Composition for cancer diagnosis for digital PCR and information providing method for cancer diagnosis

Abstract

The present invention relates to a method for diagnosing HER2-positive cancer by measuring a variation in the copy number of a HER2 oncogene using digital PCR (dPCR). According to the present invention, a method for measuring the copy number of a HER2 oncogene with high accuracy, sensitivity, and simplicity is provided, thereby improving clinical utility. In addition, the method can be readily applied as an alternative or complementary method to conventional IHC and ISHS, thereby overcoming limitations of conventional HER2 measurement methods.

Description

Composition for cancer diagnosis for digital PCR and method for providing information for cancer diagnosis

[0001] This application claims the benefit of priority based on Korean Patent Application No. 10-2024-0138461 filed on October 11, 2024, and all contents disclosed in the document of said Korean patent application are incorporated as part of this specification.

[0002] The present invention relates to a method for measuring variations in the copy number of the HER2 oncogene using digital PCR (dPCR), particularly digital real-time PCR (dPCR). Furthermore, the invention relates to a method for diagnosing HER2-positive cancer according to the above-described measurement method.

[0003] HER2 (Human epidermal growth factor receptor 2) is a cell membrane receptor with tyrosine kinase activity that primarily regulates cell proliferation, and excessive expression and activation of the HER2 protein induce unrestricted cell proliferation, leading to cancer. Protein overexpression due to HER2 gene amplification occurs in approximately 15–20% of invasive breast cancers, and this is highly associated with aggressive clinicopathological features and poor prognosis [1, 2]. Breast cancer subtypes possessing HER2 gene amplification and overexpression are classified as HER2-positive breast cancer, and patients diagnosed with HER2-positive breast cancer receive HER2-targeted therapy as a standard treatment. Since trastuzumab (brand name: Herceptin), developed as an anti-HER2 drug in 1998, was first approved by the FDA as a treatment for HER2-positive breast cancer, various types of anti-HER2 therapies have been continuously developed to date, and the success of anti-HER2 therapies has contributed to significantly improving the survival rate of patients with HER2-positive breast cancer and reducing cancer recurrence [3, 4]. In particular, recently, trastuzumab deruxtecan (T-DXd, brand name: Enhertu), an antibody-drug conjugate (ADC) developed as a new HER2 targeted therapy, received emergency FDA approval as the first treatment for HER2-low expression metastatic breast cancer, and the importance of the HER2 oncogene as a diagnostic and therapeutic target is receiving more attention.

[0004] Since trastuzumab, primarily used as a first-line treatment for HER2-positive breast cancer, is effective only in HER2-overexpressing carcinomas, HER2 companion diagnostic testing—which evaluates HER2 gene amplification or protein overexpression—is essential for breast cancer diagnosis and treatment decision-making to select patients eligible for HER2 targeted therapy. Current standard HER2 testing methods are performed using immunohistochemistry (IHC), which measures protein expression levels, and in situ hybridization (ISH), which measures gene copy number variations. HER2 IHC semi-quantitatively measures the expression level of HER2 protein on the cell surface through tissue staining using anti-HER2 antibodies. Based on the HER2 staining intensity on the cell membrane, patients are classified into four groups with IHC scores of 0, 1+, 2+, and 3+, and patients with an IHC score of 3+, indicating HER2 protein overexpression, are determined to be HER2-positive. HER2 ISH is a diagnostic method that uses DNA probes for the HER2 gene (ERBB2) and a control group (chromosome 17 centromere, CEP17) to determine whether the HER2 gene is amplified (HER2 / CEP17 ratio greater than 2). Methods such as Fluorescence In Situ Hybridization (FISH) or Silver In Situ Hybridization (SISH) are used. The standard diagnostic algorithm currently used in clinical practice involves a two-step process: first, IHC is performed to diagnose HER2 positivity (IHC score 3+) and HER2 inequality (IHC score 0, 1+); and for the borderline patient group with unclear results (HER2 IHC score 2+, equivocal), FISH and SISH are additionally performed to finally confirm whether HER2 is amplified.

[0005] Although only IHC and ISH diagnostic technologies are currently used as standard tests for HER2 evaluation, controversy regarding the testing methods and interpretation has persisted [6-8]. While HER2 IHC testing is widely used in clinical practice and is an FDA-approved method, it has the disadvantage of being heavily influenced by pre-analysis variables such as tissue fixation time and storage. Additionally, the semi-quantitative nature of the interpretation and the involvement of subjectivity by pathologists are other major factors that make accurate HER2 expression evaluation difficult. In actual cases, HER2 expression evaluation based on IHC results leads to conflicting interpretation patterns depending on the institution and the interpreter. On the other hand, ISH testing has the advantage of being relatively consistent and accurate compared to IHC testing, but due to limitations requiring high technical skills, time, and cost, it is used only in a limited manner for IHC 2+ (borderline, equivocal) patient groups. The CEP17 probe used as a control recognizes the chromosome 17p11.1-q11.1 region. However, in some breast cancer patients who possess focal amplification of the centromeric / pericentromeric region of chromosome 17, there is a possibility that they may be misdiagnosed as false negatives despite actually being HER2 positive. Furthermore, the current HER2 screening method, which is performed in two stages (IHC and ISH), consumes a significant amount of time and cost for HER2 evaluation. Additionally, since both IHC and ISH tests evaluate HER2 status through visual inspection by the interpreter, there is a major disadvantage in that objectivity may be lacking. Therefore, to overcome the shortcomings and limitations of the current HER2 screening method, the development of a new screening method capable of evaluating HER2 expression more accurately, objectively, and simply is required.

[0006] To enhance the accuracy of HER2 testing, various HER2 analytical methods have been developed and their clinical applicability evaluated. For instance, methods have been attempted to quantify HER2 protein expression levels through multiple reaction monitoring (MRM)-mass spectrometry, AI-based HER2 expression evaluation, and single-cell-based HER2 protein expression measurement, or to quantify HER2 gene copy numbers using Next Generation Sequencing (NGS) or digital PCR analysis. However, due to various limitations, such as non-standardized methods and a lack of clinical utility, no such method has yet been approved as a standard HER2 test. In particular, digital PCR—a third-generation PCR technique capable of absolute quantification of DNA and RNA without a standard curve, unlike conventional PCR methods—offers high sensitivity and precision and is optimized for evaluating genetic variations and expression. Although researchers have continuously attempted to develop HER2 diagnostic methods capable of precisely and rapidly detecting HER2 gene copy number variations by utilizing this technology, it still exhibits many limitations. All existing research on the development of digital PCR-based HER2 diagnostic methods has utilized droplet digital PCR (ddPCR), an endpoint-based digital PCR platform. However, ddPCR has limitations in widely utilizing it in clinical practice because it has technical limitations in that it applies an endpoint method, which can lead to false positives and false negatives in clinical diagnosis, and it requires a long time for droplet formation and PCR analysis, as well as high technical expertise and high costs.In addition, although ddPCR-based HER2 gene amplification tests have been reported to show a high agreement rate compared to current HER2 diagnostic methods, their efficacy has not been properly proven in multi-center clinical trials due to inconsistent results caused by the use of different cutoffs and clinically unverified controls across institutions. Even in some studies using CEP17 as a control, primer and probe sequences were non-specific, and inconsistent cut-off values ​​were used in clinical studies at each institution.

[0007] Therefore, research is needed on a rapid and accurate method for measuring the number of HER2 copies.

[0008] The objective of the present invention is to provide a method for measuring the copy number of the HER2 cancer gene and diagnosing HER2-positive cancer using digital real-time PCR (drPCR).

[0009] In order to solve the above problem, the present invention,

[0010] A polynucleotide set for detecting the HER2 gene, comprising the primer of SEQ ID NO. 1, the primer of SEQ ID NO. 2, and the probe of SEQ ID NO. 3, and

[0011] The present invention provides a composition for cancer diagnosis for digital PCR comprising any one polynucleotide set for CEP17 detection selected from the group consisting of: a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 4, a primer of SEQ ID NO. 5, and a probe of SEQ ID NO. 6; a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 7, a primer of SEQ ID NO. 8, and a probe of SEQ ID NO. 9; a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 10, a primer of SEQ ID NO. 11, and a probe of SEQ ID NO. 12; and a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 13, a primer of SEQ ID NO. 14, and a probe of SEQ ID NO. 15.

[0012] According to one embodiment, the digital PCR may be a digital real-time PCR.

[0013] According to one embodiment, the cancer may be one or more selected from the group consisting of stomach cancer, breast cancer, ovarian cancer, adenocarcinoma, endometrial cancer, prostate cancer, colorectal cancer, pancreatic cancer, lung cancer, gastroesophageal cancer, and bladder cancer.

[0014] According to one embodiment, the composition of the present invention can detect HER2 gene copy number variation (CNV).

[0015] According to one embodiment, the composition of the present invention can detect local copy number variation (CNV) and point mutation in the centromeric / pericentromeric DNA region of chromosome 17.

[0016] According to one embodiment, the 5' end of the probe of the present invention may be labeled with a fluorescent material.

[0017] According to one embodiment, the fluorescent material may be one or more selected from the group consisting of VIC, HEX, FAM, and EverGreen dye.

[0018] According to another embodiment, the present invention provides a cancer diagnostic kit comprising a composition as described above. The kit of the present invention can detect HER2 gene copy number variation (CNV).

[0019] According to another embodiment of the present invention,

[0020] A liquid sample injection step of injecting a liquid sample into an inlet and filling a plurality of wells containing a composition according to claim 1;

[0021] A step of bringing the well injected with the liquid sample into close contact with the CMOS photosensor;

[0022] A step of detecting a fluorescent signal using the above CMOS photosensor;

[0023] A Ct calculation step for calculating one or more cycle threshold values ​​(Ct) for each of the plurality of wells from the detected fluorescence signal; and

[0024] A method for providing information for the diagnosis of cancer using digital PCR is provided, comprising the step of detecting a HER2 copy number variation based on the above cycle threshold value.

[0025] According to one embodiment, the information providing method may further include the step of confirming the loss of the HER2 gene copy number.

[0026] According to one embodiment, the information providing method may further include the step of confirming a high-level amplification of the HER2 gene copy number.

[0027] According to one embodiment, the digital PCR may be a digital real-time PCR.

[0028] Specific details of other embodiments according to the present invention are included in the following detailed description.

[0029] According to the present invention, by providing a method for measuring the copy number of the HER2 cancer gene with high accuracy, sensitivity, and simplicity, clinical utility is improved, and the system of conventional HER2 measurement methods can be improved by easily applying it as a replacement or supplement to existing IHC and ISHS.

[0030] Figure 1 is a schematic diagram showing a digital real-time PCR device (LOAA) and an analysis process.

[0031] Figure 2 is a schematic diagram of the primer and probe design and optimization validation graph for drPCR of HER2 gene copy number variants.

[0032] Figure 3 is a schematic diagram of the standardization of the HER2 gene drPCR test method for breast cancer patients.

[0033] Figure 4 shows the process of establishing a drPCR test for HER2 on breast cancer tissue.

[0034] Figure 5 shows the multi-center validation of the established HER2 drPCR test method.

