A microsatellite repeat expansion burden in vitro detection method and kit for screening wrn inhibitor responders

Abstract

The application discloses a microsatellite repeat expansion load in vitro detection method and kit for screening WRN inhibitor responders, and belongs to the field of tumor precision medicine and companion diagnosis. In view of the problem that the prior art selects patients according to the binary microsatellite instability state, and regards the microsatellite highly unstable group as a homogeneous response group, and cannot enrich the real WRN dependent subgroup, the scheme is as follows: providing nucleic acid in an ex vivo tumor sample (S1); in vitro determination of a continuous score reflecting the expansion degree of the repeat units of a plurality of microsatellite sites, that is, a microsatellite repeat expansion load quantitative value (S2); comparing the quantitative value with a preset threshold value (S3); when the quantitative value is greater than or equal to the threshold value, outputting a molecular stratification result of a WRN inhibitor response candidate (S4); and the quantitative value can be further stratified in the microsatellite highly unstable or mismatch repair defect positive sample. The application replaces the binary state enrichment with a continuous load quantitative value, and significantly improves the enrichment degree of the responders.

Description

Technical Field

[0001] This invention relates to the fields of precision oncology, companion diagnostics, and patient selection, specifically to an in vitro detection method, an in vitro detection kit, and a data processing method for non-diagnostic and non-therapeutic purposes for quantifying the microsatellite repeat amplification load in tumor samples and identifying WRN inhibitor response candidates based on their threshold values. Background Technology

[0002] Precision medicine in oncology relies on screening a subgroup of patients who are truly likely to benefit from a specific treatment before medication is administered; this process is known as patient selection or companion diagnostics. For targeted drugs that exert their effects through synthetic lethal mechanisms, the accuracy of patient selection directly determines the responder enrichment of the clinical trial enrollment population, thus affecting the success or failure of drug development and the ultimate clinical benefit.

[0003] WRN helicase (Werner syndrome RecQ helicase, hereinafter referred to as WRN) is a DNA helicase belonging to the RecQ family. It has been found to produce synthetic lethality in microsatellite instability (MSI) tumors caused by mismatch repair deficiency (dMMR). In these tumors, repeat units at microsatellite sites, especially dinucleotide repeat sites, continuously amplify. The amplified repeat sequences tend to form non-B-type DNA secondary structures, causing DNA replication fork arrest. Cells rely on the unwinding activity of WRN to unwind these structures to maintain genomic integrity, and are therefore highly sensitive to WRN inhibition or deletion. Based on this mechanism, multiple research institutions are developing small molecule inhibitors (hereinafter collectively referred to as WRN inhibitors) acting on WRN helicase in parallel for the treatment of microsatellite instability solid tumors.

[0004] Existing WRN inhibitor patient selection protocols use existing commercially available detection technologies to determine microsatellite instability, which are mainly implemented in the following three ways.

[0005] The first type is immunohistochemistry (IHC). This method involves immunostaining tumor tissue sections with mismatch repair-related proteins MLH1, MSH2, MSH6, and PMS2. The presence or absence of these proteins determines whether the sample is mismatch repair deficient (dMMR) or mismatch repair normal (pMMR). The output is a qualitative conclusion regarding the presence or absence of protein expression, representing a binary state determination.

[0006] The second type is polymerase chain reaction (PCR) capillary electrophoresis. This method uses a combination of Bethesda or Promega markers to amplify a set of pre-defined microsatellite loci, and then analyzes their length distribution via capillary electrophoresis. When a predetermined number of microsatellite loci show length shift, the sample is classified as microsatellite instability-high (MSI-H); otherwise, it is classified as microsatellite stable (MSS). The output is a binary state of MSI-H or MSS.

[0007] The third category is methods combining next-generation sequencing (NGS) with microsatellite instability scoring tools. This method performs high-throughput sequencing of sample nucleic acids and uses MSIsensor-like tools to statistically analyze the proportion of sites with length variations or related signals at preset microsatellite loci, obtaining a microsatellite instability score. This score is then binary classified into MSI-H or MSS using a fixed threshold. For example, prior art with publication number WO2019236448A1 (Dana-Farber Cancer Institute Inc. and Broad Institute Inc., priority date June 4, 2018, publication date December 12, 2019) discloses a treatment scheme for microsatellite instability cancer. After obtaining the sample, it identifies whether microsatellite instability or mismatch repair defects exist. When a set number of microsatellite markers show amplification or shrinkage of repeat units, the sample is determined to be microsatellite unstable. If microsatellite instability or mismatch repair defects are determined, a WRN inhibitor is selected for treatment. While the prior art discusses in its specification that repetitive amplification is the WRN-dependent mechanism, its patient selection ultimately hinges on a binary state determination of microsatellite instability or mismatch repair deficiency. Similarly, prior art publication CN121545576A discloses a NGS-based method for screening microsatellite instability detection biomarkers, establishing a detection process including a scoring system and dynamic thresholds; however, the endpoint of this scoring and thresholding is still determining whether the sample is in a binary state of high microsatellite instability. Prior art publication CN122067600A utilizes high-throughput targeted sequencing data to obtain length and structural variation characteristics of microsatellite loci and establishes a predictive model; its output is also a binary classification result of microsatellite stability.

[0008] The aforementioned existing technologies have the following technical drawbacks. First, all of these methods use the binary state of microsatellite instability (MSI-H vs. MSS, or dMMR vs. pMMR) as the basis for selecting WRN inhibitor patients, that is, whether or not a certain microsatellite instability threshold is crossed as the inclusion criterion. However, the real driving factor of WRN dependence is not whether the sample reaches the binary threshold of microsatellite instability, but is determined by the microsatellite repeat amplification load caused by microsatellite instability (especially the degree of amplification of repeat units at dinucleotide repeat sites, which is a continuous variable). Second, since the highly microsatellite unstable population is regarded as a homogeneous WRN-responsive population, binary microsatellite detection cannot further distinguish between strongly WRN-dependent subgroups and weakly or non-WRN-dependent subgroups within the sample population already identified as MSI-H; and a considerable proportion of MSI-H tumors are actually not strongly WRN-dependent. Thirdly, therefore, enrolling WRN inhibitor clinical trials in binary microsatellite states will include a large number of non-WRN-dependent MSI-H patients, resulting in insufficient responder enrichment and low enrollment accuracy, which cannot meet the high requirements for responder enrichment of WRN inhibitors, a synthetic lethal drug.

[0009] To address the technical problem of treating highly unstable microsatellite populations as homogeneous response groups and failing to enrich true WRN-dependent subgroups within these populations, it is necessary to provide an in vitro detection protocol that can stratify patients and enrich true WRN inhibitor responders using continuously quantified microsatellite repetitive amplification load. Summary of the Invention

[0010] Purpose of the invention

[0011] The purpose of this invention is to overcome the shortcomings of existing technologies that select WRN inhibitor patients based on binary microsatellite instability and treat highly unstable microsatellite populations as homogeneous response groups, thus failing to enrich true WRN-dependent subgroups within them. This invention provides an in vitro detection method, in vitro detection kit, and data processing method for non-diagnostic and non-therapeutic purposes that uses the continuous variable of microsatellite repeat amplification load quantification to continuously stratify and threshold-enrich WRN inhibitor response candidates. This improves the enrichment of WRN inhibitor responders and the accuracy of clinical trial enrollment in tumor samples, including those with highly unstable microsatellite populations.

