Methods for mutation detection in liquid biopsy

Abstract

A method for detecting mutations by liquid biopsy of a patient, the method comprising the steps of obtaining DNA fragments by liquid biopsy, preparing a DNA library of the DNA fragments, sequencing the DNA library, identifying DNA variant fragments of the DNA fragments having the mutation, associating the identified DNA variant fragments into a first group and a second group based on the length of the DNA variant fragments, and detecting whether the mutation is neoplastic or healthy based on the presence of the DNA variant fragments in the first group and the second group.

Description

[Technical field]

[0001] The present disclosure relates to genome-related analysis. In particular, the discovery of circulating cell-free DNA (cfDNA) has provided the opportunity to analyze genetic material by liquid biopsy without the risks associated with invasive sampling methods.

[0002] More specifically, but not by way of limitation, the present disclosure refers to methods for detecting mutations from a liquid biopsy of a patient, methods for identifying, tracking, and / or tailoring treatments to match the mutation profile of a patient's tumor, and theranostic reporting of a patient's tumor. [Background technology]

[0003] Small amounts of circulating tumor DNA (ctDNA) originating from tumors, especially malignant tumors, can be found in the cfDNA in the blood of cancer patients. Assays to detect ctDNA from blood to diagnose or better identify disease are known as liquid biopsies. Currently, existing cancer diagnostic methods in liquid biopsies have some limitations, especially in early cancer stages, where ctDNA is present at very low levels compared to cfDNA in blood. This means that the sensitivity of most existing methods is not high enough to detect ctDNA at early stages or the methods are too complicated to be used in clinical practice.

[0004] In addition, it is known that the average length of ctDNA fragments in tumor cells is shorter than that of cfDNA fragments in normal, healthy cells that are not affected by cancer. Various literature is known that provides methods to increase the sensitivity of detecting specific ctDNA from liquid biopsies.

[0005] Document WO 2020 / 94775 provides a computer-implemented method for detecting variant nucleic acids from a cell-free nucleic acid-containing sample, the method comprising the steps of: a) providing data representative of fragment sizes of nucleic acid fragments obtained from the sample and / or data representative of a measure of deviation from copy number neutrality of nucleic acid fragments obtained from the sample, b) causing a computer processor to process the data from step a) according to a classification algorithm trained on a training set comprising a plurality of samples of cell-free nucleic acid containing variant nucleic acids and a plurality of samples not containing variant nucleic acids, the classification algorithm being operative to classify the sample data into one of at least two classes, the at least two classes comprising a first class containing variant nucleic acids and a second class not containing variant nucleic acids based on a plurality of cell-free nucleic acid fragment size features and / or deviation from copy number neutrality features, and c) outputting a classification of the sample from step b), thereby determining whether the sample contains variant nucleic acids or determining the probability that the sample contains variant nucleic acids.

[0006] The document US20200294624 (WO2020 / 186024) provides a sensitive, fast, and inexpensive method for classifying cancers using size-selected cell-free DNA sequences. According to the method described in this document, the fraction of cancer-derived cfDNA fragments in cancer samples of all cancer types can be enriched, including cancers with low tumor-derived cfDNA. Furthermore, this methodology facilitates the classification of cancers with very low tumor fractions (e.g., early stage cancers and low ctDNA cancers), even if the tumor fraction is less than 1%. The method proposes to take a sample from an affected area and a liquid sample from an unaffected area. A histogram of DNA fragment lengths of the two samples is created. Based on the ratio of the two histograms, the fragment size bins are ranked in order of the concentration of cfDNA fragments of the tumor sample relative to the cfDNA of the non-tumor sample. Then, a specific ctDNA known to be cancerous can be detected from the liquid sample based on the highest ranked fragment size bin.

[0007] European Patent Application Publication No. 3612964 (WO 2018 / 195483) discloses a method for determining the presence or copy number of a gene sequence variant associated with a tumor in a test sample by analyzing the size and sequence of cfDNA fragments obtained from the test sample. This document provides a liquid biopsy process and system that effectively combines size and sequence information of cfDNA to achieve high analytical sensitivity and specificity for detecting tumor-associated variants and determining cancer. According to this document, one aspect of this disclosure relates to a method for determining the presence or copy number of a gene sequence variant associated with a tumor in a test sample by analyzing the size and sequence of cfDNA fragments obtained from the test sample.

[0008] Document WO 2020 / 104670 discloses a method for detecting variant cell-free DNA (cfDNA) in a sample obtained from a subject, the analysis of the sample comprising a size selection step to separate different fragment sizes of DNA.

[0009] These publications exploit the differences in size and sequence information between cfDNA and ctDNA fragments to achieve high analytical sensitivity and specificity for detecting tumor-associated variants and determining cancer, thereby increasing the sensitivity for detecting the presence of specific variant DNA (e.g., known to be cancerous) in a liquid sample, even if the specific variant DNA is present in low concentrations.

