Diao WF, Cui M, Chen TQ, Xiao JH, Yang S, Zheng QY, Xu RY, Han XL, Hu Y. MassARRAY-based KRAS and GNAS hotspot mutation analysis of cystic fluid enables accurate classification of pancreatic cystic lesions. World J Gastroenterol 2026; 32(13): 115710 [DOI: 10.3748/wjg.v32.i13.115710]
Corresponding Author of This Article
Ya Hu, MD, Professor, Department of General Surgery, Peking Union Medical College Hospital, No. 1 Shuaifuyuan, Dongcheng District, Beijing 100730, China. huya@pumch.cn
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Observational Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Apr 7, 2026 (publication date) through Mar 27, 2026
Times Cited of This Article
Times Cited (0)
Journal Information of This Article
Publication Name
World Journal of Gastroenterology
ISSN
1007-9327
Publisher of This Article
Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Diao WF, Cui M, Chen TQ, Xiao JH, Yang S, Zheng QY, Xu RY, Han XL, Hu Y. MassARRAY-based KRAS and GNAS hotspot mutation analysis of cystic fluid enables accurate classification of pancreatic cystic lesions. World J Gastroenterol 2026; 32(13): 115710 [DOI: 10.3748/wjg.v32.i13.115710]
Wen-Fei Diao, Ming Cui, Jin-Heng Xiao, Sen Yang, Qing-Yuan Zheng, Rui-Yuan Xu, Xian-Lin Han, Ya Hu, Department of General Surgery, Peking Union Medical College Hospital, Beijing 100730, China
Wen-Fei Diao, Ming Cui, Jin-Heng Xiao, Sen Yang, Rui-Yuan Xu, Xian-Lin Han, Ya Hu, Key Laboratory of Research in Pancreatic Tumor, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
Wen-Fei Diao, Ming Cui, Jin-Heng Xiao, Sen Yang, Rui-Yuan Xu, Ya Hu, National Infrastructures for Translational Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
Wen-Fei Diao, Ming Cui, Jin-Heng Xiao, Sen Yang, Rui-Yuan Xu, Ya Hu, State Key Laboratory of Complex, Severe, and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
Tian-Qi Chen, Biomedical Engineering Facility of National Infrastructures for Translational Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China
Author contributions: Diao WF, Cui M and Chen TQ conceptualized and designed the study; Diao WF, Chen TQ, Xiao JH and Yang S collected the data and cystic fluids sample; Xu RY is responsible for data curation; Diao WF, Cui M and Zheng QY performed data analysis; Diao WF and Cui M prepared the manuscript; Hu Y and Han XL supervised the study and performed draft reviewing and editing. Since Diao WF and Cui M contributed equally for the study completion and manuscript writing, they were listed as co-first authors. Hu Y and Han XL were listed as co-correspondence as they contributed equally in study supervision and draft reviewing. All authors contributed to this study and approved the manuscript to be published.
Supported by Noncommunicable Chronic Diseases-National Science and Technology Major Project, No. 2025ZD0552402; Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences, No. 2023-I2M-2-002; National High-Level Hospital Clinical Research Funding, No. 2022-PUMCH-D-001; Peking Union Medical College Hospital Talent Cultivation Program, Category D, No. UHB12625; and Milstein Medical Asian American Partnership (MMAAP) Foundation.
Institutional review board statement: The study was approved by the Ethics Committee of Peking Union Medical College Hospital (No. I-23PJ1601), in accordance with the ethical standards of the Institutional Research Committee and with the Helsinki declaration.
Informed consent statement: Informed consents were obtained from all participants in the study.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
STROBE statement: The authors have read the STROBE Statement—checklist of items, and the manuscript was prepared and revised according to the STROBE Statement—checklist of items.
Data sharing statement: The data that support the findings of this study are available on request from the corresponding author Ya Hu, upon reasonable request at huya@pumch.cn.
Corresponding author: Ya Hu, MD, Professor, Department of General Surgery, Peking Union Medical College Hospital, No. 1 Shuaifuyuan, Dongcheng District, Beijing 100730, China. huya@pumch.cn
Received: October 24, 2025 Revised: December 1, 2025 Accepted: February 3, 2026 Published online: April 7, 2026 Processing time: 154 Days and 19.5 Hours
Abstract
BACKGROUND
Pancreatic cystic fluid analysis through mutation sequencing has proven to be a valuable approach for classifying pancreatic cystic lesions (PCLs). However, a rapid and cost-effective method for screening these lesions is still needed.
AIM
To evaluate the application of MassARRAY system in detecting hotspot mutation for the rapid diagnosis of PCLs.
METHODS
A total of 101 surgically resected PCLs were analyzed in this study. DNA was extracted from intraoperatively collected pancreatic cystic fluid. A custom panel targeting KRAS and GNAS hotspot mutations was developed, and the performance of MassARRAY-based mutation detection in classifying PCLs was evaluated.
