KR20220112450A - Composition for diagnosis of bile duct cancer using bile and ddPCR analysis and uses - Google Patents

Abstract

The present invention relates to a composition for diagnosing cancer or predicting prognosis, including a preparation capable of detecting a KRAS mutation in bile-derived circulating tumor DNA (ctDNA) using ddPCR, and a use thereof. The composition of the present invention can diagnose bile duct cancer or predict prognosis at a level similar to or superior to tissue biopsy by using bile, in particular, when ddPCR analysis is used, mutations in bile ctDNA can be stably detected, so that effects of diagnosing bile duct cancer or predicting prognosis can be significantly increased.

Description

Composition for diagnosis of bile duct cancer using bile and ddPCR analysis and uses thereof

The present invention relates to a composition for predicting cancer diagnosis or prognosis, including an agent capable of detecting a KRAS mutation in bile-derived circulating tumor DNA (ctDNA) using ddPCR (droplet digital polymerase chain reaction), and a use thereof .

Biliary tract cancers (BTC), including gallbladder cancer (GBC), ampullary carcinoma (AC), and extrahepatic cholangiocarcinoma (eCCA), are rapidly growing and fatal cancers, with 5-year survival rates of progressive or In less than 10% of metastatic biliary tract cancer patients, invasive tissue biopsy such as needle biopsy is used for diagnosis, but a non-invasive technique to detect tumor heterogeneity and molecular changes is required.

Liquid biopsy is very useful for cancer diagnosis, monitoring and prediction or prognosis. DNA fragments are cell-derived DNA, and circulating tumor DNA (ctDNA) circulating in the blood at 1 to 10 ng per 1 ml of plasma solution for normal individuals is DNA released from apoptosis, circulating or surviving tumor cells. The length is about 140 nt, and the half-life is It is about 1.5 hours. Blood has a low percentage of ctDNA, so high specificity is required for mutation detection. The accuracy of blood ctDNA diagnosis is not consistent between blood and tissue, and in particular, endoscopic or percutaneous tissue is 50% to 90% inaccurate, so ctDNA diagnosis of biliary tract cancer is very important.

KRAS is frequently mutated in tumors such as pancreatic cancer, colon cancer, and biliary tract cancer, but four locations (G12D, G12V, G13D, G12C) account for 83% of all KRAS mutations. Codon 12-position KRAS mutations further worsened the survival rate of patients with eCCA. Abnormalities in the RAS/RAF/MEK/ERK signaling pathway are also frequently observed in CCA.

NGS (Next Generation Sequencing), a method for detecting KRAS mutations, enables simultaneous mass confirmation, but has clinical limitations such as low gene detection, difficulty in distinguishing between errors and real data, and high cost. On the other hand, ddPCR (digital droplet PCR) can determine the number of copies of a specific region, and has many advantages over NGS such as short time required, low sample detection volume, and low cost. It is possible.

Republic of Korea Patent Application No. 10-2018-0110561 relates to a method for diagnosing biliary tract cancer. GCA (glycocholic acid) and TCDCA (Taurochenodeoxycholic acid) are used as biomarkers that can differentiate between patients with benign biliary disease, pancreatic cancer and biliary tract cancer. However, there have been no studies on a technology for diagnosing biliary tract cancer or predicting its prognosis by detecting KRAS mutations by ddPCR using bile.

The present inventors have completed the present invention in which bile duct cancer diagnosis and prognosis are predictable in order to provide a non-invasive method for diagnosing various biliary tract cancers with specificity similar to or more accurate than biopsy.

Accordingly, an object of the present invention is to provide a composition for diagnosing or predicting cancer, including an agent capable of detecting a KRAS mutation in bile-derived circulating tumor DNA (ctDNA), and use thereof.

The present invention provides a composition for diagnosing cancer or predicting prognosis, comprising an agent capable of detecting a KRAS mutation in bile-derived circulating tumor DNA (ctDNA).

According to a preferred embodiment of the present invention, the KRAS mutation may be any one or more selected from the group consisting of G12D, G12V, G13D and G12C.

According to a preferred embodiment of the present invention, the detection may be detection by ddPCR.

According to a preferred embodiment of the present invention, the cancer may be biliary tract cancer.

The present invention also provides a kit for diagnosing cancer or predicting a prognosis comprising the composition.