[0012] Technical solution

[0013] To achieve the above objectives, the core of the technical solution of this invention lies in replacing the binary microsatellite instability state with a continuous score of microsatellite repeat amplification load quantification value as the basis for screening WRN inhibitor responders; and based on this, further continuous stratification and enrichment are performed within the sample population that has been determined to be highly unstable microsatellites or positive for mismatch repair defects.

[0014] According to a first aspect of the present invention, a method for in vitro identification of WRN inhibitor response candidates is provided. The method is for non-disease diagnosis and non-therapeutic purposes, comprising: providing nucleic acids from an ex vivo tumor sample from a subject; in vitro measuring the microsatellite repeat amplification load (MSBLOB) quantification value of the nucleic acid, wherein the MSBLOB quantification value is a continuous score reflecting the degree of amplification of repeat units at a plurality of microsatellite sites; comparing the MSBLOB quantification value with a pre-set threshold; and when the MSBLOB quantification value is greater than or equal to the threshold, outputting a molecular stratification result identifying the ex vivo tumor sample as a WRN inhibitor response candidate. The WRN inhibitor is an inhibitor acting on WRN helicase; the MSBLOB quantification value is a continuous variable, distinct from the binary state determination result of highly unstable and stable microsatellites.

[0015] In one embodiment of the first aspect, the microsatellite repeat amplification load quantification value is a dinucleotide-specific repeat amplification load quantification value, which is a continuous score calculated for a set of sites consisting of (TA)n repeat sites and / or (AT)n repeat sites, reflecting the degree of amplification of repeat units at that site set. This embodiment further focuses the quantification on dinucleotide repeat sites that play a dominant role in WRN-dependent mechanisms, constituting a more mechanism-specific limitation.

[0016] In one embodiment of the first aspect, the method is performed on ex vivo tumor samples that have been determined to be highly unstable microsatellitely or positive for mismatch repair deficiency. Among these samples, stratification is performed based on the microsatellite repeat amplification load quantification value, and samples with a quantification value greater than or equal to the threshold are identified as WRN inhibitor response candidates. This embodiment embodies the key concept that distinguishes the present invention from the prior art: highly unstable microsatellitely is merely a threshold; within this threshold, the strength of WRN dependence is determined by the degree of repeat amplification load. Therefore, it is necessary to further enrich the true WRN-dependent subgroups within this threshold using the load quantification value.

[0017] In one embodiment of the first aspect, the microsatellite repeat amplification load quantification value is obtained by statistically analyzing the length variation and / or amplification degree of repeat units at the plurality of microsatellite sites in the high-throughput sequencing reads of the nucleic acid; the plurality of microsatellite sites are a pan-microsatellite instability scoring site set, or a site set composed of the aforementioned (TA)n repeat sites and / or (AT)n repeat sites.

[0018] In one embodiment of the first aspect, the threshold is determined based on the continuous relationship between the microsatellite repeat amplification load quantization value and WRN dependence in the reference cohort, such that the population with the quantization value greater than or equal to the threshold is enriched with WRN inhibitor response candidates relative to the population with the quantization value less than the threshold.

[0019] In one embodiment of the first aspect, the ex vivo tumor sample is derived from one or more of colorectal cancer, endometrial cancer, gastric cancer, and ovarian cancer; the ex vivo tumor sample is a tumor tissue sample or a peripheral blood cell-free DNA sample, and the nucleic acid is deoxyribonucleic acid.

[0020] According to a second aspect of the present invention, an in vitro detection kit for screening WRN inhibitor response candidates is provided. The kit comprises: nucleic acid primers and / or capture probes targeting a plurality of microsatellite loci for in vitro acquisition of nucleic acid data required to determine a microsatellite repeat amplification load quantification value, the microsatellite repeat amplification load quantification value being a continuous score reflecting the degree of amplification of repeat units at the plurality of microsatellite loci; and a determination component comprising a threshold parameter for comparing the microsatellite repeat amplification load quantification value with a pre-set threshold and identifying a WRN inhibitor response candidate when the quantification value is greater than or equal to the threshold, and a determination description. The WRN inhibitor is an inhibitor acting on WRN helicase.

[0021] In the second aspect of the implementation, the kit may be a high-throughput sequencing kit, and further includes library preparation reagents and / or sequencing adapters; the nucleic acid primers and / or capture probes may target a set of sites consisting of (TA)n repeat sites and / or (AT)n repeat sites; the WRN inhibitor may be an inhibitor of covalently binding WRN helicase or non-covalently binding WRN helicase, and the subject is a subject who receives or intends to receive a WRN inhibitor.

[0022] The WRN inhibitors mentioned above are all general concepts, that is, any inhibitor that acts on WRN helicase, and are not limited to any single specific drug, so that the technical solution of the present invention is applicable to all WRN inhibitors in this category.

[0023] Beneficial effects

[0024] Compared with the prior art, the present invention has the following beneficial effects.

[0025] First, this invention uses the microsatellite repeat amplification load quantification value as a continuous variable instead of the binary microsatellite instability state as the basis for patient selection, overcoming the systematic bias of binary detection that treats highly microsatellite unstable groups as homogeneous response groups. Retrospective calculations based on this invention have verified that within the highly microsatellite unstable group, a considerable proportion (approximately 48% in the validated dataset) of samples are not strongly WRN-dependent. Stratification using continuous load quantification values ​​and threshold enrichment can remove these non-dependent samples from the group, thereby significantly improving the enrichment of WRN inhibitor responders.

[0026] Second, the microsatellite repeat amplification load quantification value and WRN dependence strength used in this invention show a significant continuous correlation within highly unstable microsatellite populations (in the validated dataset, the correlation coefficient between load quantification value and WRN dependence within this population reaches Spearman ρ=-0.716), thus providing a more refined responder ranking and enrollment priority than the binary state. This continuous correlation is diluted after including stable microsatellite samples (the correlation coefficient for all samples is only ρ=-0.079), illustrating the technical concept of "microsatellite instability as a threshold, within which dependence is determined by the load level".

[0027] Third, the technical solution of the present invention is applicable to any inhibitor that acts on WRN helicase, and does not depend on the success or failure of any single investigational drug, thereby improving the robustness of WRN inhibitor patient selection and companion diagnostic deployment at the category level.

[0028] The purpose of the detection described in this invention is to predict and enrich the efficacy of WRN inhibitors, rather than to determine the prognosis of the disease; this invention does not advocate that the microsatellite repeat amplification load quantification value can be used as an independent prognostic biomarker. This definition is used to clarify the technical contribution boundary of this invention. Attached Figure Description

[0029] Figure 1 is a schematic diagram of the overall process of the in vitro labeling WRN inhibitor response candidate method of the present invention.