[0010] Unfortunately, although these documents state the advantages of the technology, currently there is no method that can detect whether a mutation is tumorous or healthy based on DNA taken from a patient's liquid biopsy and has a sufficiently high sensitivity to be applicable in routine clinical practice. In fact, these documents have developed methods to determine the presence of gene sequence mutations associated with tumors. However, there is no method that can identify unknown mutations from liquid biopsies and classify this unknown mutation as tumorous or healthy. In order to improve the efficacy of cancer treatment and reduce the cost of liquid biopsy analysis, it is not only important to be able to detect mutations from liquid biopsies with high sensitivity and detect the presence of tumor-associated gene sequence variants, but it is also necessary to detect whether a mutation is tumorous (originating from tumor cells) or healthy (originating from normal and healthy cells) in order to adapt cancer treatment in the most efficient way. In fact, even if the same mutation is present, the recommended cancer treatment is different in the presence of a tumor and in the healthy case. Summary of the Invention

[0011] Therefore, there is a need for a method to achieve identification of oncogenic or healthy mutations based on a liquid biopsy of a patient.

[0012] According to the present invention there is provided a method for detecting mutations from a liquid biopsy of a patient according to the independent claims.

[0013] While in the prior art, the size of the DNA variant fragments was only used to increase the sensitivity to detect a particular mutation, the present inventors have discovered that the size of the DNA variant fragments can be used to detect whether a mutation is tumorous or not. Indeed, the method according to the present invention is not only able to detect the presence of a mutation by analyzing both groups of DNA variant fragments, but also to detect whether this mutation is tumorous or healthy. There is a real need to improve the personalized cancer treatment of patients, since the detection of tumor or healthy origin of a mutation leads to different results for the patient's therapeutic treatment. Furthermore, the method according to the present invention works unsupervised, so that any mutation, even an unknown mutation, can be detected as tumorous or healthy. The method according to the present invention can also identify and detect mutations from liquid biopsies with very high sensitivity, preventing the need to resort to deep sequencing, thereby reducing the cost of the method and allowing the routine application of the method in clinical settings. Furthermore, many known methods of DNA analysis of liquid biopsies rely on the use of unique molecular identifiers (UMIs) to improve the sensitivity of the method. UMIs are short sequences that are added to DNA fragments during DNA library preparation for next generation sequencing protocols. These UMIs are specific tags to identify DNA fragments and are used to reduce errors that occur during PCR amplification before sequencing. The method according to the invention analyzes the presence of DNA variant fragments in the first and second groups, so that it can naturally complement false positives due to clonal hematopoiesis. Unlike the state of the art, the present invention is based on the presence of a specific mutation in two size groups of the DNA fragment, and does not mix different mutations in the two size groups. This further increases the sensitivity of detecting the presence of a specific mutation, which turns out to be the key to detecting whether a specific mutation is tumorous or healthy.

[0014] The DNA fragments are fragments of DNA collected by liquid biopsy. Preferably, the DNA fragments have a DNA sequence length between 150 bp and 300 bp, but it is clear that the present invention is not limited to these sizes, and DNA sequences outside this range can also be used in the present invention. The DNA fragments consist of DNA variant fragments and / or DNA wild type fragments. The DNA wild type fragments have a DNA sequence identical to the corresponding DNA sequence of the reference genome. Preferably, the DNA sequence of the reference genome is obtained by a public database such as NCBI's hg19 (e.g.: grch37). However, the reference genome can also be a genome derived from a healthy DNA sample of a patient. The DNA variant fragments refer to DNA fragments that have at least one variant from the DNA sequence of the reference genome, or are all such DNA fragments, i.e., the DNA variant fragments have a mutation. The DNA variant fragments do not distinguish between different mutations and refer to all DNA fragments having any mutation. The DNA variant fragments refer to all DNA fragments that have a certain variation of DNA (a certain DNA variant), in other words, all DNA fragments that have the same mutation. The term "mutation" refers to a specific variant in the DNA sequence of a DNA fragment compared to the corresponding DNA sequence of a reference genome. Mutation therefore refers to a certain variation of a certain portion of DNA. At least some of the DNA variant fragments all exhibit the same mutation. A DNA variant fragment can have different mutations in different portions of the DNA variant fragment. The same mutation can refer to the complete sequence of a DNA variant (fragment) (so that the entire DNA variant fragment is a realization of the same DNA variant sequence) or it can refer to only a sub-portion of DNA whose DNA sequence has the same mutation (so that the DNA variant fragments can be identical in that sub-portion of DNA and different in other portions of DNA). Thus, in the latter embodiment, a DNA variant fragment with two mutations in two different sub-portions can be considered / identified as two different specific mutations. That is, the combination of DNA variant fragments of all mutations collected by liquid biopsy forms a DNA variant fragment.Preferably, the mutation belongs to the group of variants comprising single nucleotide polymorphisms, indels, deletions, translocations, copy number variants, or combinations thereof.

