ES2971182A1 - Sequencing panel for liquid biopsy of breast cancer patients (Machine-translation by Google Translate, not legally binding) - Google Patents

Abstract

Massive sequencing panel composed of 37 genes for liquid biopsy that allows early diagnosis of breast cancer and methods for obtaining useful data for the diagnosis and monitoring of the disease. (Machine-translation by Google Translate, not legally binding)

Description

DESCRIPTION

Sequencing panel for liquid biopsy of breast cancer patients

FIELD OF INVENTION

The present invention is within the field of medicine and oncology, and refers to a sequencing panel for tumor tissues and liquid biopsy of patients with breast cancer.

STATE OF THE ART

We call breast cancer the tumors that develop in the breast tissues. There are two main types of breast cancer: ductal carcinoma, which begins in the ducts that carry milk from the breast to the nipple and is the majority, and lobular carcinoma, which begins in the lobes that produce milk. To detect breast cancer, different tests are used such as mammography, breast ultrasound with high-resolution transducers (ultrasounds) or magnetic resonance imaging. To date, the diagnosis of breast cancer could only be confirmed through a tumor biopsy.

Liquid biopsy is a strategy with great potential in the management of cancer, including diagnosis, early detection of relapses or allowing monitoring of the state of the disease throughout treatment to see its evolution. This is a test that is performed using body biofluids, most commonly blood, although they can also be analyzed in other body fluids such as the one that covers the meninges, the one that covers the lung, saliva or even urine, in order to to look for cancer cells from a tumor or DNA fragments from tumor cells, the so-called circulating tumor DNA (ctDNA). Unlike tissue biopsy, it allows molecular analysis to be performed in a much simpler and faster way, since a blood test can be done at any time, and it can also be repeated to monitor the evolution of the tumor.

What is intended to be identified in the samples are the different mutations specific to the tumor, since these mutations are not present in the rest of the normal cells in the blood.

Sequencing panels are genetic tests that encompass genes involved in the same pathology, in this case, aimed at breast cancer.

The detection and characterization of circulating tumor materials presents difficulties in an early setting due to their low quantities in patients' biofluids. However, the advance in the detection techniques currently used, very sensitive, allows that, even if there is a very small amount of ctDNA that exists in the blood or in other biofluids, it can be detected and analyzed, being able to characterize the genetics of the tumors.

The use of ctDNA together with state-of-the-art ultrasensitive NGS (next generation sequencing) technologies, also using patient-specific panels, has served to characterize tumors, evaluate responses to neoadjuvant chemotherapies and detect MRD (minimal residual disease) after Surgery.

Targeted NGS such as SafeSEQ - Safe-sequencing system have recently emerged as a new tool to increase the sensitivity of massively parallel sequencing system instruments for the identification of rare variants in plasma DNA. This sequencing strategy uses single-molecule barcodes prior to PCR amplification to reduce sequencing error and increase precision. The SafeSEQ, or secure sequencing, system can be used to distinguish between rare mutations and technical errors and is capable of detecting mutations that other technologies would miss. SafeSEQ has so far been applied to plasma samples from metastatic colorectal cancer and patients with gastrointestinal stromal tumors.

Other NGS technologies include gene region enrichment probes such as Agilent SureSelect, which can be used to design custom hybrid capture panels and mass sequence any coding or non-coding region of the genome or transcriptome.

All of this has contributed to the development of liquid biopsy techniques. Massive sequencing (NGS) procedures and different technologies are currently used to detect ctDNA at very low dilution (1). However, it is still unclear whether molecular alterations in ctDNA can be detected by NGS-based assays of plasma ctDNA in patients with primary breast cancer at screening and how they relate to the tumor tissue mutation found in the same patient.

The advantages of this type of liquid biopsy in the monitoring and prediction of metastatic breast cancer (MBC) have been proven by various studies (2), highlighting the importance of ctDNA as a potential biomarker for the evaluation, prediction and clinical management in MBC patients treated with chemotherapy. In this sense, ctDNA monitoring during treatment can significantly improve outcomes by predicting disease progression four to six months earlier than conventional methods (3).

Several studies have been carried out that try to develop a method for monitoring breast cancer using the sequencing of ctDNA molecules derived from a sample obtained from the patient and subsequent analysis of said ctDNA sequences to identify their mutations. We have an example of these sequencing panels in WO2017181146A1 where the following genes are included: AKT1, ALK, APC, ATM, BRAF, CTNNB1, EGFR, ERBB2, ESR1, FGFR2, GATA3, GNAS, IDH1, IDH2, KIT, KRAS, MET, NRAS, PDGFRA, PIK3CA, PTEN, RB1, SMAD4, STK11 and TP53y that allows detecting mutations at an allelic frequency of up to 0.025%. Another example in this sense is described in document US2021343372A1 where in a liquid biopsy sample in patients with breast cancer, mutations in TP53 are identified for 51.1% of patients, and with a lower incidence, in PIK3CA, ESR1, BRCA2, NF1 , ATMyAPC.The allele frequencies detected are of the order of 0.1%, 0.25% and 0.5%.

Other gene panels for ctDNA analysis and identification of metastatic breast cancer (MBC) include the following genes: AKT1, AR, BRCA1, BRCA2, EGFR, ERCC4, ERBB2, ERBB3, ESR1, FGFR1, KRAS, MUC16, PIK3CA , PIK3R1, PTEN, PTGFRyTGFB1(4).

The suitability of ctDNA analysis in the early diagnosis of breast cancer has also been assessed, comparing mutations in PIK3CA and TP53 between tumor biopsies and pretreatment plasma DNA (5), revealing four additional mutations, three in TP53 and one in PIK3CA, not previously identified in the tissue biopsy. of three patients.