RESULTS
Intraductal papillary mucinous neoplasms (IPMNs) exhibited GNAS mutations in 52.2% and KRAS mutations in 39.1% of cases, with 33.3% showing both mutations. KRAS mutations were detected in 30.6% of mucinous cystic neoplasms (MCNs), while serous cystadenomas and non-neoplastic neoplasms showed no detectable mutations. A logistic regression model integrating KRAS and GNAS mutations in pancreatic cystic fluid, along with serum tumor biomarkers and clinical features, achieved an area under the curve greater than 0.9 for identification of IPMNs and 0.795 for MCNs.
CONCLUSION
Targeted hotspot mutation analysis of KRAS and GNAS with MassARRAY technique in pancreatic cystic fluid offers a promising application for the molecular classification of PCLs. This method holds significant potential for improving preoperative diagnosis and assisting clinical decision-making in the management of PCLs.
Core Tip: Pancreatic cystic lesions (PCLs) are highly heterogenous, which included benign, pre-malignant and malignant lesions. Current diagnostic approaches for PCLs exhibit limited efficacy and accuracy. The implementation of MassARRAY-based KRAS and GNAS mutation analysis of pancreatic cystic fluids improves the precise identification of mucinous cysts and intraductal papillary mucinous neoplasms, thereby providing critical support for clinical decision-making.
Citation: Diao WF, Cui M, Chen TQ, Xiao JH, Yang S, Zheng QY, Xu RY, Han XL, Hu Y. MassARRAY-based KRAS and GNAS hotspot mutation analysis of cystic fluid enables accurate classification of pancreatic cystic lesions. World J Gastroenterol 2026; 32(13): 115710
Pancreatic cystic lesions (PCLs) are common clinical entities, detected in 2.4%-13.5% of asymptomatic individuals undergoing abdominal imaging[1]. The incidence of PCLs raised gradually and nearly doubled in 7 years[2]. As major components of PCLs, serous cystadenomas (SCAs), mucinous cystic neoplasms (MCNs), and intraductal papillary mucinous neoplasms (IPMNs) constituted 70.7% of resected PCLs[3]. Notably, IPMNs and MCNs are recognized as malignant precursors, while SCAs rarely transformed to malignancy[4]. Pancreatic surgeries are often considered as high-risk procedures due to its association with high incidence of surgical complications and mortality. According to current guidelines, surgical resection would be taken into consideration when symptoms are observed for SCAs. For IPMNs and MCNs, since the potential of malignancy and the risk of surgery should be taken into consideration for risk to benefit ratio evaluation, the indication for surgical resection is challenging and controversial[5-7]. Consequently, accurate identification of specific histological types of PCLs prior to surgery is essential for making clinical decision.
Currently, diagnosis of PCLs primarily depends on radiological imaging such as computed tomography (CT) scan, magnetic resonance imaging. The accuracy of these cross-section imaging modalities in identifying specific histological type of PCLs vary from 40%-95%. Endoscopic ultrasound-guided fine-needle aspiration combined with cytology demonstrates a specificity of > 90% but a relatively low sensitivity of around 60%[1]. Additionally, pancreatic cystic fluid biomarkers, such as carcinoembryonic antigen (CEA), glucose, and amylase, are valuable for the diagnosis of PCLs. Although these biomarkers demonstrated considerable diagnostic accuracy, their sensitivity or specificity alone remained suboptimal. Notably, scenarios may arise where CEA, glucose or amylase levels in pancreatic cystic fluids slightly exceed the threshold but insufficiently to establish a definitive diagnosis. Consequently, there is a compelling need for more specific biomarkers with higher diagnostic precision.
The molecular landscape of PCLs exhibits significant heterogeneity across different histological subtypes, providing critical insights for their differential diagnosis. Both IPMNs and MCNs frequently harbor KRAS mutations. Notably, GNAS mutations are distinctive features of IPMNs but are characteristically absent in MCNs. In contrast, SCAs are predominantly associated with alterations in the VHL gene. Non-neoplastic PCLs, such as pancreatic pseudocysts, typically lack these defining driver mutations[8]. Besides, the mutational patterns also differ. KRAS and GNAS mutations primarily occur as hotspot mutations, whereas VHL alterations are predominantly driven by loss of heterozygosity (LOH)[8]. According to the specific genetic background, several studies had examined the application of next-generation sequencing (NGS) on pancreatic cystic fluid to identify the specific type of PCLs[9-12]. A multicenter prospective study indicated that a 22-gene NGS panel including KRAS, GNAS and VHL showed high sensitivity and specificity for identifying mucinous cysts, SCAs as well as advanced neoplasia[12]. However, the relatively high cost, time-consuming data analysis, complex cell free DNA extraction and detection techniques, along with substantial DNA requirements, may limit the widespread use of NGS in assisting the histological diagnosis of PCLs, especially for patients with small cysts, limited cystic fluid, or low DNA content in the pancreatic cystic fluid.
