Patent Application
20240036052
The present invention relates to a cancer test method and a cancer test system for evaluating cancer in a subject.
In a conventional cancer test method using a cancer marker, first, candidate markers are comprehensively analyzed, and, from detected candidate markers, a promising candidate marker is extracted by statistical analysis and machine learning, and then a prediction formula for discriminating between healthy subjects and cancer patients is created by multivariate analysis. Next, the test is performed by a process of substituting a measured value of a marker in a subject into the prediction formula and determining a risk of cancer from an obtained predictive value.
The present inventors have also comprehensively analyzed urinary metabolites as cancer markers by means of a liquid chromatograph/mass spectrometer (LC/MS), and have multiplied the intensities of multiple urinary tumor markers by a given coefficient, and then, in accordance with a prediction formula: a predictive value=Σ(αn×In)+β[αn: coefficient, β: constant, In: intensity of urinary tumor marker], have determined that a risk of cancer is high if the predictive value is equal to or greater than 0 and that the risk is low if the predictive value is smaller than 0 (for example, PTLS 1 and 2).
PTL 1: JP 2019-105456 A
PTL 2: JP 2020-079729 A
PTL 3: WO 2007/076439 A
PTL 4: WO 2011/119772 A
PTL 5: US 2009/0004687 A1
However, in a case of using such a prediction formula, depending on a combination of a coefficient and an intensity of a cancer marker, a predictive value may be close to 0, which makes the determination difficult in some cases.
Meanwhile, the idea of a metabolite list as a marker set already exists. For example, according to PTLS 3 to 5, in a cancer test using metabolites as markers, a metabolite list including multiple metabolites is created. However, in the conventional techniques, there are neither reports nor suggestions about: at the time of evaluation of cancer risk, with respect to individual metabolite markers, setting a reference range of healthy subjects that is training data and classifying the metabolite markers into three groups; performing similar calculation on the same multiple metabolites in test data as well; or analyzing which pattern in the metabolite list created on the basis of the training data a metabolite pattern of the test data belongs to.
Accordingly, an object of the present invention is to provide a cancer test method and system for evaluating cancer with higher accuracy and precision.
The present inventors have found that in a cancer test method using cancer markers, evaluation of cancer can be made more precisely by analyzing measured values of the cancer markers by classifying them into three groups: within a reference range of healthy subjects, higher than the reference range, and lower than the reference range. Furthermore, the present inventors have found that the evaluation of cancer can be made with ease and precision by displaying the cancer markers to be visually distinguishable among the three groups and/or by assigning column values to the three groups and substituting the column values into an evaluation function indicating an estimate of the amount of variation of a marker and/or a distance function indicating a degree of similarity of a marker pattern.
The present invention encompasses, for example, the following.
(wherein,
The method according to [9], in which the calculating the differences from the patterns is performed with a distance function represented by Formula II:
(wherein,
According to the present invention, a cancer test method and system for accurately and precisely evaluating cancer are provided. The method and system of the present invention are excellent in sensitivity and specificity, and therefore reduces false positives and false negatives and helps in precise diagnosis. Therefore, the present invention may be useful in the fields of cancer diagnosis, examination, treatment evaluation, and the like.
In the present specification, “σ”, the symbol denoting standard deviation, all represents: {circumflex over (σ)}, that is, sigma-hat (a population estimate).
The present invention provides a cancer test method and a cancer test system for evaluating cancer in a subject. According to the present invention, evaluation of cancer may include determination of cancer in a subject, prediction of a risk of cancer in a subject, determination of the stage or severity of cancer in a subject, prognostication of cancer in a subject, monitoring of cancer in a subject, monitoring of efficacy in treatment of cancer present in a subject, or aid in diagnosis of cancer.
An aspect of the present invention is to provide a method for testing cancer in a subject, the method including:
In the cancer test method according to the present invention, in accordance with a measured value of each cancer marker, the multiple cancer markers may be classified into three groups: within a specific reference range, higher than the reference range, and lower than the reference range.
