Home
JournalsCollections
For Authors For Reviewers For Editorial Board Members
Article Processing Charges Open Access
Ethics Advertising Policy
Editorial Policy Resource Center
Company Information Contact Us
OPEN ACCESS

High Frequency of HES1 Loss in Colorectal Adenocarcinomas with RAS/BRAF Mutations

  • Wei Chen1,
  • Lan Zhou2 and
  • Wei Xin2,3,* 
Journal of Clinical and Translational Pathology   2022;2(4):126-130

doi: 10.14218/JCTP.2022.00020

Received:

Revised:

Accepted:

Published online:

 Author information

Citation: Chen W, Zhou L, Xin W. High Frequency of HES1 Loss in Colorectal Adenocarcinomas with RAS/BRAF Mutations. J Clin Transl Pathol. 2022;2(4):126-130. doi: 10.14218/JCTP.2022.00020.

Abstract

Background and objectives

Hes1 is the downstream target of the canonical Notch-signaling pathway, which plays an essential role in maintaining intestinal proliferative crypts and regulating enterocyte differentiation. Loss of Hes1 expression is frequently observed in right-sided colon cancers. This report presents the relationship between the dysregulated Notch pathway and the status of RAS or BRAF mutations.

Methods

Forty-three cases of primary colorectal adenocarcinomas were collected in a tertiary teaching hospital. Hes1 expression was assessed by the immunohistochemical stain. The RAS (KRAS and NRAS) and APC status were determined by the next-generation sequencing study. In addition, BRAF V600E was tested by PCR-based mutation analysis.

Results

Overall, loss of Hes1 expression was observed more frequently in colorectal cancer specimens with either RAS or BRAF mutations than in the wild type (78.6% vs. 40.0%, p < 0.05). All the right-side tumors with RAS or BRAF mutations showed loss of Hes1 expression (12/12, 100%) (p < 0.05), compared to only 62.5% (10/16) of left-sided tumors. In addition, patients with Hes1 loss in tumor tissue were less likely to have immediate metastasis (59.1%, 13/22) compared to those with preserved Hes1 expression (83.3%, 5/6) (p = 0.37).

Conclusion

The high frequency of Hes1 loss in colorectal adenocarcinoma was associated with either RAS or BRAF mutations, suggesting synergistic effects by dysregulated Notch and RAS/BRAF mutation might play a vital role in colon carcinogenesis in some forms, especially the right-sided tumors. This finding might help guide future treatments for a subset of colon cancers.

Keywords

Notch, Hes1, Colon adenocarcinoma, RAS, BRAF, Pathology

Introduction

Colorectal cancer (CRC) is a lethal disease. Globally, CRC is the third most commonly diagnosed cancer in males and the second in females. Approximately 151,030 new cases of CRC and 52,580 deaths are reported annually in the United States, accounting for approximately 9% of all cancer deaths.1

Molecular studies over the past decades have considerably improved our understanding of CRC carcinogenesis, leading to significant advances in treating CRC by introducing novel chemotherapies and targeted agents. RAS and BRAF participate in the MAPK-ERK pathway, mediating cellular responses to many extracellular signals regulating programmed cell death, cell growth, and differentiation. KRAS mutations, found in up to 50% of sporadic CRCs, result in continuous gene activation, leading to uncontrolled autonomous cell proliferation.1–3 Previous reports demonstrated that it is more common in proximal (right-sided) colon cancers than in distal (left-sided) colorectal primaries.4–6KRAS mutations have also been implicated in the process of tumor invasion and metastasis.7

BRAF and RAS are components of the MAPK-ERK pathway. The most commonly identified BRAF mutation in human CRC is a V600E amino acid substitution, seen in sporadic CRCs with a high degree of microsatellite instability (MSI-H). They are KRAS wild-type and appear to abrogate a favorable prognosis.8,9BRAF-mutated sporadic CRCs with MSI-H are widely considered to develop from serrated polyps, commonly seen in the right-sided colon as KRAS-mutated colon cancer, although KRAS and BRAF mutations are mutually exclusive to each other.