[0015] According to the present invention there is provided a method for detecting mutations by liquid biopsy of a patient, the method comprising: Obtaining DNA fragments by liquid biopsy; Preparing a DNA library of DNA fragments obtained by liquid biopsy while preserving DNA fragment lengths; Sequencing the DNA library and collecting sequencing data of DNA fragments of the DNA library, wherein the sequencing of the DNA library is performed by targeted sequencing using a targeted panel, the targeted panel defining a plurality of mutations known to be generally neoplastic or a plurality of mutations identified from a patient sample taken prior to liquid biopsy; Identifying DNA variant fragments of the DNA fragments having the same mutation based on the sequencing data; associating the identified DNA variant fragments with a first group if the DNA variant fragments are shorter than a first intermediate length threshold, the first intermediate length threshold being between 150 bp and 166 bp; associating the identified DNA variant fragments with a second group if the DNA variant fragments are longer than a second intermediate length threshold, the second intermediate length threshold being between 150 bp and 166 bp, and preferably the first and second intermediate length thresholds being equal; and detecting whether the mutation is neoplastic or healthy based on the presence of the DNA variant fragment in the first group and the second group, where the mutation is detected as neoplastic if the DNA variant fragment is more present in the first group than in the second group.

[0016] According to the present invention, there is provided a method for identifying, tracking and / or tailoring a treatment adapted to an individualized mutation profile of a patient's tumor, the method comprising: Obtaining DNA from a patient's tumor sample; sequencing DNA from the tumor sample; Identifying mutations from sequencing; designing a panel of specific DNA probes targeting the identified mutations; Implementing the method of any one of claims 1 to 5, comprising the steps of collecting DNA fragments of a patient liquid sample in a designed panel; and characterizing an individualized mutation profile of the tumor based on the detection of neoplastic and / or healthy mutations.

[0017] The present invention also relates to a theranostic report of a patient's tumor generated based on the method of identifying, tracking and / or tailoring a treatment to match the personalized mutation profile of the patient's tumor according to the present invention.

[0018] According to the present invention, it is actually possible to detect whether DNA variants from a liquid biopsy of a patient are tumor or healthy. In order to confidently establish an individualized mutation profile of a patient's tumor by liquid biopsy, not only is it necessary to detect mutations with higher sensitivity, but it is also important to be able to detect whether the mutations originate from the patient's tumor cells or from the patient's normal healthy cells. According to the present invention, it is also possible to generate a theranostic report of a patient that incorporates the characteristics of the patient's tumor. Such a report is very useful, for example, in a clinical setting, to enable a doctor to make the best medical decisions according to the available information about the patient's tumor.

[0019] The dependent claims refer to further advantageous embodiments.

[0020] Preferably, the detection of whether a mutation is neoplastic or normal is based on the presence of DNA variant fragments in the first and second groups and the mutation frequency of the mutation. It is indeed useful to cluster the mutations as normal or neoplastic based on the length of the DNA variant fragment carrying the mutation and the DNA mutation frequency.

[0021] Advantageously, a mutation is detected as neoplastic if the DNA variant fragment is more prevalent in the first group than in the second group and the mutation has a mutation frequency below a frequency threshold, the frequency threshold being between 2% and 50%.

[0022] Preferably, if the DNA variant fragment is longer than a second intermediate length threshold and shorter than an upper length threshold, the DNA variant fragment is associated with the second group. Indeed, such initial and upper length thresholds allow proper clustering of the first and second groups, which is necessary to distinguish between healthy and neoplastic DNA variants.

[0023] Preferably, the target sequencing is hybridization target sequencing using double-stranded DNA probe.Indeed, the present inventors have found that the hybridization step using double-stranded DNA probe is more efficient for capturing the target DNA for target sequencing, and allows the target DNA to be uniformly captured, and therefore allows the target DNA to be uniformly sequenced.

[0024] Advantageously, the DNA library is a single-stranded and double-stranded DNA library.Indeed, the presence of both single-stranded and double-stranded DNA libraries enriches tumor DNA fragments compared to normal and healthy DNA fragments, improving the sensitivity of tumor DNA variant detection.

[0025] According to the present invention, the term tumoral DNA fragments refers to DNA fragments that originate from tumor cells and / or from the patient's tumor.

[0026] Preferably, the targeted sequencing is performed using a personalized panel that defines the mutations identified in the patient's tumor. Preferably, the targeted sequencing is targeted sequencing of a DNA sequence having a total length of at least 0.5 Mb, preferably at least 0.8 Mb, preferably at least 1 Mb of the patient's tumor DNA, which may be a series of continuous and / or discontinuous DNA sequences having a cumulative total length of at least 0.5 Mb, preferably at least 0.8 Mb, preferably at least 1 Mb of the patient's tumor DNA. The total length of the sequenced DNA sequence refers to the cumulative number of base pairs read (cumulated for the different DNA "pieces" read). The present invention is particularly advantageous with personalized panels, since firstly, the mutations appearing in the patient's tumor are identified and defined in a targeted / personalized panel. Moreover, the method of detecting mutations from a patient's liquid biopsy using personalized panels not only detects the presence of defined and personalized mutations, but also distinguishes between tumor mutations and healthy (or non-tumor) mutations. The combination of the method of the present invention for detecting mutations from liquid biopsies of patients with a personalized panel is a very powerful tool for cancer surveillance. Alternatively or additionally, targeted sequencing can be performed with a targeted panel that defines known DNA variants that are frequently associated with the patient's tumor type and / or known DNA variants that are associated with the response or resistance of the patient's tumor type to treatment.

