Microsatellite instability detection in cell-free DNA
Patent Information
- Application Number
- JP2024060146
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2019-06-04
- Filing Date
- 2024-04-03
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2039-08-30
AI Technical Summary
Current methods for detecting microsatellite instability (MSI) in cell-free DNA (cfDNA) samples are limited, and the impact of variable tumor shedding on MSI detection has not been adequately evaluated, hindering accurate prognosis and treatment selection in cancer patients.
A method for determining MSI status in cfDNA samples by quantifying repeat lengths, generating site scores, and comparing them to trained thresholds using computer-implemented techniques, which includes probabilistic log-likelihood-based scoring and Akaike Information Criterion (AIC) to distinguish stable and unstable microsatellite loci.
The method achieves high agreement with traditional PCR-based MSI evaluation, providing accurate MSI status determination with sensitivity of at least 94% and specificity of at least 99%, enabling effective disease prognosis and treatment guidance.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[Technical field]
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Patent Application No. 62 / 726,182, filed August 31, 2018, U.S. Provisional Patent Application No. 62 / 823,578, filed March 25, 2019, and U.S. Provisional Patent Application No. 62 / 857,048, filed June 4, 2019, the entire disclosures of which are incorporated herein by reference. [Background technology]
[0002] Repetitive nucleic acid elements are patterns of nucleotides (DNA or RNA) that are present in multiple copies throughout eukaryotic and prokaryotic genomes. Examples of such repetitive elements include microsatellites, short tandem repeats (STRs), and minisatellites, among others. Microsatellites typically contain repeat units of less than 10 base pairs. STRs generally contain repeat units of 2-13 nucleotides that are often repeated hundreds of times within a given stretch of nuclear DNA. STR analysis is a common tool used in forensic analysis. Minisatellites are repetitive elements that typically have repeat units of about 10-60 base pairs.
[0003] Microsatellites, specifically, are highly polymorphic DNA repeat regions. Microsatellite instability (MSI) is a guideline-recommended biomarker used to evaluate prognosis and treatment selection, including checkpoint inhibitors recently approved for the treatment of cancers with high-grade MSI (MSI-H) status. Plasma-based next-generation DNA sequencing (NGS) assays are increasingly being used for comprehensive genomic profiling of cancers, but methods for detecting MSI status from cell-free DNA (cfDNA) data are underdeveloped. In addition, the impact of variable tumor shedding on MSI detection has not previously been evaluated. Summary of the Invention [Means for solving the problem]
[0004] There remains a need for methods and related embodiments that are useful for assessing repetitive element instability, including MSI, in a variety of samples, particularly cfDNA samples.
[0005] The present application discloses a method, computer readable medium, and system that is useful for determining the microsatellite and / or other repeat DNA instability status of cell-free DNA (cfDNA) samples from patients, and helps guide disease prognosis and treatment decisions.Typically, at least a portion of the methods disclosed herein are computer-implemented, and achieve results that have a high degree of concordance with the results obtained using more conventional polymerase chain reaction (PCR)-based MSI evaluation methods.
[0006] Further aspects and advantages of the present disclosure will become readily apparent to those skilled in the art from the following detailed description, in which merely exemplary embodiments of the present disclosure are shown and described. As will be realized, the present disclosure is capable of other and different embodiments, and its several details are capable of modifications in various obvious respects, all without departing from the present disclosure. Accordingly, the drawings and description should be regarded as illustrative in nature, and not as restrictive.
[0007] In one aspect, the disclosure provides a method for determining a repeat nucleic acid instability status of a nucleic acid sample. The method includes (a) quantifying a number of different repeat lengths present at each of a plurality of repeat nucleic acid loci from sequence information to generate a site score for each of the plurality of repeat nucleic acid loci. The sequence information is from a population of repeat nucleic acid loci in the nucleic acid sample. The method also includes (b) calling a given repeat nucleic acid locus as unstable if the site score of the given repeat nucleic acid locus exceeds a site-specific trained threshold for the given repeat nucleic acid locus to generate a repeat nucleic acid instability score that includes a number of unstable repeat nucleic acid loci from the plurality of repeat nucleic acid loci. In addition, the method also includes (c) classifying the repeat nucleic acid instability status of the nucleic acid sample as unstable if the repeat nucleic acid instability score exceeds a population trained threshold for the population of repeat nucleic acid loci in the nucleic acid sample, thereby determining the repeat nucleic acid instability status of the nucleic acid sample.
[0008] In another aspect, the present disclosure provides a method for determining the repetitive DNA instability status of a sample (e.g., a cell-free DNA (cfDNA) sample). The method includes: (a) quantifying a number of different repeat lengths present in each of a plurality of repetitive DNA loci from sequence information to generate a site score for each of the plurality of repetitive DNA loci. The sequence information is from a population of repetitive DNA loci in the sample. The method also includes: (b) comparing the site score of a given repetitive DNA locus for each of the plurality of repetitive DNA loci with a site-specific trained threshold for the given repetitive DNA locus. The method further includes: (c) calling the given repetitive DNA locus as unstable if the site score of the given repetitive DNA locus exceeds the site-specific trained threshold for the given repetitive DNA locus, to generate a repetitive DNA instability score that includes a number of unstable repetitive DNA loci from the plurality of repetitive DNA loci. In addition, the method also includes (d) classifying the repetitive DNA instability status of the sample as unstable if the repetitive DNA instability score exceeds a population-trained threshold for the population of repetitive DNA loci in the sample, thereby determining the repetitive DNA instability status of the sample. The methods disclosed herein are typically at least partially computer-implemented.
[0009] In another aspect, the disclosure provides a method of determining the microsatellite instability (MSI) status of a sample. The method includes: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, the sequence information being from a population of microsatellite loci in the sample. The method also includes (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus. The method further includes (c) calling the given microsatellite locus as unstable if the site score of the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci. In addition, the method also includes (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby determining the MSI status of the sample.
[0010] In another aspect, the disclosure provides a method for determining the microsatellite instability (MSI) status of a sample. The method includes: (a) receiving sequence information from a population of microsatellite loci in the sample; and (b) quantifying a number of different repeat lengths present at each of the plurality of microsatellite loci from the sequence information to generate a site score for each of the plurality of microsatellite loci. The method also includes (c) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus. The method further includes (d) calling the given microsatellite locus as unstable if the site score of the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci. In addition, the method also includes (e) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby determining the MSI status of the sample.
[0011] In another aspect, the present disclosure provides a method for identifying one or more customized therapies for treating a disease in a subject. The method includes: (a) quantifying a number of different repeat lengths present in each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, where the sequence information is from a population of microsatellite loci in a sample. The method also includes (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci with a site-specific trained threshold for the given microsatellite locus. The method further includes (c) calling the given microsatellite locus as unstable if the site score of the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci. The method also includes (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample to identify an unstable sample. In addition, the method also includes (e) comparing the microsatellite instability status of the sample to one or more comparator results indexed with one or more therapies to identify one or more customized therapies for treating the disease in the subject.
[0012] In another aspect, the disclosure provides a method of treating a disease in a subject. The method includes: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, the sequence information being from a population of microsatellite loci in a sample. The method also includes (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus. The method further includes (c) calling a given microsatellite locus as unstable if the site score of the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci. The method also includes (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby identifying an unstable sample. The method also includes (e) comparing the microsatellite instability status of the sample with one or more comparator results indexed with one or more therapies to identify one or more customized therapies for treating the disease in the subject. In addition, the method also includes (f) administering at least one of the identified customized therapies to the subject if the microsatellite instability status of the sample and the comparator result are substantially matched, thereby treating the disease in the subject.
[0013] In another aspect, the disclosure provides a method for treating a disease in a subject. The method includes administering one or more customized therapies to the subject, thereby treating the disease in the subject, where the customized therapies are identified by (a) quantifying a number of different repeat lengths present in each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, the sequence information being from a population of microsatellite loci in a sample. The method also includes (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus. The method further includes (c) calling the given microsatellite locus as unstable if the site score of the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci. The method also includes (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample to identify an unstable sample. The method further includes (e) comparing the microsatellite instability status of the sample with one or more comparator results indexed with one or more therapies. In addition, the method also includes (f) identifying one or more customized therapies for treating the disease in the subject if the microsatellite instability status of the sample and the comparator results are substantially matched.
[0014] In some embodiments, the site score of the plurality of microsatellite loci comprises a likelihood score. In certain of these embodiments, the likelihood score comprises a probabilistic log likelihood-based score that distinguishes the biological signal from the multiple nucleic acid fragments (in some embodiments, cfDNA fragments) from somatic cells in the sample from the noise resulting from post-sample collection artifacts in the sample. In some embodiments, the method comprises determining a probabilistic log likelihood-based score for each microsatellite locus in the sequence information from the sample using at least two parameters, where at least a first parameter comprises an allele frequency and at least a second parameter comprises at least one error mode. Typically, the allele frequency comprises the frequency of nucleic acids comprising different repeat lengths in the sequence information from the sample. In some embodiments, the at least one error mode comprises a random error mode and a strand-specific error mode. In certain embodiments, the site scores for the plurality of microsatellite loci comprise a difference or ratio between (a) a score measuring support of the observed sequence for the null hypothesis that the given microsatellite locus is stable and (b) a score measuring support of the observed sequence for the alternative hypothesis that the given microsatellite locus is unstable. In some embodiments, the site scores for the plurality of microsatellite loci are generated using one or more of a likelihood criterion, a log likelihood criterion, a posterior probability criterion, Akaike Information Criterion (AIC), a Bayesian Information Criterion, and / or the like.
[0015] In some embodiments, the site scores for the plurality of microsatellite loci include a site score based on Akaike's Information Criterion (AIC), which tests for the presence of somatic indels at the plurality of microsatellite loci. In certain of these embodiments, a given AIC-based site score is calculated according to the following formula: AIC=k-log likelihood where k is the number of parameters used in the model. Optionally, the method includes estimating the parameters of the model using maximum likelihood estimation (MLE). In some of these embodiments, the method includes determining the MLE using the Nelder-Mead algorithm. In certain embodiments, the method includes determining the MLE using the following formula: AIC 0 = k-log(Pr(obs|β,γ)) (In the formula, AIC 0 is the null hypothesis, k is the number of parameters used in the model, Pr is the probability, obs contains the repeat length of the observed sequencing reads covering a given microsatellite locus, β is at least one strand-specific error parameter, and γ is at least one random error parameter). Calculating a null hypothesis score for the model (e.g., a score that measures the support of the observed sequences for the null hypothesis that a given microsatellite locus is stable) using the following formula: In certain embodiments, obs is a number of observed sequencing reads that cover a given microsatellite locus. In some of these embodiments, the method comprises: AIC min =min α (k-log(Pr(obs|β,γ,α)) (In the formula, AIC min is the alternative hypothesis, and min α is the minimizing effect over all values of α, k is the number of parameters used in the model, Pr is the probability, obs includes the repeat length of the observed sequencing reads covering a given microsatellite locus, β is at least one strand-specific error parameter, γ is at least one random error parameter, and α is an allele frequency of at least 1, where α is one or more i is a vector of allele frequencies whose sum is equal to 1. In some embodiments, obs is a number of observed sequencing reads covering a given microsatellite locus. In certain of these embodiments, the method includes calculating the model change (i.e., ΔAIC) to determine the site score using the following formula: ΔAIC=AIC 0 -AIC min In some of these embodiments, γ comprises: (a) a read-level error rate where the observed microsatellite length in the sequencing reads is one repeat unit longer than the expected microsatellite length for the strand of the original nucleic acid molecule; and / or (b) a read-level error rate where the observed microsatellite length in the sequencing reads is one repeat unit shorter than the expected microsatellite length for the strand of the original nucleic acid molecule. In certain of these embodiments, β comprises: (a) a strand-level error rate where the expected microsatellite length of the sense strand is one repeat unit longer than the expected microsatellite length of the nucleic acid-derived molecule; (b) a strand-level error rate where the expected microsatellite length of the antisense strand is one repeat unit longer than the expected microsatellite length of the nucleic acid-derived molecule; (c) a strand-level error rate where the expected microsatellite length of the sense strand is one repeat unit shorter than the expected microsatellite length of the nucleic acid-derived molecule; and / or (d) a strand-level error rate where the expected microsatellite length of the antisense strand is one repeat unit shorter than the expected microsatellite length of the nucleic acid-derived molecule. Typically, the method comprises calling a given microsatellite locus as unstable if the site score for the given microsatellite locus statistically exceeds a site-specific trained threshold for the given microsatellite locus.
[0016] In some embodiments, the site score based on AIC is calculated using the following formula: AIC=2(k-log likelihood) where k is the number of parameters used in the model. In these embodiments, the AIC 0 and AIC min is calculated using the formula above.
[0017] For clarity, in embodiments where the AIC-based score is determined using the formula AIC=2(k-log likelihood), the site-specific threshold used to classify a site as unstable will be twice the site-specific threshold used in the previous embodiment where the AIC-based score is determined using the formula AIC=k-log likelihood.
[0018] Typically, the mutant allele fraction (MAF) of the sample (e.g., cfDNA sample) is estimated. In some of these embodiments, the tumor fraction of the sample (e.g., cfDNA sample) is estimated. In certain embodiments, the tumor fraction comprises the maximum mutant allele fraction (MAF) of all somatic mutations identified in the nucleic acid in the sample (e.g., cfDNA sample). In some embodiments, the tumor fraction is less than about 0.05%, about 0.1%, about 0.2%, about 0.5%, about 1%, about 2%, about 3%, about 4%, about 5%, about 6%, about 7%, about 8%, about 9%, about 10%, about 11%, about 12%, about 13%, about 14%, or about 15% of all nucleic acids in the sample (e.g., cfDNA sample). In some embodiments, the plurality of microsatellite loci comprises all of a population of microsatellite loci, while in other embodiments, the plurality of microsatellite loci comprises a subset of a population of microsatellite loci. In certain embodiments, the method includes determining a site-specific trained threshold and / or a population trained threshold from sequence information from a population of microsatellite loci in one or more training DNA samples. In some of these embodiments, the training DNA sample includes a non-tumor cfDNA training sample and / or DNA from one or more tumor types.
[0019] In some embodiments, the method has a sensitivity of at least about 94% at a limit of detection (LOD) of about 0.1-0.4% tumor fraction of nucleic acid in the sample. In some embodiments, the method has an analytical specificity of at least about 99% for non-tumor DNA in the sample. In certain embodiments, the determined MSI status of the sample comprises at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% concordance with the corresponding MSI status of the sample determined using PCR-based MSI assessment techniques over a tumor fraction range of about 1% to about 15%. In some of these embodiments, the concordance is 100%. In some embodiments, the method comprises classifying the MSI status of the sample as MSI-high (MSI-H) if the microsatellite instability score is a number of unstable microsatellite loci greater than about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 15, about 20, about 30, about 40, about 50 or more from the plurality of microsatellite loci. In certain embodiments, the method comprises classifying the MSI status of the sample as MSI-high (MSI-H) if the number of unstable microsatellite loci constitutes about 0.1%, about 1%, about 2%, about 3%, about 4%, about 5%, about 10%, about 15%, about 20% or about 25% of the plurality of microsatellite loci. In some embodiments, the number of different repeat lengths comprises a frequency of each different repeat length present in each of the plurality of microsatellite loci.
[0020] In various embodiments, the disclosure includes a method of selecting a customized therapy for treating a disease in a subject and / or a method of treating a disease in a subject. In some of these embodiments, the disease is biliary tract cancer, bladder cancer, transitional cell carcinoma, urothelial carcinoma, brain cancer, glioma, astrocytoma, breast cancer, metaplastic carcinoma, cervical cancer, cervical squamous cell carcinoma, rectal cancer, colorectal cancer, colon cancer, hereditary nonpolyposis colorectal cancer, colorectal adenocarcinoma, gastrointestinal stromal tumor (GIST), endometrial cancer, endometrial stromal sarcoma, esophageal cancer, esophageal squamous cell carcinoma, esophageal adenocarcinoma, ocular melanoma, uveal melanoma, gallbladder cancer, gallbladder adenocarcinoma, renal cell carcinoma, renal clear cell carcinoma, transitional cell carcinoma, urothelial carcinoma, Wilms' tumor, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML). , chronic lymphocytic (CLL), chronic myelogenous (CML), chronic myelomonocytic (CMML), liver cancer, hepatoma, hepatocellular carcinoma, cholangiocarcinoma, hepatoblastoma, lung cancer, non-small cell lung cancer (NSCLC), mesothelioma, B cell lymphoma, non-Hodgkin's lymphoma, diffuse large B cell lymphoma, mantle cell lymphoma, T cell lymphoma, non-Hodgkin's lymphoma, precursor T lymphoblastic lymphoma / leukemia, peripheral T cell lymphoma, multiple myeloma, nasopharyngeal carcinoma (NPC), neuroblastoma, oropharyngeal carcinoma, oral squamous cell carcinoma, osteosarcoma, ovarian cancer, pancreatic cancer, pancreatic ductal adenocarcinoma, pseudopapillary neoplasm neoplasm), acinic cell carcinoma, prostate cancer, prostate adenocarcinoma, skin cancer, melanoma, malignant melanoma, cutaneous melanoma, small intestine cancer, gastric cancer, gastric carcinoma, gastrointestinal stromal tumor (GIST), uterine cancer, and uterine sarcoma.
[0021] In some embodiments, the therapy includes at least one immunotherapy (e.g., a checkpoint inhibitor antibody, an autologous cytotoxic T cell, a personalized cancer vaccine, etc.). In certain embodiments, for example, the immunotherapy includes an antibody against PD-1, PD-2, PD-L1, PD-L2, CTLA-4, OX40, B7.1, B7He, LAG3, CD137, KIR, CCR5, CD27, CD40, or CD47. In some embodiments, the immunotherapy includes administration of a pro-inflammatory cytokine against at least one tumor type. Optionally, the immunotherapy includes administration of T cells against at least one tumor type.
[0022] In some embodiments, the method includes obtaining a sample from a subject. Essentially any sample type may be utilized as desired. In certain embodiments, for example, the sample is tissue, blood, plasma, serum, sputum, urine, semen, vaginal fluid, feces, synovial fluid, cerebrospinal fluid, saliva, and / or the like. Typically, the subject is a mammalian subject (e.g., a human subject). In some embodiments, the sample is blood. In some embodiments, the sample is plasma. In some embodiments, the sample is serum. In some embodiments, the sample comprises cell-free DNA (i.e., a cfDNA sample). In some embodiments, the cfDNA sample comprises circulating tumor nucleic acid.
[0023] In certain embodiments, the method includes receiving sequence information generated from a sample, the sequence information including sequencing reads from a population of microsatellite loci in the sample. In some embodiments, the method includes amplifying one or more segments of nucleic acid in the sample to generate at least one amplified nucleic acid. In certain embodiments, the method includes sequencing nucleic acid from the sample to generate sequence information. In some embodiments, the sample can be a cfDNA sample. In these embodiments, the sequence information includes cfDNA sequencing reads from a population of microsatellite loci in the cfDNA sample. In some embodiments, the sequence information is obtained from a targeted segment of nucleic acid in the sample, the targeted segment being obtained by selectively enriching one or more regions from the nucleic acid in the sample prior to sequencing. In some of these embodiments, the method includes amplifying the obtained targeted segment prior to sequencing. In these embodiments, the method typically includes attaching one or more adapters including molecular barcodes to the nucleic acid prior to amplification. In some embodiments, the method includes attaching one or more sample indexes by amplification prior to sequencing. Essentially any nucleic acid sequencing technique can be used or adapted for use in carrying out the method disclosed herein as required.For example, sequencing can be selected from targeted sequencing, intron sequencing, exome sequencing, whole genome sequencing, and / or the like as required.In some embodiments, sequencing is targeted sequencing.In some embodiments, the method comprises sequencing at least about 50, about 100, about 150, about 200, about 250, about 500, about 750, about 1,000, about 1,500, about 2,000 or more targeted genomic regions in the nucleic acid of sample to generate sequence information.
[0024] In another aspect, the disclosure, when executed by at least one electronic processor, includes at least the steps of (a) receiving sequence information from a population of microsatellite loci in a sample; (b) quantifying from the sequence information a number of different repeat lengths present at each of a plurality of microsatellite loci to generate a site score for each of the plurality of microsatellite loci; (c) comparing the site score for a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; and (d) comparing the site score for a given microsatellite locus to a site-specific trained threshold for the given microsatellite locus. The system includes a computer-readable medium containing non-transient computer-executable instructions or a controller that can access such a computer-readable medium to perform the steps of: (a) calling a given microsatellite locus as unstable if a trained threshold is exceeded, generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; and (b) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population trained threshold for a population of microsatellite loci in the sample, thereby determining the MSI status of the sample.
[0025] In some embodiments, the system includes a nucleic acid sequencer operably connected to the controller, the nucleic acid sequencer configured to provide sequence information from a population of microsatellite loci in the sample. In some of these embodiments, the nucleic acid sequencer is configured to perform pyrosequencing, single molecule sequencing, nanopore sequencing, semiconductor sequencing, sequencing by synthesis, sequencing by ligation, or sequencing by hybridization with the nucleic acid to generate sequencing reads. In certain embodiments, the system includes a sample preparation component operably connected to the controller, the sample preparation component configured to prepare a sample (in some cases a cfDNA sample) to be sequenced by the nucleic acid sequencer. In some of these embodiments, the sample preparation component is configured to selectively enrich a region from the nucleic acid in the sample. In certain embodiments, the sample preparation component is configured to bind one or more adapters comprising molecular barcodes to the nucleic acid. In some embodiments, the system includes a nucleic acid amplification component operably connected to the controller, the nucleic acid amplification component configured to amplify DNA (in some cases cfDNA). In certain of these embodiments, the nucleic acid amplification component is configured to selectively amplify a region from a nucleic acid in a sample.
[0026] In certain embodiments, the system includes a material transfer component operably connected to the controller, the material transfer component configured to transfer one or more materials between the nucleic acid sequencer and the sample preparation component. In some embodiments, the system includes a database operably connected to the controller, the database including one or more comparator results indexed with one or more therapies, and the electronic processor further performs at least (f) a comparison of the microsatellite locus status of the sample to the one or more comparator results, wherein a substantial match between the microsatellite instability score and the comparator result indicates a predicted response to the subject's therapy.
[0027] In yet another aspect, the disclosure provides a method, when executed by at least one electronic processor, comprising at least the steps of: (a) receiving sequence information from a population of microsatellite loci in a sample; (b) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from the sequence information to generate a site score for each of the plurality of microsatellite loci; (c) comparing the site score for a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; (d) quantifying the site score for the given microsatellite locus. The present invention provides a computer-readable medium comprising non-transient computer-executable instructions for performing the steps of: (a) calling a given microsatellite locus as unstable if the score exceeds a site-specific trained threshold for the given microsatellite locus, thereby generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; and (b) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population trained threshold for a population of microsatellite loci in the sample, thereby determining the MSI status of the sample.
[0028] The system and computer-readable medium disclosed herein include various embodiments. In some embodiments, for example, the site score of a plurality of microsatellite loci includes a likelihood score. In certain of these embodiments, the likelihood score includes a probabilistic log likelihood-based score that distinguishes the biological signal from a large number of nucleic acid fragments (in some embodiments, cfDNA fragments) from somatic cells in the sample from the noise generated from post-sample collection artifacts in the sample. The probabilistic log likelihood-based score for each microsatellite locus in the sequence information from the sample is typically determined using at least two parameters, where at least a first parameter includes an allele frequency and at least a second parameter includes at least one error mode. The allele frequency includes the frequency of nucleic acids that include different repeat lengths in the sequence information from the sample. The at least one error mode typically includes a random error mode and a strand-specific error mode. In some embodiments, the site scores for the plurality of microsatellite loci comprise a difference or ratio between (a) a score measuring support of the observed sequence for the null hypothesis that the given microsatellite locus is stable and (b) a score measuring support of the observed sequence for the alternative hypothesis that the given microsatellite locus is unstable. In some embodiments, the site scores for the plurality of microsatellite loci are generated using one or more statistical model selection criteria, such as a likelihood criterion, a log likelihood criterion, a posterior probability criterion, Akaike Information Criterion (AIC), Bayesian Information Criterion, and / or the like.
