Riall

T.S.

, 415 patients with adenosquamous carcinoma of the pancreas: a population-based analysis of prognosis and survival, J Surg Res 174 (2012), 12–9.

Katz

M.H.

Taylor

T.H.

Al-Refaie

W.B.

Hanna

M.H.

Imagawa

D.K.

Anton-Culver

and Zell

J.A.

, Adenosquamous versus adenocarcinoma of the pancreas: a population-based outcomes analysis, J Gastrointest Surg 15 (2011), 165–74.

Fang

Xie

Xue

Zhao

Song

Peng

Bai

and Shen

, Genomic signatures of pancreatic adenosquamous carcinoma (PASC), J Pathol 243 (2017), 155–159.

Patel

R.K.

and Jain

, NGS QC Toolkit: a toolkit for quality control of next generation sequencing data, PLoS One 7 (2012), e30619.

Andrews

, FastQC: a quality control tool for high throughput sequence data. Available online at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc, 2010.

Bolger

A.M.

Lohse

and Usadel

, Trimmomatic: a flexible trimmer for Illumina sequence data, Bioinformatics 30 (2014), 2114–20.

10.

, Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM, arXiv preprint arXiv:13033997. (2013).

11.

McKenna

Hanna

Banks

Sivachenko

Cibulskis

Kernytsky

Garimella

Altshuler

Gabriel

Daly

and DePristo

M.A.

, The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data, Genome Res 20 (2010), 1297–303.

12.

DePristo

M.A.

Banks

Poplin

Garimella

K.V.

Maguire

J.R.

Hartl

Philippakis

A.A.

del Angel

Rivas

M.A.

Hanna

McKenna

Fennell

T.J.

Kernytsky

A.M.

Sivachenko

A.Y.

Cibulskis

Gabriel

S.B.

Altshuler

and Daly

M.J.

, A framework for variation discovery and genotyping using next-generation DNA sequencing data, Nat Genet 43 (2011), 491–8.

13.

Van der Auwera

G.A.

Carneiro

M.O.

Hartl

Poplin

Del Angel

Levy-Moonshine

Jordan

Shakir

Roazen

Thibault

Banks

Garimella

K.V.

Altshuler

Gabriel

and DePristo

M.A.

, From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline, Curr Protoc Bioinformatics 43 (2013), 1110.1–11.10.33.

14.

Wang

and Hakonarson

, ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data, Nucleic Acids Res 38 (2010), e164.

15.

Flanagan

S.E.

Patch

A.M.

and Ellard

, Using SIFT and PolyPhen to predict loss-of-function and gain-of-function mutations, Genet Test Mol Biomarkers 14 (2010), 533–7.

16.

Adzhubei

I.A.

Schmidt

Peshkin

Ramensky

V.E.

Gerasimova

Bork

Kondrashov

A.S.

and Sunyaev

S.R.

, A method and server for predicting damaging missense mutations, Nat Methods 7 (2010), 248–9.

17.

Umbach

D.M.

and Weinberg

C.R.

, The use of case-parent triads to study joint effects of genotype and exposure, Am J Hum Genet 66 (2000), 251–61.

18.

Schwarz

J.M.

Cooper

D.N.

Schuelke

and Seelow

, MutationTaster2: mutation prediction for the deep-sequencing age, Nat Methods 11 (2014), 361–2.

19.

Marcus

Maitra

and Roszik

, Recent advances in genomic profiling of adenosquamous carcinoma of the pancreas, J Pathol 243 (2017), 271–272.

20.

Kyriakis

J.M.

and Avruch

, Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation, Physiol Rev 81 (2001), 807–69.

21.

Uhlik

M.T.

Abell

A.N.

Cuevas

B.D.

Nakamura

and Johnson

G.L.

, Wiring diagrams of MAPK regulation by MEKK1, 2, and 3, Biochem Cell Biol 82 (2004), 658–63.

22.

Yang

J.Y.

Zong

C.S.

Xia

Yamaguchi

Ding

Xie

Lang

J.Y.

Lai

C.C.

Chang

C.J.

Huang

W.C.