Taking into consideration the documents cited in the state of the art, it is clear the great interest that exists in the development of liquid biopsy techniques and ctDNA analysis, in order to build personalized sequencing panels for the early diagnosis of breast cancer. breast and that are useful in the evaluation and monitoring of the disease, including the possibility of obtaining predictions about the progression of the disease and making personalized therapeutic decisions.

DESCRIPTION OF THE INVENTION

In the present invention, a massive sequencing panel is proposed for the early diagnosis of breast cancer, useful in the evaluation and monitoring of the disease, composed of 37 genes (see Table 1).

These genes have been selected following the criteria set out below:

1. Genes with mutations in >=1% of breast tumor samples included in the TCGA database (https://www.cbioportal.org).

2. Genes sequenced in a seminal article on breast tumors and that presented mutations in this tumor type (6).

3. Genes with biological relevance in breast cancer and/or cancers in general and not based on their mutational frequency.

Table 1: Genes included in the custom massive sequencing panel using Agilent SureSelect XT HS technology.

All genes in the table have known sequences, which are public and accessible through public databases, so they do not require further characterization herein so that an average expert in the field can reproduce the invention.

The genomic coordinates of the regions that have been included in the panel are shown in the following Table 2.

Table 2: Genomic coordinates (human genome version=hg38):

Thanks to the use of molecular barcodes, this panel allows the ultrasensitive characterization of tumor genetic aberrations, so it can be used for the detection of mutations at allelic frequencies below 0.15%.

This high sensitivity allows detection of mutations in biofluids (liquid biopsy) from patients with localized/early tumors or those cancers with little release of tumor material. This panel has been optimized and validated using four breast tumor samples with known mutations (see example 1) and its results have been compared with other sequencing panels for this same type of cancer, breast cancer, showing greater detection capacity. of mutations (see example 2). Its use in liquid biopsies has been supported in comparative studies with tumor samples, where the high sensitivity, specificity and predictive value capacity of the breast cancer diagnosis method based on this sequencing panel is demonstrated (see example 3). Its usefulness as a marker of prognosis or progression of the disease has also been proven, through its correlation with clinicopathological variables (see example 4).

Therefore, a first aspect of the invention refers to the sequencing panel comprising the genes of Table 1.

Specifically and more preferably, to their use as biomarkers to diagnose, early diagnose, predict or forecast the evolution of breast cancer. And even more preferably, for use in liquid biopsy.

Methods of the invention

Another aspect of the invention refers to an in vitro method of obtaining data useful for diagnosing, early diagnosing, predicting or forecasting the evolution of breast cancer in an individual, hereinafter first method of the invention, which comprises:

a) obtain a biological sample isolated from the individual,

b) identify possible mutations in the genes in Table 1.

Another aspect of the invention relates to an in vitro method for diagnosing or early diagnosis of breast cancer, hereinafter second method of the invention, comprising steps (a) - (b) according to the first method of the invention. invention, and also includes:

c) assign the individual who has mutations in any of the genes in Table 1 to the group of patients suffering from breast cancer.

In a preferred embodiment of the second method of the invention it further comprises assigning the individual who has a higher median allele frequency of the mutations to the group of patients with a higher tumor grade.

Another aspect of the invention refers to an in vitro method to predict or forecast the evolution of breast cancer, which comprises the method of obtaining useful data of the invention, and which also comprises assigning the individual who has more than one mutation, to the group of patients with a probability of clinical relapse.

Another aspect of the invention refers to the in vitro method to predict or forecast the evolution of breast cancer, which includes the method of obtaining useful data of the invention and which also includes assigning the individual who has mutations in TP53 to the group of patients with a negative tumor. expression of hormone receptors.

A "biological sample" as defined herein is a small part of a subject, representative of the whole, and may consist of a biopsy or a sample of body fluid.

Preferably, the biological sample is plasma, coming from the physical separation of the blood sample taken from the patient.

Detection of mutations in the genes of Table 1 can be carried out by any means known in the state of the art.

Preferably it is carried out using next-generation sequencing (NGS) technologies. More preferably the NGS technology used is Agilent SureSelect type enrichment probes.

The method of the present invention can be applied with samples from individuals of any sex, that is, men or women, and at any age. Preferably applied to samples obtained from women.

In the present invention, "prognosis" is understood as the expected evolution of a disease and refers to the assessment of the probability according to which a subject suffers from a disease, as well as the assessment of its onset, state of development, evolution, or of its regression, and/or the prognosis of the course of the disease in the future. As experts in the field will understand, such an assessment, although preferred, may not normally be correct for 100% of the subjects to be diagnosed. The term, however, requires that a statistically significant portion of subjects can be identified as suffering from the disease or having a predisposition to it. Whether a part is statistically significant can be determined simply by the person skilled in the art using various well-known statistical evaluation tools, for example, determination of confidence intervals, determination of p-values, Student's t test, Mann test -Whitney, etc. Preferred confidence intervals are at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%. The p values are preferably 0.2, 0.1, 0.05. In turn, taking into account the method of the present invention, other subclassifications could be established within this main one, facilitating, therefore, the choice and establishment of appropriate therapeutic regimens or treatment. This discrimination, as understood by an expert in the field, is not intended to be correct in 100% of the samples analyzed. However, it requires that a statistically significant number of the analyzed samples be correctly classified. The amount that is statistically significant can be established by a person skilled in the art through the use of different statistical tools, for example, but not limited to, by determining confidence intervals, determining the significance P value, Student's test or discriminant functions. Fisher, non-parametric Mann Whitney measurements, Spearman correlation, logistic regression, linear regression, area under the ROC curve (AUC). Preferably, the confidence intervals are at least 90%, at least 95%, at least 97%, at least 98% or at least 99%. Preferably, the p value is less than 0.1, 0.05, 0.01, 0.005 or 0.0001. Preferably, the present invention allows the disease to be correctly detected differentially by at least 60%, more preferably by at least 70%, much more preferably by at least 80%, or even much more preferably by at least 90%. of the subjects of a certain group or population analyzed.