The cancer markers are not particularly limited as long as they are markers that show association with cancer. For example, a specific protein, a miRNA, and a metabolite are known as a cancer marker, and the cancer marker can be either a marker in blood or in urine. Multiple cancer markers may be preferable, and may include, for example, 2 or more, 3 or more, 4 or more, 5 or more, 10 or more, 15 or more, 20 or more, 25 or more, 30 or more, 40 or more, 50 or more, or more cancer markers. There is not particularly an upper limit to the number of the cancer markers; however, the upper limit can be, for example, up to 500, up to 400, or up to 300. The cancer markers may preferably be multiple markers that can be measured at the same time. In a preferred embodiment, the cancer markers may be urinary metabolites. A “measured value” of a cancer marker may differ depending on the type of marker used, which can be understood by those skilled in the art.
First, from data of measured values of cancer markers of multiple healthy subjects, a mean value of the healthy subjects±X×standard deviation (wherein X is an arbitrary numerical value) and/or the mean value of the healthy subjects±the standard deviation may be set as a reference range. In one embodiment, in a case where X is 2, a reference range of 95%, i.e., 2σ mode (a range sandwiched between (the mean value−2×the standard deviation) and (the mean value+2×the standard deviation)) and/or a reference range of 68% that is 1σ mode (a range sandwiched between (the mean value−the standard deviation) and (the mean value+the standard deviation)) may be set.
With respect to measured values of cancer markers of multiple cancer patients and multiple healthy subjects, discrimination information that classifies the cancer markers into three groups: within the reference range, higher than the reference range, and lower than the reference range may be obtained on the basis of the reference range. In a preferred embodiment, the measured value of each cancer marker has discrimination information that differs between the two reference ranges: the reference range based on the mean value of the healthy subjects±X×the standard deviation (wherein X is 2) and the reference range based on the mean value of the healthy subjects±the standard deviation. In the database, the measured values of the cancer markers may be stored together with the discrimination information. Therefore, the database may include, as information of the cancer markers, at least the types of the cancer markers, the measured values of the cancer markers, and the discrimination information.
The database may include a marker panel on which information of multiple cancer markers with respect to cancer patients in a specific stage or with a specific degree of severity (for example, such as a specific depth of invasion T, a specific grade, or the presence or absence of metastasis or recurrence) or cancer patients before or after treatment may be registered. Furthermore, the database may include different databases according to the types of cancers. The types of cancers may not be particularly limited, and, may include, for example, colorectal cancer, breast cancer, pediatric cancer, etc.
The information of the cancer markers may be stored as a marker panel in the database. A “metabolite list” means a set of multiple markers for each cancer in cancer patients and healthy subjects, and a table in which measured values and/or discrimination information are arranged is referred to as a marker panel. For example, there is a tabular marker panel like one shown in
In the marker panel, preferably, columns of the cancer markers may be displayed to be visually distinguishable according to three groups of discrimination information. For example, the columns can be displayed to be distinguishable by shades of color, by different colors, or by lighting. For example, as shown in the lower right part of
When the marker panel is created, preferably, the cancer markers may be shown in order of importance calculated by machine learning. For example, in
Measured values of one or more cancer markers of a subject may also be classified into three groups: within a reference range, higher than the reference range, and lower than the reference range. In a preferred embodiment, a marker panel of the subject may be created on the basis of the measured values of the cancer markers of the subject.
The marker panel including the discrimination information in the database obtained in this way may be compared with the marker panel of the cancer markers of the subject, and a correlation of the measured values of the cancer markers of the subject with the discrimination information in the database may be analyzed. This analysis can be made by a method publicly known in this technical field. In one embodiment, the three groups of discrimination information may be assigned numerical values, and, taking into account the importance calculated by machine learning, the correlation can be analyzed by a function. A specific example of that is described below.
For the purpose of analysis of the correlation with the discrimination information, it may be preferable to assign the measured values of the cancer markers a column value of 0 if it is within the reference range, a column value of +1 if higher than the reference range, or a column value of −1 if it is lower than the reference range according to the above three groups of discrimination information. For example, in a panel in the lower part of
According to the cancer test method of the present invention, a marker panel of cancer markers of a subject may be compared with the marker panel including the discrimination information in the database, and cancer can be evaluated on the basis of its homology or difference with the discrimination information of the cancer patients or the healthy subjects. In a case of a subject who cannot be determined to be a cancer patient or a healthy subject by this method, it may be preferable to make a further analysis as follows.