Huang

Kuo

H.P.

Lee

D.F.

L.Y.

Lien

H.C.

Cheng

Chang

K.J.

Hsiao

C.D.

Tsai

F.J.

Tsai

C.H.

Sahin

A.A.

Muller

W.J.

Mills

G.B.

Hortobagyi

G.N.

and Hung

M.C.

, ERK promotes tumorigenesis by inhibiting FOXO3a via MDM2-mediated degradation, Nat Cell Biol 10 (2008), 138–48.

23.

Wagner

E.F.

and Nebreda

A.R.

, Signal integration by JNK and p38 MAPK pathways in cancer development, Nat Rev Cancer 9 (2009), 537–49.

24.

Pham

T.T.

Angus

S.P.

and Johnson

G.L.

, MAP3K1: Genomic Alterations in Cancer and Function in Promoting Cell Survival or Apoptosis, Genes Cancer 4 (2013), 419–26.

25.

Kan

Jaiswal

B.S.

Stinson

Janakiraman

Bhatt

Stern

H.M.

Yue

Haverty

P.M.

Bourgon

Zheng

Moorhead

Chaudhuri

Tomsho

L.P.

Peters

B.A.

Pujara

Cordes

Davis

D.P.

Carlton

V.E.

Yuan

Wang

Eigenbrot

Kaminker

J.S.

Eberhard

D.A.

Waring

Schuster

S.C.

Modrusan

Zhang

Stokoe

de Sauvage

F.J.

Faham

and Seshagiri

, Diverse somatic mutation patterns and pathway alterations in human cancers, Nature 466 (2010), 869–73.

26.

Xie

Chen

Guo

Tang

Luo

Wang

Dai

and Wang

, Mutation screening of 10 cancer susceptibility genes in unselected breast cancer patients, Clin Genet (2017).

27.

Kanchi

K.L.

Johnson

K.J.

McLellan

M.D.

Leiserson

M.D.

Wendl

M.C.

Zhang

Koboldt

D.C.

Xie

Kandoth

McMichael

J.F.

Wyczalkowski

M.A.

Larson

D.E.

Schmidt

H.K.

Miller

C.A.

Fulton

R.S.

Spellman

P.T.

Mardis

E.R.

Druley

T.E.

Graubert

T.A.

Goodfellow

P.J.

Raphael

B.J.

Wilson

R.K.

and Ding

, Integrated analysis of germline and somatic variants in ovarian cancer, Nat Commun 5 (2014), 3156.

28.

Graves

C.A.

Jones

Reynolds

Stuart

Pirisi

Botrous

and Wells

, Neuroendocrine Merkel cell carcinoma is associated with mutations in key DNA repair, epigenetic and apoptosis pathways: a case-based study using targeted massively parallel sequencing, Neuroendocrinology 101 (2015), 112–9.

29.

T.K.

Y.F.

C.C.

Chen

C.C.

Wang

Hsieh

T.H.

Herreros-Villanueva

Chen

W.T.

Chen

Y.T.

Liu

T.C.

Chen

H.S.

and Tsai

E.M.

, Targeted next-generation sequencing for molecular diagnosis of endometriosis-associated ovarian cancer, J Mol Med (Berl) 94 (2016), 835–47.

30.

Lai

Zhou

Tang

X.J.

Gao

Z.B.

Zhou

and Chen

S.Q.

, A Tumor-Specific Neo-Antigen Caused by a Frameshift Mutation in BAP1 Is a Potential Personalized Biomarker in Malignant Peritoneal Mesothelioma, Int J Mol Sci 17 (2016).

31.

Goecks

El-Rayes

B.F.

Maithel

S.K.

Khoury

H.J.

Taylor

and Rossi

M.R.

, Open pipelines for integrated tumor genome profiles reveal differences between pancreatic cancer tumors and cell lines, Cancer Med 4 (2015), 392–403.

32.

Bedi

Zehnbauer

B.A.

Barber

J.P.

Sharkis

S.J.

and Jones

R.J.

, Inhibition of apoptosis by BCR-ABL in chronic myeloid leukemia, Blood 83 (1994), 2038–44.