Agena MassARRAY platform (Agena Bioscience, United States) is a mass spectrometry-based technique capable of detecting hotspot mutations and single nucleotide polymorphisms. It has been shown to be rapid, cost-effective and reliable for detecting somatic mutation among various tumors[13]. Previous studies have demonstrated the feasibility of applying the MassARRAY platform in detecting genetic mutations for cancer detection and surveillance[14,15]. Therefore, the aim of this study is to develop a MassARRAY-based panel targeting KRAS and GNAS hotspot mutations, evaluate its performance in PCLs classification, and introduce a rapid and cost-effective approach for the clinical diagnosis of PCLs.
MATERIALS AND METHODS
Study cohort and data collection
The workflow of this study was shown in Figure 1A, a total of 101 patients in Peking Union Medical College Hospital between April 2021 to January 2025 were included in this study prospectively. Patients preoperatively diagnosed with PCLs, received surgical resection and had definite postoperative pathological diagnosis were included in this study, while those aged < 18 years or > 80 years old and accompanied with other malignancies were excluded. Pathological diagnosis of PCLs was based on the 2019 World Health Organization Classification of Tumors of the Digestive System[16].
Figure 1 Workflow of this study and heatmap summarizing the clinicopathological, radiological characteristics and hotspot mutation analysis of pancreatic cystic lesions.
A: DNA samples were extracted from pancreatic cystic fluid samples obtained from 101 pancreatic cystic lesions (PCLs) patients for KRAS and GNAS hotspot mutation analysis by MassARRAY, and the results were utilized for PCLs subtype diagnosis; B: Clinicopathological and radiological characteristics, KRAS and GNAS mutation status of 101 PCLs. IPMN: Intraductal papillary mucinous neoplasm; MCN: Mucinous cystic neoplasm; SCA: Serous cystadenoma; ECIPAS: Epidermoid cyst within an intrapancreatic accessory spleen; MALDI-TOF: Matrix-assisted laser desorption ionization time-of-flight; ROC: Receiver operating characteristic; CEA: Carcinoembryonic antigen; CA19-9: Cancer antigen 19-9; IPMN-IC: Intraductal papillary mucinous neoplasm with invasive carcinoma; IPMN-HG: Intraductal papillary mucinous neoplasm with high grade dysplasia; IPMN-LG: Intraductal papillary mucinous neoplasm with low grade dysplasia; MCN-IC: Mucinous cystic neoplasm with invasive carcinoma.
Clinical characteristics including gender, age, tumor location, pathological diagnosis, main pancreatic duct diameter, bile duct diameter, cyst diameter, serum CEA and cancer antigen 19-9 (CA19-9) levels, serum total and direct bilirubin levels were recorded. Main pancreatic duct dilation and bile duct dilation were defined as diameters of ≥ 5 mm and ≥ 7 mm, respectively. The thresholds for serum CEA, CA19-9, total bilirubin and direct bilirubin were 5 ng/mL, 34 U/mL, 22.2 μmol/L and 6.8 μmol/L, respectively.
Sample collection and DNA extraction
For each patient enrolled in this study, pancreatic cystic fluid was collected with syringe during surgery for DNA extraction or stored at -80 °C immediately for further analysis. Genomic DNA was extracted from pancreatic cystic fluid using Magnetic Serum/Plasma DNA Maxi Kit (DP710, TIANGEN, China) according to the manufacturer's instructions. Briefly, pancreatic cystic fluid samples were firstly mixed with proteinase K, Buffer GHH and magnetic beads at room temperature for 20 minutes. Then, magnetic beads were precipitated and supernatant were removed, followed by mixing with Buffer GDF to magnetic beads samples for 30 seconds. After another removal of supernatant, Buffer PWG was added to wash the magnetic beads twice and finally Buffer TBC was added and mixed at 56 °C for 5 minutes to elute DNA. Extracted DNA was then stored at -80 °C for further analysis.
Hotspot mutation analysis by MassARRAY
Specific primers were designed according to the genomic sequence of hotspot mutations in KRAS (codons 12, 13, 61, 146) and GNAS (codon 201) (Supplementary Table 1). Multiplex PCR amplification was performed on each genomic DNA samples and a single-base extension reaction was then conducted using sequence-specific extension primers and mass-modified dideoxynucleotides (Supplementary Table 2). After that, the extension products were then analyzed by matrix-assisted laser desorption ionization time-of-flight mass spectrometry, and allele-specific peaks were analyzed by Typer Analyzer software (version 4.1).
Statistical analysis
Continuous variables were analyzed using Student t-tests and expressed as mean ± SD, while categorical variables were assessed with χ2 tests and presented as n (%). Statistical analysis was performed using the SPSS software (version 24.0, IBM). Logistic regression models to predict the diagnosis of PCLs according to clinical characteristics (whether pancreatic cyst diameter ≥ 30 mm, main pancreatic duct dilation, bile duct dilation or not) and serum markers (whether serum CEA ≥ 5 ng/mL, CA19-9 ≥ 34 U/mL, serum total bilirubin ≥ 22.2 μmol/L and direct bilirubin ≥ 6.8 μmol/L or not) and KRAS/GNAS mutation status (wild type or mutations) were constructed using R software (version 4.4.1). Receiver operating characteristic (ROC) curves were generated and area under the curve (AUC) values were computed using R software (version 4.4.1) with pROC package. Sensitivity and specificity were calculated with reportROC package.