[0027] Advantageously, the method comprises a step of carrying out a sequencing of the patient's tumor sample to identify the mutations of the individualized or targeted panel.Preferably, the identification of the mutations from the first group and / or the mutations from the second group is carried out by a comparison of the mutations in the first group and the second group, and (i) if the mutation is present only in the second group, the mutation is detected as a healthy mutation, and / or (ii) if the DNA mutation frequency of the DNA mutation is superior to a frequency threshold, the mutation is detected as a healthy mutation, and / or (iii) if the mutation is present more in the second group than in the first group, the mutation is detected as a healthy mutation.In fact, besides detecting tumor mutations, it is also advantageous to detect healthy mutations in order to better define the patient's tumor.

[0028] Advantageously, liquid biopsy is performed on patients in at least one of the following stages: (i) patients with tumors before treatment, (ii) patients with tumors during treatment, (iii) patients with tumors after the end of treatment, (iv) patients in tumor remission stage, (v) patients in tumor recurrence stage. Indeed, to better track the evolution of a patient's tumor and / or to better analyze the effectiveness of tumor treatment, it is useful to perform the method according to the invention on DNA samples obtained from liquid biopsy, where liquid biopsies are performed at different stages of the patient's tumor evolution and / or at different stages of the patient's treatment. In particular, this method is particularly advantageous at the tumor recurrence stage, since it allows the detection of tumor recurrence with high sensitivity and low false positive rate.

[0029] Preferably, for each mutation in the plurality of mutations defined in the targeted panel: identifying DNA variant fragments of each of the DNA fragments having a mutation based on the sequencing data; associating the identified DNA variant fragments of each mutation with a first group if the DNA variant fragments of each mutation are shorter than a first intermediate length threshold, the first intermediate length threshold being between 150 bp and 166 bp; associating the identified DNA variant fragments of each mutation with a second group if the DNA variant fragments of the respective mutation are longer than a second intermediate length threshold, the second intermediate length threshold being between 150 bp and 166 bp, preferably the first intermediate length threshold and the second intermediate length threshold being equal; detecting whether each mutation is neoplastic or healthy based on the presence of DNA variant fragments of each mutation in the first group and the second group, where each mutation is detected as neoplastic if the DNA variant fragments are more prevalent in the first group than in the second group.

[0030] Advantageously, the liquid sample is derived from a patient in a stage of tumor remission. Indeed, it is advantageous to carry out the method according to the invention on a patient in a stage of tumor remission in order to non-invasively analyze the possibility of re-emergence of the patient's tumor.

[0031] Preferably, identifying, tracking and / or adapting a treatment is adapted to the characterization of the individualized mutation profile of a tumor according to the present invention. Indeed, identifying, tracking and adapting a treatment is beneficial to the characterization of the individualized mutation profile of a patient's tumor.

[0032] Advantageously, a treatment is applied to a patient that is adapted to the characterization of the individualized mutation profile of the patient's tumour.

[0033] Other characteristics, details and advantages of the invention will become apparent from the following non-limiting description, made with reference to the accompanying drawings, in which: [Brief description of the drawings]

[0034] [Figure 1] 1 is a flow chart illustrating the procedure for designing a target panel of specific DNA probes. [Diagram 2]The present invention provides a method for detecting mutations from a patient's liquid biopsy and a method for identifying, tracking and / or tailoring a treatment that matches the personalized mutation profile of a patient's tumor according to the present invention. [Diagram 3] 1 shows how a DNA variant fragment satisfies the first or second group of criteria. [Figure 4] FIG. 1 is a schematic diagram showing clustering of DNA variant fragments carrying a first mutation into a first group and a second group. [Diagram 5] FIG. 1 is a schematic diagram showing the clustering of DNA variant fragments carrying a second DNA mutation into a first group and a second group. [Figure 6] 1 is a flow chart illustrating the steps for identifying whether a mutation is neoplastic or healthy. [Figure 7] FIG. 1 is a schematic diagram showing that liquid biopsy can be performed at different stages of a tumor's evolution and / or a patient's clinical treatment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0035] In the drawings, the same reference numbers are assigned to the same or analogous elements.

[0036] Other characteristics and advantages of the invention will become apparent from the following non-limiting description and by reference to the drawings and examples.