[0035] Figures 6 and 7 verify the accuracy of the HER2 drPCR test method using NGS.

[0036] Figure 8 verifies the false positives of HER2 IHC and the accuracy of drPCR testing through re-evaluation of HER2 IHC reading results.

[0037] Figure 9 confirms the effect of tumor purity on drPCR measurements for HER2.

[0038] Figure 10 shows microscopic images and interpretation results of a case with intratumoral HER2 heterogeneity, and confirms the effect of intratumoral HER2 expression heterogeneity on drPCR measurements for HER2.

[0039] Figures 11 to 13 relate to the measurement of drPCR of the HER2 gene and the centromeric / pericentromeric DNA region of chromosome 17 using CEP17 and Alt CEP17.

[0040] Figure 14 confirms the detection of partial deletion of the HER2 gene copy number using drPCR.

[0041] Figure 15 shows the drPCR measurement results for HER2 in tissue and plasma.

[0042] Figure 16 illustrates an algorithm model for the HER2 screening method using drPCR.

[0043] Figure 17 shows the drPCR test results for the HER2 gene on gastric cancer tissue.

[0044] Figure 18 is a schematic diagram showing a comparison of the analysis times of ddPCR and drPCR.

[0045] Figure 19 is a schematic diagram showing a comparison of analysis times when FISH or SISH is applied and when drPCR is applied.

[0046] Figure 20 is a schematic diagram showing the advantages (speed, simplicity, accuracy, precision) of the HER2 drPCR test method and ways to utilize it in cancer diagnosis and research.

[0047] The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention.

[0048] Diagnosis of HER2 is essential for selecting HER2-positive patients with HER2 gene amplification or protein overexpression who may benefit from anti-HER2 therapy in breast cancer or gastric cancer.

[0049] Conventionally, HER2 is measured using immunohistochemistry (IHC) and in situ hybridization (ISH), which are standard diagnostic methods for HER2. However, since IHC and ISH tests rely on visual inspection, subjectivity is involved, leading to differing interpretation results between readers and institutions. Furthermore, there is a problem in that it is difficult to accurately evaluate HER2 expression due to various variables that occur depending on the condition of the tissue. In addition, there is a disadvantage in that a significant amount of time and cost is incurred during the process of proceeding with a two-stage testing algorithm designed to mutually compensate for the shortcomings of IHC and ISH.

[0050] Therefore, a new, objective HER2 testing method is needed to quickly and accurately diagnose HER2 positive / negative status.

[0051] In this study, we aimed to develop a clinically reliable HER2 testing method based on drPCR analysis by using the LOAA digital real-time PCR (drPCR) system, a new in vitro diagnostic platform that combines digital PCR technology with real-time PCR to improve performance compared to conventional digital PCR and enable ultra-fast gene mutation analysis. This method allows for the rapid and sensitive detection of HER2 gene amplification with high accuracy. To accurately detect HER2 gene copy number variations (CNV / copy number alteration, CNA) through drPCR analysis, we designed primer-probe sets specific to DNA regions adjacent to the HER2 gene and the centromere of chromosome 17 (CEP17). Through performance testing, we derived optimal drPCR conditions and a cut-off (HER2 / CEP17 copy number ratio 1.9) capable of determining the presence of HER2 CNA in breast or gastric cancer cell lines and patient-derived tissues. To validate the clinical validity of drPCR-based HER2 status assessment and standardize the evaluation methodology, three independent breast cancer cohorts from different institutions were utilized. A multi-center clinical trial was conducted by designating each institution's cohort as one training set (SCH Hospital, n = 103) and two validation sets (SNU Hospital, n = 200; CNU Hospital, n = 60). In the training cohort (n = 103), a high correlation was observed between the HER2 / CEP17 ratio values ​​calculated from drPCR and FISH analysis using the current standard HER2 CNA assay (r 2= 0.910; P < 0.001), drPCR-based HER2 evaluation showed a high agreement rate of 98.1% with HER2 status defined by IHC and / or FISH (sensitivity: 92.6%, specificity: 100%). Consistently, high agreement rates were observed between the results of current HER2 tests and the results of HER2 drPCR-based HER2 evaluations in two independent validation cohorts. In particular, the optimal HER2 drPCR cutoff (HER2 / CEP17 ratio ≥ 1.9, AUC 0.963) established based on training cohort testing was verified to be equally applicable to two independent validation cohorts with high sensitivity, specificity, and accuracy, which supports the possibility of it being utilized as a standardized HER2 testing method. In addition, the accuracy of drPCR-based HER2 diagnosis was verified by comparing the results of HER2 drPCR evaluations performed on clinical specimens with the results of targeted next-generation sequencing (NGS)-based HER2 CN analysis and HER2 IHC scoring values ​​objectified by artificial intelligence (AI) analysis. In some cases where there was a discrepancy in readings between drPCR-based HER2 evaluations and current HER2 tests, low tumor purity (≤25%) was observed. In these cases, the HER2 diagnosis agreement rate improved when macrodissection was performed or when a cut-off (HER2 / CEP17 ratio 1.385) for specimens with low tumor purity was applied. In addition, by devising a method to detect breast cancer exhibiting a shallow deletion (HER2 / CEP17 ratio < 1) of HER2 CN through HER2 drPCR analysis, a new auxiliary method was presented to distinguish between HER2-zero and low-expression breast cancer patient groups, and the potential for utilizing drPCR-based HER2 diagnosis in plasma as well as clinical tissue specimens was identified.

[0052] The present invention aims to provide the following means to solve the problems of conventional IHC and ISH tests for measuring HER2 gene copy number variations.

[0053] 1. Development of Primers and Probes for Precise Measurement of HER2 Gene Copy Number Variants Based on Digital Real-Time PCR (drePCR)

[0054] 2. drPCR-based HER2 gene amplification diagnostic method

[0055] 3. drPCR-based diagnostic method for HER2 copy number deletion

[0056] 4. Diagnostic method for HER2 gene amplification of clinical specimens with low tumor purity

[0057] 5. HER2 Gene Amplification Diagnostic Method Using an Alternative CEP17 Primer-Probe for Precise HER2 Diagnosis in CEP17 Mutant Carcinomas

[0058] 6. Diagnosis of HER2 gene amplification based on drPCR analysis in plasma-derived cell-free DNA (cfDNA)

[0059]

[0060] Specifically, the present invention is,

[0061] A polynucleotide set for detecting the HER2 gene, comprising the primer of SEQ ID NO. 1, the primer of SEQ ID NO. 2, and the probe of SEQ ID NO. 3, and

[0062] The present invention provides a composition for cancer diagnosis for digital PCR comprising any one polynucleotide set for CEP17 detection selected from the group consisting of: a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 4, a primer of SEQ ID NO. 5, and a probe of SEQ ID NO. 6; a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 7, a primer of SEQ ID NO. 8, and a probe of SEQ ID NO. 9; a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 10, a primer of SEQ ID NO. 11, and a probe of SEQ ID NO. 12; and a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 13, a primer of SEQ ID NO. 14, and a probe of SEQ ID NO. 15.

[0063] According to one embodiment, the digital PCR may be a digital real-time PCR.

[0064] According to one embodiment, the present invention can detect HER2 gene copy number variation (CNV).

[0065] According to one embodiment, the present invention can measure HER2 gene copy number variations using dPCR or drPCR and specific primers and probes for HER2 gene amplification, and thereby diagnose cancer. Examples of cancers may include breast cancer, gastric cancer, adenocarcinoma, thyroid cancer, lung cancer, ovarian cancer, endometrial cancer, liver cancer, colorectal cancer, pancreatic cancer, prostate cancer, gastroesophageal cancer, biliary tract cancer, gallbladder cancer, kidney cancer, oral cancer, basal cell carcinoma, brain tumor, colon cancer, anal cancer, multiple myeloma, melanoma, bladder cancer, retinoblastoma, peritoneal cancer, adrenal cancer, small intestine cancer, esophageal cancer, lymphoma, meningioma, heart cancer, duodenal cancer, bone tumor, urethral cancer, uterine cancer, spinal cord tumor, laryngeal cancer, thymic cancer, etc. According to one aspect, the present invention can be used for the diagnosis of breast cancer or gastric cancer.

[0066] According to one embodiment, the 5' end of the probe may be labeled with a fluorescent material. For example, the fluorescent material may be one or more selected from the group consisting of VIC, HEX, FAM, and EverGreen dye.

[0067] According to another embodiment, the present invention provides a cancer diagnostic kit comprising a composition for cancer diagnosis for digital PCR.

[0068] According to another embodiment of the present invention,

[0069] A liquid sample injection step of injecting a liquid sample into an inlet and filling a plurality of wells containing a composition according to claim 1;

[0070] A step of bringing the well injected with the liquid sample into close contact with the CMOS photosensor;

[0071] A step of detecting a fluorescent signal using the above CMOS photosensor;

[0072] A Ct calculation step for calculating one or more cycle threshold values ​​(Ct) for each of the plurality of wells from the detected fluorescence signal; and

[0073] A method for providing information for the diagnosis of cancer using digital PCR is provided, comprising the step of detecting a HER2 copy number variation based on the above cycle threshold value.

[0074]

[0075] Examples

[0076] Materials and Methods

[0077] Research Design and Research Participants

[0078] For this multicenter study, a total of 363 breast cancer patients were enrolled at three institutions in Korea. By institution, 200 patients were recruited from Seoul National University Hospital (SNU Hospital; SNUH), 103 from Soonchunhyang University Cheonan Hospital (SCH Hospital; SCHH), and 60 from Chonnam National University Hwasun Hospital (CNU Hospital, CNUH). All patients underwent surgical resection between 2017 and 2022 and were diagnosed with breast cancer via biopsy. Through a review of medical records, age, tumor subtype, nuclear grade, tumor grade, AJCC stage, grade, presence of lymphatic invasion, ER, PR, HER2 expression, and HER2 immunohistochemical staining results were collected. Slides from all relevant cases were reviewed by pathologists, and representative slides were selected to evaluate the tumor cell ratio on hematoxylin-eogene (HE) slides and to identify formalin-fixed-paraffin-embedded (FFPE) blocks. The clinical and pathological characteristics of each institution are summarized in Table 1. This study was approved by the Institutional Review Boards (IRBs) of SNU, SCH, and CNU Hospitals. The biological samples and data from CNU Hospital used in this study were provided by the Human Biological Material Bank of CNU Hospital, a member of the KBN Network.

[0079] The subjects included in the study were divided into one training cohort and two validation cohorts. 103 patients from SCH Hospital were selected for the training cohort. In this group, FISH and drPCR were performed on all cases to evaluate and calculate the HER2 amplification ratio. Based on this, an appropriate cutoff value was proposed to distinguish between positive and negative drPCR results, which was applied to each validation set consisting of 200 cases from SNU Hospital and 60 cases from CNU Hospital. To verify the accuracy of the drPCR HER2 test, it was compared with the collected IHC and FISH results.