RESULTS
Patient characteristics
A total of 101 patients were included in this study, which consisting 23 IPMNs, 36 MCNs, 32 SCAs, 9 pancreatic pseudocysts and 1 epidermoid cyst within an intrapancreatic accessory spleen (ECIPAS) (Figure 1B). Among the 23 IPMNs, 7 were diagnosed as IPMN with invasive carcinoma, 4 IPMN with high grade dysplasia and 12 IPMN with low grade dysplasia. Among 36 MCNs, 1 was diagnosed as MCN with invasive carcinoma. Nine pseudocysts and 1 ECIPAS were characterized as non-neoplastic cysts.
The demographic characteristics of patients with PCLs are summarized in Table 1. Significant intergroup differences were observed in gender distribution (P < 0.001), age (P < 0.001), lesion location (P = 0.007), main pancreatic duct dilation (P < 0.001), bile duct dilation (P = 0.041), and serum CEA levels (P = 0.012). Patients with IPMN were predominantly male and older, whereas MCNs, SCAs, and non-neoplastic cysts occurred more frequently in younger females. 47.8% IPMNs were found located in the pancreatic head, while other PCLs mainly located in the body or tail. Although cysts ≥ 30 mm predominated across all groups in this study, IPMNs exhibited smaller mean diameters compared to MCNs, SCAs, and non-neoplastic cysts (43.09 ± 17.81 mm vs 56.79 ± 33.5 mm, 48.88 ± 18.49 mm, and 68.10 ± 27.63 mm, respectively). IPMNs also demonstrated higher rates of main pancreatic duct dilation (69.6%), bile duct dilation (17.4%), and elevated serum CEA levels (21.7%). No significant differences were observed in serum CA19-9, total bilirubin or direct bilirubin levels among the four PCLs subtypes.
MassARRAY-based detection of hotspot mutations in pancreatic cyst fluid
Hotspot mutations at KRAS codons 12, 13, 61, 146 and GNAS codon 201 were analyzed in 101 PCLs (Table 2). KRAS hotspot mutations were detected in 9 (39.1%) of 23 IPMN cases, predominantly comprising G12D (13%), G12V (17.4%) mutations, Q61H (4.3%) and A146V (17.4%) mutations. GNAS hotspot mutation was detected in 12 (52.2%) of 23 IPMNs, whereas 34.8% and 17.4% were GNAS R201H and GNAS R201C mutations. Among MCNs, KRAS hotspot mutation were found in 11 (30.6%) of 36 cases. Specifically, KRAS codon 12 hotspot mutation predominated (G12V: 11.1%, G12D: 8.3%, G12R: 2.8%), while KRAS A146V and Q61H accounted for 8.3% and 5.6%. All SCAs and non-neoplastic cysts harbored no KRAS or GNAS mutations. Additionally, no KRAS codon 13 hotspot mutations were detected.
Table 2 Summary of hotspot mutation analysis, n (%).
Diagnostic performance of clinical, molecular, and integrated models in PCLs
Given that IPMNs and MCNs represent precursors to pancreatic adenocarcinoma, distinguishing these mucinous neoplasms from non-mucinous cysts are clinically critical. Therefore, we evaluated the effect of clinical characteristics and serum markers, KRAS/GNAS mutation panel and an integrated model, which combined clinical characteristics, serum markers and KRAS/GNAS mutation status, in identifying specific histological subtypes of PCLs.
As shown in Figure 2A, detecting KRAS/GNAS mutation in pancreatic cystic fluids could effectively identify mucinous neoplasms with AUC of 0.703 (95%CI: 0.640-0.767), sensitivity of 40.7% (95%CI: 28.1%-53.2%) and specificity of 100%. However, when based on clinical characteristics and serum markers only, it achieved an AUC of 0.674 (95%CI: 0.577-0.770), sensitivity of 37.3% (95%CI: 24.9%-49.6%) and specificity of 95.2% (95%CI: 88.8%-100%). The integrated model showed better performance with AUC of 0.795 (95%CI: 0.711-0.878), sensitivity of 62.7% (95%CI: 50.4%-75.1%) and specificity of 95.2% (95%CI: 88.8%-100%).
Figure 2 Receiver operating characteristic curves for identification of mucinous neoplasms and intraductal papillary mucinous neoplasms among all pancreatic cystic lesions and pancreatic cystic lesions located in the pancreatic head.