[0037] FIG. 1 illustrates the various steps for designing a targeted panel of specific DNA probes from a collection of DNA from patient tumor samples.

[0038] The first step S1 consists of obtaining DNA from a patient tumor sample. In some embodiments, the patient tumor sample is derived from a solid biopsy. In some embodiments, the patient tumor sample is derived from a liquid biopsy. The DNA from the patient tumor sample can be DNA fragments from the tumor and / or genomic DNA, chromosomal DNA from the tumor.

[0039] In step S2, the DNA from the tumor sample is sequenced. Preferably, the DNA is sequenced at least 100000 base pairs (bp) or 100 kilobp (kb), preferably at least 500 kb, preferably at least 700 kb, preferably at least 800 kb, preferably at least 900 kb, preferably at least 1000 kb or 1 megabase (Mb) of DNA from the patient's tumor sample. Preferably, the DNA is sequenced at less than 10 Mb, preferably less than 5 Mb, preferably less than 3 Mb of DNA from the patient's tumor sample. Advantageously, the sequencing is performed by a NextSeq500 / 550 sequencer or any sequencer from Illumina or MGI. Preferably, the sequencing is paired end sequencing, but may also be single end sequencing.

[0040] Next, in step S3, mutations are identified from the tumor sample based on comparison of the sequencing DNA of the tumor sample with the corresponding DNA sequence of the reference genome.

[0041] The next step S4 consists of designing a targeted panel of specific DNA probes targeting the identified mutations.

[0042] 2 discloses an embodiment of a method for detecting mutations from a liquid biopsy of a patient and identifying a treatment that matches the mutation profile of the patient's tumor according to the present invention. Steps S11 to S13 describe rather laboratory steps of the method, whereas steps 22 to 27 describe rather in silico steps. However, it is also possible to perform some steps S11 to S13 in silico and / or one or more steps 22 to 27 non-in silico.

[0043] Step S11 consists of obtaining DNA fragments from the liquid biopsy.

[0044] The next step S12 consists in preparing a DNA library of DNA fragments, preserving the length of the DNA fragments. Advantageously, the DNA library is prepared on the basis of DNA fragments having single-stranded and / or double-stranded DNA fragments. Preferably, the DNA library is prepared on the basis of DNA fragments having single-stranded and double-stranded DNA fragments.

[0045] Next, in S13, the DNA library is sequenced based on the target panel. The target panel defines multiple mutations. The target panel can sequence DNA fragments in the liquid biopsy or library with defined mutations. Thus, the sequencing data resulting from sequencing the DNA library based on the target panel will read or sequence all DNA fragments with the mutations defined in the target panel (while DNA fragments that do not show the defined mutations will not be read or sequenced). The target panel can also define wild-type DNA, so that the target panel also sequences DNA fragments that do not contain mutations. The sequencing data thus obtained contains DNA sequences of DNA fragments of the liquid sample with defined mutations (which may belong to wild-type DNA). Preferably, the target panel is a panel targeted to the patient's cancer, i.e. a patient-specific panel designed based on the patient's sample. Such a patient-specific target panel can be created as described in FIG. 1. Advantageously, targeted sequencing is performed to enrich DNA with (i) known mutations frequently associated with the patient's tumor type, and / or (ii) known mutations associated with the response or resistance to treatment of the patient's tumor type, and / or (iii) individualized mutations identified in the patient's tumor. However, targeted panels do not have to be patient-specific. Targeted panels can, for example, include multiple mutations known to cause a specific cancer type, in other words, a cancer-specific panel. Specific cancer types are, for example, lung cancer, breast cancer, skin cancer, etc. Targeted sequencing is hybridization targeted sequencing.

[0046] The sequencing step provides (digital) DNA sequencing data (short sequencing data). Preferably, the sequencing data provides for each (sequenced) DNA fragment sequence information of this DNA fragment. This sequence information preferably comprises the nucleotide sequence of the (sequenced) DNA fragment. Figures 4 and 5 show sequencing data 30 and sequenced DNA fragments. To simplify the figures, only a wild type DNA fragment 1, first DNA variant fragments 1.1, 1.2 with a first specific mutation (shown as a grey box) and second DNA variant fragments 1.1', 1.2' with a second specific mutation (shown as a white circle) are shown. The example is not limiting of the invention and the sequencing data may comprise DNA variant fragments with two or more different mutations.

[0047] In step S22, among the DNA fragments of the DNA sequencing data, DNA variant fragments having a specific mutation are identified. That is, in step S22, among all DNA fragments in the DNA sequencing data, DNA variant fragments (having a specific mutation) are selected. This selection is preferably performed in silico, in other words, based on the digital sequencing data. Thus, the DNA variant fragments referred to in the subsequent steps S23 and S24 always refer to DNA variant fragments of the same specific mutation. In the example of FIG. 4, the DNA fragments of the first DNA variant fragment 1.1, 1.2 are identified / selected. In the example of FIG. 5, the DNA fragments of the second DNA variant fragment 1.1', 1.2' are identified / selected. As a result of step S22, a set of DNA variant fragments is obtained, i.e. a set of DNA fragments belonging to the same specific mutation.