Kit and uses

Another aspect of the invention relates to a kit or device, hereinafter kit or device of the invention, which comprises the elements necessary to detect mutations in the genes of Table 1 simultaneously.

Automation of the method of the invention by implementing it in a computer program

Another aspect of the invention relates to a computer program comprising instructions for carrying out the procedure according to any of the methods of the invention.

In particular, the invention encompasses computer programs arranged on or in a carrier. The carrier can be any entity or device capable of supporting the program. As a variant, the carrier could be an integrated circuit in which the program is included and which has been adapted to execute, or to be used in the execution of, the corresponding processes.

For example, the programs could be embedded in a storage medium, such as a ROM memory, a CD ROM memory or a semiconductor ROM memory, a USB memory, or a magnetic recording medium, for example, a floppy disk or a disk. hard. Alternatively, the programs could be supported on a transmissible carrier signal; For example, it could be an electrical or optical signal that could be transported via electrical or optical cable, by radio or by any other means.

The invention also extends to computer programs adapted so that any processing means can implement the methods of the invention. Computer programs also include cloud applications based on this procedure.

Other aspects of the invention relate to the readable storage medium and the transmissible signal comprising program instructions necessary for the execution of the inventive method by a computer.

Throughout the description and claims the word "comprises" and its variants are not intended to exclude other technical characteristics, additives, components or steps. For those skilled in the art, other objects, advantages and features of the invention will emerge partly from the description and partly from the practice of the invention. The following examples and drawings are provided by way of illustration, and are not intended to be limiting of the present invention.

DESCRIPTION OF THE FIGURES

Figure 1. Mutations found in plasma samples from patients with localized breast cancer. A: Mutations identified in plasma samples in both the tumor genetics-informed and non-informed substudies. B: clinicopathological characteristics of the patients showing the current clinical status (clinical relapse).

Figure 2. Mutations identified in the tumor DNA sequencing of the samples included in the study. Oncoplot. A: mutations observed in tumor samples indicating whether they were found in plasma. B: clinicopathological characteristics of the patient.

EXAMPLES OF THE INVENTION

Example 1: Optimization and validation of the gene panel in Table 1

The panel of 37 genes involved in breast cancer in Table 1 was evaluated by generating sequencing libraries with the DNA obtained from 4 breast tumors and comparing the results with another panel, which uses a different technology.

Next generation ultrasensitive technologies (NGS - next generation sequencing) were used for sequencing, specifically SureSelect XT HS. It was verified that the coverage of the genomic areas was correct and known mutations were detected in these tumors, which had been previously sequenced with another technology, called TruSeq from Illumina.

Table 3: Mutations observed in the four tumors tested using the custom panel of 37 genes (Sure Select

ND, Not detected; NT, Not tested, the panel does not include the gene, VAF-Variant allelic frequency of a variant.

Comparing the mutations found using both panels in the 4 test tumors, completely comparable allelic frequencies were observed for the mutations that were detected by both panels, therefore validating the panel of the invention.

In addition, a mutation with low allelic and pathogenic frequency was detected in the CDH1 gene that was not detected in the previous sequencing.

It should be noted that using the panel of the invention new mutations were detected because this panel includes the sequencing of many more genes.

Example 2: Sensitivity of the gene panel in Table 1 for use as breast cancer biomarkers in liquid biopsy.

Study design and patient population

The study included 29 patients with BI-RADS 4C/5 mammography findings and subsequent diagnosis of primary breast cancer at the Virgen de la Victoria Hospital in Malaga and the University Clinical Hospital in Valencia. These patients were the same as those studied in Jimenez Rodriguez, B. et al in “Detection of TP53 and PIK3CA Mutations inCirculating Tumor DNA Using Next-Generation Sequencing in the Screening Process for Early Breast Cancer Diagnosis”(5)

Blood samples were collected immediately (less than one hour) before tissue biopsy and before receiving any type of treatment. Matching tumor tissues were collected through core needle biopsies of breast tumors, which were subsequently fresh frozen.

Immunohistochemical (IHC) analysis was performed to quantify the expression of human epidermal growth factor receptor 2 (HER2), hormone receptors (HR), and Ki67.

Estrogen receptor (ER) and progesterone receptor (PR) were considered positive if tumors had more than 1% cells with nuclear staining.

HER2 staining was graded on a scale of 0 to 3+, according to HercepTest guidelines. HER2 status was considered positive when it was scored as 3+, while 0 to 1+ was negative and 2+ was an inconclusive result and in those cases silver in situ hybridization (SISH) was performed to confirm positivity.

Hormone receptor-positive tumor plus a Ki 67 index of <14% was considered a luminal A tumor, while >14% was considered a luminal B tumor.

IHC breast cancer subtypes were defined by a combination of these IHC markers as follows: luminal A (PR positive, HER2 negative and Ki-67 < 14%), luminal B (ER positive and/or PR positive, HER2 negative and Ki-67 > 14%), HER2 positive (ER negative, PR negative, HER2 positive) and triple negative (ER, PR and HER2 negative).