A: Among all pancreatic cystic lesions (PCLs), the integrated prediction model, KRAS and GNAS mutation status, clinical characteristics and serum markers achieved area under the curve (AUC) of 0.795, 0.703 and 0.674 for identification of mucinous neoplasms, respectively; B: For identification of intraductal papillary mucinous neoplasms (IPMNs) across all pancreatic cysts, the integrated prediction model yielded an AUC of 0.922, while KRAS and GNAS mutation status alone yielded 0.773 and clinical characteristics with serum markers yielded 0.849; C: Among pancreatic head PCLs, mucinous neoplasms detection demonstrated AUCs of 0.977 (integrated prediction model), 0.801 clinical characteristics and serum markers), and 0.781 (KRAS and GNAS mutation status); D: For IPMNs in pancreatic head cysts, AUC values reached 0.978 (integrated prediction model), 0.852 (clinical characteristics and serum markers), and 0.731 (KRAS and GNAS mutation status). PCLs: Pancreatic cystic lesions; AUC: Area under the curve; IPMNs: Intraductal papillary mucinous neoplasm.
In the identification of IPMNs, the KRAS/GNAS mutation panel achieved an AUC of 0.773 (95%CI: 0.664-0.882) (Figure 2B), with a sensitivity of 52.2% (95%CI: 31.8%-72.6%) and a specificity of 100%. In contrast, depending on clinical characteristics and serum markers solely yielded an AUC of 0.849 (95%CI: 0.753-0.944) (Figure 2B), with sensitivity of 78.3% (95%CI: 61.4%-95.1%) and specificity of 80.8% (95%CI: 72.0%-89.5%). The integrated model demonstrated superior diagnostic performance, achieving an AUC of 0.922 (95%CI: 0.839-1.000) (Figure 2B), sensitivity of 82.6% (95%CI: 67.1%-98.1%), and specificity of 100% (Table 3).
Table 3 Performance of different diagnostic modalities.
Sensitivity (95%CI)
Specificity (95%CI)
Identification of mucinous neoplasms among all PCLs
Integrated prediction model
62.7% (50.4%-75.1%)
95.2% (88.8%-100%)
KRAS/GNAS mutation panel
40.7% (28.1%-53.2%)
100% (100%-100%)
Clinical characteristics and serum markers
37.3% (24.9%-49.6%)
95.2% (88.8%-100%)
Identification of IPMNs among all PCLs
Integrated prediction model
82.6% (67.1%-98.1%)
100% (100%-100%)
KRAS/GNAS mutation panel
52.2% (31.8%-72.6%)
100% (100%-100%)
Clinical characteristics and serum markers
78.3% (61.4%-95.1%)
80.8% (72.0%-89.5%)
Identification of mucinous neoplasms among PCLs in pancreatic head
Site-specific discrimination of mucinous neoplasms and IPMNs in pancreatic head cystic lesions
Currently, the diagnosis of PCLs in the head of the pancreas is a significant clinical challenge, as these cysts typically require pancreaticoduodenectomy, which may lead to severe complications. Therefore, identifying those patients truly needed for surgery are critical for making clinical decision.
Of the 101 PCLs enrolled in this study, 27 PCLs involved the pancreatic head. Additionally, there were 13 IPMNs, 3 MCNs, 9 SCAs and 2 pseudocysts (Figure 1B). In the discrimination of mucinous neoplasms among pancreatic head PCLs (Figure 2C), the MassARRAY-based KRAS/GNAS mutation panel achieved an AUC of 0.781 (95%CI: 0.656-0.907), with a sensitivity of 56.2% (95%CI: 31.9%-80.6%) and specificity of 100%. The model based solely on clinical characteristics and serum markers achieved an AUC of 0.801 (95%CI: 0.632-0.970), a sensitivity of 62.5% (95%CI: 38.8%-86.2%), and specificity of 100%. Notably, the integrated model demonstrated significantly improved performance, achieving an AUC of 0.977 (95%CI: 0.942-1.000), sensitivity of 87.5% (95%CI: 71.3%-100%), and specificity of 100%.
For distinguishing IPMNs within pancreatic head PCLs (Figure 2D), the MassARRAY-based KRAS/GNAS mutation panel achieved an AUC of 0.731 (95%CI: 0.590-0.872), sensitivity of 46.2% (95%CI: 19.1%-73.3%), and specificity of 100%. Utilizing clinical characteristics and serum markers achieved an AUC of 0.852 (95%CI: 0.703-1.000), sensitivity of 62.5% (95%CI: 38.8%-86.2%), and specificity of 100%. Critically, the integrated prediction model achieved the highest performance with an AUC of 0.978 (95%CI: 0.943-1.000), sensitivity of 84.6% (95%CI: 65.0%-100%) and specificity of 100%.