The present invention also comprises the steps of i) obtaining ctDNA from bile isolated from a subject;

ii) detecting KRAS mutations in ctDNA;

iii) determining the presence of KRAS mutations as cancer;

It provides an information providing method for cancer diagnosis or prognosis prediction comprising a.

The composition of the present invention can diagnose or predict biliary tract cancer to a degree similar to or superior to tissue biopsy using bile. In particular, when used for ddPCR analysis, it is possible to stably detect a mutation in bile ctDNA, thereby diagnosing or prognosing biliary tract cancer. The predictive effect can be significantly increased.

1 shows the bile KRAS ctDNA concentration (A) and the 2-year survival rate (B) of mutated or wild-type KRAS patients according to clinical stages. The 2-year survival rate was measured using the Kaplan-Meier model.
Figure 2a shows the size distribution of DNA fragments and histograms of KRAS mutations in bile (A), tissue (B) and plasma (C), and numbers indicate the size of the DNA.
2B shows a 2D plot of the results of ddPCR KRAS multiplex analysis with bile ctDNA in negative control (D), positive control (E) and biliary tract cancer (F) with KRAS mutations. Black clusters in the plot represent negative droplets, green clusters represent droplets positive only for wild-type DNA, blue clusters represent droplets positive only for mutant DNA, and orange clusters represent droplets positive for both mutant and wild-type DNA.
Figure 3 shows KRAS double analysis and ctDNA concentration in all bile samples. The blue line represents the cutoff value for KRAS mutation in bile (1500 copy/ml). The concentration of DNA mutations (variant DNA copies per droplet) was estimated with a Poisson distribution, and all panels show FAM amplitudes of up to 12000 and HEX amplitudes of up to 12000.
Figure 4 shows KRAS multiplex analysis and DNA concentration in bile and FFPE. The frequencies of G12D and G12V mutations among 20 patients with KRAS mutations in biliary ctDNA (B) and the frequencies of G12D and G12V mutations among 10 patients with KRAS mutations in FFPE tissues (C) are shown. Concentrations of mutant DNA (mutant DNA copies per droplet) were estimated with a Poisson distribution, and all panels show FAM amplitudes up to 12000 and HEX amplitudes up to 12000.
Figure 5 shows KRAS double assay and DNA concentration in bile and FFPE samples. Variation agreement in KRAS between bile and FFPE samples (N = 20) is shown. The blue line shows the cutoff value for KRAS mutation in bile (1500 copy/ml), and the orange line shows the cutoff value for FFPE (4500 copy/ml). Concentrations of variant DNA (mutant DNA copies per droplet) were estimated with a Poisson distribution, and all panels show FAM amplitudes up to 12000 and HEX amplitudes up to 12000.
FIG. 6A is a transcript sequencing result of bile, tissue, and plasma samples, and shows representative electrophoretic diagrams of juice, tissue, and plasma mRNA detected with the Agilent 2100 Bioanalyzer.
6B is a Venn diagram for the number of KRAS-associated mRNA expressed as a result of transcriptome sequencing of bile, tissue and plasma samples from BTC patients analyzed, respectively (B) and evaluated in bile, tissue and plasma samples (patient CA34). The distribution of the correlation matrix for the expression of 66 KRAS-related oncogenes is shown (C). Each circle represents the TPM (transcripts per million reads).

In the present invention, it was confirmed that a liquid biopsy of bile analyzed by ddPCR can be used for the detection of DNA mutations, and the expression of many genes encoding oncogenic signaling proteins, including KRAS, by transcriptome sequencing was confirmed in the bile and tumor tissues of BTC patients. was confirmed to be detected at the same level. This means that bile accurately reflects the biological and oncological properties of the tumor tissue. Clinically, mutant KRAS transcripts were frequently found in end-stage (III/IV) tumor bile, and these patients had significantly lower survival rates than patients with wild-type KRAS transcripts. Therefore, the present invention suggests that ctDNA analysis can use a liquid biopsy of bile instead of a tissue biopsy, and that the KRAS mutation can serve as a prognostic biomarker in BTC patients. In addition, ctDNA monitoring can confirm treatment response in real time and provide future therapies.