[0029] In some embodiments of the system or computer readable medium, the site scores for the plurality of microsatellite loci include site scores based on Akaike's Information Criterion (AIC), which tests for the presence of somatic indels at the plurality of microsatellite loci. In certain embodiments, for example, a given AIC-based site score is calculated using the following formula: AIC=k-log likelihood where k is the number of parameters used in the model. Optionally, the parameters of the model are estimated using maximum likelihood estimation (MLE). In some of these embodiments, the MLE is determined using the Nelder-Mead algorithm. In certain embodiments, the null hypothesis score of the model is calculated using the following formula: AIC 0 = k-log(Pr(obs|β,γ)) (In the formula, AIC 0 is the null hypothesis, k is the number of parameters used in the model, Pr is the probability, obs contains the repeat length of the observed sequencing reads covering a given microsatellite locus, β is at least one strand-specific error parameter, and γ is at least one random error parameter). In some embodiments, the alternative hypothesis score of the model is calculated using the following formula: AIC min =min α (k-log(Pr(obs|β,γ,α)) (In the formula, AIC min is the alternative hypothesis, and min α is the minimizing effect over all values of α, k is the number of parameters used in the model, Pr is the probability, obs includes the repeat length of the observed sequencing reads covering a given microsatellite locus, β is at least one strand-specific error parameter, γ is at least one random error parameter, and α is an allele frequency of at least 1, where α is one or more i is a vector of allele frequencies whose sum is equal to 1. In these embodiments, to determine the site score, the site score is calculated using the following formula: ΔAIC=AIC 0 -AIC min Changes in the model are typically detected using
[0030] In some embodiments, γ comprises: (a) the read-level error rate where the observed microsatellite length in the sequencing read is one repeat unit longer than the expected microsatellite length for the strand of the original nucleic acid molecule; and / or (b) the read-level error rate where the observed microsatellite length in the sequencing read is one repeat unit shorter than the expected microsatellite length for the strand of the original nucleic acid molecule. In certain embodiments, β comprises: (a) a strand-level error rate where the expected microsatellite length of the sense strand is one repeat unit longer than the expected microsatellite length of the nucleic acid derived molecule; (b) a strand-level error rate where the expected microsatellite length of the antisense strand is one repeat unit longer than the expected microsatellite length of the nucleic acid derived molecule; (c) a strand-level error rate where the expected microsatellite length of the sense strand is one repeat unit shorter than the expected microsatellite length of the nucleic acid derived molecule; and / or (d) a strand-level error rate where the expected microsatellite length of the antisense strand is one repeat unit shorter than the expected microsatellite length of the nucleic acid derived molecule.
[0031] In certain embodiments of the system or computer readable medium, a given microsatellite locus is called unstable if the site score of the given microsatellite locus statistically exceeds the site-specific trained threshold of the given microsatellite locus.Typically, the tumor fraction is estimated to include the maximum mutant allele fraction (MAF) of all somatic mutations identified in the nucleic acid in the sample.In certain embodiments, the tumor fraction is less than about 0.05%, about 0.1%, about 0.2%, about 0.5%, about 1%, about 2%, about 3%, about 4%, about 5%, about 6%, about 7%, about 8%, about 9%, about 10%, about 11%, about 12%, about 13%, about 14% or about 15% of all nucleic acids in the sample. In some embodiments, the plurality of microsatellite loci comprises all of a population of microsatellite loci, while in other embodiments, the plurality of microsatellite loci comprises a subset of a population of microsatellite loci. In certain embodiments, the site-specific trained threshold and / or population trained threshold are determined from sequence information from a population of microsatellite loci in one or more training DNA samples. Optionally, the MSI status of the sample is classified as high-MSI (MSI-H) if the microsatellite instability score is greater than about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 15, about 16, about 17, about 18, about 19, about 20, about 30, about 40, about 50 or more unstable microsatellite loci from the plurality of microsatellite loci. In some embodiments, the MSI status of a sample is classified as MSI-high (MSI-H) if the number of unstable microsatellite loci constitutes about 0.1%, about 1%, about 2%, about 3%, about 4%, about 5%, about 10%, about 15%, about 20% or about 25% of the plurality of microsatellite loci.
[0032] In yet another aspect, the disclosure provides a system including a communications interface that obtains sequencing information from one or more nucleic acids in a sample from a subject via a communications network; and a computer in communication with the communications interface, the computer comprising at least one computer processor and a method, when executed by the at least one computer processor, for performing the steps of: (a) receiving sequence information from a population of microsatellite loci in the sample; (b) quantifying from the sequence information a number of different repeat lengths present at each of a plurality of microsatellite loci to generate a site score for each of the plurality of microsatellite loci; (c) determining a site score for a given microsatellite locus for each of the plurality of microsatellite loci. (d) calling the given microsatellite locus as unstable if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; and (e) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population trained threshold for a population of microsatellite loci in the sample, thereby determining the MSI status of the sample.
[0033] In some embodiments, the sequence information is provided by a nucleic acid sequencer. Typically, the nucleic acid sequencer performs pyrosequencing, single molecule sequencing, nanopore sequencing, semiconductor sequencing, sequencing by synthesis, sequencing by ligation, sequencing by hybridization, and / or another sequencing technique with the nucleic acid to generate sequencing reads. In some embodiments, the nucleic acid sequencer generates sequencing reads using a clonal single molecule array derived from a sequencing library. In certain embodiments, the nucleic acid sequencer includes a chip having an array of microwells to sequence the sequencing library to generate sequencing reads.
[0034] The computer-readable medium of the system disclosed herein typically includes a memory, a hard drive, or a computer server. In some embodiments, the communication network includes one or more servers capable of distributed computing. In some embodiments, the distributed computing is cloud computing. In some embodiments, the computer is located on a computer server that is remotely located from the nucleic acid sequencer. In some embodiments, the system disclosed herein includes an electronic display in communication with the computer via a network, the electronic display including a user interface for displaying results upon execution of (i)-(iv). In some of these embodiments, the user interface is a graphical user interface (GUI) or a web-based user interface. In some embodiments, the electronic display is mounted on a personal computer. In certain embodiments, the electronic display is mounted on a computer capable of connecting to the Internet. In some of these embodiments, the computer capable of connecting to the Internet is located remotely from the computer. Typically, the computer-readable medium includes a memory, a hard drive, or a computer server. In some embodiments, the communication network includes a telecommunications network, the Internet, an extranet, or an intranet.
[0035] In some embodiments, the results of the systems and methods disclosed herein are used as input data to generate a report. The report may be in paper or electronic format. For example, the MSI score and / or MSI status obtained by the methods and systems disclosed herein may be displayed directly in such a report. Alternatively or additionally, diagnostic information or one or more customized therapies based on the MSI status may be included in the report. In some embodiments, the report is communicated to a subject (e.g., a patient) or a healthcare provider.
[0036] In some embodiments, the method, system, or computer readable medium further comprises classifying the repeat nucleic acid instability status of the nucleic acid sample as stable if the repeat nucleic acid instability score is at or below a population-trained threshold for the population of repeat nucleic acid loci in the nucleic acid sample.
[0037] In some embodiments, the method, system, or computer readable medium further comprises classifying the repetitive DNA instability status of the sample as stable if the repetitive DNA instability score is at or below a population-trained threshold for the population of repetitive DNA loci in the sample.
[0038] In some embodiments, the method, system, or computer readable medium further comprises classifying the microsatellite instability status of the sample as stable if the microsatellite instability score is at or below a population-trained threshold for the population of microsatellite loci in the sample.
[0039] Various steps of the methods disclosed herein, or steps performed by the systems disclosed herein, may be performed at the same or different times, in the same or different geographic locations, e.g., countries, and / or by the same or different persons.
[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate certain embodiments and, together with the present written description, serve to explain certain principles of the methods, computer-readable media and systems disclosed herein. The description provided herein is better understood when read in conjunction with the accompanying drawings, which are included by way of example and not by way of limitation. It should be understood that the same components are identified by the same reference numerals throughout the drawings unless otherwise indicated by the context. It should also be understood that some or all of the figures may be schematic for purposes of illustration and do not necessarily depict the actual relative size and position of the elements shown. [Brief description of the drawings]
[0041] [Figure 1] FIG. 1 is a flow chart that generally depicts exemplary method steps for determining microsatellite instability (MSI) status according to some embodiments of the present invention.
[0042] [Diagram 2] FIG. 2 is a schematic diagram of an exemplary system suitable for use with certain embodiments of the present invention.
[0043] [Diagram 3] FIG. 3 is a plot of limit of detection (LoD) (probability of detection (y-axis); mutant allele fraction (MAF) (x-axis)) for simulated samples.
[0044] [Figure 4] 4A (MSI score (y-axis); flow cell (x-axis)) and 4B (MSI score (y-axis); reference sample (x-axis)) are plots of data from the repeatability and reproducibility analysis.
[0045] [Diagram 5] Figures 5A (somatic max-MAF (y-axis); MSI score (x-axis)) and 5B (somatic max-MAF (y-axis); MSI score (x-axis)) are plots showing that tumor proportion does not correlate with MSI score.
[0046] [Figure 6]Figures 6A and 6B are plots showing technical characteristics of microsatellite detection. Figure 6A is a plot showing hierarchical clustering of Akaike Information Criterion scores for 99 candidate microsatellite loci from cfDNA sequencing results from 84 healthy donors. Loci with poor unique molecular coverage are shown in black, while loci with excess technical artifacts are shown in dark grey. Robust but consistent measurements of microsatellite repeat length, defining informative sites, are shown in light grey. Arrows indicate the three Bethesda loci included in this study. Figure 6B is a plot showing the observed error rate reduction associated with each component of digital sequencing.
[0047] [Figure 7] Figures 7A-7C are plots showing analytical validation of ctDNA MSI detection. Observed MSI detection rates were plotted by titration level (grey dots), and probit regression was used to determine the 95% detection limits for cfDNA inputs of 5 ng (Figure 7A) and 30 ng (Figure 7B). Figure 7C is a plot showing sample-level MSI scores for 499 independent replicates of two microsatellite stable (MSS) and two MSI-H devices run across 499 separate sequencing runs. The dashed line indicates the sample-level threshold for MSI detection.
[0048] [Figure 8] Figure 8 is a plot showing a precision study using design samples at three input levels (5, 10 and 30 ng) processed in triplicate within and between trials. Each greyscale shade represents a different trial.
[0049] [Figure 9] Figure 9 is a plot showing the tumor types represented in the clinical validation cohort. In particular, only tumor types with at least 5 representative samples are shown. All others (n=25 distinct tumor types) are grouped into the category "other."
[0050] [Figure 10] Figures 10A-10C show ctDNA MSI status and tissue testing concordance data. Figure 10A is a plot showing sample-level MSI scores for 1137 cfDNA samples grouped by tissue testing result and observed tumor fraction. The dashed line indicates the sample-level threshold for MSI detection. Figure 10B is a plot showing concordance results grouped by tissue testing methodology. Figure 10C is a table showing descriptive statistics for the evaluable unique patient cohort.
[0051] [Figure 11] Figures 11A-11C are plots showing the ctDNA MSI landscape across 28,459 clinical samples. Figure 11A is a plot showing that the positive axis reports ctDNA MSI prevalence across the 16 most prevalent tumor types in the sample set. The negative axis reports tissue MSI prevalence across the same based on Hause et. al (52). The total number of samples is reported for each with the number of MSI-H samples in parentheses. Figure 11B is a plot showing the sample-level MSI score by tumor type for tumor types with ≧5 MSI-H samples. The dashed line indicates the sample-level threshold for MSI detection. Figure 11C is a plot showing the frequency of individual microsatellite sites contributing to MSI-H samples by tumor type for tumor types with ≧5 MSI-H samples. UCEC, endometrial cancer of the uterus; STAD, gastric adenocarcinoma; COAD, colon adenocarcinoma; PRAD, prostate adenocarcinoma; COUP, cancer of unknown primary origin; BLCA, bladder cancer; CHCA, cholangiocellular carcinoma; HNSC, head and neck squamous cell carcinoma; LUSC, lung squamous cell carcinoma; BRST, breast cancer; Ovarian cancer; LUAD, lung adenocarcinoma.
[0052] [Figure 12]Figures 12A and 12B are plots showing tumor mutation burden by MSI status. The number of single nucleotide variants (SNVs) (Figure 12A) and indels (Figure 12B) detected per sample broadly stratified by MSI status across 278 MSI-H and 28,181 MSS samples.
[0053] [Figure 13] Figures 13A-13E show clinical outcome data for immune checkpoint blockade (ICB) therapy in ctDNA MSI-H patients. Figure 13A shows swimmer plots of pembrolizumab therapy duration in months. Baseline (Figures 13B and 13C) and post-treatment (Figures 13D and 13E), CT (Figures 13B and 13D) and gastroscopy (Figures 13C and 13E) for patient 2. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0054] definition In order that this disclosure may be more readily understood, certain terms are first defined below. Additional definitions for the following terms, as well as other terms, may be set forth throughout this specification. In the event that a definition of a term set forth below conflicts with a definition in an application or patent incorporated by reference, the definition set forth in this application should be used to understand the meaning of the term.
[0055] As used in this specification and the appended claims, the singular forms "a," "an," "an "An" and "the" are used interchangeably unless the context clearly dictates otherwise. A reference to "a method" includes one or more The present invention includes methods, methods, and / or steps of the type described herein and / or that will become apparent to those of skill in the art upon reading this disclosure or otherwise.
[0056] It should also be understood that the terminology used herein is for the purpose of describing specific embodiments only and is not intended to be limiting.Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this disclosure belongs.In describing and claiming the method, computer readable medium, and system, the following terminology and their grammatical variants will be used in accordance with the definitions set forth below.
[0057] About: As used herein, "about" or "approximately" when applied to one or more values or elements of interest refers to a value or element that is similar to the stated reference value or element. In certain embodiments, the term "about" or "approximately" refers to a range of values or elements that fall within 25%, 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1% or less in either direction (greater or less) of the stated reference value or element, unless otherwise stated or otherwise clear from the context (except when such number exceeds 100% of possible values or elements).
[0058] Adapter: As used herein, "adapter" refers to a short nucleic acid (e.g., less than about 500 nucleotides, less than about 100 nucleotides, or less than about 50 nucleotides in length), typically at least partially double-stranded, used to ligate to either or both of a given sample nucleic acid molecule. The adapter can include a nucleic acid primer binding site that allows amplification of the nucleic acid molecule flanked by the adapter at both ends, and / or a sequencing primer binding site that includes a primer binding site for sequencing applications, such as various next generation sequencing (NGS) applications. The adapter can also include a binding site for a capture probe, such as an oligonucleotide bound to a flow cell support or the like. The adapter can also include a nucleic acid tag, as described herein. The nucleic acid tag is typically located relative to the amplification primer and sequencing primer binding sites such that the nucleic acid tag is included in the amplicon and sequence reads of a given nucleic acid molecule. The same or different adapters can be ligated to each end of the nucleic acid molecule. In some embodiments, adapters of the same sequence, except for the different nucleic acid tags, are ligated to each end of the nucleic acid molecule. In some embodiments, the adaptor is a Y-shaped adaptor with one end blunted or tailed as described herein for binding to a nucleic acid molecule that is also blunted or tailed with one or more complementary nucleotides. In yet another exemplary embodiment, the adaptor is a bell-shaped adaptor with a blunt or tailed end for binding to a nucleic acid molecule to be analyzed. Other examples of adaptors include T-tailed adaptors and C-tailed adaptors.
[0059] Administer: As used herein, "administering" or "administering" a therapeutic agent (e.g., an immunological therapeutic agent) to a subject means giving, applying, or contacting a composition with the subject. Administration can be accomplished by any of a number of routes, including, for example, topical, oral, subcutaneous, intramuscular, intraperitoneal, intravenous, intrathecal, and intradermal.
[0060] Akaike Information Criterion: As used herein, "Akaike Information Criterion" or "AIC" refers to a criterion for selecting a statistical model from among a finite set of models, and includes a penalty term for the number of parameters in the model. In some embodiments, the model with the lowest AIC is selected.
[0061] Allele frequency: As used herein, "allele frequency" refers to the relative frequency of an allele at a particular locus in a population or in a given subject. Allele frequency is typically expressed as a proportion or percentage.
[0062] Amplify: As used herein, "amplify" or "amplification" in the context of nucleic acids refers to the production of multiple copies of a polynucleotide or a portion of a polynucleotide, typically starting from a small amount of the polynucleotide (e.g., a single polynucleotide molecule), where an amplification product or amplicon is generally detectable. Amplification of polynucleotides encompasses a variety of chemical and enzymatic processes.
[0063] Barcode: As used herein, "barcode" or "molecular barcode" in the context of nucleic acids refers to a nucleic acid molecule that contains a sequence that can serve as a molecular identifier. For example, individual "barcode" sequences are typically added to each DNA fragment during next generation sequencing (NGS) library preparation, thus allowing each sequencing read to be identified and sorted prior to final data analysis.
[0064] Cancer type: As used herein, "cancer," "cancer type," or "tumor type" refers to a type or subtype of cancer, as defined, for example, by histopathology. A cancer type can be defined by any conventional criteria, for example, the type or subtype of cancer in a given tissue (e.g., hematological cancer, central nervous system (CNS), brain cancer, lung cancer (small cell and non-small cell), skin cancer, nasal cancer, throat cancer, liver cancer, bone cancer, lymphoma, pancreatic cancer, colon cancer, rectal cancer, thyroid cancer, bladder cancer, kidney cancer, oral cancer, stomach cancer, breast cancer, prostate cancer, ovarian cancer, lung cancer, intestinal cancer, soft tissue cancer, neuroendocrine cancer, gastroesophageal cancer, head and neck cancer, gynecological cancer, colorectal cancer, urothelial cancer, etc. Cancers can be defined based on their appearance in tumors (e.g., tumors of unknown primary origin, solid tumors, heterogeneous tumors, homogeneous tumors), unknown primary, and the like, and / or based on the same cell lineage (e.g., carcinoma, sarcoma, lymphoma, cholangiocarcinoma, leukemia, mesothelioma, melanoma, or glioblastoma) and / or cancers that exhibit cancer markers such as Her2, CA15-3, CA19-9, CA-125, CEA, AFP, PSA, HCG, hormone receptors, and NMP-22. Cancers can also be classified by stage (e.g., stage 1, 2, 3, or 4) and whether they are primary or secondary.
[0065] Cell-free nucleic acid: As used herein, "cell-free nucleic acid" refers to nucleic acid that is not contained in or otherwise bound to a cell, or in some embodiments, nucleic acid that naturally remains in a sample after removal of intact cells. Cell-free nucleic acid can include, for example, all unencapsulated nucleic acid provided by a subject's bodily fluids (e.g., blood, plasma, serum, urine, cerebrospinal fluid (CSF), etc.). Cell-free nucleic acid includes DNA (cfDNA), RNA (cfRNA), and hybrids thereof, including genomic DNA, mitochondrial DNA, circulating DNA, siRNA, miRNA, circulating RNA (cRNA), tRNA, rRNA, small nucleolar RNA (snoRNA), Piwi-binding RNA (piRNA), long non-coding RNA (long ncRNA), and / or any fragments thereof. Cell-free nucleic acid can be double-stranded, single-stranded, or hybrids thereof. Cell-free nucleic acid can be released into bodily fluids by secretion or cell death processes, such as cell necrosis, apoptosis, or the like. Some cell-free nucleic acids are released from cancer cells into body fluids, for example, circulating tumor DNA (ctDNA). Others are released from healthy cells. ctDNA can be unencapsulated tumor-derived fragmented DNA. Another example of cell-free nucleic acid is fetal DNA that circulates freely in maternal bloodstream, also called cell-free fetal DNA (cffDNA). Cell-free nucleic acid can have one or more epigenetic modifications, for example, cell-free nucleic acid can be acetylated, 5-methylated, ubiquitinated, phosphorylated, sumoylated, ribosylated and / or citrullinated.
[0066] Comparator result: As used herein, "comparator result" refers to a result or set of results that can be compared to a given test sample or test result to identify one or more likely characteristics of the test sample or result, and / or one or more possible prognostic results, and / or one or more customized therapies for the subject from which the test sample was taken or otherwise derived. Comparator results are typically obtained from a set of reference samples (e.g., from a subject with the same disease or cancer type as the test subject, and / or from a subject that is undergoing or has undergone the same therapy as the test subject). In certain embodiments, for example, the microsatellite instability status of a sample (e.g., an unstable cfDNA sample) is compared to the comparator result to identify a substantial match between the microsatellite instability status of the cfDNA test sample and the microsatellite instability status determined for the set of reference samples. The microsatellite instability score determined for the set of reference samples is typically indexed with one or more customized therapies. Thus, if a substantial match is identified, a corresponding customized therapy is also thereby identified as a potential treatment route for the subject from whom the test sample was taken.
[0067] Control Sample: As used herein, a "control sample" or "control DNA sample" refers to a sample of known composition and / or having known characteristics and / or parameters (e.g., known tumor fraction, known coverage, known microsatellite instability score, and / or the like) that is analyzed along with or compared to a test sample to assess the accuracy of an analytical procedure.
[0068] Coverage: As used herein, "coverage" refers to the number of nucleic acid molecules occupying a particular base position.
[0069] Customized Therapy: As used herein, "customized therapy" refers to a therapy with a desired treatment outcome for a subject or population of subjects selected based on given criteria, e.g., having a given microsatellite instability status or within a defined range of microsatellite instability scores.
[0070] Deoxyribonucleic acid or ribonucleic acid: As used herein, "deoxyribonucleic acid" or "DNA" refers to natural or modified nucleotides that have a hydrogen group at the 2' position of the sugar moiety. DNA typically comprises a chain of nucleotides that contains four types of nucleotide bases: adenine (A), thymine (T), cytosine (C) and guanine (G). As used herein, "ribonucleic acid" or "RNA" refers to natural or modified nucleotides that have a hydroxyl group at the 2' position of the sugar moiety. RNA typically comprises a chain of nucleotides that contains four types of nucleotides: A, uracil (U), G and C. As used herein, the term "nucleotide" refers to natural or modified nucleotides. Certain pairs of nucleotides specifically bind to each other in a complementary manner (this is called complementary base pairing). In DNA, adenine (A) pairs with thymine (T) and cytosine (C) pairs with guanine (G). In RNA, adenine (A) pairs with uracil (U) and cytosine (C) pairs with guanine (G). When a first nucleic acid strand binds to a second nucleic acid strand that is made up of nucleotides that are complementary to the nucleotides in the first strand, the two strands combine to form a duplex. As used herein, "nucleic acid sequencing data," "nucleic acid sequencing information," "sequence information," "nucleic acid sequence," "nucleotide sequence," "genomic sequence," "gene sequence," or "fragment sequence," or "nucleic acid sequencing read" refers to any information or data that indicates the order and identity of nucleotide bases (e.g., adenine, guanine, cytosine, and thymine or uracil) in a molecule (e.g., a whole genome, whole transcriptome, exome, oligonucleotide, polynucleotide, or fragment) of a nucleic acid such as DNA or RNA.It should be understood that the present teachings contemplate sequence information being obtained using all available types of techniques, platforms or technologies, including, but not limited to, capillary electrophoresis, microarrays, ligation-based systems, polymerase-based systems, hybridization-based systems, direct or indirect nucleotide identification systems, pyrosequencing, ion or pH-based detection systems, and electronic signature-based systems.
