Method for diagnosing minimal residual focus by detecting structural variation sequence of cfDNA

Abstract

The present invention relates to a novel WGS (Whole Genome Sequence) analysis system and method for the diagnosis of minimal residual lesions in which a structural variation sequence is detected with high sensitivity using a cfDNA sample derived from a cancer patient, and the cancer patient is diagnosed using the detected structural variation data. In the present invention, even when a cancer cell line and NA12878 are mixed at a ratio of 1: 12, 800, a structurally variable sequence has been detected, and a structurally variable sequence has been sensitively confirmed in cfDNA of a metastatic lung cancer patient. Therefore, by means of the method for detecting a structural variation sequence in a cfDNA sample derived from a cancer patient according to the present invention, a specific structural variation sequence of a patient can be detected with high sensitivity even if cancer-derived cfDNA is present in blood or the like at a low purity, and after cancer treatment, the specific structural variation sequence of the patient can be detected with high sensitivity. As a result, cancer cells remaining on a patient can be detected even with a small amount of sample, and the method can be generally used in diagnosis of minimal residual lesions regardless of cancer types.

Description

Technical Field

[0001] The present invention relates to a new whole-genome sequencing (WGS) system and method for detecting structural variation sequences in cfDNA (Cell-free DNA) samples from cancer patients with high sensitivity, and using the same to diagnose minimal residual disease (MRD) in cancer patients.

[0002] This application claims priority based on Korean Patent Application No. 10-2022-0131806 filed on October 13, 2022, and all contents disclosed in the specification and drawings of this application are incorporated herein by reference. Background Art

[0003] Next Generation Sequencing (NGS) is a high-speed genome analysis method that divides the genome into numerous fragments, reads them, and compares the obtained base sequence fragments to analyze the genome sequence. Whole-genome sequencing (WGS) performed by NGS is practical for detecting almost all types of somatic variants, and is widely used in various fields due to this practicality, especially playing a very important role in cancer genomics.

[0004] Gene sequencing is developing rapidly around the world, and this second-generation gene sequencing technology is also being actively used in the fields of clinical genomics, pharmacogenomics, and translational medicine.

[0005] Structural variants (SV) play an important role in the development of cancer, and therefore, various bioinformatics algorithms and tools have been developed to detect somatic structural variants in cancer genomes. Structural variants that occur when specific locations in the genome are cut and combined with each other generate fusion DNA sequences at the binding site, which are specific intrinsic sequences that cannot be confirmed in normal tissues. In cancer cells, cfDNA is known to flow into the blood due to apoptosis, and if fusion DNA sequences caused by specific structural mutations of primary cancer are detected in cfDNA, it is expected that it can be used to diagnose minimal residual lesions (MRD) remaining in the body after cancer treatment.

[0006] Targeted high-depth sequencing has been widely used as a method for discovering mutations in cfDNA. This method has the advantage of being able to sensitively discover mutations, but in order to achieve high-depth, it is necessary to prepare a targeted panel suitable for individual cancer types and an experimental pre-processing step to reduce noise.

[0007] Therefore, a new method is needed to effectively detect individual cancer-specific structural variation sequences using WGS of cfDNA samples, which can be generally used to diagnose minimal residual cancer lesions. Summary of the invention

[0008] Technical issues

[0009] In order to sensitively detect structural variation sequences in cfDNA samples, the inventors used WGS of primary cancer to search for structural variations, and by matching the sequences of the structural variations in the WGS of cfDNA, reads with a small amount of structural variation sequence information can be detected in cfDNA, confirming that they can be used for the diagnosis of minimal residual lesions in patients, thereby completing the present invention.

[0010] Technical Solution

[0011] In order to achieve the above-mentioned purpose, the present invention provides a method for detecting structural variation sequences in cfDNA samples derived from cancer patients, comprising the following steps: 1) analyzing whole genome sequencing (WGS) of cancer tissues derived from patients by using two or more structural variation analysis software to obtain consensus structural variation data of commonly confirmed primary cancer samples; and 2) matching the reference sequence of the structural variation site obtained in the step 1) with the cfDNA whole genome sequencing (WGS) of the corresponding patient, and obtaining supporting reads of the structural variation sequence present in the consensus structural variation site, thereby matching the specific sequence generated by the structural variation present in the cfDNA sample.

