Exosomal circular RNA hsa_circ_0006220, and hsa_circ_0001666 as biomarkers in the diagnosis of pancreatic cancer

Abstract Background Pancreatic cancer is a highly malignant tumor of the digestive system. Objective Exosomal circular RNA can be used as a biomarker for the early diagnosis of pancreatic cancer. Methods The expression of various differentially expressed circRNAs in pancreatic cancer tissues was analyzed by gene chip, exosome expression was verified by electron microscopy and Western blotting, and the expression of exosomal circRNA in pancreatic cancer cells, tissues, and plasma were determined by quantitative reverse transcription‐polymerase chain reaction (qRT‐PCR). Results Compared with healthy controls, hsa_circ_0006220 and hsa_circ_0001666 were highly expressed in exosomes in the plasma of pancreatic cancer patients. The AUC values were 0.7817 for hsa_circ_0006220, 0.8062 for hsa_circ_0001666, and 0.884 for the combined diagnosis. In addition, clinicopathological features revealed that the expression of hsa_circ_0006220 in plasma exosomes from pancreatic cancer patients was associated with CA19‐9 levels (p = 0.0001) and lymph node metastasis (p = 0.0005). The expression of hsa_circ_0001666 was correlated with both tumor size (p = 0.0157) and CA19‐9 level (p = 0.0001). Conclusions The high expression of exosomal hsa_circ_0001666 and hsa_circ_0006220 suggests that these can be used as new biomarkers for the diagnosis and treatment of pancreatic cancer.

prognoses of malignant tumors with an overall 5-year survival rate of only about 8%. 2,3 Therefore, the key to improving the survival rate of pancreatic cancer lies in the early diagnosis and early treatment.
The search for relevant molecular markers for the detection of highrisk pancreatic cancer will help patients achieve early detection and timely intervention and treatment.
Exosomes are membrane vesicles with diameters of 40-100 nm that originate from endocytosis. They are secreted by various cells and contain large amounts of nucleic acids (circRNA, lncRNA, miRNA, mRNA, and DNA), proteins, enzymes, and other substances. They are usually extracellular and play important roles in communication. Since exosomes released by cells into the body fluids show different protein and RNA contents in healthy subjects and patients with different diseases, they have the potential for being used as diagnostic markers. 4 Tumor-derived exosomes are rich in non-coding RNAs that may be used as tumor markers. 5 For example, in the plasma of gastric cancer patients, the expression of exosomes circ-KIAA1244 was found to be correlated with the TNM staging, lymph node metastasis, and overall survival time of the patients, suggesting their use as a new circulating biomarker for gastric cancer. 6 Therefore, this method could act like a "liquid biopsy" as a marker for the early diagnosis of pancreatic cancer.
Circular RNA (circRNA) is an RNA with a continuous covalent closed-loop structure. CircRNAs can be divided into non-coding and coding circRNAs according to the ability to encode protein. 7 With the continuous development of sequencing technologies, circRNAs have been found to be both species-specific and tissue-specific 8,9 and are widely expressed in the cytoplasm and nucleus. In addition, they play important regulatory roles in the occurrence and development of many diseases (especially cancer). Since circRNA has a stable circular structure and no unstable 3' and 5' ends, the molecule is not easily degraded by nucleases, making it an ideal biomarker for disease detection. 10 These findings also suggest that circRNA may be an ideal diagnostic marker for pancreatic cancer.
However, there are few studies on the diagnostic value of exosomal circRNAs for pancreatic cancer. In this study, we screened five expressed circRNA molecules through the GEO database, collected clinical plasma samples, and verified the expression levels of these molecules in exosomes. The experimental results showed that hsa_circ_0001666 and hsa_circ_0006220 are located in the plasma of pancreatic cancer patients. The expression of these circRNAs in the body and their diagnostic potential were analyzed by receiver operating characteristic (ROC) curves. The findings of this study will assist the development of screening tests using plasma exosomal cir-cRNA for the diagnosis of patients with pancreatic cancer.

| Clinical samples
In this study, from June 2018 to June 2020, 30 cases of pancreatic cancer tissues and their corresponding paired adjacent tissues were collected in the East District Hospital of Ningbo Li Huili Hospital, China. At the same time, the plasma of 62 pancreatic cancer patients and 62 healthy subjects were collected. This study was approved by the hospital ethics committee. All patients provided written informed consent.

| Exclusion criteria
Patients with pancreatic ductal adenocarcinoma undergoing preoperative neoadjuvant chemotherapy or radiotherapy; patients with benign or borderline malignant tumors of the pancreas; patients with pancreatic metastases (including kidney cancer and pancreatic cancer that had undergone metastasis to the pancreas); serous patients with cystadenocarcinoma or mucinous cystadenocarcinoma; patients with chronic pancreatitis that had experienced acute pancreatitis within the previous three months; patients with chronic pancreatitis with a tendency to become malignant within six months of follow-up, and patients with blood specimens showing hemolysis of grade 5 or above.

