Circulating microRNA signatures for pancreatic cancer

Abstract

Provided herein is a method and kit for measuring the amount of specific microRNA biomarkers in samples collected from subjects.MicroRNA biomarkers are associated with pancreatic cancer.Also provided is a method for screening subjects for the amount of specific microRNA biomarkers, as well as a method for diagnosing, treating or monitoring subjects suspected of having pancreatic cancer.

Description

[Technical field]

[0001] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT Related Applications This application claims priority to U.S. Provisional Patent Application No. 63 / 220,195, filed July 9, 2021, the entire disclosure of which is incorporated herein by reference in its entirety. [Background technology]

[0002] Pancreatic cancer survival rates depend on the size of the tumor at the time of diagnosis and the extent of metastasis. The earlier pancreatic cancer is treated, the better the prognosis. Unfortunately, pancreatic cancer usually produces few or no symptoms until it has progressed and metastasized. Up to 80% of cases are diagnosed at a later stage, which is more difficult to treat (Pancreatic Cancer Prognosis, John Hopkins Medicine: available at https: / / www.hopkinsmedicine.org / health / conditions-and-diseases / pancreatic-cancer / pancreatic-cancer-prognosis). Compared to other cancers, pancreatic cancer has a very low 5-year survival rate of only 5-10%, due to a high percentage of people being diagnosed at stage IV, when the cancer has metastasized.

[0003] MicroRNAs (miRNAs) are small regulatory RNA molecules that control gene expression through RNA silencing and post-transcriptional regulation. They are often tissue specific and dysregulated in many cancers. MicroRNAs have double-stranded hairpin structures and are more stable than messenger RNAs. Some miRNAs have been detected in blood, and their amounts remain stable in blood samples for years or decades, providing the feasibility of using them as biomarkers for non-invasive cancer diagnosis. However, most studies have focused on miRNAs that are aberrantly expressed in tumor samples rather than blood samples.

[0004] There is a need in the art to identify circulating miRNAs for the accurate and reliable diagnosis of early stage pancreatic cancer. Summary of the Invention

[0005] The present disclosure provides a method for determining the presence or absence and / or amount of a microRNA in a sample (e.g., a blood sample) from a subject (e.g., a human subject), as well as a kit that includes a probe for the miRNA. The present disclosure also describes a method for treating a subject, as well as a method for screening a subject's blood sample for the presence or absence of a particular miRNA.

[0006] The present disclosure provides, in one aspect, a method for diagnosing pancreatic cancer in a subject, comprising: (a) obtaining a sample collected from a subject; (b) detecting and quantifying one or more test microRNAs selected from the group consisting of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-190a-3p, and hsa-miR-26b-5p in the sample; (c) comparing the amount of the test microRNA determined in step (b) with a statistical model; thereby diagnosing pancreatic cancer in a subject.

[0007] In some embodiments, the method further includes (d) detecting and quantifying one or more normalizing microRNAs selected from the group consisting of hsa-miR-17-5p, hsa-miR-199a-3p, hsa-miR-28-3p, and hsa-miR-92a-3p in the sample, and (e) normalizing the amount of the test microRNA using the amount of the normalizing microRNA quantified in step (d).

[0008] In some embodiments, methods are provided that include detecting and quantifying hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, and hsa-miR-26b-5p.

[0009] In exemplary embodiments, methods are provided that include detecting and quantifying hsa-miR-192-5p, hsa-let-7g-5p, hsa-let-7a-5p, hsa-miR-194-5p, hsa-miR-122-5p, hsa-miR-340-5p, and hsa-miR-26b-5p. In some embodiments, methods are provided that consist of detecting and quantifying hsa-miR-192-5p, hsa-let-7g-5p, hsa-let-7a-5p, hsa-miR-194-5p, hsa-miR-122-5p, hsa-miR-340-5p, and hsa-miR-26b-5p.

[0010] In other embodiments, methods are provided that include detecting and quantifying hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, and hsa-miR-194-5p.

[0011] In yet other embodiments, methods are provided that include detecting and quantifying any of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, and hsa-let-7g-5p; or hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, and hsa-miR-26b-5p.

[0012] In some embodiments, a method is provided comprising detecting and quantifying hsa-miR-323a-5p, hsa-miR-190a-3p, hsa-miR-192-5p, and hsa-let-7d-5p.

[0013] In another embodiment, a method is provided comprising detecting and quantifying hsa-miR-192-5p and hsa-miR-194-5p.

[0014] In yet other embodiments, methods are provided comprising detecting and quantifying hsa-miR-192-5p, hsa-let-7a-5p, hsa-miR-194-5p, hsa-let-7f-5p, hsa-miR-122-5p, hsa-miR-340-5p, and hsa-miR-26b-5p.

[0015] In some embodiments, methods are provided that include detecting and quantifying at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 test microRNAs.

[0016] In other embodiments, a method is provided that includes detecting and quantifying reference microRNAs hsa-miR-17-5p, hsa-miR-199a-3p, hsa-miR-28-3p, and hsa-miR-92a-3p.

[0017] In a further embodiment, a method is provided comprising detecting and quantifying the reference microRNAs hsa-miR-17-5p, hsa-miR-199a-3p, and hsa-miR-92a-3p.

[0018] In yet other embodiments, a method is provided that includes detecting and quantifying at least two or three normalizing microRNAs.

[0019] In other embodiments, a method is provided that includes detecting and quantifying microRNA by detecting the binding of the sample to at least one probe that can specifically hybridize to each of the microRNAs or their cDNAs. In some embodiments, at least one of the probes comprises a detectable label. In other embodiments, each of the probes comprises a detectable label.

[0020] In further embodiments, a method is provided that includes detecting and quantifying microRNA by a nucleic acid detection assay. In some embodiments, the assay is selected from the group consisting of microarray, RT-PCR, and RT-qPCR.

[0021] In additional embodiments, a method is provided that includes detecting and quantifying microRNAs by reverse transcribing microRNA molecules in a sample, thereby obtaining a cDNA sample, and sequencing the cDNA sample. In some embodiments, the method further includes amplifying DNA molecules in the cDNA sample before sequencing the cDNA sample. In some embodiments, the method of detecting and quantifying microRNAs is performed using miRNA-seq.

[0022] In yet another aspect, the present disclosure provides a kit comprising at least one test probe capable of specifically hybridizing to a microRNA selected from the group consisting of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-190a-3p, and hsa-miR-26b-5p, or their cDNA.

[0023] In some embodiments, a test probe that specifically hybridizes to hsa-miR-192-5p, a test probe that specifically hybridizes to hsa-miR-98-5p, a test probe that specifically hybridizes to hsa-let-7g-5p, a test probe that specifically hybridizes to hsa-let-7f-5p, a test probe that specifically hybridizes to hsa-let-7a-5p, a test probe that specifically hybridizes to hsa-miR-122-5p, Kits are provided that include test probes that specifically hybridize to hsa-let-7d-5p, test probes that specifically hybridize to hsa-miR-340-5p, test probes that specifically hybridize to hsa-miR-194-5p, and test probes that specifically hybridize to hsa-miR-26b-5p, or test probes that specifically hybridize to the cDNAs of those microRNAs.

[0024] In another embodiment, Kits are provided that include test probes that specifically hybridize to hsa-miR-192-5p, test probes that specifically hybridize to hsa-let-7g-5p, test probes that specifically hybridize to hsa-let-7a-5p, test probes that specifically hybridize to hsa-miR-122-5p, test probes that specifically hybridize to hsa-miR-340-5p, test probes that specifically hybridize to hsa-miR-194-5p, and test probes that specifically hybridize to hsa-miR-26b-5p, or their cDNAs.

[0025] In some embodiments, a kit is provided that comprises a test probe that specifically hybridizes to hsa-miR-192-5p, a test probe that specifically hybridizes to hsa-let-7g-5p, a test probe that specifically hybridizes to hsa-let-7a-5p, a test probe that specifically hybridizes to hsa-miR-122-5p, a test probe that specifically hybridizes to hsa-miR-340-5p, a test probe that specifically hybridizes to hsa-miR-194-5p, and a test probe that specifically hybridizes to hsa-miR-26b-5p, or its cDNA.

[0026] In additional embodiments, kits are provided that include test probes that specifically hybridize to hsa-miR-192-5p, test probes that specifically hybridize to hsa-miR-98-5p, test probes that specifically hybridize to hsa-let-7f-5p, test probes that specifically hybridize to hsa-let-7a-5p, test probes that specifically hybridize to hsa-miR-122-5p, test probes that specifically hybridize to hsa-let-7d-5p, test probes that specifically hybridize to hsa-miR-340-5p, and test probes that specifically hybridize to hsa-miR-194-5p, or test probes that specifically hybridize to the cDNAs of those microRNAs.

[0027] In a further embodiment, a kit is provided that includes a test probe that specifically hybridizes to hsa-let-7g-5p, or a test probe that specifically hybridizes to hsa-miR-26b-5p, or a test probe that specifically hybridizes to the cDNA of that microRNA.

[0028] In some embodiments, kits are provided that include test probes that specifically hybridize to hsa-miR-192-5p and test probes that specifically hybridize to hsa-miR-194-5p, or test probes that specifically hybridize to the cDNAs of those microRNAs.

[0029] In further embodiments, kits are provided that include test probes that specifically hybridize to hsa-miR-192-5p, test probes that specifically hybridize to hsa-let-7a-5p, test probes that specifically hybridize to hsa-miR-194-5p, test probes that specifically hybridize to hsa-let-7f-5p, test probes that specifically hybridize to hsa-miR-122-5p, test probes that specifically hybridize to hsa-miR-340-5p, and test probes that specifically hybridize to hsa-miR-26b-5p, or test probes that specifically hybridize to the cDNAs of those microRNAs.

[0030] In some embodiments, kits are provided that include at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 test probes.

[0031] In a further embodiment, a kit is provided comprising at least one normalization probe capable of specifically hybridizing to a microRNA selected from the group consisting of hsa-miR-17-5p, hsa-miR-199a-3p, hsa-miR-28-3p, and hsa-miR-92a-3p, or its cDNA.

[0032] In some embodiments, a kit is provided that includes a normalization probe capable of specifically hybridizing to hsa-miR-17-5p, a normalization probe capable of specifically hybridizing to hsa-miR-199a-3p, a normalization probe capable of specifically hybridizing to hsa-miR-28-3p, and a normalization probe capable of specifically hybridizing to hsa-miR-92a-3p, or a normalization probe capable of specifically hybridizing to the cDNA of those microRNAs.

[0033] In additional embodiments, kits are provided that include normalization probes capable of specifically hybridizing to hsa-miR-17-5p, normalization probes capable of specifically hybridizing to hsa-miR-199a-3p, and normalization probes capable of specifically hybridizing to hsa-miR-92a-3p, or normalization probes capable of specifically hybridizing to the cDNAs of those microRNAs.

[0034] In other embodiments, a kit is provided that includes at least two or three normalization probes.

