INTRODUCTION
Among pancreatic neoplasms, serous cystadenoma (SCN) accounts for approximately 1%-2% of all pancreatic tumors, while pancreatic ductal adenocarcinoma (PDAC) is even less frequent[1-4]. The coexistent tumors in the pancreas are even rarer. To our knowledge, there have been only seven case reports[5-11]. The incidence rate of pancreatic cancer has been increasing annually, accompanied by exceedingly high mortality rates and poor prognostic outcomes[12]. With advancements in next-generation sequencing (NGS) technology, sequencing has increasingly been applied to patients with pancreatic tumors, in hopes of discovering key driving genes. The widely recognized driver genes associated with pancreatic cancer include KRAS, TP53, CDKN2A, and SMAD4[13-15]. In recent years, MUC16 and several other genes have also been identified as risk genes for malignancy and poor prognosis in pancreatic cancers. However, there remains a lack of research focusing on the coexistence of pancreatic cancer and other pancreatic tumors. Therefore, it is necessary to discuss the prognosis and intrinsic genomics in this subtype of pancreatic cancer, which could provide valuable insights for future clinical diagnosis and treatment.
With the continuous advancements in medical imaging technology, abdominal imaging has witnessed a growing application in patients with pancreatic tumors, especially when presenting with nonspecific symptoms[16], while the complexity of tumor coexistence may pose potential diagnostic challenges. Among the fundamental imaging modalities, computed tomography (CT) and ultrasound (US) are widely employed. For the evaluation of dilated common bile ducts or pancreatic duct (PD), magnetic resonance cholangiopancreatography (MRCP) has been established as the gold standard for several years[17]. Subsequently, endoscopic retrograde cholangiopancreatography and endoscopic US (EUS) are often selected as complementary tools for further assessment[18,19]. However, it is essential to acknowledge that histopathology remains the definitive gold standard for diagnosis, particularly in the context of rare diseases.
This study presented a rare case and aimed to identify critical gene mutations in the context of pancreatic dual tumors with spatial distance, through the technique of whole-genome sequencing. Meanwhile, we highlighted the challenges encountered in the differential diagnosis of intraductal papillary mucinous neoplasm (IPMN), PDAC, SCN, and chronic pancreatitis. Although these entities are known to exhibit distinct manifestations in various imaging modalities, it is important to recognize that the actual clinical scenario can be more complex and intriguing than anticipated.
CASE PRESENTATION
The coexistent pancreatic tumors were discovered during routine medical care, with continuous follow-up extending from March 2015 to May 2016 until death. Detailed information regarding clinical symptoms, physical examinations, and medical history was collected. Auxiliary examinations included laboratory, imaging, and pathological profiles, all of which were obtained from the medical record data system of Peking Union Medical College Hospital. Laboratory indicators included blood routine, liver and kidney function, and tumor markers. The serum normal range reference values applied in our study were reported as follows: (1) Carbohydrate antigen (CA) 19-9: 0-34 U/mL; (2) Carcinoembryonic antigen: 0-5 ng/mL; and (3) CA24-2: 0-20 U/mL. The imaging characteristics of this disease often resembled those of other conditions, complicating the differential diagnosis. Preoperative imaging studies included CT, positron emission tomography-CT (PET-CT), MRCP, and EUS. Image analysis was conducted and verified by two experienced radiologists through a retrospective consistent review. Pathological profiles included macroscopic specimens and pathological staining sections embedded in paraffin, examined and reported by two experienced pathologists. This study was approved by the institutional ethics review board, and informed consent was obtained.
Chief complaints
In March 2015, a 62-year-old male presented with right upper abdominal discomfort for one year.
History of present illness
He denied any concomitant symptoms or prior treatment.
History of past illness
Heavy tobacco and alcohol consumption, and cerebral infarction history were reported.
Physical examination upon admission
Nothing special was found in his physical examination.
