Peripheral blood microRNA biomarkers for early diagnosis of pancreatic cancer and their applications

Abstract

The present invention relates to peripheral blood microRNA biomarkers for the early diagnosis of pancreatic cancer and their applications, belonging to the field of medical molecular biology. Four novel blood microRNA biomarkers are provided for the early diagnosis of pancreatic cancer. These markers are used non-invasively for early diagnosis of pancreatic cancer, providing a risk warning three years before a patient is diagnosed. They are highly accurate, with high sensitivity and specificity for pancreatic cancer.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical molecular biology, and more specifically, relates to peripheral blood microRNA biomarkers for early diagnosis of pancreatic cancer and applications thereof, and in particular to four peripheral blood microRNA biomarkers for early diagnosis of pancreatic cancer. Background Art

[0002] Pancreatic cancer is the third leading cause of death among common malignant tumors. As one of the most lethal malignancies, the five-year relative survival rate for pancreatic cancer patients is only approximately 10%. Because pancreatic cancer lacks specific symptoms in its early stages, approximately 80% of patients already have localized lesions or distant metastases at the time of diagnosis. Compared with patients in the late stages, patients diagnosed early can improve their survival by approximately 30%. Furthermore, pancreatic cancer evolves slowly, with an average of 11.7 years between onset and tumor invasion, providing a long window for early detection (S. Yachida, S. Jones, I. Bozic, et al. Distant metastasis occurs late during the genetic evolution of pancreatic cancer. Nature, 2010, 467(7319): 1114-1117). Therefore, early screening and intervention are crucial for improving the survival rate of pancreatic cancer patients.

[0003] MicroRNAs are a class of noncoding, single-stranded RNA molecules approximately 22 nucleotides in length. They can interact with their specific target genes and serve as important post-transcriptional regulatory factors. During normal human physiology, microRNAs are released from cells into the peripheral blood, acting as signaling molecules to regulate intercellular communication. Therefore, abnormal microRNA secretion may indicate the onset and progression of diseases, including tumorigenesis. Furthermore, microRNAs are stably present in the blood in a form resistant to RNase activity, further highlighting their great potential as disease biomarkers.

[0004] Over the past few decades, numerous studies have reported circulating microRNA markers associated with pancreatic cancer. However, most studies have been limited to a few to dozens of microRNAs, representing only a small fraction of the approximately 2,600 mature microRNAs recorded in miRBase. Only a few studies have systematically quantified microRNA expression using microRNA assays, microarrays, or sequencing. Existing microRNA markers can significantly improve the predictive accuracy of pancreatic cancer compared to CA19-9. However, these findings have not yet been validated in prospective studies. To date, only miR-10b, miR-21-5p, miR-30c, and miR-106b have been validated in a nested case-control study based on a European cohort, and their role in predicting pancreatic cancer risk was only evident within 2 years before diagnosis (EJ Duell, L. Lujan-Barroso, N. Sala, et al. Plasma micrornas as biomarkersof pancreatic cancer risk in a prospective cohort study. International Journal of Cancer, 2017, 141(5): 905-915). Summary of the Invention

[0005] To address the aforementioned issues, the present invention addresses the existing lack of microRNA biomarkers for the early diagnosis of pancreatic cancer. Based on a prospective cohort, four new blood microRNA biomarkers are provided for the early diagnosis of pancreatic cancer. These markers are used non-invasively for the early diagnosis of pancreatic cancer, potentially providing early warning of pancreatic cancer risk up to six years before a patient is diagnosed.

[0006] According to a first aspect of the present invention, there is provided a microRNA biomarker for diagnosing pancreatic cancer, characterized in that the microRNA biomarker is at least one of miR-139-3p, miR-92b-5p, miR-1976 and miR-1908-5p;

[0007] The nucleotide sequence of the miR-139-3p is shown in SEQ ID NO: 1, the nucleotide sequence of the miR-92b-5p is shown in SEQ ID NO: 2, the nucleotide sequence of the miR-1976 is shown in SEQ ID NO: 3, and the nucleotide sequence of the miR-1908-5p is shown in SEQ ID NO: 4.

[0008] According to another aspect of the present invention, a kit for diagnosing pancreatic cancer is provided, wherein the kit comprises the microRNA biomarker.

[0009] According to another aspect of the present invention, there is provided a use of the microRNA biomarker in preparing a diagnostic reagent for pancreatic cancer.

[0010] In general, the above technical solutions conceived by the present invention have the following technical advantages compared with the existing technology:

[0011] (1) The present invention is based on human blood samples. The detection method is safe and non-invasive, and is highly accepted among asymptomatic people, which is conducive to the early diagnosis of pancreatic cancer.