[0071] Immunotherapy: As used herein, "immunotherapy" refers to treatment with one or more agents that act to stimulate the immune system to kill cancer cells or at least inhibit the growth of cancer cells, preferably to reduce further growth of the cancer, reduce the size of the cancer, and / or eliminate the cancer. Some such agents bind to targets present on cancer cells; some bind to targets present on immune cells but not on cancer cells; some bind to targets present on both cancer cells and immune cells. Such agents include, but are not limited to, checkpoint inhibitors and / or antibodies. Checkpoint inhibitors are inhibitors of pathways in the immune system that maintain self-tolerance and modulate the duration and amplitude of physiological immune responses in peripheral tissues to minimize collateral tissue damage (see, e.g., Pardoll, Nature Reviews Cancer 12, 252-264 (2012)). Exemplary agents include PD-1, PD Antibodies for the tumor include antibodies against any of PD-L1, PD-L2, CTLA-4, OX40, B7.1, B7He, LAG3, CD137, KIR, CCR5, CD27, CD40, or CD47. Other exemplary agents include proinflammatory cytokines, such as IL-1β, IL-6, and TNF-α. Other exemplary agents are T cells activated against tumors, such as T cells activated by expressing chimeric antigens that target tumor antigens recognized by T cells.
[0072] Indel: As used herein, "indel" refers to a mutation involving the insertion or deletion of one or more nucleotides in the genome of a subject.
[0073] Indexed: As used herein, "indexed" refers to a first element (e.g., a microsatellite instability score) linked to a second element (e.g., a given therapy).
[0074] Instability status: As used herein, "instability status" or "instability score" in the context of repetitive nucleic acids (e.g., repetitive nucleic acid / repetitive DNA instability status or score, microsatellite instability status or score) refers to a measure or determination of whether a given repetitive nucleic acid locus or population of repetitive nucleic acid loci in one or more nucleic acid samples exhibits a level or degree of mutation (e.g., variable repeat length) above a threshold level determined for that locus or population of loci, a level or degree of mutation (e.g., variable repeat length) that is at that threshold level, or a level or degree of mutation (e.g., variable repeat length) that is below that threshold level. For clarity, instability status and instability score are not interchangeable, but rather related concepts. Instability status is based on instability score. For example, if the sample's instability score is at or below the population-trained threshold, the sample is classified as a stable sample (e.g., for MSI -MSS or low MSI), and if the sample's instability score is higher than the population-trained threshold, the sample is classified as an unstable sample (e.g., for MSI -high MSI).
[0075] Limit of Detection (LoD): As used herein, "limit of detection" or "LoD" refers to the smallest amount of a substance (e.g., nucleic acid) in a sample that can be measured by a given assay or analytical procedure.
[0076] Maximum MAF: As used herein, "maximum MAF" or "max MAF" refers to the maximum MAF of all somatic variants in a sample.
[0077] Microsatellite: As used herein, "microsatellite" refers to a repetitive nucleic acid having repeat units less than about 10 base pairs or nucleotides in length.
[0078] Minisatellite: As used herein, "minisatellite" refers to a repetitive nucleic acid having a repeat unit of about 10 to about 60 base pairs or nucleotides in length.
[0079] Mutant allele fraction: As used herein, "mutant allele fraction", "mutation dose", or "MAF" refers to the fraction of nucleic acid molecules that carry an allelic change or mutation at a given genomic location. MAF is generally expressed as a fraction or percentage. For example, MAF is typically less than about 0.5, 0.1, 0.05 or 0.01 (i.e., less than about 50%, 10%, 5% or 1%) of all somatic variants or alleles present at a given locus.
[0080] Mutation: As used herein, "mutation" refers to a mutation from a known reference sequence, including, for example, single nucleotide variants (SNVs), copy number variants or variations (CNVs) / abnormalities, insertions or deletions (indels), gene fusions, transversions, translocations, frameshifts, duplications, repeat expansions, and epigenetic variants. Mutations may be germline mutations or somatic mutations. In some embodiments, the reference sequence for comparison is the wild-type genomic sequence of the species of the subject that provides the test sample, typically the human genome.
[0081] Neoplasm: As used herein, the terms "neoplasm" and "tumor" are used interchangeably. They refer to the abnormal growth of cells in a subject. A neoplasm or tumor may be benign, potentially malignant, or malignant. A malignant tumor is referred to as a cancer or cancerous tumor.
[0082] Next generation sequencing: As used herein, "Next generation sequencing" or "NGS" refers to sequencing technology that has increased throughput compared to older Sanger-based and capillary electrophoresis-based techniques, for example, has the ability to simultaneously generate hundreds of thousands of relatively short sequence reads.Some examples of next generation sequencing techniques include, but are not limited to, sequencing by synthesis, sequencing by ligation, and sequencing by hybridization.
[0083] Nucleic acid tag: As used herein, "nucleic acid tag" refers to a short nucleic acid (e.g., less than about 500, 100, 50, or 10 nucleotides in length) that is used to distinguish between nucleic acids of different types or that have been treated differently from different samples (e.g., representing a sample index) or between different nucleic acids of different types or that have been treated differently in the same sample (e.g., representing a molecular barcode). Nucleic acid tags include a predetermined, fixed, non-random, random, or semi-random oligonucleotide sequence. Such nucleic acid tags can be used to label different nucleic acid molecules or different nucleic acid samples or subsamples. Nucleic acid tags can be single-stranded, double-stranded, or at least partially double-stranded. Nucleic acid tags can have the same length or various lengths, as appropriate. Nucleic acid tags can include double-stranded molecules with one or more blunt ends, can include 5' or 3' single-stranded regions (e.g., overhangs), and / or can include one or more other single-stranded regions in other regions in a given molecule. Nucleic acid tags can be attached to one or both ends of other nucleic acids (e.g., sample nucleic acids to be amplified and / or sequenced). Nucleic acid tags can be decoded to reveal information such as the sample of origin, form, or processing of a given nucleic acid. For example, nucleic acid tags can also be used to enable pooling and / or parallel processing of multiple samples containing nucleic acids with different molecular barcodes and / or sample indexes, which are then deconvoluted by detecting (e.g., reading) the nucleic acid tags. Nucleic acid tags can also be referred to as identifiers (e.g., molecular identifiers, sample identifiers). Additionally or alternatively, nucleic acid tags can be used as molecular barcodes (e.g., to distinguish between different molecules in the same sample or secondary samples, or between amplicons of different parent molecules). This includes, for example, uniquely tagging different nucleic acid molecules in a given sample, or non-uniquely tagging such molecules.For non-unique tagging applications, a limited number of tags (i.e., molecular barcodes) can be used to tag nucleic acid molecules, such that different molecules can be distinguished based on their endogenous sequence information (e.g., the start and / or end positions where they are located in a selected reference genome, a subsequence at one or both ends of the sequence, and / or the length of the sequence) in combination with at least one molecular barcode. Typically, a sufficient number of different molecular barcodes are used such that the probability that any two molecules can have the same endogenous sequence information (e.g., the start and / or end positions, a subsequence at one or both ends of the sequence, and / or the length) and also have the same molecular barcode is low (e.g., less than about 10%, less than about 5%, less than about 1%, or less than about 0.1% chance).
[0084] Polynucleotide: As used herein, "polynucleotide", "nucleic acid", "nucleic acid molecule", or "oligonucleotide" refers to a linear polymer of nucleosides (including deoxyribonucleosides, ribonucleosides, or analogs thereof) linked by internucleoside linkages. Typically, a polynucleotide contains at least three nucleosides. Oligonucleotides often range in size from a few monomeric units, e.g., 3-4, to hundreds of monomeric units. Whenever a polynucleotide is represented by a sequence of letters, such as "ATGCCTG", it will be understood that the nucleotides are in 5'→3' order from left to right, unless otherwise noted, and that in the case of DNA, "A" denotes deoxyadenosine, "C" denotes deoxycytidine, "G" denotes deoxyguanosine, and "T" denotes deoxythymidine. The letters A, C, G, and T may be used, as is common in the art, to refer to the bases themselves, to refer to the nucleosides, or to refer to the nucleotides that contain the bases.
[0085] Population trained threshold: As used herein, a "population trained threshold" in the context of repetitive nucleic acids refers to a separately determined maximum total number of unstable repetitive nucleic acid loci (e.g., a number of unstable microsatellite loci) expected to be observed in training DNA samples (e.g., non-tumor samples, tumor samples, etc.) that contain those loci. Population trained thresholds are typically used to characterize the experimentally determined repetitive nucleic acid instability scores for a particular sample.
[0086] Processing: As used herein, the terms "processing", "calculating" and "comparing" are used interchangeably. In certain embodiments, these terms refer to determining the difference, for example, the difference in number or sequence. For example, the value or sequence of repeat DNA instability score (e.g., microsatellite instability score), gene expression, copy number variation (CNV), indel, and / or single nucleotide variant (SNV) can be processed.
[0087] Reference sequence: As used herein, "reference sequence" refers to a known sequence used for the purpose of comparison with an experimentally determined sequence. For example, the known sequence may be a whole genome, a chromosome, or any segment thereof. A reference sequence typically comprises at least about 20, at least about 50, at least about 100, at least about 200, at least about 250, at least about 300, at least about 350, at least about 400, at least about 450, at least about 500, at least about 1000, or more nucleotides. A reference sequence can be aligned with a single continuous sequence of a genome or chromosome, or can comprise discontinuous segments that align with different regions of a genome or chromosome. Exemplary reference sequences include human genomes, such as, for example, hG19 and hG38.
[0088] Repeat length: As used herein, "repeat length" in the context of a repetitive nucleic acid refers to the number of repeat units present at a given repetitive nucleic acid locus. By way of example, the following single stranded nucleic acid strand has a repeat length of 8: [ka]
[0089] Repeat unit: As used herein, a "repeat unit" in the context of a repetitive nucleic acid refers to a particular nucleotide pattern or motif (e.g., homopolymer or heteropolymer) that is repeated at a given repetitive nucleic acid locus. To illustrate, the following single stranded nucleic acid strand has a repeat unit of "ATT": [ka]
[0090] Repetitive nucleic acid: As used herein, "repetitive nucleic acid" or "repetitive element" refers to a recurring pattern of nucleotides that are present in multiple copies throughout a given genome and / or population of genomes. Repetitive nucleic acids include repetitive DNA and repetitive RNA. Non-limiting examples of repetitive nucleic acids include microsatellites, terminal repeats, tandem repeats, minisatellites, satellite DNA, interspersed repeats, transposable elements (e.g., DNA transposons, retrotransposons (e.g., LTR-type retrotransposons (HERVs) and LTR-type retrotransposons (HERVs)), etc.), clustered regularly interspaced palindromic repeats (CRISPR), direct repeats, inverted repeats, mirror repeats, and inverted repeats.
[0091] Repeat nucleic acid instability score: As used herein, "repeat nucleic acid instability score" in the context of repeat nucleic acids (e.g., repeat DNA instability score, microsatellite instability score, etc.) refers to the total number of repeat nucleic acid loci from a population of repeat nucleic acid loci in a given sample that are called or otherwise determined to be unstable. This repeat nucleic acid instability score is a sample-level score (or sample score) and is distinct from the site score, which is locus-specific.
[0092] Sample: As used herein, "sample" means anything that can be analyzed by the methods and / or systems disclosed herein.
[0093] Sensitivity: As used herein, "sensitivity" refers to the probability of detecting the presence of a mutation at a given MAF and coverage.
[0094] Sequencing: As used herein, "sequencing" refers to any of the numerous techniques used to determine the sequence (e.g., identity and order of monomeric units) of biomolecules, e.g., nucleic acids such as DNA or RNA. Exemplary sequencing methods include targeted sequencing, single molecule real-time sequencing, exon or exome sequencing, intron sequencing, electron microscopy-based sequencing, panel sequencing, transistor-mediated sequencing, direct sequencing, random shotgun sequencing, Sanger dideoxytermination sequencing, whole genome sequencing, sequencing by hybridization, pyrosequencing, capillary electrophoresis, duplex sequencing, cycle sequencing, single-base extension sequencing, These include, but are not limited to, solid-phase sequencing, high-throughput sequencing, massively parallel signature sequencing, emulsion PCR, co-amplification-PCR at lower denaturation temperatures (COLD-PCR), multiplex PCR, reversible dye terminator sequencing, paired-end sequencing, near-term sequencing, exonuclease sequencing, sequencing by ligation, short-read sequencing, single molecule sequencing, sequencing by synthesis, real-time sequencing, reverse terminator sequencing, nanopore sequencing, 454 sequencing, Solexa Genome Analyzer sequencing, SOLiD™ sequencing, MS-PET sequencing, and combinations thereof. In some embodiments, sequencing can be performed by a genetic analyzer, such as those commercially available from Illumina, Inc., Pacific Biosciences, Inc., or Applied Biosystems / Thermo Fisher Scientific, among others.
[0095] Sequence information: As used herein, "sequence information" in the context of a nucleic acid polymer means the order and identity of the monomeric units (eg, nucleotides) in that polymer.
[0096] Site Score: As used herein, "site score" refers to a measure of the likelihood of the presence of additional repeat lengths other than the germline repeat length at a given repetitive nucleic acid locus in a sample. In certain embodiments, the site score is determined for a given locus by calculating the delta Akaike information criterion (ΔAIC) for that locus.
[0097] Site-specific trained threshold: As used herein, a "site-specific trained threshold" refers to the separately determined maximum value of site scores for a given repetitive nucleic acid locus (e.g., a given microsatellite locus) such that the locus is stable.
[0098] Somatic mutation: As used herein, "somatic mutation" refers to a mutation in the genome that occurs after conception. Somatic mutations can occur in any cell of the body except germ cells, and therefore cannot be passed on to offspring.
[0099] Specificity: As used herein, "specificity" in the context of a diagnostic analysis or assay refers to the degree to which the analysis or assay detects the intended target analyte and not other components of a given sample.
[0100] Substantial match: As used herein, "substantial match" means that at least a first value or element is at least approximately equal to at least a second value or element. In certain embodiments, for example, a customized therapy is identified when the microsatellite instability score and the comparator result are at least substantially or approximately matched.
[0101] Subject: As used herein, "subject" refers to an animal, such as a mammalian species (e.g., human) or avian (e.g., avian) species, or other organism, such as a plant. More specifically, the subject may be a vertebrate, such as a mammal, such as a mouse, a primate, a monkey, or a human. Animals include livestock (e.g., production cattle, dairy cattle, poultry, horses, pigs, and the like), sport animals, and companion animals (e.g., pets or support animals). The subject may be a healthy individual, an individual with or predisposed to a disease, or an individual suspected of having or predisposed to a disease, or an individual in need of treatment or suspected of needing treatment. The terms "individual" or "patient" are intended to be interchangeable with "subject."
[0102] For example, the subject may be an individual who has been diagnosed with cancer, who is going to undergo cancer therapy, and / or who has undergone at least one type of cancer therapy.The subject may be in remission from cancer.As another example, the subject may be an individual who is diagnosed with autoimmune disease.As another example, the subject may be a female individual who is pregnant or planning to become pregnant, and may have been diagnosed with or may be suspected of having a disease, such as cancer, autoimmune disease.
[0103] Threshold: As used herein, "threshold" refers to an independently determined value used to characterize or classify an experimentally determined value.
[0104] Training DNA sample: As used herein, "training DNA sample" refers to a DNA sample used to estimate site-specific trained thresholds and population trained thresholds. A training DNA sample dataset includes one or more training DNA samples. The training DNA sample includes one or more normal DNA samples and / or tumor DNA samples. In some embodiments, the training DNA sample includes one or more samples with high MSI and / or low MSI / MSS status.
[0105] Tumor proportion: As used herein, "tumor proportion" refers to the estimation of the proportion of nucleic acid molecules that are derived from tumor in a given sample.For example, the tumor proportion of a sample can be a measure derived from the max MAF of the sample or the coverage of the sample or the length of the cfDNA fragments in the sample or any other selected feature of the sample.In some embodiments, the tumor proportion of a sample is equal to the max MAF of the sample.
[0106] Unstable: As used herein, "unstable" or "instability" in the context of repetitive nucleic acids refers to a level of mutations (e.g., indels or the like) observed at a given repetitive nucleic acid locus or in a given population of repetitive nucleic acid loci in a nucleic acid sample (e.g., a cfDNA sample) that exceeds a threshold (e.g., site-specific trained threshold-locus level; population trained threshold-sample level; or the like).
[0107] Detailed Description Introduction
[0108] Cancer encompasses a large group of genetic diseases that have the common characteristic of abnormal cell growth and can metastasize beyond the site of cellular origin in the body. The molecular basis of this disease is mutations and / or epigenetic changes that lead to a transformed cell phenotype, whether these deleterious changes are acquired by inheritance or have a somatic basis. Complicating matters, these molecular changes typically differ not only between patients with the same cancer type, but even within a given patient's own tumor.
[0109] Given the variability of mutations observed in most cancers, one of the challenges in cancer treatment is to identify the therapy to which a patient is most likely to respond given their individual cancer type. Various biomarkers are used to match the right treatment, including cancer immunotherapy, to cancer patients. One biomarker of response is microsatellite instability (MSI), a state of or predisposition to genetic hypermutation caused by a decline in DNA mismatch repair (MMR) mechanisms. Cancer patients with microsatellite instability classified as high (MSI-H or high MSI) often exhibit an accumulation of somatic mutations in tumor cells that result in a range of molecular and biological changes, including high tumor mutation burden, increased expression of neoantigens, and abundant tumor-infiltrating lymphocytes. Changet al. "Microsatellite Instability: A PredictiveBiomarker for CancerImmunotherapy," Appl Immunohistochem Mol Morphol,26(2):e15-e21 (2018). These changes have been associated with increased sensitivity to checkpoint inhibitors such as pembrolizumab (Keytruda®), which are used to treat advanced melanoma, head and neck squamous cell carcinoma, non-small cell lung cancer (NSCLC), and classical Hodgkin's lymphoma. To date, the application of this response biomarker has been essentially limited to the assessment of MSI status in solid tumor samples using standard PCR-based techniques.
[0110] The present disclosure provides a method, computer readable medium and system that are useful for determining and analyzing MSI in patient samples, particularly cell-free DNA (cfDNA) samples.The MSI status determined using these methods and related embodiments can help guide disease prognosis and treatment decisions.The results achieved by the methods and related embodiments disclosed herein generally have a high degree of agreement with the results obtained using, for example, more conventional PCR-based MSI evaluation methods.
[0111] Methods for determining microsatellite instability status
[0112] The present application discloses various methods for accurately determining the microsatellite instability (MSI) status and / or other repeat DNA instability status of a sample, particularly a cell-free DNA (cfDNA) sample. In certain embodiments, the method for assessing MSI status includes targeted sequencing of cfDNA, for example using a digital sequencing platform from Guardant Health, Inc. (Redwood City, CA, USA), which allows for broad coverage of simple repeats where microsatellite instability may exist across a wide range of cancer types. The digital sequencing platform is an NGS panel of cancer-related genes that utilizes high-quality sequencing of cell-free DNA (which may include circulating tumor DNA) isolated by simple non-invasive blood sampling. Digital sequencing utilizes pre-sequencing preparation of a digital library of individually tagged cfDNA molecules combined with post-sequencing bioinformatics reconstruction to eliminate nearly all false positives. By way of example, FIG. 1 provides a flow chart that generally depicts an exemplary method step for determining MSI status according to some embodiments of the present invention. As shown, in step 110, the method 100 quantifies a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci. The sequence information is typically obtained from a population of microsatellite loci in a cfDNA sample. As described further herein, the number of different repeat lengths present at a given microsatellite locus is quantified in some embodiments using a site score based on a probabilistic log likelihood. As also described further herein, other quantification techniques are optionally utilized, provided that they are highly accurate in distinguishing the relatively small number of cfDNA fragments from somatic cells in the sample from, for example, noise resulting from post-sample collection artifacts (e.g., amplification artifacts, sequencing artifacts, and the like).
[0113] Method 100 also includes, at step 112, comparing the site score of a given microsatellite locus to a site-specific trained threshold for each of the particular microsatellite loci. The experimentally determined site score of a particular locus and its corresponding site-specific trained threshold are typically compared for each of a plurality of microsatellite loci. The site-specific trained threshold for a given locus is generally a pre-defined value for that particular locus derived from a population of training DNA samples, such as a cohort of normal or non-tumor cfDNA samples. As shown, method 100 includes, at step 114, calling a given microsatellite locus as unstable if the site score (e.g., likelihood score or the like) of the given microsatellite locus exceeds (e.g., is statistically greater than) the site-specific trained threshold for the given microsatellite locus. Based on these comparisons, a microsatellite instability score is generated, which includes the number of microsatellite loci called as unstable from the plurality of microsatellite loci (e.g., the total or aggregate MSI score of the sample). In addition, the method 100 also includes, in step 116, classifying the MSI status of the cfDNA sample as unstable if the microsatellite instability score exceeds a population-trained threshold of the population of microsatellite loci in the cfDNA sample, thereby identifying an unstable cfDNA sample (e.g., scoring or predicting the sample as having high MSI). In other words, the MSI status of the sample is determined, in certain embodiments, by the presence of a minimum number of unstable microsatellite loci. The population-trained threshold is generally a predetermined value derived from a population of training DNA samples, such as a cohort of normal or non-tumor cfDNA samples.
[0114] In some embodiments, the thresholds (e.g., site-specific trained thresholds, population trained thresholds, and the like) are determined from at least one training DNA sample dataset or are otherwise derived from such a dataset. The training DNA sample dataset typically comprises at least about 25 to at least about 30,000 or more training samples. In some embodiments, the training DNA sample dataset comprises about 50, 75, 100, 150, 200, 300, 400, 500, 600, 700, 800, 900, 1,000, 2,500, 5,000, 7,500, 10,000, 15,000, 20,000, 25,000, 50,000, 100,000, 1,000,000, or more training DNA samples.
[0115] In certain embodiments, method 100 includes additional upstream and / or downstream steps. In some embodiments, for example, method 100 begins in step 102 with providing a sample from a subject in step 104 (e.g., providing a blood sample drawn from the subject). In these embodiments, the workflow of method 100 also typically includes amplifying nucleic acids in the sample to generate amplified nucleic acids in step 106, and sequencing the amplified nucleic acids to generate sequence information in step 108, followed by quantifying the number of different repeat lengths present at each of the multiple microsatellite loci from the sequence information in step 110. Nucleic acid amplification (including associated sample preparation), nucleic acid sequencing, and associated data analysis are further described herein.
[0116] In some embodiments, method 100 includes various steps downstream from identifying an unstable cfDNA sample in step 116. Some examples of these include comparing the microsatellite instability status of the cfDNA sample to a therapy-indexed comparator result in step 118 to identify a customized therapy for treating a disease (e.g., cancer or another genetic disease, disorder, or condition) in the subject. In other exemplary embodiments, method 100 also includes administering at least one of the identified customized therapies to the subject if the microsatellite instability status of the sample and the comparator result in step 120 substantially match prior to termination in step 122 (e.g., to treat the subject's cancer or another disease, disorder, or condition).