[0035] In an additional embodiment, a kit is provided that does not include normalization probes.

[0036] In further embodiments, a kit is provided that includes at least one probe that includes a detectable label. In some embodiments, each of the probes includes a detectable label.

[0037] In some embodiments, a kit is provided that includes reagents for reverse transcription of microRNA molecules.

[0038] In yet another aspect, the disclosure provides a method for treating a subject suspected of having pancreatic cancer, comprising: (a) obtaining a sample collected from a subject; (b) detecting and quantifying one or more test microRNAs selected from the group consisting of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-190a-3p, and hsa-miR-26b-5p in the sample; (c) comparing the amount of the test microRNA determined in step (b) to a statistical model; (d) selecting subjects for more invasive testing and / or investigation of pancreatic cancer based on the comparison of step (c), and optionally administering a treatment to the subject with pancreatic cancer. The present invention provides a method comprising:

[0039] In some embodiments, the more invasive test is selected from the group consisting of magnetic resonance imaging (MRI), computed tomography (CT) scan, x-ray, positron emission tomography-computed tomography (PET-CT) scan, endoscopy, ultrasound, nuclear scan, and biopsy.

[0040] In other embodiments, the pancreatic cancer surveillance comprises periodic imaging tests selected from the group consisting of magnetic resonance imaging (MRI), computed tomography (CT) scans, x-rays, positron emission tomography-computed tomography (PET-CT) scans, endoscopy, ultrasound, and nuclear scans. In some embodiments, periodic imaging is performed every 3, 6, or 12 months.

[0041] In yet other embodiments, the subject is receiving a treatment for pancreatic cancer. In some embodiments, the treatment is selected from the group consisting of surgery, chemotherapy, immunotherapy, and radiation therapy. In some embodiments, the treatment includes the immunotherapy pembrolizumab. In some embodiments, the treatment includes surgery. In some embodiments, the treatment includes chemotherapy and immunotherapy. In some embodiments, the treatment includes chemotherapy and radiation.

[0042] In additional embodiments, the subject is administered chemotherapy treatment for pancreatic cancer. In some embodiments, the chemotherapy is selected from the group consisting of taxanes, antimetabolites, platinum chemotherapy, alkylating agents, agents that inhibit DNA replication, PARP inhibitors, and antitumor chemotherapy. In some embodiments, the chemotherapy comprises a taxane selected from the group consisting of paclitaxel, docetaxel, and albumin-bound paclitaxel. In some embodiments, the chemotherapy comprises an antimetabolite selected from the group consisting of gemcitabine hydrochloride, 5-fluorouracil (5-FU), and capecitabine. In some embodiments, the chemotherapy comprises the platinum chemotherapy oxaliplatin. In some embodiments, the chemotherapy comprises the alkylating agent cisplatin. In some embodiments, the chemotherapy comprises an agent that inhibits DNA replication selected from the group consisting of irinotecan and liposomal irinotecan. In some embodiments, the chemotherapy comprises the PARP inhibitor olaparib. In some embodiments, the chemotherapy comprises an anti-tumor chemotherapy selected from the group consisting of everolimus, erlotinib hydrochloride, sunitinib, and mitomycin.

[0043] In further embodiments, the subject is administered a combination drug treatment for pancreatic cancer.In some embodiments, the combination drug is selected from the group consisting of FOLFIRINOX (folinic acid, fluorouracil, irinotecan hydrochloride, and oxaliplatin), GEMCITABINE-CISPLATIN (gemcitabine hydrochloride and cisplatin), GEMCITABINE-OXALIPLATIN (gemcitabine hydrochloride and oxaliplatin), and OFF (oxaliplatin, fluorouracil, and folinic acid).

[0044] In an additional aspect, the present disclosure provides an assay method for diagnosing pancreatic cancer in a subject, comprising: a) detecting and quantifying one or more test microRNAs selected from the group consisting of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-190a-3p, and hsa-miR-26b-5p in a sample from the subject; b) analyzing in a neural network the amount of one or more test microRNAs quantified in step a) to determine the probability that the subject has pancreatic cancer; c) assigning the subject as likely to have pancreatic cancer based on the analysis of step b); The present invention provides a method comprising:

[0045] In some embodiments, the assignment of a subject as likely to have pancreatic cancer has an accuracy rate of more than 50%, 60%, 70%, 80%, or 90%. In some embodiments, the assignment of a subject as likely to have pancreatic cancer has a specificity rate of more than 50%, 60%, 70%, 80%, or 90%. In some embodiments, the assignment of a subject as likely to have pancreatic cancer has a sensitivity rate of more than 50%, 60%, 70%, 80%, or 90%.

[0046] In further embodiments, the method includes obtaining a sample. In some embodiments, the sample is a blood sample or a pancreatic sample. In some embodiments, the blood sample is selected from the group consisting of plasma, serum, and whole blood.

[0047] In yet other embodiments, the method includes obtaining a sample collected from a subject. In some embodiments, the subject is a human subject. In some embodiments, the subject is at a higher risk of developing pancreatic cancer. In some embodiments, the subject has diabetes. In some embodiments, the subject has pancreatitis. In some embodiments, the subject has a family history of pancreatic cancer or pancreatitis. In some embodiments, the subject is at a higher risk of developing pancreatic cancer due to a genetic mutation. In some embodiments, the subject has a mutation in a gene selected from the group consisting of BRCA1, BRCA2, PALB2, TP53, MLH1, CDKN2A, and ATM.

[0048] In further embodiments, the method comprises the use of a statistical model. In some embodiments, the statistical model comprises one or more models selected from the group consisting of linear discriminant analysis, logistic regression, multivariate adaptive regression spline, naive Bayes, neural network, support vector machine, decision tree, K nearest neighbors, feature tree, least absolute deviation (LAD) tree, Bayesian network, elastic net regression, and random forest. [Brief description of the drawings]

[0049] [Figure 1] Figure 1 shows the schematic diagram of the study design of Example 1 for generating circulating miRNA signature from human serum using two independent cohorts of patients.The schematic diagram also shows how patients are assigned to training set, test set and validation set.The schematic diagram also shows the steps of using a series of statistical tools to create an algorithm, creating a final set of miRNAs, calibrating the model, and validating the model using qPCR. [Diagram 2]Figure 2 shows variable selection study of the training set of Example 1. Ten miRNAs were selected as having a Family-Wise Error Rate (FWER) p-value < 0.05 (with Bonferroni adjusted p-value). The Volcano plot (Figure 2A) and results table (Figure 2B) show that for these 10 miRNAs, three were upregulated and seven were downregulated. [Figure 3-1] Figure 3 shows the development of the classification model in Example 1, with logistic regression performance further refined by backward stepwise logistic regression. Figure 3A shows the plot of specificity vs. sensitivity for the miRNA models tested. Figure 3B shows the calculated values ​​for the four miRNAs used in the final model. These results show that the model fits very well to the training and test sets with Hosmer Lemeshow value=4.4927 and p-value=0.810161. Figure 3C shows the sensitivity and specificity for detecting cancer in cancer-positive samples versus controls. The final logistic regression model of the four miRNAs showed a sensitivity of 79.3% and a specificity of 84.1%. [Figure 3-2] (As stated above.) [Figure 4] Figure 4 shows the development of the classification model in Example 1 using an artificial neural network with sensitivity analysis used to reduce the number of miRNAs in the analysis. Figure 4A shows a plot of specificity versus sensitivity for the model miRNA models tested. The artificial neural network requires the following eight miRNAs: hsa-miR-192-5p, hsa-let-7a-5p, hsa-let-7d-5p, hsa-miR-194-5p, hsa-miR-98-5p, hsa-let-7f-5p, hsa-miR-122-5p, and hsa-miR-340-5p. Figure 4B shows the sensitivity and specificity of cancer detection in cancer-positive samples versus controls. The final artificial neural network showed a sensitivity of 71.4% and a specificity of 90.9%. [Diagram 5]5A and 5B show the partitioning of the dataset from Example 1 into training, test and validation sets for creating diagnostic models using logistic regression, artificial neural networks on raw datasets and artificial neural networks on SMOTE balanced datasets. As shown in FIG. 5A, the datasets of two groups of Polish samples were partitioned for modeling. FIG. 5B shows the use of SMOTE technology to create a balanced dataset from the training set. [Figure 6] Figure 6B shows the use of logistic regression of two miRNA (hsa-miR-192-5p and hsa-miR-194-5p) models in Example 1 to evaluate the performance of both test and validation datasets. Figure 6B shows that the model has a sensitivity rate of 66% and a specificity rate of 74% in predicting cancer in observed cancer samples versus controls. [Figure 7-1] Figure 7 shows the results of the neural network model of classical (non-SMOTE modified) data in Example 1. These results were generated using clinical data including age and gender, and all 10 miRNAs. The following miRNAs were used for normalization: hsa-miR-17-5p, hsa-miR-92a-3p, and hsa-miR-199a-3p. Figure 7A shows that the area under the ROC curve (AUC) of the training set is 0.8475. Figure 7B provides the values ​​of 82.57% accuracy, 59.72% sensitivity, and 93.84% specificity for the training and test datasets. Figure 7B provides the values ​​of 83.02% for accuracy, 64.71% for sensitivity, and 91.67% for specificity for the validation dataset. [Figure 7-2] (As stated above.) [Figure 8-2]Figure 8 shows the results of the neural network model on the SMOTE balanced dataset in Example 1. These results were generated using a clinical dataset and a trimmed miRNA set including hsa-miR-192-5p, hsa-let-7a-5p, hsa-miR-194-5p, hsa-let-7f-5p, hsa-miR-122-5p, hsa-miR-340-5p, and hsa-miR-26b-5p. The following miRNAs were used for normalization: hsa-miR-17-5p, hsa-miR-92a-3p, and hsa-miR-199a-3p. Figure 8A shows that the AUC of the dataset is 0.8971. Figure 8B provides the values ​​of accuracy of 84.86%, sensitivity of 79.17%, and specificity of 87.67% for the training and test datasets. FIG. 8B provides values ​​for accuracy of 86.79%, sensitivity of 76.47%, and specificity of 91.67% for the validation dataset. [Figure 8-2] (As stated above.) DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0050] The present disclosure provides methods and kits for measuring the amount of specific miRNA biomarkers in samples collected from a subject. The miRNA biomarkers are associated with cancer (e.g., pancreatic cancer). Specifically, unique combinations of miRNA testing and normalization are described that are used to predict an increased probability that a subject has cancer (e.g., pancreatic cancer) in a statistically relevant manner. The use of these miRNA combinations provides a non-invasive cancer detection method that is useful for monitoring an individual's susceptibility to disease. The detection methods can be used alone or in combination with other known diagnostic methods. The methods described herein are particularly useful for detecting or diagnosing pancreatic cancer. For example, these methods are effective in differentiating pancreatic cancer from pancreatitis, which is the most common differential diagnosis. Also provided herein are methods for investigating and treating subjects diagnosed with cancer (e.g., pancreatic cancer).