[0030] Figure 2 is a schematic diagram of the continuous relationship between the microsatellite repeat amplification load quantization value and WRN dependence in this invention.

[0031] Figure 3 is a schematic diagram of the in vitro detection kit and supporting data processing module structure for screening WRN inhibitor response candidates according to the present invention.

[0032] For ease of understanding, the reference numerals and their meanings used in the above figures are explained uniformly as follows. In Figure 1: S1 is the step of providing nucleic acids from an ex vivo tumor sample from the subject; S2 is the step of determining the microsatellite repeat amplification load quantification value of the nucleic acid in vitro; S3 is the step of comparing the microsatellite repeat amplification load quantification value with a pre-set threshold; S4 is the step of outputting molecular stratification results identifying the sample as a WRN inhibitor response candidate when the quantification value is greater than or equal to the threshold; S5 is the step of outputting molecular stratification results identifying the sample as a non-WRN inhibitor response candidate when the quantification value is less than the threshold; S21 is the step of calculating the quantification value for the pan-microsatellite instability scoring site set; S22 is the step of calculating the quantification value for the (TA)n and / or (AT)n dinucleotide specific repeat site set; S31 is the step of determining whether the sample is microsatellite highly unstable or mismatch repair defect positive; S32 is the step of stratifying samples within microsatellite highly unstable or mismatch repair defect positive samples based on quantification values. In Figure 2: 201 is the coordinate axis for the quantified value of microsatellite repeat amplification load; 202 is the coordinate axis for WRN dependence; 203 is the continuous relationship curve between the quantified value and WRN dependence within the highly unstable microsatellite sample population; 204 is the relationship curve of all samples after including stable microsatellite samples; 205 is the pre-set threshold position; 206 is the candidate region for WRN inhibitor response enriched above the threshold; 207 is the non-response candidate region below the threshold. In Figure 3: M1 is the nucleic acid processing module; M2 is the microsatellite repeat amplification load quantification module; M3 is the threshold comparison and identification module; M4 is the result output module; 301 is the nucleic acid primer and / or capture probe; 302 is the library preparation reagent and / or sequencing adapter; 303 is the threshold parameter and judgment explanation; 304 is the quantified value of microsatellite repeat amplification load; 305 is the WRN inhibitor response candidate identification result. Detailed Implementation

[0033] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to explain the present invention and are not intended to limit the scope of protection of the present invention. Those skilled in the art can reproduce the present invention based on the content described in the following embodiments without creative effort, and can make adaptive substitutions for specific parameters, site sets, detection platforms, tumor types and sample types.

[0034] In this specification, unless otherwise stated, "microsatellite repeat amplification load quantification value" uniformly refers to a continuous score reflecting the degree of amplification of repeat units at multiple microsatellite sites. It is a continuous variable, distinct from the binary state determination result that classifies a sample as microsatellite highly unstable (MSI-H) or microsatellite stable (MSS); "(TA) / (AT) dinucleotide specific repeat amplification load quantification value" uniformly refers to the microsatellite repeat amplification load quantification value calculated for the site set consisting of (TA)n repeat sites and / or (AT)n repeat sites; "WRN inhibitor" uniformly refers to any inhibitor that acts on WRN helicase, not limited to any single specific drug; "WRN inhibitor response candidate" uniformly refers to the molecular stratification intermediate result corresponding to an ex vivo tumor sample identified by the method of this invention as potentially benefiting from WRN inhibitor treatment. This identification is an intermediate result output after processing the obtained detection data and does not constitute a diagnostic conclusion for the disease, nor does it contain any drug administration or treatment steps. In this specification, steps are designated by numbers such as S1, S2, etc., and modules are designated by numbers such as M1, M2, etc. The reference numerals used are consistent with those shown in the accompanying drawings.

[0035] The calculations involved in the various embodiments of this invention can be implemented in software on general-purpose computing devices, and the required sequencing data can be obtained from a high-throughput sequencing platform. In a typical implementation environment, the computing device is equipped with a processor and a memory, the memory storing instructions for performing microsatellite repeat amplification load quantization, threshold comparison, and result labeling; the high-throughput sequencing platform is used to sequence nucleic acids in ex vivo tumor samples to generate sequencing reads. The objects processed in each step are nucleic acids in ex vivo tumor samples or sequencing data generated therefrom, and steps involving obtaining samples from living organisms are not involved.

[0036] Example 1: Complete method flow for in vitro labeling of WRN inhibitor response candidates

[0037] This embodiment provides a complete method flow for identifying WRN inhibitor response candidates in vitro, as shown in Figure 1. The flow includes steps S1 to S4 in sequence, and steps S31 and S32 can be additionally performed in the subset of samples that are determined to be MSI-H / dMMR positive.

[0038] Step S1 involves providing nucleic acids from an ex vivo tumor sample obtained from the subject. The ex vivo tumor sample can be an ex vivo tumor tissue sample or a peripheral blood cell-free DNA (circulating tumor DNA, ctDNA) sample. For tumor tissue samples, fresh frozen tissue or formalin-fixed paraffin-embedded (FFPE) tissue can be used. Deoxyribonucleic acid (DNA) is extracted after tissue lysis and protease digestion. For peripheral blood cell-free DNA samples, cell-free DNA is extracted after plasma separation. In a specific example, the total amount of extracted DNA is not less than 10 nanograms, and after quality assessment confirms that the fragment distribution and concentration are within acceptable limits, it proceeds to subsequent steps. The extracted nucleic acid is the target for measurement in subsequent steps; this step does not include obtaining samples from a living organism.

[0039] Step S2 involves in vitro determination of the microsatellite repeat amplification load (MSBLoad) quantification value of the nucleic acid. The MSBLoad quantification value is a continuous score reflecting the degree of amplification of repeat units at multiple microsatellite loci. In this step, the nucleic acid obtained in Step S1 is first used to construct a high-throughput sequencing library. Library construction includes DNA fragmentation, end repair, adapter ligation, and target region enrichment for the multiple microsatellite loci. Sequencing is then performed on a high-throughput sequencing platform to obtain sequencing reads covering the multiple microsatellite loci. Finally, the sequencing reads are aligned, and the length variation and / or amplification degree of repeat units at the multiple microsatellite loci are statistically analyzed to calculate the MSBLoad quantification value. In a specific example, target region enrichment uses a capture probe designed for the microsatellite loci for hybridization capture, with a sequencing target depth of at least 200-fold. After read alignment, the variation distribution of repeat unit length relative to a reference length is statistically analyzed for each microsatellite locus, and the amplification signals at each locus are summarized into a continuous score as the MSBLoad quantification value.

[0040] This step can be implemented in two ways to obtain the quantized value, corresponding to steps S21 and S22 in Figure 1, respectively.