Laboratory examinations
The genomic DNAs (gDNA) of the tumor tissue were extracted using the standard protocols with the QIAamp DNA FFPE Tissue Kit (Qiagen, 56404). DNA concentrations were determined by Qubit dsDNA HS Assay Kit (Invitrogen, Q32854). The gDNAs were fragmented to sizes ranging from 150 bp to 250 bp using the Covaris S220. A next-generation sequencing gDNA library was prepared for whole-genome sequencing (WGS) using 150-300 ng of fragmented gDNA. In brief, the library was prepared by KAPA Hyper Prep Kit (KAPA, KK8504) according to the manual. The concentrations of the library were determined using Qubit, and the size distributions of the library were analyzed using the Bioanalyzer 2100 (Agilent Technologies).
The pooled DNAs of the library were mixed with 1 μL of DNA blocker (Integrated DNA Technologies) and 5 μL of SureSelect QXT Fast Blocker Mix (Agilent) before being dried by a vacuum concentrator. The dried mixture was then dissolved in a 13 μL hybridization buffer, supplied by the hybridization of SureSelect QXT Reagent kit (Agilent). Sequencing of the whole-genome libraries was performed using 150bp paired-end runs on the Illumina HiSeq × 10 (Illumina). The 150 bp paired-end sequencing data were aligned to the human reference genome hg19 (GRCh37) by bwa v0.7.17[20], and variant calling analysis was performed using VarScan v2.3.8[21]. Calls were annotated using ANNOVAR v2015.06[22] and COSMIC v70[23]. To identify relevant gene targets for pancreatic cancer, a Hotspot Panel was selected based on common variants from pancreatic cancer samples through WGS. A Literature review was also conducted to help narrow down the gene list. And laboratory examinations, including CA19-9 (14.2 U/mL), carcinoembryonic antigen (4.78 ng/mL), and CA242 (6.4 U/mL) revealed no significant abnormalities.
Imaging examinations
MRCP revealed dilation of PD and common bile ducts, along with a cystic lesion at the tail. The high-resolution CT scan with three-dimensional reconstruction revealed atrophy in the pancreatic neck, body, and tail. It also documented a dilated PD and a low-density lesion at the tail, measuring 3.0 cm × 2.0 cm, with well-defined borders and multiple small cystic-like foci, some connected to the main PD, further supporting the diagnosis of IPMN (Figure 1). PET-CT confirmed pancreatic atrophy and identified a mass (1.7 cm × 2.0 cm) at the tail with increased radioactive uptake (Figure 2). In the pancreatic tail, EUS displayed a cystic PD dilation and an irregularly enhanced echo (2.5 cm × 1.8 cm) containing non-echoic dots. At the head, the PD was measured 0.6 cm in width, and a hyperechoic nodule was identified in the neck region of the PD (Figure 3). These findings collectively pointed towards a strong suspicion of IPMN.
Figure 1 High-resolution computed tomography with three-dimensional reconstruction of the patient.
Figure 2 Positron emission tomography-computed tomography of the patient.
Figure 3 Endoscopic ultrasound of the patient.
DISCUSSION
This article focused on the intrinsic genomic profiles, prognostic characteristics, and imaging diagnosis of coexistent pancreatic tumors. Our study presented a mass in the pancreatic tail with PD dilation in imaging examinations, while ultimately confirmed pathologically as PDAC in the head and SCN in the tail. Coexistent PDAC and SCN, spatially distant from each other, are exceedingly rare. Whole genome sequencing suggested that MUC16 is associated with dual tumor coexistence and poor prognosis.