[0051] It is to be understood that the methods described in this disclosure are not limited to the particular methods and experimental conditions disclosed herein, as such methods and conditions may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting.

[0052] Furthermore, the experiments described herein use molecular cell biology and immunological techniques that are familiar to those skilled in the art, unless otherwise specified.Such techniques are well known to those skilled in the art and are fully described in the literature.See, for example, Ausubel, et al., ed., Current Protocols in Molecular Biology, John Wiley & Sons, Inc., NY, NY (1987-2008) (including all supplements), Molecular Cloning: A Laboratory Manual (Fourth Edition) by MR Green and J. Sambrook and Harlow et al., Antibodies: A Laboratory Manual, Chapter 14, Cold Spring Harbor Laboratory, Cold Spring Harbor (2013, 2nd edition).

[0053] Unless otherwise defined herein, scientific and technical terms used herein have the meanings commonly understood by those skilled in the art. In case of any potential ambiguity, the definitions provided herein take precedence over dictionary or exogenous definitions. Unless otherwise required by context, singular words include plurals and plural words include the singular. The use of "or" means "and / or" unless otherwise specified. The use of the term "comprising" and other forms such as "including" and "included" is not limiting.

[0054] In general, the nomenclature used in connection with cell and tissue culture, molecular biology, immunology, microbiology, genetics, and protein and nucleic acid chemistry, and hybridization described herein is well known and commonly used in the art. The methods and techniques provided herein are generally performed according to conventional methods well known in the art and described in various general and more specific references cited and discussed throughout the specification, unless otherwise stated. Enzymatic reactions and purification techniques are performed according to manufacturer's specifications as commonly accomplished in the art or as described herein. The nomenclature used in connection with the analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry experimental procedures and techniques described herein are well known and commonly used in the art. Standard techniques are used for chemical synthesis, chemical analysis, pharmaceutical preparation, formulation, and delivery, and patient treatment.

[0055] RNA and DNA Detection Methods The present disclosure provides compositions and methods for cancer diagnosis and treatment.In exemplary embodiments, herein, a diagnostic test for pancreatic cancer is provided that is highly sensitive and specific.This diagnostic test relies on the detection and quantification of nucleic acid.In particular, a diagnostic test is provided that relies on the detection and quantification of microRNA.

[0056] Provided herein is a method for quantifying and detecting nucleic acid.As used herein, the term "nucleic acid" refers to a polymer of two or more nucleotides or nucleotide analogs (e.g., ribonucleic acid with methylene bridge between 2'-O and 4'-C atom of ribose ring) that can hybridize with complementary nucleic acid.As used herein, this term includes, but is not limited to, DNA, RNA, LNA, and PNA.

[0057] Specifically, methods are provided herein for the detection and quantification of microRNAs. As used herein, the term "microRNA" or "miRNA" refers to small non-coding ribonucleic acid (RNA) gene products of 19-26 nucleotides in length that form hairpin secondary structures. The microRNAs described herein are named using the nomenclature described in Ambros et al., RNA. 2003 Mar;9(3):277-9, which is incorporated herein by reference, and sequences can be found at mirbase.org.

[0058] In some embodiments, the detection and quantification of miRNAs comprises the use of test microRNAs.As used herein, the term "test microRNA" refers to the microRNA whose presence or absence and / or amount is determined (e.g., using an algorithm), for example, for diagnostic purposes.In some embodiments, the presence or absence and / or amount of one or more test microRNAs can additionally be used for normalization purposes.

[0059] As used herein, the phrase "detect and quantify one or more test microRNAs" encompasses any method that can be used to measure the concentration, absolute value or presence of microRNA.Exemplary methods for determining the amount of microRNA include sequencing (e.g., Gilbert sequencing, Sanger sequencing, SMRT sequencing or next-generation sequencing), microarray detection, PCR, RT-PCR, real-time qPCR, and real-time RT-qPCR.

[0060] In some embodiments, the detection and quantification of miRNAs is performed using normalization. As used herein, the term "normalize" or "normalization" refers to adjusting a first measurement (e.g., the level of a gene of interest) to a second measurement (e.g., the level of a housekeeping gene), where the first and second measurements are measured from the same sample (e.g., different parts of the same homogenous sample), and the second measurement correlates to the amount and / or quality of the sample. Normalization allows for a relative amount of the first value to be obtained that is not affected by the amount and / or quality of the sample, which may vary with individual sample preparations. As used herein, the term "normalizing microRNA" or "reference microRNA" refers to a microRNA that is known to have a stable amount in a sample (e.g., a blood sample) and is used to normalize the measurements of a test microRNA in the sample. A single normalizing microRNA can be used to normalize the measured amount of a target microRNA in a sample, or it can be used to normalize the average value of multiple microRNAs. In certain embodiments, normalization may be calculated by the number of amplification cycles (average of normalizer microRNA) minus the number of amplification cycles (miR of interest).

[0061] In an exemplary embodiment, one or more, or various combinations of the miRNAs in Tables 1 and 2 are quantified using the methods disclosed herein. Table 1 provides reference miRNAs for normalizing results. Table 2 provides test miRNAs that serve as biomarkers for cancer (e.g., pancreatic cancer).

[0062] [Table 1]

[0063] [Table 2]

[0064] In some embodiments, the levels of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 of the test miRNAs in Table 2 are detected and quantified. In an exemplary embodiment, the levels of 7 of the test miRNAs in Table 2 are detected and quantified.

[0065] In some embodiments, the levels of one, two, three, or four of the reference miRNAs in Table 1 are detected and quantified. In an exemplary embodiment, the levels of all four of the reference miRNAs in Table 1 are detected and quantified. In an additional exemplary embodiment, the levels of four of the reference miRNAs in Table 1 and seven of the test miRNAs in Table 2 are detected and quantified.

[0066] In an exemplary embodiment, the miRNA to be detected and quantified is obtained from a sample from a subject.The term "subject" as used herein refers to a mammal, for example, a human, a domestic animal or livestock, for example, a cat, a dog, a cow, or a horse.The term "sample" as used herein refers to a biological specimen of material from a subject, such as tissue or body fluid.

[0067] As used herein, "obtaining a sample collected from a subject" encompasses any suitable means disclosed herein, as well as routine clinical methods for removing biological specimens from a subject. Samples can be taken directly from a subject or obtained from a third party. Sample collection can be performed by a healthcare provider, such as, for example, a physician, physician assistant, nurse, veterinarian, dentist, chiropractor, emergency physician, dermatologist, oncologist, gastroenterologist, or surgeon. Samples include, but are not limited to, blood, mucous membranes (e.g., saliva), lymph, urine, feces, and solid tissue samples. In an exemplary embodiment, a bodily fluid sample is collected from a subject. Techniques for obtaining bodily fluid samples from a subject are well known, including techniques for collecting and processing whole blood and lymph.

[0068] In an exemplary embodiment, the sample obtained from the subject is a blood sample. As used herein, the term "blood sample" refers to a volume of blood taken from a subject, such as whole blood, or a component part of blood taken from a subject, such as plasma, which lacks cells normally contained in whole blood (e.g., red blood cells, white blood cells, and platelets), or serum, which is plasma lacking fibrinogen and some clotting factors. In some embodiments, the sample is a tissue sample. The tissue sample can be obtained by biopsy. In some embodiments, the sample is a tissue biopsy (e.g., needle biopsy, CT-guided needle biopsy, aspiration biopsy, endoscopic biopsy, bronchoscopy biopsy, bronchial lavage, incisional biopsy, excision biopsy, punch biopsy, slice biopsy, skin biopsy, bone marrow biopsy, or electrochemical loop excision).

[0069] In some embodiments, the detection and quantification of miRNA comprises a step or steps of binding between nucleic acids. As used herein, the term "binding" or "binding" refers to the non-covalent or covalent interaction between two molecules, for example, between two complementary nucleic acids.

[0070] In some embodiments, the detection and quantification of miRNAs includes a step or steps of hybridization between nucleic acids. The term "hybridize" as used herein refers to annealing a first single-stranded nucleic acid to a second complementary single-stranded nucleic acid, where the complementary nucleotides of the first and second nucleic acids form a pair by hydrogen bonds. As used herein, the term "specifically hybridize" refers to a non-covalent interaction between a first nucleic acid molecule (e.g., a nucleic acid probe having a specific nucleotide sequence) and a second nucleic acid molecule (e.g., a microRNA having a nucleotide sequence complementary to that of the nucleic acid probe). Hybridization conditions are reported in the art and known to those skilled in the art. In some embodiments, the conditions for detecting hybridization are appropriate conditions for a nucleic acid detection assay (e.g., microarray, RT-PCR, or RT-qPCR). The possibility of hybridization between two nucleic acids correlates with the complementary nucleotide sequence between the two nucleic acids. An oligonucleotide "specifically hybridizes" to a target polynucleotide if the oligonucleotide hybridizes to the target under physiological conditions at a Tm of greater than 37°C, greater than 45°C, preferably at least 50°C, and typically 60°C to 80°C or higher. The "Tm" of an oligonucleotide is the temperature at which 50% hybridizes to a complementary polynucleotide. Tm is determined under standard conditions in saline, for example, as described in Miyada et al., Methods Enzymol. 154:94-107 (1987).

[0071] In some embodiments, the detection and quantification of miRNA comprises the step of complementary nucleic acid interaction.Polynucleotides are described as "complementary" to each other when hybridization occurs between two single-stranded polynucleotides in antiparallel configuration.Complementarity (the degree to which a polynucleotide is complementary to another polynucleotide) can be quantified by the proportion of bases in opposing strands that are expected to form hydrogen bonds with each other according to the generally accepted rules of base pairing.

[0072] In some embodiments, the detection and quantification of miRNA is carried out by PCR. The term "PCR" as used herein refers to polymerase chain reaction to amplify a quantity of target DNA. PCR relies on thermal cycling, which consists of repeated heating and cooling cycles of the reaction for DNA denaturation, annealing and enzymatic extension of the amplified DNA. First, the strands of DNA are separated at high temperature in a process called melting or denaturation of DNA. Then, the temperature is lowered, allowing the primer and the strands of target DNA to selectively bind or anneal, creating a template for DNA polymerase to amplify the target DNA. Then, at the operating temperature of DNA polymerase, template-dependent DNA synthesis occurs. These steps are repeated to make many copies of the target DNA.

[0073] In some embodiments, the detection and quantification of miRNA involves the use of primers. As used herein, "primer" refers to a short, single-stranded DNA sequence that selectively binds to a target DNA sequence and allows the addition of a new deoxyribonucleotide by DNA polymerase at the 3' end. According to certain embodiments, the forward primer is 18-35, 19-32 or 21-31 nucleotides in length. The nucleotide sequence of the forward primer is not limited as long as it specifically hybridizes to a part or all of the target site and its Tm value can be in the range of 50°C to 72°C, particularly in the range of 58°C to 61°C, and can be in the range of 59°C to 60°C. The nucleotide sequence of the primer can be manually designed to confirm the Tm value using a primer Tm prediction tool. The primer nucleotides can include nucleotide analogs and / or modified nucleotides, such as LNA or PNA.