The hairy enhancer of split-1 (HES1) is a downstream target of the Notch signaling pathway, which plays an essential role in promoting cell survival.10 Activation of the Notch pathway leads to the release of the Notch intracellular domain, which translocates to the nucleus and activates transcription of numerous downstream target genes, including HES1, HES2, HEY1, HEY2, and DTX1. HES1 is expressed in the nuclei of normal intestinal epithelial cells. It has been associated with maintaining intestinal proliferative crypts and regulating enterocyte differentiation. We reported previously that loss of HES1 expression is frequently observed in the right-sided colon cancers in mice, which commonly harbors RAS mutations and BRAF mutation.11 Very recently, Jackstadt et al. showed that enforced Notch signaling in the murine intestinal epithelium leads to highly penetrant metastasis (100% metastasis; with >80% liver metastases) in KRAS mutation-driven serrated colon cancer.10 Based on the above findings, it is exciting to explore the relationship between the dysregulated Notch pathway and the status of RAS and BRAF mutations in human CRC, which, to our best knowledge, has never been studied and reported.

Materials and methods

Patients and clinicopathological data

This study was approved by the Institutional Review Board of University Hospitals Cleveland Medical Center/Case Western Reserve University. HIPPA waiver was approved as this is a discarded tissue and chart review study with no patient-identifiable data collected or stored after the screening. A total of 43 patients with primary CRC who underwent tumor resection or biopsy between 2/23/2009 and 5/2/2017 and received molecular workup between 1/1/2016 and 12/31/2017 at the University Hospitals Cleveland Medical Center were enrolled in this study. All specimens were formalin-fixed and paraffin-embedded tissue sections. Hematoxylin and eosin-stained slides were reviewed. All the cases were screened by the junior pathologists and then independently reviewed by two experienced gastrointestinal pathologists. Only the cases with a consensus of CRC diagnosis among three pathologists were included in the study.

Detailed clinicopathological data were retrieved from the respective pathology reports/clinical records, including tumor diagnosis, staging, locations, and patient follow-up status. Survival time was correlated with the medical record chart in the data analysis.

DNA extraction

Formalin-fixed, paraffin-embedded tissues were retrieved for DNA extraction. Unstained slides were accompanied by an H&E-stained slide to allow the identification of cancer cells (>20% viable cancer cell nuclei) by a pathologist. QIAamp FFPE DNA Isolation Kit (Qiagen, Valencia, CA) was used for genomic DNA isolation following the suggested procedures.

RAS (KRAS and NRAS) and APC status evaluation

Gene status of RAS (KRAS and NRAS) and APC were determined by the next-generation sequencing (NGS) at University Hospitals Cleveland Medical Center’s Clinical Laboratory Improvement Amendments-certified Translational Laboratory (UHTL). Ion AmpliSeqTM Cancer Hotspot Panel v2 (CHPv2) library kit was used to establish DNA library preparation following the standard protocol (Thermo Fisher, Carlsbad, CA) with the amount of 30 ng of genomic DNA. The template amplification and enrichment were prepared using the One Touch 2 system. The OT2 200 sequencing kit was used for the sequencing, which was performed on Ion Torrent PGM with 8–12 samples per 318-Chip.

BRAF V600E status evaluation

The c.1799T>A (V600E) point mutation was detected by the allele-specific hybridization of PCR amplification. The reagents used for the Applied Biosystems 7500 Fast Real-Time PCR System allele-specific Taqman® probes, fluorescently labeled with VIC-(wild-type) or FAM-(V600E mutant). Delta Ct value was calculated as the difference in Ct value between wild-type and mutant BRAF allele. Results were interpreted based on the patient delta Ct value of V600E relative to the corresponding 5% sensitivity controls.

Immunohistochemistry (IHC) evaluation of HES1

Protein expression of HES1 was evaluated by IHC on all cases included in the study, as described in our previous paper.12 The IHC was performed by the Immunohistochemistry Diagnostic Laboratory of University Hospitals Cleveland Medical Center. Tissue slides were processed using a BenchMark Ultra automated immunostainer (Ventana, Tucson, AZ). Slides were deparaffinized, antigen retrieved with standard Cell Conditioning 1 (Ventana Medical Systems, AZ), a tris-based buffer pH 8.3 solution for 64 minutes at 95°C, then incubated at 37°C with the primary antibody HES1 rabbit monoclonal (1:200 dilutions, clone EPR4226 from Epitomics, CA, part of Abcam, Cambridge, MA) for 24 minutes. Detection was performed with Ultraview-DAB from Ventana and subsequently counterstained.