[0041] In the basic implementation corresponding to step S21, the plurality of microsatellite loci is a pan-microsatellite instability scoring locus set. This locus set covers microsatellite loci widely distributed in the genome. For each locus, the degree of length variation or amplification of its repeat units is statistically analyzed, and the proportion of loci with significant length variation or the weighted sum of the amplification degree of each locus is calculated as a pan-microsatellite repeat amplification load quantification value. In a specific example, this quantification method can be implemented based on the statistical framework of MSIsensor-like tools, that is, comparing the length distribution of each microsatellite locus in the locus set of tumor samples and controls, and outputting a continuous score reflecting the amplification degree of each locus; unlike the prior art that uses a fixed threshold to binary classify the score into MSI-H / MSS, this implementation retains the score as a continuous quantification value for subsequent continuous stratification.

[0042] To facilitate reproduction by those skilled in the art, the calculation of the quantification value of the microsatellite repeat amplification load is further explained below. Assume the microsatellite instability scoring locus set contains N microsatellite loci. For the i-th locus, the copy number of the repeat units corresponding to all reads covering that locus is extracted from the aligned sequencing reads, yielding the repeat length distribution of that locus in the sample. This sample distribution is compared with the reference length distribution or normal control distribution for the same locus. A distribution distance metric (e.g., chi-square statistic or Kolmogorov–Smirnov distance) is used to determine whether the locus has undergone significant length variation. If variation is determined, the net increment of the repeat units relative to the reference length is further calculated as the amplification degree of that locus. Based on this, the quantification value of the microsatellite repeat amplification load can be calculated using the following formula: the quantification value equals the proportion of the number of loci with significant length variation to the total number of effective loci assessed, or equals the normalized value obtained by weighting the amplification degree of each locus according to its locus weight. The locus weight can be set based on the frequency of amplification of each locus in the target tumor type and the correlation strength between its amplification degree and WRN dependence. Regardless of the aggregation method used, the resulting quantification values ​​are continuous, thus preserving the intra-population resolution lost in binary decision-making. In a specific example, a valid site refers to a site in the sample whose sequencing coverage depth reaches a set lower limit (e.g., not less than 20-fold); only valid sites are used in the above calculations to avoid bias introduced by low-coverage sites.

[0043] In the advanced implementation corresponding to step S22, the plurality of microsatellite sites are a set of sites composed of (TA)n repeat sites and / or (AT)n repeat sites, and the quantification value is the quantification value of dinucleotide-specific repeat amplification load. This implementation limits the quantification object to the (TA) / (AT) dinucleotide repeat sites that play a dominant role in the WRN-dependent mechanism: the amplified (TA)n repeats tend to form non-B-type DNA secondary structures, which relieve the highly WRN-dependent unwinding activity. Therefore, the correlation between the amplification degree of repeat units at (TA) / (AT) dinucleotide repeat sites and the WRN dependence strength is more direct than that of pan-microsatellite sites. In a specific example, firstly, (TA)n and (AT)n dinucleotide repeat sites are screened from the genome to form a set of specific sites. Capture probes targeting this set of specific sites are designed for target region enrichment and sequencing. Then, the length variation and amplification degree of repeat units on this set of specific sites are statistically analyzed to calculate the quantification value of dinucleotide-specific repeat amplification load. The selection of the specific locus set can be optimized based on the distribution of (TA)n / (AT)n repeat sites in the reference genome, the frequency of amplification of each site in the target tumor type, and the correlation strength between the amplification degree of each site and WRN dependence, so as to control the number of sites while preserving discriminative performance.

[0044] The calculation of the dinucleotide-specific repeat amplification load quantification value can be performed in accordance with the summarization method described in step S21, the difference being that its site set consists only of (TA)n and (AT)n dinucleotide repeat sites. In a specific example, the construction of the specific site set includes the following sub-steps: First, sites in the reference genome with repeat units of TA or AT and consecutive repeat counts not less than a set lower limit (e.g., not less than 8 times) are selected as candidate sites; second, the amplification frequency and amplification amplitude of repeat units at each candidate site are evaluated in the reference cohort, and sites that rarely amplify in the target tumor type or have high length polymorphism in normal samples and are prone to introducing background noise are removed; finally, among the remaining sites, those with a strong correlation between their amplification degree and WRN dependence are retained to form the specific site set. For each specific site, the copy number distribution of (TA) / (AT) repeat units is extracted from the covered read, and its net increment relative to the reference length is calculated as the amplification degree of the site, and the dinucleotide-specific repeat amplification load quantification value is obtained according to the weighted summarization method described above. Since (TA) / (AT) dinucleotide repeats are the main source of the aforementioned non-B-type DNA secondary structures, it is possible to retain or even enhance the ability to distinguish WRN-dependent structures while controlling the number of sites. This specific quantification method constitutes an implementation of the present invention in terms of mechanism specificity, and is also one of the key differences between the present invention and existing technologies that determine binary states solely based on microsatellite sites.

[0045] Step S3 involves comparing the quantified microsatellite repeat amplification load (MSBL) value with a pre-set threshold. The threshold is determined based on the continuous relationship between the MSBL quantified value and WRN dependence in the reference cohort, ensuring that individuals with MSBL values ​​greater than or equal to the threshold are enriched with WRN inhibitor response candidates compared to individuals with MSBL values ​​less than the threshold. The method for determining the threshold will be specifically explained in Example 2 with validation data.

[0046] Step S4: When the quantified value of the microsatellite repeat amplification load is greater than or equal to the threshold, a molecular stratification result identifying the ex vivo tumor sample as a candidate for a WRN inhibitor response is output. Correspondingly, as shown in step S5 of Figure 1, when the quantified value is less than the threshold, a molecular stratification result identifying the ex vivo tumor sample as a non-WRN inhibitor response candidate is output. This output is an intermediate molecular stratification result obtained after processing the acquired detection data; it does not include any drug administration or treatment steps and does not constitute a diagnostic conclusion for the disease.

[0047] In a preferred embodiment, the method is performed on a subset of ex vivo tumor samples that have been identified as highly unstable microsatellite activity or positive for mismatch repair deficiency, as shown in steps S31 and S32 of Figure 1. Step S31 involves determining whether the ex vivo tumor sample is highly unstable microsatellite activity or positive for mismatch repair deficiency; this can be pre-determined using existing binary microsatellite detection methods (such as IHC, PCR capillary electrophoresis, or NGS binary classification). Step S32 involves stratifying the samples identified as highly unstable microsatellite activity or positive for mismatch repair deficiency within this positive subset based on the microsatellite repeat amplification load quantification value obtained in step S2. Samples with quantification values ​​greater than or equal to the threshold are enriched from the highly unstable microsatellite activity or mismatch repair deficiency population and identified as WRN inhibitor response candidates. This preferred embodiment directly corresponds to the core concept that distinguishes this invention from the prior art: microsatellite high instability only constitutes a threshold for WRN dependence, and the strength of WRN dependence is determined by the degree of microsatellite repeated amplification load within the threshold; therefore, even within a population that has been determined to be highly unstable microsatellites, it is still necessary to further stratify using continuous load quantification values ​​to eliminate samples with low loads and not strong WRN dependence in the population, thereby achieving enrichment of truly WRN-dependent subpopulations.