| Cell culture
Human embryonic kidney HEK 293Tcells, human pancreatic duct cells hTERT-HPNE, and two human pancreatic cancer cell lines Panc2, MIA PaCa-2 were cultured in Dulbecco's modified Eagle medium (DMEM) containing 10% fetal bovine serum in a cell culture incubator containing 5% CO 2 at 37°C. The medium was changed every 36 h based on observed alterations in color and the growth state of the cells. Cells were passaged when confluent.

| Isolation of exosome
The exosomes were isolated using the Total Exosome Isolation Kit

| Electron microscopy
Freshly separated exosomal droplets (20 μl) were fixed on a copper network for 5 min for electron microscopy. The sample was stained with a 2% phosphotungstic acid solution for 5 min at room temperature. Dry filter paper was used to remove the excess liquid, and the sample was then baked under a 60-W incandescent lamp for 3-5 min. The image was observed using a transmission electron microscope (Tecnai Spirit G2 BioTWIN, FEI).

| Western blotting
The total protein was collected through lysis with radioimmunoprecipitation assay (RIPA) buffer supplemented with 1:100 protease inhibitors (Life Technologies, USA) and phosphatase inhibitor cocktail I and II (Sigma-Aldrich, USA). The protein was quantified using a Bicinchoninic Acid Protein assay kit (Life Technologies) and separated on SDS polyacrylamide gels. The separated proteins were

| RNA extraction and qRT-PCR
The RNA was extracted from the tissue samples using TRIzol (Invitrogen, USA), following the manufacturer's instructions. TRIzol LS (Invitrogen) was used to isolate total RNA from the plasma. A Prime Script TM II cDNA Synthesis Kit (TaKaRa, Japan) was employed to reverse-transcribe 1 µg total RNA to cDNA. qRT-PCR (Roche, Switzerland) combined with SYBR ® GreenⅠmethod (Roche, Switzerland) was used for amplification. The reaction conditions used were as follows: pre-denaturation at 95°C for 30 s, PCR reaction at 95°C for 5 s, 60°C for 20 s, for 40 cycles, and annealing at 50°C for 30 s. The 2 −△△ct method was applied to determine the relative circRNA expression. SPSS 23.0 and GraphPad 8.0 were used to perform statistical analysis on all the experimental data (each experiment was repeated at least three times). In addition, the count data used cases (percentage) to indicate the potential expression of circRNA (2 −Δ△ct value) in pancreatic cancer and corresponding adjacent tissues, and paired t-tests were used to determine differences. The expression of cir-cRNA in the blood of pancreatic cancer patients and healthy subjects was determined by non-parametric tests. Fisher's exact test and Spearman correlation coefficient were used to analyze the correlation between circRNA expression and clinicopathological parameters. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficiency of circRNA in pancreatic cancer samples. The Youden index was used to calculate the critical value, and the Z test was used to compare the area under the curve (AUC).

| Statistical analysis
p < 0.05 was considered statistically significant.
The results showed that hsa_circ_0006220 and hsa_circ_0001666 could effectively distinguish between the normal control group and pancreatic cancer patients. Among them, hsa_circ_0001666 had the highest AUC value, suggesting a good diagnostic value, while hsa_ circ_0006220 had a somewhat lower diagnostic value.
Evidence shows that a combination of tumor markers can improve the accuracy of diagnosis. In this study, we combined hsa_ circ_0006220 and hsa_circ_0001666. The results showed that in distinguishing normal people from pancreatic cancer patients, The AUC value of the combination was 0.884 (95% CI = 82.7% to 94.1%, with a sensitivity = 0.742% and a specificity = 0.871%, Figure 3F), which was significantly higher than that of a single molecule. This indicated that the combination of hsa_circ_0006220 and hsa_ circ_0001666 had improved diagnostic value and high accuracy.
The above data showed that the expression of hsa_circ_0006220 and hsa_circ_0001666 in plasma exosomes samples of pancreatic cancer was significantly upregulated. Therefore, we analyzed the association between hsa_circ_0006220 and hsa_circ_0001666 and clinicopathological factors in patients with pancreatic cancer. As shown in Table 1, the expression of hsa_circ_0006220 in plasma exosomes from pancreatic cancer patients was closely related to CA19-9 (p = 0.0001) and lymph node metastasis (p = 0.0005).
However, there was no significant correlation between the expression of hsa_circ_0006220 and other clinicopathological factors, including age (p = 0.0730), sex (p = 0.3074), tumor size (p = 0.4208), clinical stage (p = 0.0384), and degree of differentiation (p = 0.0684). As shown in Table 2

| Stability of hsa_circ_0006220 and hsa_ circ_0001666
Generally, storage temperature, freeze-thaw cycles, and time are the main problems encountered in clinical blood sample testing.
Therefore, in order to study the stability of hsa_circ_0006220 and hsa_circ_0001666 in plasma exosomes, fresh plasma samples from three patients with pancreatic cancer were divided into equal parts and stored at different temperatures (25, 4, −20, and −80°C), stored for 24 h, or exposed to different incubation times (4, 8, 16, and 32 h)