DISCUSSION
Evaluating a patient with PCLs is critical for making clinical decision, deciding follow-up or receiving surgical resection. The first key point of evaluating a PCLs is defining whether it is a mucinous cyst or a non-mucinous cyst, since mucinous cysts have the potential to be malignant and closer follow-up or surgical resection are recommended according to guidelines[5-7]. Among those mucinous cysts, MCNs usually located in the pancreatic body or tail and did not involve the main pancreatic duct, it would be easily distinguished from mucinous neoplasms[17]. Therefore, the second key point is to identify IPMNs that may require surgical resection, especially those located in pancreatic head. SCAs located in pancreatic head could also lead to pancreatic duct dilation and atrophy, which were thought to be the characteristics of IPMNs and may finally lead to misdiagnosis and unnecessary surgery[18,19]. Besides, pancreatoduodenectomy, a procedure with high rate of complication and mortality, is often introduced to patients with PCLs in the pancreatic head and surgical resection is needed[20,21]. Thus, misdiagnosis of specific histological types of PCLs and unnecessary surgery may contribute to poor diagnosis. Therefore, it is essential to identify the histological types of PCLs located in the pancreatic head and correctly identify those PCLs that are truly require surgical resection.
KRAS mutations are frequently observed in several human solid tumors including colorectal cancer, non-small cell lung cancer, and pancreatic cancer[22]. Mutation of KRAS activates the RAS-mitogen-activated protein kinase and PI3K-AKT pathways, therefore promoting tumor growth[23]. As a critical oncogenic driver, therapeutic strategies targeting KRAS mutations, such as pan-KRAS inhibitors, mutation-specific KRAS inhibitors, and proteolysis-targeting chimeras, are currently under investigation[24,25]. Meanwhile, detection of KRAS mutations has been extensively investigated through liquid biopsy for cancer screening. In IPMNs and MCNs, hotspot mutations represent the most prevalent types of KRAS mutations, notably G12D and G12V. Similarly, GNAS mutations are commonly found in IPMNs rather than MCNs[8]. The presence of these GNAS mutations contribute to mucin production and cystic lesions formation. Within IPMNs, mutations at the GNAS R201 codon, particularly R201C and R201S, are most frequent[26,27]. Consequently, the detection of KRAS or GNAS mutations in the cyst fluid of PCLs provides direct impression to specific biological characteristics. These molecular characteristics are critical for clarifying the histological types of PCLs and guiding clinical management.
MassARRAY genotyping technique has been widely used in liquid biopsy of cancer detection or monitoring. Genomic mutation could be detected with the use of MassARRAY genotyping technique from low content of DNA samples, such as circulating tumor DNA and circulating cell-free DNA samples[15,28,29]. On the other hand, since different histological subtypes of PCLs shared different genomic background, previous studies have shown that using NGS in pancreatic cystic fluids to identified genomic mutation were effective in PCLs classification[10-12,30]. Based on the advantage of MassARRAY genotyping technique and specific genetic background of PCLs, we assessed the effect of adapting MassARRAY genotyping technique in evaluating KRAS/GNAS mutation status in pancreatic cystic fluids for PCLs diagnosis. In this study, 9 (39.13%) cases harbored KRAS mutation and 12 (52.17%) harbored GNAS mutation among 23 IPMNs, whereas 11 (30.6%) cases harbored KRAS mutation among 36 MCNs. It was reported that 50%-57% harbored KRAS mutation while 47%-79% harbored GNAS mutation with NGS in tumor tissue of IPMNs[31-33]. Between different histological types of IPMNs, it was reported that KRAS mutation ranged from 39%-83% and GNAS mutation ranged from 33%-100%[33]. As for MCNs, it was reported 35.7%-94.4% harbored KRAS mutation according to previous small-scale studies[17,34]. The KRAS and GNAS mutation rate with MassARRAY in pancreatic cystic fluids remained similar to the mutation rate in tumor tissues reported in previous studies, which indicating that MassARRAY-based KRAS and GNAS mutation analysis in pancreatic fluids was an effective method to identify genomic mutation in PCLs. It has been reported that only 93%-98% cases had sufficient DNA content in pancreatic cystic fluids that are satisfied for NGS testing, while $750 cost per specimen and 10 days for report on average were required for NGS testing[11,12]. In this study, among a total of 101 cases and 8 detection sites, only 1 case at KRAS codon 61 and 2 cases at KRAS codon 146 reported ‘No Call’ because of insufficient DNA content in cystic fluids, which refers to the inability to differentiate between wild type and hotspot mutation. Besides, these 3 cases yielded definitive results at other codons of KRAS and GNAS. Given that the majority of KRAS mutations occur at codon 12, these 3 ‘No Call’ results had a nearly negligible impact on the overall diagnostic sensitivity. In general, less DNA content from pancreatic cystic fluids was required for KRAS and GNAS hotspot mutation analysis with MassARRAY technique and the results could be reported within 3 days with a cost of about $50 per specimen. Therefore, MassARRAY-based KRAS and GNAS hotspot mutation analysis in pancreatic cystic fluids could serve as an effective method for assisting PCLs diagnosis, particularly when rapid and cost-effective analysis are required or when evaluating PCLs with small cyst size.