[0080]

[0081] Immunohistochemistry (IHC)

[0082] To analyze HER2 protein levels in tissue samples obtained from breast cancer patients, tumor sections of paraffin-embedded blocks were stained using an antibody against HER2 / NEU (clone 4B5, Ventana), and expression was evaluated by two pathologists (Si-Hyung Jang and Han-Seok Yoo) according to ASCO / CAP guidelines. All immunohistochemical slides were scanned using an Aperio ScanScope CS (Leica biosystems, Inc., Wetzlar, Germany) and quantified according to the following criteria: in invasive cancer, staining of more than 10% of the cell membrane was considered significant, and based on staining intensity, 0 or 1+ was interpreted as negative, 2+ as equivocal, and 3+ as positive. Interpretation was performed by pathologists and the ImageScope computer image analysis system (Aperio Technologies) using the 'Cell+membrane detection' algorithm.

[0083]

[0084] Fluorescence in situ hybridization (FISH)

[0085] For FISH analysis, two markers (probes) from the PathVysion DNA Probe Kit (Vysis, Downers, Grove, IL)—HER2 / new (orange) and CEP17 (green)—were used. FISH analysis was performed on 103 cases from SCH Hospital in accordance with the manufacturer's guidelines. Representative slides were selected based on a review by a pathologist (Dr. Si-Hyung Jang), marked with a 0.5 x 0.5 cm area, and the corresponding regions were sectioned from FFPE blocks to a thickness of 2 micrometers. After removing paraffin, the samples were treated with a pretreatment solution at 80°C for 12 minutes, followed by treatment with a protease solution at 37°C for 60 minutes. Subsequently, the samples were washed with purified water and dried, then fixed with formalin for 10 minutes, followed by ethanol washing and air drying for 5 minutes. For hybridization, the HER2 / CEP17 dual marker was applied, and the samples were incubated overnight at 37°C. After washing, the slides were stained with DAPI and covered with cover slides. FISH analysis was performed by a pathologist (Dr. Si-Hyung Jang) using a fluorescence microscope. Fifty tumor cells were evaluated in each case, and the interpretation was based on the criteria used in the ASCO / CAP 2018 guidelines. Although the original ASCO / CAP guidelines suggested evaluating 20 tumor cells, this study aimed to improve the accuracy of the HER2 / CEP17 ratio by evaluating a larger number of cells.

[0086]

[0087] Silver In Situ Hybridization (SISH)

[0088] SISH analysis of tissue sections was performed using the VENTANA HER2 Dual ISH DNA Probe Cocktail kit (Roche Diagnostics, Indianapolis, IN, USA) and an automated immunostainer (Benchmark ULTRA automated slide stainer, Roche Diagnostics) according to the manufacturer's instructions. The HER2 DNA probe labeled with dinitrolphenol (DNP) and the control probe labeled with digoxigenin (DIG) (chr17, chromosome 17 centromere) were denatured at 80°C for 10 minutes, accustomed to the tissue at 44°C for 6 hours, and then washed three times at 72°C. To detect the HER2 signal in the tissue, the color reaction of the HER2 DNP probe was confirmed using the ultraView™ SISH Detection Kit, which contained a rabbit anti-DNP primary antibody and a goat anti-rabbit IgG antibody conjugated with horse radish peroxidase (HRP). After the primary and secondary antibody reactions, sequential reaction with Silver A (silver acetate), Silver B (hydroquinone), and Silver C (H2O2) resulted in the deposition of silver deposits in the nucleus, and each pair of HER2 genes was identified as a black dot. The control probe reacted with the mouse anti-DIG primary antibody followed by a goat anti-mouse IgG antibody conjugated with alkaline phosphatase (AP); subsequently, after reaction with Fast Red and naphthol phosphate, a red dot representing CEP17 was identified. Each tissue was background stained with Harris hematoxylin and Bluing agent.

[0089]

[0090] breast cancer cell line culture

[0091] Human breast epithelial cell line MCF10A and breast cancer cell lines MCF7, BT474, SKBR3, HCC-1954, HCC-1569, MDA-MB-453, BT549, MDA-MB-231, and MDA-MB-468 were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and the Korean Cell Line Bank (KCLB, Seoul, Republic of Korea). All cell lines were cultured in incubators at 37°C and 5% CO2. MCF10A cell lines were placed in DMEM / F-12 medium (Welgene) supplemented with 5% horse serum (Gibco, Grand Island, NY, USA), 20 ng / ml EGF, 10 μg / ml insulin, 0.5 μg / ml hydrocortisone, and 1% penicillin-streptomycin (Welgene, Daegu, Korea); MCF7, MDA-MB-231, and MDA-MB-468 cell lines were placed in DMEM medium (Welgene) supplemented with 10% fetal bovine serum (FBS; Welgene, Daegu, Korea) and 1% penicillin-streptomycin (Welgene); and BT474, SKBR3, HCC-1954, HCC-1569, JIMT-1, MDA-MB-453, and BT549 cell lines were placed in 10% FBS (Welgene, Daegu, Korea) and 1% It was cultured in RPMI1640 (Welgene) with added Penicillin-Streptomycin (Welgene).

[0092]

[0093] DNA extraction

[0094] 1) Extraction of genomic DNA (gDNA) from breast cancer cell lines

[0095] For drPCR-based comparative analysis of HER2 gene copy numbers in HER2-positive and HER2-negative breast cancer cell lines, gDNA was extracted from each cell line according to the manufacturer's guidelines using the QIAamp DNA Mini Kit (51304, QIAGEN, Hilden, Germany) or the AccuPrep Genomic DNA Extraction Kit (K-3032, Bioneer, Daejeon, Republic of Korea). The gDNA extracted from the cell lines was cleaved into short DNA fragments by reacting with EcoRI (R0101, New England Biolabs, Ipswich, MA, USA) restriction enzyme at 37°C for 1 hour for drPCR analysis, and then inactivated by reacting at 65°C for 20 minutes. The cleaved gDNA was diluted 1:5 with nuclease-free water to dilute the salts present in the digestion buffer, and then used for drPCR analysis.

[0096]

[0097] 2) Extraction of DNA from tissues and plasma derived from breast cancer patients

[0098] gDNA extraction from FFPE tissues was performed using the PANAMAX FFPE DNA extraction kit and the PANAMAX 16 system in accordance with the manufacturer's guidelines. FFPE tissues were pretreated with FIB solution and proteinase K, followed by heat treatment at 70°C overnight. Subsequently, samples were mixed with FLB solution and placed in wells containing binding buffer, after which DNA extraction and purification were performed using the automated PANAMAX 16 system.

[0099] DNA extraction from plasma was similar, and the PANAMAX Plasma ccfDNA extraction kit and the PANAMAX16 system were used for the extraction. 2 ml of plasma and 100 µl (microliter) of proteinase K reagent were added to the wells along with the dissolution and binding buffer. Then, DNA was extracted using the automated PANAMAX16 system.

[0100]

[0101] Primer and probe design for digital real-time PCR

[0102] To measure the copy number of the HER2 gene based on drPCR, primers and probes (PnPs) for drPCR were designed to specifically bind to HER2 and CEP17 in the DNA region coding for the HER2 gene (ERBB) located at chromosome 17q12 and in the DNA region near the q arm of the chromosome 17 centromere (CEP17). In addition, to assess the potential impact of genetic variations in the CEP17 q arm on the HER2 copy number detection results, three alternative CEP17 primer-probe sets were designed for the distal region of the CEP17 q arm and near the CEP17 p arm. For duplex drPCR, the HER2 probe was labeled with FAM fluorescence, while the CEP17 and alternative CEP17 probes were labeled with Chamel610 fluorescence. DNA sequence information for each PnP is summarized in Table 2.

[0103]

[0104] Digital real-time PCR (drPCR)

[0105] A LOAA on-point (OPTOLANE, Seongnam, Republic of Korea) digital real-time PCR instrument was used to measure HER2 gene copy number variation (CNV) in cancer. FFPE derived from breast cancer patients, plasma, or DNA derived from cell lines (25 ng, 50 ng, or 10 ng, respectively) were mixed with 1 μM forward / reverse primers targeting HER2 and CEP17 DNA regions, 0.17 μM HER2 target probe, 0.5 μM CEP17 target probe, and 2x Dr.PCR mastermix to prepare a total of 30 μl of mixture for the drPCR reaction, which was then injected into the Genotizer chip. After mounting the chip injected with the mixture into a chip case, a PCR reaction was performed using a LOAA digital real-time PCR instrument, consisting of a reaction at 50°C for 3 minutes and 95°C for 15 minutes, followed by 45 cycles of 95°C for 10 seconds and 60°C for 40 seconds. Subsequently, the concentrations of HER2 and CEP17 and the HER2 / CEP17 ratio were analyzed to calculate the HER2 gene copy number. For cell lines and tissue samples, HER2 gene amplification (HER2 positive) was determined when the HER2 / CEP17 ratio was 1.9 or higher, and for plasma, when the HER2 / CEP17 ratio was 1.2 or higher; a HER2 gene copy number deletion was determined when the HER2 / CEP17 ratio was 0.95 or lower.

[0106]

[0107] Droplet digital PCR (ddPCR)

[0108] To compare the performance of LOAA drPCR and droplet digital PCR (ddPCR) in detecting HER2 gene copy number mutations, 22 FFPE tissue samples from breast cancer patients were analyzed by ddPCR using PnP sets for HER2 and CEP17. Since the fluorescence detection systems between drPCR and ddPCR are different, the fluorescent dye of the CEP17 probe (chamel610) was replaced with the hexane of ddPCR. A 20 μl PCR mixture was prepared containing 25 ng gDNA, 2X ddPCR SuperMix (1863024, Bio-Rad Laboratories, Hercules, CA, USA), primers targeting HER2 or CEP17 DNA regions (forward primer, reverse primer, 0.9 μM each), and a probe (0.25 μM). Droplets were generated using a Bio-Rad QX200™ droplet generator, and the reaction mixture was dispensed. The generated droplets were transferred to a 96-well PCR plate and subjected to PCR reaction using a SimpliAmp Thermal cycler (Applied Biosystems by Thermo Fisher Scientific, Carlsbad, CA, USA). The thermal cycling conditions for PCR consisted of a 10-minute reaction at 95°C, 40 cycles of 30 seconds at 94°C and 1 minute at 60°C, followed by a 10-minute reaction at 98°C and maintenance at 4°C. After the PCR reaction was completed, the PCR plates were read using a Bio-Rad QX200 droplet reader, and the results were subsequently analyzed using software provided by Bio-Rad.

[0109]

[0110] Next-generation sequencing (HER2 target DNA sequencing)

[0111] 1) DNA and RNA extraction for Next-Generation Sequencing (NGS)

[0112] Two types of NGS platforms were used for HER2-targeted sequencing analysis. For NGS utilizing ThermoFisher's Ion Torrent Genexus NGS platform, 12 samples were selected from tissue specimens derived from patients at SCH Hospital, and Recover All Multi-Sample DNA / RNA Workflows (Invitrogen A26069 and A26135) were used to extract DNA and RNA from FFPE tissues. Tissue sections were paraffin removed using xylene, washed twice with 100% ethanol, and dried for 15 minutes. Subsequently, the samples were treated with protease K. Each sample was passed through a PureLink column for DNA and RNA separation. DNA remaining on the column was separated through column washing, and the filtrate was filtered through a new PureLink column and treated with DNase to separate RNA. DNA / RNA concentrations were measured using Qubit. Forty-one clinical tissue samples from SCHH and SNUH were selected for HER2 target sequencing analysis in the CancerScreen Core Panel (Celemics, Inc, Seoul, Republic of Korea) utilizing Illumina’s NextSeq 500 NGS platform, and gDNA was extracted from each sample according to the manufacturer’s guidelines using the PANAMAX FFPE DNA extraction kit and PANAMAX16 system or the QIAamp DNA FFPE tissue kit (Qiagen).