In one embodiment, a correlation with the discrimination information may be made on the basis of a value of an evaluation function shown in
(wherein,
A value obtained by the evaluation function is an estimate of the amount of variation of a cancer marker. On the basis of respective values of the evaluation function for the cancer patients and the healthy subjects in the database, an evaluation function value that can give an evaluation of being at risk for cancer and an evaluation function value that can give an evaluation of being healthy can be calculated. For example, when the evaluation function value that can give an evaluation of being at risk for cancer is denoted by α, and the evaluation function value that can give an evaluation of being healthy is denoted by β, a subject may be evaluated to be at risk for “cancer” if an evaluation function value of the subject is above α, and may be evaluated to be “healthy” if the evaluation function value is less than β.
In one embodiment, a difference from the pattern can be calculated on the basis of a value of a distance function shown in
(wherein,
A value obtained by the distance function is a degree of similarity of a cancer marker to the pattern. As with the determination algorithm 1, an evaluation function may be calculated on the basis of discrimination information with a mean value of the healthy subjects±X×standard deviation (wherein X is 2) as a reference range. Specifically, first, an evaluation function may be calculated in 2σ mode. As described above, for example, when an evaluation function value that can give an evaluation of being at risk for cancer is denoted by α, and an evaluation function value that can give an evaluation of being healthy is denoted by β, a subject may be evaluated to be at risk for “cancer” if an evaluation function value of the subject is above α, and may be evaluated to be “healthy” if the evaluation function value is less than β. Here, in a case where the evaluation function value is equal to or more than β and equal to or less than α, a distance function may further be calculated. That is, a pattern of the subject may be compared with a pattern of a typical cancer patient (for example, a cancer patient with high severity) and a pattern of a typical healthy subject (for example, such as a healthy subject whose cancer markers are all within the reference range), and whether or not there is a deviation (a difference) may be calculated. The one having a smaller deviation (difference) may be an evaluation result. That is, the subject may be evaluated to be “close to cancer” if the pattern of the subject is close to the pattern of the cancer patients, and may be evaluated to be “close to healthy” if not close to the pattern of the cancer patients (the right side of
In the cancer test method using the metabolite list according to the present invention, four types of evaluations, “cancer”, “healthy”, “close to cancer”, and “close to healthy”, can be made, thus a cancer test becomes more precise, and it becomes possible to evaluate a subject who has been previously determined to be falsely positive or falsely negative with more precise and high accuracy.
The “evaluation” made by the cancer test method or the cancer test system according to the present invention is intended to be able to evaluate a statistically significant proportion of subjects. Therefore, the “evaluation” made by the method and the system according to the present invention also includes a case where all (i.e., 100%) of subjects cannot always have a correct result. The statistically significant proportion can be determined by means of a variety of well-known statistical evaluation tools, for example, such as determination of confidence interval, determination of p-value, Student's t-test, and Mann Whitney test. A preferred confidence interval may be at least 90%. The p-value may preferably be 0.1, 0.01, 0.05, 0.005, or 0.0001. More preferably, at least 60%, at least 80%, or at least 90% of subjects can be appropriately evaluated by the method or the system according to the present invention.
Below is described an embodiment in which urinary metabolites are used as cancer markers, and a cancer test is performed on the basis of a metabolite list.
First, urinary metabolites in a urine sample of a subject may be measured. The urinary metabolites to be measured are not particularly limited as long as they can be used as cancer markers. For example, urinary metabolites exemplified in the Examples and urinary metabolites reported in WO 2017/213246 A (colorectal cancer and breast cancer), JP 2019-168319 A, PTL 2 (pediatric cancer), etc. can be used. Multiple urinary metabolites may be used in combination, preferably, 3 or more, 5 or more, 10 or more, or 20 or more urinary metabolites may be used in combination, thereby more precise and highly accurate evaluation and monitoring of efficacy in treatment may become possible. The combination of urinary metabolites is not particularly limited, and can be appropriately selected according to the type of cancer, the sex and age of a subject, the purpose, including determination of cancer or cancer risk, follow-up (monitoring of cancer), and monitoring of treatment, etc.