[0117] The methods described herein include various alternative embodiments. For example, the site score of a microsatellite locus optionally includes a likelihood score. In some embodiments, the likelihood score includes a probabilistic log likelihood-based score. In some of these embodiments, the method includes determining a probabilistic log likelihood-based score for each microsatellite locus in the sequence information obtained from the sample using various parameters such as allele frequency or one or more error modes (e.g., random error mode, strand-specific error mode, and / or the like). The allele frequency generally includes the observed frequency of nucleic acids with different repeat lengths at a given microsatellite locus in the sequence information obtained from the sample. In some embodiments, the site score of a particular microsatellite locus includes the difference or ratio between (a) a score that measures the support of the observed nucleic acid sequence to the null hypothesis that the given microsatellite locus is stable and (b) a score that measures the support of the observed nucleic acid sequence to the alternative hypothesis that the given microsatellite locus is unstable. The null hypothesis is the one with the smallest AIC score among all hypotheses under the assumption that the site is stable, and the alternative hypothesis is the one with the smallest AIC score among all hypotheses under the assumption that the site is unstable. Typically, site scores are generated using various measures of model accuracy, such as likelihood criteria, log likelihood criteria, posterior probability criteria, Akaike Information Criterion (AIC), Bayesian Information Criterion, and / or the like. Further details regarding statistical modeling, including measures of statistical model accuracy as appropriate for use in carrying out the methods disclosed herein, can be found, for example, in Bruce, Practical Statistics for Data Scientists: 50 Essential Concepts, 1 st Ed., O'Reilly Media (2017), Freedman et al., Statistics, 4 thEd., WW Norton & Company (2007), James et al., An Introduction to Statistical Learning: with Applications in R, 1 st Ed., Springer (2013), and Hastieet al., The Elements of Statistical Learning:Data Mining, Inference, andPrediction, 2 nd Ed., Springer (2016), each of which is incorporated by reference herein in its entirety.
[0118] To further illustrate, the site score for a given microsatellite locus optionally includes a site score based on an AIC that tests for somatic indels at that microsatellite locus. In some of these embodiments, the site score based on a given AIC is calculated using the following formula: AIC=k-log likelihood where k is the number of parameters used in the model. In some embodiments, the method includes estimating the parameters of the model using maximum likelihood estimation (MLE) (e.g., using the Nelder-Mead algorithm or another simplex search algorithm). The method includes estimating the parameters of the model using the following formula: AIC 0 = k-log(Pr(obs|β,γ)) (In the formula, AIC 0 is the null hypothesis, k is the number of parameters used in the model, Pr is the probability, obs is the number of observed sequencing reads covering a given microsatellite locus, β is at least one strand-specific error parameter, and γ is at least one random error parameter. In some of these embodiments, the method optionally includes calculating a null hypothesis for the model using the following formula: AIC min =min α (k-log(Pr(obs|β,γ,α)) (In the formula, AIC min is the alternative hypothesis, and min α is the minimizing effect over all values of α, k is the number of parameters used in the model, Pr is the probability, obs is the number of observed sequencing reads covering a given microsatellite locus, β is at least one strand-specific error parameter, γ is at least one random error parameter, and α is an allele frequency of at least 1, where α is one or more i is a vector of allele frequencies whose sum is equal to 1. The change in the model used to determine the site score (ΔAIC) is calculated using the following formula: ΔAIC=AIC 0 -AIC min is typically detected using
[0119] In certain embodiments, the parameter γ comprises: (a) the read-level error rate where the observed microsatellite length in the sequencing read is one repeat unit longer than the expected microsatellite length for the strand of the original nucleic acid molecule; and / or (b) the read-level error rate where the observed microsatellite length in the sequencing read is one repeat unit shorter than the expected microsatellite length for the strand of the original nucleic acid molecule. In some embodiments, parameter β comprises: (a) a strand-level error rate where the expected microsatellite length of the sense strand is one repeat unit longer than the expected microsatellite length of the nucleic acid derived molecule; (b) a strand-level error rate where the expected microsatellite length of the antisense strand is one repeat unit longer than the expected microsatellite length of the nucleic acid derived molecule; (c) a strand-level error rate where the expected microsatellite length of the sense strand is one repeat unit shorter than the expected microsatellite length of the nucleic acid derived molecule; and / or (d) a strand-level error rate where the expected microsatellite length of the antisense strand is one repeat unit shorter than the expected microsatellite length of the nucleic acid derived molecule.
[0120] In some embodiments, the site score based on AIC is calculated using the following formula: AIC=2(k-log likelihood) where k is the number of parameters used in the model. In these embodiments, the AIC 0 and AIC min is calculated using the formula above.
[0121] For clarity, in embodiments where the AIC-based score is determined using the formula AIC=2(k-log likelihood), the site-specific threshold used to classify a site as unstable will be twice the site-specific threshold used in the previous embodiment where the AIC-based score is determined using the formula AIC=k-log likelihood.
[0122] Samples analyzed using the methods described herein typically contain a variety of mutant allele fractions (MAFs) (e.g., sample fractions that exhibit different repeat lengths, specific microsatellite loci, or other allele changes). In addition, samples in some embodiments contain tumor fractions. In certain embodiments, the maximum MAF (max MAF) serves as an approximation of the tumor fraction in a given sample. The tumor fraction is typically less than about 0.05%, about 0.1%, about 0.2%, about 0.5%, about 1%, about 2%, about 3%, about 4%, about 5%, about 6%, about 7%, about 8%, about 9%, about 10%, about 11%, about 12%, about 13%, about 14% or about 15% of all nucleic acids in the sample.
[0123] In some embodiments, the methods disclosed herein typically include a sensitivity of at least about 94% at a limit of detection (LOD) of about 0.2% tumor fraction of nucleic acid in a given sample. The methods also typically have a specificity of at least about 99% for non-tumor DNA in the sample. The MSI status determined for the sample also typically has at least about 95%, 96%, 97%, 98% or 99% concordance with the corresponding MSI status of the sample determined using standard PCR-based MSI assessment techniques over a tumor fraction range of about 1.4% to about 15%. In some embodiments, the concordance is 100%.
[0124] In certain embodiments, the MSI status of a particular sample is classified as high-grade MSI (MSI-H) if the microsatellite instability score for the sample is about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 11, about 12, about 13, about 14, about 15, about 16, about 17, about 18, about 19, about 20, about 30, about 40, about 50, about 60, about 70, about 80, about 90, about 100 or more than 100 unstable microsatellite loci in the sample. In certain embodiments, the population trained threshold used to determine the instability status (e.g., MSI status) of a sample is about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 11, about 12, about 13, about 14, about 15, about 16, about 17, about 18, about 19, about 20, about 30, about 40, about 50, about 60, about 70, about 80, about 90, about 100, or more than 100 unstable repetitive nucleic acid (e.g., microsatellite) loci. In some embodiments, the population trained threshold of the sample is about 5 unstable microsatellite loci. In some embodiments, the population trained threshold of the sample is about 6 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold of the sample is about 10 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold of the sample is about 15 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 16 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 20 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 25 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 26 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 30 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 35 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 36 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 40 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 45 unstable repetitive nucleic acid loci.In some embodiments, the population trained threshold for the sample is about 46 unstable repetitive nucleic acid loci. In some embodiments, the population trained threshold for the sample is about 50 unstable repetitive nucleic acid loci. In some embodiments, the repetitive nucleic acid loci can be microsatellite loci. In some embodiments, the MSI status of a given sample is classified as MSI-H if the number of unstable microsatellite loci constitutes about 0.1%, about 1%, about 2%, about 3%, about 4%, about 5%, about 10%, about 15%, about 20%, or about 25% of all microsatellite loci evaluated in that sample. In some embodiments, about 50, about 60, about 70, about 80, about 90, about 100, about 200, about 300, about 400, about 500, about 600, about 700, about 800, about 900, about 1000, about 1100, about 1200, about 1300, about 1400, about 1500, about 1600, about 1700, about 1800, about 1900, about 2000, or more than 2000 repeat nucleic acid (e.g., microsatellite) loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 50 repeat nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 60 repeat nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 70 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 80 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 90 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 100 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 200 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 300 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample.In some embodiments, about 400 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 500 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 1000 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 1100 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 1200 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 1300 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 1400 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, at least 1500 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, about 1600 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, at least 1700 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, at least 1800 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, at least 1900 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, at least 2000 repetitive nucleic acid loci are used to determine the repeat nucleic acid instability (e.g., MSI) status of a given sample. In some embodiments, the repetitive nucleic acid loci can be microsatellite loci. In some embodiments, the repetitive nucleic acid instability status can be MSI status.
[0125] In some embodiments, the method includes obtaining a sample from a subject. Essentially any sample type may be utilized as desired. In certain embodiments, for example, the sample is tissue, blood, plasma, serum, sputum, urine, semen, vaginal fluid, feces, synovial fluid, spinal fluid, saliva, and / or the like. Additional exemplary sample types that may be utilized as desired are described further herein. Typically, the subject is a mammalian subject (e.g., a human subject). Essentially any type of nucleic acid (e.g., DNA and / or RNA) may be evaluated according to the methods disclosed herein. Some examples include cell-free nucleic acid (e.g., cfDNA, whether tumor-derived, fetal-derived, maternal-derived, and / or the like), cellular nucleic acid, including circulating tumor cells (e.g., obtained by lysis of intact cells in the sample), circulating tumor nucleic acid, and the like. In some embodiments, the sample includes cell-free DNA (cfDNA sample). In some embodiments, the cfDNA sample includes circulating tumor nucleic acid.
[0126] The methods disclosed herein generally include obtaining sequence information from nucleic acid in a sample taken from a subject. In certain embodiments, the sequence information is obtained from a targeted segment of nucleic acid. Essentially any number of genomic regions may be targeted as desired. The targeted segments can include at least 10, at least 50, at least 100, at least 500, at least 1000, at least 2000, at least 5000, at least 10,000, at least 20,000 or at least 50,000 (e.g., 25, 50, 75, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1,000, 2,000, 3,000, 4,000, 5,000, 6,000, 7,000, 8,000, 9,000, 10,000, 15,000, 25,000, 30,000, 35,000, 40,000, 45,000) different or overlapping genomic regions. In some embodiments, the targeted segment comprises selected regions of at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, at least 200, at least 300, at least 400, at least 500, at least 600, or at least 700 genes. In some embodiments, the targeted segment comprises selected regions of at least 70 genes. In some embodiments, the targeted segment comprises regions of at least 500 genes.
[0127] In these embodiments, the methods typically include various sample or library preparation steps to prepare the nucleic acids for sequencing. Many different sample preparation techniques are well known to those of skill in the art. Essentially any of these techniques may be used or adapted for use in carrying out the methods described herein. For example, in addition to various purification steps to isolate the nucleic acids from other components in a given sample, typical steps for preparing nucleic acids for sequencing include tagging the nucleic acids with molecular identifiers or barcodes, adding adapters (e.g., which may include barcodes), amplifying the nucleic acids one or more times, enriching targeted segments of the nucleic acids (e.g., using various target capture strategies, etc.), and / or the like. Exemplary library preparation processes are described further herein. Further details regarding nucleic acid sample / library preparation can be found, for example, in vanDijk et al., Library preparation methods for next-generation sequencing: Tonedown the bias, Experimental Cell Research, 322(1):12-20 (2014); Micic (Ed.), Sample Preparation Techniques for Soil, Plant, and Animal Samples (SpringerProtocols Handbooks), 1 st Ed., Humana Press (2016), and Chiu, Next-Generation Sequencing and Sequence Data Analysis, Bentham Science Publishers (2018), each of which is incorporated by reference in its entirety.
[0128] The microsatellite and / or other repeat nucleic acid instability status determined by the methods disclosed herein is optionally used to diagnose the presence of a disease or condition, particularly cancer, in a subject, to characterize such a disease or condition (e.g., to stage a given cancer, to determine the heterogeneity of the cancer, and the like), to monitor the response to treatment, to assess the potential risk of developing a given disease or condition, and / or to assess the prognosis of the disease or condition. The microsatellite and / or other repeat nucleic acid instability status is also optionally used to characterize specific forms of cancer. Because cancers are often heterogeneous in terms of both composition and staging, the microsatellite and / or other repeat nucleic acid instability status can allow the characterization of specific subtypes of cancer, thereby aiding in diagnosis and treatment selection. This information can also provide clues to the subject or medical practitioner regarding the prognosis of a particular type of cancer, allowing the subject and / or medical practitioner to tailor treatment options as the disease progresses. Some cancers become more aggressive and genetically more unstable as they progress. Other tumors remain benign, inactive or dormant.
[0129] Microsatellite and / or other repeat nucleic acid instability status can also be useful for determining disease progression and / or monitoring recurrence. In certain cases, for example, successful treatment initially increases observed microsatellite and / or other repeat nucleic acid instability as an increase in the number of cancer cell deaths and shed nucleic acid. In these cases, as treatment progresses, microsatellite and / or other repeat nucleic acid instability will usually subsequently decrease as the size of the tumor continues to decrease. In other cases, successful treatment can also decrease microsatellite and / or other repeat nucleic acid instability without an initial increase in such instability. In addition, when cancer is found to be in remission after treatment, microsatellite and / or other repeat nucleic acid instability status can be used to monitor residual disease or recurrence of disease in patients.
[0130] sample
[0131] The sample can be any biological sample isolated from a subject. Samples can include body tissue, whole blood, platelets, serum, plasma, feces, red blood cells, white blood cells or leucocytes, endothelial cells, tissue biopsies (e.g., biopsies from known or suspected solid tumors), and the like. ), cerebrospinal fluid, synovial fluid, lymphatic fluid, ascites, interstitial or extracellular fluid (e.g., fluid from intercellular spaces), gingival exudate, crevicular fluid, bone marrow, pleural exudate, cerebrospinal fluid, saliva, mucus, sputum, semen, sweat, urine. The sample is preferably a body fluid, in particular blood and its fractions, and urine. Such samples include nucleic acids shed from tumors. Nucleic acids can include DNA and RNA, and can be in double-stranded and single-stranded forms. The sample may be in the form in which it was originally isolated from the subject, or may have been subjected to further processing to remove or add components such as cells, to enrich one component relative to another, or to convert one form of nucleic acid into another, for example, RNA into DNA, or single-stranded nucleic acid into double-stranded nucleic acid. Thus, for example, the body fluid sample for analysis is serum or plasma, which contains cell-free nucleic acid, for example, cell-free DNA (cfDNA).
[0132] In some embodiments, the sample volume of bodily fluid taken from the subject depends on the desired read depth for the region to be sequenced. Exemplary volumes are about 0.4-40 ml, about 5-20 ml, about 10-20 ml. For example, the volume can be about 0.5 ml, about 1 ml, about 5 ml, about 10 ml, about 20 ml, about 30 ml, about 40 ml, or more milliliters. The volume of plasma sampled is typically between about 5 ml and about 20 ml.
[0133] Samples can contain varying amounts of nucleic acid. Typically, the amount of nucleic acid in a given sample is considered to be equivalent to a number of genome equivalents. For example, a sample of about 30 ng of DNA contains about 10,000 (10 4 ) individual polynucleotide molecules in cfDNA, and approximately 200 billion (2 × 10 11 Similarly, a sample of about 100 ng of DNA contains the equivalent of about 30,000 haploid human genomes, or about 600 billion individual molecules in the case of cfDNA.
[0134] In some embodiments, the sample comprises nucleic acid from different sources, for example, from cells and from acellular sources (such as blood samples). Typically, the sample comprises nucleic acid having mutations. For example, the sample optionally comprises DNA having germline mutations and / or somatic mutations. Typically, the sample comprises DNA having cancer-associated mutations (e.g., cancer-associated somatic mutations).
[0135] Exemplary amounts of cell-free nucleic acid in a sample prior to amplification typically range from about 1 femtogram (fg) to about 1 microgram (μg), e.g., from about 1 picogram (pg) to about 200 nanograms (ng), from about 1 ng to about 100 ng, from about 10 ng to about 1000 ng. In some embodiments, the sample contains about 600 ng or less, about 500 ng or less, about 400 ng or less, about 300 ng or less, about 200 ng or less, about 100 ng or less, about 50 ng or less, or about 20 ng or less of cell-free nucleic acid molecules. Optionally, the amount is at least about 1 fg, at least about 10 fg, at least about 100 fg, at least about 1 pg, at least about 10 pg, at least about 100 pg, at least about 1 ng, at least about 10 ng, at least about 100 ng, at least about 150 ng, or at least about 200 ng of cell-free nucleic acid molecules. In certain embodiments, the amount is about 1 fg, about 10 fg, about 100 fg, about 1 pg, about 10 pg, about 100 pg, about 1 ng, about 10 ng, about 100 ng, about 150 ng, or about 200 ng of cell free nucleic acid molecules or less. In certain embodiments, the method includes obtaining between about 5 ng and about 30 ng of cell free nucleic acid molecules from the sample. In certain embodiments, the method includes obtaining between about 5 ng and about 100 ng of cell free nucleic acid molecules from the sample. In certain embodiments, the method includes obtaining between about 5 ng and about 150 ng of cell free nucleic acid molecules from the sample. In certain embodiments, the method includes obtaining between about 5 ng and about 200 ng of cell free nucleic acid molecules from the sample. In some embodiments, the amount is about 100 ng or less of cell free nucleic acid molecules from the sample. In some embodiments, the amount is about 150 ng or less of cell free nucleic acid molecules from the sample. In some embodiments, the amount is about 200 ng or less of cell free nucleic acid molecules from the sample. In some embodiments, the amount is about 250 ng or less of cell-free nucleic acid molecules from the sample. In some embodiments, the amount is about 300 ng or less of cell-free nucleic acid molecules from the sample. In certain embodiments, the method includes obtaining between about 1 fg and about 200 ng of cell-free nucleic acid molecules from the sample.
[0136] The cell-free nucleic acid molecules typically have a size distribution between about 100 nucleotides in length and about 500 nucleotides in length, with molecules between about 110 nucleotides in length and about 230 nucleotides in length representing about 90% of the molecules in the sample, with a mode at about 168 nucleotides in length and a second minor peak in the range between about 240 and about 440 nucleotides in length. In certain embodiments, the cell-free nucleic acid is between about 160 and about 180 nucleotides in length, or between about 320 and about 360 nucleotides in length, or between about 440 and about 480 nucleotides in length.
[0137] In some embodiments, the cell-free nucleic acid is isolated from the body fluid by a partitioning step, in which the cell-free nucleic acid as found in solution is separated from intact cells and other non-soluble components of the body fluid. In some of these embodiments, the partitioning includes techniques such as centrifugation or filtration. Alternatively, the cells in the body fluid are lysed and the cell-free and cellular nucleic acids are processed together. Generally, after the addition of a buffer and a washing step, the cell-free nucleic acid is precipitated, for example with alcohol. In certain embodiments, an additional purification step, such as a silica-based column to remove contaminants or salts, is used. For example, non-specific bulk carrier nucleic acid is added as needed throughout the reaction to optimize certain aspects of the exemplary procedure, such as yield. After such processing, the sample typically contains various forms of nucleic acid, including double-stranded DNA, single-stranded DNA and / or single-stranded RNA. If necessary, the single-stranded DNA and / or single-stranded RNA is converted to a double-stranded form so that it can be included in subsequent processing and analysis steps.
[0138] Nucleic Acid Tags
[0139] In some embodiments, nucleic acid molecules (from a sample of polynucleotides) can be tagged with a sample index and / or a molecular barcode (commonly referred to as a "tag"). The tag can be incorporated or otherwise attached to an adapter by chemical synthesis, ligation (e.g., blunt-end ligation or sticky-end ligation), or overlap extension polymerase chain reaction (PCR), among other methods. Such an adapter can eventually be attached to a target nucleic acid molecule. In other embodiments, one or more rounds of amplification cycles (e.g., PCR amplification) are applied to introduce a sample index to the nucleic acid molecule using conventional nucleic acid amplification methods. Amplification can be performed in one or more reaction mixtures (e.g., multiple microwells in the form of an array). The molecular barcode and / or sample index can be introduced simultaneously or in any order. In some embodiments, the molecular barcode and / or sample index are introduced before and / or after the sequence capture step is performed. In some embodiments, only the molecular barcode is introduced before the capture of the probe, and the sample index is introduced after the sequence capture step is performed. In some embodiments, both the molecular barcode and the sample index are introduced before the probe-based capture step is performed. In some embodiments, the sample index is introduced after the sequence capture step is performed. In some embodiments, the molecular barcode is incorporated into the nucleic acid molecule (e.g., cfDNA molecule) in the sample via an adaptor by ligation (e.g., blunt-end ligation or sticky-end ligation). In some embodiments, the sample index is incorporated into the nucleic acid molecule (e.g., cfDNA molecule) in the sample by overlap extension polymerase chain reaction (PCR). Typically, the sequence capture protocol involves introducing a single-stranded nucleic acid molecule complementary to a targeted nucleic acid sequence, e.g., a coding sequence of a genomic region, where mutations in such region are associated with a cancer type.
[0140] In some embodiments, the tag can be located at one end or both ends of the sample nucleic acid molecule.In some embodiments, the tag is a predetermined or random or semi-random sequence oligonucleotide.In some embodiments, the tag is less than about 500, 200, 100, 50, 20, 10, 9, 8, 7, 6, 5, 4, 3, 2 or 1 nucleotide in length.The tag can be randomly or randomly linked to the sample nucleic acid.
[0141] In some embodiments, each sample is uniquely tagged with a sample index or a combination of sample indexes. In some embodiments, each nucleic acid molecule of a sample or subsample is uniquely tagged with a molecular barcode or a combination of molecular barcodes. In other embodiments, multiple molecular barcodes (e.g., non-unique molecular barcodes) may be used, where the molecular barcodes in the multiple molecular barcodes are not necessarily unique to each other. In these embodiments, the molecular barcodes are generally attached (e.g., by ligation) to individual molecules, resulting in unique sequences that can be tracked individually by the combination of the molecular barcode and the sequence to which it can be attached. Detection of the non-uniquely tagged molecular barcodes in combination with endogenous sequence information (e.g., beginning (start) and / or end (termination) portions corresponding to the sequence of the original nucleic acid molecule in the sample, partial sequences at one or both ends of the sequence read, length of the sequence read, and / or length of the original nucleic acid molecule in the sample) allows a unique identity to be given to a particular molecule. The length of the individual sequence reads, or the number of base pairs of the individual sequence reads, are also used as necessary to give a unique identity to a given molecule. As described herein, fragments from a single strand of nucleic acid can be given a unique identity thereby allowing subsequent identification of the fragment from the parental and / or complementary strand.
[0142] In some embodiments, molecular barcodes are introduced to molecules in a sample in an expected ratio of a set of identifiers (e.g., a combination of unique and non-unique molecular barcodes). One exemplary format uses about 2 to about 1,000,000 different molecular barcodes, or about 5 to about 150 different molecular barcodes, or about 20 to about 50 different molecular barcodes. Alternatively, about 25 to about 1,000,000 different molecular barcodes may be used. Molecular barcodes can be ligated to both ends of a target molecule. For example, 20-50 x 20-50 molecular barcodes can be used. In some embodiments, 20-50 different molecular barcodes can be used. In some embodiments, 5-100 different molecular barcodes can be used. In some embodiments, 5-150 molecular barcodes can be used. In some embodiments, 5-200 different molecular barcodes can be used. Such a number of identifiers is typically sufficient so that different molecules with the same start and end points have a high probability (e.g., at least 94%, 99.5%, 99.99%, or 99.999%) that they will receive different combinations of identifiers. In some embodiments, about 80%, about 90%, about 95%, or about 99% of the molecules have the same combination of molecular barcodes.
[0143] In some embodiments, the assignment of unique or non-unique molecular barcodes during the reaction can be performed using methods and systems described, for example, in U.S. Patent Application Publication Nos. 20010053519, 20030152490, and 20110160078, and U.S. Patent Nos. 6,582,908, 7,537,898, 9,598,731, and 9,902,992, each of which is hereby incorporated by reference in its entirety. Alternatively, in some embodiments, only intrinsic sequence information (e.g., start and / or end positions, partial sequences at one or both ends of the sequence, and / or length) can be used to identify different nucleic acid molecules of a sample.
[0144] Nucleic Acid Amplification
[0145] The sample nucleic acid flanked by adaptors is typically amplified by PCR and other amplification methods using nucleic acid primers that bind to the primer binding sites in the adaptors that are flanked by the DNA molecules to be amplified.In some embodiments, the amplification method includes cycles of extension, denaturation and annealing that occur as a result of thermal cycling, or may be isothermal amplification, such as in the case of transcription-mediated amplification.Other exemplary amplification methods that can be used as needed include ligase chain reaction, strand displacement amplification, nucleic acid sequence-based amplification, and self-sustained sequence replication, among other techniques.