[0012] In addition, the present invention provides a method for providing information on minimal residual disease (MRD), and the device includes the following steps: 1) analyzing whole genome sequencing (WGS: Whole-Genome Sequencing) of cancer tissues derived from patients by using two or more structural variation analysis software to obtain consensus structural variation site data of commonly confirmed primary cancer samples; and 2) matching the reference sequence of the structural variation site obtained in the step 1) with the cfDNA whole genome sequencing obtained after cancer treatment of the corresponding patient, and obtaining the structural variation sequence supporting reads present in the consensus structural variation site, thereby matching the specific sequence generated by the structural variation present in the cfDNA sample.

[0013] Technical Effects

[0014] In the present invention, even when the cancer cell line and NA12878 were mixed at a ratio of 1:12,800, the structural variation sequence was detected, and the structural variation sequence was also sensitively confirmed in the cfDNA of the metastatic lung cancer patient. Therefore, if the method for detecting the structural variation sequence in the cfDNA sample derived from the cancer patient according to the present invention is used, even if the cfDNA derived from the cancer is present in the blood or the like at a low purity, the patient's specific structural variation sequence can be detected with high sensitivity, and after cancer treatment, the cancer cells remaining in the patient's body can be detected even with a small amount of sample, and therefore, it can be generally used for the diagnosis of tiny residual lesions regardless of the type of cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1: is a diagram showing a flow chart of the method for detecting structural variation sequences in cfDNA of the present invention.

[0016] Figure 2 Schematic diagram of the process of confirming the structural variation sequence by comparing the corresponding reference sequence with each supporting read segment, which is performed in the second step of the detection method of the present invention.

[0017] Figure 3 This is a diagram schematically showing the analysis and verification results using a cell line mixing experiment.

[0018] Figure 4 and Figure 5 The results show that in the clinical validation experiment, the structural variation confirmed in the primary cancer tissue derived from the patient in case 1 was confirmed in both cfDNA-1 and cfDNA-2 samples, but the structural variation was not observed in the normal control group.

[0019] Figure 6 The figure shows the results of detecting KRAS mutation (36%) in the primary cancer tissue of case 1 patient, and detecting the same mutation in cfDNA-1 and cfDNA-2 samples at 19% and 6%, respectively.

[0020] Figure 7 and Figure 8 The results show that in the clinical validation experiment, the structural variation confirmed in the primary cancer tissue from case 2 patient was also confirmed in cfDNA-1 and cfDNA-2 samples, but the structural variation was not observed in the normal control group.

[0021] Fig. 9 This figure confirms that TP53 mutation (45%) was detected in the primary cancer tissue of case 2 patient, and the same mutation was also detected in cfDNA-1 and cfDNA-2 samples, with the results being 2% and 0%, respectively.

[0022] Fig.10 This figure confirms that SMARCA4 mutation (27%) was detected in the primary cancer tissue of case 2 patient, and the same mutation was also detected in cfDNA-1 and cfDNA-2 samples, with the results being 5% and 2%, respectively. DETAILED DESCRIPTION

[0023] The present invention relates to a method for detecting structural variation sequences in cfDNA samples derived from cancer patients and a method for using the method to provide information on minimal residual disease (MRD).

[0024] According to the present invention, structural variations pre-confirmed from primary cancer tissues of cancer patients can be used to quickly compare and analyze structural variation sequences present in cfDNA of cancer patients after treatment with high sensitivity, thereby monitoring cancer cells remaining in the patient's body after treatment and providing information on the cancer treatment process, prognosis, possibility of recurrence, etc.

[0025] Hereinafter, the present invention will be described in detail.

[0026] The present invention relates to a method for detecting structural variation sequences in cfDNA samples derived from cancer patients, comprising the following steps: 1) obtaining consensus structural variation data of commonly confirmed primary cancer samples by analyzing WGS of cancer tissues derived from patients using two or more structural variation analysis software; and 2) matching the reference sequence of the structural variation site obtained in step 1) with the cfDNA WGS of the corresponding patient to obtain supporting reads of the structural variation sequence present at the consensus structural variation site, thereby matching specific sequences generated by the structural variation present in the cfDNA sample.

[0027] Step 1) of the present invention is a step of obtaining consensus structural variation data of primary cancer samples by analyzing WGS of cancer tissues derived from patients and WGS of normal control groups using two or more structural variation analysis data.