Although the genetic profile of ctDNA in bile and plasma can be evaluated in tumor tissues, the present applicant experimentally confirmed that the detection rate of ctDNA in plasma is very low. In several previous studies, the detection rate of ctDNA (20% to 75%) and the concordance rate of tissue and plasma results (25% to 79%) were not high. was only 48.0%. Also, the concordance rate of KRAS mutations detected in bile and tissue (80.0%) was higher than that of plasma and tissue (42.9%). Therefore, it was confirmed that bile is a better sample than blood for BTC patient ctDNA analysis. In the case of obstructive jaundice due to BTC, the concentration of ctDNA derived from the tumor tissue may be higher because the exposure frequency of the bile duct epithelium and cancer cells is high.

Bile collection requires an invasive procedure for endoscopic retrograde cholangiopancreatography or percutaneous transhepatic biliary drainage (PTBD). However, in the case of obstructive jaundice due to BTC or benign biliary stenosis, biliary decompression is often required, and it is still clinically difficult to distinguish BTC from benign biliary stenosis.

Hereinafter, the present invention will be described in more detail.

In order to check whether somatic cell mutation or liquid biopsy can replace tissue biopsy in biliary tract cancer (BTC), the present applicant has submitted ddPCR (formalin-fixed paraffin-embedded) and frozen plasma droplet digital polymerase chain reaction) was performed. Samples were taken from 42 patients with BTC, 20 patients for FFPE and 16 patients for plasma.

Bile separation for ctDNA analysis followed the protocol of Abi Zabron (Imperial College, London, England), and the bile was centrifuged at 16000 g at 4 °C for 10 minutes to obtain a supernatant (Bile). Bile was stored at -80 °C after centrifugation with Avanti J-25I (Beckman, CA, USA). Bile ctDNA isolation was isolated according to the manual of the NucleoSpin® cfDNA XS kit (GmbH & Co. KG). 270 μl of bile was used and analyzed with an Agilent 2100 Bioanalyzer.

Formalin fixed paraffin embedded (FFPE) was obtained from 20 patients who underwent surgical resection or endoscopic biopsy, and as a result of histological examination, it was confirmed that there were more than 30% of tumor cells. After FFPE (5mm) section, DNA was isolated according to the NucleoSpin® FFPE DNA Kit (GmbH & Co. KG) manual, and the DNA was stored at -20°C before analysis before use.

KRAS mutations were identified with bile ctDNA in 20 of 42 BTC patients (48%). Patients with KRAS mutations had worse survival than those with wild-type KRAS (2-year survival: 0% vs 55.5%, respectively; P = 0.018), the concordance between biliary ctDNA and FFPE DNA was 80.0%, and between plasma and FFPE 42.9%. The bile and FFPE tissue transcriptome sequencing results showed a high positive correlation between the expression of KRAS-related oncogenes in bile and tissues (r = 0.991, Po0.001).

Next generation sequencing (NGS) was performed on bile, tissue and plasma of a patient (CA34) for mRNA-Seq data analysis, and then, the detected gene expression profiles were compared and analyzed. Total RNA concentration was measured with Quant-iT™ RiboGreen™ RNA Assay Kit (Invitrogen) and DV200 (% of RNA fragments > 200 bp) values were measured with TapeStation RNA Screen Tape (Agilent Technologies, Inc.). Total RNA (100 ng) was used to construct a sequencing library with the SureSelectXT RNA Direct system (Agilent Technologies, Inc.) according to the manual. Total RNA was first fragmented with divalent cations at high temperature, and the cleaved RNA fragment was copied into first-strand cDNA using reverse transcriptase and random primers, followed by second-strand cDNA synthesis. These cDNA fragments were then ligated to adapters after a final repair process by adding a single 'A' base. For human exon capture, the Agilent SureSelect XT Human All Exon v6+UTRs kit (Agilent Technologies, Inc.) was used according to the standard Agilent SureSelect Target Enrichment protocol (Agilent Technologies, Inc.). 250 ng of cDNA library was mixed with hybridization buffer, blocking mix, RNase block and 5 μl of SureSelect XT Human All Exon v6+UTR capture library. Hybridization to capture baits was performed at 65 °C using the thermal cycler lid option heated at 105 °C for 24 h in a PCR thermal cycler. The captured library was washed and a second PCR amplification was performed. The final purified product was qPCR quantified according to the qPCR quantification protocol guide (KAPA Library Quantification kits for Illumina Sequencing platforms) and verified with TapeStation DNA Screen Tape D1000 (Agilent). The indexed library was Illumina NovaSeq (Illumina, Inc., San Diego, CA, USA). Paired-end (2×100 bp) sequencing was performed by Macrogen Incorporated (Seoul, Korea).