The rarity and poor prognosis of coexistent pancreatic tumors
SCN is a relatively uncommon and benign neoplasm, representing approximately 25% of all pancreatic cystic tumors[7]. Its coexistence with other pancreatic neoplasms, such as ductal adenocarcinoma and endocrine tumors, is infrequent[5]. Recently, the World Health Organization updated the classification of SCN, introducing the category of mixed serous neuroendocrine neoplasms[24]. Some reported cases have documented the presence of SCN concomitant with endocrine tumors. However, our literature search revealed that only 6 cases of coexistent PDAC and SCN had been reported before our study[5,6,8-10], in which the dual tumors all had spatial distance between them. Therefore, the coexistence of PDAC and SCN is exceptionally rare, rendering prognosis analysis challenging.
PDAC is a highly aggressive carcinoma, and epidemiological projections indicate a rise in the incidence of pancreatic cancer in the coming decade[1,2]. The simultaneous occurrence of PDAC and SCN may confer a more unfavorable prognosis compared to either PDAC or SCN in isolation, given their presumed shared origins. As proposed by Nitta and Montag, aberrations in pancreatic exocrine cells may contribute to the development of both lesions[7,9]. Consequently, SCN coexisting with PDAC may possess a greater malignant potential, with an increased likelihood of evolving into serous cystadenocarcinoma. Furthermore, patients with dual neoplasms, especially two distinct lesions located in separate regions of the pancreas, may experience more extensive organ damage, resulting in a greater loss of normal pancreatic tissue and an enlarged surgical resection field. Consequently, the prognosis for the coexistence of PDAC and SCN is anticipated to be worse.
Our study raised two questions: What is the reason for the coexistence of PDAC and SCN, and what is the relationship between the two lesions? Precursor lesions of pancreatic cancer include pancreatic intraepithelial neoplasia (PanIN), IPMN, and mucinous cystic neoplasm. PanINs develop within the exocrine ducts of pancreatic lobules, which are categorized into PanIN-1, PanIN-2, and PanIN-3 based on the degree of dysplasia. Approximately 5%-10% of non-invasive cystic lesions may advance to invasive adenocarcinoma, referred to as intraductal papillary mucinous carcinoma, accounting for 10% of all the resected PDACs[25]. Differentiate from IPMN, mucinous cystic neoplasm lacks communication with the PD, carrying an estimated 10% risk of malignant transformation[26]. Additionally, a small subset of SCNs may progress into serous cystadenocarcinomas, posing a risk of invasive malignancy[27]. However, regardless of precancerous types, coexistent precursor and cancer lesions typically share an adjacent spatial location, a microscopically benign-malignant transition zone, and similar immunohistochemical characteristics[11]. Since our two lesions were located at opposite ends of the pancreas, we believed that they were geographically independent, with a much lower likelihood of sharing the same origin. One potential explanation could be that the SCN induced PD obstruction and chronic pancreatitis, thereby increasing the risk of PDAC occurrence. Andea et al[28] indicated that PanIN or PDAC occurred more frequently in the context of chronic pancreatitis, compared to normal pancreatic tissue. Another possibility is that this patient may harbor unidentified genetic predispositions to coexisting tumors. Therefore, we hoped that genome-wide sequencing could provide some clues.
The correlation between MUC16 with tumor coexistence and poor prognosis
The updated report from the American Cancer Society ranks pancreatic cancer as the third leading cause of cancer-related mortality, trailing behind lung cancer, prostate or breast cancer. In the United States, the mortality rate of pancreatic cancer has been increasing annually by 0.3%. Pancreatic cancer accounts for around 3% of newly diagnosed cancers, ranking 10th among males and 7th among females[12].
From an evolutionary perspective, the development of pancreatic cancer can be divided into three stages: (1) Initiation; (2) Clonal expansion; and (3) Invasive dissemination[29]. In the initiation stage, induced by external exposures or DNA repair deficiencies, driver gene mutations occur in normal pancreatic cells. Subsequently, during the clonal expansion phase, mutation-carrying cells proliferate into tumor clones, which could be described in a stepwise progression model[30] or a punctuated evolution model[31]. In the invasive dissemination stage, it is crucial to emphasize interactions between tumor cells and their surrounding microenvironment, including factors like desmoplastic stroma, oxygen concentrations, and immunological landscape[29].