[0113]

[0114] 2) Library preparation, sequencing, and data analysis

[0115] For NGS analysis based on the Ion Torrent Genexus (ThermoFisher) platform, at least 10 ng of nucleic acid was loaded onto an Ion Torrent GX5 chip, and the chip was then inserted into an Ion Torrent Genexus sequencer. Using the Genexus platform, all NGS processes, including library preparation, template preparation, sequencing, and bioinformatics analysis, were performed automatically. The OCA version 3, the solid tumor panel used in this study, can detect mutations in 161 major genes, including 87 hotspot mutations, 43 local copy number changes (CNV gain), 48 CDS mutations, and 51 fusion mutations. After sequencing, the data were analyzed using Genexus software and interpreted using Oncomine Kowlegebase Reporter v.5.6. For HER2 target sequencing analysis based on the NextSeq 500 NGS (Illumina) platform, a library was constructed using the Celemics library prep kit (Celemics), and the NGS library and capture probe were hybridized using the CancerScreen Core Panel (Celemics) to capture all target regions. The target capture library was amplified after PCR to enrich the sample volume and sequenced as 2 x 150bp paired-end reads on the Illumina NextSeq 500 sequencing system. The HER2 copy number variation (CNV) value was calculated by dividing the normalized read depth of the test sample by the average normalized read depth of the control sample. A CNV value more than twice as high as that of the control sample was considered as gene amplification.

[0116]

[0117] Public data analysis

[0118] To explore genetic variations and mRNA expression levels of representative genes located near the chromosome 17 centromere and at the end of the q arm in breast cancer patients and cell lines, we used cBioportal (http: / www.cbioportal.org) [17, 18] to analyze the large-scale breast cancer patient dataset TCGA (The Cancer Genome Atlas) and the cancer cell line analysis dataset CCLE (Cancer Cell Line Encyclopedia).

[0119]

[0120] Statistical analysis

[0121] To confirm statistical significance between groups, unpaired Student's t-test was used for the analysis between two groups, and analysis of variance (ANOVA) was used for the analysis between multiple groups. Pearson's correlation analysis was used to determine the r-value for correlation testing. A P-value < 0.05 indicated statistical significance. To determine the appropriate cutoff for drPCR-based HER2 evaluation in clinical specimens, ROC curves were constructed for continuous and categorical data for the training and validation cohorts using the R program, and sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated as HER2 amplification diagnostic performance indicators for the measured optimal cutoff.

[0122]

[0123] Results

[0124] 1. Principles of the Chip-Based LOAA Digital Real-Time PCR (drPCR) System (Overview of LOAA digital real-time PCR system)

[0125] This study utilized the LOAA digital real-time PCR (drPCR, Optorain) instrument, a new digital PCR platform that offers improved performance compared to conventional digital PCR. LOAA drPCR is a new digital real-time PCR (drPCR) system that combines real-time PCR and digital PCR technologies, featuring a digitizer (Genotizer TMIt consists of a chip, a chip case, and equipment (Fig. 1A). LOAA drPCR is a detachable chip-based dPCR created by adding a well structure based on MEMS (Micro Electro Mechanical System) technology to a CMOS semiconductor chip, distributing and injecting equal amounts of target molecules into approximately 20,000 wells within the chip (Figs. 1B and C). Reducing the PCR volume using MEMS technology results in a smaller thermal mass, which accelerates the heating-cooling cycle. Unlike conventional real-time PCR methods that amplify genes by controlling temperature through a thermal controller separate from the optical system, Optorain's PCR system amplifies genes by controlling electrical signals using a thermal sensor embedded in the chip case (Fig. 1B). Optical systems used in conventional Real-Time PCR or Digital PCR measure the fluorescence emitted from the sample after reflecting wavelengths passed through an Excitation Filter onto a dichroic mirror. However, while the light emitted from the sample spreads evenly across all angles, optical sensors can only detect light from limited angles, and the remaining fluorescence is lost. Furthermore, since fluorescence intensity is proportional to the square of the distance from the fluorescence source, detection becomes more sensitive the closer the object is to the source. Focusing on these issues, LOAA drPCR maximized fluorescence detection capabilities by constructing a bio-reaction chamber directly above the optical sensor (Figs. 1B and C). Because CMOS photosensors receive not only excitation light but also fluorescence from the PCR reaction, it is necessary to develop a fluorescence filter that blocks as much excitation light as possible while transmitting as much fluorescence as possible. Accordingly, Optorain integrated a fluorescence filter for FAM, the most widely used fluorescent material in PCR, into the CMOS sensor using its proprietary technology. Additionally, to add fluorescence channels, two color filters—green and red—were coated onto each pixel of the semiconductor sensor, allowing the fluorescence values ​​of each color pixel to be collected and the two fluorescences to be output separately (Fig. 1C).The well structure fabricated by Optorain consists of a 1152 x 1152 2D photosensor array (Pixel) with column addresses (Cj) and row addresses (Rj), perfectly separated by a special membrane tens of μm thick (Fig. 1C). Optorain's MEMS partition is positioned as close as possible to the light source, and the partition reacts only to the photosensor array inside the partition to measure changes in fluorescence signals, thereby generating a real-time amplification graph and displaying the gene amplification process in every cycle. This real-time digital PCR offers many advantages over conventional endpoint digital PCR technology. First, the real-time PCR graph for each partition reduces the possibility of false positives and allows for the distinction between negatives and even one copy, enabling accurate test results even at very low concentrations. Second, as the real-time PCR graph for each partition can be verified, quantification can be performed using the CT values ​​of the real-time PCR even if the target fills all partitions. In quantitative measurements, the peak of the Poisson distribution is nearly 3 to 4 times higher compared to conventional digital PCR, and accordingly, the dynamic range of measurement increases by more than 4 times compared to conventional digital PCR. In particular, the dynamic range of diagnosis is directly related to sensitivity in cancer diagnostic applications where cancer cells must be detected by comparison with normal cells. Third, high-resolution digital melting curve analysis is possible in samples with various methylations through digital melting curve analysis using real-time digital PCR technology. Another advantage of real-time dPCR is that the experimental procedure is simple and intuitive. Since the process of drawing a standard curve, as in conventional real-time PCR, is not required, the test time is shortened to within 1.5 hours (Fig. 1D). In conclusion, compared to conventional digital PCR or real-time PCR, this real-time dPCR reduces false positives, increases sensitivity, and provides convenience to users, making it a device that can be utilized in various research and diagnostic tasks.

[0126]

[0127] 2. Design of primers and probes and optimization of drPCR conditions for HER2 copy number detection based on drPCR analysis

[0128] To accurately measure the HER2 copy number using the drPCR analysis technique, a drPCR HER2 and CEP17 (q11.1) primer-probe set (PnP) was designed to specifically bind only to the DNA sequences in the corresponding regions of the DNA of ERBB2 (the gene coding for HER2) located at chromosome 17q12 (NC_000017, 39,708,411 - 39,708,474 bp based on GRCh38 / hg38) and the DNA region near the q arm of chromosome 17 centromere (CEP17) as a reference control (NC_000017, 26,982,730 - 26,982,791 bp) (Fig. 2A, Table 2). In addition, three additional sets of alternative CEP17 PnPs were designed as alternative CEP17 PnPs for a group of breast cancer patients with genetic mutations in the CEP17(q11.1) region, in the distal region of the CEP17 q arm (Alt CEP17(q11.1-distal); NC_000017, 27,083,670 - 27,083,733 bp), near the CEP17 p arm region (Alt CEP17(p11.1); NC_000017, 22,705,258 - 22,705,325 bp), and in the chromosome 17p11.2 region (Alt CEP17(p11.2); NC_000017, 22,532,067 - 22,532,132 bp) (Fig. 2A, Table 2). For Duplex drPCR, FAM fluorescence was labeled on the HER2 probe, and Chamel610 fluorescence was labeled on CEP17 and alternative CEP17 probes.The specificity of the designed HER2 and CEP17 PnP was verified through real-time PCR analysis (Fig. 2B), and it was verified through an interference test using breast cancer cell lines that there was no interference effect on the derivation of results and fluorescence efficacy between HER2 and CEP17 in duplex drPCR (Fig. 2C). In addition, the HER2 and CEP17 drPCR conditions established through optimization experiments were tested on cell lines of various breast cancer subtypes and normal epithelial cells, and it was verified that HER2 gene amplification could be accurately detected via drPCR in HER2-positive breast cancer cell lines (BT-474, SKBR3, HCC-1954, HCC-1569, JIMT1, MDA-MB-453) known to possess HER2 gene amplification according to the CCLE dataset and various literature (Fig. 2DF).

[0129]

[0130] 3. Multicenter study design to evaluate HER2 status using drPCR in patients with breast cancer

[0131] The design of a multicenter clinical trial for the development and standardization of a drPCR-based HER2 diagnostic method in breast cancer is as follows (Fig. 3). 103 FFPE cases from SCH Hospital were selected for the training group, and FISH and drPCR were performed on all cases. In the FISH analysis, 50 tumor cells were evaluated—more than the 20 tumor cells suggested by the ASCO / CAP guidelines—to calculate the HER2 / CEP17 ratio in various regions of the tumor. The drPCR HER2 / CEP17 ratio was compared with the HER2 / CEP17 ratio from FISH performed on the same tissue block. Subsequently, an appropriate cutoff was determined through ROC curve analysis, and the agreement between the drPCR results and previously collected clinical information (HER2 results determined via IHC and FISH) was evaluated. NGS was performed on 40 samples, and the agreement with the drPCR and FISH results was evaluated and analyzed.

[0132] The validation group consisted of two institutions, and drPCR was performed on 200 FFPE samples from SNU Hospital and 60 from CNU Hospital. The results of drPCR, applying cutoffs derived from the training group, were compared with the IHC and ISH results from the collected clinical records. In cases where discrepancies occurred between cases, they were reconfirmed through NGS, ISH, and IHC. Validation and verification were performed using appropriate cutoffs from the training set through ROC curve analysis. To compare the results between conventional droplet digital PCR (ddPCR) and drPCR, 17 samples from SCHH and 5 samples from SNUH (a total of 22 samples) were analyzed and compared.

[0133] In addition, to evaluate the utility of drPCR for HER2 testing in blood samples, the drPCR HER2 amplification results of plasma DNA and tissue DNA from 60 patients obtained from CNU Hospital were compared, and a cutoff for plasma was derived by referring to the tissue results.