[0028] In the step, it is characterized in that, in order to obtain more accurate structural variation data of patients, more than two structural variation analysis software are used. In this step, WGS and alignment are performed using cancer tissue (i.e., primary cancer sample) obtained from cancer patients and normal control group, and the mapping (Mapping) of the position of sequencing reads on the reference genome is performed to confirm which chromosome and which site the sequencing reads are located at. If the mapping is completed, the chromosome number and site information on the reference genome for each sequencing read (sequencing read) are provided, and the BAM (binary alignment map) format of the aligned base fragments including these contents can be obtained. Using the first structural variation analysis software, "structural variation call (SV call: Structural variant call)" is performed, and the structural variation call is confirmed by comparing and analyzing each sequencing read in the "tumor BAM" file obtained from the tumor tissue and the "control group BAM" file. The presence of a variant of a structural variation generated at a specific site that is different from the reference genome sequence, thereby obtaining "first structural variation site data".

[0029] Next, in order to obtain the "consensus structural variation site data" of the present invention, the second structural variation analysis software is used to perform structural variation calling. In this case, in order to quickly confirm the consensus structural variation, the "first structural variation site data" obtained by using the first structural variation analysis software can be input into the second structural variation analysis software to perform structural variation calling.

[0030] That is, in the present invention, "consensus structural variation site data" refers to information corresponding to structural variation sites confirmed to be commonly present in SV calls by multiple structural variation analysis software.

[0031] The step 1) can be used for the purpose of constructing a database (DB) for structural variations present in primary cancer samples of cancer patients. As described above, for the consensus structural variation site data that has been obtained once, it can be repeatedly used for the purpose of monitoring when comparing and analyzing the cfDNA structural variations of the same cancer patient obtained from multiple subsequent time periods.

[0032] The structural variation analysis software used in the present invention can use a variety of software known in the art that can detect somatic structural variations by whole genome sequencing without restriction, for example, two or more selected from the group consisting of DELLY, BRASS, SvABA, dRanger, Pindell, BreakDancer, GASV, Hydra, CNV nator and JuLI. Two or more of the analysis software can be used in sequence, and the first WGS analysis software analyzes each sequencing read by structural variation calling (SV call) to confirm the presence of a variant of a structural variation different from the reference genome sequence generated at a specific position, and provides "first structural variation position data". Then, if the first structural variation site data obtained from the first WGS analysis software is input to the second WGS analysis software in VCF, BED format, etc., and the tumor BAM file as the analysis object is input, then based on the pre-confirmed first structural variation site data, the "consensus structural variation site data" that exists in both WGS analysis software can be quickly obtained. The selection of WGS analysis software is not limited thereto. Preferably, the first WGS analysis software is a software that can quickly process a large amount of information. In the present invention, as a preferred example, DELLY (version: 0.8.7, https: / / github.com / dellytools / delly) is used. In addition, as a second WGS analysis software that quickly explores consensus structural variations based on the first structural variation site data provided by DELLY, the public software JuLI (https: / / github.com / sgilab / JuLI, J Mol Diagn. 2020Mar;22(3):304-318) is used.

[0033] Therefore, the method of detecting structural variations in cfDNA samples derived from cancer patients of the present invention can be as follows: the step 1) includes the following steps performed in sequence: 1-1) obtaining the first structural variation site data of the primary cancer sample by using an analysis software selected from the group consisting of DELLY, BRASS, SvABA, dRanger, Pindell, BreakDancer, GASV, Hydra and CNV nator; and 1-2) inputting the first structural variation site data obtained in the step 1-1) into JuLI to obtain the consensus structural variation site data of the commonly confirmed primary cancer sample.

[0034] In addition, the present invention includes the following steps: 2) matching the reference sequence of the structural variation site obtained in step 1) with the cfDNA WGS of the corresponding patient to obtain the structural variation sequence supporting reads present in the consensus structural variation site, thereby matching the specific sequence generated by the structural variation present in the cfDNA sample.

[0035] The present invention is characterized in that, regardless of the type of cancer, a small amount of structural variations commonly present in cfDNA samples can be detected with high sensitivity. In the present invention, the cfDNA may be cfDNA derived from cerebrospinal fluid, pleural effusion, pericardial effusion, peritoneal effusion, urine, whole blood, plasma or serum of a cancer patient, but is not limited thereto.

[0036] In the present invention, as with conventional variant detection methods, the detection sensitivity can be adjusted according to the sequencing depth.

[0037] In addition, the cfDNA of the cancer patient who is the subject of analysis in step 2) can be obtained from the patient during or after cancer treatment. In the present invention, a database (DB) of consensus structural variation sites present in the primary cancer tissue of the cancer patient is constructed in step 1), and then compared with the structural variation site sequence in the patient's cfDNA during or after the patient receives cancer treatment, thereby providing information on the tiny amount of residual cancer cells remaining in the patient after treatment.