33.

Goldman

J.M.

and Melo

J.V.

, Targeting the BCR-ABL tyrosine kinase in chronic myeloid leukemia, N Engl J Med 344 (2001), 1084–6.

34.

Hruban

R.H.

Iacobuzio-Donahue

Wilentz

R.E.

Goggins

and Kern

S.E.

, Molecular pathology of pancreatic cancer, Cancer J 7 (2001), 251–8.

35.

Brody

J.R.

Costantino

C.L.

Potoczek

Cozzitorto

McCue

Yeo

C.J.

Hruban

R.H.

and Witkiewicz

A.K.

, Adenosquamous carcinoma of the pancreas harbors KRAS2, DPC4 and TP53 molecular alterations similar to pancreatic ductal adenocarcinoma, Mod Pathol 22 (2009), 651–9.

36.

Witkiewicz

A.K.

McMillan

E.A.

Balaji

Baek

Lin

W.C.

Mansour

Mollaee

Wagner

K.U.

Koduru

Yopp

Choti

M.A.

Yeo

C.J.

McCue

White

M.A.

and Knudsen

E.S.

, Whole-exome sequencing of pancreatic cancer defines genetic diversity and therapeutic targets, Nat Commun 6 (2015), 6744.

37.

Rachagani

Senapati

Chakraborty

Ponnusamy

M.P.

Kumar

Smith

L.M.

Jain

and Batra

S.K.

, Activated KrasG(1)(2)D is associated with invasion and metastasis of pancreatic cancer cells through inhibition of E-cadherin, Br J Cancer 104 (2011), 1038–48.

38.

Thiery

J.P.

, Epithelial-mesenchymal transitions in tumour progression, Nat Rev Cancer 2 (2002), 442–54.

39.

Sjoblom

Jones

Wood

L.D.

Parsons

D.W.

Lin

Barber

T.D.

Mandelker

Leary

R.J.

Ptak

Silliman

Szabo

Buckhaults

Farrell

Meeh

Markowitz

S.D.

Willis

Dawson

Willson

J.K.

Gazdar

A.F.

Hartigan

Liu

Parmigiani

Park

B.H.

Bachman

K.E.

Papadopoulos

Vogelstein

Kinzler

K.W.

and Velculescu

V.E.

, The consensus coding sequences of human breast and colorectal cancers, Science 314 (2006), 268–74.

40.

Balakrishnan

Bleeker

F.E.

Lamba

Rodolfo

Daniotti

Scarpa

van Tilborg

A.A.

Leenstra

Zanon

and Bardelli

, Novel somatic and germline mutations in cancer candidate genes in glioblastoma, melanoma, and pancreatic carcinoma, Cancer Res 67 (2007), 3545–50.

41.

Price

N.D.

Trent

El-Naggar

A.K.

Cogdell

Taylor

Hunt

K.K.

Pollock

R.E.

Hood

Shmulevich

and Zhang

, Highly accurate two-gene classifier for differentiating gastrointestinal stromal tumors and leiomyosarcomas, Proc Natl Acad Sci U S A 104 (2007), 3414–9.

Identification of germline and somatic mutations in pancreatic adenosquamous carcinoma using whole exome sequencing

Abstract

OBJECTIVES:

METHODS:

RESULTS:

CONCLUSION:

Keywords

Abbreviations

1. Introduction

2. Materials and methods

2.1 Patients

2.2 Sample acquisition and preparation

2.3 Whole exome sequencing

Table 1 Demographic and clinical characteristics of the 12 patients with pancreatic adenosquamous carcinoma

2.5 Statistical analysis

3. Results

3.1 Subjects characteristics

3.2 Identified germline and somatic mutations

Table 2 Association between gene mutations and clinical characteristics and outcome

5. Conclusions

Supplementary data

sj-xlsx-1-cbm-10.3233_CBM-190236.xlsx - Supplemental material

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Demographic and clinical characteristics of the 12 patients with pancreatic adenosquamous carcinoma

Table 2
Association between gene mutations and clinical characteristics and outcome