[0048] In step S23, the identified DNA variant fragments are associated with a first group and a second group based on the length of the DNA variant fragments. The DNA variant fragments of the first group are shorter than the DNA variant fragments of the second group, and at least the majority of the DNA variant fragments of the first group are shorter than the DNA variant fragments of the second group. Preferably, all DNA variant fragments of the first group are shorter than all DNA variant fragments of the second group. DNA variant fragments of the first / second group are taken to mean DNA variant fragments associated with the first / second group. The result of step S23 is a first set / group of shorter DNA variant fragments 10 and a second set / group of longer DNA variant fragments 20. The first and second groups 10, 20 only have DNA fragments of a specific mutation, i.e. only have one (and / or the same) mutation. Preferably, DNA variant fragments 1.1 shorter than a first intermediate length threshold are associated with the first group 10, and DNA variant fragments 1.2 longer than a second intermediate length threshold are associated with the second group 20. Most preferably, the first and second intermediate length thresholds are the same, in other words the (common) intermediate length threshold is as shown in the example of FIG. 3. However, the first and second intermediate length thresholds may be different. In this case, the first and second intermediate length thresholds are preferably differentiated by less than 5 base pairs (bp), preferably less than 3 bp. The intermediate length threshold, the first intermediate length threshold and / or the second intermediate length threshold are preferably less than 167 bp, preferably less than 166 bp, preferably less than 165 bp, preferably less than 164 bp, preferably less than 163 bp, preferably less than 162 bp, preferably less than 161 bp and / or preferably greater than 150 bp, preferably greater than 155 bp, preferably greater than 157 bp, preferably greater than 158 bp, preferably greater than 159 bp. In a preferred embodiment, the intermediate length threshold is 160 bp in length. There may be additional criteria for associating DNA fragments to the first group 10 or the second group 20.Preferably, DNA variant fragments 1.2 that are longer than the second intermediate length threshold (or intermediate length threshold) and shorter than the upper length threshold are associated with a second group 20. The upper length threshold is greater than the (second) intermediate length threshold. Preferably, the upper length threshold is less than 250 bp, preferably less than 200 bp, preferably less than 190 bp, preferably less than 185 bp. In a preferred embodiment, the upper length threshold is 180 bp.

[0049] FIG. 3 is a diagram illustrating how the (identified) DNA variant fragments are associated with the first and second groups based on their length. This is achieved by performing the method of FIG. 3 for each identified DNA variant fragment. Thus, step S23 of FIG. 2 can be achieved by performing the method shown in FIG. 3 for each (identified) DNA variant fragment. If the DNA variant fragment is smaller than the intermediate length threshold, the DNA variant fragment satisfies the condition of the first group 10 and / or is associated with the first group 10 in step S32. If the DNA variant fragment is not smaller than the intermediate length threshold or is larger than the intermediate length threshold, the process proceeds to step S33, where it is further checked whether the DNA variant fragment is smaller than the upper length threshold. In this case (i.e., if the DNA variant fragment has a length between the intermediate length threshold and the upper length threshold), the DNA variant fragment is associated with the second group 20 in step S34. As in this procedure, the order of steps S32 to S34 and the checks can be rearranged as desired as long as the same DNA variant fragments are associated with the first and second groups 10 and 20. That is, as shown in FIG. 4, the short DNA variant fragment 1.1 is associated with the first group 10, and the long DNA variant fragment 1.2 is associated with the second group 20.

[0050] The order of steps S22 and S23 is not important for the present invention: either the DNA variant fragments belonging to one mutation are first identified and then associated into two groups 10, 20 based on their length, or the entire DNA (variant) fragments are first associated into two groups 10, 20 based on their length, and then the DNA variant fragments belonging to one mutation are identified in the first and second groups 10, 20. Note that when it is stated that the identified DNA variant fragments are associated into the first group 10 and / or the second group 20, it does not mean that the identification step S22 is necessarily performed before the association step S23.

[0051] In step S24, a specific mutation (of the DNA variant fragments) is detected as neoplastic or healthy. This detection is preferably based on the presence of the DNA variant fragments of the specific mutation in the first and second groups 10, 20. Preferably, a specific mutation is detected as neoplastic if the DNA variant fragments (of the specific mutation) are more present in the first group 10 than in the second group 20, preferably statistically significantly more present than in the second group 20. This detection is preferably based on the frequency of the mutation. This detection is preferably based on the presence of the DNA variant fragments of the specific mutation in the first and second groups 10, 20, based on the mutation frequency. Preferably, a specific mutation is detected as neoplastic if it cumulatively fulfills two conditions: the mutation frequency of the mutation is below a frequency threshold, and the DNA variant fragments (of the specific mutation) are more present in the first group 10 than in the second group 20, preferably statistically significantly more present than in the second group 20. The statistical significance can be tested, for example, by a chi-square test.