RNA isolated from each sample constituted a sequencing library according to the SMARTer smRNA-Seq Kit (Illumina Inc.) manual. Briefly, the input RNA was first polyadenylated to provide the priming sequence for the oligo (dT) primer. cDNA synthesis was primed with 3' smRNA dT primers incorporating adapter sequences at the 5' end of each RNA template to add non-template nucleotides. These nucleotides were bound by SMART smRNA Oligo and enhanced with locked nucleic acid (LNA) technology for greater sensitivity. In the template conversion step, PrimeScript RT uses the SMART smRNA Oligo as a template for adding a second adapter sequence to the 3′ end of each first strand cDNA molecule. Next, add a full-length Illumina adapter (with index sequence for sample multiplexing) during PCR amplification. Forward PCR primers bind to sequences added by SMART smRNA Oligo while reverse PCR primers bind to sequences added by 3' smRNA dT primers. The library cDNA molecules contained the sequences required for clustering on the Illumina flow cell. Libraries were gel purified with blue pippin and analyzed for size, purity and concentration in an Agilent Bioanalyzer. Libraries were pooled in equal molar amounts and sequenced on an Illumina HiSeq 2500 instrument to generate 51 primary reads. Image decomposition and quality value calculations were performed using modules in the Illumina pipeline.

In order to remove low quality and adapter sequences before analysis, raw reads were preprocessed in sequencing and processed reads were aligned to Homo sapiens genome assembly (GRCh38) using HISAT v2.1.0(22). HISAT uses two types of indexes for alignment: global, whole-genome indexes, and tens of thousands of small local indices. These two types of index are constructed using the same Burrows-Wheeler transform (BWT) and graph FM index (GFM) as Bowtie2. Because of the use of these efficient data structures and algorithms, HISAT generates junctional alignments several times faster than the widely used Bowtie and BWA. The reference genome sequence and annotation data of Homo sapiens (GRCh38) were downloaded from NCBI. Assembly of known transcripts was processed with StringTie v2.1.3b (23 [Pertea, 2016 # 1074,24). Based on these results, transcripts and expressed genes were counted as the number of reads per sample.

The present invention provides a composition for diagnosing cancer or predicting prognosis, comprising an agent capable of detecting a KRAS mutation in bile-derived circulating tumor DNA (ctDNA).

The term “diagnosis” of the present invention, in a broad sense, means judging the actual condition of a patient's disease in all aspects. The content of the judgment is the disease name, etiology, disease type, severity, detailed mode of the disease, and the presence or absence of complications. Diagnosis in the present invention is preferably to determine the risk of cancer and its development.

The term “prognosis” of the present invention refers to a prospect or preliminary evaluation of the medical outcome of a condition, for example, predicting a poor or good outcome (eg, the likelihood of long-term survival). Negative prognosis or poor outcome includes prediction of recurrence, disease progression (eg, cancer growth or metastasis, or drug resistance), or likelihood of death, and positive prognosis or good outcome includes cure of disease (eg, disease absence), remission (eg, combating cancer), or prediction of stabilization. In the present invention, the prognosis means cancer recurrence, overall survival, or disease-free survival. Prediction of prognosis (or diagnosis of prognosis) can provide clues about future cancer treatment, particularly whether early cancer patients are treated with chemotherapy. Prognosis prediction also includes prediction of the patient's response to cancer treatment and the course of treatment.

According to a preferred embodiment of the present invention, the KRAS mutation may be any one or more selected from the group consisting of G12D, G12V, G13D and G12C.

According to a preferred embodiment of the present invention, the detection may be detection by ddPCR.

According to a preferred embodiment of the present invention, the cancer may be biliary tract cancer.

The present invention also provides a kit for diagnosing cancer or predicting a prognosis comprising the composition.

The kit of the present invention may be composed of one or more other component compositions, solutions or devices suitable for commonly used expression level analysis methods. For example, a kit for measuring protein expression may include a substrate, a suitable buffer, a secondary antibody labeled with a chromogenic enzyme or a fluorescent substance, a chromogenic substrate, and the like for immunological detection of the antibody.