[0012] (2) Based on a prospective cohort sample with an average follow-up of approximately 5.7 years, this study identified and validated four blood microRNA biomarkers for the early diagnosis of pancreatic cancer. These biomarkers can provide early warning of pancreatic cancer risk three years before the diagnosis, offering the advantage of a long prediction window.

[0013] (3) The microRNA biomarkers of the present invention are highly accurate and have high sensitivity and specificity for pancreatic cancer. The AUC of the combined four microRNAs of the present invention can reach 0.847, with a sensitivity and specificity of 82.8% and 72.0%, respectively. The microRNA biomarkers of the present invention can be combined with other clinical diagnostic methods to provide a more accurate method for the early diagnosis of pancreatic cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 These are the expression levels of the four biomarkers at different early stages of diagnosis.

[0015] Figure 2 are the expression levels of the four biomarkers in different tumor stages.

[0016] Figure 3 Figure 3 is the ROC analysis of four biomarkers and their combination for differentiating pancreatic cancer from healthy controls. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0018] The purpose of this study was to identify and validate blood microRNA biomarkers associated with incident pancreatic cancer in a prospective cohort. The blood microRNA biomarkers included miR-139-3p (MIMAT0004552), miR-92b-5p (MIMAT0004792), miR-1976 (MIMAT0009451), and miR-1908-5p (MIMAT0007881).

[0019] The nucleotide sequence of the miR-139-3p is shown in SEQ ID NO: 1 (UGGAGACGCGGCCCUGUUGGAGU), the nucleotide sequence of the miR-92b-5p is shown in SEQ ID NO: 2 (AGGGACGGGACGCGGUGCAGUG), the nucleotide sequence of the miR-1976 is shown in SEQ ID NO: 3 (CCUCCUGCCCUCCUUGCUGU), and the nucleotide sequence of the miR-1908-5p is shown in SEQ ID NO: 4 (CGGCGGGGACGGCGAUUGGUC).

[0020] The blood expression levels of miR-139-3p, miR-92b-5p, miR-1976 and miR-1908-5p of the present invention are positively correlated with the risk of developing new pancreatic cancer in the general population.

[0021] According to a specific embodiment of the present invention, the reagent materials and / or instrument equipment for detecting the level of microRNA in a sample from an individual to be tested include the reagent materials and / or instrument equipment used in the following methods:

[0022] microRNA second-generation sequencing detection method and microRNA chip detection method.

[0023] Furthermore, the present invention evaluates the application of the blood microRNA biomarker in the early diagnosis of pancreatic cancer.

[0024] Preferably, the differential blood expression of miR-139-3p, miR-92b-5p, miR-1976 and miR-1908-5p in patients with newly diagnosed pancreatic cancer can be detected three to six years before the diagnosis of pancreatic cancer.

[0025] Preferably, the differential blood expression of miR-139-3p, miR-92b-5p, miR-1976 and miR-1908-5p can be found in pancreatic cancer patients with tumor classification stage III and below.

[0026] Preferably, the combination marker of the four microRNAs can effectively predict pancreatic cancer patients in the validation data set, and the model AUC is 0.847.

[0027] The following are specific embodiments

[0028] Example 1: Collection of research subjects and data generation

[0029] This study is based on the baseline data of the Dongfeng-Tongji (DFTJ) cohort collected in 2013, with an average follow-up time of 5.7 years. Each case was matched with a control according to the same gender, the same enrollment hospital, age ±1.2 years and blood collection date difference ±7 days. A nested case-control study of 44 pairs of newly diagnosed pancreatic cancer was designed and established.

[0030] The plasma samples used in this study were fasting blood samples collected from the DFTJ cohort in 2013. Plasma, serum, and clots were immediately centrifuged and separated and aliquoted. The samples were stored at -80°C and entered into an electronic database. Approximately 500 μL of plasma was first extracted using TRIzol® LS lysis buffer pretreatment combined with the miRNeasy Serum / Plasma Kit (catalog no. 217184) for total RNA extraction. Then, using 5 μL of plasma total RNA as the starting RNA template, plasma microRNA libraries were constructed using the QIAseq microRNA Library Kit (catalog no. 331502, QIAGEN). The experimental procedures were strictly followed as described in the QIAseq microRNA Library Kit Handbook. Finally, small RNA sequencing (single-end sequencing, 75 bp) was performed on the Illumina NextSeq500 platform.