[0134]

[0135] 4. Establishment of a drPCR-based assessment of HER2 status in FFPE tumor tissues from breast cancer patients

[0136] The SCHH cohort was selected as the training set, consisting of 19 cases with a HER2 IHC score of 0, 22 cases with an IHC score of 1+, 43 cases with an IHC score of 2+, and 19 cases with an IHC score of 3+ (Fig. 4A). drPCR was performed on all 103 cases, and when comparing the HER2 / CEP17 ratio based on the IHC score group, the HER2 / CEP17 ratio in drPCR increased statistically significantly as HER2 expression in IHC increased (Fig. 4B). FISH results were obtained in 102 out of 103 cases, while results could not be obtained in one case due to DNA degradation. The HER2 / CEP17 ratio in drPCR was also compared between the FISH-positive and FISH-negative groups, and a statistically significant difference was observed between the two groups as well (P < 0.001) (Fig. 4C). Next, the correlation between the HER2 / CEP17 ratio in HER2 FISH and the HER2 / CEP17 ratio values ​​in drPCR was evaluated, and a very high correlation was observed. (r 2(=0.910, P < 0.001) (Fig. 4D). In the representative case, drPCR showed a HER2 / CEP17 ratio value similar to FISH and detected high-level and low-level amplification of the HER2 gene (Fig. 4E). Using ROC curves, drPCR cutoff values ​​that showed the most agreement with IHC and FISH results, which are currently standard methods for determining HER2 expression, were presented. The cutoff calculated using continuous variables was 1.825, with an AUC of 0.993. ROC curves were plotted with representative cutoffs, and when evaluating the AUCs at 2.00, 1.95, 1.90, 1.825, and 1.80, the most suitable cutoffs were 1.90, 1.825, and 1.80, respectively (AUC=0.963) (Figs. 4F and G). Table 3 summarizes the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) for the cutoff used. At a cutoff value of 1.90, the sensitivity was 92.6%, specificity 100%, accuracy 98.1%, positive predictive value (PPV) 100%, and negative predictive value (NPV) 97.4%.

[0137]

[0138] 5. Multicenter validation of the accuracy and clinical utility of the established drPCR HER2 assay (High diagnostic accuracy of drPCR-based HER2 assessment in two independent validation cohorts)

[0139] The drPCR-based HER2 assay established in the training cohort was tested by applying the same method in two independent institutions (validation cohorts). The SNUH cohort (Validation set 1) included a total of 200 cases composed of IHC score groups of 0 (n = 32), 1+ (n = 33), 2+ (n = 105), and 3+ (n = 30). Among these groups, FISH was performed on the IHC 2+ group (n = 105); IHC scores of 0, 1+, and 2+ (negative FISH analysis) were classified as negative (no HER2 gene amplification), while IHC scores of 2+ (positive FISH analysis) and 3+ were classified as positive (HER2 gene amplification). gDNA was extracted from a total of 200 FFPE patient samples, drPCR was performed, and the results were compared with IHC or FISH readings. Samples inconsistent with IHC results were reanalyzed using FISH or SISH and compared with the original readings. For ambiguous samples that were inconsistent with IHC readings even in FISH / SISH, or showed different results between IHC and ISH tests, the results of IHC / ISH and drPCR were compared and verified using Next-Generation Sequencing (NGS) (Fig. 5A). In the SNUH cohort, the group with low HER2 expression identified by IHC staining showed a low HER2 / CEP17 ratio in the drPCR results, while the group with high HER2 expression showed a high HER2 / CEP17 ratio (Fig. 5B). When the ISH result was negative, the drPCR result showed a low HER2 / CEP17 ratio, and when the ISH result was positive, the drPCR result showed a high HER2 / CEP17 ratio (Fig. 5C). The HER2 / CEP17 ratio confirmed by ISH analysis showed a statistically significant linear correlation with the HER2 / CEP17 ratio confirmed by drPCR analysis, and a very strong positive correlation was confirmed (r.2 =0.772, P < 0.001) (Fig. 5D). As a result of ROC analysis to evaluate the usefulness and accuracy of the drPCR analysis results, the AUC value was 0.917 at HER2 / CEP17 ratio cutoff values ​​of 1.9 and 1.95, and the highest AUC value of 0.920 was found when the HER2 / CEP17 ratio cutoff was 2.0 (Fig. 5E). In the SNUH cohort, at a cutoff value of 1.90, the sensitivity was 84.1%, specificity 99.4%, accuracy 96%, positive predictive value (PPV) 97.4%, and negative predictive value (NPV) 95.7% (Table 3).

[0140] In the CNUH cohort (Validation set 1), another independent institution, a comparative analysis was also conducted between HER2 test results based on conventional IHC / SISH analysis and HER2 interpretation results evaluated via drPCR analysis. The CNUH cohort, comprising a total of 60 cases, consisted of 15 cases with a HER2 IHC score of 0, 8 cases with an IHC score of 1+, 15 cases with an IHC score of 2+, and 22 cases with an IHC score of 3+ (Fig. 5F). Similar to the previous two institutions, when comparing the HER2 / CEP17 ratio based on the IHC score group, the HER2 / CEP17 ratio from drPCR increased statistically significantly as HER2 expression in IHC increased (P < 0.001) (Fig. 5G), and a statistically significant difference in the distribution of the HER2 / CEP17 ratio from drPCR was also observed between the SISH-positive and SISH-negative groups (P = 0.005) (Fig. 5H). When evaluating the AUC at various cutoffs (2.00, 1.95, 1.90, 1.825, and 1.80) on the ROC curve, the highest AUC value of 0.963 was obtained at a cutoff of 1.90 or lower (Fig. 5I). In the CNUH cohort, the sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) for a cutoff value of 1.9 were 92.6%, 100%, 96.7%, 100%, and 94.3%, respectively (Table 3). The usefulness and accuracy of drPCR results for a total of 363 patients were further evaluated by aggregating results from three independent institutions; ROC analysis revealed that the highest AUC value of 0.942 was observed when the HER2 / CEP17 ratio cutoff was 1.9 (Fig. 5J). For a total of 363 cases across the three institution cohorts combined, at a cutoff value of 1.9, the sensitivity was 88.8%, specificity 99.6%, accuracy 96.7%, positive predictive value (PPV) 98.9%, and negative predictive value (NPV) 96.0% (Table 3).In summary, the results of this study suggest the possibility that the established drPCR-based HER2 assay can be applied as a standardized, objective, and accurate assay across multiple institutions.

[0141]

[0142] 6. Evaluation of Concordance Rate between drPCR and FISH, IHC, and NGS Results (High concordance of drPCR to FISH, IHC, and NGS results in the evaluation of HER2 status in breast cancer patients)

[0143] 1) To further validate the accuracy of the drPCR-based HER2 assay by comparing it with NGS, 40 samples from the SCHH cohort with matching FISH and drPCR results were selected, and targeted NGS was performed on them (Fig. 6A). Of these, 12 samples were tested using amplicon-based NGS (ThermoFisher Genexus platform), and the remaining 28 were tested using hybrid capture-based NGS (Illumina NextSeq platform). Among the 40 cases, 16 showed HER2 amplification in both FISH and drPCR analyses, while the remaining 24 cases did not show HER2 amplification. The cases with HER2 amplification were further classified into two groups. Nine cases with an IHC score of 3+ and a high HER2 / CEP17 ratio of 3.5 or higher in FISH / drPCR were classified as the high-level amplification group, while seven cases with an IHC score of 2+ or a HER2 / CEP17 ratio between 2.0 and 3.5 in FISH were classified as the low-level amplification group. When comparing NGS and FISH / drPCR results, agreement was found in 38 out of 40 cases, and discrepancies were observed in 2 cases, resulting in an overall agreement rate of 95.0%. Characteristically, in the high-level amplification group, NGS showed perfect agreement with FISH / drPCR results, whereas discrepancies were observed in the low-level amplification group or in cases without amplification (Fig. 6B, C). In other words, NGS showed a false negative in 1 out of 7 cases in the low-level HER2 amplification group and a false positive in 1 out of 24 cases without HER2 amplification. In addition, all discrepancies occurred in the hybrid capture-based NGS method.When these results are taken together, high agreement rates are observed among HER2 drPCR, FISH, and NGS, and it is shown that drPCR is more sensitive and accurate than NGS, particularly for low levels of HER2 gene amplification.

[0144]

[0145] 2) In the SNUH cohort, NGS analysis was performed on a total of 13 representative cases, including 8 cases that had discrepancies or disputes regarding the results of standard HER2 detection methods (IHC and ISH) and drPCR analysis, to compare the HER2 readings according to each test method (Fig. 7A). As can be seen in Fig. 7 and Table 4, among the total 13 cases in which NGS was performed, the HER2 readings were consistent between the NGS and drPCR test results in 12 cases, excluding 1 case (SNUH 7-17). In addition, regarding 8 cases (SNUH 2-4, 2-8, 3-10, 3-11, 7-17, 7-18, 7-19, 9-3) that showed discrepancies in interpretation between HER2 IHC and ISH tests and drPCR tests, the results of HER2 ISH retesting showed that in most cases, the original ISH reader and the new reader gave different HER2 diagnoses. When comprehensively considering the results of ISH retesting and NGS analysis in these cases, the objectivity of drPCR results compared to ISH could be verified (Fig. 7B, Table 4). These results confirmed, by comparing the agreement rate with NGS results, that subjective interpretation by readers can cause false positives and false negatives in IHC and ISH-based HER2 tests, and that misdiagnosis of HER2 caused by such factors can be compensated for by using the drPCR analysis method. In addition, unlike drPCR and ISH, NGS failed to detect HER2 gene loss, which indicates the relative accuracy of drPCR compared to NGS in HER2 CNV analysis (Table 4).

[0146]

[0147] 7. Detection of IHC false positives and verification of drPCR accuracy through re-evaluation of HER2 IHC results based on artificial intelligence (AI) analysis and comparative analysis with drPCR results

[0148] Among cases exhibiting discrepancies in interpretation between the current HER2 standard test and HER2 drPCR analysis-based HER2 evaluation results, cases with a HER2 IHC score of 3+ (positive) but determined to be HER2 negative by drPCR occurred at a frequency of approximately 20% within the HER2 IHC 3+ group at each institution. To analyze the cause of the discrepancy between IHC interpretation and drPCR results, the IHC results for these specimens were re-evaluated using an automated HER2 IHC score calculation system utilizing independent interpreters and an artificial intelligence (AI)-based algorithm analysis method (Fig. 8A). In the SNUH cohort, the results of the IHC tests were fully reviewed for a total of 200 cases, and unlike conventional visual assessments, differences were observed in the distribution pattern of patient numbers by HER2 IHC score (Figs. 8B and C). In visual assessment (some readings were modified from the original reading during the sample collection process), IHC Grade 0 was distributed as 16% (n = 32), IHC Grade 1+ as 16.5% (n = 33), IHC Grade 2+ as 52.5% (n = 105), and IHC Grade 3+ as 15% (n = 30). However, in the IHC re-evaluation through AI-based algorithm analysis, the grades of HER2 expression were interpreted as IHC Grade 0 as 23.5% (n = 47), IHC Grade 1+ as 35% (n = 70), IHC Grade 2+ as 31% (n = 62), and IHC Grade 3+ as 10.5% (n = 21) (Fig. 8C). This demonstrates that the objectivity of the evaluation results may be lacking due to subjective judgments among readers, which is a limitation of visual IHC testing.