The study was approved by the research ethics committees of the Hospital Virgen de la Victoria and the Hospital Clínico de Valencia; all patients gave their written informed consent. The research was carried out in accordance with Good Clinical Practice and the Declaration of Helsinki.

The clinicopathological characteristics of the tumors of the enrolled breast cancer patients are shown in Table 4. The average age of the patients was 64 years (range 44-92 years). The 29 primary breast tumors examined included 24 invasive ductal carcinomas (IDC) (82.7%), two invasive lobular carcinomas (ILC) (6.9%), two papillary carcinomas (6.9%), and one tubular carcinoma ( 3.4%). Fourteen tumors had diameters less than 2 cm (48.3%), while 15 had diameters greater than 2 cm (51.7%). Axillary lymph node metastasis was detected by core needle biopsy in nine patients (31.0%).

Ten patients without lymph node metastases had tumors with diameters of more than 2 cm (34.5%). Immunohistochemistry was used to determine IHC tumor subtypes as follows, 14 luminal A tumors (48.3%), 11 luminal B tumors (37.9%), three triple-negative tumors (10.3%), and one tumor HER2 positive (3.4%). Regarding hormone receptor status, 25 of 29 BC patients were estrogen receptor (ER) positive (86.2%) and 17 of 29 patients were progesterone receptor (PR) negative (58.6%).

Regarding tumor grade, the majority of patients had grade 2 tumors (16 patients, 55%) while seven patients had grade 3 tumors (24.5%) and five patients had grade 1 tumors (17%). . The tumor grade was unknown in one patient (3.5%).

Table 4: Clinicopathological characteristics of the patients.

CDI: Invasive ductal carcinoma; ILC: invasive lobular carcinoma; HER2: human epidermal growth factor receptor 2; IHC: Immunohistochemistry; BIRADS: Breast Imaging Data and Reporting System.

Extraction of DNA

Tumor DNA was isolated from fresh frozen tissue samples using the DNeasy blood and tissue kit (Qiagen, Valencia, CA, USA) following the manufacturer's instructions.

Plasma DNA samples were collected from 10 ml blood plasma samples in STRECK tubes immediately before tissue biopsy. Within 2 h of collection, plasma was separated from whole blood samples by centrifugation for 10 min at 3000 rpm at room temperature and stored at −80°C until further use. Plasma samples were thawed and centrifuged again for 10 minutes at 13,000 rpm at room temperature before DNA extraction to remove debris. DNA isolation from 2 ml of plasma was performed with the QlAamp circulating nucleic acid kit (Qiagen) according to the manufacturer's instructions and eluted in 140 ^L AVE buffer (Qiagen). The total amount of amplifiable human genomic DNA isolated from plasma samples was quantified using a modified version of the human long interspersed element 1 (LINE-1) real-time PCR assay and reported as genome equivalents (GE).

Targeted sequencing in tissue and plasma DNA samples

The NGS (next generation sequencing) study of the samples was carried out with a personalized next generation sequencing design (NGS - next generation sequencing, specifically SureSelectXT HS) for the 37 genes in Table 1.

The preparation of the sequencing library and the processing of sequencing data was carried out following the methodology described in the respective sections of example 3.

Statistic analysis

Statistical analysis was performed with SPSS (19.0, SPSS Inc., Chicago, IL, USA). For association analyzes between clinicopathological variables and mutations derived from the genes in Table 1 in blood, Spearmen's correlation and Fisher's exact test (for categorical variables) were used. The threshold for statistical significance was set at p < 0.05.

The results using the 37-gene panel for both tumor biopsies and plasma samples are shown in Table 5. A comparison is also presented with the results obtained for these same 29 patients included in the article by Jimenez Rodriguez, B. et al“Detection of TP53 and PIK3CA Mutations in Circulating Tumor DNA Using Next-Generation Sequencing in the Screening Process for Early Breast Cancer DiagnosisJ. Clin Med. 2019 (5), where another NGS technique was used, Plasma SafeSEQ (Sysmex Inostics S.A.).

Table 5: Comparative results of the mutations detected in tumor biopsy and in plasma samples using the gene panel in Table 1; and comparison with plasma mutations located by other state-of-the-art methods.

The "VAF (%)" column expresses the allele frequencies observed using the panel of Table 1, both in solid tumor and in plasma, where allele frequencies of 0.1% are detected in liquid biopsies.

The method of the invention is capable of detecting the concordant variants or mutations between plasma and tumor that are described in reference 5.

Furthermore, thanks to the gene panel in Table 1, it is possible to detect 7 variants in 7 different plasma samples, in the PIK3CAyTP53 genes (which are the ones tested with both panels) and that were not detected in reference 5. These results are noted. in bold, indicating YES in tested/NO in detected in plasma.

By including many more genetic regions in the panel of Table 1, it is possible to detect 3 more mutations in 3 different plasma samples (015MS, 016MS, 044MS).

Example 3: Sensitivity and detection capacity and diagnostic capacity of the custom sequencing panel.

To analyze the sensitivity of the sequencing panel of the invention, a comparison was made between samples from healthy patients (control) and patients who developed breast cancer.

The study was approved by the local ethics committee and was conducted in accordance with the Good Clinical Practice guidelines and the Declaration of Helsinki. Informed consent was obtained from all women who were recruited at the Virgen de la Victoria Hospital in Malaga and at the Hospital Clínico Universitario in Valencia before performing any assessment required according to the protocol.

The following samples have been used:

- 75 plasma samples from patients with breast cancer, of which 74 were analysable. These samples were obtained from patients with BIRADS 4C/5 mammography findings just before the tissue biopsy, before diagnosis and before undergoing treatment for cancer.