[0048] In the method described in this embodiment, the WRN inhibitor is any inhibitor that acts on WRN helicase, including inhibitors that covalently bind to WRN helicase and inhibitors that do not covalently bind to WRN helicase. Investigative inhibitors of this type include HRO761 and VVD-214 (also known as RO7589831 or VVD-133214). The subject is a subject who has received or intends to receive a WRN inhibitor. The method is applicable to one or more of colorectal cancer, endometrial cancer, gastric cancer, and ovarian cancer, where microsatellite instability is common, and there is a need for WRN-dependent applications stratified by microsatellite repeat amplification load.

[0049] When implementing this method on different sample types, steps S1 and S2 can be adaptively adjusted. For tumor tissue samples, the tumor cell content can be confirmed by pathological evaluation before nucleic acid extraction in step S1 to ensure that the measured signal mainly originates from the tumor. When the tumor cell content is low, the measured site amplification signal can be corrected by combining tumor purity to reduce the dilution of the quantification value by the normal cell background. For peripheral blood cell-free DNA samples, since the proportion of tumor-derived DNA is usually low and the degree of fragmentation is high, a higher sequencing target depth can be used in step S2, combined with molecular tag deduplication, to improve the detection sensitivity of low abundance amplification signals. At the same time, a minimum effective tumor fraction threshold can be set, indicating that the reliability of the quantification value is limited when the tumor fraction of the sample is lower than this threshold. The above adjustments do not change the core concept of stratification based on continuous load quantification values, but are only used to ensure the reliability of quantification values ​​under different sample types.

[0050] When implementing this method on different tumor types, reference cohorts can be constructed for each tumor type to calibrate corresponding thresholds, and the site composition of the (TA) / (AT) dinucleotide-specific site set can be optimized for each tumor type, as the microsatellite site profiles showing significant amplification may differ across tumor types. By calibrating thresholds and optimizing site sets for each tumor type, this method can maintain its enrichment capacity for WRN-dependent subpopulations in all target tumor types.

[0051] In a complete application example, this method was applied to an ex vivo tumor tissue sample from a colorectal cancer patient who had been identified as having high microsatellite instability by existing binary microsatellite detection: DNA was extracted in step S1; in step S2, target region enrichment, sequencing, and quantification were performed using a (TA) / (AT) dinucleotide-specific site set to obtain the quantified value of the sample's dinucleotide-specific repeat amplification load; in step S3, this quantified value was compared with a threshold defined for colorectal cancer; if the quantified value was greater than or equal to the threshold, step S4 output the molecular stratification result identifying the sample as a WRN inhibitor response candidate, suggesting that the patient was more likely to benefit from WRN inhibitor treatment, thus further enriching them as a priority enrollment candidate within the already identified high microsatellite instability population; if the quantified value was less than the threshold, step S5 output the molecular stratification result identifying the sample as a non-WRN inhibitor response candidate. All the above outputs are intermediate molecular stratification results; whether to administer WRN inhibitor treatment to the patient based on these results is a separate clinical decision and not part of the steps in this method.

[0052] To illustrate the adjustable range of each parameter in this embodiment, the values ​​of several key parameters are further explained below. In the target region enrichment and sequencing in step S2, the sequencing target depth can be between 100x and 1000x, for example, 200x, 500x, or 800x. For peripheral blood cell-free DNA samples, since the proportion of tumor-derived DNA is relatively low, a higher sequencing target depth can be used, for example, not less than 500x, combined with molecular tag deduplication to improve the detection of low-abundance signals. In the construction of the dinucleotide-specific site set in step S22, the number of sites in the specific site set can be between tens and thousands, for example, 50, 200, 1000, or more. Increasing the number of sites is beneficial to improving the stability of the quantification value, while reducing the number of sites is beneficial to reducing detection costs and improving adaptability to low-quality samples. A trade-off can be struck between these two factors depending on the application scenario. In the threshold determination in step S3, the threshold depends on the quantization method, site set, and target enrichment level used. It should be calibrated based on the continuous relationship between the quantization value and WRN dependence in the reference queue, rather than using the fixed threshold used to determine the unstable binary state of microsatellites.

[0053] Example 2: Retrospective computation verification based on publicly available queue data

[0054] This embodiment provides the process and results of retrospective calculation verification of the technical solution described in Embodiment 1 based on publicly available cohort data. This demonstrates the beneficial effect of continuous stratification using microsatellite repeat amplification load quantification values ​​compared to binary microsatellite state selection, and explains the basis for determining the threshold. The verification in this embodiment is based on in vitro computational analysis of a publicly available dataset. The data and results were calculated by the inventors themselves and are sufficient to support the feasibility of the technical solution of this invention.

[0055] Datasets and Samples. This validation used publicly available tumor cell line functional genomics datasets from the DepMap (Cancer Dependency Map) project, specifically the 26Q1 version of CRISPR gene knockout gene effect data and the 24Q4 version of microsatellite instability score data. After intersecting the two datasets by cell line identifier and removing cell lines missing key variables, a total of 1192 tumor cell lines were included, of which 75 were microsatellite unstable and 1117 were microsatellite stable. The microsatellite status of each cell line was characterized by a microsatellite instability score (MSIScore calculated by a tool like MSIsensor2), with an MSIScore greater than 20 used as the cutoff value for microsatellite instability. The dependence of each cell line on WRN was characterized by a gene effect value based on CRISPR gene knockout; the lower the gene effect value, the stronger the inhibition of cell viability by WRN knockout, i.e., the stronger the WRN dependence. The gene effect value and microsatellite instability score mentioned are both quantitative indicators already provided in publicly available datasets. This validation does not conduct separate sequencing or library construction experiments; it only performs statistical analysis on the aforementioned publicly available quantitative indicators. In this validation, the microsatellite instability score (MSIScore) is used as a representative implementation of the microsatellite repeat amplification load quantification value (corresponding to the pan-microsatellite quantification implementation method in step S21 of Example 1), and the WRN gene effect value is used as a measure of WRN dependence. This validation uses the microsatellite instability score as a representative implementation of the microsatellite repeat amplification load quantification value because the score itself reflects the degree of variation in repeat unit length at the pan-microsatellite locus and can be used as a continuous measure of the pan-microsatellite repeat amplification load. In practical implementation, the quantitative value can also be directly calculated from the sequencing reads according to the summarization method described in step S21 or step S22 of Example 1. The two are essentially consistent in reflecting the degree of repeat amplification with continuous quantitative values.

[0056] Validation Result 1: Overall difference in WRN dependence between microsatellite unstable and microsatellite stable populations. Among the 1192 cell lines included, the median WRN gene effect value was -0.562 for microsatellite unstable cell lines and -0.115 for microsatellite stable cell lines. The difference between the two groups was statistically significant according to the Mann-Whitney test (P = 1.75 × 10⁻¹²). This result confirms the association between microsatellite instability and WRN dependence at the population level, indicating that microsatellite instability is a threshold condition for WRN dependence. This is consistent with the general understanding in existing techniques that use microsatellite instability as an inclusion threshold.