Statistical analysis

Statistical analyses were performed with SPSS 16.0 (SPSS Inc., Chicago, IL). Categorical data were analyzed by the Fisher exact test (2-tailed). Cohen’s Kappa (κ) analysis was used to examine the agreement for HES1 loss and RAS/BRAF mutations. p value less than 0.05 was considered statistically significant.

Results

Patients

A total of 43 cases with primary CRC were enrolled in our study, with a mean patient age of 65.7 years (ranging from 31 to 83 years) and a male-to-female ratio of 1.9:1. The ethnic distribution was as follows: Caucasian 21 (48.8%), African American 18 (41.9%), Asian 2 (4.7%) and others/unknown 2 (4.7%). The majority were resection specimens (35/43, 81.4%), while 8 were biopsies of the primary colon masses. Fourteen specimens (32.6%) were from the right colon (cecum and ascending colon), and 29 specimens (67.4%) were from the left colon (transverse colon, descending colon, sigmoid colon, and rectum).

RAS, BRAF V600E, and APC mutations of CRC

RAS, BRAF V600E, and APC gene status were identified by chart review or examined using the same methods if they were not performed previously. Successful molecular evaluations were acquired from all 43 specimens by NGS.

Mutations of either KRAS or NRAS were considered RAS mutation-positive, which account for 55.8% of all tested specimens (24/43). Among them, twenty-two (22/24, 91.7%) had KRAS mutation, and the other two (2/24, 8.3%) had NRAS mutation. Right-side tumors harbored a higher rate of RAS mutations (10/14, 71.4%) compared to left-side tumors (14/29, 48.3%) (p < 0.05, chi-square test). The most frequent mutation observed was a KRAS point mutation in codon 12 (16/24, 66.7%), including KRAS p. G12V (c.35G>T) (6/24, 25.0%), followed by KRAS p. G12D (c.35G>A) (5/24, 20.8 %). A KRAS point mutation in codon 13 [p. G13D (c.38G>A)] was observed in four cases (4/24, 16.7%).

BRAF V600E mutation was identified in four cases (4/43, 9.3%), which were mutually exclusive to KRAS mutations. Two specimens (50.0%) were from the right-side colon, and the other two (50.0%) were from the left side. In addition, DNA mismatch repair (MMR) gene deficiency was detected by IHC and confirmed by NGS in two cases harboring BRAF V600E mutation and one case without BRAF V600E mutation.

APC mutation was found in 12 specimens (12/43, 27.9%), with eight from the left-sided colon and four from the right-sided colon.

HES1 expression, RAS/BRAF V600E, and APC mutations

Overall, loss of HES1 nuclear expression was found in two-thirds of the cancer specimens (28/43, 65.1%), with a much higher frequency in right-sided tumors (13/14, 92.9%), compared to the left-sided (15/29, 51.7%) (p < 0.05) (Fig. 1). We also noticed that HES1 absence was more frequently identified in tumor tissue with either RAS or BRAF V600E mutations than cases that remained in wild-type (78.6% vs. 40.0%, p < 0.05) (Table 1). Importantly, further analysis with tumor location and RAS/BRAF V600E mutations showed that all the specimens from the right-side colon with RAS or BRAF mutations had a loss of HES1 expression (12/12, 100%) (p < 0.05), while only 62.5% (10/16) of left-sided tumors with RAS or BRAF mutation had lost HES1 expression (Table 1). The concordance of HES1 loss and RAS/BRAF V600E mutations was 92.8% (13/14, κ = 0.89; p < 0.01) in right-sided tumors.

HES1 expression by IHC.
Fig. 1  HES1 expression by IHC.

(a) Nuclei expression of HES1 is identified in one case with RAS and BRAF wide type (200×); (b) It is usually lost in cases with mutated RAS or BRAF gene. Lymphocytes serve as an internal control in negative cases (200×).

Table 1

RAS or BRAF V600E mutation and HES1 expression in the right/left colorectal adenocarcinoma

Right side (n = 14)
Left side (n = 29)
HES1RAS/BRAF*
pRAS/BRAF*
p
Wild-typeMutatedWild-typeMutated
Present1 (50.0%)0 (0)0.018 (61.5%)6 (37.5%)0.19
Loss1 (50.0%)12 (100%)5 (38.5%)10 (62.5%)

Among 3 MMR deficiency cases, HES1 loss was identified in one case (1/3, 33.3%) and preserved in the other two cases (2/3, 66.7%), including one LYNCH patient. However, there is no reliable conclusion due to the minimal case number here.