[0031] The raw sequencing data were quality-controlled using FastQC software (version 0.11.8) to confirm the absence of abnormal samples and that all samples had more than 9 million non-repetitive sequence reads for subsequent sequence alignment and quantitative analysis. Low-quality reads (quality score less than 20) were removed, and adapter sequences were removed using miRge 3.0 software. Sequences were annotated with unique molecular identifiers (UMIs) to avoid duplicate counts or bias in subsequent data due to repetitive DNA fragments generated during PCR amplification. Sequence alignment of mature microRNAs was performed using the miRBase database (version 22, http: / / www.mirbase.org / ), and microRNA quantification was performed using the default settings in miRge 3.0. Expression levels of 1214 expressed microRNAs were obtained. Because the alignment ratio (aligned reads / total reads) of five samples was less than 40%, these samples and their corresponding matched pairs were removed, resulting in a total of four sample pairs. Finally, 40 pairs of newly diagnosed pancreatic cancer case-control samples were retained for downstream analysis.

[0032] This study describes and compares baseline differences in basic characteristics, lifestyle, and glycemic profiles between newly diagnosed pancreatic cancer cases and healthy controls, including age, sex, body mass index (BMI), waist-to-height ratio, smoking status, alcohol consumption, fasting blood glucose, and HbA1c. As shown in Table 1, the mean (standard deviation) age of newly diagnosed pancreatic cancer cases was 69.38 (8.02) years, and 42% were male. Baseline differences primarily occurred in glycemic profiles, including fasting blood glucose, glycated hemoglobin, and type 2 diabetes, all of which were significantly elevated in the newly diagnosed pancreatic cancer group.

[0033] Table 1 Baseline information and comparison of new pancreatic cancer cases and matched controls

[0034]

[0035] Note: Type 2 diabetes is defined as a self-reported history of type 2 diabetes, use of glucose-lowering medications or insulin, fasting blood glucose ≥ 7.0 mmol / L, or HbAlc ≥ 6.5%. Continuous variables are expressed as mean ± standard deviation, and categorical variables are expressed as number (percentage). P Variables are tested for differences between the new case group and the matched control group. Continuous variables were analyzed using paired t-tests, and categorical variables were analyzed using Cochran-Mantel-Haenszel tests.

[0036] Example 2: Screening and validation of pancreatic cancer-related peripheral blood microRNA markers

[0037] (1) Identification of peripheral blood microRNA markers associated with new pancreatic cancer

[0038] First, the ComBat model based on negative binomial regression was used to correct for batch effects. Then, 746 microRNAs expressed in less than 25% of the samples (counts per million, CPM>1) were removed, and a total of 771 microRNAs were included for subsequent analysis. The DESeq2 method was used to identify differentially expressed microRNAs in new pancreatic cancer samples and control samples. The model was adjusted for age, sex, admission hospital, and type 2 diabetes. The differences between the groups were quantified using the log2-transformed fold change (log2FC) indicator. The Benjamini-Hochberg method was used to calculate the FDR to correct for P The significance level was defined as FDR < 0.05.

[0039] As shown in Table 2 , a total of 21 microRNAs were associated with new pancreatic cancer, including miR-6852-5p, miR-6764-5p, miR-769-3p, miR-139-3p, miR-1270, miR-92b-5p, miR-486-5p, miR-1180-3p, miR-106b-3p, miR-1976, and miR- R-12136, miR-1908-5p, miR-501-3p, miR-5187-5p, miR-3138, and miR-1306-3p were positively correlated with the risk of incident pancreatic cancer, while miR-141-3p, miR-30e-5p, miR-338-3p, miR-199b-5p, and miR-128-3p were negatively correlated with the risk of incident pancreatic cancer.

[0040] Table 2 Preliminary screening of 21 microRNA biomarkers associated with new pancreatic cancer

[0041]

[0042] (2) Verification of peripheral blood microRNA markers associated with new pancreatic cancer

[0043] The 21 biomarkers screened in (1) were validated based on data from a Japanese clinical study. The data included 100 pancreatic cancer patients admitted to the National Cancer Center East Hospital in Japan between 2010 and 2012. All patients were histologically confirmed to have adenocarcinoma by pathological diagnosis. In addition, 150 healthy individuals recruited from Toray Industries Inc.'s Japanese subsidiary in 2013 were included as controls. The inclusion criteria for healthy control individuals were age over 60 years, no history of cancer, and no hospitalization in the past 3 months. 2555 mature microRNAs were measured in the serum samples of each participant using the Human microRNA Oligo Chip (Toray Industries, Inc.). The present invention removed one pancreatic cancer patient with missing demographic information, and a total of 99 pancreatic cancer patients and 150 healthy controls were included for subsequent validation analysis.

[0044] Among the 21 candidate biomarkers, a total of 9 microRNAs, miR-30e-5p, miR-769-3p, miR-139-3p, miR-92b-5p, miR-486-5p, miR-106b-3p, miR-1976, miR-1908-5p, and miR-3138, were expressed in more than 25% of the samples (signal values ​​greater than 2 6 To avoid the influence of extreme values, this study performed a 90% winsorization process on each microRNA based on the standardized microRNA matrix, replacing the expression values ​​greater than the 95th percentile with the 95th percentile value, and replacing the signal values ​​less than the 5th percentile with the 5th percentile value. Then, for each microRNA, Z The data were converted and standardized. A logistic regression model was used to adjust for gender and age at diagnosis. The odds ratio (OR) and its 95% confidence interval (CI) were reported to reflect the effect size of microRNA on pancreatic cancer. The significance level was adjusted by the Bonferroni method. P The value is less than 0.05.