[0038] More specifically, step 2) is as follows: First, WGS and alignment are performed on the cfDNA (Cell-free DNA) of cancer patients, and the cfDNA BAM (binary alignment map) format (i.e., cfDNA mapping data) recording the chromosome number and site information on the reference genome is obtained from each sequencing read. Then, the cfDNABAM file obtained as cfDNA mapping data and the consensus structural variation site data obtained in step 1) are input into the structural variation analysis software, and supporting read calls are performed. For the consensus structural variation sites confirmed in step 1), the corresponding reference sequence (counter reference sequence) and the supporting reads are aligned at the same time, thereby matching specific sequences generated by structural variations present in the cfDNA sample.

[0039] The corresponding reference sequence can be used in the same meaning as the reference genome data, and can be hg19 or hg38, etc.

[0040] The method of the present invention can be widely used for all cancers regardless of the type of cancer. Unlike existing image or blood tests and targeted high-depth sequencing methods, it does not require additional test items or pre-designed procedures based on the type of cancer. Therefore, the cancers targeted by the present invention may include, without limitation, cancers known in the art, such as gastric cancer, lung cancer, non-small cell lung cancer, breast cancer, ovarian cancer, liver cancer, bronchial cancer, nasopharyngeal cancer, laryngeal cancer, pancreatic cancer, bladder cancer, colorectal cancer, colon cancer, cervical cancer, bone cancer, non-small cell bone cancer, blood cancer, skin cancer (melanoma, etc.), brain cancer or neck cancer, uterine cancer, rectal cancer, anal cancer, colon cancer, fallopian tube cancer, endometrial cancer, vaginal cancer, vulvar cancer, Hodgkin's disease, esophageal cancer, small intestine cancer, endocrine gland cancer, thyroid cancer, parathyroid cancer, adrenal cancer, soft tissue sarcoma, urethral cancer, penile cancer, prostate cancer, chronic or acute leukemia, lymphocytic lymphoma, kidney or ureteral cancer, renal cell carcinoma, renal pelvic cancer, polyploid cancer, carcinoma, salivary gland cancer, sarcoma, pseudomyxoma, hepatoblastoma, testicular cancer, glioblastoma, lip cancer, ovarian germ cell tumor, basal cell carcinoma, multiple myeloma, gallbladder cancer, choroidal melanoma, ampulla of Barthes carcinoma, peritoneal carcinoma, adrenal cancer, tongue cancer, small cell carcinoma, pediatric lymphoma, neuroblastoma, duodenal cancer, ureteral cancer, astrocytoma, meningioma, renal pelvic cancer, vulvar cancer, thymic carcinoma, central nervous system (CNS) tumors, primary CNS lymphoma, spinal cord tumors, brain stem gliomas, or pituitary adenomas.

[0041] In the present invention, the structural variation that becomes the detection object may be a somatic structural variation, may be one or more selected from the group consisting of duplication, deletion, translocation and insertion of the gene, and refers to a situation in which a specific difference is presented in the sequence compared with the corresponding reference sequence as the comparison object. Most cancers undergo dozens to hundreds of structural variations in the early stages of cancer, and maintain such structural variations during the onset of cancer. In cancer cells, cfDNA flows out of the blood due to apoptosis. If the structural variation sequence found in cancer cells is confirmed in the cfDNA present in the blood, it can be diagnosed that the cancer remains in the patient's body and has the possibility of inducing tiny residual lesions. Therefore, the structural variation sequence that becomes the detection object of the present invention is a sequence of structural variations that are confirmed to be present in the patient's cancer cells and are present in cfDNA.

[0042] As another example of the present invention, the present invention provides a method for providing information on minimal residual disease (MRD), the method comprising the following steps: 1) obtaining consensus structural variation site data of commonly confirmed primary cancer samples by analyzing whole genome sequencing (WGS) of cancer tissues derived from patients using two or more structural variation analysis software; 2) matching the reference sequence of the structural variation site obtained in step 1) with the cfDNA WGS obtained after cancer treatment of the corresponding patient to obtain supporting reads of the structural variation sequence present in the consensus structural variation site, thereby matching specific sequences generated by structural variations present in the cfDNA sample.