HES1 loss was found in 83.3% (10/12) cases with APC mutations and 58.1% (18/31) cases without APC mutations. However, our studies observed no statistically significant differences between HES1 expression and APC mutations.

HES1 expression and distant metastasis

A clinical chart review for distant metastasis was performed on all the cases. More than a half of our patients presented with stage IV disease (24/43, 55.8%), including metastasis to liver (n = 19), lung (n = 3), ovary (n = 1) and peritoneum (n = 1). Twelve patients (12/43, 27.9%) lived without metastasis in our observation window. One case (1/43, 2.3%) developed local recurrence. The rest cases (6/43, 14.0%) developed metastasis during the follow-up after initial diagnosis with time-to-metastasis ranging from 2 months to 48 months. The relationship between HES1 loss and distant metastasis was also analyzed in patients harboring RAS/BRAF V600E mutations. Patients with HES1 loss in tumor tissue were less likely to have immediate metastasis (at initial diagnosis or within six months) (59.1%, 13/22) compared to patients with preserved HES1 expression (83.3%, 5/6) (p = 0.37) (Table 2).

Table 2

HES1 mutation and distance metastasis in patients with RAS/BRAF V600E mutations

HES1Immediate metastasis (within 6 months)
p
Yes (n = 18)No (n = 10)
Absent (n = 22)13 (59.1%)9 (40.9%)0.37
Present (n = 6)5 (83.3%)1 (16.7%)

Discussion

HES1 protein is expressed in the nuclei of epithelial cells of the intestine in humans and mice. However, the expression of HES1 in CRC shown in the literature is inconsistent. In a study of Notch effectors in intestinal tumorigenesis, HES1 expression was identified by IHC on all 14 random CRC specimens.13 Fre et al. reported that HES1 was strongly expressed in adenomas but not expressed in 14 cases of human colon adenocarcinomas, indicating that loss of HES1 could be associated with the carcinogenic transformation of adenoma.14 Some studies showed the heterogeneous nature of HES1 expression in cancer tissue. Gao et al. reported that HES1 was associated with the differentiation of the adenocarcinoma, with a stronger expression in poorly differentiated adenocarcinoma.15 Our previous study of an animal model with genetic defects in Notch signaling indicated that suppressed HES1 expression was associated with adenocarcinoma development in mouse colon.11

Previous reports have shown differences between the right-side colon (including the caecum, ascending colon, and transverse colon) and the left-side colon (including the descending colon, sigmoid colon, and rectum) in epidemiologic incidence, morphology, and molecular alterations.16–18 Our study, for the first time, showed that the loss of HES1 expression was highly associated with right-sided CRC (92.9%) and was absent in all the right-sided cases with either RAS or BRAF V600E-activated mutations (100.0%). This finding highlights the question, “does HES1, or the Notch pathway, have a role in carcinogenesis with the MAPK-ERK pathway?” Activation mutations of RAS and BRAF would lead to upregulation of the MAPK-ERK pathway, which was the most common oncogenic event in colorectal tumorigenesis.19 Suppression of HES1 in the tumor tissue with activated MAPK-ERK pathway suggested a possible role of HES1 in the carcinogenesis mediated by MAPK-ERK pathway. Very recently, HES1 was reported to be involved in the initiation and progression of KRAS-driven pancreatic tumorigenesis in mice.20 In their study, activation of MAPK signaling via mutant KRAS activation induced sustained HES1 expression in pancreatic acinar cells. Interestingly, our finding indicates HES1 loss, rather than retained expression in the above study, associated with RAS-mutant CRC, which enhances the complexity of the role of HES1 in the MAPK-ERK pathway.