The urine sample means urine collected from a subject and a sample obtained by treating the urine (for example, urine added with a preservative, such as toluene, xylene, or hydrochloric acid).
Furthermore, subjects may be humans and other mammals, for example, primates, domestic animals, animals for pets, and experimental animals, and further, may be reptiles, birds, or other things. In particular, the subject may preferably be a human. For example, the present invention may be applied to mass screening in a medical examination or a cancer test, or may be applied at the time of additional checking after such mass screening.
Measurement of a urinary metabolite means measuring the amount or concentration of the metabolite in a urine sample, preferably semiquantitatively or quantitatively, and the amount of the metabolite may be either an absolute amount or a relative amount. The measurement can be directly or indirectly made. The direct measurement may include measuring the amount or concentration of the metabolite on the basis of a signal directly correlated with the number of molecules of the urinary metabolite present in the sample. Such a signal may be based on, for example, a specific physical or chemical characteristic of the urinary metabolite. The indirect measurement may be measurement of a signal obtained from a secondary component (i.e., a component other than the urinary metabolite), for example, a ligand, a label, or an enzymatic product.
A method of measuring a urinary metabolite is not particularly limited, and a method or means publicly known in this technical field can be used. For example, measurement of a urinary metabolite can be made by a means for measuring a physical or chemical characteristic specific to the urinary metabolite, for example, a means for measuring a precise molecular weight or NMR spectrum, or the like. Means for measuring a urinary metabolite may include analyzers such as a mass spectrometer, an NMR spectrometer, a two-dimensional electrophoresis apparatus, a chromatograph, and a liquid chromatography mass spectrometer (LC/MS). These analyzers may be used independently to measure urinary tumor markers, or urinary tumor markers may be measured by a plurality of the analyzers.
Alternatively, in a case where a reagent for detecting a metabolite to be measured, for example, an immunoreaction reagent, an enzyme reaction reagent, or the like can be used, the metabolite in urine can be measured using such a reagent.
As above, a urinary metabolite contained in a urine sample collected from a subject may be measured, and a measured value of the urinary metabolite may be applied to the above cancer test method of the present invention, thereby it is possible to evaluate cancer in the subject. Furthermore, the urinary metabolite in each of urine samples collected at multiple points of time from the subject may be measured.
Then, with respect to the above measured value of the urinary metabolite of the subject, a correlation with the discrimination information in the database may be analyzed. The database here has stored therein a metabolite list on which information of multiple urinary metabolites with respect to multiple healthy subjects and cancer patients may be registered. The database may include, with respect to the measured value of each urinary metabolite, discrimination information that classifies the urinary metabolite, with
In one embodiment, with respect to the measured value of the urinary metabolite of the subject, a correlation with three groups of discrimination information may be analyzed. As described above, the analysis of the correlation with the discrimination information can be made, for example, by comparison of metabolite lists by means of the determination algorithm 1 or the determination algorithm 2. On the basis of a result of this analysis, cancer in the subject may be evaluated.
According to the cancer test method of the present invention, the presence or progression of cancer can be determined at an early stage with high accuracy, and cancer can be precisely evaluated and subdivide into cancer or close to cancer, or healthy or close to healthy. For example, it may also be possible to evaluate the stage or severity of cancer, and this helps in determining an extensive examination and a treatment plan. If a simple test makes it possible to diagnose whether or not cancer is present or if a subject is at risk for cancer, it can be expected that not only treatment but also invasion risk caused by a test can be prevented. The subject can receive treatment for cancer early and a follow-up after the treatment. Furthermore, it may be possible to monitor the efficacy in the treatment of cancer, and it may be possible to consider the discontinuance, continuance, or change of the treatment according to the efficacy in the treatment. Moreover, since a urine sample is used, a minimally invasive evaluation of cancer can be made with ease and at low cost.