[0149] In particular, re-examinations were conducted on cases originally classified as HER2 IHC 3+ that were inconsistent with the drPCR results, and it was confirmed that there were instances where IHC was overestimated in each cohort. Specifically, it was found that there were cases classified as HER2 positive despite being HER2 negative. Among the cases (n = 22) classified as 3+ by the standard IHC staining method but negative by drPCR, re-examinations revealed that all but two cases were re-re-classified as 2+, and the results analyzed using an AI-based algorithm analysis were also classified as 2+ or lower (Figs. 8D and 8E). When ISH was performed on these results, all but three cases with low tumor purity were analyzed as HER2 negative, which was confirmed to be exactly consistent with the drPCR results (Fig. 8E, Table 5). Consequently, among the HER2 IHC original reading grade 3+ cases (n = 94) in the three-institution cohort, 20.2% of the cases were determined to be HER2 IHC false positives (Figure 8F), and these results indicate that the drPCR method has higher accuracy and specificity than the standard IHC staining method, which can be interpreted as a false positive.

[0150]

[0151] 8. Effect of tumor purity on the drPCR-based HER2 copy number measurement

[0152] In previous studies on the development of HER2 diagnostic methods using digital PCR, it was reported that low tumor purity can affect digital PCR results. Therefore, the accuracy of drPCR-based HER2 testing was evaluated based on tumor purity. Tumor purity was measured by reviewing HE slides of tissues used for drPCR, and the ratio of tumor cells to total cells was evaluated using an optical microscope. Among a total of 363 cases in the three-institution cohort, tumor purity exceeded 25% in 307 cases and was 25% or less in 56 cases (Table 6). In the group with tumor purity of 25% or less, the rate of discrepancy increased when compared to the results of conventional FISH / IHC (Table 6, Fig. 9A). In cases where tumor purity exceeded 25%, the sensitivity was 91.8%, specificity 99.6%, accuracy 97.4%, positive predictive value (PPV) 98.7%, and negative predictive value (NPV) 96.9%. On the other hand, in cases with a tumor purity of 25% or less, sensitivity was 69.2%, specificity 100%, accuracy 92.9%, positive predictive value (PPV) 100%, and negative predictive value (NPV) 91.5%, confirming that sensitivity and accuracy decrease with lower tumor purity (Table 6).

[0153]

[0154] We measured the effect of selectively resecting (macrodissection) tumor portions and testing with drPCR in cases of low tumor purity. In two out of four SCHH cases and one out of three CNUH cases exhibiting low tumor purity and inconsistent drPCR and FISH results (FISH positive / drPCR negative), the drPCR results became positive after macrodissection, aligning with the FISH results (Table 7). We reviewed the slides of two SCHH cases where results remained unchanged after macrodissection. In one case, the number of tumor cells observed differed significantly between the HE slide and the subsequently produced HER2 IHC slide; the number of tumor cells present at the time of the initial HE slide production was greatly reduced by the time of the later HER2 IHC slide production. Considering that drPCR was performed after IHC, it is possible that there were almost no tumor cells in the specimen at the time of drPCR. The other inconsistent case showed severe lymphocyte infiltration surrounding the tumor cells. Even with macrodissection, there was a possibility that the results could not accurately reflect the HER2 status of the tumor cells due to the high number of lymphocytes infiltrating around the tumor (Fig. 9B). Additionally, in cases where tumor purity exceeded 25%, macrodissection did not significantly affect the HER2 / CEP17 ratio (Table 7). Furthermore, in cases with tumor purity of 25% or less, an ROC curve analysis was performed to determine the optimal cutoff for HER2 drPCR, yielding a cutoff of 1.385 (AUC = 0.942) (Fig. 9C). When this cutoff was applied to cases with tumor purity of 25% or less, among the 7 discrepancies caused by low tumor purity in the SCHH and CNUH cohorts, 4 drPCR readings were corrected from negative to positive, while 1 showed a borderline reading (Fig. 9D).

[0155] Overall, the results demonstrate that there is no problem in performing HER2 drPCR testing when the tumor purity exceeds 25%. Additionally, for clinical specimens with a tumor purity of 25% or less, we suggest methods to improve the accuracy of interpretation by performing macrodissection or applying a cut-off optimized for the criteria of the low tumor purity group (Fig. 9E). However, in order to establish optimal guidelines applicable in clinical practice, it is considered necessary to collect a sufficient number of HER2-positive specimens with low tumor purity for further verification.

[0156]

[0157] 9. Effect of intratumoral HER2 heterogeneity on the drPCR-based HER2 copy number measurement

[0158] Intratumoral HER2 heterogeneity, which is commonly observed in clinical practice, has been suggested to cause misinterpretations in conventional HER2 assays or to be the cause of discrepancies between digital PCR test results and conventional HER2 assay results. In the results of this study as well, high HER2 heterogeneity was observed in cases where there was a discrepancy between the standard HER2 assay and the drPCR analysis results (Figures 10A and 10B). When evaluating the distribution of tumor cells by HER2 IHC grade using an AI-based algorithm analysis on cases where there was a discrepancy between the HER2 expression grade confirmed by the standard assay and the drPCR analysis results, unlike cases with low HER2 heterogeneity where the proportion of a specific HER2 IHC grade was high (more than half), the IHC grades 0, +1, and +2 showed a diverse distribution pattern, and this HER2 heterogeneity can be considered one of the causes of the discrepancy with the drPCR analysis results.

[0159] In these samples, due to high levels of HER2 expression heterogeneity, discrepancies in interpretation were observed between existing IHC / ISH readings and drPCR test results, or conflicting evaluation results were shown between IHC and ISH tests depending on the reader, which suggests that HER2 expression heterogeneity may lead to HER2 IHC false positives or unclear ISH interpretations (Figs. 10 A and B).

[0160] Unlike IHC or ISH tests that interpret only a portion of tissue through a microscope, drPCR can quantify the amount of HER2 in the entire tumor, so drPCR-based HER2 testing is expected to provide more accurate results for cases with high HER2 expression heterogeneity.

[0161]

[0162] 10. Design of HER2 diagnostic methods using alternative CEP17 PnPs in patient populations with CEP17 genetic mutations (Alternative CEP17 reference controls for accurate measurement of HER2 copy number status in cases with CEP17 abnormalities)

[0163] Although clinically rare, abnormalities in the number of chromosome 17 or DNA copy number variations near the centromere, which occur in some patients, can affect the results of CEP17 used as a control in the diagnosis of HER2-induced HER2-induced HSH, potentially leading to misdiagnosis of HER2. In particular, while polysomy 17, in which the number of chromosome 17 itself increases, is known to occur almost infrequently in clinical practice, localized DNA copy number amplification around the chromosome 17 centromere has been reported at a frequency of 1-10% across various institutions and can lower the HER2 / CEP17 ratio, thereby affecting the diagnosis of actual HER2-positive cancers with HER2 overexpression. When analyzing the genetic mutation patterns of representative genes in each region of Chromosome 17 (17p11.2, 17q11.1, 17q11.2, 17q21) in the TCGA breast cancer cohort, gene amplification of genes near the centromere, excluding MTND4LP8 (17p11.2) and TUFMP1 (17q11.1), was found in approximately 1-5% of all breast cancer patients (Fig. 11A). In addition, the RAI1 (17p11.2), RARA (17q21), and EFTUD2 (17q21) genes on chromosome 17 have been proposed as controls for HER2 CNV analysis in HER2 ISH and drPCR assays, but the mutation frequency of these genes in breast cancer reached 1-6%. Among the genes located on chromosomes other than chromosome 17, the EIF2C1 (1p34), TFF3 (21q22), AP3B1 (5q14), PPIA (7p13), and RPP30 (10q23) genes, which were proposed as control groups for HER2 CNV analysis, also showed mutations of approximately 1-2% in breast cancer (Fig. 11B).However, no mutations were observed in the regions of the genes (MTND4LP8, TUFMP1) located in the alternative CEP17 PnP designed in this study (Figures 11A and B), which demonstrates the suitability of the newly derived CEP17 and alternative CEP17 PnP as HER2 CNV measurement controls.

[0164] In addition, this study was able to easily detect local variations around the centromere of chromosome 17 by utilizing CEP17 and alternative CEP17 PnP. In the three-institution cohort used in this study, cases with an increased CEP17 copy number of 3.5 or higher were also found, although at a low frequency of between 1–3% (Fig. 11C). In addition, we newly discovered a specific pattern of point mutations (A>G, chr17:26,982,760 or chr17:26,982,767bp) in the CEP17(q11.1) probe region designed by our research team in 16 cases (4.4%) out of a total of 363 cases in the three-institutional breast cancer cohort and in some breast cancer cell lines (HCC-1954, SKBR3) (Fig. 11D, Fig. 12A and B). In these samples, while the CEP17 conc value was not significantly affected in drPCR analysis, differences in fluorescence intensity due to mutations within the probe sequence appeared divided into two groups (Fig. 11D, Fig. 12B). We verified that using a probe in which the mutated sequence within the probe was replaced with the wild type in the SKBR3 cell line improved the issue of the probe's fluorescence intensity groups being divided during drPCR (Fig. 12C, D).

[0165] To devise a more accurate drPCR-based HER2 assay for breast cancer patients with CEP17 genetic mutations in various localized sites, the status of HER2 and CEP17 in these cases was precisely evaluated using alternative CEP17 PnPs that detect different CEP17 regions. In HER2-positive HCC-1954 and SKBR3 breast cancer cell lines, DNA copy number loss in the CEP17 p arm was observed along with point mutations in the CEP17 q arm (Fig. 12-E). In the case of one patient from the SCHH cohort (SCHH 4-9) who showed distinct CEP17 amplification in FISH testing, evaluation of HER2 / CEP17 drPCR results using four types of CEP17 PnPs verified that while there was localized gene amplification in the CEP17 q arm, there were no genetic mutations in the CEP17 p arm (Fig. 13-A). For other clinical cases suspected of CEP17 gene amplification, a more accurate assessment of HER2 status was possible by inferring the pattern of CEP17 using alternative PnP (Fig. 13A). In addition, for clinical samples with a point mutation in the CEP17(q11.1) region, it was verified that using CEP17 (Alt CEP17(p11.1)) in the p-arm region without the corresponding mutation improved the probe's fluorescence intensity issue and yielded more accurate results (Fig. 13B).