Traditionally, the widely recognized driver genes for pancreatic cancer include KRAS, CDKN2A, TP53, and SMAD4. Under the influence of these genes, normal pancreatic cells develop into PanIN, eventually progressing to invasive cancer[13,14,32]. Recent genomic analyses[30,33] have revealed that KRAS mutations are present in over 99% of PanIN-1 Lesions, CDKN2A inactivation can be detected as early as PanIN-2, while TP53 and SMAD4 mutations occur in PanIN-3. In the human genome of PDACs, KRAS mutation is the most common (90%-95%), while CDKN2A, TP53, and SMAD4 are inactivated in 50%-80% of cases[15,34]. These four key driver genes act synergistically, forming a complex network that promotes tumorigenesis[29].
With the continuous iteration of genomic sequencing technology, the research scope has expanded to include point mutations, gene copy number variations, and chromosome structure variations. Consequently, several genes or pathways beyond the core genes have been newly discovered, such as GATA6[31], KMT2C[15], SF3B1[15,35], and SLIT/ROBO axon signal pathway[35]. Moreover, germline mutations associated with DNA stability, including BRCA1/2 and PALB2, are also involved[36].
Our whole genome sequencing also exhibited these four gene mutations, which played a driving role in carcinoma formation. Notably, a special gene mutation, MUC16, was also detected in our patient. In 2017, Balachandran et al[37] reported the unique expression of MUC16 in PDAC with heterogeneous immunology, sparking widespread discussion and attention. MUC16, situated on chromosome 19p13.2, spans 374k base pairs and contains 92 exons. It encodes a high-molecular-weight mucin-like glycoprotein, commonly found in the epithelial cells of adult body cavities, also known as CA125, comprising a large number of O-glycosylated tandem repeat domains and extension structures.
It was demonstrated that MUC16 was the third most common mutated gene, following KRAS and TP53, in two independent PDAC cohorts[37]. High levels of CA125 were observed in 50%-70% of PDAC patients[38]. Numerous studies have shown that elevated expression and abnormal O-glycosylation of MUC16 were independent risk factors for tumor progression, increased metastasis, and poor survival in PDAC patients[38-40]. However, to date, there have been no reports or articles focused on the correlation between genetics and the coexistence of pancreatic tumors. In our study, MUC16 was pronouncedly highlighted through genome-wide sequencing. In consistency with the previous studies, we believe that MUC16 played an important role in the poor prognosis of pancreatic cancer. Since the coexistence of pancreatic tumors is exceedingly rare, only one set of sequencing data was available in this study. Moreover, PDAC inherently carries an extremely poor prognosis. Therefore, we cannot exclude the possibility that the short survival observed in this case may have been influenced by other factors, such as resistance to radiotherapy or chemotherapy. Nevertheless, based on our findings and a review of the existing literature, we propose that MUC16 may be associated with both the coexistence and poor prognosis of dual or multiple pancreatic tumors. Further studies with larger sample sizes are warranted to validate this potential association.
In addition to the common features of pancreatic cancer, inter-individual heterogeneity has become a hot topic in recent research. Heterogeneity implies differences in etiology, clinical manifestations, treatment responses, and prognosis. Stratification or grouping based on this heterogeneity aids in personalized diagnosis and treatment of pancreatic cancer. Collisson et al[41] were the first to report genomic variations among PDAC patients, categorizing them into classical, quasi-mesenchymal, and exocrine-like subtypes. Subsequently, Moffitt et al[42] identified two subtypes, classical and basal-like; while Waddell et al[15] determined four subtypes: Stable, locally rearranged, scattered, and unstable. Lately, through genomic sequencing, Bailey et al[43] classified PDACs into four characteristic subgroups: Squamous, pancreatic progenitor, immunogenic, and aberrantly differentiated endocrine-exocrine. There exists correspondence or overlap among all these reported classifications.