[0057] Verification Result 2: The microsatellite highly unstable population is not a homogeneous WRN-dependent population. Further analysis of the 75 microsatellite unstable cell lines revealed that approximately 48% of these cell lines had WRN gene effect values ​​of at least -0.5, indicating that they did not constitute strong WRN dependence. This result directly demonstrates that treating the microsatellite highly unstable population as a homogeneous WRN response population introduces a systematic bias. Including the population in a binary microsatellite state will include nearly half of the non-strongly dependent samples, thereby diluting the enrichment of responders. This result provides the data basis for the beneficial effect described in this invention: "Approximately 48% of non-dependent samples can be removed from the microsatellite highly unstable population."

[0058] Validation Result 3: Continuous Correlation Between Microsatellite Repeat Expansion Load Quantification and WRN Dependence within Highly Instable Microsatellite Populations. Within the 75 microsatellite unstable cell lines mentioned above, a significant continuous correlation was found between the microsatellite repeat expansion load quantification value (characterized by MSIScore) and WRN dependence (characterized by WRN gene effect value), with a Spearman correlation coefficient ρ of -0.716 and a p-value of 5.3 × 10⁻¹³. Since a lower WRN gene effect value indicates stronger dependence, this negative correlation indicates that a higher microsatellite repeat expansion load quantification value corresponds to stronger WRN dependence. This continuous correlation is shown as curve 203 in Figure 2. This result demonstrates that within the threshold of microsatellite instability, the strength of WRN dependence is continuously determined by the degree of microsatellite repeat expansion load, thus providing a direct basis for stratification and ranking based on continuous quantification values. This fine stratification is impossible to achieve with existing techniques that only provide MSI-H / MSS binary results.

[0059] Verification Result 4: The continuous correlation is diluted in all samples by microsatellite stable samples. When 1117 microsatellite stable cell lines were included in all 1192 cell lines for the same correlation analysis, the Spearman correlation coefficient ρ between microsatellite repeat amplification load quantification value and WRN dependence was only -0.079, as shown by curve 204 in Figure 2. This result contrasts with Verification Result 3, indicating that the continuous relationship between load quantification value and WRN dependence mainly exists in samples that cross the microsatellite instability threshold, and does not hold in microsatellite stable samples that do not cross the threshold. If all samples are sorted directly by load quantification value without distinguishing the threshold, this continuous relationship will be diluted by microsatellite stable samples. This contrasting result supports the technical concept of the present invention, "microsatellite instability as the threshold, and dependence determined by the load level within the threshold," and also explains why the present invention can stratify by load quantification value while retaining stratification within the positive subset of highly unstable microsatellites or mismatch repair defects (Example 1, steps S31, S32).

[0060] Summary of Beneficial Effects. Based on the four verification results above, it is evident that while existing techniques using binary microsatellite state-based patient selection can identify the overall association between microsatellite instability and WRN dependence at the population level (Verification Result 1), they treat highly microsatellite-instable populations as homogeneous response groups. Therefore, they cannot identify the approximately 48% of non-strongly dependent samples within this population (Verification Result 2), nor can they utilize the significant continuous relationship between the quantified load value and WRN dependence within this population (Verification Result 3) for fine-grained ranking. In contrast, this invention uses continuous quantified load values ​​for stratification and threshold enrichment within a threshold, effectively eliminating non-strongly dependent samples and ranking candidate samples based on this continuous relationship, thus significantly improving the enrichment of WRN inhibitor responders compared to binary selection. Verification Result 4 indicates that this stratification must be performed within the microsatellite instability threshold to avoid dilution of the continuous relationship by samples outside the threshold, which is precisely the strategy adopted in the preferred embodiment of this invention. Therefore, the beneficial effects of this invention compared to existing technologies are supported by sufficient data, rather than merely inference.

[0061] Basis and Examples for Threshold Determination. Based on the above verification results, the threshold should not be determined using the fixed cutoff value used to determine the binary state of microsatellite instability. Instead, it should be calibrated based on the continuous relationship between the load quantification value and WRN dependence in the reference cohort, so that the population above the threshold is enriched with WRN inhibitor response candidates relative to the population below the threshold. The WRN dependence ratio (or expected response rate) of samples above the threshold under different candidate thresholds can be plotted in the reference cohort, and the quantification value that makes this ratio reach the preset enrichment target can be selected as the threshold. The preset enrichment target can be set according to the accuracy requirements of clinical trial enrollment. In this verification, the MSIScore greater than 20 is used as the cutoff value for determining microsatellite instability only for characterizing the threshold of microsatellite state, and is not the threshold used in this invention for enriching WRN inhibitor response candidates; the latter should be determined separately according to the above-mentioned calibration method based on continuous relationship, and varies with the quantification method, site set, and target enrichment degree.

[0062] The specific steps for threshold calibration can be implemented as follows in one example: First, determine the microsatellite repeat amplification load (WRN) quantification value for each sample in the reference cohort, and use its WRN dependence measure (e.g., defining WRN dependence as a WRN gene effect value not exceeding a set threshold) as a reference label. Second, set a series of candidate thresholds within the range of quantification values. For each candidate threshold, calculate the proportion of WRN-dependent samples in the subset of samples with quantification values ​​greater than or equal to that candidate threshold; this proportion is the enriched positive prediction ratio at that candidate threshold. Simultaneously, record the proportion of samples included in the subset at that candidate threshold relative to the reference cohort; this proportion reflects the enrollment coverage. Third, weigh the enriched positive prediction ratio against the enrollment coverage: a higher candidate threshold results in a higher enriched positive prediction ratio but lower enrollment coverage; a lower candidate threshold results in higher enrollment coverage but weaker enrichment. Fourth, select the candidate threshold that achieves the preset target enriched positive prediction ratio while meeting the preset lower limit of enrollment coverage as the final threshold. The thresholds obtained through the above steps significantly increase the WRN dependence ratio of the population above the threshold compared to the population below the threshold, thereby enriching the response candidates.

[0063] Parameter sensitivity explanation. The threshold values ​​defined in this invention and the resulting enrichment effects vary depending on the quantification method and locus set. Those skilled in the art can choose according to application requirements. Taking the microsatellite quantification implementation verified in this embodiment as an example, when the threshold is set higher, the proportion of WRN-dependent samples in the subset above the threshold increases, enhancing the enrichment effect, but reducing the number of included samples; when the threshold is set lower, the number of included samples increases, but the proportion of non-dependent samples mixed in increases, weakening the enrichment effect. When the (TA) / (AT) dinucleotide-specific quantification implementation described in step S22 of Example 1 is used instead of the microsatellite quantification implementation, since the specific locus set focuses more on the direct driving factors of WRN dependence, the positive prediction ratio after enrichment is expected to be further improved under the same enrollment coverage. The corresponding threshold also needs to be separately defined within the range of the specific quantification value. The above parameter sensitivity indicates that the technical effect of this invention does not depend on any fixed quantification method, locus set, or threshold value, but on the technical concept itself of stratifying with continuous load quantification values ​​and defining the threshold according to the continuous relationship.