[0043] In the present invention, microresidual lesions are also referred to as minimal residual lesions, which refer to the state in which a small number of malignant cells remaining in a patient during or after treatment are detected at the molecular level. Microresidual lesions have become the subject of follow-up for various blood cancers and solid cancers, and the patient's responsiveness to treatment can be confirmed by confirming microresidual lesions, and the risk of recurrence can be predicted. Therefore, the information on microresidual lesions of the present invention can be information on whether cancer cells remain during or after treatment, the possibility of cancer recurrence, or the prognosis of cancer treatment. If the structural variation sequence confirmed in the primary cancer sample is confirmed in the patient's cfDNA sample by executing the method of the present invention, it can be predicted that the patient has residual cancer cells, low responsiveness to treatment, high risk of recurrence, or poor prognosis after cancer treatment.

[0044] In the present invention, the treatment of cancer patients includes but is not limited to all treatment methods known in the art, such as radiotherapy, immunotherapy, hormone therapy, chemotherapy or surgical resection therapy.

[0045] In addition, the present invention provides information on minimal residual lesions by comparing the consensus structural variation of the patient's own primary cancer with the cfDNA structural variation sequence after treatment, so the present invention can be used as a patient-customized information provision method.

[0046] If the present invention is used, structural variant sequences can be detected with high sensitivity even in blood, plasma or serum samples containing very low purity cfDNA from patients after treatment. In one specific example of the present invention, it was confirmed that a sufficient number of supporting reads can be confirmed even in samples diluted to 1:12,800, thereby confirming the structural variant sequences of patient cfDNA samples and providing information on minimal residual lesions.

[0047] The method of detecting structural variation sequences in cfDNA samples from cancer patients and the method of providing information on micro-residual lesions of the present invention can both be methods performed in silico by a computer system. Therefore, the base sequence variation information can be received / obtained by a computer system, and in this regard, the method of the present invention can also include the step of receiving gene variation information by a computer system.

[0048] The contents of the present invention described above are applicable to each other as long as there is no contradiction, and contents implemented by those skilled in the art with appropriate modifications are also included in the scope of the present invention.

[0049] Hereinafter, the present invention will be described in detail by way of examples, but the scope of the present invention is not limited only to the following examples.

[0050] Example 1. Construction of a method for detecting structural variation-specific sequences in plasma cfDNA

[0051] To identify specific sequences of structural variants in plasma cfDNA from cancer patients, a two-step process was developed:

[0052] 1. Confirmation steps of consensus structural variants in patients

[0053] In order to find the consensus structural variation of the patient, two analysis tools were used. After performing WGS using cancer tissues and control group samples from patients, alignment was performed using the default setting of BWA. As analysis tools, JuLI (https: / / github.com / sgilab / JuLI), a structural variation detection software that can detect DNA somatic variation and DNA fusion, and DELLY (DELLY version: 0.8.7, https: / / github.com / dellytools / delly), which is other structural variation detection software, were used. First, structural variation calls using tumor BAM files and control group BAM files were performed by DELLY to confirm the structural variation sites (first structural variation site data) present in the patient's primary cancer sample. The confirmed patient structural variation site results are input into JuLI in BED format, and the tumor BAM file is input, and then the structural variation call is performed to confirm the sites suspected of being structural variations in DELLY and JuLI, thereby obtaining consensus structural variation site data. The confirmed consensus structural variation site data is used as the JuLI output format for the next two steps.

[0054] 2. cfDNA structural variation sequence detection steps through read-supported detection

[0055] Plasma samples were obtained from patients after cancer onset, DNA was extracted from the samples, WGS analysis was performed, and then alignment was performed using the default setting of BWA to obtain a cfDNABAM file. In order to confirm in the cfDNABAM file whether there are supporting reads representing structural variation-specific sequences at the same site as the consensus structural variation site ensured by the first step, the "callread" function of the JuLI program was applied. If the "callread" function is used, specific sequences resulting from structural variations can be matched at consensus structural variation sites by comparing corresponding reference sequences such as hg19 or hg38 with each supporting read. In this case, if the default value of splitratio in the JuLI parameters is adjusted, the specificity can be adjusted, and the specificity is increased by setting it to 0.95.

[0056] If this method is used, even with only a small number of reads, it is possible to detect with high sensitivity whether there are cfDNA structural variation-supporting reads with specific sequences supporting consensus structural variations at the fusion site to be confirmed.

[0057] The flowchart of the method for detecting structural variation sequences in plasma cfDNA of the present invention consisting of the two steps is shown in Figure 1 , and the comparison between the corresponding reference sequence and supporting reads of the two steps is patterned as shown in Figure 2 middle.