Although patients diagnosed with localized early-stage tumors can be treated with surgery and adjuvant regimens, most patients diagnosed with distant metastasis succumb to cancer.21 It is, therefore, critical to understand the mechanisms that enable CRC to metastasize and develop effective strategies to prevent metastasis. So far, there is little evidence for activating Notch gene mutations in human CRC metastasis. Recently, Jackstadt et al.10 showed that enforced Notch signaling led to highly penetrant metastasis in KRASG12D mutation-driven serrated colon cancer in the murine intestinal epithelium, which is consistent with our finding that patients with HES1 loss were less likely to have immediate metastasis, compared to patients with preserved HES1 expression. In our study, the most frequent mutation of KRAS is p. G12V (c.35G>T) (6/24, 25.0%), followed by KRAS p. G12D (c.35G>A) (5/24, 20.8 %). Unfortunately, due to the limitation of the case number, our study did not reach statistical significance. However, it would be interesting to further explore the topic with a larger sample volume.

RAS and BRAF mutation status were essential markers closely related to patient management and prognosis prediction, which clinicians frequently requested. Currently, IHC can only detect BRAF V600E in limited organizations, while the RAS status exam needs molecular diagnostic study, which is usually expensive and takes a longer turn-around time compared to IHC. Therefore, finding markers that could predict the status and be easily examined with IHC is demanded. Based on our data, it is reasonable to propose HES1 as one of the candidates with excellent concordance in the right-sided tumor cases. While NGS is the current “golden standard” method, it would be helpful if a quick IHC test could serve as a predictor in local labs.

Conclusions

High frequency of HES1 loss in colorectal adenocarcinoma is associated with either RAS or BRAF mutations, suggesting synergistic effects by dysregulated Notch and RAS/BRAF mutation may play an important role in colon carcinogenesis in some forms, especially the right-sided tumors. This finding may help guide future treatments for a subset of colon cancers. Although the loss of Hes1 in RAS or BRAF mutations showed increased immediate metastasis compared to those with the normal Hes1 expression cases, the case number is not large enough to warrant a statistical difference. To explore the possibility, a large-scale study may be warranted to establish the relationship.

Abbreviations

CRC: 

colorectal carcinoma

HES1: 

Hairy enhancer of split-1

NGS: 

next generation sequencing

Notch1: 

Neurogenic locus notch homolog protein 1

Declarations

Acknowledgement

Part of the research was presented at the 107th United States and Canadian Academy of Pathology annual meeting (2019).

Ethical statements

The research protocol, including the HIPPA waiver, was approved by the institutional review committee at University Hospital Cleveland Medical Center/Case Western Reserve University. The project was finished before both authors left the original institutions. The IRB is still active under Dr. Lan Zhou as the new PI.

Data sharing statement

All data used to support the findings of this study are included in the article.

Funding

The research was partially funded by a faculty startup fund (Wei Xin and Lan Zhou), Department of Pathology, Case Western Reserve University.

Conflict of interest

The authors have no conflicts of interest related to this publication.

Authors’ contributions

Performance of experiments, analysis and interpretation of data, manuscript writing, statistical analysis (WC); study design, analysis and interpretation of data, critical revision, statistical analysis, critical funding, administration, technical or material support (LZ); study design, experimentation, analysis and interpretation of data, critical revision, statistical analysis, critical funding, administration, technical or material support (WX). All authors have contributed significantly to this study and approved the final manuscript.