[0074] In some embodiments, the detection and quantification of miRNA is carried out using RT-PCR.As used herein, the term "RT-PCR" refers to reverse transcription polymerase chain reaction, which is a process for amplifying RNA.RNA molecules are reverse transcribed into complementary DNA (cDNA) using reverse transcriptase enzyme, and then the resulting cDNA is amplified using PCR.

[0075] In some embodiments, the detection and quantification of miRNAs is performed using RT-qPCR. As used herein, the term "RT-qPCR" refers to reverse transcription quantitative polymerase chain reaction, a variant of RT-PCR in which the amplification of cDNA during the RT-PCR process is quantitatively detected in real time using a probe that detects the amplified target DNA. For example, in some embodiments, a self-quenching nucleic acid probe is added to the reaction mixture. The self-quenching nucleic acid probe fluoresces only when bound to the target sequence. Upon completion of each cycle of PCR, the self-quenching probe binds to the amplified DNA, is unquenched, and fluoresces upon exposure to a light excitation source. As the DNA is amplified, the increase in probe and target binding results in increased fluorescence of the self-quenching nucleic acid probe. Detection of the fluorescent probe after each amplification cycle allows for real-time measurement of the amplification process, since it binds to the amplified target DNA and increases the amount of the nucleic acid probe that fluoresces. In some embodiments, an intercalating dye probe is added to the reaction mixture that fluoresces upon interaction with double-stranded nucleic acid. The increase in dye fluorescence during the amplification process allows for measurement of DNA amplification in real time, as the amount of dye probe intercalation increases as the amount of target DNA being amplified increases.

[0076] In some embodiments, the detection and quantification of miRNA involves the use of a probe. As used herein, the term "probe" refers to a molecule or complex used to determine the presence or absence and / or amount of a microRNA in a sample (e.g., a blood sample). In certain embodiments, the probe comprises a nucleic acid portion (e.g., DNA, modified DNA, or modified RNA) that can specifically hybridize to a microRNA or its complementary DNA (cDNA). In certain embodiments, the probe comprises a sequence of at least 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or 25 consecutive nucleotides that are identical or complementary to the microRNA. In certain embodiments, the probe further comprises a detectable label that is covalently or non-covalently conjugated to the nucleic acid portion. Exemplary detectable labels include, but are not limited to, a fluorophore, a small molecule (e.g., avidin family small molecule), an enzyme, an antibody or antibody fragment, or a nucleic acid sequence that is not present in the subject in a form that is linked to the microRNA (e.g., a barcode sequence). Thus, the probe can be a fluorophore-labeled nucleic acid having a nucleotide sequence that is complementary to the nucleotide sequence of a microRNA.

[0077] In some embodiments, the detection and quantification of miRNA comprises the use of normalization probe. As used herein, the term "normalization probe" refers to a probe used to determine the presence or absence and / or amount of normalization microRNA in a sample (e.g., blood sample). In certain embodiments, the normalization probe comprises a nucleic acid moiety (e.g., DNA, modified DNA, or modified RNA) that can specifically hybridize to normalization microRNA or its complementary DNA (cDNA).

[0078] In some embodiments, the detection and quantification of miRNAs involves the use of test probes. As used herein, the term "test probe" refers to a probe used to determine the presence or absence and / or amount of a test microRNA in a sample (e.g., a blood sample). In certain embodiments, the test probe comprises a nucleic acid moiety (e.g., DNA, modified DNA, or modified RNA) that can specifically hybridize to the test microRNA or its complementary DNA (cDNA).

[0079] In some embodiments, the detection and quantification of miRNA involves the use of reagents for amplifying DNA sequences. The phrase "reagents for amplifying DNA sequences" includes, but is not limited to, (1) thermostable DNA polymerase; (2) deoxynucleotide triphosphates (dNTPs); (3) buffer that provides the appropriate chemical environment for optimal activity, binding kinetics, and stability of DNA polymerase; (4) divalent cations such as magnesium or manganese ions; and (5) monovalent cations such as potassium ions. The reagents can be provided in the form of a solution, a concentrated solution, or a powder.

[0080] In some embodiments, the detection and quantification of miRNA comprises the use of a reagent for reverse transcription of RNA molecules.The phrase "reagent for reverse transcription of RNA molecules" includes, but is not limited to, reverse transcriptase; RNase inhibitor; primer hybridized to nucleic acid sequence (such as RNA or DNA); primer hybridized to adenosine oligonucleotide; and buffer that provides the appropriate chemical environment for optimal activity, binding kinetics, and stability of reverse transcriptase.The reagent can be provided in the form of solution, concentrated solution, or powder.

[0081] In some embodiments, the detection and quantification of miRNA is preformed using next-generation sequencing.As used herein, the term "next-generation sequencing" refers to high-throughput parallel sequencing of short pieces of single-stranded nucleic acid attached to slides or beads, such as ILLUMINA, ROCHE (454 sequencing), or ION TORRENT, THERMOFISHER technology.The incorporation of individual nucleotides into single-stranded nucleic acid can be detected optically (through the fluorescence of the incorporated nucleotide) or by detecting hydrogen ions released during the incorporation of nucleotide (e.g., ion semiconductor sequencing).

[0082] In some embodiments, the detection and quantification of miRNA is preformed using microarray detection.As used herein, the term "microarray detection" refers to a method of detecting target nucleic acid using single-stranded nucleic acid probes attached to separate areas of a solid surface (e.g., spots on a slide or beads in a microwell).The hybridization of the probe to a specific nucleic acid can be detected by various methods, such as using optical detection (e.g., fluorophores, chemiluminescent molecules) or radiological detection.

[0083] In some embodiments, the detection and quantification of miRNAs involves the use of non-natural labels. As used herein, the term "non-natural labels" includes, but is not limited to, one or more labeling molecules that can bind to, attach to, or associate with a biological molecule (e.g., a nucleic acid, nucleotide, protein, peptide, amino acid, carbohydrate, lipid, primary / secondary metabolite, or chemical product produced by a living organism) and allow the detection of the molecule when associated with the biological molecule; non-natural labels are not normally associated with a biological molecule. Exemplary non-natural labels include, but are not limited to, antigenic tags (e.g., digoxigenin); radioisotopes (e.g., 32P); enzymes that catalyze chemiluminescent or colorimetric chemical reactions (e.g., horseradish peroxidase or alkaline phosphatase); nucleic acid dyes (e.g., Hoechst 33342, DAPI, ethidium bromide); organic fluorophores (e.g., 6-carboxyfluorescein, tetrachlorofluorescein, fluorescein, rhodamine, or cyanine); fluorophore quenchers (e.g., tetramethylrhodamine, dimethylaminoazobenzenesulfonic acid, BLACK HOLE QUENCHERS, or IOWA BLACK dye); protein fluorophores (e.g., green fluorescent protein); donor and acceptor fluorophores for fluorescence resonance energy transfer (e.g., fluorescein and tetramethylrhodamine, or NowGFP and mOrange); quantum dot fluorophores (e.g., metal chalcogenides, core-shell semiconductor nanocrystals, or alloy semiconductor quantum dots); and immune system-based molecules bound, attached, or associated with the non-natural labels described herein (e.g., antibodies or antibody fragments labeled with fluorophores or catalytic enzymes).

[0084] Statistical Models for Diagnosis The present disclosure provides compositions and methods for cancer diagnosis and treatment. In an exemplary embodiment, the applicant provides a highly sensitive and specific cancer (e.g., pancreatic cancer) diagnostic test. This diagnostic test relies on the use of statistical models. In an exemplary embodiment, one or more or various combinations of the miRNAs in Tables 1 and 2 are quantified using the methods disclosed herein, and then further analyzed using statistical models to determine a diagnosis. As used herein, the term "diagnosis" refers to identifying or recognizing that an individual may have a particular disease, such as cancer (e.g., pancreatic cancer).

[0085] In an exemplary embodiment, the diagnostic method of the present disclosure is carried out using a statistical model. As used herein, the term "statistical model" refers to a mathematical representation of observed data. The statistical model used in the method disclosed herein can be any statistical model known in the art. Exemplary statistical models include one or more models selected from the group consisting of linear discriminant analysis, logistic regression, multivariate adaptive regression spline, naive Bayes, artificial neural network, support vector machine, decision tree, K nearest neighbor classifier, feature tree, least absolute deviation (LAD) tree, Bayesian network, elastic net regression, and random forest.

[0086] Disclosed herein are groups of miRNAs that can be used individually and in groups or subsets to enhance the diagnosis of cancer (e.g., pancreatic cancer). An example of a miRNA that is useful in the diagnostic methods disclosed herein is shown in Table 2. The test miRNAs presented in Table 2 can be used as biomarkers to predict the likelihood that an individual has cancer (e.g., pancreatic cancer). In some embodiments, the levels of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 of the test miRNAs in Table 2 are determined and compared to a statistical model. In an exemplary embodiment, the levels of 7 of the test miRNAs in Table 2 are determined and compared to a statistical model. In an exemplary embodiment, the levels of one or more test miRNAs from Table 2 are normalized. The normalization involves the use of a reference miRNA that serves as a baseline for determining the relative quantification of the test miRNA. In an exemplary embodiment, the test miRNAs include one or more miRNAs in Table 2, and the reference miRNAs include one or more miRNAs in Table 1. In some embodiments, one, two, three, or four of the reference miRNAs in Table 1 are used to compare the normalized amount of the test miRNA to the statistical model. In an exemplary embodiment, four of the reference miRNAs in Table 1 are used to compare the normalized amount of the test miRNA to the statistical model.

[0087] In an exemplary embodiment, provided herein is a method for diagnosing pancreatic cancer in a subject, the method comprising: (a) obtaining a sample collected from the subject; (b) detecting and quantifying one or more test microRNAs selected from the group consisting of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-190a-3p, and hsa-miR-26b-5p in the sample; and (c) comparing the amount of the test microRNA determined in step (b) with a statistical model, thereby diagnosing pancreatic cancer in the subject.

[0088] In an exemplary embodiment, the statistical model used is logistic regression.As used herein, the term "logistic regression" is a statistical model used to determine whether independent variables affect binary dependent variables.Logistic regression is used to describe data and explain the relationship between one dependent binary variable and one or more nominal, ordinal, interval or ratio level independent variables.

[0089] In an exemplary embodiment, the statistical model used is a neural network. As used herein, the term "artificial neural network" or "neural network" refers to a predictive model based on a linked collection of neural units in silico that loosely models a simple mathematical model of the brain. An artificial neural network allows for the identification of complex nonlinear relationships between its response variables and its predictor variables. An artificial neural network can have one or more hidden layers, each layer containing one or more neurons that interact to generate a prediction given two or more variables.