In another embodiment, a urine sample may be collected from a subject at multiple points of time, a urinary metabolite contained in the urine sample at each point of time of measurement may be measured, and, with respect to a measured value of the urinary metabolite at each point of time of measurement, a correlation with discrimination information in the database may be analyzed. The measurement can be made at least 2 times, 3 times, 4 times, 5 times, 10 times, 15 times, 20 times, 30 times, or more than 30 times with time, for example, at intervals of 1 day, 2 days, 5 days, 1 week, 2 weeks, 3 weeks, 1 month, 2 months, 3 months, half a year, 1 year, 2 years, 3 years, 5 years, or more than 5 years. By this analysis, time-course monitoring can be performed, and the progress of cancer, the metastasis or recurrence of cancer, the onset of cancer from no abnormalities, etc. can be evaluated.
In still another embodiment, efficacy in treatment (a therapeutic agent or method) of cancer in a subject having cancer can be monitored. Specifically, it includes:
In the above method, a urine sample may be collected from a subject having cancer before the subject undergoes a treatment with a therapeutic agent or method, and urinary metabolites in the urine sample may be measured. After the treatment with the therapeutic agent or method is performed on the subject having cancer, a urine sample may be collected at an appropriate time, and the urinary metabolites in the urine sample may be measured. For example, a urine sample may be collected immediately, 30 minutes, 1 hour, 3 hours, 5 hours, 10 hours, 15 hours, 20 hours, 24 hours (1 day), 2 to 10 days, 10 to 20 days, 20 to 30 days, and 1 month to 6 months after the treatment. Measurement of urinary metabolites in the urine sample can be made in a similar way to the above. By measuring urinary metabolites before and after the treatment, it becomes possible to monitor the efficacy in the treatment with the therapeutic agent or method. This helps to consider the discontinuance, continuance, or change of the treatment on the basis of a result of the monitoring.
Furthermore, the cancer test method may be performed in combination with other conventional publicly known methods for the diagnosis of cancer. Such publicly known methods for the diagnosis of cancer include an imaging test (for example, such as ultrasonography, computer tomography (CT), X-ray radiography, and positron CT (PET)), endoscopy, a pathological examination with a biopsy, measurement of cancer markers in blood, etc.
On the basis of a result of the above evaluation, a doctor can make a diagnosis of the subject's cancer and perform an appropriate treatment. That is, the present invention also relates to a method of examining cancer in a subject and treating it. For example, in a case where cancer in a subject has been determined in accordance with the method according to the present invention, and the subject has been evaluated to be highly likely to have cancer, a treatment for treating cancer or preventing the progression of cancer in the subject may be performed. Furthermore, in a case where it has been evaluated that the stage of cancer in the subject has progressed or that cancer is highly likely to have a poor prognosis, the treatment may be continued, or a change in the method may be considered if necessary. Moreover, in a case where it has been evaluated that there is a high possibility that cancer is present in the subject, the presence of cancer may be confirmed by performing another method for the diagnosis of cancer such as the above methods. Furthermore, on the basis of results of the evaluations before and after the treatment, the efficacy in the treatment may be monitored, and the discontinuance, continuance, or change of the treatment may be determined. Moreover, in a case where it has been determined that there are no abnormalities, measurement of urinary metabolites can be made with time to follow up.
As a method for the treatment of cancer, surgery, radiotherapy, chemotherapy, immunotherapy, proton beam therapy, heavy ion radiotherapy, etc. can be performed either alone or in combination appropriately. The treatment of cancer can be appropriately selected by those skilled in the art in consideration of the type of cancer, the stage, the severity, the malignancy, the sex, the age and condition, the responsiveness to treatment, the genetic polymorphism (SNP) carried, etc.
The cancer test method of the present invention can be easily and simply performed by using a system. A cancer test system according to the present invention includes the following means:
The system of the present invention may preferably be a system in which the storage unit, the input unit, the analysis unit, and the evaluation unit described above are operably connected to one another so that the cancer test method of the present invention can be implemented.