In this study, depending on the results of KRAS/GNAS mutation analysis alone reached 100% specificity but only 40.7% and 56.2% sensitivity for identifying mucinous neoplasms and IPMNs. In comparison, though clinical characteristics and serum markers showed relatively lower specificity, it showed higher sensitivity in identifying IPMNs than using the results of KRAS/GNAS mutation analysis alone. Then we developed a prediction model for PCLs classification, which integrating clinical characteristics, serum markers and KRAS/GNAS mutation analysis. The integrated prediction model reached 62.7% sensitivity and 82.6% specificity in identifying mucinous neoplasms, while it achieved 87.5% sensitivity and 84.6% specificity in identifying IPMNs. Besides, among the PCLs located in pancreatic head, integrating KRAS/GNAS mutation analysis improved the sensitivity and specificity in identification of mucinous neoplasms and IPMNs. Therefore, the AUC of the integrated prediction model reached 0.795 for identification of mucinous neoplasms among all PCLs and exceeded 0.9 for identification of IPMNs among all PCLs. Among PCLs in the pancreatic head, the integrated prediction model also showed AUC over 0.9 in identifying both mucinous neoplasms and IPMNs. Besides, as a consequence of higher KRAS/GNAS mutations detection rate, previous studies utilizing NGS to examine KRAS/GNAS mutation status achieved nearly 90% sensitivity and 100% specificity in assisting classification of PCLs subtypes[9-12]. These results indicated that KRAS/GNAS mutation analysis in pancreatic cystic fluids may add values to diagnosis of PCLs.
This study had some limitations. Firstly, selection bias might have occurred since only patients receiving surgical resection in a single center were included. The diagnostic performance of MassARRAY-based KRAS and GNAS hotspot mutation analysis for patients without surgical indication and under longitudinal surveillance remains to be validated through large-scale, prospective and multicenter studies. Secondly, since other relevant genes including TP53, VHL and RNF43 are generally altered by LOH and frameshift mutation rather than hotspot mutation, only KRAS and GNAS hotspot mutation were included in the MassARRAY-based panel in this study. More hotspot mutations of other specific driver genes are expected to facilitate wider applicability. Despite these limitations, this KRAS and GNAS hotspot mutation panel has the potential for clinical application in identifying mucinous cysts and IPMNs.
CONCLUSION
In conclusion, although MassARRAY-based KRAS and GNAS hotspot mutation analysis showed slightly lower sensitivity compared to NGS, it demonstrated good resolution in identifying mucinous neoplasms and IPMNs, especially in small cysts and located in pancreatic head. Therefore, MassARRAY-based KRAS and GNAS hotspot mutation analysis of pancreatic cystic fluids could serve as a supplementary method for PCLs classification, particularly for those unsatisfactory for NGS or when rapid and cost-effective methods are required.
ACKNOWLEDGEMENTS
We extend our sincere gratitude to all patients participated in this study. We also thank Bio Miao Biological Technology (Beijing, China) for support in usage of MassARRAY platform.
Singhi AD, McGrath K, Brand RE, Khalid A, Zeh HJ, Chennat JS, Fasanella KE, Papachristou GI, Slivka A, Bartlett DL, Dasyam AK, Hogg M, Lee KK, Marsh JW, Monaco SE, Ohori NP, Pingpank JF, Tsung A, Zureikat AH, Wald AI, Nikiforova MN. Preoperative next-generation sequencing of pancreatic cyst fluid is highly accurate in cyst classification and detection of advanced neoplasia.Gut. 2018;67:2131-2141.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 210][Cited by in RCA: 294][Article Influence: 36.8][Reference Citation Analysis (0)]
Paniccia A, Polanco PM, Boone BA, Wald AI, McGrath K, Brand RE, Khalid A, Kubiliun N, O'Broin-Lennon AM, Park WG, Klapman J, Tharian B, Inamdar S, Fasanella K, Nasr J, Chennat J, Das R, DeWitt J, Easler JJ, Bick B, Singh H, Fairley KJ, Sarkaria S, Sawas T, Skef W, Slivka A, Tavakkoli A, Thakkar S, Kim V, Vanderveldt HD, Richardson A, Wallace MB, Brahmbhatt B, Engels M, Gabbert C, Dugum M, El-Dika S, Bhat Y, Ramrakhiani S, Bakis G, Rolshud D, Millspaugh G, Tielleman T, Schmidt C, Mansour J, Marsh W, Ongchin M, Centeno B, Monaco SE, Ohori NP, Lajara S, Thompson ED, Hruban RH, Bell PD, Smith K, Permuth JB, Vandenbussche C, Ernst W, Grupillo M, Kaya C, Hogg M, He J, Wolfgang CL, Lee KK, Zeh H, Zureikat A, Nikiforova MN, Singhi AD. Prospective, Multi-Institutional, Real-Time Next-Generation Sequencing of Pancreatic Cyst Fluid Reveals Diverse Genomic Alterations That Improve the Clinical Management of Pancreatic Cysts.Gastroenterology. 2023;164:117-133.e7.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 61][Cited by in RCA: 134][Article Influence: 44.7][Reference Citation Analysis (1)]