[0045] As shown in Table 3, miR-139-3p, miR-92b-5p, miR-1976, and miR-1908-5p had the same effect direction with the risk of pancreatic cancer in both the discovery data and the validation data, and were all significantly correlated.

[0046] Table 3 Validation of four miRNA markers associated with the risk of pancreatic cancer

[0047]

[0048] Example 3: Detectability of four biomarkers at different progression stages

[0049] Based on the discovery cohort, the present invention explored the changes in microRNA markers with the time interval of pancreatic cancer diagnosis, wherein the time interval is represented by the follow-up time from baseline to pancreatic cancer diagnosis. Specifically, according to the follow-up time from baseline to pancreatic cancer diagnosis, the new pancreatic cancer patients were divided into a group every 3 years, and a total of two groups were obtained, of which 26 were diagnosed with pancreatic cancer within 3 years after baseline, and the remaining 14 were diagnosed after 3 years. For each microRNA marker, linear regression was used to correct for gender, baseline age and admission hospital, and the residual value was taken, and the Wilcoxon test was used to compare the differences between each new pancreatic cancer subgroup and the control group. Figure 1 As shown, based on the discovery cohort data, miR-139-3p ( P = 0.034) and miR-92b-5p ( P =0.031) can indicate the occurrence of pancreatic cancer more than 3 years after the baseline, miR-1908-5p ( P = 0.038) and miR-1976 ( P = 0.013) showed significant differences within the case subgroups within 3 years.

[0050] Based on the validation dataset, the present invention explored the expression of microRNA markers in different pancreatic cancer tumor stages. Among the 99 pancreatic cancer patients, there were 16 stage II pancreatic cancer patients, 27 stage III patients, 54 stage IV patients and 2 patients with only neoadjuvant pathologic stages (ypStage). After removing the 2 ypStage patients, linear regression was used to correct for gender and diagnosis age to obtain the residual value of each microRNA, and the Wilcoxon test was used to compare the differences between the case subgroups and the control group at different stages. Figure 2 As shown, compared with the control samples, miR-1908-5p ( P = 0.003), miR-139-3p ( P = 0.009) and miR-1976 ( P = 9×10 -4 ) is significantly increased in stage II tumors.

[0051] Example 4: Predictive Effects of Four Biomarkers and Their Combination on Pancreatic Cancer

[0052] This chapter, based on the JNCCHE study, employed leave-one-out cross-validation (LOOCV) to evaluate the predictive performance of miRNA markers by plotting receiver operating characteristic (ROC) plots and calculating the area under the ROC curve (AUC). Furthermore, the predictive performance of miRNA markers was evaluated by calculating the corresponding accuracy, sensitivity, positive predictive value, and negative predictive value, fixing the specificity at 0.80.

[0053] Figure 3 Table 4 and Table 4 show the predictive performance of the four miRNA markers in the validation dataset. The results showed that the AUCs for all four microRNAs were above 0.7, with the combined marker achieving the highest AUC of 0.847 (0.798, 0.896). Using the maximum Youden index as the cutoff value, the sensitivity and specificity of the combined marker were 82.8% and 72.0%, respectively. Furthermore, when the specificity was fixed at 0.80, the predictive accuracy, sensitivity, positive predictive value, and negative predictive value of the combined marker reached 0.771 (0.769, 0.772), 0.727 (0.639, 0.815), 0.705 (0.617, 0.794), and 0.816 (0.753, 0.878), respectively.

[0054] Table 4 Predictive effects of four miRNA markers in the JNCCHE study

[0055]

[0056] Note: The estimation results in the above table are based on a fixed specificity of 0.80.

[0057] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A microRNA biomarker for early diagnosis of pancreatic cancer, characterized in that: The microRNA biomarkers consist of miR-139-3p, miR-92b-5p, miR-1976, and miR-1908-5p; The nucleotide sequence of the miR-139-3p is shown in SEQ ID NO: 1, the nucleotide sequence of the miR-92b-5p is shown in SEQ ID NO: 2, the nucleotide sequence of the miR-1976 is shown in SEQ ID NO: 3, and the nucleotide sequence of the miR-1908-5p is shown in SEQ ID NO:

4.

2. Use of a reagent for detecting the microRNA biomarker according to claim 1 in preparing a kit for early diagnosis of pancreatic cancer.

Images