Our study focused on an important clinical subtype of pancreatic cancer: Coexistent PDAC and IPMN. There presents important reference value in sharing clinical features, imaging findings, pathological profiles, and prognosis of this specific subtype. Furthermore, exploring the correlation between these features and the intrinsic genomic information holds indicative significance.
Challenges in imaging identification and differential diagnosis
Imaging examinations are crucial for diagnosing pancreatic masses, as pancreatic diseases often remain asymptomatic until tissue damage and malignancies have advanced or metastasized. Contrast-enhanced multidetector CT and PET-CT play central roles in diagnosing, staging, and treatment planning for PDAC patients[44]. Besides, it is reported that MRI could better detect cystic lesions, while EUS could be more accurate for small solid lesions and convenient for fine needle aspiration[3,10].
The uniqueness of our study lies in its distinctive imaging presentation, which posed challenges in differential diagnosis: (1) The radiological findings of the pancreatic tail mass and PD dilation strongly suggested IPMN, contradicting the final pathology of SCN; and (2) The mass in the pancreatic head displayed iso-density and iso-enhancement with surrounding normal tissue, complicating the identification of PDAC and differential diagnosis of PD dilation.
IPMN originates from the epithelium of the main PD or its branches and often presents variable duct dilation. Neoplastic cells in IPMN typically exhibit a papillary morphology, although flat epithelium can also be observed[45]. Diagnostic modalities such as MRCP, CT, and EUS are commonly employed to evaluate IPMN. Key indicators of IPMN include a dilated PD with a connected lesion, a patulous or “fish-eye” papilla, and mucin presented at the papillae[45]. Furthermore, analyzing protein expressions in various cyst fluids may also contribute to diagnosis[46].
In this study, determining the etiology of PD dilation was a crucial aspect of differential diagnosis. According to our imaging findings, IPMN appeared to be the most plausible etiology. However, the possibility of chronic pancreatitis or a mass at the pancreatic head should also be considered, as both conditions could lead to PD dilation. Simultaneously, it is important to emphasize that the ultimate diagnosis confirmation relies on pathological examination. Additionally, in patients with IPMN, the coexistence of distinct pancreatic cancer is possible, particularly when early-stage cancer may not be readily detectable by CT, MRI, or EUS[47]. In certain instances, concurrent chronic pancreatitis and IPMN may also occur[16]. Consequently, it is essential to consider all possibilities, including IPMN, chronic pancreatitis, and pancreatic mass at the head, when arriving at a comprehensive diagnosis. Given the challenging differential diagnosis of coexistent pancreatic tumors, it is imperative to improve the accuracy of preoperative imaging examinations. Nevertheless, there is limited evidence regarding the accuracy of imaging screening, highlighting a research gap that needs to be filled.
Our article indicated the association between MUC16 mutation and poor prognosis, along with coexistent pancreatic tumors, paving a new avenue for future studies. There have been several studies exploring the application of this correlation in clinical settings. MUC16 fluorescent probes were created to aid in lesion marking during surgery[48], while immune imaging probes were expected to improve early recognition of pancreatic cancer in imaging examinations[49]. Additionally, anti-MUC16 nanoantibodies might serve as a novel immunotherapeutic agent in PDAC patients[38].
There are some limitations in this study. The sample size of coexistent PDAC and SCN was limited, and the evidence strength was not sufficient enough. However, we presented the seventh case worldwide of coexistent pancreatic tumors with spatially distant locations. The clinical and genomic profiles are extremely rare and valuable, especially in this subtype of pancreatic cancer. Consistent with the results from Balachandran et al[37], our findings hold considerable credibility and suggestive significance. Future studies with larger samples are required for further validation. Additionally, MRCP imaging was not provided as a figure, due to examinations being conducted at a local hospital instead of our center, and no electronic archives were retained.