[0090] In some embodiments, the amount of miRNA is compared to a statistical model to calculate the probability that a subject has cancer (e.g., pancreatic cancer). In an exemplary embodiment, a neural network is used to calculate the probability that a subject has cancer (e.g., pancreatic cancer). Disclosed herein is an analysis method for diagnosing cancer (e.g., pancreatic cancer), in which a subject is assigned as likely to have cancer based on the calculation of the probability that the subject has cancer (e.g., pancreatic cancer) in a statistical model. As used herein, "likely to have cancer" means that a subject is more likely to have cancer than the statistical occurrence of cancer in the general population. In some embodiments, a subject is assigned as likely to have cancer if the likelihood that the subject has cancer is calculated to be at least about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, or about 100%.As used herein, "likely to have pancreatic cancer" means that the subject is more likely to have pancreatic cancer than the statistical occurrence of pancreatic cancer in the general population. In some embodiments, if the likelihood that the subject has pancreatic cancer is calculated to be at least about 5%, about 10%, about 15%, about 20%, about 25%, about 30%, about 35%, about 40%, about 45%, about 50%, about 55%, about 60%, about 65%, about 70%, about 75%, about 80%, about 85%, about 90%, about 95%, or about 100%, the subject is assigned as having a high likelihood of having pancreatic cancer.In an exemplary embodiment, if the likelihood that the subject has pancreatic cancer is calculated to be at least about 50%, the subject is assigned as having a high likelihood of having pancreatic cancer.

[0091] The method disclosed herein can determine with high accuracy whether a subject has cancer from a sample derived from the subject.In some embodiments, the accuracy of the diagnosis of cancer is at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or 100%.In some embodiments, the accuracy of the diagnosis of pancreatic cancer is at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or 100%.In an exemplary embodiment, the accuracy of the diagnosis of pancreatic cancer is at least about 80%.

[0092] The method disclosed herein can determine whether a subject has cancer with a high degree of specificity from a sample derived from the subject. In other words, the method disclosed herein can detect cancer vs. general differential diagnosis with a high degree of accuracy. In some embodiments, the specificity of the diagnosis of cancer is at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%. In some embodiments, the specificity of the diagnosis of pancreatic cancer is at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%. In an exemplary embodiment, the specificity of the diagnosis of pancreatic cancer is at least 80%.

[0093] The method disclosed herein can determine whether a subject has cancer with high sensitivity from a sample derived from the subject.In some embodiments, the sensitivity of diagnosing cancer is at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%.In some embodiments, the sensitivity of diagnosing pancreatic cancer is at least about 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%.In an exemplary embodiment, the sensitivity of diagnosing pancreatic cancer is at least 80%.

[0094] In an exemplary embodiment, provided herein is an analytical method for diagnosing pancreatic cancer in a subject, comprising: (a) detecting in a sample from the subject hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-323b-5p, hsa-miR-323c-5p, hsa-miR-323d-5p, hsa-miR-323e-5p, hsa-miR-323f ... The method includes (a) detecting and quantifying one or more test microRNAs selected from the group consisting of iR-190a-3p, and hsa-miR-26b-5p; (b) analyzing in a neural network the amount of the one or more microRNAs quantified in step a) to determine the probability that the subject has pancreatic cancer; and (c) assigning the subject as likely to have pancreatic cancer based on the analysis of step b).

[0095] Cancer Diagnosis and Treatment The methods disclosed herein can be used alone or in combination with other methods for the diagnosis and treatment of cancer.

[0096] Cancers that can be diagnosed using the methods disclosed herein include, but are not limited to, solid tumors, hematological cancers (e.g., leukemia, lymphoma, myeloma, e.g., multiple myeloma), and metastatic lesions. In one embodiment, the cancer is a solid tumor. Examples of solid tumors include malignant tumors, e.g., sarcomas and carcinomas, e.g., adenocarcinomas of various organ systems, e.g., those affecting the lung, breast, ovary, lymphatic system, gastrointestinal tract (e.g., colon), anus, genital and genitourinary tract (e.g., kidney, urothelium, bladder cells, prostate), pharynx, CNS (e.g., brain, neural or glial cells), head and neck, skin (e.g., melanoma), and pancreas, as well as adenocarcinomas, including malignant tumors such as colon cancer, rectal cancer, renal cell carcinoma, liver cancer, lung cancer (e.g., non-small cell lung cancer or small cell lung cancer), small intestine cancer, and esophageal cancer. Cancers can be early stage, intermediate stage, late stage, or metastatic cancer.

[0097] In an exemplary embodiment, the cancer is pancreatic cancer. As used herein, the term "pancreatic cancer" refers to a group of malignant tumors that affect the pancreas. Adenocarcinoma of the pancreas is the most common type of pancreatic cancer and begins as a cancerous proliferation of exocrine cells. Pancreatic neuroendocrine tumors, or islet cell tumors, begin in endocrine cells, are less common. Approximately 95% of exocrine pancreatic cancers are adenocarcinomas, which usually begin in the pancreatic ducts (Pancreatic Cancer, American Cancer Society, available at: https: / / www.cancer.org / cancer / pancreatic-cancer / about / what-is-pancreatic-cancer.html). Rarely, cancers arise from cells that make pancreatic enzymes and are known as acinar cell carcinomas. Other rare types of exocrine cancers include adenosquamous carcinoma, squamous cell carcinoma, signet ring cell carcinoma, undifferentiated carcinoma, and undifferentiated carcinoma with giant cells. (Ibid.)

[0098] All pancreatic cancers are classified according to the TNM staging system, which is based on an assessment of the primary tumor, lymph node status, and the presence of metastatic disease (Ansari et al. Pancreatic Cancer: Yesterday, Today, and Tomorrow, Future Oncology 2016 Aug;12(16):1929-46). The TNM staging system classifies pancreatic cancer into stages 0, IA, IB, IIA, IIB, III, and IV. Stage 0 indicates a primary tumor with carcinoma in situ, no regional lymph node involvement, and no distant metastasis. Stage IA indicates a primary tumor confined to the pancreas, less than or equal to 2 cm (T1), no regional lymph node involvement (N0), and no distant metastasis (M0). Stage IB indicates a primary tumor confined to the pancreas, more than 2 cm (T2), no regional lymph node involvement (N0), and no distant metastasis (M0). Stage IIA indicates primary tumor extension beyond the pancreas but without involvement of the celiac artery or superior mesenteric artery (T3), without regional lymph node involvement (N0), and without distant metastasis (M0). Stage IIB indicates primary tumor progression at any level (T1-T3), regional lymph node involvement (N1), and without distant metastasis (M0). Stage III indicates primary tumor involvement in the celiac artery or superior mesenteric artery (unresectable primary tumor) (T4), lymph node involvement at any level (N0-N1), and without distant metastasis (M0). Stage IV indicates primary tumor at any level (T0-T4), lymph node involvement at any level (N0-N1), and distant metastasis (M1).

[0099] As used herein, "higher risk of developing pancreatic cancer" refers to a subject who is predisposed to or statistically more likely to develop pancreatic cancer than the general population due to factors that may include genetics, age, comorbidities, and the like. In some embodiments, an individual is at higher risk of developing pancreatic cancer because they have diabetes. The term "diabetes" as used herein refers to a chronic metabolic disease characterized by elevated levels of blood glucose. Diabetes can be classified as either type 1 or type 2. Type 1 diabetes, or insulin-dependent diabetes, is a chronic condition in which the pancreas produces little or no insulin. Type 2 diabetes occurs when the body becomes resistant to insulin or does not produce enough insulin. In some embodiments, an individual is at higher risk of developing pancreatic cancer because they have pancreatitis. In other embodiments, an individual is at higher risk of developing pancreatic cancer because of a family history of pancreatic cancer or pancreatitis.

[0100] In an exemplary embodiment, an individual is at higher risk of developing pancreatic cancer due to genetic mutation.As used herein, "high risk of developing pancreatic cancer due to genetic mutation" refers to an individual who has a DNA mutation that makes the development of pancreatic cancer statistically more likely than the general population.These genetic mutations can include any mutations known in the art to be specifically correlated with cancer or pancreatic cancer.Exemplary genes whose mutations are associated with pancreatic cancer include BRCA1, BRCA2, PALB2, TP53, MLH1, CDKN2A, and ATM.

[0101] In exemplary embodiments, the methods disclosed herein are used to select subjects for more invasive testing to confirm a cancer diagnosis (e.g., pancreatic cancer). As used herein, the term "more invasive testing" refers to any type of testing beyond detecting levels of miRNAs as described herein. In some embodiments, "more invasive testing" can include testing in which a subject's sample is collected and analyzed to detect cancer, such as a biopsy or blood draw. In some embodiments, "more invasive testing" includes endoscopy or other exploratory techniques to detect cancer. In some embodiments, "more invasive testing" includes imaging to detect cancer. Exemplary imaging techniques include magnetic resonance imaging (MRI), computed tomography (CT) scans, x-rays, positron emission tomography and computed tomography (PET-CT) scans, ultrasound, endoscopy, and nuclear scans.

[0102] In an exemplary embodiment, the method disclosed herein is used to select a subject for cancer investigation to monitor whether the subject develops cancer (e.g., pancreatic cancer). As used herein, the term "cancer investigation" can include the use of any of the above-mentioned "more invasive test" methods, which are repeated periodically. In some embodiments, the above tests are repeated in the following time frames: about once every 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 weeks. In some embodiments, the above tests are repeated in the following time frames: about once every 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, or 24 months.

[0103] In an exemplary embodiment, the method disclosed herein is used to select a subject for the treatment of cancer (e.g., pancreatic cancer). As used herein, the term "treating" or "treatment" refers to alleviating, reducing, or relieving at least one symptom in a subject, or resulting in the delay of disease progression. For example, treatment can be the reduction of one or more symptoms of a disorder, or the complete eradication of a disorder, e.g., cancer. In the sense of the present disclosure, the term "treat" also means stopping and / or reducing the risk of aggravation of a condition, disease, or disorder associated with or caused by the condition being prevented, or preventing at least one symptom. For example, treatment can alleviate, reduce, or alleviate at least one symptom of cancer (e.g., pancreatic cancer).

[0104] Surgery is an exemplary treatment for cancer (e.g., pancreatic cancer). Surgery for cancer localized to the head of the pancreas is called a Whipple procedure (pancreaticoduodenectomy). A surgery called a pancreatectomy can be performed to remove the left side of the pancreas. Another treatment for pancreatic cancer is a surgery to remove the entire pancreas, called a total pancreatectomy.