- 71 pre-treatment solid biopsies from the aforementioned patients.

- 22 plasma samples from women who were included in the study because they showed suspicious mammography but whose diagnosis was ultimately negative for the presence of breast cancer, and who will serve here as a negative control.

Table 6: Clinicopathological characteristics of the 74 patients from whom plasma samples were extracted.

CDI: invasive ductal carcinoma; ILC: invasive lobular carcinoma; DCIS: ductal carcinoma in situ; PC: papillary carcinoma; CT: tubular carcinoma; MC: mucinous carcinoma; HER2: human epidermal growth factor receptor 2; IHC: Immunohistochemistry; BIRADS: Breast Imaging Data and Reporting System.

Tumor biopsies were removed using thick biopsy needles and subsequently frozen. Immunohistochemical analysis was performed to quantify the expression of human epidermal growth factor receptor 2 (HER2), hormone receptors (HR), and Ki67. Estrogen receptor (ER) and progesterone receptor (PR) were considered positive in tumors that had more than 1% stained cells with nuclei. HER2 staining was graded on a scale of 0 to 3+, according to HercepTest guidelines17. HER2 status was considered positive when it was scored as 3+, while 0 to 1+ was negative and 2+ was an inconclusive result and in those cases silver in situ hybridization (SISH) was performed to confirm positivity. Hormone receptor-positive tumor plus a Ki67 index of <14% was considered a luminal A tumor, while >14% was considered a luminal B tumor. IHC BC subtypes were defined by a combination of immunohistochemical markers as follows: Luminal A (PR positive, HER2 negative and Ki67 < 14%), luminal B (ER positive and/or PR positive, HER2 negative and Ki67 > 14%), HER2 positive (ER positive/negative, PR

positive/negative, HER2 positive) and triple negative (ER, PR and HER2 negative).

Processing of blood samples

10 ml of plasma was obtained from each individual in STRECK tubes. Within 2 h of collection, plasma was isolated from blood by centrifugation for 10 min at 3000 rpm at room temperature and stored at −80°C until extraction of circulating free DNA (cfDNA).

DNA extraction and quantification from plasma and solid biopsies

cfDNA was extracted from plasma samples using the QlAamp circulating nucleic acid kit (Qiagen) according to the manufacturer's instructions. Tumor DNA was isolated from fresh frozen tissue samples using the DNeasy blood and tissue kit (Qiagen) following the manufacturer's instructions. cfDNA and solid tumor DNA were quantified by droplet digital PCR (ddPCR) and the RNAseP assay (Life Technologies) as indicated in Garcia-Murillas I.etal.(9).

Design of the sequencing panel for breast cancer

The custom NGS panel for breast cancer was designed using SureDesign software (Agilent) with the following settings: 5x for tiling, less stringent for masking, XTHSBoosting for boosting, and a value of 30 for extension in repeats.

The panel included the coding regions from the list of genes in Table 1.

Preparation of the sequencing library

The SureSelectXTHS methodology (Agilent) was used to generate sequencing libraries. Libraries were made using a median plasma DNA input of 39.78 ng (max 173.91 ng - min 5.01 ng) from breast cancer patients and 21.78 ng (max 113.52 ng - min. 1.71 ng) from healthy individuals and a median tissue DNA of 199.5 ng (max. 200 ng - min. 6.95 ng) from tumors. Tissue DNA was fragmented using the SureSelect Enzymatic Fragmentation Kit (Agilent) and libraries were prepared using the SureSelectXT Target Enrichment System Kits (Agilent) following the manufacturer's instructions. All PCR steps were carried out in the C1000 Touch Thermocycler (Bio-Rad).

Fragment ranges from the libraries were analyzed with Bioanalyzer high-sensitivity DNA chips (Agilent) and quantified with the KAPA Library Quantification Kit (Roche). For DNA sequencing of tumor tissue, 8 pools containing 8 to 9 library samples per pool were prepared and sequenced. For plasma DNA from breast cancer patients, 8 pools containing 9 to 10 library samples per pool and 3 pools containing 7 to 8 library samples per pool of plasma DNA from controls were also prepared and sequenced. healthy. 19 lanes (1 lane per group) were used to sequence the libraries with the objective of obtaining ultra-deep sequencing of around 20,000X before de-duplication on the DNBseq-G400 (MGI) platform in 100 paired reads following the instructions from the manufacturer for sequencing molecular barcodes, also called unique molecular identifiers (UMI).

Sequencing data processing

A custom method was created for processing SureSelectXTHS (Agilent) sequencing data. Quality control of sequencing data was initially performed using fastQC v0.11.9. Adapter reads were then trimmed and quality filtered using trim-galore v0.6.7. To perform the processing steps involving barcoded data, a subset of fgbio v1.5.1 tools was used. Data were mapped to the GRCh38 reference genome using bwa v0.7.17. fgbio GroupReadsByUmi was then used to group by barcode using the Identity option to account for SureselectXT-HS barcodes being degenerate. Consensus reads were then generated using fgbio CallMolecularConsensusReads. The generated consensus reads were mapped again with bwa v0.7.17. These aligned consensus reads were then filtered using fgbio FilterConsensusReads by requiring a minimum base quality of 30 and keeping the consensus reads supported by at least a minimum number of reads. Then, fgbio ClipBam was used to remove overlapping regions of forward and reverse reads.

Finally, variant identification was carried out with MuTect2 (gatk v4.2.2.0-1), including a non-tumor DNA panel and a germline variant annotation file for the GRCh38 genome, obtained from the resource package gatk, which is used to annotate variants to filter and only consider regions included in the SureSelect panel. Variants were annotated with ANNOVA v20200608 with information about their presence in the public databases COSMIC version 95 and “The Cancer Genome Atlas” (TCGA).