[0064] Performance Enhancement and Boundary Clarification. The retrospective calculations based on the public cohort described above are sufficient to support the feasibility of the technical solution of this invention. When implementing this invention into a specific detection product, the performance of the detection platform can be further enhanced through wet experiments, including verifying the detection limit, precision, and repeatability of the adopted quantification method on the target platform, and further confirming the correspondence between the quantification value and the WRN-dependent phenotype on the matching samples; such performance enhancement is used in the product implementation stage and does not affect the feasibility of the technical solution of this invention. In addition, as stated in the invention summary section, the purpose of the detection described in this invention is to predict and enrich the efficacy of WRN inhibitors, rather than to judge the prognosis of the disease. This invention does not advocate that the quantification value of the microsatellite repeat amplification load can be used as an independent prognostic biomarker; this definition is used to clarify the boundary of the technical contribution of this invention.

[0065] Literature support. The above-mentioned technical concept of characterizing WRN dependence by microsatellite repeat amplification load, especially the degree of (TA)n dinucleotide repeat amplification, is consistent with the trends reported in relevant studies. Studies have reported a positive correlation between the degree of (TA)n dinucleotide repeat amplification and the drug sensitivity index (IC50) of WRN inhibitors in preclinical models, with a goodness of fit r² reaching 0.72 in all colorectal cancer organoid models (r² is approximately 0.6 in the microsatellite instability model subset) (Picco G et al., Cancer Discovery, 2024, Vol. 14, p. 1457). It should be noted that this correlation describes the association between the degree of (TA)n repeat amplification and drug sensitivity IC50, not the association with CRISPR gene effect values; the two have different scopes. Based on this, the study proposes that microsatellite instability detection should be combined with quantitative analysis of (TA)n repeat amplification to more accurately identify WRN dependence. This corresponds to the advanced implementation method of this invention, which uses the quantitative value of (TA) / (AT) dinucleotide-specific repeat amplification load for fine stratification (Example 1, Step S22). Other studies have reported that the large-scale repetitive amplification of (TA)n and other repeatings in microsatellite unstable tumors constitutes a solidified damage load in the genome, and its release depends on the unwinding activity of WRNs (van Wietmarschen N et al., Nature, 2020, Vol. 586, p. 292). Based on this mechanism, it can be inferred that WRN dependence is determined by the established repetitive amplification load, rather than by the mismatch repair defect state itself. This aligns with the present invention's concept of using the quantitative value of the repetitive amplification load, rather than the mismatch repair state, as the stratification basis. The aforementioned literature serves to illustrate the mechanism and quantitative background of the technical concept of this invention, while the beneficial effects of this invention are directly based on the aforementioned retrospective calculation verification results based on a public cohort.

[0066] Example 3: In vitro detection kit and associated data processing implementation for screening WRN inhibitor response candidates

[0067] This embodiment provides an in vitro detection kit for screening WRN inhibitor response candidates, and a corresponding data processing implementation method for non-diagnostic and non-therapeutic purposes, the module structure of which is shown in Figure 3.

[0068] The in vitro detection kit includes nucleic acid primers and / or capture probes 301 and a decision component. The nucleic acid primers and / or capture probes 301 target multiple microsatellite sites for obtaining nucleic acid data required to determine the microsatellite repeat amplification load quantification value in vitro. In one embodiment, the nucleic acid primers and / or capture probes 301 are designed for a site set consisting of (TA)n repeat sites and / or (AT)n repeat sites to obtain nucleic acid data required to calculate the dinucleotide-specific repeat amplification load quantification value. When the kit is a high-throughput sequencing kit, the kit also includes library preparation reagents and / or sequencing adapters 302. The library preparation reagents are used to fragment, end-repair, and ligate adapters to nucleic acids from ex vivo tumor samples to construct a sequencing library. The decision component includes a threshold parameter and a decision description 303. The threshold parameter is a threshold determined based on the continuity of load quantification values ​​and WRN dependence in a reference cohort. The decision description describes a decision rule for comparing the measured microsatellite repeat amplification load quantification value with the threshold and identifying WRN inhibitor response candidates when the quantification value is greater than or equal to the threshold. The determination component may be provided in the form of a documentation document containing threshold parameters, or in the form of data processing instructions that carry the above determination rules.

[0069] Data processing in conjunction with the kit can be implemented by several functional modules as shown in Figure 3. The nucleic acid processing module M1 receives nucleic acid from ex vivo tumor samples, uses nucleic acid primers and / or capture probes 301 and library construction reagents and / or sequencing adapters 302 to complete target region enrichment and library construction, and obtains sequencing reads through sequencing. The microsatellite repeat amplification load quantification module M2 compares the sequencing reads, statistically analyzes the length variation and / or amplification degree of repeat units at multiple microsatellite sites, and calculates the microsatellite repeat amplification load quantification value 304. The threshold comparison and identification module M3 compares the microsatellite repeat amplification load quantification value 304 with the threshold recorded in the threshold parameter 303. When the quantification value is greater than or equal to the threshold, the sample is identified as a WRN inhibitor response candidate. The result output module M4 outputs the WRN inhibitor response candidate identification result 305, which is an intermediate result of molecular stratification and does not include drug administration or treatment steps.

[0070] In one embodiment, the data processing can be implemented independently as an in vitro, non-diagnostic, and non-therapeutic data processing method. This method processes the obtained sequencing data from the plurality of microsatellite loci, calculates quantification values ​​via a microsatellite repeat amplification load quantification module M2, compares and identifies the samples via a threshold comparison and identification module M3, and finally outputs molecular stratification identification results for WRN inhibitor response candidates via a result output module M4. The output of this method is limited to intermediate molecular stratification results and does not directly provide a disease diagnosis conclusion, nor does it include any drug administration or treatment steps. This data processing method may include: receiving obtained sequencing reads covering the plurality of microsatellite loci or alignment results obtained therefrom; statistically analyzing the length variation and / or amplification degree of repeat units at the plurality of microsatellite loci and calculating the microsatellite repeat amplification load quantification value according to the aforementioned summarization method; reading the threshold value recorded in the threshold parameter and comparing it with the quantification value; when the quantification value is greater than or equal to the threshold, outputting a molecular stratification result identifying the corresponding sample as a WRN inhibitor response candidate; otherwise, outputting a molecular stratification result identifying the sample as a non-WRN inhibitor response candidate. The data processed in each of the above steps are all data that have been ex vivo and sequenced. It does not involve obtaining samples from living organisms, nor does it involve administering any treatment to the subjects corresponding to the identified samples. Therefore, the data processing method described is an in vitro method that is not for diagnostic or therapeutic purposes.