The kit of the present invention may include a sample extraction means for obtaining a sample from the subject to be evaluated. The sample extraction means may include a needle or syringe or the like. The kit may include a sample collection container for receiving the extracted sample, which may be liquid, gaseous or semi-solid. The kit may further include instructions for use. The sample may be any body sample in which a KRAS mutation may be present. For example, the sample can be urine, feces, hair, sweat, saliva, bodily fluid, blood, cells or tissue. Determination of KRAS mutations in body samples can be performed on whole samples or on processed samples.

The present invention also comprises the steps of i) obtaining ctDNA from bile isolated from a subject;

ii) detecting KRAS mutations in ctDNA;

iii) determining the presence of KRAS mutations as cancer;

It provides an information providing method for cancer diagnosis or prognosis prediction comprising a.

Hereinafter, the present invention will be described in more detail through examples. These examples are only for illustrating the present invention, and it is obvious to those of ordinary skill in the art that the scope of the present invention is not to be construed as a limitation due to these examples.

patient and bile collection

From August 2018 to March 2020 at Keimyung University Dongsan Hospital Bile was collected from 42 patients with biliary tract cancer (eCCA) and one patient with benign biliary tract disease (distal bile duct (CBD) stones). The diagnosis of biliary tract cancer was based on tissue diagnosis from endoscopic retrograde cholangiopancreatography (ERCP) or surgical resection by percutaneous transhepatic biliary drainage (PTBD) or needle biopsy. Patients with other malignancies or with pancreatic or hepatocellular carcinoma on tissue biopsy were excluded. Finally, 42 patients with pathologically proven biliary tract cancer were screened. Therapeutic surgical resection was completed when the eCCA patient isolated the CCA without metastatic lesions and the general condition was tolerable. However, chemoradiation was performed when the patient had metastatic lesions or refused surgical resection. Tumor stages were classified according to the 7th edition of the American Joint Cancer Commission (AJCC) Classification of Malignant Tumors, and the AJCC stages were classified according to clinical (patients treated with chemoradiation or conservative therapy) or pathological data (therapeutic surgical resection). was evaluated based on patients treated with This prospective study was approved by the institutional review committee of Keimyung University Dongsan Medical Center (DSMC 2018-06-054) and included secondary use of human-derived substances. To relieve cholangitis or obstructive jaundice (in the case of biliary tract cancer) and to relieve cholangitis by removing CBD stones (in the case of the control group), bile from two groups was collected during PTBD. After resolution of cholangitis in both groups, bile was collected at least 3 days after PTBD procedure. FFPE samples from 20 of 42 biliary tract cancer patients were used. Before double surgery, plasma from 16 patients was collected and stored at -80°C before use. The collected bile was centrifuged at 16,000 g for 10 min at 4 °C to obtain a supernatant (Bile), and each pellet was resuspended in cold PBS (2x) according to the protocol of Abi Zabron (Imperial College, London, England). Then, 16,000 g of bile pellets were obtained after centrifugation at 4° C. for 5 minutes. All experiments were centrifuged with Avanti J-25I (Beckman, CA, USA), and bile and pellets were stored at -80 °C and used in all experiments. Specific patient characteristics are shown in Table 1 below (AJCC; American Joint Committee on Cancer, eCCA; extrahepatic cholangiocarcinoma, GBC; gallbladder cancer, AC; ampulla vater cancer).

Bile (N = 42) FFPE (N = 20) Plasma (N = 16)
wild type
(N=22) KRAS mutation
(N=20) P wild type
(N=10) KRAS mutation
(N=10) P wild type
(N=13) KRAS mutation
(N=3) P age
(Mean, SD) 74.4±9.6 71.9±9.5 0.408 75.3±8.3 65.7±8.8 0.022 73.1±8.3 74.3±1.5 0.802 Gender (Male,%) 10 (45.5) 13 (65.0) 0.337 6 (60.0) 7 (70.0) 0.774 7 (53.8) 2 66.7) 1.000 AJCC Steps
(I, Ⅱ/Ⅲ, Ⅳ) 5 (22.7)/17 (77.3) 3 (15.0)/17 (85.0) 0.808 3(30.0)/7(70.0) 5(50.0)/5(50.0) 0.478 7(53.8)/6(46.2) 0(0.0)/3(100.0) 0.238 cancer type 0.544 0.589 0.809 eCCA 19 (86.4) 17 (85.0) 9 (90.0)) 8 (80.0) 12 (92.3) 3 (100.0) GBC 3 (13.6) 2 (10.0) 1 (10.0) 1 (10.0) 0 (0.0) 0 (0.0) AC 0 (0.0) 1 (5.0) 0 (0.0) 1 (10.0) 1 (7.7) 0 (0.0) How to obtain bile 0.476 1.000 0.809 Operation 10 (45.5) 6 (30.0) 7 (70.0) 8 (80.0) 12 (92.3) 2 (66.7) Palliative care 12 (54.5) 14 (70.0) 3 (30.0) 2 (20.0) 1 (7.7) 1 (33.3)