[0058] Experimental Example 1. Verification Analysis

[0059] The following describes the results of analysis and verification using samples from actual patients and cell lines.

[0060] 1.1 Preparation of analytical samples

[0061] In order to confirm the DNA structural variation present in cfDNA, the following samples were prepared. Cancer cells and reference substances were used as samples for analytical verification experiments. When performing the variation definition, NA12878 was designated as the reference substance, and structural variation was detected based on this comparison. NA12878 was purchased from the Coriell institute and used. Five cancer cell lines (WM2664, A375, SNU16, HCC1954, HCC95) were purchased from the Korean Cell Line Bank and used.

[0062] The DNA collected from NA12878 and five cancer cell lines (WM2664, A375, SNU16, HCC1954, HCC95) was evenly divided into 150bp to 170bp, which is known as the universal cfDNA slice size. The five cancer cell lines were diluted at 10ng / ul and mixed in the same volume of 50μl, and finally a mixed sample of 10ng / ul and 250μl was prepared as the initial sample. The initial sample volume of NA12878 was 50ng / ul. Afterwards, the mixed samples of these five cancer cells and NA 12878 were prepared in different ratios of cell line: NA12878 = 1:100, 1:200, 1:400, 1:800, 1:1600, 1:3200, 1:6400 and 1:12800, respectively, and used in experiments to confirm the sensitivity of the present invention. All experiments were repeated 3 times.

[0063] As clinical samples from patients, formalin fixed paraffin embedded (FFPE) cancer tissues and buffy coats of blood (control tissue) were obtained from patients with metastatic lung cancer who visited Inha University Hospital, and plasma samples collected at different time points were prepared to collect cfDNA.

[0064] 1.2 Analytical validation using cell line mixing experiments

[0065] The analysis verification using the method of Example 1 was performed. First, NA12878 and five cancer cell lines (WM2664, A375, SNU16, HCC1954, HCC95) and their mixed samples were analyzed three times at a mixing ratio of cell line: NA 12878 = 1:100, 1:200, 1:400, 1:800, 1:1600, 1:3200, 1:6400 and 1:12800, respectively. The results are shown in Tables 1 to 8. Each table shows the value of repeated detection of structural variations in five cancer cell lines three times, and the same sample ID represents one BAM file. CD_21_16908 to CD_21_16912 are the data corresponding to each cancer cell line as follows: CD_21_16908 (A375SM), CD_21_16909 (HCC95), CD_21_16910 (HCC1954), CD_21_16911 (SNU16), CD_21_16912 (WM2664). The cancer cell lines generated data at an average of 22X, and the mixed samples generated data at an average of 70X. The data used to confirm whether the variation of each cell line was detected in the WGS used as a reference only is shown in Table 9.

[0066] [Table 1]

[0067]

[0068] [Table 2]

[0069]

[0070]

[0071] [Table 3]

[0072]

[0073]

[0074] [Table 4]

[0075]

[0076] [Table 5]

[0077]

[0078]

[0079] [Table 6]

[0080]

[0081]

[0082] [Table 7]

[0083]

[0084]

[0085] [Table 8]

[0086]

[0087] [Table 9]

[0088]

[0089]

[0090] The "Total fusion number of each cell-line (corresponding to the structural variation of the primary cancer sample in Example 1)" indicating the number of structural variations present in cancer cell lines was 171, 292, 475, 194, and 247, respectively, after removing the germline structural variations that were repeated in other cell lines in each cancer type. As shown in Tables 1 to 8, structural variations (cfDNAFusion Number) present in various cancer cell lines were also detected in the cfDNA BAM files, and the supporting reads (read counts) supporting the structural variations were confirmed. In addition, it was confirmed that the structural mutation sequences of each cell line were not detected in the reference WGS that was not mixed with the cell line (Table 9).

[0091] The results of the sensitivity and specificity of the experiments using the cell lines are summarized in Table 10 and Figure 3 middle.

[0092] [Table 10]

[0093]

[0094] As shown in Table 10, as the dilution ratio increases, the number of supporting reads shows a trend of gradually decreasing, but even at a dilution of 1:12,800, supporting sequences that can confirm structural variation sequences were confirmed, and it was confirmed that they can be detected with a high sensitivity of about 73.3%. In particular, in the case of HCC1954 with 475 structural variations, 100% of them can be detected even at a dilution of 1:12,800. This means that if the method of the present invention is used, structural variation sequences present in cfDNA samples can be very effectively detected, thereby providing information on minimal residual lesions in cancer patients.