[0146] One or more rounds of amplification cycles are generally applied to introduce a sample index into a nucleic acid molecule using conventional nucleic acid amplification methods. Amplification is typically performed in one or more reaction mixtures. Molecular tags and sample indexes / tags can be introduced simultaneously or in any order, as needed. In some embodiments, molecular tags and sample indexes / tags are introduced before and / or after a nucleic acid molecule capture step (i.e., nucleic acid enrichment) is performed. In some embodiments, only molecular tags are introduced before probe capture, and sample indexes / tags are introduced after a sequence capture step is performed. In certain embodiments, both molecular tags and sample indexes / tags are introduced before a probe-based capture step is performed. In some embodiments, sample indexes / tags are introduced after a sequence capture step is performed. Typically, a sequence capture protocol involves introducing a single-stranded molecule complementary to a targeted nucleic acid sequence, e.g., a coding sequence of a genomic region, and a mutation in such region that is associated with a cancer type. Typically, the amplification reaction produces a plurality of nucleic acid amplicons, non-uniquely or uniquely tagged with molecular tags and sample indexes / tags, in a size range of about 200 nucleotides (nt) to about 700 nt, 250 nt to about 350 nt, or about 320 nt to about 550 nt. In some embodiments, the amplicons have a size of about 300 nt. In some embodiments, the amplicons have a size of about 500 nt.
[0147] Nucleic acid enrichment
[0148] In some embodiments, sequences are enriched prior to sequencing the nucleic acids. Enrichment is optionally performed for specific target regions ("target sequences"). In some embodiments, differential tiling and capture mechanisms can be used to enrich target regions of interest using nucleic acid capture probes ("baits") selected for one or more bait set panels. Differential tiling and capture mechanisms generally use different relative concentrations of bait sets to differentially tile (e.g., at different "resolutions") across genomic regions associated with the baits that are subject to a set of constraints (e.g., sequencer constraints such as sequencing load, availability of each bait, etc.) and capture a desired level of target nucleic acid for downstream sequencing. These targeted genomic regions of interest optionally include natural or synthetic nucleotide sequences of nucleic acid constructs. In some embodiments, biotin-labeled beads with probes for one or more regions of interest can be used to capture target sequences, and optionally, the regions can then be amplified to enrich the regions of interest.
[0149] Sequence capture typically involves the use of oligonucleotide probes that hybridize to target nucleic acid sequences. In certain embodiments, the probe set strategy involves tiling probes across the region of interest. Such probes can be, for example, about 60 to about 120 nucleotides in length. Sets can have a depth of about 2x, 3x, 4x, 5x, 6x, 8x, 9x, 10x, 15x, 20x, 30x, 40x, 50x, or greater than 50x. The effectiveness of sequence capture generally depends in part on the length of the sequence within the target molecule that is complementary (or nearly complementary) to the sequence of the probe.
[0150] Nucleic Acid Sequencing
[0151] The sample nucleic acid, which may or may not be pre-amplified, and may be flanked by adapter, will generally be subjected to sequencing.Optionally, the sequencing method or commercially available format used includes, for example, Sanger sequencing; high-throughput sequencing; pyrosequencing; sequencing by synthesis; single molecule sequencing; nanopore-based sequencing; semiconductor sequencing; sequencing by ligation; sequencing by hybridization; RNA-Seq (Illumina); Digital Gene Expression (Helicos); next-generation sequencing (NGS); Single Molecule Sequencing by Synthesis (SMSS) (Helicos); massively parallel sequencing; Clonal Single Molecule Array (Solexa); shotgun sequencing; Ion Torrent; Oxford Nanopore; Roche Genia; Maxim-Gilbert sequencing; primer walking; sequencing using PacBio, SOLiD, Ion Torrent or nanopore platform. Sequencing reactions can be performed in a variety of sample processing units, which may include multiple lanes, multiple channels, multiple wells, or other means of processing multiple sample sets substantially simultaneously. The sample processing units may also include multiple sample chambers that allow multiple runs to be processed simultaneously.
[0152] The sequencing reaction can be performed with one or more nucleic acid fragment types or regions known to contain markers of cancer or other diseases (e.g., microsatellites and / or other repetitive nucleic acid elements). The sequencing reaction can also be performed with any nucleic acid fragment present in the sample. The sequence reaction can produce a sequence coverage of the genome of at least about 5%, 10%, 15%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99%, 99.9% or 100% of the genome. In other cases, the sequence coverage of the genome can be less than about 5%, 10%, 15%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99%, 99.9% or 100% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.01%, 0.02%, 0.05%, 0.1%, 0.2%, 0.5%, 1%, 2% or 5% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.01% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.02% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.05% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.1% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.2% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 0.5% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 1% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 2% of the genome. In some embodiments, the sequence coverage of the genome may be less than about 5% of the genome. In some embodiments, the sequence coverage of the genome may be at least about 5% of the genome. In some embodiments, the sequence coverage of the genome may be at least about 10% of the genome. In some embodiments, the sequence coverage of the genome may be at least about 20% of the genome.
[0153] Multiplex sequencing techniques can be used to carry out simultaneous sequencing reactions.In some embodiments, cell-free polynucleotides are sequenced using at least about 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 50000 or 100,000 sequencing reactions.In other embodiments, cell-free polynucleotides are sequenced using less than about 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 50000 or 100,000 sequencing reactions.Sequencing reactions are typically carried out sequentially or simultaneously.Subsequent data analysis is generally carried out for all or part of sequencing reactions. In some embodiments, data analysis is performed on at least about 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 50000, or 100,000 sequencing reactions. In other embodiments, data analysis may be performed on less than about 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 50000, or 100,000 sequencing reactions. Exemplary read depths are about 1000 to about 50,000 reads per locus (base position), or >50,000 reads per locus.
[0154] In some embodiments, a nucleic acid population is prepared for sequencing by enzymatically forming blunt ends on double-stranded nucleic acids with single-stranded overhangs at one or both ends. In these embodiments, the population is typically treated with an enzyme having 5'-3' DNA polymerase activity and 3'-5' exonuclease activity in the presence of dNTP forms of nucleotides (e.g., A, C, G, and T or U). Exemplary enzymes or their catalytic fragments that are optionally used include large Klenow fragment and T4 polymerase. In the 5' overhang, the enzyme typically extends the buried 3' end of the opposing strand until it overlaps the 5' end to generate a blunt end. In the 3' overhang, the enzyme generally digests from the 3' end to the 5' end of the opposing strand, and may also digest beyond the 5' end. If this digestion proceeds beyond the 5' end of the opposing strand, the enzyme with the same polymerase activity used for the 5' overhang can fill the gap. The formation of blunt ends on a double-stranded nucleic acid can, for example, facilitate the attachment of adapters and subsequent amplification.
[0155] In some embodiments, the nucleic acid population will undergo further processing, such as converting single stranded nucleic acids to double stranded and / or converting RNA to DNA. These forms of nucleic acid are also optionally ligated to adaptors and amplified.
[0156] The nucleic acid, which may or may not have been previously amplified, will be subjected to the process of forming blunt ends described above, and, if necessary, other nucleic acids in the sample can be sequenced to generate sequenced nucleic acids. A sequenced nucleic acid can refer to the sequence (i.e., sequence information) of a nucleic acid, or to a nucleic acid whose sequence has been determined. Sequencing can be performed to obtain sequence data for individual nucleic acid molecules directly or indirectly from the consensus sequence of the amplification products of the individual nucleic acid molecules in the sample.
[0157] In some embodiments, after blunt end formation, both ends of the double-stranded nucleic acid with single-stranded overhang in the sample are ligated to an adapter containing a barcode, and the nucleic acid sequence as well as the barcodes introduced by the adapter arranged in series are determined by sequencing. The blunt-ended DNA molecule is optionally ligated to the blunt end of an at least partially double-stranded adapter (e.g., Y-shaped or bell-shaped adapter). Alternatively, tails with complementary nucleotides can be added to the blunt ends of the sample nucleic acid and the adapter to facilitate ligation (e.g., sticky end ligation).
[0158] A nucleic acid sample is typically contacted with a sufficient number of adapters so that the probability that any two identical nucleic acids will receive the same combination of adapter barcodes from adapters ligated to both ends is low (e.g., <1 or <0.1%). The use of adapters in this manner allows for the identification of a family of nucleic acid sequences that have the same start and end points as on the reference nucleic acid and are ligated to the same combination of barcodes. Such a family represents the pre-amplification sequence of the amplification product of the nucleic acid in the sample. The sequences of the family members can be compiled to derive the consensus nucleotide or the complete consensus sequence of the nucleic acid in the original sample modified by blunt end formation and adapter ligation. In other words, a nucleotide occupying a particular position of a nucleic acid in a sample is determined to be the consensus nucleotide of the nucleotide occupying its corresponding position in the family member sequence. A family may include sequences of one strand of a double-stranded nucleic acid, or may include sequences of both strands. When family members comprise the sequences of both strands of double-stranded nucleic acid, the sequences of one strand are converted to their complementary sequences to compile all sequences and derive consensus nucleotide or sequence.Some families only comprise a single member sequence.In this case, this sequence can be considered as the sequence of the nucleic acid in the sample before amplification.Alternatively, families that only have a single member sequence can be excluded from further analysis.
[0159] Nucleotide variations in sequenced nucleic acids can be determined by comparing the sequenced nucleic acids to a reference sequence. The reference sequence is often known, e.g., the entire or partial genomic sequence from a subject (e.g., the entire genomic sequence of a human subject). The reference sequence can be, for example, hG19 or hG38. The sequenced nucleic acid can represent a sequence determined directly for a nucleic acid in a sample, as described above, or can represent a consensus sequence of sequences of amplification products of such nucleic acids. Comparison can be performed at one or more designated positions on the reference sequence. A subset of sequenced nucleic acids can be identified that includes a position that matches a designated position of the reference sequence when the respective sequences are maximally aligned. Within such a subset, it can be determined which, if any, sequenced nucleic acids include a nucleotide variation at a designated position, and which, if any, include a reference nucleotide (i.e., the same as in the reference sequence), if necessary. If the number of sequenced nucleic acids in the subset that includes a nucleotide variant exceeds a selected threshold, a variant nucleotide can be called at the designated position. The threshold value may be a simple number, such as at least 1, 2, 3, 4, 5, 6, 7, 9, or 10, of the sequenced nucleic acids in the subset that includes the nucleotide variant, or may be a ratio, such as at least 0.5, 1, 2, 3, 4, 5, 10, 15, or 20, of the sequenced nucleic acids in the subset that includes the nucleotide variant, among other possibilities. Comparisons can be repeated for any designated position of interest in the reference sequence. Sometimes comparisons can be performed for designated positions occupying at least about 20, 100, 200, or 300 consecutive positions on the reference sequence, such as about 20-500 or about 50-300 consecutive positions.
[0160] Further details regarding nucleic acid sequencing, including the formats and applications described herein, can be found in, for example, Levy et al., Annual Review of Genomics and Human Genetics, 17:95-115. (2016), Liu et al., J. of Biomedicine and Biotechnology, Volume 2012, Article ID 251364:1-11 (2012), Voelkerding et al., Clinical Chem., 55: 641-658(2009), MacLeanet al., Nature Rev. Microbiol., 7: 287-296 (2009), Astier et al.,J Am Chem Soc.,128(5):1705-10 (2006), U.S. Patent No. 6,210,891, U.S. Patent No. 6,258,568, U.S. Patent No. 6,833,246, U.S. Patent No. 7,115,400, U.S. Patent No. 6,969,488, U.S. Patent No. 5,912,148, U.S. Patent No. 6,130,073, U.S. Patent No. 7,169,560, U.S. Patent No. 7,282,337, U.S. Patent No. 7,482,120, U.S. Patent No. 7,501,245 Nos. 6,818,395, 6,911,345, 7,501,245, 7,329,492, 7,170,050, 7,302,146, 7,313,308, and 7,476,503, each of which is incorporated herein by reference in its entirety.
[0161] Comparator Results
[0162] The microsatellite and / or other repeat nucleic acid instability status of a given subject determined according to the methods disclosed herein is typically compared to a database of comparator results from a reference population to identify customized or targeted therapies for the subject. In some embodiments, the microsatellite and / or other repeat nucleic acid instability status of the test subject and the comparator results are measured, for example, across the entire genome or exome, while in other embodiments, these markers are measured, for example, based on a subset or target region of the genome or exome, and extrapolated as needed to determine, for example, microsatellite instability for the entire genome or exome. Typically, the reference population includes patients with the same cancer type as the test subject, and / or patients who are undergoing or have undergone the same therapy as the test subject. In some embodiments, the test subject microsatellite and / or other repeat nucleic acid instability status and the comparator microsatellite and / or other repeat nucleic acid instability status are measured by determining the number or weight of mutations in a given or selected set of genes or genomic regions. Essentially any gene (e.g., cancer gene) can be selected for such analysis as needed. In certain of these embodiments, the selected genes or genomic regions include at least about 50, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1,500, 2,000 or more selected genes or genomic regions. In some of these embodiments, the selected genes or genomic regions optionally include one or more genes listed in Table 1. [Table 1]
[0163] Cancer and other diseases
[0164] In certain embodiments, the methods and systems disclosed herein are used to identify customized therapies for treating a given disease, disorder or condition in a patient. Typically, the disease of interest is some type of cancer. Non-limiting examples of such cancers include biliary tract cancer, bladder cancer, transitional cell carcinoma, urothelial carcinoma, brain cancer, glioma, astrocytoma, breast cancer, metaplastic carcinoma, cervical cancer, cervical squamous cell carcinoma, rectal cancer, colorectal cancer, colon cancer, hereditary nonpolyposis colorectal cancer, colorectal adenocarcinoma, gastrointestinal stromal tumor (GIST), endometrial cancer, endometrial stromal sarcoma, esophageal cancer, esophageal squamous cell carcinoma, esophageal adenocarcinoma, ocular melanoma, uveal melanoma, gallbladder cancer, gallbladder adenocarcinoma, renal cell carcinoma, renal clear cell carcinoma, transitional cell carcinoma, urothelial carcinoma, Wilms' tumor, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic (CLL), chronic myelogenous (CML), chronic myelomonocytic (CMML). , liver cancer, hepatoma, hepatocellular carcinoma, cholangiocarcinoma, hepatoblastoma, lung cancer, non-small cell lung cancer (NSCLC), mesothelioma, B cell lymphoma, non-Hodgkin's lymphoma, diffuse large B cell lymphoma, mantle cell lymphoma, T cell lymphoma, non-Hodgkin's lymphoma, precursor T lymphoblastic lymphoma / leukemia, peripheral T cell lymphoma, multiple myeloma, nasopharyngeal carcinoma (NPC), neuroblastoma, oropharyngeal carcinoma, oral squamous cell carcinoma, osteosarcoma, ovarian cancer, pancreatic cancer, pancreatic ductal adenocarcinoma, pseudopapillary neoplasm, acinar cell carcinoma, prostate cancer, prostate adenocarcinoma, skin cancer, melanoma, malignant melanoma, cutaneous melanoma, small intestine cancer, gastric cancer, gastrointestinal stromal tumor (GIST), uterine cancer, and uterine sarcoma.
[0165] Non-limiting examples of other genetic diseases, disorders, or conditions that may be optionally evaluated using the methods and systems disclosed herein include achondroplasia, alpha-1 antitrypsin deficiency, antiphospholipid syndrome, autism, autosomal dominant polycystic kidney disease, Charcot-Marie-Tooth (CMT), meowing, Crohn's disease, cystic fibrosis, Dercum's disease, Down's syndrome, Duane's syndrome, Duchenne muscular dystrophy, factor V Leiden thrombophilia, familial hypercholesterolemia, familial Mediterranean fever, fragile X syndrome, Gaucher disease, Hemochromatosis, hemophilia, holoprosencephaly, Huntington's disease, Klinefelter's syndrome, Marfan's syndrome, myotonic dystrophy, neurofibromatosis, Noonan's syndrome, osteogenesis imperfecta, Parkinson's disease, phenylketonuria, Poland's anomaly, porphyria, progeria, retinitis pigmentosa, severe combined immunodeficiency (SCID), sickle cell disease, spinal muscular atrophy, Tay-Sachs, thalassemia, trimethylaminuria, Turner's syndrome, palatocardiofacial syndrome, WAGR syndrome, Wilson's disease, or the like.
[0166] Customized Therapy and Related Administration
[0167] In some embodiments, the methods disclosed herein relate to identifying and administering a customized therapy to patients with a given microsatellite and / or other repeat nucleic acid instability status. Essentially any cancer therapy (e.g., surgical therapy, radiation therapy, chemotherapy, and / or the like) is included as part of these methods. Typically, the customized therapy includes at least one immunotherapy (or immunotherapeutic agent). Immunotherapy generally refers to a method of enhancing immune response to a given cancer type. In certain embodiments, immunotherapy refers to a method of enhancing T cell response to tumor or cancer.
[0168] In some embodiments, the immunotherapy or immunotherapeutic agent targets immune checkpoint molecules. Certain tumors can evade the immune system by exploiting immune checkpoint pathways. Thus, targeting immune checkpoints has emerged as an effective approach to combat the ability of tumors to evade the immune system and to activate antitumor immunity against certain cancers. Pardoll, Nature Reviews Cancer, 2012, 12:252-264.
[0169] In certain embodiments, the immune checkpoint molecule is an inhibitory molecule that reduces the signal involved in T cell response to antigen. For example, CTLA4 is expressed on T cells and is involved in downregulation of T cell activation by binding to CD80 (also known as B7.1) or CD86 (also known as B7.2) on antigen presenting cells. PD-1 is another inhibitory checkpoint molecule expressed on T cells. PD-1 limits the activity of T cells in peripheral tissues during inflammatory responses. In addition, the ligand of PD-1 (PD-L1 or PD-L2) is commonly upregulated on the surface of many different tumors, resulting in downregulation of anti-tumor immunity in the tumor microenvironment. In certain embodiments, the inhibitory immune checkpoint molecule is CTLA4 or PD-1. In other embodiments, the inhibitory immune checkpoint molecule is a ligand of PD-1, such as PD-L1 or PD-L2. In other embodiments, the inhibitory immune checkpoint molecule is a ligand of CTLA4, such as CD80 or CD86. In other embodiments, the inhibitory immune checkpoint molecule is lymphocyte activation gene 3 (LAG3), killer cell immunoglobulin-like receptor (KIR), T cell membrane protein 3 (TIM3), galectin 9 (GAL9), or adenosine A2a receptor (A2aR).
[0170] Antagonists targeting these immune checkpoint molecules can be used to enhance antigen-specific T cell responses to certain cancers. Thus, in certain embodiments, the immunotherapy or immunotherapeutic agent is an inhibitory immune checkpoint molecule antagonist. In certain embodiments, the inhibitory immune checkpoint molecule is PD-1. In certain embodiments, the inhibitory immune checkpoint molecule is PD-L1. In certain embodiments, the inhibitory immune checkpoint molecule antagonist is an antibody (e.g., a monoclonal antibody). In certain embodiments, the antibody or monoclonal antibody is an anti-CTLA4, anti-PD-1, anti-PD-L1, or anti-PD-L2 antibody. In certain embodiments, the antibody is a monoclonal PD-1 antibody. In some embodiments, the antibody is a monoclonal PD-L1 antibody. In certain embodiments, the monoclonal antibody is a combination of an anti-CTLA4 antibody and an anti-PD-1 antibody, an anti-CTLA4 antibody and an anti-PD-L1 antibody, or an anti-PD-L1 antibody and an anti-PD-1 antibody. In certain embodiments, the anti-PD-1 antibody is one or more of pembrolizumab (Keytruda®) or nivolumab (Opdivo®). In certain embodiments, the anti-CTLA4 antibody is ipilimumab (Yervoy®). In certain embodiments, the anti-PD-L1 antibody is one or more of atezolizumab (Tecentriq®), avelumab (Bavencio®), or durvalumab (Imfinzi®).
[0171] In certain embodiments, the immunotherapy or immunotherapeutic agent is an antagonist (e.g., an antibody) against CD80, CD86, LAG3, KIR, TIM3, GAL9, or A2aR. In other embodiments, the antagonist is a soluble version of an inhibitory immune checkpoint molecule, such as a soluble fusion protein comprising the extracellular domain of an inhibitory immune checkpoint molecule and the Fc domain of an antibody. In certain embodiments, the soluble fusion protein comprises the extracellular domain of CTLA4, PD-1, PD-L1, or PD-L2. In some embodiments, the soluble fusion protein comprises the extracellular domain of CD80, CD86, LAG3, KIR, TIM3, GAL9, or A2aR. In one embodiment, the soluble fusion protein comprises the extracellular domain of PD-L2 or LAG3.
[0172] In certain embodiments, the immune checkpoint molecule is a costimulatory molecule that amplifies the signal involved in T cell response to antigen. For example, CD28 is a costimulatory receptor expressed on T cells. When T cells bind to antigen through their T cell receptor, CD28 binds to CD80 (also known as B7.1) or CD86 (also known as B7.2) on antigen-presenting cells to amplify T cell receptor signaling and promote T cell activation. Because CD28 binds to the same ligands (CD80 and CD86) as CTLA4, CTLA4 can prevent or regulate the costimulatory signaling mediated by CD28. In certain embodiments, the immune checkpoint molecule is a costimulatory molecule selected from CD28, inducible T cell costimulatory molecule (ICOS), CD137, OX40 or CD27. In other embodiments, the immune checkpoint molecule is a ligand for a costimulatory molecule, including, for example, CD80, CD86, B7RP1, B7-H3, B7-H4, CD137L, OX40L, or CD70.
[0173] Agonists that target these costimulatory checkpoint molecules can be used to enhance antigen-specific T cell responses to certain cancers. Thus, in certain embodiments, the immunotherapy or immunotherapeutic agent is an agonist of a costimulatory checkpoint molecule. In certain embodiments, the agonist of a costimulatory checkpoint molecule is an agonist antibody, preferably a monoclonal antibody. In certain embodiments, the agonist antibody or monoclonal antibody is an anti-CD28 antibody. In other embodiments, the agonist antibody or monoclonal antibody is an anti-ICOS, anti-CD137, anti-OX40, or anti-CD27 antibody. In other embodiments, the agonist antibody or monoclonal antibody is an anti-CD80, anti-CD86, anti-B7RP1, anti-B7-H3, anti-B7-H4, anti-CD137L, anti-OX40L, or anti-CD70 antibody.
[0174] Therapeutic options for treating a particular genetic disease, disorder or condition other than cancer are generally known to those of skill in the art and will be apparent given the particular disease, disorder or condition under consideration.
[0175] In certain embodiments, the customized therapy described herein is typically administered parenterally (e.g., intravenously or subcutaneously). Pharmaceutical compositions containing immunotherapeutic agents are typically administered intravenously. Certain therapeutic agents are administered orally. However, the customized therapy (e.g., immunotherapeutic agent, etc.) can also be administered by any method known in the art, including, for example, buccal, sublingual, rectal, vaginal, urethral, topical, intraocular, intranasal, and / or intraauricular, and this administration can include tablets, capsules, granules, aqueous suspensions, gels, sprays, suppositories, salves, ointments, or the like.