BTC patients were 23 male and 19 female, with a mean age of 73.2 ± 9.5 years (range 50-88 years). Of the BTC (N = 42) patients, 5 had GBC, 1 had AC, and 36 had eCCA. Among 42 patients, FFPE tumor tissue was measurable in 20 patients, and frozen silver plasma collected before surgery was measurable in 16 patients. 16 patients underwent curative surgical resection and 26 patients underwent palliative chemoradiotherapy or conservative therapy. The mean concentration of bile ctDNA in late stage (III/IV) stage BTC was higher than in early stage (I/II) stage BTC, and was significant (16,100 ± 18,700 vs 2300 ± 800 copy/ml, respectively, P = 0.083); 1A) is shown.

In addition, in Kaplan-Meier model analysis, patients with KRAS mutations in bile ctDNA had worse survival rates than patients with wild-type KRAS (2-year survival rates: 0% vs. 55.5%, respectively, P = 0.018) (Fig. 1B) . These results indicated that KRAS mutation in bile ctDNA can be used as a prognostic biomarker for BTC.

DNA fragment size analysis and KRAS mutation histogram

To compare the sizes of ctDNA molecules in bile, plasma and FFPE tissues, and to evaluate mutated genes, ddPCR (Droplet digital PCR) KRAS duplex assay was applied to bile, plasma and tissues.

ddPCR was performed with a DNA library in the Bio-Rad QX200 ddPCR system using the ddPCR™ KRAS Screening Multiplex Kit and the ddPCR™ KRAS G12/G13 Screening Kit (Bio-Rad Laboratories; Hercules, CA, USA). 2D plots generated with QuantaSoft™ software show mixed mutations and wild-type. The 2D plot is the result of ddPCR KRAS multiplex analysis applied to bile-derived ctDNA from negative control NTC (non-template control), positive control, and BTC patients. Amplification products were detected using a C1000 Tough Thermal Cycler (Bio-Rad Laboratories). The reaction was incubated at 95 °C for 10 min, then repeated 40 cycles at 94 °C for 30 sec, 55 °C for 60 sec, and 98 °C for 10 min. The size distribution of bile ctDNA and FFPE DNA was analyzed with an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.).

As a result, as shown in [Fig. 2a] and [Fig. 2b], most of the DNA fragments in bile were less than 100bp and a periodic peak of 10bp was observed (Fig. 2a A, 35-70bp). FFPE (0-300 bp)DNA had a larger peak than the bile DNA fragment (Fig. 2a B), and almost no peak was observed in plasma (Fig. 2a C). Histograms of bile negative control, positive control and BTC mutations are shown in [Fig. 2b] D-F. All figures are shown in scale of 12,000 (FAM) on the y-axis and 12,000 (HEX) on the x-axis.

Whether KRAS mutations are consistent between ctDNAs

The degree of KRAS DNA mutation (mutant DNA copy per droplet) was estimated by the Poisson distribution. The average concentration of mutant DNA in the bile ctDNA of BTC patients (14,995 copy/ml) was increased about 10-fold compared to the positive control group of patients with CBD absent (1000 copy/ml; FIG. 3A), and the KRAS mutation cutoff value was > 1500 copy/ml. ml (Table 2). KRAS mutations were identified in the bile ctDNA of 20 (48%) of BTC patients (N = 42) by dual multiplex assay, and G12D, G12V, G13D, G12A, G12C, G12R and G12S were included. Among the 20 patients with KRAS mutations in bile ctDNA, the frequencies of G12D and G12V were 20/20 (100%) and 9/20 (45%), respectively (Fig. 4B).

The average concentration of FFPE DNA (24,515 copy/ml) was increased about 5 times compared to the positive control group (4100 copy/ml). Therefore, the cutoff value for the KRAS mutation was set to > 4500 copy/ml (Table 2). The KRAS mutation in FFPE was confirmed to be G12D or G12V in most of 10 patients (Fig. 4C).