[0095] 1.3 Clinical validation using clinical samples

[0096] The method of the present invention is used to conduct clinical validation experiments to confirm whether mutations can also be effectively detected in the blood and cfDNA of actual patients.

[0097] The samples were primary cancer tissue, blood, and cfDNA1 and cfDNA2 obtained from two different time points of the patient, and the same analysis method as in Example 1 was used.

[0098] Each cancer patient was a metastatic lung cancer patient, and was recorded as Case 1 and Case 2. About 30X WGS data was generated using FFPE cancer tissues obtained from the patients, and an average of about 48X WGS data was generated from the remaining cfDNA samples.

[0099] Specifically, the patient's FFPE cancer tissue was used as a sample, WGS analysis using DELLY was performed, and alignment was performed using the default setting of BWA, and the consensus structural variations of the patient confirmed in both analysis tools were confirmed using JuLI. This was used as the JuLI output format. Then, the cfDNA structural variation sequence was confirmed by a 2-step supporting read call that compared the BAM file obtained by WGS of plasma cfDNA samples obtained from different time points and the consensus structural variation data confirmed above with the corresponding reference sequence. Through this, the cfDNA structural variations present in the patient's plasma obtained at different time points can be detected more sensitively, and the tiny residual lesions of cancer remaining in the patient can be diagnosed.

[0100] Patient Case 1

[0101] Figure 4 and Figure 5 The analysis results of case 1 show the results of confirming the structural variation sequence using cfDNA obtained from different time points (2006.11.07 and 2007.04.12) as samples. Figure 4 and Figure 5 In the study, it was confirmed that the structural variations confirmed in the primary cancer tissues derived from the patients were also confirmed in the cfDNA-1 and cfDNA-2 samples.

[0102] In order to verify the effect of the present invention, cancer mutations were confirmed by performing panel sequencing on the patient's primary cancer, and the cancer mutations were confirmed in cfDNA. Through panel sequencing, it was confirmed that case 1 patient was a patient who showed about 30% KRAS mutations, and the same mutations were actually confirmed in WGS of primary cancer, cfDNA-1, and cfDNA-2. The results of confirming the sites of KRAS mutations in WGS are shown in Figure 6 middle.

[0103] exist Figure 6 In the results, it was confirmed that KRAS mutations (36%) were detected in primary cancer tissues at a similar level to that of panel sequencing, and 19% and 6% KRAS mutations were also detected in cfDNA-1 and cfDNA-2 samples, respectively, verifying that if the method of the present invention is used, micro-residual cancer can be confirmed by sequence analysis of structural mutations in cfDNA samples. The number of reads detected in the corresponding patients is shown in Table 11 below.

[0104] [Table 11]

[0105]

[0106] According to Table 11, based on the results of analysis in primary cancer samples, the patient in Case 1 was a patient with a fusion sequence of a total of 131 structural variation sequences, and 115 or 112 structural variations were also confirmed in cfDNA by the method of the present invention, 4122 supporting reads were confirmed in the cfDNA-1 sample showing 19% KRAS variation, and 1926 supporting reads were confirmed in the cfDNA-2 sample showing 6% KRAS variation.

[0107] Compared to the existing method of analyzing plasma cfDNA by WGS, which shows low sensitivity and difficulty in detecting mutations below 5%, the present invention shows more than 1,900 reads even for mutations of around 6%, thus indicating that the result is a result showing that it has very excellent detection capabilities.

[0108] Patient Case 2

[0109] Figure 7 and Figure 8 The analysis results of case 2 show the results of confirming the structural variation sequence using cfDNA obtained from different time points (2010.10.29 and 2011.01.04) as samples. Figure 7 and Figure 8 In the study, we confirmed that structural variants identified in primary cancer tissues from patients were also identified in cfDNA- and cfDNA-2 samples.

[0110] Through panel sequencing, the mutation of case 2 was actually confirmed in the WGS of primary cancer, cfDNA-1, and cfDNA-2, and is shown in Fig. 9 and Fig.10 middle.

[0111] exist Fig. 9 In the study, it was confirmed that TP53 mutations were 2% and 0% in cfDNA-1 and cfDNA-2, respectively. Fig.10 However, as shown in Table 12, 128 reads were confirmed in cfDNA-1 showing 2-5% variation, and 118 reads were confirmed in cfDNA-1 showing 0-2% variation, confirming that supporting reads showing the presence of structural variation can be fully detected even in samples showing low allele frequencies of 2% to less than 5%.