[0071] In another embodiment, the above-mentioned functional modules can be integrated into a computing device, which includes a processor and a memory. The memory stores computer-executable instructions, which, when executed by the processor, realize the functions of the microsatellite repeat amplification load quantification module M2, the threshold comparison and identification module M3, and the result output module M4. The target region enrichment, library construction, and sequencing completed by the nucleic acid processing module M1 are completed by the supporting experimental operations and high-throughput sequencing platform, and its product is the sequencing data processed by the computing device. The data flow between the functional modules is as follows: the nucleic acid processing module M1 provides sequencing reads to the microsatellite repeat amplification load quantification module M2; the microsatellite repeat amplification load quantification module M2 provides microsatellite repeat amplification load quantification value 304 to the threshold comparison and identification module M3; the threshold comparison and identification module M3 compares according to the threshold parameter 303 and provides identification judgment to the result output module M4; the result output module M4 outputs WRN inhibitor response candidate identification results 305.

[0072] In one specific instance, the components of the in vitro diagnostic kit may include: nucleic acid primers and / or capture probes designed for the plurality of microsatellite sites or (TA) / (AT) dinucleotide-specific site sets; fragmentation enzymes, terminal repair enzymes, ligases, and sequencing adapters for library construction; hybridization buffers and washing reagents for target region enrichment; and a decision component containing threshold parameters and decision rules. The threshold parameters in the decision component may be provided separately for different tumor types or sample types to adapt to corresponding application scenarios.

[0073] The WRN inhibitors used in the kit and associated data processing are any inhibitors that act on WRN helicases, including both covalently bound and non-covalently bound WRN helicase inhibitors; the subjects are those who have received or intend to receive WRN inhibitors.

[0074] Explanation of Alternative Solutions

[0075] The technical solution of this invention can be implemented in various alternative ways without departing from its core concept. Regarding the quantification method, the quantification value of microsatellite repeat amplification load can be implemented using the method based on the pan-microsatellite instability scoring site set (step S21) and the method based on the (TA) / (AT) dinucleotide specific repeat site set (step S22), or a weighted combination of the quantification results of the above two types of site sets. Regarding the detection platform, in addition to high-throughput sequencing, any in vitro detection platform capable of quantifying the length variation or amplification degree of repeat units at multiple microsatellite sites and calculating continuous quantification values ​​can be used to implement this invention. Regarding the sample type, in addition to tumor tissue samples and peripheral blood cell-free DNA samples, other in vitro samples containing tumor-derived nucleic acids can also be used. Regarding tumor types, in addition to colorectal cancer, endometrial cancer, gastric cancer, and ovarian cancer, this invention is also applicable to other tumor types that can experience microsatellite instability and microsatellite repeat amplification. Regarding thresholds, corresponding thresholds can be calibrated for different quantification methods, different site sets, different tumor types, and different sample types. All of the above-mentioned alternative solutions fall within the technical concept of this invention.

[0076] Supplementary Explanation

[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention. The scope of protection of the present invention is defined by the appended claims.

Claims

1. A method for in vitro labeling of WRN inhibitor response candidates, characterized in that... The method is for non-disease diagnosis and non-treatment purposes, including: Provide nucleic acids from ex vivo tumor samples from the subjects; The microsatellite repeat amplification load (MSBLoad) of the nucleic acid was determined in vitro. The MSBLoad is a continuous score reflecting the degree of amplification of repeat units at multiple microsatellite sites. The quantized value of the microsatellite repeat amplification load is compared with a preset threshold; and When the quantified value of the microsatellite repeat amplification load is greater than or equal to the threshold, the molecular stratification result identifying the ex vivo tumor sample as a candidate for WRN inhibitor response is output; Wherein, the WRN inhibitor is an inhibitor that acts on WRN helicase; the quantification value of the microsatellite repeat amplification load is a continuous variable, which is different from the binary state determination result of microsatellite highly unstable and microsatellite stable.

2. The method according to claim 1, characterized in that... The microsatellite repeat amplification load quantification value is a dinucleotide-specific repeat amplification load quantification value, which is a continuous score calculated for a set of sites consisting of (TA)n repeat sites and / or (AT)n repeat sites, reflecting the degree of amplification of repeat units at that site set.

3. The method according to claim 1, characterized in that... The method is performed on ex vivo tumor samples that have been identified as having high microsatellite instability or positive mismatch repair defects. Among the samples with high microsatellite instability or positive mismatch repair defects, the samples are stratified by the quantification value of the microsatellite repeat amplification load, and those with the quantification value greater than or equal to the threshold are identified as WRN inhibitor response candidates.

4. The method according to claim 1, characterized in that... The microsatellite repeat amplification load quantification value is obtained by statistically analyzing the length variation and / or amplification degree of repeat units at the plurality of microsatellite sites in the high-throughput sequencing reads of the nucleic acid; the plurality of microsatellite sites are a pan-microsatellite instability scoring site set, or a site set composed of (TA)n repeat sites and / or (AT)n repeat sites as described in claim 2.

5. The method according to claim 1, characterized in that... The threshold is determined based on the continuous relationship between the microsatellite repeat amplification load quantification value and WRN dependence in the reference cohort, so that the population with the quantification value greater than or equal to the threshold is enriched with WRN inhibitor response candidates relative to the population with the quantification value less than the threshold.

6. The method according to claim 1, characterized in that... The ex vivo tumor samples were derived from one or more of colorectal cancer, endometrial cancer, gastric cancer, and ovarian cancer.

7. The method according to claim 1, characterized in that... The ex vivo tumor sample is a tumor tissue sample or a peripheral blood cell-free DNA sample, and the nucleic acid is deoxyribonucleic acid.

8. An in vitro detection kit for screening WRN inhibitor response candidates, characterized in that... The kit includes: Nucleic acid primers and / or capture probes, said nucleic acid primers and / or capture probes targeting a plurality of microsatellite loci, are used to obtain nucleic acid data required for in vitro determination of microsatellite repeat amplification load quantification, said microsatellite repeat amplification load quantification is a continuous score reflecting the degree of amplification of repeat units at said plurality of microsatellite loci; and The determination component includes a threshold parameter and a determination description that compares the quantized value of the microsatellite repeat amplification load with a pre-set threshold and identifies WRN inhibitor response candidates when the quantized value is greater than or equal to the threshold. The WRN inhibitor is an inhibitor that acts on WRN helicase.

9. The in vitro diagnostic kit according to claim 8, characterized in that... The kit is a high-throughput sequencing kit, and also includes library preparation reagents and / or sequencing adapters.

10. The in vitro diagnostic kit according to claim 8, characterized in that... The nucleic acid primers and / or capture probes target a set of sites consisting of (TA)n repeat sites and / or (AT)n repeat sites.

11. The in vitro diagnostic kit according to claim 8, characterized in that... The WRN inhibitor is an inhibitor of covalently bound WRN helicase or non-covalently bound WRN helicase; the subject is a subject who has received or intends to receive a WRN inhibitor.

Images