[0052] Mutation frequency is often also called allele frequency or gene frequency. Mutation frequency is determined based on the frequency of a particular mutation in liquid biopsy and / or sequencing data obtained from liquid biopsy. The mutation frequency of a mutation can be determined based on the ratio between the presence of the mutation and the presence of wild type DNA. Preferably, this is determined by the ratio #V / #R between the number of DNA variant fragments in the sequencing data #V and the number of DNA fragments in the sequencing data #R. The number of DNA fragments in the sequencing data #R corresponds to the number of reads, i.e. the total number of DNA fragments read in the sequencing process. However, other definitions are possible. For example, the mutation frequency can also be defined by the corresponding ratios defined above within a certain size range of DNA fragments, or other definitions that represent the mutation frequency. The frequency threshold is preferably greater than 5%, preferably greater than 10%, preferably greater than 15%, preferably greater than 18%. The frequency threshold is preferably less than 40%, preferably less than 30%, preferably less than 25%, preferably less than 22%. In a preferred embodiment, the frequency threshold is 20%.

[0053] Preferably, a mutation is detected as healthy or non-neoplastic if the above defined conditions for a neoplastic mutation are not met and / or if the mutation frequency of the mutation is greater than a frequency threshold and / or if the DNA variant fragment is more prevalent in the second group 20 than in the first group 10. However, the method can also determine only whether a mutation is neoplastic. In this case, the detection when the mutation is not detected as neoplastic is considered as a patent where the mutation is detected as healthy.

[0054] FIG. 6 is an exemplary flow chart illustrating how a mutation is detected as neoplastic or healthy. FIG. 6 is an example method for implementing step S24 of FIG. 2. In step S42, a check is made as to whether the mutation frequency of the mutation is below a frequency threshold. In S44, if the mutation frequency is not below the frequency threshold, the mutation is detected as healthy. If the mutation frequency is below the frequency threshold, a check is performed in S43 as to whether the DNA variant fragments are more prevalent in the first group 10 than in the second group 20. If the DNA variant fragments with the mutation are more prevalent in the first group 10 than in the second group 20, the mutation is detected as neoplastic in step S45, whereas if the DNA variant fragments with the mutation are not more prevalent in the first group 10 than in the second group 20 and / or are more prevalent in the second group 20 than in the first group 10, the mutation is detected as healthy in step S46.

[0055] The method according to the invention may be applied for only one mutation. In this case, the method ends after step S24 or proceeds to step S27. Preferably, the method according to the invention is applied for a plurality of mutations such that for each of the plurality of mutations, whether the DNA variant is healthy or neoplastic is detected. Each of the plurality of mutations is defined in a target panel. In this case, by repeating steps S22 to S24 for each mutation of the plurality of mutations, it is possible to determine for each mutation of the plurality of mutations whether the mutation is neoplastic or healthy in S24. If all mutations of the plurality of mutations have been processed, the method proceeds to step S26. Figure 4 shows the step of identifying a first DNA variant fragment S22 of a first mutation in the DNA fragment 30 and associating the first DNA variant fragment with the first group 10 and the second group S23, while Figure 5 shows the step of identifying a second DNA variant fragment S22 of a second mutation in the DNA fragment 30 and associating the second DNA variant fragment with the first group 10 and the second group S23. The first mutation and the second mutation are different.

[0056] Step S26 is optional and consists of characterizing the tumor based on the result of step S24 of the detection of the mutation or mutations. Characterizing the tumor based on the detection of neoplastic and / or healthy mutations means, for example, identifying tumor subtypes that are particularly known to be sensitive and / or resistant to a particular tumor treatment. That is to say, characterizing the tumor allows the identification of a particular tumor treatment that is well adapted to the patient's tumor.

[0057] Step S27 is also optional and consists of identifying, tracking and / or adapting a treatment that matches the individualized mutation profile of the patient's tumor based on the tumor characterization in step S26 or based on the result(s) of the detection in step S24. Preferably, the theranostic report of the patient's tumor may comprise the tumor's medical information and / or sequencing data and a list of identified mutations that are neoplastic and / or healthy. The theranostic report of the tumor may also include (i) data on the likelihood of response to the treatment based on the tumor characterization, (ii) a list of treatments that may be associated with clinical benefit and / or have characteristics that are not associated with clinical benefit, (iii) a list of clinical assays associated with the tumor characterization, and / or a list of scientific publications related to the tumor characterization.

[0058] Figure 7 shows different time points at which liquid biopsies can be performed on patients to monitor the evolution of the tumor. Preferably, liquid biopsies are performed on patients in at least one of the following stages: (i) a patient 2.1 with a tumor before treatment, (ii) a patient 2.2 with a tumor under treatment, (iii) a patient 2.3 with a tumor after the end of treatment, (iv) a patient 2.4 in tumor remission stage, (v) a patient 2.4 in tumor recurrence stage. Preferably, liquid biopsies are taken from patients in tumor remission stage. It is to be understood that the invention is not limited to the described embodiments, but variations are possible without departing from the scope of the claims.