[0112] [Table 12]

[0113]

Claims

1. A method for detecting structural variation sequences from cfDNA samples derived from cancer patients, comprising the following steps: 1) Analyzing whole genome sequencing (WGS) of cancer tissues from patients using two or more structural variation analysis software to obtain consensus structural variation data of commonly confirmed primary cancer samples; and 2) matching the reference sequence of the structural variation site obtained in step 1) with the cfDNA whole genome sequencing of the corresponding patient, and obtaining the structural variation sequence supporting reads present in the consensus structural variation site, thereby matching the specific sequence generated by the structural variation present in the cfDNA sample.

2. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The method adjusts the detection sensitivity by adjusting the sequencing depth of whole genome sequencing.

3. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The whole genome sequencing analysis software is selected from two or more of the group consisting of DELLY, BRASS, SvABA, dRanger, Pindell, BreakDancer, GASV, Hydra, CNV nator and JuLI.

4. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The step 1) comprises the following steps performed in sequence: 1-1) obtaining first structural variation site data of the primary cancer sample by using an analysis software selected from the group consisting of DELLY, BRASS, SvABA, dRanger, Pindell, BreakDancer, GASV, Hydra and CNV nator; as well as 1-2) Inputting the first structural variation site data obtained in step 1-1) into JuLI to obtain consensus structural variation site data of the commonly confirmed primary cancer samples.

5. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The cfDNA is derived from cerebrospinal fluid, pleural effusion, pericardial effusion, peritoneal effusion, urine, whole blood, plasma or serum of a cancer patient.

6. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The cfDNA of the cancer patient in step 2) is obtained from the patient during or after cancer treatment.

7. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The structural variation is one or more selected from the group consisting of gene duplication, deletion, translocation and insertion.

8. The method for detecting structural variation sequences in cfDNA samples from cancer patients according to claim 1, wherein: The cancer is selected from the group consisting of gastric cancer, lung cancer, non-small cell lung cancer, breast cancer, ovarian cancer, liver cancer, bronchial cancer, nasopharyngeal cancer, laryngeal cancer, pancreatic cancer, bladder cancer, large intestine cancer, colon cancer, cervical cancer, bone cancer, non-small cell bone cancer, blood cancer, skin cancer (melanoma, etc.), brain cancer or neck cancer, uterine cancer, rectal cancer, anal cancer, colon cancer, fallopian tube cancer, endometrial cancer, vaginal cancer, vulvar cancer, Hodgkin's disease, esophageal cancer, small intestine cancer, endocrine gland cancer, thyroid cancer, parathyroid cancer, adrenal cancer, soft tissue sarcoma, urethral cancer, penile cancer, prostate cancer, chronic or acute leukemia, lymphocytic lymphoma, kidney or ureteral cancer, renal cell carcinoma, renal pelvic cancer, polyploid cancer, carcinoma, salivary gland cancer, sarcoma, pseudomyxoma, hepatoblastoma, testicular cancer, glioblastoma, lip cancer, ovarian germ cell tumor, basal cell carcinoma, multiple myeloma, gallbladder cancer, choroidal melanoma, ampulla of Barthes carcinoma, peritoneal carcinoma, adrenal cancer, tongue cancer, small cell carcinoma, pediatric lymphoma, neuroblastoma, duodenal cancer, ureteral cancer, astrocytoma, meningioma, renal pelvic cancer, vulvar cancer, thymic carcinoma, central nervous system (CNS) tumors, primary CNS lymphoma, spinal cord tumors, brain stem gliomas, or pituitary adenomas.

9. A method for providing information on minimal residual disease (MRD), comprising the following steps: 1) Analyzing whole genome sequencing (WGS) of cancer tissues from patients using two or more structural variation analysis software to obtain consensus structural variation site data of commonly confirmed primary cancer samples; and 2) matching the reference sequence of the structural variation site obtained in step 1) with the whole genome sequencing of cfDNA obtained after cancer treatment of the corresponding patient, and obtaining supporting reads of the structural variation sequence present in the consensus structural variation site, thereby matching the specific sequence generated by the structural variation present in the cfDNA sample.

10. The method for providing information on minimal residual disease according to claim 9, wherein: The information on the minimal residual lesions is information on whether cancer cells remain during or after treatment, the possibility of cancer recurrence, or the prognosis of cancer treatment.

11. The method for providing information on minimal residual disease according to claim 9, wherein: The treatment in step 2) is radiotherapy, immunotherapy, hormone therapy, chemotherapy or surgical resection.

Images