[0176] System and computer readable medium
[0177] The present disclosure also provides various systems and computer program products or machine-readable media. In some embodiments, for example, the methods described herein are at least partially performed or facilitated as needed using systems, distributed computing hardware and applications (e.g., cloud computing services), electronic communication networks, communication interfaces, computer program products, machine-readable media, electronic storage media, software (e.g., machine-executable code or logic instructions), and / or the like. By way of example, FIG. 2 provides a schematic diagram of an exemplary system suitable for use in carrying out at least aspects of the methods disclosed herein. As shown, the system 200 includes at least one controller or computer, such as a server 202 (e.g., a search engine server), including a processor 204 and a memory, storage device, or memory component 206, and one or more other communication devices 214 and 216 (e.g., client-side computer terminals, phones, tablets, laptops, other mobile devices, etc.) that are remote from the remote server 202 and in communication with the remote server 202 via an electronic communication network 212, such as the Internet or other internetwork. The communications devices 214 and 216 typically include, for example, an electronic display (e.g., an internet-enabled computer or the like) in communication with the server 202 computer over the network 212, the electronic display including a user interface (e.g., a graphical user interface (GUI), a web-based user interface, and / or the like) for displaying results upon execution of the methods described herein. In certain embodiments, the communications network also encompasses the physical transfer of data from one location to another, for example, using a hard drive, thumb drive, or other data storage mechanism.The system 200 also includes a program product 208 stored on a computer or machine readable medium, such as one or more of various types of memory, such as memory 206 of the server 202, readable by the server 202, for example, to facilitate a guided search application or the like executable by one or more other communication devices, such as 214 (shown generally as a desktop or personal computer) and 216 (shown generally as a tablet computer). In some embodiments, the system 200 also optionally includes at least one database server, such as server 210, associated with an online website on which data (e.g., control sample or comparator result data, indexed customized therapies, etc.) is stored, searchable directly or by the search engine server 202. The system 200 also optionally includes one or more other servers located remotely from the server 202, each of which is optionally associated with one or more database servers 210, either remotely located or located in the same area as each of the other servers. These other servers can beneficially provide services to remote users and can facilitate geographically distributed operations.
[0178] As will be appreciated by those skilled in the art, the memory 206 of the server 202 may include volatile and / or non-volatile memory, including, for example, RAM, ROM, and magnetic or optical disks, among others, as appropriate. Although illustrated as a single server, those skilled in the art will also appreciate that the illustrated configuration of the server 202 is provided merely as an example, and that other types of servers or computers configured according to various other methodologies or configurations may also be used. The server 202 shown diagrammatically in FIG. 2 represents a server or a server cluster or a server farm, and is not limited to any individual physical server. The server site may be deployed as a server farm or server cluster managed by a server hosting provider. The number of servers and their structure and configuration may be increased based on the usage, demand, and capacity requirements for the system 200. As will also be appreciated by those skilled in the art, for example, the other user communication devices 214 and 216 in these embodiments may be laptops, desktops, tablets, personal digital assistants (PDAs), mobile phones, servers, or other types of computers. As known and understood by those skilled in the art, the network 212 may include the Internet, an intranet, a telecommunications network, an extranet, or the World Wide Web of multiple computers / servers in communication with one or more other computers via a communications network and / or via portions of a local or other area network.
[0179] Further, as will be appreciated by those skilled in the art, the exemplary program product or machine-readable medium 208, where appropriate, is in the form of microcode, programs, cloud computing formats, routines and / or symbolic languages that effect one or more sets of ordered operations that control and direct the operation of the hardware. According to an exemplary embodiment, the program product 208 also need not be entirely in volatile memory, but may be selectively loaded as required according to various methodologies as will be known and understood by those skilled in the art.
[0180] Further, as will be appreciated by those skilled in the art, the term "computer-readable medium" or "machine-readable medium" refers to any medium that participates in providing instructions to a processor for execution. By way of example, the term "computer-readable medium" or "machine-readable medium" encompasses distribution media, cloud computing formats, intermediate storage media, computer execution memory, and any other media or devices that can store the program product 208 that performs the functions and processes of the various embodiments of the present disclosure, for example, for reading by a computer. The "computer-readable medium" or "machine-readable medium" can take many forms, including, but not limited to, non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical or magnetic disks. Volatile media include, for example, dynamic memory, such as the main memory of a given system. Transmission media include coaxial cables, copper wire and fiber optics, electrical wires including, for example, bus bars, and the like. Transmission media can also take the form of acoustic or light waves, such as those generated during radio wave and infrared data communications, among others. Exemplary forms of computer-readable media include a floppy disk, a flexible disk, a hard disk, a magnetic tape, a flash drive, or any other magnetic medium, a CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with a pattern of holes, RAM, PROM and EPROM, FLASH-EPROM, any other memory chip or cartridge, a carrier wave, or any other medium from which a computer can read.
[0181] The program product 208 is copied from the computer readable medium to a hard disk or similar intermediate storage medium as needed. When attempted, the program product 208 or portions thereof are loaded from their distribution medium, their intermediate storage medium, or the like into one or more computer executable memories that configure the computer to operate according to the functions or methods of the various embodiments. All such operations are well known to those skilled in the art of, for example, computer systems.
[0182] To further illustrate, in certain embodiments, the present application provides a system including one or more processors and one or more memory components in communication with the processor. The memory components typically include one or more instructions that, when executed, provide information that causes the processor to display (e.g., via communication devices 214, 216 or the like) sequence information, microsatellite and / or other repeat nucleic acid instability status, comparator results, customized therapy, and / or the like, and / or receive information from other system components and / or from a system user (e.g., via communication devices 214, 216 or the like).
[0183] In some embodiments, the program product 208, when executed by the electronic processor 204, includes at least the steps of (i) receiving sequence information from a population of microsatellite loci in a sample; (ii) quantifying from the sequence information a number of different repeat lengths present at each of a plurality of microsatellite loci to generate a site score for each of the plurality of microsatellite loci; (iii) comparing the site score for a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; and (iv) comparing the site score for a given microsatellite locus to a site-specific trained threshold for the given microsatellite locus if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus. The method comprises the steps of: (a) calling a microsatellite locus as unstable to generate a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; (b) classifying the MSI status of the sample as unstable to identify an unstable sample if the microsatellite instability score exceeds a population-trained threshold for a population of microsatellite loci in the sample; and, optionally, (c) comparing the microsatellite instability score of the unstable sample with one or more comparator results, wherein a substantial match between the microsatellite instability score of the unstable sample and the comparator results indicates a predicted response to the subject's therapy.
[0184] System 200 also typically includes additional system components that are configured to perform various aspects of the methods described herein. In some of these embodiments, one or more of these additional system components are located remotely from remote server 202 and are in communication with remote server 202 via electronic communications network 212, while in other embodiments, one or more of these additional system components are located in the same local area and are in communication with server 202 (i.e., in the absence of electronic communications network 212) or are in direct communication, for example, with desktop computer 214.
[0185] In some embodiments, for example, additional system components include a sample preparation component 218 that is operably connected to the controller 202 (either directly or indirectly (e.g., via electronic communication network 212)). The sample preparation component 218 is configured to prepare nucleic acids in the sample to be amplified and / or sequenced by a nucleic acid amplification component (e.g., a thermal cycler, etc.) and / or a nucleic acid sequencer (e.g., prepare a library of nucleic acids). In certain of these embodiments, the sample preparation component 218 is configured to separate nucleic acids in the sample from other components for attaching one or more adapters that include barcodes to the nucleic acids as described herein, for selectively enriching one or more regions from a genome or transcriptome prior to sequencing, and / or the like.
[0186] In certain embodiments, the system 200 also includes a nucleic acid amplification component 220 (e.g., a thermal cycler, etc.) operably connected (directly or indirectly, e.g., via electronic communication network 212) to the controller 202. The nucleic acid amplification component 220 is configured to amplify nucleic acids in a sample from a subject. For example, the nucleic acid amplification component 220 is optionally configured to amplify selectively enriched regions from a genome or transcriptome in the sample as described herein.
[0187] The system 200 also typically includes at least one nucleic acid sequencer 222 operably connected (directly or indirectly (e.g., via electronic communication network 212)) to the controller 202. The nucleic acid sequencer 222 is configured to provide sequence information from nucleic acids (e.g., amplified nucleic acids) in a sample from a subject. Essentially any type of nucleic acid sequencer can be adapted for use in these systems. For example, the nucleic acid sequencer 222 is optionally configured to perform pyrosequencing, single molecule sequencing, nanopore sequencing, semiconductor sequencing, sequencing by synthesis, sequencing by ligation, sequencing by hybridization, or other sequencing techniques with the nucleic acids to generate sequencing reads. Optionally, the nucleic acid sequencer 222 is configured to group sequence reads into families of sequence reads, each family including sequence reads generated from nucleic acids in a given sample. In some embodiments, the nucleic acid sequencer 222 generates sequencing reads using clonal single molecule arrays derived from a sequencing library. In certain embodiments, the nucleic acid sequence sequencer 222 includes at least one chip having an array of microwells for sequencing a sequencing library to generate sequencing reads.
[0188] To facilitate full or partial system automation, the system 200 also typically includes a material transfer component 224 that is operably connected (directly or indirectly (e.g., via electronic communications network 212)) to the controller 202. The material transfer component 224 is configured to transfer one or more materials (e.g., nucleic acid samples, amplicons, reagents, and / or the like) to and / or from the nucleic acid sequencer 222, the sample preparation component 218, and the nucleic acid amplification component 220.
[0189] Further details regarding computer systems and networks, databases, and computer program products can be found in, for example, Peterson, Computer Networks: A Systems Approach, Morgan Kaufmann,5th Ed. (2011), Kurose, Computer Networking: ATop-Down Approach, Pearson, 7 th Ed. (2016), Elmasri, Fundamentals ofDatabase Systems, Addison Wesley, 6thEd.(2010), Coronal, Database Systems:Design, Implementation, & Management,Cengage Learning, 11 th Ed. (2014), Tucker, Programming Languages, McGraw-Hill Science / Engineering / Math, 2nd Ed. (2006), and Rhoton, Cloud Computing Architected: Solution Design Handbook, Recursive Press (2011), each of which references are incorporated herein by reference in their entirety. EXAMPLES
[0190] Example 1 Using non-tumor samples as background, high-grade MSI (MSI-H) samples were simulated in silico with variable tumor fraction and number of unstable sites. The distribution observed in a cohort of 3000 samples of different cancer types was used as the prior for the number of unstable sites. The analysis was performed with a limit of detection (LoD) of 0.2% of tumor content. ) demonstrated a sensitivity of 94%. The expected specificity of the method for determining MSI status according to the embodiments described herein on non-tumor donor samples was 99.999%. Comparison of these results with standard or conventional PCR-based MSI assessment showed 100% concordance across a tumor content range of 1.4-15%. In addition, the performance of this analysis was tested on 155 clinical samples from three cancer types for which standard PCR-based assessment of MSI status was available (10 MSI-H, 145 microsatellite stable (MSS)). MSI calls generated according to the embodiments described herein showed 100% concordance with standard PCR-based MSI assessment.
[0191] Example 2 The digital sequencing clinical platform (Guardant Health, Inc., Redwood City, CA, USA) was used to assess the MSI status of 82 samples. The digital sequencing platform is an NGS panel of cancer-associated genes that utilizes high-quality sequencing of cell-free DNA (which may contain circulating tumor DNA) isolated by a simple noninvasive blood draw. Digital sequencing utilizes pre-sequencing preparation of a digital library of individually tagged cfDNA molecules combined with post-sequencing bioinformatics reconstruction to eliminate nearly all false positives. Targeted sequencing of cfDNA in the samples was used to obtain sequence information. Site scores (ΔAIC) were determined for the 61 most informative microsatellite loci in each sample. The tumor fraction of the samples ranged from 0.5% to 15%. Site scores were compared to the corresponding site-specific trained thresholds for each sample to identify unstable microsatellite loci in each sample. The number of unstable microsatellite loci identified in a given sample was used as the microsatellite instability score for that particular sample (i.e., MSI sample score). A population-trained threshold for 61 microsatellite loci in a sample was determined, where a microsatellite instability score greater than or equal to 5 predicted the sample to be classified as highly MSI (MSI-H), while a microsatellite instability score less than or equal to 4 predicted the sample to be classified as microsatellite stable (MSS). Nine of the 82 samples were classified as MSI-H. The remaining 73 samples were classified as MSS. In all 82 samples, the predicted stability status matched the expected stability status. The MSI status was confirmed based on orthogonal validation.
[0192] Example 3 Introduction
[0193] Microsatellite instability (MSI) is a guideline-recommended biomarker for prognostic significance in various tumor types as well as predictive significance for treatment with immune checkpoint inhibitors. Traditionally, microsatellite instability detection has relied on testing of tumor tissue by PCR or immunohistochemistry. More recently, next-generation sequencing (NGS) methods have been developed, but these methods also rely on the availability of tumor tissue. In contrast, plasma-based MSI detection methods could provide a non-invasive, real-time assessment of MSI status. Guardant Health's large panel cell-free DNA (cfDNA) NGS assay evaluates 500 cancer-related genes to identify genomic alterations and tumor mutation burden (TMB). In addition to single nucleotide variants (SNVs), indels, copy number amplifications (CNAs), fusions, and TMB, the panel can detect high microsatellite instability (high MSI) status based on somatic alterations at >1,000 MSI sites. The analytical validation presented in this example has four major components that serve to determine the performance of a cfDNA NGS assay of 500 cancer-associated genes for the detection of high-grade MSI: accuracy, limit of detection (LoD), precision, and limit of blank (LoB).
[0194] method
[0195] Accuracy analysis used 258 samples from three sources and compared MSI status predicted by cfDNA NGS assays of 500 cancer-related genes with true based on tissue MSI status determined by orthogonal methods. Thirty-six participant samples with tissue MSI status (true: tissue MSI status) and 121 healthy donors (true: microsatellite stable, MSS) were used, as well as 101 samples sequenced by cfDNA NGS assays of 500 cancer-related genes (large panel assay) and cfDNA GNS assays of 73 cancer-related genes (small panel, MSI status as true). Reproducibility and repeatability analysis used two sets of replicates (56 replicates in total). MSI status and MSI scores were compared within and between trials. LoD was obtained both by simulation. LoB was calculated using healthy donor samples and known MSS samples.
[0196] result
[0197] 1.Accuracy analysis Twelve of the 13 high-grade MSI samples based on tissue MSI status were called high-grade MSI by the large panel assay. All MSS / low-grade MSI (MSI-L) samples were correctly detected (Table 2). When restricted to the microsatellite regions covered by the small panel assay (approximately 90 sites), the 12 samples detected as high-grade MSI by the large panel assay also met the small panel assay threshold for being called high-grade MSI. [Table 2]
[0198] 2.LoD analysis
[0199] Simulations with >1,000 sites used to detect MSI across five tumor fractions ranging from 0.05% to 1% show an LoD of 0.1% (Figure 3).
[0200] 3. Detection of high-grade MSI with a cfDNA NGS assay of 500 cancer-associated genes is reproducible and repeatable
[0201] 24 high-grade MSI repeats were tested in two runs. All repeats are detected as high-grade MSI in the cfDNA NGS assay of 500 cancer-related genes. The MSI numerical score is ±4 within and between runs (Figure 4A). 10 MSS / low-grade MSI samples have 2–3 repeats (32 repeats in total) tested on the same flow cell. All repeats are detected as MSS / low-grade MSI in the cfDNA NGS assay of 500 cancer-related genes. The MSI score is ±3 within each sample (Figure 4B).
[0202] 4.LoB analysis
[0203] 121 healthy donor samples and 25 volunteer samples with known MSS status were used for LoB analysis. All 146 samples were correctly classified as MSS / low-grade MSI by the large panel assay, indicating a false positive rate of 0%.
[0204] 5. Tumor percentage and MSI score
[0205] MSI scores were plotted against maximum mutant allele fraction (MAF) of somatic calls in over 2,000 large panel assay samples (Fig. 5A and B), revealing that tumor fraction (measured by MAF) did not correlate with MSI status.
[0206] conclusion
[0207] Detection of high-grade MSI with the 500 cancer-associated genes cfDNA NGS assay showed high sensitivity (>90%) and specificity (100%). Repeatability and reproducibility within and across trials were high. The LoD for detection of high-grade MSI was 0.1% MAF. LoB studies showed a 0% false positive rate. The 500 cancer-associated genes cfDNA NGS assay provides reliable prediction of high-grade MSI status in cfDNA without the need for tissue samples, which should provide treatment value to physicians.
[0208] Example 4 Introduction
[0209] Microsatellite instability (MSI) is a National Comprehensive Cancer Network (NCCN) Clinical Practice Guideline-recommended biomarker in at least nine cancer types—cervical, cholangiocarcinoma, colorectal, endometrial, esophageal and esophagogastric, gastric, ovarian, pancreatic, and prostate cancers (1–9)—because of its importance as a predictive biomarker for response to immune checkpoint blockade (ICB) as exemplified by the pan-cancer approval of pembrolizumab (10, 11). Detection of MSI in a patient may also alert clinicians to evaluate asymptomatic family members of the patient for hereditary cancer risk.
[0210] MSI is a typical manifestation of DNA mismatch repair deficiency (dMMR), which results in a dramatic increase in mutation rates throughout the genome, including the gain and / or loss of nucleotides within repeated motifs known as microsatellite tracts, from which the entity is named. MSI is most prevalent in endometrial, colorectal and gastroesophageal cancers, where it may be a sequela of sporadic mutations in MMR-related genes or a manifestation of Lynch syndrome, an inherited cancer predisposition syndrome most often caused by germline mutations in MLH1, MSH2, MSH6, PMS2 or EPCAM (12). However, despite its increased prevalence in these cancer types, landscape analyses have shown that MSI is also present at non-negligible rates in most other solid tumors, including common tumor types such as lung, prostate and breast cancer (13).
[0211] Recent studies have shown that MSI predicts clinical benefit from ICB with PD-1 / PD-L1 inhibitors, leading to the approval of these agents in several indications when MSI is present, such as nivolumab ± ipilimumab for high-grade MSI (MSI-H, positive for MSI) metastatic colorectal cancer and pembrolizumab for unresectable or metastatic MSI-H solid tumors after progression on previously approved therapy (14). In addition to its value as a predictive biomarker for ICB benefit, MSI also has prognostic significance, most notably in colorectal cancer (CRC), where testing is recommended in clinical practice guidelines for all patients (3, 15).
[0212] Nowadays, MSI testing is most commonly performed by polymerase chain reaction (PCR) and / or immunohistochemistry (IHC) analysis of tumor tissue specimens. The former evaluates the five standard microsatellite loci originally recommended by the Bethesda panel (16, 17), compares their length in tumor DNA compared to the germline genotype evaluated in matched nontumor DNA, and uses the length instability of each microsatellite tract as direct evidence of MSI. However, this limited microsatellite panel was developed primarily for CRC and has more limited sensitivity in other cancer types (18). In contrast, IHC methods evaluate the levels of four MMR proteins, and the absence of expression of one or more (MMR-deficient, dMMR) strongly correlates with MSI status. However, approximately 5-11% of MSI-H cases demonstrate intact MMR staining and localization (proficient MMR, pMMR), due to the retention of antigenicity and intracellular trafficking of otherwise nonfunctional proteins (19). Recent publications (20, 21) have demonstrated that next-generation sequencing (NGS) can also accurately characterize MSI status in tumors, thereby enabling comprehensive profiling of not only MSI status but also targetable genomic biomarkers with a single NGS assay.
[0213] Despite recommendations across many cancer types in the NCCN guidelines and accompanying FDA-approved treatment options, current MSI testing rates outside of CRC and gastroesophageal cancer remain very low (22). Even in CRC, where MSI testing recommendations have been implemented since 2005 (17, 23), less than 50% of patients are tested (24), resulting in missed opportunities for ICB procedures and failure to recognize patients who may have high family risk for cancer. Although multifactorial, such MSI undergenotyping is often due to barriers associated with tissue acquisition and complex testing recommendations / algorithms. For example, testing of archival diagnostic specimens can incur significant delays associated with locating and acquiring this material, and may also result in inaccurate assessment of MSI status due to tumor extension and / or heterogeneity. Similarly, testing of freshly obtained tissue specimens can also result in significant delays associated with biopsy scheduling and failure, with the additional attendant risks and costs due to procedural complexity. Thus, invasive tissue acquisition procedures are contraindicated for many patients who are heavily pretreated and / or frail. Additionally, the exploding number of biomarkers and the diversity of testing options creates daunting complexity for already overburdened physicians.
[0214] Cell-free circulating tumor DNA (ctDNA) assays ("liquid biopsies") successfully address such obstacles in genotyping for many of the indications by enabling minimally invasive profiling of contemporaneous tumor DNA. Liquid biopsies therefore broaden the opportunity for patients to access standard-of-care targeted therapies, including ICB, by identifying patients whose tumors harbor biomarkers of interest that cannot be identified otherwise due to tissue sampling limitations, and by doing so more quickly than typical tissue testing (25). Furthermore, comprehensive liquid biopsies can provide all guideline-recommended somatic genomic biomarker information for all adult solid tumors in a single test. In this study, we sought to enhance the utility of a previously validated ctDNA-based genotyping test with the addition of MSI detection. We describe here the design and validation of MSI assessment on this platform, report its performance in the largest yet described ctDNA-tissue MSI validation cohort (n=1145), and evaluate response prediction in 16 advanced gastric cancer patients treated with ICB. We also report here the MSI-H landscape of >28,000 consecutive solid tumor patients tested in a Clinical Laboratory Improvement Amendment (CLIA)-certified, College of American Pathologists (CAP)-vetted, and New York State Department of Health-licensed laboratory.
[0215] material and method
[0216] 1. Microsatellite loci selection
[0217] The Guardant Health small panel cell-free DNA (cfDNA) NGS assay is a 74-gene panel (26, 27) previously validated for the detection of SNVs, indels, CNAs, and fusions in all guideline-recommended indications for advanced solid tumors. The assay initially incorporated 99 putative microsatellite loci consisting of seven or more short tandem repeats (STRs) in length, selected to include sites prone to instability across multiple tumor types, including three of the five Bethesda panel sites (BAT-25, BAT-26, and NR-21). The remaining two Bethesda sites (NR-24 and MONO-27) were not included due to extremely low mappability of the regions. Sequencing data from a set of 84 healthy donor samples was used to assess coverage and noise profiles at these sites to exclude uninformative sites from the final MSI detection algorithm.
[0218] 2. Model Description
[0219] MSI detection is based on incorporating the observed read sequences with molecular barcoding information into a single probability model, which compares the likelihood of the observed data assuming PCR and sequencing noise with the likelihood of the observed data assuming somatic MSI instability. Each individual site is scored independently using the Akaike Information Criterion (AIC) (28). The AIC model generates a locus score (ranging from 0 to infinity) that reflects the likelihood that the variability observed at any given microsatellite locus is due to biological instability relative to noise, and a locus is considered unstable if its score (i.e., site score) is above a site-specific trained threshold. The number of affected loci is calculated across the final 90 sites, and a sample is called positive if the number of unstable loci ("MSI score") is above a population trained threshold (n=6). The thresholds for individual loci and for the total MSI score per sample were established using permutation-based simulations on data from healthy donor samples, varying the frequency of molecules with different repeat lengths and error parameters at individual loci as well as the total number of unstable loci in the simulation samples. Simulations from this approach were used here to examine 100,000 combinations of microsatellite length and number of unstable loci. This allows for the evaluation of a diverse landscape of scenarios, some of which may not be represented in the unsimulated dataset. The algorithm does not distinguish between microsatellite stable (MSS) and low-grade MSI (a category defined by the observation of a single unstable Bethesda locus using PCR methods), but groups these into a single category. This is based on previous reports that MSI-L status is not a distinct phenotype, but an artifact of testing a small number of microsatellites, and thus previously characterized MSI-L samples mimic the MSS phenotype within the total MSI load when a large number of microsatellite loci are tested.
[0220] 3. Sample
[0221] A set of 84 healthy donor samples was used to develop and train the MSI algorithm as well as simulated data. The clinical validation study included 1145 archival samples (residual plasma and / or cell-free DNA) collected and processed as part of routine standard of care clinical trials in the Guardant Health CLIA laboratory as previously described (26) or archival patient plasma samples collected in EDTA blood collection tubes. Twenty healthy donor samples were also used in the analytical specificity study. Design samples used in the analytical validation study included cell line supernatants and cfDNA pools extracted from healthy donor plasma. Cell-free DNA prepared from culture supernatants from the following cell lines was used (ATCC, Inc.): KM12, NCI-H660, HCC1419, NCI-H2228, NCI-H1650, NCI-H1648, NCI-H1975, NCI-H1993, NCIH596, HCC78, GM12878, and MCF-7. cfDNA isolated from cell line culture supernatants mimics the fragment size and mechanism of extracellular release (29), library conversion, and sequencing characteristics of patient-derived cfDNA, while also providing a renewable source of well-defined material in sufficient quantities to support advanced research material requirements, such as detection limits and precision.