[0105] Chemotherapy is an exemplary treatment for cancer (e.g., pancreatic cancer). A variety of exemplary chemotherapeutic agents can be used to treat cancer (e.g., pancreatic cancer). Taxanes can also be used to treat pancreatic cancer, such as paclitaxel (Taxol®), docetaxel (Taxotere®), and albumin-bound paclitaxel (Abraxane®). Antimetabolites can be used to treat pancreatic cancer, such as gemcitabine hydrochloride (Gemzar® or Infugem®), 5-fluorouracil (5-FU or Adrucil®), or capecitabine (Xeloda®). Platinum chemotherapy can be used to treat pancreatic cancer, such as oxaliplatin (Eloxatin®). Alkylating agents can be used to treat pancreatic cancer, such as cisplatin (PLATINOL®). Agents that inhibit DNA replication can also be used to treat pancreatic cancer, including, for example, irinotecan (Camptosar®) and liposomal irinotecan (Onivyde®). PARP inhibitors can also be used to treat pancreatic cancer, including, for example, olaparib (Lynparza®). Antitumor chemotherapeutic agents can also be used to treat pancreatic cancer, including, for example, everolimus (Afinitor®), erlotinib hydrochloride (Tarceva®), sunitinib (Sutent®), mitomycin.

[0106] Drug combinations are exemplary treatments for cancer (e.g., pancreatic cancer). An exemplary drug combination is FOLFIRINOX (folinic acid, fluorouracil, irinotecan hydrochloride, and oxaliplatin). Another exemplary drug combination is GEMCITABINE-CISPLATIN (gemcitabine hydrochloride and cisplatin). Another exemplary drug combination is GEMCITABINE-OXALIPLATIN (gemcitabine hydrochloride and oxaliplatin). Yet another exemplary drug combination is OFF (oxaliplatin, fluorouracil, and folinic acid).

[0107] Radiation therapy is an exemplary treatment for cancer (e.g., pancreatic cancer). Radiation therapy utilizes high-energy beams, such as X-rays or protons, to destroy cancer cells. The radiation therapy used can be external beam radiation, where an external beam of radiation comes from a machine and directs the radiation to the patient's cancer. Alternatively, in internal radiation therapy, a radiation source, such as a solid or liquid, is injected into the patient's body. Chemotherapy and radiation can be used in combination, which is called chemoradiotherapy.

[0108] Immunotherapy is an exemplary treatment for cancer (e.g., pancreatic cancer). Immunomodulators such as pembrolizumab (Keytruda®) are a type of immunotherapy that can be used to treat pancreatic cancer.

[0109] In an exemplary embodiment, provided herein is a method for treating a subject suspected of having pancreatic cancer, comprising the steps of: (a) obtaining a sample collected from the subject; and (b) detecting in the sample hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-340 ... The method includes (a) detecting and quantifying one or more test microRNAs selected from the group consisting of miR-23a-5p, miR-190a-3p, and miR-26b-5p; (c) comparing the amount of the test microRNA determined in step (b) with a statistical model; and (d) selecting a subject for more invasive testing and / or investigation of pancreatic cancer based on the comparison in step (c), and optionally administering a treatment to the subject with pancreatic cancer.

[0110] In some embodiments, the treatment is a combination of various treatment methods described herein. An exemplary treatment method is a combination of chemotherapy and immunotherapy. Another exemplary treatment method is a combination of chemotherapy and radiation. Yet another exemplary treatment method is a combination of immunotherapy and radiation. Further exemplary treatment methods include the use of any of the treatment methods disclosed herein in combination with surgery (e.g., surgery and chemotherapy, surgery and immunotherapy, surgery and radiation, etc.).

[0111] kit Kits are also provided for carrying out the diagnostic and treatment methods disclosed herein. The kits may optionally further comprise instructions on how to use the various components of the kit.

[0112] In an exemplary embodiment, the kit comprises at least one test probe capable of specifically hybridizing to a microRNA selected from the group consisting of hsa-miR-192-5p, hsa-miR-98-5p, hsa-let-7g-5p, hsa-let-7f-5p, hsa-let-7a-5p, hsa-miR-122-5p, hsa-let-7d-5p, hsa-miR-340-5p, hsa-miR-194-5p, hsa-miR-323a-5p, hsa-miR-190a-3p, and hsa-miR-26b-5p, or its cDNA.

[0113] The kit of the present invention may include a carrier that is compartmentalized to accommodate one or more containers, such as vials, test tubes, ampoules, bottles, etc. in a sealed manner. Each such container includes a component or mixture of components as described herein (primers, test probes, normalization probes, test or normalization probes with non-natural labels (e.g., fluorescent labels such as TaqMan®, Scorpions®, and LightCycler®), fluorescent dyes (e.g., SYBR® Green), solvents or buffers, reagents for amplifying DNA sequences, reagents for reverse transcription of RNA molecules, etc.). In general, the kit may also contain one or more buffers, control samples, etc.

[0114] In certain embodiments, the kit comprises one or more containers containing the test probes disclosed herein. In some embodiments, the test probes contain detectable labels. In some embodiments, the kit comprises one or more containers containing normalization probes. In some embodiments, the normalization probes contain detectable labels.

[0115] In exemplary embodiments, the kit includes one or more containers containing reagents for reverse transcription of miRNA molecules. In some embodiments, the kit includes one or more containers containing a reverse transcriptase. In some embodiments, the kit includes one or more containers containing an oligo-dT primer. In some embodiments, the kit includes one or more containers containing dNTPs. In some embodiments, the kit includes one or more containers containing an RNase inhibitor. In some embodiments, the kit includes one or more containers containing a primer that hybridizes to RNA. In some embodiments, the kit includes one or more containers containing a buffer solution that provides a suitable chemical environment for the reverse transcriptase. In some embodiments, all reagents for reverse transcription of miRNA molecules are contained within a single container.

[0116] In exemplary embodiments, the kit comprises one or more containers containing reagents for qPCR. In some embodiments, the kit comprises one or more containers containing DNA polymerase. In some embodiments, the kit comprises one or more containers containing DNA binding dye. In some embodiments, the kit comprises one or more containers containing probes containing non-natural labels. In some embodiments, the kit comprises one or more containers containing dNTPs.

[0117] It will be readily apparent to those skilled in the art that other suitable modifications and adaptations of the methods described herein may be made using appropriate equivalents without departing from the scope of the embodiments disclosed herein. Now, certain embodiments have been described in detail, but the same will be more clearly understood by reference to the following examples, which are included for illustrative purposes only and are not intended to be limiting.

[0118] As used herein, the terms "comprising," "including," "having," and grammatical variations thereof are taken as specifying the stated features, integers, steps, or components, but do not exclude the addition of one or more additional features, integers, steps, components, or groups thereof. These terms encompass the terms "consisting of" and "consisting essentially of." EXAMPLES

[0119] [Example 1] Neural networks to generate diagnostic circulating miRNA signatures from human serum Summary: This example describes the development of a diagnostic test for pancreatic cancer detection that relies on miRNA expression, and an advanced AI-based algorithm that calculates the probability of disease through an artificial neural network. The method includes a set of 10 miRNAs that can be measured using miRNA sequencing or quantitative PCR (qPCR). For both methods, an appropriately weighted algorithm was prepared that uses the input of miRNA expression data and provides the user with the probability of the sample being from a patient with pancreatic cancer. The model was developed from 182 samples from Boston and Poland using miRNA-seq and validated on retested Polish samples and an additional 150 samples from Poland. Test performance was evaluated using samples from healthy patients and patients with pancreatitis, the most common differential diagnosis for pancreatic cancer. In both cases, samples were randomly split into sets used to train the classification model, and 20% of the samples were kept as an independent validation set to evaluate the performance of the model. Results of the miRNA-seq-based test showed a sensitivity of about 71% and a specificity of about 91%. The qPCR-based neural network slightly improved this performance with a sensitivity of about 76% and a specificity of about 92%. These diagnostic values ​​make it suitable for repeat testing in patients at high risk for pancreatic cancer, such as those diagnosed with diabetes.

[0120] To generate diagnostic circulating miRNA signatures from human serum, we assembled a study population of pre-treatment subjects with 182 patients from two independent cohorts. As shown in Figure 1, one cohort was located in the United States at the Dana-Farber Cancer Institute (DFCI) and the second cohort was located in Poland at the Lotz University of Medicine. Patients in the DFCI cohort had advanced stage pancreatic cancer (n=30) and were age- and sex-matched with 30 healthy controls. Polish patients had pancreatic cancer (n=44: 8 early, 27 advanced, 9 unknown), pancreatitis (n=28), or were clinically healthy (n=50). Microsequencing (miRNA-seq) was used to detect all previously described known and predicted miRNAs in ovarian cancer (Elias et al. Diagnostic Potential for a Serum miRNA Neural Network for Detection of Ovarian Cancer, Elife. 2017;6:e28932. Published October 31, 2017. doi:10.7554 / eLife.28932) or radiation exposure (Fendler et al. Evolutionarily conserved serum microRNAs predict radiation-induced fatality in nonhuman primates. Sci Transl Med. 2017;9(379):eaal2408. doi:10.1126 / scitranslmed.aal2408). As shown in Figure 1, patients were then randomly assigned to three groups: 1) a training set for variable selection and model development, 2) a test set for calibration of diagnostic cutoffs for classification models, and 3) a validation set for performance testing of diagnostic models on new data. A series of statistical tools, including machine learning approaches, were then deployed to analyze the miRNA-seq data and generate the best performing algorithm for discriminating between pancreatic cancer and pancreatitis patients and healthy controls. Once the final set of miRNAs was established and model calibration was performed, they were analyzed for validation subgroups.Despite the limited number of samples available, the performance of the neural network analysis exceeded 85% accuracy. Notably, both early and advanced cancers were identified with similar performance using the neural network approach.

[0121] Four reference miRNAs, shown in Table 1, were identified for qPCR validation purposes.

[0122] [Table 3]

[0123] Ten miRNAs shown in Table 2 were selected and used in the miRNA signature for pancreatic cancer diagnosis.

[0124] [Table 4]

[0125] Figure 2A-B shows the variable selection process for the training set described above. Ten miRNAs were selected as having a Family-Wise Error Rate (FWER) p-value < 0.05 (with Bonferroni adjusted p-value). The Volcano plot and result table show that for these 10 miRNAs, three were upregulated and seven were downregulated.

[0126] The 10 miRNAs were used to develop classification models using two main methods: logistic regression (with backward stepwise variable selection to reduce the number of miRNAs in the analysis) and artificial neural network (with sensitivity analysis used to reduce the number of miRNAs in the analysis).

[0127] The results of the logistic regression analysis are shown in Figure 3A-B. Figure 3A shows a plot of specificity versus sensitivity for the model miRNA models tested. Figure 3B shows the calculated values ​​of the four miRNAs used in the final model. The results were calculated using log based family-wise error rate (FWER) with a cutoff value of 50%. These results showed that the model fit the training and test sets very well with Hosmer Lemeshow value = 4.4927 and p-value = 0.810161. Figure 3C shows the sensitivity and specificity for detecting cancer in samples vs. controls. The final logistic regression model of the four miRNAs showed a sensitivity of 79.3% and a specificity of 84.1%.