Here, the input unit may be configured to receive an input of a measured value of a cancer marker; a user may input the measured value through any input device, or may import the measured value obtained by another device (for example, an apparatus for measuring a cancer marker, such as a mass spectrometer, an NMR spectrometer, or a liquid chromatography mass spectrometer (LC/MS) apparatus as an example), or may retrieve the measured value input to another system. The input unit may include a data analysis unit that is configured to analyze a measured value of a cancer marker of which the input has been received. For example, it may be configured to analyze whether data of the type of cancer marker, a measured value, subject information, etc. is suitable for subsequent analysis, and/or to perform data pretreatment.
The database of the storage unit has stored a marker panel on which information of multiple cancer markers with respect to multiple healthy subjects and cancer patients may be registered. The database may include, as information of the cancer markers, with respect to each of the cancer patients and the healthy subjects, at least the types of the cancer markers, the measured values of the cancer markers, and the discrimination information. Preferably, a cancer marker panel of the cancer patients and the healthy subjects may be stored in the database. Furthermore, the database may store data of multiple cancer markers with respect to cancer patients in a specific stage or with a specific degree of severity, or may store data of previous measured values or data of cancer patients before and after treatment. Furthermore, the database may include different databases according to the types of cancers.
The analysis unit may include a data analysis unit including software for processing a measured value obtained from the input unit and a calculator. The data analysis unit may be configured to create a marker panel of a subject on the basis of the measured value obtained from the input unit. The data analysis unit can include, for example, a signal display portion, a unit for analyzing a measured value, a computer unit, etc.
Furthermore, the analysis unit may be configured to read out information (for example, the discrimination information) of the cancer markers of the cancer patients and the healthy subjects from a storage device (a database) or something, and to analyze a correlation with measured values of cancer markers of a subject of which the inputs have been received by the input unit. At this time, the analysis unit may be configured to selectively read out an appropriate database according to the type of the cancer markers. Alternatively, in a case of temporal monitoring in the same subject, the analysis unit may be configured to read out the previous measured values from the storage device (the database) or something, and to compare the previous measured values with the measured values of the cancer markers of which the inputs have been received by the input unit.
Furthermore, the evaluation unit may be configured to evaluate cancer in the subject on the basis of a result of the analysis of the correlation made by the analysis unit. Here, the evaluation unit may be configured to acquire information indicating the presence or absence of cancer in the subject, the stage of cancer, no abnormalities, etc. For example, information that the subject is “cancer”, “healthy”, “close to cancer”, or “close to healthy” may be acquired. A preferred system may be a system that a user can use even without knowledge of a specialized clinician, and, for example, when data of measured values has been input to an input unit, the data is automatically analyzed, and a result of evaluation of cancer in a subject is displayed. At that time, together with the result of evaluation of cancer, a marker panel (a metabolite list) of the analyzed cancer markers may be displayed together.
The cancer test system of the present invention may further have a data storage unit, a data output and display unit, etc.
As an example of application of the present invention, a cancer test in a test center is described. In the test center, information of a cancer test is provided in response to a request or the like from a test subject. The test subject may make a choice about the number of markers for a test when applying for a primary test. This can also be used as an overall cancer test (various cancers are analyzed at a time) in combination with other markers.
Then, the test center delivers the test subject a test kit necessary for collection of urine. The test kit is sent by mail or the like as necessary. After the test subject receives the test kit, the test subject delivers or sends a specimen to the test center, or does it in some other way. In the test center, the specimen is cryonically preserved at about −80° C. as necessary for a subsequent test. In the test center, a primary test is performed according to the cancer test method described above, and a test result is sent to the test subject.
The test subject receives the result of the primary test, and may apply for a secondary test according to the content, or may have a more detailed diagnosis. This makes it possible to confirm the suspicion of cancer in the primary test and further identify the stage of cancer.
The present invention is specifically described below with examples; however, these examples are merely provided for description of the present invention, and do not limit or restrict the scope of the invention disclosed in the present application.
As an example showing the efficacy of the present method, evaluations of colorectal cancer in subjects of Mongoloid descent (30 colorectal cancer patients and 30 healthy subjects) were made. Urine samples were collected from the subjects, and urinary metabolites were detected by means of a liquid chromatography mass spectrometry (LC/MS).
(1) Creation of Metabolite List
With 60 subjects (30 colorectal cancer patients and 30 healthy subjects) on the vertical axis and 17 types of urinary tumor markers (metabolites) on the horizontal axis, a metabolite list was created.