Variant filtering and analysis

For tumor, a more stringent approach was used in order to generate a robust reference for comparison with ctDNA findings. Consensus reads were generated that required a minimum of 3 reads per molecular barcode family. Only variants with VAF > 0.05 that were also present in COSMIC or TCGA (increasing the VAF threshold to VAF > 0.2 for formalin-fixed paraffin-embedded (FFPE) tissues) were accepted as valid calls.

In the case of ctDNA, mutations were identified using two methods: i) Strict; using the same approach as described above, but filtering for a minimum of 1 read per molecular barcode, without applying a VAF threshold. To consider mutations not found in the tumor as detected in plasma, they were required to have a duplex configuration, with at least two fragments mapped to different coordinates and to be present in both COSMIC and TCGA-BC. The same processing approach was applied to the control samples. ii) Exploratory; visualization of the alignments in the IGV Genome Browser (Robinson, Thorvaldsdottir, Wenger, Zehir, & Mesirov, 2017) to identify mutations found in the corresponding tumors but not detected by automated molecular barcoding. For these manually identified mutations, the number of reads that carried the mutation in that sample and the number of reads that did not carry the mutation were counted. They were then compared to the same readout ratios in controls and breast cancer plasma samples without the corresponding mutation using a Fisher test.

Statistical analysis and data visualization

Statistical analysis and graphing of the data were performed using R (https://www.R-project.org/). Fisher's exact test or Chi-square test was applied when appropriate, both to test the association. between clinicopathological variables and plasma sequencing data as well as in sequencing data analyses. The Wilcoxon test was also applied to test for differences in sequencing coverage between cases and controls. The statistical significance threshold was set at p<0.05.

Sensitivity, specificity and positive predictive value (PPV) values were calculated using the caret v6.0.93 package. The oncoplot function of the maftools package (Mayakonda, Lin, Assenov, Plass, & Koeffler, 2018) v2.12.0 was used to plot mutations and clinicopathological data.

Results

First, tumor DNA was sequenced using Agilent SureSelect XT HS technology, following protocol recommendations, as noted above. Tumor sequencing was performed with a median coverage of 15.483X. Post-bioinformatic processing using molecular barcoding to minimize sequencing errors provided a final median coverage of 1.698X. Among the captured regions, only 3 presented less than 100X in more than 10% of the sequenced bases. Furthermore, all genes presented homogeneous coverage in all samples. Custom filtering was then performed using information from public genomic databases to identify somatic mutations.

61 mutations were identified in 40/71 (56.33%) of the tumor samples. Among them, 33 were located in the PIK3CA gene (54.09%), 12 in TP53 (19.67%) and 4 in GATA3 (6.55%) (Figure 2, Table 7), which represent the most frequently mutated genes in our set of tumors

To study the relationship between the mutations observed in tumors and those also detected in plasma, the custom sequencing panel in Table 1 was applied together with high-depth sequencing with a median coverage of 17,704X using plasma DNA from the corresponding patients. In total, 74 plasma samples from the patients were sequenced. After processing of the molecular barcodes, the median coverage was 2.525X. Among the sequenced gene regions, 3 showed low coverage and all genes showed homogeneous coverage. After bioinformatic analyzes using the established mutation identifier, 13/61 (21.31%) mutations were found in plasma that were also present in the corresponding tumors, 7 mutations in the TP53 gene (53.84%) and 3 in PIK3CA ( 27.27%) as the most frequently mutated genes.

Additionally, all mutations observed in the tumors were manually inspected in the raw plasma sequencing data. The aligned data were used to identify supporting reads for the alternative alleles using IGV software. Mutations that were found in at least 2 reads with different genomic coordinates were passed to the next step of analysis.

To consider the variants as valid, a Fisher's exact test was applied using sequencing data from 22 healthy plasma controls and patients who did not carry said mutation in their tumors. In total, 5 mutations from 4 different patients were rescued from plasma sequencing by manual inspection. Among them, 3 mutations were located in the TP53 gene and 2 in GATA3. Interestingly, the 2 structural variants in GATA3 with robust sequencing statistics were recovered by manual inspection.

Taking into account the total number of variants detected, 18 (13 5) in plasma and 61 in tumor, 29.50% of the variants found in the tumor tissue were also located in the plasma samples. Therefore, the sensitivity of this method would be around 30%.

Table 7: Mutations detected in tumor and plasma samples.

The method of the invention has a sensitivity (30%) that far exceeds other methods known to date, especially if we compare it with the two main studies where technologies/methodologies have been developed that use NGS panels (magnification and capture). ) in breast cancer screening, that is, as a diagnostic method. On the one hand, studies have been carried out using PAN-cancer panels as a diagnostic method, where a sensitivity in breast cancer of 33% has been achieved, similar to that of the present invention, but for this it has been necessary to include also other analytes in blood circulation, such as proteins, as additional biomarkers (7).

On the other hand, when only ultrasensitive and amplification patient-specific panels have been used, the sensitivity obtained has only been 14.3%, and prior sequencing of the tumor is also necessary to be able to design them, which is not necessary according to the method. of the present invention. It should be noted that this low sensitivity occurs despite the use of one of the newest technologies in the field, Ampliseq HD (ThermoFisher).

Therefore, we can conclude that it is the specific selection of the genes in Table 1 that allows us to obtain a method for diagnosing breast cancer that overcomes the limitations in terms of sensitivity presented by the methodologies known to date, also based on panels. sequencing.