[0128] The results of the artificial neural network are shown in Figures 4A-B. Figure 4A shows a plot of specificity versus sensitivity for the model miRNA models tested. The artificial neural network requires eight miRNAs: hsa-miR-192-5p, hsa-let-7a-5p, hsa-let-7d-5p, hsa-miR-194-5p, hsa-miR-98-5p, hsa-let-7f-5p, hsa-miR-122-5p, and hsa-miR-340-5p. Figure 4B shows the sensitivity and specificity for detecting cancer in samples versus controls. The final artificial neural network showed a sensitivity of 71.4% and a specificity of 90.9%.

[0129] Next, qPCR validation of the diagnostic model was completed, as shown in Figure 1. The results of the classification model were reproduced using qPCR. A set of 10 miRNAs and 4 reference miRNAs was quantified in the Polish samples used for miRNA-seq and in the additional 150 samples shown in Table 3 below. QPCR-based validation was performed using custom-made arrays in which the 10 miRNAs shown in Table 2 were selected to be significantly differentially expressed after Bonferroni correction. The 4 miRNAs shown in Table 1 were then selected as normalization factors using the normiRazor tool (Grabia et al. NormiRazor: Tool Applying GPU-Accelerated Computing for Determination of Internal References in MicroRNA Transcription Studies. BMC Bioinformatics 21, 425 (2020). https: / / doi.org / 10.1186 / s12859-020-03743-8).

[0130] [Table 5]

[0131] qPCR analysis included cycle threshold (Ct) preprocessing. Three preprocessing steps for the original Ct values ​​included: 1) background filtering (values ​​above the Ct value measured in the blank sample were treated as nondetections), 2) nondetection imputation using the expectation maximization (EM) algorithm, and 3) removal of duplicate Ct values ​​(the average Ct value from the two measurements was obtained). Finally, normalization (dCt calculation) was performed using the following formula: Cq = average of the top three normalization factors (hsa-miR-17-5p, hsa-miR-92a-3p, and hsa-miR-199a-3p).

[0132] Then, three different techniques were used to design the final diagnostic model. These techniques included logistic regression (with backward stepwise variable selection), an artificial neural network on the raw dataset, and an artificial neural network with synthetic minority oversampling technique (SMOTE) to balance the dataset. In this analysis, it was determined that miRNAs are able to distinguish cancer from controls, and clinical data including age and gender data were also used to improve the performance of the final model, in line with the objective of creating the best possible classification tool. The data is further divided into a training / test set to evaluate overfitting in the construction of the artificial neural network, and a validation set to evaluate the model performance in an unbiased manner. Finally, the SMOTE oversampling technique was used to balance the settings and improve the performance of the classification model.

[0133] As shown in Figure 5A, two sets of data sets of Polish samples were split for modeling. For the development of predictive models, the data sets were split into training, testing, and validation groups. As shown in Figure 5B, to counteract the imbalance problem, a balanced data set was created using the SMOTE technique for the training set, while the testing and validation data sets remained the same.

[0134] As shown in Figure 6A-B, logistic regression was used to evaluate the performance of the two miRNA (hsa-miR-192-5p and hsa-miR-194-5p) model on both the test and validation datasets. Figure 6B shows that the model has a sensitivity rate of 66% and a specificity rate of 74% in predicting cancer in observed cancer samples versus controls.

[0135] As shown in Figure 7A-B, the neural network model of classical (non-SMOTE-modified) data was tested using clinical data including age and gender, and all 10 miRNAs. The following miRNAs were used for normalization: hsa-miR-17-5p, hsa-miR-92a-3p, and hsa-miR-199a-3p. The results of this neural network model are shown in Figure 7A-B. Figure 7A shows that the area under the ROC curve (AUC) of the training set is 0.8475. Figure 7B provides values ​​of 82.57% accuracy, 59.72% sensitivity, and 93.84% specificity for the training and test datasets. Figure 7B provides values ​​of 83.02% for accuracy, 64.71% for sensitivity, and 91.67% for specificity for the validation dataset.

[0136] As shown in Figure 8A-B, the neural network model on the SMOTE balanced dataset was tested with the clinical dataset and the trimmed miRNA set including hsa-miR-192-5p, hsa-let-7a-5p, hsa-miR-194-5p, hsa-let-7f-5p, hsa-miR-122-5p, hsa-miR-340-5p, and hsa-miR-26b-5p. The following miRNAs were used for normalization: hsa-miR-17-5p, hsa-miR-92a-3p, and hsa-miR-199a-3p. Figure 8A shows that the AUC of the dataset is 0.8971. Figure 8B provides values ​​of 84.86% accuracy, 79.17% sensitivity, and 87.67% specificity for the training and testing datasets. FIG. 8B provides values ​​for accuracy of 86.79%, sensitivity of 76.47%, and specificity of 91.67% for the validation dataset.

[0137] These results show that the identified miRNA sets can be used to identify patients with pancreatic cancer and distinguish them from healthy controls or patients with pancreatitis with a sensitivity of 76-82% and a specificity of 83-91%, depending on the selection tool. The artificial networks can be used in combination or separately, depending on the a priori risk of pancreatic cancer and whether the physician prefers to use a miRNA-based test as a screening or confirmatory test.

Claims

**Claim 1**: A kit comprising a microRNA selected from the group consisting of hsa-miR-192-5p (SEQ ID NO: 5), hsa-miR-98-5p (SEQ ID NO: 6), hsa-let-7g-5p (SEQ ID NO: 7), hsa-let-7f-5p (SEQ ID NO: 8), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-122-5p (SEQ ID NO: 10), hsa-let-7d-5p (SEQ ID NO: 11), hsa-miR-340-5p (SEQ ID NO: 12), hsa-miR-194-5p (SEQ ID NO: 13), hsa-miR-323a-5p (SEQ ID NO: 15), hsa-miR-190a-3p (SEQ ID NO: 16), and hsa-miR-26b-5p (SEQ ID NO: 14), or at least one test probe capable of specifically hybridizing to the cDNA thereof. **Claim 2**: i) A test probe that specifically hybridizes to hsa-miR-192-5p (SEQ ID NO: 5) or the cDNA thereof, A test probe that specifically hybridizes to hsa-miR-98-5p (SEQ ID NO: 6) or the cDNA thereof, A test probe that specifically hybridizes to hsa-let-7g-5p (SEQ ID NO: 7) or the cDNA thereof, A test probe that specifically hybridizes to hsa-let-7f-5p (SEQ ID NO: 8) or the cDNA thereof, A test probe that specifically hybridizes to hsa-let-7a-5p (SEQ ID NO: 9) or the cDNA thereof, A test probe that specifically hybridizes to hsa-miR-122-5p (SEQ ID NO: 10) or the cDNA thereof, A test probe that specifically hybridizes to hsa-let-7d-5p (SEQ ID NO: 11) or the cDNA thereof, A test probe that specifically hybridizes to hsa-miR-340-5p (SEQ ID NO: 12) or the cDNA thereof, A test probe that specifically hybridizes to hsa-miR-194-5p (SEQ ID NO: 13) or the cDNA thereof, and A test probe that specifically hybridizes to hsa-miR-26b-5p (SEQ ID NO: 14) or the cDNA thereof including; or ii) A test probe that specifically hybridizes to hsa-miR-192-5p (SEQ ID NO: 5) or the cDNA thereof, A test probe that specifically hybridizes to hsa-let-7g-5p (SEQ ID NO: 7) or the cDNA thereof, A test probe that specifically hybridizes to hsa-let-7a-5p (SEQ ID NO: 9) or its cDNA, A test probe that specifically hybridizes to hsa-miR-122-5p (SEQ ID NO: 10) or its cDNA, A test probe that specifically hybridizes to hsa-miR-340-5p (SEQ ID NO: 12) or its cDNA, A test probe that specifically hybridizes to hsa-miR-194-5p (SEQ ID NO: 13) or its cDNA, and A test probe that specifically hybridizes to hsa-miR-26b-5p (SEQ ID NO: 14) or its cDNA comprising; or iii) A test probe that specifically hybridizes to hsa-miR-192-5p (SEQ ID NO: 5) or its cDNA, A test probe that specifically hybridizes to hsa-let-7g-5p (SEQ ID NO: 7) or its cDNA, A test probe that specifically hybridizes to hsa-let-7a-5p (SEQ ID NO: 9) or its cDNA, A test probe that specifically hybridizes to hsa-miR-122-5p (SEQ ID NO: 10) or its cDNA, A test probe that specifically hybridizes to hsa-miR-340-5p (SEQ ID NO: 12) or its cDNA, A test probe that specifically hybridizes to hsa-miR-194-5p (SEQ ID NO: 13) or its cDNA, and A test probe that specifically hybridizes to hsa-miR-26b-5p (SEQ ID NO: 14) or its cDNA comprising; or iv) A test probe that specifically hybridizes to hsa-miR-192-5p (SEQ ID NO: 5) or its cDNA, A test probe that specifically hybridizes to hsa-miR-98-5p (SEQ ID NO: 6) or its cDNA, A test probe that specifically hybridizes to hsa-let-7f-5p (SEQ ID NO: 8) or its cDNA, A test probe that specifically hybridizes to hsa-let-7a-5p (SEQ ID NO: 9) or its cDNA, A test probe that specifically hybridizes to hsa-miR-122-5p (SEQ ID NO: 10) or its cDNA, A test probe that specifically hybridizes to hsa-let-7d-5p (SEQ ID NO: 11) or its cDNA, A test probe that specifically hybridizes to hsa-miR-340-5p (SEQ ID NO: 12) or its cDNA, and A test probe that specifically hybridizes to hsa-miR-194-5p (SEQ ID NO: 13) or its cDNA Including, optionally, A test probe that specifically hybridizes to hsa-let-7g-5p (SEQ ID NO: 7) or its cDNA, or A test probe that specifically hybridizes to hsa-miR-26b-5p (SEQ ID NO: 14) or its cDNA Further comprising; or v) A test probe that specifically hybridizes to hsa-miR-192-5p (SEQ ID NO: 5) or its cDNA, and A test probe that specifically hybridizes to hsa-miR-194-5p (SEQ ID NO: 13) or its cDNA Including; Or vi) A test probe that specifically hybridizes to hsa-miR-192-5p (SEQ ID NO: 5), A test probe that specifically hybridizes to hsa-let-7a-5p (SEQ ID NO: 9) or its cDNA, A test probe that specifically hybridizes to hsa-miR-194-5p (SEQ ID NO: 13) or its cDNA, A test probe that specifically hybridizes to hsa-let-7f-5p (SEQ ID NO: 8) or its cDNA, A test probe that specifically hybridizes to hsa-miR-122-5p (SEQ ID NO: 10) or its cDNA, A test probe that specifically hybridizes to hsa-miR-340-5p (SEQ ID NO: 12) or its cDNA, and A test probe that specifically hybridizes to hsa-miR-26b-5p (SEQ ID NO: 14) or its cDNA Including, The kit according to claim 1.