[0222] 4. Sample Processing and Bioinformatics Analysis
[0223] Cell-free DNA was extracted from plasma samples or cell line supernatants (QIAmp Circulating Nucleic Acid Kit, Qiagen, Inc.), and ∼30 ng of extracted cfDNA was labeled with nonrandom oligonucleotide barcodes (IDT, Inc.), followed by library preparation, hybrid capture enrichment (Agilent Technologies, Inc.), and sequencing by paired-end synthesis (NextSeq 500 / 550 or HiSeq 2500, Illumina, Inc.) as previously described (26). Bioinformatics analysis and variant detection were performed as previously described (26).
[0224] 5. Analytical Verification Methods
[0225] The analytical validation studies conducted were based on established CLIA, Nex-StoCT Working Group, and Association published guidelines for performance characteristics and validation principles. of Molecular Pathologist / CAP guidance. To determine the sensitivity of the assay for MSI status, cfDNA from cell line supernatants from an MSI-H cell line (KM12) (29) was diluted with cfDNA from a microsatellite stable (MSS) cell line (NCI-H660) (30, 31) and tested at both standard (30 ng) and low (5 ng) cfDNA inputs. The dilution series targeted maximum mutant allele fractions (max MAFs) of 0.03–2% for the 5 ng input and 0.01–1% for the 30 ng input. Target tumor fractions were validated using known germline variants specific to the titration and dilution materials. Evaluation of repeatability (precision within a run) and reproducibility (match between runs) was based on clinical and design model samples. Six of the clinical samples for accuracy (three MSI-H, and three MSS) were selected, which had max MAF values of 1-2%, corresponding to approximately 2-3 times the expected LoD at 5 ng. MSI assay specificity was determined by analyzing 20 healthy donor samples and 245 conjugate samples of known MSS.
[0226] 6. Clinical validation methods
[0227] Archived plasma or cfDNA from clinical samples from patients for whom standard-of-care tissue-based MSI test results were available were assayed using the ctDNA MSI algorithm (n=1145). Tissue-based MSI status was derived from IHC, PCR, or less frequently NGS. Clinical outcome data were extracted from patient medical records by the primary physician and de-identified.
[0228] 7. Landscape analysis of plasma MSI status from 28,459 advanced cancer patient samples
[0229] The cohort included 28,459 consecutive advanced cancer patient samples tested using a 73 cancer-associated gene cfDNA NGS assay (small panel) over the course of their clinical care. All analyses were performed according to an IRB-approved protocol using de-identified data. Prevalence of MSI-H in this cohort was assessed across 16 primary tumor types: bladder cancer, breast cancer, cholangiocarcinoma, colon adenocarcinoma, cancer of unknown primary, head and neck squamous cell carcinoma, hepatocellular carcinoma, lung adenocarcinoma, lung cancer not otherwise specified, lung squamous cell carcinoma, "other" cancer diagnoses, pancreatic adenocarcinoma, prostate adenocarcinoma, gastric adenocarcinoma, and uterine endometrial carcinoma.
[0230] 8. Statistics
[0231] Statistical analysis used the Student's t test for analysis of the number of variants per sample and the Fisher exact test for comparison of proportions. Lower and upper 95% confidence intervals (CIs) for binomial proportions were calculated using Wilson's score interval with continuity correction.
[0232] 9. Code of Ethics
[0233] This study was conducted using de-identified data in accordance with a protocol approved by the Quorum Institutional Review Board.
[0234] result
[0235] 1.MSI algorithm development
[0236] Traditional challenges of ctDNA genotyping using NGS include efficient molecular capture due to low input and low tumor fraction in circulation (26, 27), as well as correction of sequencing and other technical artifacts. MSI detection presents additional challenges as it requires 1) efficient molecular capture, sequencing and mapping of repetitive genomic regions that accurately represent MSI status; 2) error correction and variant detection within repetitive regions; and 3) differentiation of signal from MSI and strong PCR slippage artifacts due to non-MSI somatic mutations at sites typically affected by somatic instability. Indeed, technical PCR errors are typically at least an order of magnitude larger than typical sequencing error rates at homopolymeric sites, thus necessitating optimal use of repetitive site selection and molecular barcoding to achieve adequate signal-to-noise detection ratios across a large number of candidate microsatellite sites.
[0237] While tissue sequencing panels often contain sufficient informative microsatellite loci simply due to the large panel size and longer DNA fragment lengths (13, 32), the medium size of the ctDNA panel utilized here, and the short cell-free DNA (cfDNA) fragment lengths, require deliberate microsatellite selection and inclusion. To accomplish this, candidate sites were evaluated using an iterative approach informed by the literature and tissue sequencing compendium to perform pan-cancer MSI detection with minimal background noise. The list of candidate loci was further refined based on the performance criteria described above using cfDNA from healthy donors.
[0238] Based on performance evaluation in training healthy donor samples, informative loci were defined as those that were effectively captured, sequenced, and mapped, and were associated with little mutation in the MSS samples (shown in light gray in Figure 6A). Uninformative loci either could not be captured, sequenced, or mapped, resulting in inappropriate molecular representation (shown in black in Figure 6A), or demonstrated substantial mutation in the MSS samples, resulting in excess artifactual signal (shown in dark gray in Figure 6A). Interestingly, the BAT-25, BAT-26, and NR-21 Bethesda loci utilized in legacy MSI tissue testing (16, 17) and in some ctDNA panels (33) performed poorly compared to other candidates and were excluded from the final marker set (shown in Figure 6A with arrows).
[0239] Using this approach, 90 microsatellite loci, including 89 mononucleotide and a single trinucleotide repeat, all with a repeat length of 7 or greater, were selected for inclusion in the final test version. Assessment of the unique molecular coverage distribution demonstrated that 65% of these loci had coverage greater than 0.5 times the median sample coverage.
[0240] In addition to effective molecular capture and mapping, MSI detection also entails highly accurate differentiation of cancer-associated signals from background noise due to sequencing and polymerase errors at the very low allele fractions where ctDNA is typically found (26, 27, 34). Importantly, the same repetitive genomic context that makes microsatellite candidates informative for MSI detection due to polymerase slippage during cell replication in vivo also makes them particularly susceptible to the same polymerase slippage during library preparation and sequencing in vitro, resulting in high technical noise levels. To address this, digital sequencing error correction was used to define true biological insertion-deletion events with high fidelity at microsatellite loci as previously described (26, 27). The digital sequencing platform is an NGS panel of cancer-associated genes that utilizes high-quality sequencing of cell-free DNA (which may contain circulating tumor DNA) isolated by a simple non-invasive blood draw. Digital sequencing utilizes pre-sequencing preparation of digital libraries of individually tagged cfDNA molecules combined with post-sequencing bioinformatics reconstruction to eliminate nearly all false positives.
[0241] Among these high background error repeats, digital sequencing allowed efficient and accurate reconstruction of microsatellite sequences of individual unique molecules present in the original patient blood sample, with a 100-fold reduction in sequencing errors per molecule compared to standard sequencing approaches (Figure 6B). Threshold simulations based on permutations of healthy donor samples were then used to establish site-specific and aggregate sample-level MSI status determination thresholds. Combining these site- and sample-specific thresholds with the effect of digital sequencing corrections, the false positive rate per sample was estimated to be approximately 10-7.3. In addition, titration simulations adjusted for the distribution of clinical input predicted robust MSI detection down to approximately 0.2% tumor fraction, with a marked decrease in detection efficiency thereafter. Therefore, samples with a circulating tumor fraction (defined by the maximum somatic variant allele fraction) of <0.2% were considered unevaluable for MSI status.
[0242] 2. Analytical validation study
[0243] To evaluate the analytical sensitivity of MSI detection, cfDNA from the supernatant of the MSI-H cell line KM12 was diluted into MSS cfDNA targeting five titration points, which included 15 independently processed replicates, bracketing the limit of detection (LoD) predicted by the computational simulation described above. Each titration series was analyzed at both the minimum acceptable cfDNA input of 5 ng, and the maximum most common cfDNA input of 30 ng. Using probit analysis, the 95% LoD (LOD95) was calculated to be 0.4% at the 5 ng input (Figure 7A) and 0.1% at the 30 ng input (Figure 7B).
[0244] To assess intermediate precision of the analysis, replicates of four different engineered materials were analyzed, two MSS and two MSI-H (Figure 7C). Across 499 replicates, categorical agreement for MSI status was 100% (499 / 499, 95% CI 99-100%), and the coefficient of variation for quantitative MSI scores was 6.3-7.2% for MSI-H samples (Table 3). Replicativity and input robustness were also assessed by testing replicates of MSS and MSI-H engineered materials at cfDNA inputs of 5ng, 10ng and 30ng, which also demonstrated 100% agreement (27 / 27, 95% CI 85-100%, Figure 8 and Table 4). Clinical accuracy was confirmed in replicates of 72 independent patient samples representing a range of MSI scores and tumor fractions processed across three independent batches, dates, operators, and reagent lots, demonstrating 100% qualitative agreement (72 / 72, 95% CI 94–100%) with coefficients of variation ranging from 2.0 to 15.2% for the underlying quantitative MSI scores (Table 5). [Table 3] [Table 4] [Table 5]
[0245] To assess analytical specificity, healthy donor plasma samples (distinctly different from those used for training), MSS conjugates, and MSS patient samples were analyzed for false MSI-H calls. Analytical specificity was 100% across healthy donor samples (20 / 20, 95% CI 83-100%), conjugate samples (245 / 245, 95% CI 98-100%), and patient samples (48 / 48, 95% CI 92-100%).
[0246] 3. Clinical validation studies
[0247] As no orthogonal cfDNA-based methods were available to use as a comparator, clinical accuracy was determined by comparing ctDNA MSI assessment with MSI status from medical records determined using standard of care tissue testing (a mix of IHC, PCR, and NGS methods) for 1,145 samples representing 40 different cancer types, 15 of which had at least five representative specimens (Figure 9). In 949 unique evaluable patients, ctDNA detected 87% (71 / 82, 95% CI 77–93%) of patients reported as MSI-H and 99.5% (863 / 867, 95% CI 98.7–99.8%) of patients reported as MSS / MSI-L, with an overall accuracy of 98.4% (934 / 949, 95% CI 97.3–99.1%) and a positive predictive value (PPV) of 95% (71 / 75, 95% CI 86–98%) (Figure 10C, Table 8). Consistent with the computational modeling studies, MSI-H detection was rare (0 / 19) in samples classified as non-evaluable due to low tumor fraction (Figure 10A), which explained 57% (16 / 28) of the ctDNA-tissue discordance observed in the entire set of unique patient samples (Tables 6–9). For samples with tumor fraction >1%, ctDNA PPA increased to 93% (54 / 58, 95% CI 82–98%, Table 9). [Table 6] [Table 7] [Table 8] [Table 9]
[0248] Interestingly, despite the high degree of correlation between IHC and PCR histologic testing reported in the literature (23, 35), the concordance between ctDNA and tissue MSI status here varied by tissue testing methodology (97.4% (450 / 462) by PCR, 98.0% (239 / 244) by NGS, and 83.0% (93 / 112) by IHC; Figure 10B and Tables 10 and 11). On further investigation, this discordance was noted to be due to both an increase in the tissue IHC-positive, ctDNA-negative population (2.4% by PCR, 2.0% by NGS, and 12.5% by IHC; Fisher's exact test p<0.001 for IHC-PCR and IHC-NGS) and an increase in the tissue IHC-negative, ctDNA-positive population (0.2% by PCR, 0% by NGS, and 4.5% by IHC; Fisher's exact test p<0.01 for both comparisons). These discrepancies led us to investigate whether IHC limitations could contribute to the observed IHC-ctDNA discordances rather than reduced ctDNA accuracy. Of the 25 samples for which IHC and alternative tissue testing results were available, 12 samples demonstrated IHC-ctDNA discordances. Importantly, PCR and / or NGS tissue testing supported the ctDNA NGS results rather than the tissue IHC in 5 of the 12 discordances. Taken together, these data support previous reports that IHC may not be as reliable as PCR diagnostic prototypes for MSI determination (36). [Table 10] [Table 11]
[0249] 4. ctDNA MSI status in 28,459 consecutive patients with advanced cancer
[0250] Numerous studies have assessed MSI prevalence across different tumor types in tissues (13, 32, 37), but to date, no landscape analysis of ctDNA MSI status across cancer types has been published. To accomplish this, we applied the MSI algorithm described above to 28,459 consecutive clinical samples from patients with advanced cancer examined at the Guardant Health Clinical Laboratory. In this cohort, 278 samples (median tumor fraction 6.55%, range 0.09–89%) comprising 16 different tumor types were identified as MSI-H by ctDNA. This corresponds to an overall pan-cancer prevalence of approximately 1%, similar to that previously reported for tissues (13, 32, 37). Similarly, MSI-H prevalence across tumor types also closely mirrored that observed in the tissue-based analysis (Figure (Figure11A); 11A), as expected, MSI-H was most prevalent in endometrial, colorectal, and gastric cancers, while other tumors such as lung, bladder, and head and neck cancers manifested lower prevalence. Specific deviations from previous MSI-H prevalence estimates included slightly lower prevalence in endometrial, colorectal, and gastric cancers, and slightly higher prevalence in prostate cancer.
[0251] Given the pan-solid tumor nature of the ctDNA intended use population and the immunotherapies approved for MSI-H tumors, the microsatellite loci for this panel were purposefully selected to provide MSI status information across all solid tumor types. Consistent with this design intent, analysis of the distribution of sample-level and locus-level MSI scores - i.e., MSI scores and site scores, respectively (Figures 11B and 11C) demonstrated consistent performance across tumor types, with MSI-H samples demonstrating signals significantly higher than the threshold. Furthermore, the diagnostic yield of MSI assessment outside of tumor types where MSI is commonly tested was substantial, with more than half of the identified cases (143 / 278) present in tumor types where MSI testing is very rare, thus identifying patients who would never have been tested otherwise.
[0252] Consistent with what has been reported in tissues (38), the number of indels and SNVs (including nonsynonymous and synonymous variants) is significantly increased in MSI-H samples compared to those characterized as having MSS status (Figure 12). Specifically, the median number of SNVs in MSI-H samples was 6.3 compared to 1.4 in MSS (chi-squared p<0.0001), and the median number of indels in MSI-H samples was 2.6 compared to 0.4 in MSS (chi-squared p<0.0001).
[0253] 5. ctDNA MSI status predicts immunotherapy response
[0254] Currently, the most notable utility of MSI status is its ability to select patients for immunotherapy. Despite this and the obstacles to obtaining tissue from many patients, the ability of ctDNA MSI status to predict response to immunotherapy has not been reported. To establish clinical validity for this biomarker, we present clinical outcomes for 16 ctDNA MSI-H metastatic gastric cancer patients treated with pembrolizumab (n=15) or nivolumab (n=1) after failure of standard care chemotherapy in a phase II pembrolizumab trial in gastric cancer (NCT#02589496). cfDNA and tissue PCR MSI assessment in pretreatment samples were 100% concordant for MSI-H (16 / 16, 95%CI 76-100%). Ten of 16 patients achieved a complete response (n=3) or partial response (n=7) as objectively assessed by the investigator using RECIST 1.1 criteria, and an additional 3 patients had stable disease (Figure 13A), representing an objective response rate of 63% (10 / 16, 95% CI 36–84%) and a disease control rate of 81% (13 / 16, 95% CI 0.01–0.29). The response rates were 54–95%), which was similar to previously reported responses in patients with MSI-H defined by histological testing (39). Importantly, even in this pretreated population, these responses were durable, with a median treatment duration of 39 weeks. Indeed, patient 21, for example, experienced complete disease regression following pembrolizumab treatment after failure of fluoropyrimidine / platinum chemotherapy and remains disease-free for more than 6 months after completion of 35 cycles of treatment (Figures 13B–13E).
[0255] Consideration
[0256] A novel cfDNA-based targeted NGS approach was validated for MSI detection - by using a large panel of microsatellites, the approach achieved high sensitivity compared to tissue-based methods while maintaining very high specificity. MSI-H prevalence detected in plasma across 16 common solid tumors was similar to published tissue-based compendiums, thus demonstrating the pan-tumor performance intended in the MSI detection algorithm design. Furthermore, by showing that patients with MSI-H detected by cfDNA benefited from ICB therapy similar to those reported for tissue-defined populations (39), we demonstrated the clinical utility of extending the availability of MSI detection to all patients, regardless of tissue availability or the need to undergo invasive tissue acquisition procedures.
[0257] This example demonstrates robust analytical performance for MSI detection in a ctDNA panel previously validated for detection of four other variants in all guideline-recommended indications (26). Notably, analytical sensitivity for MSI detection in design samples demonstrated reproducible detection down to 0.1%, consistent with previous reports of similar sensitivity for indels and SNVs (26). Importantly, this example evaluated the performance of ctDNA MSI testing in 1145 samples with orthogonal tissue MSI, constituting the largest ctDNA-tissue MSI matched cohort yet described. Compared to standard of care tissue MSI testing for the same patients, ctDNA MSI assessment demonstrated a high PPV (95%) comparable to the reported PPV of 90-92% reported for local vs. central tissue-based MSI assessment (36), as well as a high PPA (87%) in the evaluable population consistent with previous studies investigating concordance of plasma and tissue genotyping for other variants (25, 26, 45, 46). Factors that may contribute to imperfect concordance include tumor heterogeneity, primary lesion Differential shedding due to metastatic disease, discordance in time of tissue and plasma sampling, and less tumor shedding due to some tumors can be cited (40, 44, 47-49). Interestingly, gastric cancer patients identified as MSI-H by plasma and by pentaplex PCR in this report were previously reported to constitute separate tumor populations of MSS and MSI-H disease as assessed by both IHC and PCR performed on tissue (40). The same study found a 9% discordance in MSI-H between paired tissue biopsy samples in the same patient (40), thereby highlighting the potential contribution of intertumor heterogeneity to discordance in MSI status. Furthermore, the observation of non-trivial discordance between PCR and IHC histology in this report highlights the importance of accurate MSI testing, reported as a major cause of ICB failure (36). Consistent with the challenges presented by tissue genotyping in advanced solid tumors, studies in metastatic non-small cell lung cancer (NSCLC) have shown that, compared with tissue, plasma-based tests not only increase successful tumor genotyping outcomes, but also increase the frequency of detecting targetable mutations while delivering results at least 1 week faster than typical tissue genotyping results ( 25 , 45 ).
[0258] This example presents the first ctDNA-based landscape analysis of MSI in a large advanced pan-cancer cohort. Overall, the relative prevalence across tumor types in a set of >28,000 consecutive clinical samples is consistent with those reported for tissues (13, 32, 37), with few differences. For example, prevalence in CRC and endometrial cancer is lower than those reported for tissues (13), most likely reflecting the fact that tissue-based landscape analysis includes a large number of early MSI-H tumors that have a better prognosis (15) and are unlikely to be part of the advanced cancer population tested with ctDNA. On the other hand, the higher than expected prevalence of MSI-H prostate cancer is due to the increased representation of MSI-H disease in advanced patients, and two recent studies focusing on MSI status in advanced prostate cancer showed MSI-H prevalence of 3.1% and 3.8% in their patient populations, similar to our 2.6% observed in this study (50, 51). Not surprisingly, given the design intent of pan-cancer MSI detection, landscape analysis did not reveal tumor-type-specific patterns of microsatellite instability, but this does not exclude the possibility that tumor-type-specific patterns in plasma, similar to those shown in tissues (37, 52), may emerge by evaluation of a larger number of microsatellite loci and a larger number of representative MSI-H samples.
[0259] Although the clinical outcomes reported here are limited to gastric cancer, the observed objective response rates are nevertheless consistent with expectations from tissue-based studies, suggesting that ICB treatment based on cfDNA MSI results should achieve the expected outcomes across solid tumor types. In addition, the paucity of germline dMMR data prevented conclusions from being drawn about the familial impact of cfDNA-detected MSI. Finally, although treatment data were not available for the majority of patients with cfDNA- / tissue+ discrepancies, it is expected that at least some patients received ICB therapy based on their tumor results, which may suppress MSI-H disease and lead to the lack of MSI detection by cfDNA. Thus, the 87% sensitivity for MSI-H detection compared with tissue may be higher in treatment-naïve or treatment-naïve patients. Further studies should be pursued to address these issues.
[0260] In summary, we developed and validated a cfDNA-based targeted NGS panel that accurately assesses MSI status and also provides comprehensive tumor genotyping, thus enabling testing that meets the guidelines for pan-solid tumors with high sensitivity, specificity, and accuracy in a single peripheral blood draw. Clinical validation using both MSI and comparison with the originally reported outcomes of MSI-H patients supported the clinical accuracy and clinical relevance of this approach.Such simultaneous characterization of MSI status and tumor genotype from a simple peripheral blood draw has the potential to expand access to both targeted and immunotherapies to all patients with advanced cancer, including those for whom current tissue-based testing paradigms are inappropriate.
[0261] References [ka] [ka] [ka] [ka] [ka]
[0262] Although the above disclosure has been described in some detail by way of illustration and example for purposes of clarity and understanding, it will be apparent to those skilled in the art upon reading this disclosure that various changes in form and detail may be made without departing from the true scope of the disclosure and may be practiced within the scope of the appended claims. For example, the methods, systems, computer readable media, and / or all of their component features, steps, elements or other aspects may be used in various combinations.