Example 4: Next Generation Sequencing (NGS) Panel Ability to Detect Breast Cancer Patients After Suspicious Mammograms Using Liquid Biopsy

To investigate the ability of the next generation sequencing (NGS) panel to non-invasively detect breast cancer after suspicious mammograms, a bioinformatics analysis was developed from the data obtained in the samples of example 3. On this occasion the analysis It is aimed at locating all possible mutations that may occur in plasma, unlike the previous example where an analysis is carried out informed by the genetics of the tumor, that is, where only the mutations that appear in the tumor are searched in plasma.

Mutect2 was used to detect mutations in plasma samples using 2 families of molecular barcodes and without filters at variant allele frequencies (VAF). Variants were considered tumor-associated if i) they affected exonic regions, ii) were annotated in the COSMIC, TCGA BC, and TCGA databases that include all cancer types, and if iii) they presented reads aligned at 2 or more coordinates. different genomics manually visualized using the IGV software.

In total, 25/74 (33.78%) patients had tumor mutations detected in plasma (Table 8, Figure 1), 16 of the mutations were not observed in the previous analysis reported by tumor genetics. The genes most frequently affected by the 25 different mutations detected were TP53 (14 mutations, 48.27%), PIK3CA (4 mutations, 13.79%) and GATA3 (3 mutations, 10.34%). Furthermore, ctDNA mutations were observed in 8 plasma samples in whose corresponding tumor biopsies no mutations were detected. Additionally, a mutation was observed in a plasma sample with no tumor tissue available.

Table 8: Results in plasma samples from 74 patients who have developed breast cancer, using the gene panel in Table 1.

In total, mutations have been found in 25 of the 74 samples, which represents 33.78% of the total.

Regarding the controls, the following mutations have been located.

Table 9: Results in plasma samples from 22 healthy subjects, using the gene panel in Table 1.

Mutations have only been found in 3 of the 22 control samples analyzed, that is, in 13.63% of the samples. This is a mutation of the MAP3K1(p.N1125D) gene described in the COSMIC database in a breast cancer tumor sample, a mutation in the ERBB2(p.V842I) gene that has been observed with a substantially higher frequency in cancers. colon and endometrium and another mutation located in the geneSMAD4(p.R361H) also common in colon adenocarcinoma; and on the other hand, that the method of the invention has a capacity to detect allele frequencies of up to 0.03%, which equals, if not exceeds, other previous studies previously cited in the background of the invention.

With these same data, the diagnostic capacity of the sequencing panel in Table 1 has been studied using a statistical analysis, specifically: MedCalc Software Ltd. Diagnostic test evaluation calculator. https://www.medcalc.org/calc/diagnostic_test.php.

The results of said analysis are shown below.

Table 10: Results.

(*) Values dependent on the prevalence of the disease.

Sensitivity: the probability that a test result will be positive when the disease is present (true positive rate).

Specificity: the probability that a test result will be negative when the disease is not present (true negative rate).

Positive likelihood ratio: Ratio between the probability of a positive test result given the presence of the disease and the probability of a positive test result given the absence of the disease.

Negative likelihood ratio: Ratio between the probability of a negative test result given the presence of the disease and the probability of a negative test result given the absence of the disease.

Positive predictive value: probability that the disease is present when the test is positive.

Negative predictive value: probability that the disease is not present when the test is negative.

Precision: overall probability that a patient is correctly classified.

As can be seen from this analysis, the use of the personalized panel together with ultra-deep sequencing and a personalized bioinformatic analysis not informed by the tumor led to a sensitivity of 31.08%, the method of the invention has a very high specificity , of 86.36% and a positive predictive value of 88.46%, so, if a mutation occurs in any of the genes in the panel in Table 1, there is a probability of almost 90% of present breast cancer.

With all this we can conclude that the present invention provides, thanks to the selected sequencing panel, a methodology that allows the diagnosis of breast cancer, even from liquid biopsies, with sensitivity, capacity to detect low allele frequencies and specificity. and positive predictive value, which overcomes the limitations of other similar methods known to date.

The association of clinicopathological variables with the detection of mutations in plasma was also investigated. Specifically, ctDNA positivity in plasma, the median allelic frequency (FA) of the mutations, the number of mutations per sample, and samples with TP53 mutations were studied to determine their association with clinical characteristics (Tables 11 and 12). In general, a higher median FA was associated with a higher tumor grade (p= 0.04639), the presence of more than 1 mutation in plasma with probability of clinical relapse (p= 0.02372), and mutations in TP53 were more frequently observed in tumors negative for the expression of hormone receptors (HR), specifically estrogen receptor (ER) negative (p= 0.0316) and progesterone receptor (PR) negative (P=0.0257). Furthermore, the association of clinical relapse and plasma mutations with high median allele frequency (AF), defined as mutations with >0.05% in AF, was interestingly close to significance (p = 0.059).

It should be noted that 38.35% of the patients included in this study were asymptomatic and were diagnosed by the breast cancer early detection program. Of them, 28.57% had plasma mutations, a percentage similar to the 33.33% of symptomatic women with mutations.

Table 11: Relationship between clinical characteristics and positive ctDNA and median allele frequency (FA) in plasma.

Table 12: Relationship between clinical characteristics and 2 mutations in plasma and Mutated TP53.

In conclusion, we observed a mutation concordance between tumor and plasma of 29.50% with a sensitivity below 0.075% in the frequency of mutant alleles. In the tumor-uninformed approach, we detected mutations in 33.78% of plasma samples by observing 8 patients with mutations only in plasma. We determined a specificity of 86.36% and a positive predictive value of 88.46%. Furthermore, we demonstrated an association between higher median FA and higher tumor grade, the presence of multiple plasma mutations with probability of clinical relapse, and TP53 mutations with estrogen receptor positivity. It is therefore a unique sequencing panel with a technology that has not been previously used in early breast cancer, demonstrating ultra-high sensitivity and paving the way for its use in the detection of breast cancer.