3. The kit according to claim 1, comprising at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 test probes.

4. The kit according to claim 1 or 2, further comprising at least one normalization probe that can specifically hybridize to a microRNA selected from the group consisting of hsa-miR-17-5p (SEQ ID NO: 1), hsa-miR-199a-3p (SEQ ID NO: 2), hsa-miR-28-3p (SEQ ID NO: 3), and hsa-miR-92a-3p (SEQ ID NO: 4), or their cDNA.

5. (i) A normalization probe capable of specifically hybridizing to hsa-miR-17-5p (SEQ ID NO: 1) or its cDNA, a normalization probe capable of specifically hybridizing to hsa-miR-199a-3p (SEQ ID NO: 2) or its cDNA, a normalization probe capable of specifically hybridizing to hsa-miR-28-3p (SEQ ID NO: 3) or its cDNA, and a normalization probe capable of specifically hybridizing to hsa-miR-92a-3p (SEQ ID NO: 4) or its cDNA further comprised; or (ii) A normalization probe capable of specifically hybridizing to hsa-miR-17-5p (SEQ ID NO: 1) or its cDNA, a normalization probe capable of specifically hybridizing to hsa-miR-199a-3p (SEQ ID NO: 2) or its cDNA, and a normalization probe capable of specifically hybridizing to hsa-miR-92a-3p (SEQ ID NO: 4) or its cDNA further comprised, the kit according to Claim 1 or 2.

6. The kit according to Claim 1, comprising at least two or three kinds of normalization probes.

7. The kit according to Claim 1 or 2, not comprising a normalization probe.

8. The kit according to Claim 1 or 2, wherein at least one of the probes comprises a detectable label.

9. The kit according to Claim 8, wherein each of the probes comprises a detectable label.

10. The kit according to Claim 1 or 2, further comprising a reagent for reverse transcription of microRNA molecules.

11. A method for analyzing the possibility of having pancreatic cancer in a subject, (a) Detecting and quantifying one or more test microRNAs selected from the group consisting of hsa-miR-192-5p (SEQ ID NO: 5), hsa-miR-98-5p (SEQ ID NO: 6), hsa-let-7g-5p (SEQ ID NO: 7), hsa-let-7f-5p (SEQ ID NO: 8), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-122-5p (SEQ ID NO: 10), hsa-let-7d-5p (SEQ ID NO: 11), hsa-miR-340-5p (SEQ ID NO: 12), hsa-miR-194-5p (SEQ ID NO: 13), hsa-miR-323a-5p (SEQ ID NO: 15), hsa-miR-190a-3p (SEQ ID NO: 16), and hsa-miR-26b-5p (SEQ ID NO: 14) in the sample derived from the subject; (b) In a neural network, analyzing the amounts of the one or more test microRNAs quantified in step (a) to determine the probability that the subject has pancreatic cancer; (c) Assigning the subject as having a high likelihood of having pancreatic cancer based on the analysis of step (b) An analysis method for diagnosing pancreatic cancer in the subject, comprising:

12. Step (b) is i) hsa-miR-192-5p (SEQ ID NO: 5), hsa-miR-98-5p (SEQ ID NO: 6), hsa-let-7g-5p (SEQ ID NO: 7), hsa-let-7f-5p (SEQ ID NO: 8), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-122-5p (SEQ ID NO: 10), hsa-let-7d-5p (SEQ ID NO: 11), hsa-miR-340-5p (SEQ ID NO: 12), hsa-miR-194-5p (SEQ ID NO: 13), and hsa-miR-26b-5p (SEQ ID NO: 14); or ii) hsa-miR-192-5p (SEQ ID NO: 5), hsa-let-7g-5p (SEQ ID NO: 7), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-194-5p (SEQ ID NO: 13), hsa-miR-122-5p (SEQ ID NO: 10), hsa-miR-340-5p (SEQ ID NO: 12), and hsa-miR-26b-5p (SEQ ID NO: 14), or Optionally, hsa-miR-192-5p (SEQ ID NO: 5), hsa-let-7g-5p (SEQ ID NO: 7), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-194-5p (SEQ ID NO: 13), hsa-miR-122-5p (SEQ ID NO: 10), hsa-miR-340-5p (SEQ ID NO: 12), and hsa-miR-26b-5p (SEQ ID NO: 14); or iii) hsa-miR-192-5p (SEQ ID NO: 5), hsa-miR-98-5p (SEQ ID NO: 6), hsa-let-7f-5p (SEQ ID NO: 8), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-122-5p (SEQ ID NO: 10), hsa-let-7d-5p (SEQ ID NO: 11), hsa-miR-340-5p (SEQ ID NO: 12), and hsa-miR-194-5p (SEQ ID NO: 13), or Optionally, hsa-miR-192-5p (SEQ ID NO: 5), hsa-miR-98-5p (SEQ ID NO: 6), hsa-let-7f-5p (SEQ ID NO: 8), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-122-5p (SEQ ID NO: 10), hsa-let-7d-5p (SEQ ID NO: 11), hsa-miR-340-5p (SEQ ID NO: 12), hsa-miR-194-5p (SEQ ID NO: 13), and hsa-let-7g-5p (SEQ ID NO: 7), or hsa-miR-192-5p (SEQ ID NO: 5), hsa-miR-98-5p (SEQ ID NO: 6), hsa-let-7f-5p (SEQ ID NO: 8), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-122-5p (SEQ ID NO: 10), hsa-let-7d-5p (SEQ ID NO: 11), hsa-miR-340-5p (SEQ ID NO: 12), hsa-miR-194-5p (SEQ ID NO: 13), and hsa-miR-26b-5p (SEQ ID NO: 14) either; or iv) hsa-miR-323a-5p (SEQ ID NO: 15), hsa-miR-190a-3p (SEQ ID NO: 16), hsa-miR-192-5p (SEQ ID NO: 5), and hsa-let-7d-5p (SEQ ID NO: 11); or v) hsa-miR-192-5p (SEQ ID NO: 5) and hsa-miR-194-5p (SEQ ID NO: 13); or vi) Detecting and quantifying hsa-miR-192-5p (SEQ ID NO: 5), hsa-let-7a-5p (SEQ ID NO: 9), hsa-miR-194-5p (SEQ ID NO: 13), hsa-let-7f-5p (SEQ ID NO: 8), hsa-miR-122-5p (SEQ ID NO: 10), hsa-miR-340-5p (SEQ ID NO: 12), and hsa-miR-26b-5p (SEQ ID NO: 14) The method according to claim 11, comprising the step of detecting and quantifying

13. The method according to claim 11, wherein step (b) comprises the step of detecting and quantifying at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 of said test microRNAs

14. Step (a) further comprises the step of detecting and quantifying one or more normalization microRNAs selected from the group consisting of hsa-miR-17-5p (SEQ ID NO: 1), hsa-miR-199a-3p (SEQ ID NO: 2), hsa-miR-28-3p (SEQ ID NO: 3), and hsa-miR-92a-3p (SEQ ID NO: 4) in said sample; Step (b) further comprises the step of normalizing the amount of said test microRNA using the amount of said normalization microRNA The method according to any one of claims 11 to 13

15. Step (a) is i) hsa-miR-17-5p (SEQ ID NO: 1), hsa-miR-199a-3p (SEQ ID NO: 2), hsa-miR-28-3p (SEQ ID NO: 3), and hsa-miR-92a-3p (SEQ ID NO: 4) which are normalization miRNAs; or ii) hsa-miR-17-5p (SEQ ID NO: 1), hsa-miR-199a-3p (SEQ ID NO: 2), and hsa-miR-92a-3p (SEQ ID NO: 4) which are normalization miRNAs The method according to claim 14, comprising the step of detecting and quantifying

16. The method according to claim 14, wherein step (a) comprises the step of detecting and quantifying at least 2 or 3 of said normalization microRNAs

17. The method according to any one of claims 11 to 13, wherein step (a) is performed by detecting the binding of said sample to at least one probe capable of specifically hybridizing to each of said microRNAs or their cDNAs **Claim 18**: The method according to claim 17, wherein step (a) is performed using a nucleic acid detection assay, and optionally, the assay is selected from the group consisting of microarray, RT-PCR, and RT-qPCR. **Claim 19**: The method according to claim 18, wherein step (a) is performed using RT-qPCR. **Claim 20**: The method according to claim 17, wherein at least one of the probes comprises a detectable label. **Claim 21**: The method according to claim 20, wherein each of the probes comprises a detectable label. **Claim 22**: The method according to any one of claims 11 to 13, wherein step (a) is performed by reverse transcribing the microRNA molecules in the sample to obtain the cDNA sample, and sequencing the cDNA sample. **Claim 23**: The method according to claim 22, wherein step (a) further comprises amplifying the DNA molecules in the cDNA sample before sequencing the cDNA sample. **Claim 24**: The method according to claim 23, wherein step (a) is performed using miRNA-seq. **Claim 25**: The assignment of the subject as having a high likelihood of pancreatic cancer is i) a precision rate exceeding 50%, 60%, 70%, 80%, or 90%, optionally having a precision rate exceeding 80%; or ii) a specificity rate exceeding 50%, 60%, 70%, 80%, or 90%, optionally having a specificity rate exceeding 80%; or iii) a sensitivity rate exceeding 50%, 60%, 70%, 80%, or 90%, optionally having a sensitivity rate exceeding 80%, The method according to any one of claims 11 to 13. **Claim 26**: The method according to any one of claims 1, 2, and 11 to 13, wherein the sample is a blood sample or a pancreatic sample. **Claim 27**: The method according to claim 26, wherein the blood sample is selected from the group consisting of plasma, serum, and whole blood. **Claim 28**: The method according to any one of claims 1, 2, and 11 to 13, wherein the subject is a human subject. **Claim 29**: The method according to any one of claims 1, 2, and 11 to 13, wherein the subject has a higher risk of developing pancreatic cancer. **Claim 30**: The method according to any one of claims 1, 2, and 11 to 13, wherein the subject has diabetes. **Claim 31**: The method according to any one of claims 1, 2, and 11 to 13, wherein the subject has pancreatitis. **Claim 32**: The method according to any one of claims 1, 2, and 11 to 13, wherein the subject has a family history of pancreatic cancer or pancreatitis. **Claim 33**: The method according to any one of claims 1, 2, and 11 to 13, wherein the subject has a higher risk of developing pancreatic cancer due to gene mutations. **Claim 34**: The method according to claim 33, wherein the subject has a mutation in a gene selected from the group consisting of BRCA1, BRCA2, PALB2, TP53, MLH1, CDKN2A, and ATM. **Claim 35**: The method according to any one of claims 1 or 2, wherein the statistical model includes one or more models selected from the group consisting of linear discriminant analysis, logistic regression, multivariate adaptive regression splines, naive Bayes, neural network, support vector machine, decision tree, k-nearest neighbor method, functional tree, least absolute deviation (LAD) tree, Bayesian network, elastic net regression, and random forest. **Claim 36**: The method according to claim 35, wherein the statistical model includes a neural network or logistic regression.

Images