[0263] All patents, patent applications, websites, other publications or documents, accession numbers, and the like cited herein are incorporated herein by reference in their entirety for all purposes, to the same extent as if each individual item was specifically and individually indicated to be so incorporated by reference. Even if different versions of a sequence were linked to an accession number at different times, the version linked to that accession number as of the effective filing date of this application is intended. By effective filing date, we mean the actual filing date or the filing date of a priority application, whichever is earlier, with reference to the accession number, if available. Similarly, even if different versions of a publication, website, or the like were published at different times, the version published closest to the effective filing date of this application is intended, unless otherwise indicated. The present invention provides, for example, the following items. (Item 1) 1. A method for determining repeat nucleic acid instability status of a nucleic acid sample, comprising: (a) quantifying a number of different repeat lengths present in each of a plurality of repetitive nucleic acid loci from sequence information to generate a site score for each of the plurality of repetitive nucleic acid loci, wherein the sequence information is from a population of repetitive nucleic acid loci in the nucleic acid sample; (b) calling a given repetitive nucleic acid locus as unstable if the site score of the given repetitive nucleic acid locus exceeds a site-specific trained threshold for the given repetitive nucleic acid locus, generating a repetitive nucleic acid instability score that includes a number of unstable repetitive nucleic acid loci from the plurality of repetitive nucleic acid loci; and (c) classifying the repeat nucleic acid instability status of the nucleic acid sample as unstable if the repeat nucleic acid instability score exceeds a population-trained threshold for the population of repeat nucleic acid loci in the nucleic acid sample, thereby determining the repeat nucleic acid instability status of the nucleic acid sample. A method comprising: (Item 2) 1. A method for determining repetitive DNA instability status of a sample, comprising: (a) quantifying a number of different repeat lengths present in each of a plurality of repetitive DNA loci from sequence information to generate a site score for each of the plurality of repetitive DNA loci, wherein the sequence information is from a population of repetitive DNA loci in the sample; (b) comparing the site score of a given repetitive DNA locus for each of the plurality of repetitive DNA loci to a site-specific trained threshold for the given repetitive DNA locus; (c) calling the given repetitive DNA locus as unstable if the site score for the given repetitive DNA locus exceeds the site-specific trained threshold for the given repetitive DNA locus, generating a repetitive DNA instability score that includes a number of unstable repetitive DNA loci from the plurality of repetitive DNA loci; and (d) classifying the repetitive DNA instability status of the sample as unstable if the repetitive DNA instability score exceeds a population-trained threshold for the population of repetitive DNA loci in the sample, thereby determining the repetitive DNA instability status of the sample. A method comprising: (Item 3) 1. A method for determining the microsatellite instability (MSI) status of a sample, comprising: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, wherein the sequence information is from a population of microsatellite loci in the sample; (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; (c) calling the given microsatellite locus as unstable if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; and (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby determining the MSI status of the sample. A method comprising: (Item 4) 1. A method for determining the microsatellite instability (MSI) status of a sample, comprising: (a) receiving sequence information from a population of microsatellite loci in said sample; (b) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from the sequence information to generate a site score for each of the plurality of microsatellite loci; (c) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; (d) calling the given microsatellite locus as unstable if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; and (e) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby determining the MSI status of the sample. A method comprising: (Item 5) 1. A method for identifying one or more customized therapies for treating a disease in a subject, comprising: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, wherein the sequence information is from a population of microsatellite loci in a sample; (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; (c) calling the given microsatellite locus as unstable if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby identifying an unstable sample; and (e) comparing the microsatellite instability status of the sample with one or more comparator results indexed with one or more therapies to identify one or more customized therapies for treating the disease in the subject. A method comprising: (Item 6) 1. A method of treating a disease in a subject, comprising: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, wherein the sequence information is from a population of microsatellite loci in a sample; (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; (c) calling the given microsatellite locus as unstable if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby identifying an unstable sample; (e) comparing the microsatellite instability status of the sample with one or more comparator results indexed with one or more therapies to identify one or more customized therapies for treating the disease in the subject; and (f) administering at least one of the identified customized therapies to the subject if the microsatellite instability status of the sample and the comparator result substantially match, thereby treating the disease in the subject. A method comprising: (Item 7) 1. A method of treating a disease in a subject, comprising administering to the subject one or more customized therapies, thereby treating the disease in the subject, wherein the customized therapies comprise: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate a site score for each of the plurality of microsatellite loci, wherein the sequence information is from a population of microsatellite loci in a sample; (b) comparing the site score of a given microsatellite locus for each of the plurality of microsatellite loci to a site-specific trained threshold for the given microsatellite locus; (c) calling the given microsatellite locus as unstable if the site score for the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, generating a microsatellite instability score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; (d) classifying the MSI status of the sample as unstable if the microsatellite instability score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby identifying an unstable sample; (e) comparing the microsatellite instability status of the sample with one or more comparator results indexed with one or more therapies; and (f) identifying one or more customized therapies for treating the disease in the subject if the microsatellite instability status of the sample and the comparator result substantially match. The method is identified by (Item 8) The method of any one of the preceding paragraphs, wherein the site scores for the plurality of microsatellite loci comprise likelihood scores. (Item 9) 9. The method of claim 8, wherein the likelihood score comprises a probabilistic log-likelihood based score that distinguishes biological signals derived from a large number of cfDNA fragments from somatic cells in the sample from noise arising from post-sample collection artifacts in the sample. (Item 10) 10. The method of claim 9, comprising determining the probabilistic log-likelihood-based score for each microsatellite locus in the sequence information from the sample using at least two parameters, where at least a first parameter comprises an allele frequency and at least a second parameter comprises at least one error mode. (Item 11) 11. The method of claim 10, wherein the allele frequency comprises a frequency of nucleic acids comprising different repeat lengths in the sequence information from the sample. (Item 12) 11. The method of claim 10, wherein the at least one error mode comprises a random error mode and a strand-specific error mode. (Item 13) the site scores for the plurality of microsatellite loci are (a) a score that measures the support of the observed sequence for the null hypothesis that the given microsatellite locus is stable; and (b) a score that measures the support of the observed sequence for the alternative hypothesis that the given microsatellite locus is unstable; and The method according to any one of the preceding claims, comprising the difference or ratio of: (Item 14) The method of any one of the preceding paragraphs, wherein the site scores for the plurality of microsatellite loci are generated using one or more of a likelihood criterion, a log likelihood criterion, a posterior probability criterion, Akaike Information Criterion (AIC), and a Bayesian Information Criterion. (Item 15) 2. The method of any one of the preceding paragraphs, wherein the site scores for the plurality of microsatellite loci comprise site scores based on Akaike Information Criterion (AIC), which tests for the presence of somatic indels at the plurality of microsatellite loci. (Item 16) The site score based on a given AIC is calculated according to the following formula: AIC=k-log likelihood 16. The method according to claim 15, wherein k is the number of parameters used in the model. (Item 17) 17. The method of claim 10 or 16, comprising estimating parameters of the model using maximum likelihood estimation (MLE). (Item 18) 18. The method of claim 17, comprising determining the MLE using the Nelder-Mead algorithm. (Item 19) The following formula: AIC 0 = k-log(Pr(obs|β,γ)) Calculating a null hypothesis score for the model using the formula: AIC 017. The method of claim 16, wherein x is the null hypothesis, k is the number of parameters used in the model, Pr is a probability, obs comprises the repeat length of observed sequencing reads covering the given microsatellite locus, β is at least one strand-specific error parameter, and γ is at least one random error parameter. (Item 20) The following formula: AIC min =min α (k-log(Pr(obs|β,γ,α)) Calculating an alternative hypothesis score for said model using the formula: AIC min is the alternative hypothesis, and min α is the minimizing effect over all values of α, k is the number of parameters used in the model, Pr is a probability, obs comprises the repeat length of the observed sequencing reads covering the given microsatellite locus, β is at least one strand-specific error parameter, γ is at least one random error parameter, and α is an allele frequency of at least 1, where α is one or more i is a vector of allele frequencies whose sum equals 1, Item 17. The method according to item 16. (Item 21) The following formula: ΔAIC=AIC 0 -AIC min 4. The method of any one of the preceding claims, comprising detecting the site score using (Item 22) γ is (a) a read-level error rate in which the observed microsatellite length in a sequencing read is one repeat unit longer than the expected microsatellite length for the strand of the nucleic acid molecule from which it originates; and / or (b) the read-level error rate in which the observed microsatellite length in a sequencing read is one repeat unit shorter than the expected microsatellite length for the strand of the original nucleic acid molecule; The method according to any one of the preceding items, comprising: (Item 23) β is, (a) strand-level error rate where the predicted microsatellite length of the sense strand is one repeat unit longer than the predicted microsatellite length of the nucleic acid-derived molecule; (b) a strand-level error rate in which the predicted microsatellite length of the antisense strand is one repeat unit longer than the predicted microsatellite length of the nucleic acid-derived molecule; (c) a strand-level error rate in which the predicted microsatellite length of the sense strand is one repeat unit shorter than the predicted microsatellite length of the nucleic acid-derived molecule; and / or (d) a strand-level error rate in which the predicted microsatellite length of the antisense strand is one repeat unit shorter than the predicted microsatellite length of the nucleic acid-derived molecule; The method according to any one of the preceding items, comprising: (Item 24) 2. The method of any one of the preceding items, comprising calling the given microsatellite locus as unstable if the site score for the given microsatellite locus statistically exceeds the site-specific trained threshold for the given microsatellite locus. (Item 25) The method of any one of the preceding claims, wherein the sample comprises a mutant allele proportion. (Item 26) The method of any one of the preceding claims, wherein the sample comprises a tumor fraction. (Item 27) 27. The method of claim 26, wherein the tumor fraction comprises the maximum mutated allele fraction (MAF) of all somatic mutations identified in the nucleic acid in the sample. (Item 28) 28. The method of claim 27, wherein the tumor fraction is less than about 0.05%, about 0.1%, about 0.2%, about 0.5%, about 1%, about 2%, about 3%, about 4%, about 5%, about 6%, about 7%, about 8%, about 9%, about 10%, about 11%, about 12%, about 13%, about 14%, or about 15% of the total nucleic acid in the sample. (Item 29) 2. The method of any one of the preceding claims, wherein the plurality of microsatellite loci comprises all of the population of microsatellite loci. (Item 30) 29. The method of any one of items 1 to 28, wherein the plurality of microsatellite loci comprises a subset of the population of microsatellite loci. (Item 31) The method of any one of the preceding claims, comprising determining the site-specific trained threshold and / or the population trained threshold from sequence information from a population of microsatellite loci in one or more training DNA samples. (Item 32) 32. The method of claim 31, wherein the training DNA sample comprises a non-tumor cfDNA sample. (Item 33) 32. The method of claim 31, wherein the training DNA sample comprises DNA from one or more tumor types. (Item 34) The method of any one of the preceding paragraphs, having a sensitivity of at least about 94% at a limit of detection (LOD) of about 0.2% tumor fraction of nucleic acid in the sample. (Item 35) 2. The method of any one of the preceding paragraphs, having a specificity of at least about 99% for non-tumor DNA in the sample. (Item 36) The method of any one of the preceding claims, wherein the determined MSI status of the sample comprises at least about 95%, 96%, 97%, 98% or 99% concordance with the corresponding MSI status of the sample determined using a PCR-based MSI assessment technique over a tumor fraction range of about 1.4% to about 15%. (Item 37) 37. The method of claim 36, wherein the match is 100%. (Item 38) 20. The method of any one of the preceding items, comprising classifying the MSI status of the sample as MSI-high (MSI-H) if the microsatellite instability score is greater than about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 15, about 20, about 30, about 40, about 50 or more unstable microsatellite loci from the plurality of microsatellite loci. (Item 39) 20. The method of any one of the preceding paragraphs, comprising classifying the MSI status of the sample as MSI-high (MSI-H) if the number of unstable microsatellite loci constitutes about 0.1%, about 1%, about 2%, about 3%, about 4%, about 5%, about 10%, about 15%, about 20% or about 25% of the plurality of microsatellite loci. (Item 40) The disease is selected from the group consisting of biliary tract cancer, bladder cancer, transitional cell carcinoma, urothelial carcinoma, brain cancer, glioma, astrocytoma, breast cancer, metaplastic carcinoma, cervical cancer, cervical squamous cell carcinoma, rectal cancer, colorectal cancer, colon cancer, hereditary nonpolyposis colorectal cancer, colorectal adenocarcinoma, gastrointestinal stromal tumor (GIST), endometrial cancer, endometrial stromal sarcoma, esophageal cancer, esophageal squamous cell carcinoma, esophageal adenocarcinoma, ocular melanoma, uveal melanoma, gallbladder cancer, gallbladder adenocarcinoma, renal cell carcinoma, renal clear cell carcinoma, transitional cell carcinoma, urothelial carcinoma, Wilms' tumor, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic (CLL), chronic myelogenous (CML), chronic myelomonocytic (CMML), liver cancer, liver cancer, hepatoma, hepatocellular carcinoma, cholangiocarcinoma, liver The method according to any one of items 5 to 7, comprising a cancer comprising at least one tumor type selected from the group consisting of blastoma, lung cancer, non-small cell lung cancer (NSCLC), mesothelioma, B cell lymphoma, non-Hodgkin's lymphoma, diffuse large B cell lymphoma, mantle cell lymphoma, T cell lymphoma, non-Hodgkin's lymphoma, precursor T lymphoblastic lymphoma / leukemia, peripheral T cell lymphoma, multiple myeloma, nasopharyngeal carcinoma (NPC), neuroblastoma, oropharyngeal cancer, oral squamous cell carcinoma, osteosarcoma, ovarian cancer, pancreatic cancer, pancreatic ductal adenocarcinoma, pseudopapillary neoplasm, acinar cell carcinoma, prostate cancer, prostate adenocarcinoma, skin cancer, melanoma, malignant melanoma, cutaneous melanoma, small intestine cancer, gastric cancer, gastrointestinal stromal tumor (GIST), uterine cancer, and uterine sarcoma. (Item 41) 8. The method according to any one of items 5 to 7, wherein the therapy comprises at least one immunotherapy. (Item 42) 42. The method of claim 41, wherein the immunotherapy comprises at least one checkpoint inhibitor antibody. (Item 43) 42. The method of claim 41, wherein the immunotherapy comprises an antibody against PD-1, PD-2, PD-L1, PD-L2, CTLA-4, OX40, B7.1, B7He, LAG3, CD137, KIR, CCR5, CD27, CD40, or CD47. (Item 44) 42. The method of claim 41, wherein the immunotherapy comprises administration of a pro-inflammatory cytokine to at least one tumor type. (Item 45) 42. The method of claim 41, wherein the immunotherapy comprises administration of T cells against at least one tumor type. (Item 46) 20. The method of any one of the preceding claims, further comprising obtaining the sample from a subject. (Item 47) 47. The method of claim 46, wherein the sample is selected from the group consisting of tissue, blood, plasma, serum, sputum, urine, semen, vaginal fluid, feces, synovial fluid, spinal fluid and saliva. (Item 48) 47. The method of claim 46, wherein the subject is a mammalian subject. (Item 49) 49. The method of claim 48, wherein the mammalian subject is a human subject. (Item 50) 47. The method of claim 46, wherein the sample contains cell-free nucleic acid. (Item 51) 51. The method of claim 50, wherein the sample contains circulating tumor nucleic acid. (Item 52) 2. The method of any one of the preceding claims, further comprising receiving the sequence information generated from the sample, wherein the sequence information comprises cfDNA sequencing reads from the population of microsatellite loci in the sample. (Item 53) 53. The method of any one of items 46 to 52, comprising amplifying one or more segments of nucleic acid in the sample to generate at least one amplified nucleic acid. (Item 54) 54. The method of any one of items 46 to 53, further comprising a step of sequencing nucleic acid from the sample to generate the sequence information. (Item 55) 55. The method of any one of items 46 to 54, wherein the sequence information is obtained from a targeted segment of a nucleic acid in the sample, the targeted segment being obtained by selectively enriching one or more regions from the nucleic acid in the sample prior to sequencing. (Item 56) 56. The method of claim 55, further comprising amplifying the obtained targeted segment prior to sequencing. (Item 57) 57. The method of any one of items 46 to 56, further comprising attaching one or more adapters comprising a barcode to the nucleic acid prior to sequencing. (Item 58) 58. The method of any one of items 54 to 57, wherein the sequencing is selected from the group consisting of targeted sequencing, intron sequencing, exome sequencing, and whole genome sequencing. (Item 59) 59. The method of any one of items 54-58, comprising sequencing at least about 50, about 100, about 150, about 200, about 250, about 500, about 750, about 1,000, about 1,500, about 2,000, or more targeted genomic regions within the nucleic acid of the sample to generate said sequence information. (Item 60) 2. The method of any one of the preceding paragraphs, wherein the number of different repeat lengths comprises a frequency of each different repeat length present in each of the plurality of microsatellite loci. (Item 61) 13. The method of any one of the preceding items, wherein at least a portion of the method is computer-implemented. (Item 62) The method of any one of the preceding items, further comprising a step of generating a report optionally including information regarding the instability status of the sample, information regarding the instability score of the sample, and / or information regarding the one or more customized therapies for treatment of the disease in the subject. (Item 63) Item 63. The method or system of item 62, further comprising the step of communicating the report to the subject or to a third party, such as a health care provider. (Item 64) 2. The method of any one of the preceding claims, further comprising classifying the repeat nucleic acid instability status of the nucleic acid sample as stable if the repeat nucleic acid instability score is at or below the population trained threshold for the population of the repeat nucleic acid loci in the nucleic acid sample. (Item 65) 2. The method of any one of the preceding claims, further comprising classifying the repetitive DNA instability status of the sample as stable if the repetitive DNA instability score is at or below the population trained threshold for the population of the repetitive DNA loci in the sample. (Item 66) 2. The method of any one of the preceding claims, further comprising classifying the microsatellite instability status of the sample as stable if the microsatellite instability score is at or below the population trained threshold for the population of microsatellite loci in the sample.
Claims
1. A method for obtaining a site score (SS) generated from sequence information of a number of different repeat lengths present at each of a number of microsatellite loci as an index for determining the microsatellite instability (MSI) status of a sample, the sample comprising cell-free nucleic acid, the method comprising: (a) quantifying a number of different repeat lengths present at each of a plurality of microsatellite loci from sequence information to generate the SS for each of the plurality of microsatellite loci, wherein the sequence information is from a population of microsatellite loci in the sample; The SS of the plurality of microsatellite loci is (i) (A) a score that measures the support of the observed sequence for the null hypothesis that a given microsatellite locus is stable; and (B) A score that measures the support of the observed sequence for the alternative hypothesis that a given microsatellite locus is unstable; and (ii) generated using one or more of a likelihood criterion, a log likelihood criterion, a posterior probability criterion, and a Bayesian information criterion; The method includes determining a probabilistic log likelihood-based score for each microsatellite locus in the sequence information from the sample using at least two parameters, the at least two parameters being: (iii) an allele frequency (AF) comprising the frequency of nucleic acids comprising different repeat lengths in the sequence information from the sample; and (iv) at least one error mode, including a random error mode and a strand-specific error mode; and (b) comparing the SS of a given microsatellite locus to a site-specific trained threshold for the given microsatellite locus, for each of the plurality of microsatellite loci; (c) calling the given microsatellite locus as unstable if the SS of the given microsatellite locus exceeds the site-specific trained threshold for the given microsatellite locus, and generating a microsatellite instability (MI) score that includes a number of unstable microsatellite loci from the plurality of microsatellite loci; and (d) classifying the MSI status of the sample as unstable if the MI score exceeds a population-trained threshold for the population of microsatellite loci in the sample, thereby determining the MSI status of the sample; A method comprising: (e) comparing the MSI status of the sample to one or more comparator results indexed with one or more therapies; 13. The method of claim 1, wherein if the MSI status of the sample and the comparator result substantially match, the subject is a candidate for one or more customized therapies to treat disease in the subject.
3. At least one error mode including a random error mode and a strand-specific error mode, (1) the read-level error rate where the observed microsatellite length within a sequencing read is one repeat unit longer than the expected microsatellite length for the strand of the original nucleic acid molecule; (2) the read-level error rate where the observed microsatellite length within a sequencing read is one repeat unit shorter than the expected microsatellite length for the strand of the original nucleic acid molecule; (3) a strand-level error rate in which the predicted microsatellite length of the sense strand is one repeat unit longer than the predicted microsatellite length of the nucleic acid-derived molecule; (4) a strand-level error rate in which the predicted microsatellite length of the antisense strand is one repeat unit longer than the predicted microsatellite length of the nucleic acid-derived molecule; (5) a strand-level error rate in which the predicted microsatellite length of the sense strand is one repeat unit shorter than the predicted microsatellite length of the nucleic acid-derived molecule; and / or (6) a strand-level error rate in which the predicted microsatellite length of the antisense strand is one repeat unit shorter than the predicted microsatellite length of the nucleic acid-derived molecule; The method according to any one of claims 1 to 2, comprising:
4. determining the site-specific trained threshold and / or the population trained threshold from sequence information from a population of microsatellite loci in one or more training DNA samples; 4. The method of any one of claims 1-3, wherein the training DNA samples comprise non-tumor cfDNA samples, DNA from one or more tumor types, or both.
5. A method according to any one of claims 1 to 4, having a sensitivity of at least 94% at a limit of detection (LOD) of 0.2% tumor fraction of nucleic acid in the sample.
6. A method described in any one of claims 1 to 5, having a specificity of at least 99% for non-tumor DNA in the sample.
7. A method according to any one of claims 1 to 6, comprising a step of determining the MSI status of the sample having at least 95%, 96%, 97%, 98% or 99% concordance with the corresponding MSI status of the sample determined using a PCR-based MSI assessment technique across a tumor fraction range of 1.4% to 15%, e.g. the concordance is 100%.
8. The method of claim 1, further comprising classifying the MSI status of the sample as MSI-H based on an MI score exceeding 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 30, 40, 50 or more unstable microsatellite loci from the plurality of microsatellite loci.
9. The method of claim 1, further comprising classifying the MSI status of the sample as MSI-H based on a number of unstable microsatellite loci comprising 0.1%, 1%, 2%, 3%, 4%, 5%, 10%, 15%, 20% or 25% of the plurality of microsatellite loci.
10. (a) The disease is selected from the group consisting of biliary tract cancer, bladder cancer, transitional cell carcinoma, urothelial carcinoma, brain cancer, glioma, astrocytoma, breast cancer, metaplastic carcinoma, cervical cancer, cervical squamous cell carcinoma, rectal cancer, colorectal cancer, colon cancer, hereditary nonpolyposis colorectal cancer, colorectal adenocarcinoma, gastrointestinal stromal tumor (GIST), endometrial cancer, endometrial stromal sarcoma, esophageal cancer, esophageal squamous cell carcinoma, esophageal adenocarcinoma, ocular melanoma, uveal melanoma, gallbladder cancer, gallbladder adenocarcinoma, renal cell carcinoma, renal clear cell carcinoma, transitional cell carcinoma, urothelial carcinoma, Wilms' tumor, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic (CLL), chronic myelogenous (CML), chronic myelomonocytic (CMML), liver cancer, liver cancer, hepatoma, cell carcinoma, cholangiocarcinoma, hepatoblastoma, lung cancer, non-small cell lung cancer (NSCLC), mesothelioma, B cell lymphoma, non-Hodgkin's lymphoma, diffuse large B cell lymphoma, mantle cell lymphoma, T cell lymphoma, non-Hodgkin's lymphoma, precursor T lymphoblastic lymphoma / leukemia, peripheral T cell lymphoma, multiple myeloma, nasopharyngeal carcinoma (NPC), neuroblastoma, oropharyngeal carcinoma, oral squamous cell carcinoma, osteosarcoma, ovarian cancer, pancreatic cancer, pancreatic ductal adenocarcinoma, pseudopapillary neoplasm, acinar cell carcinoma, prostate cancer, prostate adenocarcinoma, skin cancer, melanoma, malignant melanoma, cutaneous melanoma, small intestine cancer, gastric cancer, gastrointestinal stromal tumor (GIST), uterine cancer, and uterine sarcoma; (b) the therapy comprises at least one immunotherapy, e.g., the immunotherapy comprises: at least one checkpoint inhibitor antibody; antibodies against PD-1, PD-2, PD-L1, PD-L2, CTLA-4, OX40, B7.1, B7He, LAG3, CD137, KIR, CCR5, CD27, CD40 or CD47; Administration of a pro-inflammatory cytokine to at least one tumor type; or Administration of T cells against at least one tumor type The method of claim 2 , comprising:
11. The method of claim 1, wherein the sample is obtained from a subject.
12. The method of any one of claims 1 to 11, wherein the sample is selected from the group consisting of tissue, blood, plasma, serum, sputum, urine, semen, vaginal fluid, feces, synovial fluid, spinal fluid and saliva.
13. The method of any one of claims 1 to 12, wherein the subject is a mammalian subject.
14. The method of any one of claims 1 to 13, wherein the sample contains circulating tumor nucleic acid.
15. A method according to any one of claims 1 to 14, wherein the method comprises a step of receiving the sequence information generated from the sample, the sequence information comprising cfDNA sequencing reads from the population of microsatellite loci in the sample.
16. A method according to any one of claims 1 to 15, wherein the sequence information is obtained from a targeted segment of nucleic acid in the sample.
17. The method of claim 1, wherein the sample comprises a mutant allele proportion and / or the sample comprises a tumor proportion.
18. The method of claim 17, wherein the tumor fraction comprises a maximum mutated allele fraction (MAF) of all somatic mutations identified in the nucleic acid in the sample, and the tumor fraction is less than 0.05%, 0.1%, 0.2%, 0.5%, 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14% or 15% of all nucleic acid in the sample.
19. A method according to any one of claims 1 to 18, wherein the number of different repeat lengths comprises the frequency of each different repeat length present in each of the plurality of microsatellite loci.