Taking the above into account, the inventive plasma sequencing panel showed improved capabilities to detect ctDNA in breast cancer patients located in the early stages of diagnosis, improving the detection sensitivity of previous research and adding evidence that ctDNA could assist in the diagnosis process of the asymptomatic population. This is also supported by the similar percentages in plasma mutation identification between symptomatic and asymptomatic women of 33.33% and 28.57% respectively. In this sense, we developed a personalized bioinformatics methodology to identify plasma mutations without tumor information, demonstrating a high PPV and suggesting that similar approaches could constitute a screening tool for breast cancer. It has also been demonstrated that by early sequencing the plasma DNA of patients with breast cancer, it is feasible to obtain important information about the disease, as well as predict the clinical evolution in these patients.

References

1. Alfonso, Alba-Bernalet al.;Challenges and achievements of liquid biopsy technologies employed in early breast cancer.eBioMedicine-Volume 62, 103100. https://doi.org/10.1016/¡.ebiom.2020.10310Q.

2. Sun, M.-Y.;et al,.Targeted Next-Generation Sequencing of Circulating Tumor DNA Mutations among Metastatic Breast Cancer Patients.Curr. Oncol.2021,28,2326-2336. https://doi.org/10.3390/curroncol28040214.

3. Priskin, K.;et al.,BC-Monitor: Towards a Routinely Accessible Circulating Tumor DNA-Based Tool for Real-Time Monitoring Breast Cancer Progression and Treatment Effectiveness.Cancers2021,13,3489. https://doi.org/10.3390/cancers13143489.

4. Keup, C., Benyaaet al.Targeted deep sequencing revealed variants in cell-free DNA of hormone receptor-positive metastatic breast cancer patients.Cell. Mol. Life Sci.77, 497-509 (2020). https://doi.org/10.1007/s00018-019-03189-z.

5- Jimenez Rodriguez, B.;et al.,Detection ofTP53andPIK3CAMutations in Circulating Tumor DNA Using Next-Generation Sequencing in the Screening Process for Early Breast Cancer Diagnosis.J. Clin. Med.2019, 8, 1183. https://doi.org/10.3390/icm8081183 unique molecular identifiers 6. Pereira, B.,et al.,The somatic mutation profiles of 2,433 breast cancers refine their genomic and transcriptomic landscapes.Nat Commun7 , 11479 (2016). https://doi.org/10.1038/ncomms11479.

7. Cohen JDet al.,Detection and localization of surgically resectable cancers with a multianalyte blood test.Science.2018 Feb 23;359(6378):926-930. doi: 10.1126/science.aar3247.

8. Page K,et al.,Prevalence of ctDNA in early screen-detected breast cancers using highly sensitive and specific dual molecular barcoded personalized mutation assays.Ann Oncol.

2021 Aug;32(8):1057-1060. doi: 10.1016/j.annonc.2021.04.018. Epub 2021 Apr 29. PMID: 33932505.

9. Garcia-Murillas, I.etal.Mutation tracking in circulating tumor DNA predicts relapse in early breast cancer. Sci Transí Med7, 302ra133 (2015).

Claims (16)

r e iv in d i c a t i o n s 1. Sequencing panel comprising the genes in Table 1 2. Use of the sequencing panel in Table 1 for the diagnosis or prognosis of breast cancer. 3. Use of the sequencing panel according to the previous claim in liquid biopsies. 4. An in vitro method for obtaining useful data to diagnose, predict or forecast the evolution of breast cancer, from a biological sample previously isolated from an individual, which includes identifying possible mutations in the genes in Table 1. 5. An in vitro method for diagnosing or early diagnosing breast cancer that comprises the method of obtaining useful data of claim 4 and which also comprises assigning the individual who has mutations in any of the genes in Table 1 to the group of patients who suffers from breast cancer. 6. An in vitro method for diagnosing breast cancer according to the preceding claim, which further comprises assigning the individual who has a higher median allelic frequency of the mutations of the genes in Table 1 to the group of patients with a higher tumor grade. . 7. An in vitro method to predict or forecast the evolution of breast cancer that comprises the method of obtaining useful data of claim 4 and that also comprises assigning the individual who has more than one mutation in the genes of Table 1, to the group of patients with a probability of clinical relapse. 8. An in vitro method to predict or forecast the evolution of breast cancer that comprises the method of obtaining useful data of claim 4 and that also comprises assigning the individual who has mutations in TP53 to the group of patients with a tumor negative for the expression of receptors. hormonal. 9. The method according to any of claims 4 to 8 wherein the biological sample is plasma. 10. The method according to any of claims 4 to 9, wherein the mutations in the genes of Table 1 are detected by next generation sequencing technologies onext generation sequencing. 11. The method according to the previous claim where the next-generation sequencing technology used is Agilent Sure Select type enrichment probes. 12. A kit that includes the elements necessary for the simultaneous detection of mutations in any of the genes in Table 1. 13. Use of the kit according to claim 12 to diagnose or predict the evolution of breast cancer. 14. A computer program adapted so that any processing means can implement the method according to any of claims 4 to 11. 15. A computer-readable storage medium comprising program instructions capable of causing a computer to carry out the steps of the method according to any of claims 4 to 11. 16. A transmissible signal comprising program instructions capable of causing a computer to carry out the steps of the method according to any of claims 4 to 11.