F I G U R E 3
CircRNA expression in plasma exosomes. qRT-PCR analysis of (A) circ_0000977, (B) hsa_circ_0006220, and (C) hsa_circ_0001666 in the plasma exudates of 62 patients with pancreatic cancer and healthy individuals; (D) diagnostic value of hsa_ circ_0006220 in patients with pancreatic cancer; (E) diagnostic value of hsa_circ_0001666 in patients with pancreatic cancer; (F) diagnostic value of the combination of hsa_circ_0006220 and hsa_circ_0001666 in patients with pancreatic cancer at room temperature, or treated with 1, 2, 3, or 4 repeated freezethaw cycles. The results showed that the relative expression levels of hsa_circ_0006220 and hsa_circ_0001666 were not statistically different at different temperatures ( Figure 4A, B), different incubation times ( Figure 4C, D), and after repeated freeze-thaw cycles ( Figure 4E, F). Therefore, the stability of hsa_circ_0006220 and hsa_circ_0001666 indicates that they may become new diagnostic markers for pancreatic cancer.

| DISCUSS ION
Pancreatic cancer is one of the most malignant tumors of the digestive system. Although the prognosis has improved slightly recently, the 5-year survival rate is still less than 8%. Its incidence rate is still on the rise all over the world, and the case fatality rate is expected to rank second among all malignant tumors in Western countries in 2030. 11,12 At the time of diagnosis, about 80-85% of patients have unresectable or metastatic pancreatic cancer. 13 Even if a small percentage of patients are diagnosed with locally resectable tumors, the prognosis is still very poor, and only about 20% of patients survive five years after surgery. In the past ten years, advances in the diagnosis methods, perioperative management, radiotherapy technology, and systemic therapy for advanced pancreatic cancer patients have made some progress, but the treatment of pancreatic cancer is still very limited. There is an urgent need to find new methods for screening high-risk patients to detect pancreatic tumors at an early stage, which will have far-reaching significance for the clinical treatment of pancreatic cancer.
At present, the most commonly used method for preoperative diagnosis of pancreatic cancer is imaging examination combined with the tumor markers CEA and CA19-9. 1,14,15 Commonly used clinical imaging methods for pancreatic cancer include ultrasound, CT, MRI, and endoscopic ultrasound. All these methods have difficulty in diagnosing early pancreatic cancer with tumor diameters of <1 cm.
The specificity of CA19-9 in pancreatic cancer screening is about 70%, but this improvement is not obvious in early-stage patients.
At the same time, about 5-10% of pancreatic cancer patients do not express CA19-9 and for CEA, the average specificity is 79%, and the average sensitivity is 54%. 16 Studies have shown significantly higher activity of class III alcohol dehydrogenase (ADH) in pancreatic cancer tissue and plasma in comparison with healthy tissue and serum. In addition, the assessment of the diagnostic ability of ADH III showed a diagnostic sensitivity and specificity of 70% and 76%, respectively, TA B L E 1 Clinicopathological factors of plasma exosomes samples of patients with pancreatic cancer and expression of hsa_circ_0006220  24 In addition. exosomal circ-PDE8A derived from pancreatic cancer cells was found to be able to enter the blood circulation and was related to the progression and prognosis of F I G U R E 4 Stability of hsa_circ_0001666 and hsa_circ_0006220. Expression of hsa_circ_0001666 and hsa_circ_0006220 was detected by qRT-PCR at different storage temperatures (A-B), (C-D) repeated freeze-thaw cycles, and storage times (E-F) pancreatic cancer. 22 In our study, the exosomal hsa_circ_0001666 and hsa_circ_0006220 performed well in distinguishing pancreatic cancer patients from healthy subjects, with a higher diagnostic efficacy when used in combination. In addition, hsa_circ_0001666 and hsa_circ_0006220 also have value in guiding the diagnosis of pancreatic cancer and helping to determine the tumor stage.
However, the independent and combined diagnostic value of hsa_ circ_0001666 and hsa_circ_0006220 require further optimization using clinical large-scale cohort, multi-sample, prospective, and multi-center studies.
At the same time, exosomal circRNAs show good diagnostic performance, not only because of their high specificity and sensitivity but also because of their structure and stability in the blood. 22 We observed through RNase R and ActD that the expression of exosomal circRNA was not significantly affected, and neither long-term storage at room temperature nor repeated freezing and thawing altered the expression level of the exosomal

circRNAs.
Recently, several studies have reported that exosomal cir-cRNA mediates the progression of pancreatic cancer through various mechanisms such as promoting the invasive growth of cells, regulating the permeability of the endothelial monolayer, promoting metastasis, and enhancing glycolysis to promote drug resistance. 22,[25][26][27] In this study, the exosomes hsa_circ_0001666 and hsa_circ_0006220 were found to be effective for pancreatic cancer screening and were shown to be related to the cancer stage, suggesting that they may be involved in the occurrence and development of pancreatic cancer. This mechanism remains to be further explored.

CO N FLI C T O F I NTE R E S T
All authors declare no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
LH, LX, and YCZ contributed to the experimental design. LFJ, LX, WWT, and XDQ contributed to clinical diagnosis and sample collection. LH, KWX, and LFJ contributed to testing. JHW and LH contributed to the data analysis, result interpretation, and writing. All authors have reviewed and approved the final study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used in the current study are available from the corresponding author upon reasonable request.