(2) Creation of Evaluation Function Plot
As shown in
(3) Analysis by Determination Algorithm 1
On the basis of the plot in 2σ mode, an evaluation function plot in 1σ mode was created (the right side of
In 2σ mode, when an area where the evaluation function was greater than 22 was determined to be “cancer”, and an area where the evaluation function was smaller than 10 was determined to be “healthy”, the key point was what to make of subjects 11, 17, 18, 26, 27, 36, 51, and 60 of which the evaluation function was between 22 and 10 (the left side of
The above determination results are summarized in
Sensitivity=(30/30)×100=100%,
Specificity=(30/30)×100=100%.
Therefore, by the present method, cancer was able to be evaluated with high accuracy in both sensitivity and specificity.
There is described an example where colorectal cancer was evaluated using a distance function in addition to the evaluation function. In a case where whether a subject is a cancer patient or a healthy subject is determined by the distance function based on the determination algorithm 2, flows shown in
First, as with the determination algorithm 1, in the evaluation function plot calculated in 2σ, an area showing a high value considered to correspond to a cancer patient (the area where the evaluation function value was greater than 22) was determined to be “cancer”, and an area showing a low value considered to correspond to a healthy subject (the area where the evaluation function value was smaller than 10) was determined to be “healthy”.
Next, as for an area (22≥the evaluation function value≥10) where it was considered that cancer patients and healthy subjects were mixed, a distance function shown in
Also in this algorithm, the results were:
Sensitivity=(30/30)×100=100%,
Specificity=(30/30)×100=100%.
Therefore, by the present method, cancer was able to be evaluated with high accuracy in both sensitivity and specificity.
There may be a case where information about the severity of cancer is obtained by inspecting a metabolite list.
The characteristic feature of a cancer test based on an evaluation function is that it can be detected regardless of T1, T2, T3, and T4 indicating the depth of invasion according to the TNM classification.
As an example showing the efficacy of the present method, evaluations of breast cancer (in 30 breast cancer patients (25 pretreatment patients and 5 posttreatment patients) and 210 healthy subjects) were made. Urine samples were collected from the subjects, and urinary metabolites were detected by means of a liquid chromatography mass spectrometry (LC/MS). In the determination algorithm 1 and the determination algorithm 2, the efficacy similar to that of the case of colorectal cancer was obtained.
The determination results by the determination algorithm 1 are summarized in
Sensitivity=(28/30)×100=93.3%,
Specificity=(208/210)×100=99.0%.
However, it was determined that as for subjects 18 and 25 who are cancer patients determined to be close to healthy in the sensitivity, remeasurement is required for subject 18 because there was missing data, and subject 25 is under postoperative chemotherapy, thus a future evaluation is required. Therefore, by the present method, cancer was able to be evaluated with high accuracy in both sensitivity and specificity.
In an example where breast cancer is evaluated using a distance function in addition to the evaluation function, a flow shown in
First, as with the determination algorithm 1, in the evaluation function plot calculated in 2σ, an area showing a high value considered to correspond to a cancer patient (an area where the evaluation function value was greater than 40) was determined to be “cancer”, and an area showing a low value considered to correspond to a healthy subject (an area where the evaluation function value was smaller than 20) was determined to be “healthy”. Next, as for an area (40≥the evaluation function value≥20) where it was considered that cancer patients and healthy subjects were mixed, the distance function shown in
Results of actual analysis according to the determination algorithm 2 are shown in
The results of determination by the determination algorithm 2 (
Sensitivity=(28/30)×100=93.3%,
Specificity=(204/210)×100=97.1%.
However, it was determined that as for subjects 18 and 25 who are cancer patients determined to be close to healthy in the sensitivity, remeasurement is required for subject 18 because there was missing data, and subject 25 is under postoperative chemotherapy, thus a future evaluation is required. Therefore, by the present method, cancer was able to be evaluated with high accuracy in both sensitivity and specificity.
All publications, patents, and patent applications cited herein are hereby incorporated by reference in their entirety.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2020/034198 | 9/9/2020 | WO |