References

  1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin 2020;70(1):7-30 View Article PubMed/NCBI
  2. Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, Leppert M, et al. Genetic alterations during colorectal-tumor development. N Engl J Med 1988;319(9):525-532 View Article PubMed/NCBI
  3. Takayama T, Ohi M, Hayashi T, Miyanishi K, Nobuoka A, Nakajima T, et al. Analysis of K-ras, APC, and beta-catenin in aberrant crypt foci in sporadic adenoma, cancer, and familial adenomatous polyposis. Gastroenterology 2001;121(3):599-611 View Article PubMed/NCBI
  4. Frattini M, Balestra D, Suardi S, Oggionni M, Alberici P, Radice P, et al. Different genetic features associated with colon and rectal carcinogenesis. Clin Cancer Res 2004;10(12 Pt 1):4015-4021 View Article PubMed/NCBI
  5. Harada K, Hiraoka S, Kato J, Horii J, Fujita H, Sakaguchi K, et al. Genetic and epigenetic alterations of Ras signalling pathway in colorectal neoplasia: analysis based on tumour clinicopathological features. Br J Cancer 2007;97(10):1425-1431 View Article PubMed/NCBI
  6. Samowitz WS, Curtin K, Schaffer D, Robertson M, Leppert M, Slattery ML. Relationship of Ki-ras mutations in colon cancers to tumor location, stage, and survival: a population-based study. Cancer Epidemiol Biomarkers Prev 2000;9(11):1193-1197 View Article PubMed/NCBI
  7. Miranda E, Destro A, Malesci A, Balladore E, Bianchi P, Baryshnikova E, et al. Genetic and epigenetic changes in primary metastatic and nonmetastatic colorectal cancer. Br J Cancer 2006;95(8):1101-1107 View Article PubMed/NCBI
  8. Noffsinger AE. Serrated polyps and colorectal cancer: new pathway to malignancy. Annu Rev Pathol 2009;4:343-364 View Article PubMed/NCBI
  9. Spring KJ, Zhao ZZ, Karamatic R, Walsh MD, Whitehall VL, Pike T, et al. High prevalence of sessile serrated adenomas with BRAF mutations: a prospective study of patients undergoing colonoscopy. Gastroenterology 2006;131(5):1400-1407 View Article PubMed/NCBI
  10. Jackstadt R, van HooffSR, Leach JD, Cortes-Lavaud X, Lohuis JO, Ridgway RA, et al. Epithelial NOTCH Signaling Rewires the Tumor Microenvironment of Colorectal Cancer to Drive Poor-Prognosis Subtypes and Metastasis. Cancer Cell 2019;36(3):319-336.e7 View Article PubMed/NCBI
  11. Wang Y, Huang D, Chen KY, Cui M, Wang W, Huang X, et al. Fucosylation Deficiency in Mice Leads to Colitis and Adenocarcinoma. Gastroenterology 2017;152(1):193-205.e10 View Article PubMed/NCBI
  12. Cui M, Awadallah A, Liu W, Zhou L, Xin W. Loss of Hes1 Differentiates Sessile Serrated Adenoma/Polyp From Hyperplastic Polyp. Am J Surg Pathol 2016;40(1):113-119 View Article PubMed/NCBI
  13. Peignon G, Durand A, Cacheux W, Ayrault O, Terris B, Laurent-Puig P, et al. Complex interplay between β-catenin signalling and Notch effectors in intestinal tumorigenesis. Gut 2011;60(2):166-176 View Article PubMed/NCBI
  14. Fre S, Pallavi SK, Huyghe M, Laé M, Janssen KP, Robine S, et al. Notch and Wnt signals cooperatively control cell proliferation and tumorigenesis in the intestine. Proc Natl Acad Sci U S A 2009;106(15):6309-6314 View Article PubMed/NCBI
  15. Gao F, Zhang Y, Wang S, Liu Y, Zheng L, Yang J, et al. Hes1 is involved in the self-renewal and tumourigenicity of stem-like cancer cells in colon cancer. Sci Rep 2014;4:3963 View Article PubMed/NCBI
  16. Azzoni C, Bottarelli L, Campanini N, Di Cola G, Bader G, Mazzeo A, et al. Distinct molecular patterns based on proximal and distal sporadic colorectal cancer: arguments for different mechanisms in the tumorigenesis. Int J Colorectal Dis 2007;22(2):115-126 View Article PubMed/NCBI
  17. Okamoto M, Kawabe T, Yamaji Y, Kato J, Ikenoue T, Togo G, et al. Flat-type early colorectal cancer preferentially develops in right-sided colon in older patients. Dis Colon Rectum 2005;48(1):101-107 View Article PubMed/NCBI
  18. Gonzalez EC, Roetzheim RG, Ferrante JM, Campbell R. Predictors of proximal vs. distal colorectal cancers. Dis Colon Rectum 2001;44(2):251-258 View Article PubMed/NCBI
  19. Domingo E, Niessen RC, Oliveira C, Alhopuro P, Moutinho C, Espín E, et al. BRAF-V600E is not involved in the colorectal tumorigenesis of HNPCC in patients with functional MLH1 and MSH2 genes. Oncogene 2005;24(24):3995-3998 View Article PubMed/NCBI
  20. Nishikawa Y, Kodama Y, Shiokawa M, Matsumori T, Marui S, Kuriyama K, et al. Hes1 plays an essential role in Kras-driven pancreatic tumorigenesis. Oncogene 2019;38(22):4283-4296 View Article PubMed/NCBI
  21. AJCC (American Joint Committee on Cancer). In: Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A (eds). Cancer Staging Manual. New York: Springer; 2010, 143 View Article PubMed/NCBI