Claims

1. 1. A method for detecting mutations by liquid biopsy of a patient, the method comprising: collecting DNA fragments by the liquid biopsy; Preparing a DNA library of the DNA fragments obtained by the liquid biopsy while preserving the DNA fragment lengths; sequencing the DNA library and collecting sequencing data for the DNA fragments in the DNA library, wherein the sequencing of the DNA library is performed by targeted sequencing using a targeted panel, the targeted panel defining a plurality of mutations known to be commonly neoplastic or a plurality of mutations identified in a patient sample taken prior to the liquid biopsy; Identifying DNA variant fragments of the DNA fragments that have the same mutation based on the sequencing data; associating the identified DNA variant fragments with a first group if the DNA variant fragments are shorter than a first intermediate length threshold, the first intermediate length threshold being between 150 bp and 166 bp; associating the identified DNA variant fragments with a second group if the DNA variant fragments are longer than a second intermediate length threshold, the second intermediate length threshold being between 150 bp and 166 bp, and preferably the first and second intermediate length thresholds being equal; and detecting whether the mutation is neoplastic or healthy based on the presence of the DNA variant fragment in the first group and the second group, wherein the mutation is detected as neoplastic if the DNA variant fragment is more prevalent in the first group than in the second group. method.

2. 10. The method of claim 1, Detecting whether the mutation is neoplastic or normal based on the presence of the DNA variant fragment in the first group and the second group and the mutation frequency of the mutation. method.

3. In the method according to claim 1, detecting the mutation as neoplastic if the DNA variant fragment is more prevalent in the first group than in the second group and the mutation has a mutation frequency below a frequency threshold, the frequency threshold being between 2% and 50%. method.

4. In the method according to claim 1, the DNA variant fragment is associated with the second group if the DNA variant fragment is longer than the second intermediate length threshold and shorter than the upper length threshold; method.

5. In the method according to claim 1, The DNA libraries are single-stranded DNA libraries and double-stranded DNA libraries. method.

6. In the method according to claim 1, Identifying mutations from said first group and / or said second group is done by comparing said mutations in said first group and said second group, (i) the mutation is detected as a healthy mutation if the mutation is present only in the second group; and / or (ii) if the DNA mutation frequency of said mutation in DNA is greater than a frequency threshold, said mutation is detected as a healthy mutation; and / or (iii) if the mutation is more prevalent in the second group than in the first group, the mutation is detected as a healthy mutation; method.

7. The method according to claim 1, The liquid biopsy is performed on a patient in at least one of the following stages: (i) a patient with a tumor before treatment; (ii) a patient with a tumor during treatment; (iii) a patient with a tumor after completion of treatment; (iv) a patient in tumor remission; or (v) a patient in tumor recurrence. method.

8. The method according to claim 1, For each mutation in the plurality of mutations defined in the targeted panel: identifying DNA variant fragments of the DNA fragments carrying the respective mutations based on the sequencing data; associating the identified DNA variant fragments of each mutation with a first group if the DNA variant fragments of each mutation are shorter than a first intermediate length threshold, wherein the first intermediate length threshold is between 150 bp and 166 bp; associating the identified DNA variant fragments of each mutation with a second group if the DNA variant fragments of each mutation are longer than a second intermediate length threshold, wherein the second intermediate length threshold is between 150 bp and 166 bp, and preferably the first and second intermediate length thresholds are equal; detecting whether each mutation is neoplastic or healthy based on the presence of the DNA variant fragment of each mutation in the first group and the second group, wherein the each mutation is detected as neoplastic if the DNA variant fragment is more prevalent in the first group than in the second group. method.

9. 1. A method for identifying, tracking, and / or tailoring a treatment to match an individualized mutation profile of a patient's tumor, said method comprising: obtaining DNA from a tumor sample from said patient; sequencing the DNA from the tumor sample; identifying mutations from said sequencing; designing a panel of specific DNA probes targeting the identified mutations; 9. A method according to claim 1, wherein the DNA fragments of a liquid sample from the patient are collected in the designed panel; characterizing the individualized mutation profile of the tumor based on the detection of neoplastic and / or healthy mutations. method.

10. The method according to claim 9, The liquid sample originates from the patient in tumor remission. method.

11. 10. The method of claim 9, and identifying, tracking, and / or adapting the treatment is adapted to the characterization of the individualized mutation profile of the tumor. method.

12. 10. The method of claim 9, a treatment is administered to the patient that is compatible with the characterization of the individualized mutation profile of the patient's tumor; method.

13. 10. The method of claim 9, The liquid sample originates from the patient in tumor remission. method.

14. 10. A theranostic report generated according to the method of claim 9.

Images