[0166]

[0167] 11. Detection of HER2 copy number loss in breast cancer patients using drPCR

[0168] Recently, trastuzumab deruxtecan (T-DXd, brand name: Enhertu), an antibody-drug conjugate (ADC) developed as a new HER2 targeted therapy, received FDA Emergency Approval after its efficacy was proven as the first treatment for HER2-low-expression metastatic breast cancer. Consequently, significant attention is being focused not only on the diagnosis of HER2-positive breast cancer but also on the accurate diagnosis of HER2-low-expression breast cancer. To diagnose HER2-low-expression breast cancer for the purpose of selecting patients for T-DXd treatment, it is necessary to clearly distinguish between the HER2-zero group and the HER2-low group; however, current IHC-based HER2 expression assessments lack objectivity in distinguishing between HER2 0 (zero) and 1+ (low), leading to inter-reader discrepancies and making accurate diagnosis difficult. Although the correlation between the HER2 gene copy number and T-DXd reactivity has not yet been clearly established, a recent report claimed that HER2 hemizygous deletions are found in some breast cancer patients and are highly associated with T-DXd drug resistance. Therefore, this study explored the possibility of detecting not only the amplification but also the loss of HER2 copy number based on drPCR.

[0169] In the TCCA breast cancer cohort, the HER2 gene copy number showed a very high correlation with mRNA expression levels (Fig. 14A). Although complete loss of the HER2 copy number (homozygous deletion, deep deletion) was not found in breast cancer, shallow deletions of the HER2 copy number were detected at a relatively high frequency of over 24% (Fig. 14B). In the three-institution cohort used in this study, when analyzing cases presumed to have a HER2 shallow deletion (defined as a HER2 / CEP17 ratio of 0.95 or less) based on HER2 drPCR analysis results, 23 out of a total of 363 cases showed a low HER2 / CEP17 ratio of 0.95 or less, and 73.9% of these were determined to have an IHC score of 0 in the AI-based HER2 IHC scoring (Fig. 14C). When comparing the HER2 / CEP17 ratios calculated by FISH and drPCR analysis in these cases presumed to have HER2 shallow deletion, most specimens showed HER2 / CEP17 ratios of less than 1, which were similar between the two tests (Fig. 14D, Table 8). Therefore, these results suggest that analyzing the loss of HER2 copy number through FISH or drPCR testing can be utilized as an auxiliary means to distinguish between the HER2-zero and HER2-low groups for the precise prediction of T-DXd drug responsiveness. In particular, unlike ISH, which is limited to IHC2+ testing due to the high time and cost involved, we suggest that if drPCR testing—which can measure HER2 status quickly and easily—is performed on all HER2 IHC score groups, it would be possible to correct false positives and false negatives in IHC and more accurately diagnose breast cancer with HER2 gene amplification or loss.

[0170]

[0171] 12. Comparative Analysis of drPCR-based HER2 Testing Methods in Matched FFPE Tissue and Plasma Samples from Breast Cancer Patients

[0172] In 60 breast cancer patients at CNU Hospital, plasma samples were obtained in addition to tissue specimens. To explore the potential for HER2 diagnosis via drPCR in liquid biopsies as well as tissues, the HER2 status of tissue and plasma specimens from the same patients was compared using drPCR. When HER2 status was evaluated using IHC and SISH, there were 27 HER2-positive patients and 33 negative patients (Fig. 15A). When comparing the accuracy of drPCR results with IHC / ISH of plasma specimens for each group, the positive group showed agreement in 4 out of 27 patients (14.8%), while the negative group showed 29 negative results out of 33 patients (87.9%), resulting in a total agreement rate of 55% for all patients (Figs. 15B and D). In one case, the IHC result was 3+ positive, but the drPCR performed on the tissue was negative. HER2 amplification was detected in drPCR performed on the plasma of the same patient (Fig. 15D). A review of the case revealed that lymphocytes had severely infiltrated around the tumor cells. Compared to the tissue sample, where the cutoff was 1.9, the highest agreement was observed in the plasma when the cutoff was set to 1.2 (Figs. 15C, D).

[0173] According to previous studies, the detection rate of HER2 gene amplification in cell-free DNA (cfDNA) of plasma is highest in metastatic HER2-positive breast cancer (approx. 33–96%) and relatively low in early-stage HER2-positive breast cancer (approx. 9–31%). The CNUH cohort utilized for plasma analysis in this study consisted of 29 patients in stage 1, 21 in stage 2, and 10 in stage 3, while stage 4 and metastatic breast cancer were excluded. The concordance rates between HER2 IHC / ISH and drPCR for stage 1, 2, and 3 patients were 62.1%, 52.4%, and 40.0%, respectively, indicating no difference in detection rates based on stage (Figures 15 E and F). These results suggest the feasibility of liquid biopsy-based HER2 drPCR diagnosis; however, since the probability of detecting HER2-positive cases is lower compared to HER2-negative cases, further research is needed to develop methods to technically compensate for this. In addition, since the cohort of this study consisted only of stage 1–3 cases with low HER2 gene amplification frequency in plasma, it is necessary to validate the liquid biopsy-based HER2 drPCR test method in a stage 4 metastatic HER2 breast cancer cohort.

[0174]

[0175] 13. Proposed HER2 testing algorithm using digital real-time PCR

[0176] In summary, based on the results of this study, HER2 drPCR testing can be proposed as a solution to address the shortcomings of the current two-step standard HER2 testing method (Fig. 16A), which is based on IHC and ISH tests. Both ISH and drPCR are testing methods with the same objective of diagnosing HER2 gene amplification, and since the equivalence between ISH and drPCR was proven in this clinical study, drPCR can be utilized as a substitute for ISH testing (Fig. 16B). However, unlike ISH, which is performed only in the HER2 IHC2+ (borderline) group, it is proposed to leverage the advantages of drPCR testing to perform drPCR diagnosis as an adjunct for all HER2 IHC score groups to correct false positive and false negative results that commonly occur in IHC diagnosis (Fig. 16B). Alternatively, a method can also be proposed in which drPCR is used as a one-step HER2 test that replaces both IHC and ISH tests, and the existing IHC / ISH test is used as an auxiliary test only when the drPCR test shows borderline results (Fig. 16 C).

[0177]

[0178] 14. Verification of the Applicability of drPCR HER2 Testing Using Tumor Tissue from Gastric Cancer Patients

[0179] In gastric cancer, the presence of HER2 gene amplification is essential for selecting patients for targeted therapy. Currently, patients with HER2 positivity in advanced gastric cancer use combination therapy of an anti-HER2 agent (trastuzumab) and chemotherapy as first-line treatment, and recently, the trastuzumab deruxtecan drug (T-Dxd) has also received FDA approval and is being used for treatment. Due to the characteristics of the tumor, gastric cancer exhibits severe heterogeneity in HER2 expression; it is reported that approximately 50% of cases show different HER2 amplification characteristics within the same tumor, and HER2 amplification results differ by 2–14% between the primary tumor and metastatic / recurrent lesions, making HER2 interpretation in gastric cancer patients extremely important. Accordingly, to verify whether the drPCR HER2 assay established for breast cancer is applicable to HER2 interpretation in gastric cancer patients, our research team compared the HER2 IHC and FISH interpretation results with the HER2 drPCR interpretation results for four FFPE cases of gastric cancer tumor tissues from SCH Hospital. The cohort consisted of HER2 IHC 0 (n = 1), 1+ (n = 1), and 2+ (n = 2) groups, and among them, FISH was performed on the IHC 2+ group to confirm and interpret the HER2 gene copy number. gDNA was extracted from four FFPE gastric cancer patient samples, drPCR was performed, and the previously established drPCR HER2 assay was applied for interpretation. The results are shown in Figure 17.

[0180] As shown in Figure 17, it was confirmed that all readings matched when compared with IHC or FISH readings.

[0181]

[0182] 15. Verification of Rapidity

[0183] Figure 18 illustrates a comparison of the analysis times of the existing Targeted NGS (NGS), Droplet digital PCR (ddPCR), and Digital real-time PCR (drPCR) according to the present invention. The NGS analysis method takes about 4 days, ddPCR takes about 5 hours, and drPCR takes about 1 hour.

[0184] In addition, the time required for HER2 detection is compared between the conventional method and the present invention, as shown in FIG. 19. When FISH or SISH is performed after IHC as in the conventional method, it takes about 3 to 4 days. On the other hand, the present invention takes about 1 to 2 days by performing drPCR after IHC.

[0185]

[0186] 16. Accuracy Verification

[0187] 16-1. Verification of drPCR accuracy compared to ISH for HER2

[0188] Among the Seoul National University Hospital cohort (total 200 cases), for cases where the interpretation of the HER2 result was indeterminate (ID) due to discrepancies between the original IHC / ISH-based reading and the drPCR result, or inter-variability in the IHC / ISH test, IHC and ISH re-examinations (visual reading: participation of multiple readers / utilization of AI-based IHC scoring) were performed, and NGS was also conducted as an auxiliary means for HER2 CNV evaluation to accurately predict the HER2 status. The negative or positive status is shown in Figure 7 and Table 4.

[0189] Among the 8 total cases in the HER2 IHC-equivocal (2+) group where there was a discrepancy between the ISH and drPCR results based on the original reading, 6 cases showed different interpretations among readers during the ISH re-examination (SNUH 2-4, 2-8, 3-11, 7-17, 7-18, 7-19). These cases, which were HER2 negative in drPCR, were positive in the original ISH reading but were determined to be ISH negative in a re-examination by an independent reader. In the NGS analysis, all were verified to have no HER2 gene amplification except for one case (7-17, ambiguous positive in NGS), so they were presumed to be ISH false-positive cases. 3-10 and 9-3 are the two cases where the original ISH reading and the re-examination results were consistent, but the drPCR results were ultimately inconsistent; however, since the NGS analysis was consistent with the drPCR results, it suggests the possibility of ISH false-positives.

[0190] When predicting the actual status of HER2 by considering both ISH retest and NGS results, the accuracy of drPCR in the Seoul National University Hospital cohort increased from 96% (192 / 200 matches) to 98.5% (197 / 200 matches) (Fig. 7B).

[0191] Most of the cases showing different interpretations among readers in the HER2 ISH test had intratumoral HER2 heterogeneity. It was confirmed that the likelihood of different interpretations increases depending on the area observed under the microscope, and the related results are shown in Figure 10.

[0192] A major problem with the current standard HER2 diagnostic methods, IHC / ISH, is that objectivity may be lacking due to the nature of these tests relying on visual interpretation. Furthermore, the subjectivity of pathologists can intervene, leading to conflicting interpretation patterns across institutions, readers, and samples (inter- and intra-observer variability). It is known that this variability is particularly high in the IHC borderline group (2+) and in groups possessing intratumoral HER2 heterogeneity. The Seoul National University Hospital (SNUH) cohort utilized in this study had a higher proportion of the IHC borderline group (2+ / equi) compared to the cohorts of the other two institutions (105 of 200, 52.5%; Fig. 3), and inter-reader discrepancies in interpretation were observed in several cases during ISH testing (especially in cases with high intratumoral heterogeneity) (Figs. 7, 10, Table 4).

[0193] Compared to the ISH test, which observes a portion of the tumor under a microscope and counts only 20 cells, drPCR, which observes the pattern of HER2 in the entire tumor, can yield relatively more accurate results in cases with high intratumoral HER2 heterogeneity, thus demonstrating the possibility of misdiagnosis in ISH and the objectivity of drPCR results.