We assessed variation agreement by comparing the concentration of bile ctDNA with the concentration of 20 pairs of FFPE DNA. There was 80% agreement between the same bile ctDNA and FFPE (16/20) (Fig. 5D). As a result, mutation agreement between bile ctDNA and FFPE DNA was confirmed in biliary tract cancer, indicating that KRAS mutation had a significant effect on prognosis.

The mutation concentration of plasma KRAS ctDNA was very low, and KRAS mutation was detected only in 3 samples (18.8%) with a cutoff value of 60 copy/ml (Table 2). Of the 16 plasmas, 14 samples were identical to the FFPE, and the variation agreement was 42.9% (6/14).

Bile (N=42) FFPE (N=20) Plasma (N=16) N(%) DNA concentration N(%) DNA concentration N(%) DNA concentration KRAS multiple mutations 20 (47.6) 14995.0 ± 17602.6 10 (50.0) 24515.0 ± 17128.9 3 (18.7) 12.5 ± 26.9 KRAS point mutation p. G12D 20 (100.0) 7917 ± 15436.7 9 (90.0) 6982.0 ± 14951.6 1 (33.3) 4.3 ± 17.5 p. G12V 9 (45.0) 1775.5 ± 3782.3 10 (100.0) 5647.0 ± 8185.9 3 (100.0) 11.8 ± 25.6 p. G12D + p. G12V 20 (100.0) 9692.5 ± 15287.9 10 (100.0) 12632.0 ± 15207.5 3 (100.0) 16.2 ± 37.5

Transcriptome sequencing of the same patient bile, plasma and tissue

Transcriptome sequencing was performed to confirm the level of mDNA expression in bile, tissue and plasma of the same patient.

Template size distribution was confirmed with an Agilent Technologies 2100 Bioanalyzer using a DNA 1000 chip (RNA fragment > 200 bp). Poorly expressed mRNAs with less than 10 transcripts per million read in bile, tissue and plasma from the same patient were excluded from the analysis. Gene annotations for Homo sapiens (GRCh38; relaease 109.20190607) were retrieved from the National Center for Biotechnology Information (NCBI). The profile of KRAS gene expression was determined by RNA-Seq by Expectation-Maximization (RSEM) (v1.3.1) with the following options: estimate, rspd; seed-length, 15; strandedness, forward.

Representative electrophoretic diagrams for bile, tissue and plasma mRNAs from the same patient are shown in FIG. 6A . A total of 46,425 genes and 66 KRAS-related genes including RAS and RAF were observed in three groups. A Venn diagram showing the number of KRAS genes and 57.6% of KRAS-related oncogenes were expressed in both bile and tissue (Fig. = 0.985, P < 0.001) or tissue and plasma (r = 0.991, P < 0.001) showed a strong positive correlation (r = 0.991, P < 0.001) ( FIG. 6B C ). This result demonstrates KRAS gene expression detected in bile by mRNA-Seq. In addition, many oncogenes expressed in BTC, such as BRAF and NRAS, were also detected in bile (data not shown). These data suggest that both gene expression and mutation can be detected in bile, suggesting that bile reflects the gene expression profile of BTC tissues.

[statistical analysis]

All statistical analyzes were performed with SigmaStat statistical software (ver. 14.0-ESD, USA). Continuous values for statistical analysis results were expressed as mean ± SD. For comparison between the two groups, an independent t test was used for continuous variables and a χ ² test was used for categorical variables. Differences with a P value less than 0.05 were considered statistically significant.

Claims (6)

A composition for diagnosing cancer or predicting prognosis, comprising an agent capable of detecting a KRAS mutation in bile-derived circulating tumor DNA (ctDNA).
The composition of claim 1, wherein the KRAS mutation is at least one selected from the group consisting of G12D, G12V, G13D and G12C.
The composition for diagnosing cancer or predicting a prognosis according to claim 1, wherein the detection is performed by ddPCR.
The composition for diagnosis or prognosis of cancer according to claim 1, wherein the cancer is biliary tract cancer.
A kit for diagnosing cancer or predicting a prognosis comprising the composition of claim 1.
i) obtaining ctDNA from bile isolated from a subject;
ii) detecting a KRAS mutation in the ctDNA;
iii) determining the presence of a KRAS mutation as cancer;
Information providing method for cancer diagnosis or prognosis prediction comprising a.

Images