Leest PV, Janning M, Rifaela N, Azpurua MLA, Kropidlowski J, Loges S, Lozano N, Sartori A, Irwin D, Lamy PJ, Hiltermann TJN, Groen HJM, Pantel K, Kempen LCV, Wikman H, Schuuring E. Detection and Monitoring of Tumor-Derived Mutations in Circulating Tumor DNA Using the UltraSEEK Lung Panel on the MassARRAY System in Metastatic Non-Small Cell Lung Cancer Patients.Int J Mol Sci. 2023;24:13390.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 12][Reference Citation Analysis (0)]
Pea A, Bevere M, Gkountakos A, Pasini D, Fiorini D, Mafficini A, Golovco S, Simbolo M, Pedron S, Sciammarella C, Mattiolo P, Mombello A, Villanova M, Franzina C, Masetto F, Ciulla C, Sperandio N, Fujikura K, Ahadi MS, Samra JS, Johns AL, Verheij J, Stommel MWJ, van Santvoort H, Schubert Santana L, Malleo G, Milella M, Brosens LAA, Wood LD, Chang DK, De Robertis R, D'Onofrio M, Gill AJ, Salvia R, Corbo V, Lawlor RT, Scarpa A, Luchini C. Mucinous cystic neoplasms and simple mucinous cysts are two distinct precursors of pancreatic cancer: clinicopathological, genomic, and transcriptomic characterization.J Pathol. 2025;266:421-434.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 1][Cited by in RCA: 9][Article Influence: 9.0][Reference Citation Analysis (0)]
Eslami-S Z, Cortés-Hernández LE, Sinoquet L, Gauthier L, Vautrot V, Cayrefourcq L, Avoscan L, Jacot W, Pouderoux S, Viala M, Thomas QD, Lamy PJ, Quantin X, Gobbo J, Alix-Panabières C. Circulating tumour cells and PD-L1-positive small extracellular vesicles: the liquid biopsy combination for prognostic information in patients with metastatic non-small cell lung cancer.Br J Cancer. 2024;130:63-72.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 1][Cited by in RCA: 35][Article Influence: 17.5][Reference Citation Analysis (0)]
Springer S, Wang Y, Dal Molin M, Masica DL, Jiao Y, Kinde I, Blackford A, Raman SP, Wolfgang CL, Tomita T, Niknafs N, Douville C, Ptak J, Dobbyn L, Allen PJ, Klimstra DS, Schattner MA, Schmidt CM, Yip-Schneider M, Cummings OW, Brand RE, Zeh HJ, Singhi AD, Scarpa A, Salvia R, Malleo G, Zamboni G, Falconi M, Jang JY, Kim SW, Kwon W, Hong SM, Song KB, Kim SC, Swan N, Murphy J, Geoghegan J, Brugge W, Fernandez-Del Castillo C, Mino-Kenudson M, Schulick R, Edil BH, Adsay V, Paulino J, van Hooft J, Yachida S, Nara S, Hiraoka N, Yamao K, Hijioka S, van der Merwe S, Goggins M, Canto MI, Ahuja N, Hirose K, Makary M, Weiss MJ, Cameron J, Pittman M, Eshleman JR, Diaz LA Jr, Papadopoulos N, Kinzler KW, Karchin R, Hruban RH, Vogelstein B, Lennon AM. A combination of molecular markers and clinical features improve the classification of pancreatic cysts.Gastroenterology. 2015;149:1501-1510.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 378][Cited by in RCA: 346][Article Influence: 31.5][Reference Citation Analysis (0)]
Amato E, Molin MD, Mafficini A, Yu J, Malleo G, Rusev B, Fassan M, Antonello D, Sadakari Y, Castelli P, Zamboni G, Maitra A, Salvia R, Hruban RH, Bassi C, Capelli P, Lawlor RT, Goggins M, Scarpa A. Targeted next-generation sequencing of cancer genes dissects the molecular profiles of intraductal papillary neoplasms of the pancreas.J Pathol. 2014;233:217-227.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 298][Cited by in RCA: 263][Article Influence: 21.9][Reference Citation Analysis (0)]
Pflüger MJ, Fujikura K, Braxton AM, Lee JW, Zucha DM, Pedro BA, Goodman D, Lu J, Jiang L, Wang X, Zhu J, Dal Molin M, Wang H, Brosens LAA, He J, Kawamoto S, Jiao Y, Wood LD. Multiregion Genomic Analysis of Human Pancreatic Mucinous Cystic Neoplasms.Mod Pathol. 2025;38:100759.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 4][Reference Citation Analysis (0)]
Footnotes
Peer review: Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: China
Peer-review report’s classification
Scientific quality: Grade A, Grade A, Grade A, Grade B, Grade B
Novelty: Grade A, Grade A, Grade B, Grade B, Grade B
Creativity or innovation: Grade A, Grade A, Grade B, Grade B, Grade B
Scientific significance: Grade A, Grade A, Grade B, Grade B, Grade B
P-Reviewer: Li JT, MD, Assistant Professor, China; Qin SL, PhD, Full Professor, China; Tian L, Assistant Professor, Principal Investigator, China S-Editor: Li L L-Editor: A P-Editor: Wang WB