[0194]

[0195] 16-2. Verification of HER2 IHC False Positives & drPCR Accuracy

[0196] In the current IHC / ISH-based standard diagnostic method for HER2, the HER2 IHC 3+ group is classified as HER2-positive without performing ISH. Although ISH is relatively more accurate than IHC in HER2 evaluation, due to the inconvenience, high cost, and time-consuming nature of ISH, the current standard diagnostic method performs ISH only on the IHC 2+ group (Fig. 16A). These results suggest that false positives can occur frequently in the IHC 3+ group (leading to wrong anti-HER2 treatment) and that drPCR is more accurate than IHC in HER2 diagnosis (Fig. 8, Table 5). Furthermore, considering the rapidity and convenience of drPCR compared to ISH, drPCR is expected to be very useful as a HER2 diagnostic tool capable of correcting false positives in IHC.

[0197] Therefore, to prevent IHC 3+ false positives, it is appropriate to perform drPCR simultaneously with IHC and utilize it as an auxiliary means to verify the results of IHC (Fig. 16B).

[0198]

[0199] 16-3. Comparison of drPCR and ddPCR Results

[0200] To compare the HER2 CNV detection accuracy between ddPCR and drPCR, 22 representative cases with various levels of HER2 CNV (high level amp, low level amp, no alteration, deletion) were selected (Table 9).

[0201]

[0202] As shown in Table 9, for all cases analyzed in comparison (n=22), drPCR accurately detected HER2 CNV, whereas ddPCR failed to detect HER2 CNV in one case (SCHH 6-12) with low HER2 amplification. Additionally, in cases where the HER2 gene was excessively amplified (SNUH 9-5, HER2 / CEP17=54.99), drPCR accurately detected HER2 over-amplification without changing PCR conditions, whereas ddPCR required changes in PCR parameters and DNA dilution to detect it (data not shown).

[0203] Since the agreement rate between ddPCR and drPCR, known as the gold standard of digital PCR (dPCR), is high (21 of 22; over 95%), the present invention can be applied to HER2 and CEP17 P&P not only to drPCR but also to various dPCR platforms. In some cases (low HER2 amp levels or abnormally high HER2 amp levels), drPCR shows higher performance compared to ddPCR. Table 9 confirms the superior speed and sensitivity of dePCR compared to ddPCR.

[0204] Therefore, the relative accuracy of drPCR, along with its speed, is proven compared to ddPCR.

[0205]

[0206] 16-4. Comparison of HER2 drPCR vs. NGS

[0207] To cross-validate the accuracy of drPCR-based HER2 CNV detection with NGS, 40 representative cases (HER2 status criteria: HER2-pos, n=16; HER2-neg, n=24 / HER2 amp level criteria: high level amp, n=9; low level amp, n=7; no amp, n=24) were selected from the training cohort (Soonchunhyang University Hospital, SCHH; total 103 cases) and NGS was performed, after which the results were compared with ISH and drPCR results (Fig. 6A). In the SNUH cohort, a total of 13 representative cases, including those with controversial determinations, were selected and NGS was performed to evaluate the reliability of drPCR (Fig. 7).

[0208] Among the 40 cases of NGS cross-validation in SCHH, HER2 readings were consistent across ISH, drPCR, and NGS in 38 cases, showing a high concordance rate of 95% (Figs. 6B and C). The remaining two cases showed a pattern of agreement between ISH and drPCR, but disagreement between ISH / drPCR and NGS. NGS failed to detect HER2 low amps detected by ISH and drPCR, and NGS showed positive results in cases interpreted as HER2 negative by ISH / drPCR. Among the 13 cases analyzed in the NGS analysis of the SNUH cohort, drPCR and HER2 positive / negative results were consistent in 12 cases, but NGS failed to detect CN deletions in the HER2-negative group (Fig. 7B, Table 4).

[0209] NGS misdiagnosis: In representative cases of the SCHH cohort, NGS detected HER2 CNV in 100% of the high HER2 amp group (9 / 9, 100%). However, an NGS false negative was found in one case with low HER2 amp (1 of 6), and an NGS false positive was found in one HER2-negative case without HER2 amp (1 of 24) (Fig. 6B and C).

[0210] In addition, for two representative cases (SNUH 2-10, 5-1) in the SNUH cohort that were proven to have a HER2 CN deletion by ISH and drPCR tests, NGS failed to detect the HER2 CN deletion (Table 4).

[0211] NGS-based HER2 CNV analysis has been proposed as an auxiliary or alternative diagnostic method for HER2 IHC / ISH, but NGS has several limitations, such as high cost, long processing time, non-standardized HER2 cut-off, and lack of clinical validity, so it has not yet been clinically approved as a HER2 diagnostic method.

[0212] The present invention proposes the advantages of drPCR compared to NGS, such as low cost, rapid and simple analysis, the establishment of a standardized drPCR analysis method through multi-center clinical validation in this study, and higher accuracy than NGS in HER2 CNV detection, as well as the potential for clinical application.

[0213]

[0214] 16-5. Verification of Sensitivity, Specificity, PPV, NPV, and Accuracy of HER2 drPCR

[0215] When evaluating drPCR-based HER2 diagnostic methods conducted at three independent institutions (comparative analysis of concordance rates with current HER2 diagnostic IHC / ISH), all three institutions showed high levels of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy (Table 3). Although Seoul National University Hospital showed relatively lower accuracy compared to the other two institutions, this was based on the original readings of IHC / ISH; if the verification results of discrepancies (ISH / ISH retesting, NGS verification) were also taken into account, it would show a higher level of accuracy (Table 3 and Fig. 7).

[0216] According to Table 3, since the accuracy of the HER2 drPCR test method according to the present invention is confirmed, a standardized HER2 drPCR test method can be established through a multi-center clinical trial.

[0217]

[0218] Overall, the HER2 drPCR test method invented by this institute demonstrates superiority in terms of speed, convenience, and accuracy compared to existing HER2 test methods. Furthermore, by precisely identifying the genetic mutation patterns of HER2 and CEP17 through newly derived HER2, CEP17, and alternative CEP17 PnPs, it presents a method to prevent misdiagnosis of HER2 testing that may be caused by localized DNA amplification of CEP17. This drPCR-based precise analysis of HER2 and CEP17 copy numbers can be utilized in cancer diagnosis for classifying HER2-positive and HER2-negative groups for the selection of patients for HER2 targeted therapy, classifying HER2-low expression and HER2-zero groups within the HER2-negative group, and classifying HER2-high amplification groups within the HER2-positive group. It is also expected to be utilized for analyzing abnormal patterns on chromosome 17 (Fig. 20).

[0219]

[0220] As mentioned above, this study developed a novel HER2 diagnostic method capable of simply, rapidly, and precisely detecting HER2 gene amplification using LOAA digital real-time PCR (drPCR), a new digital PCR platform that enables ultra-high-speed, precise gene mutation detection through real-time analysis that overcomes the limitations of existing digital PCR technology. Furthermore, the clinical utility of the drPCR-based HER2 diagnostic method developed in this study was verified through a multi-center clinical trial, thereby establishing a standardized testing method. All three independent institutions involved in the verification demonstrated a high agreement rate in readings compared to existing HER2 testing methods, and the method was also effective in correcting cases of false positives or inter-reader discrepancies in evaluations associated with existing methods. Additionally, by significantly reducing the testing time compared to existing methods while providing objective and quantified HER2 copy number values, the method demonstrated the advantage of easily and quickly identifying HER2-positive cancer patients with HER2 gene amplification. Therefore, the ultra-fast HER2 precision diagnostic method using the drPCR technique developed in this study is equivalent to current HER2 testing methods and is expected to be widely used in clinical practice as a new diagnostic method that can replace or complement existing testing methods.

Claims

1. A polynucleotide set for detecting the HER2 gene, comprising the primer of SEQ ID NO. 1, the primer of SEQ ID NO. 2, and the probe of SEQ ID NO. 3, and A composition for cancer diagnosis for digital PCR comprising any one polynucleotide set for CEP17 detection selected from the group consisting of: a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 4, a primer of SEQ ID NO. 5, and a probe of SEQ ID NO. 6; a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 7, a primer of SEQ ID NO. 8, and a probe of SEQ ID NO. 9; a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 10, a primer of SEQ ID NO. 11, and a probe of SEQ ID NO. 12; and a polynucleotide set for CEP17 detection comprising a primer of SEQ ID NO. 13, a primer of SEQ ID NO. 14, and a probe of SEQ ID NO.

15.

2. In Paragraph 1, A composition for cancer diagnosis in which the above digital PCR is a digital real-time PCR.

3. A cancer diagnostic composition according to claim 1 or 2, wherein the cancer is one or more selected from the group consisting of gastric cancer, breast cancer, ovarian cancer, adenocarcinoma, endometrial cancer, prostate cancer, colorectal cancer, pancreatic cancer, lung cancer, gastroesophageal cancer, and bladder cancer.

4. A cancer diagnostic composition according to claim 1 or 2, wherein the composition detects a HER2 gene copy number variation (CNV).

5. A cancer diagnostic composition according to claim 1 or 2, wherein the 5' end of the probe is labeled with a fluorescent material.

6. A cancer diagnostic composition according to claim 5, wherein the fluorescent material is one or more selected from the group consisting of VIC, HEX, FAM, and EverGreen dye.

7. A cancer diagnostic kit comprising the composition of claim 1 or 2.

8. A cancer diagnostic kit according to claim 7 that detects HER2 gene copy number variation (CNV).

9. A cancer diagnostic kit according to claim 7, wherein the cancer is one or more selected from the group consisting of stomach cancer, breast cancer, ovarian cancer, adenocarcinoma, endometrial cancer, prostate cancer, colorectal cancer, pancreatic cancer, lung cancer, gastroesophageal cancer, and bladder cancer.

10. A liquid sample injection step of injecting a liquid sample into an inlet and filling a plurality of wells containing the composition according to claim 1; A step of bringing the well injected with the liquid sample into close contact with the CMOS photosensor; A step of detecting a fluorescent signal using the above CMOS photosensor; A Ct calculation step for calculating one or more cycle threshold values ​​(Ct) for each of the plurality of wells from the detected fluorescence signal; and A method for providing information for the diagnosis of cancer using digital PCR, comprising the step of detecting a HER2 copy number variation based on the above cycle threshold value.

11. In Paragraph 10, A method for providing information for the diagnosis of cancer using digital PCR, wherein the above-mentioned information providing method further includes a step of confirming the loss of HER2 gene copy number or high-level amplification of HER2 gene copy number.

12. In Paragraph 10 or 11, A method for providing information for the diagnosis of cancer using digital PCR, wherein the above digital PCR is a digital real-time PCR.

13. In Paragraph 10 or 11, A method for providing information for the diagnosis of cancer, wherein the above cancer is one or more selected from the group consisting of stomach cancer, breast cancer, ovarian cancer, adenocarcinoma, endometrial cancer, prostate cancer, colorectal cancer, pancreatic cancer, lung cancer, gastroesophageal cancer, and bladder cancer.

14. A cancer diagnostic composition according to claim 1 or 2, wherein the composition detects local copy number variation (CNV) and point mutation in the centromeric / pericentromeric DNA region of chromosome 17.

Images