JP2021131345A - Cancer identification method, identification device, and program - Google Patents

Abstract

To provide a cancer identification method, identification device, and program, which reduce burden on subjects and offer high accuracy in determining positive and negative.SOLUTION: A cancer identification method is provided, comprising acquiring amounts of substances of multiple compounds volatilizing from urine of a subject, and determining whether the subject is positive or negative for cancer on the basis of the amounts of the substances.SELECTED DRAWING: Figure 1

Description

The present invention relates to a discrimination method, a discrimination device and a program for discriminating positive or negative cancer based on a compound volatilized from urine.

Conventionally, as a method for examining / diagnosing cancer (for example, breast cancer), fine needle aspiration cytopathology or histology, which collects a small amount of cells or tissue fragments, and mammography (mammography) examination, which is a screening test (see Patent Document 1). ), Or a tumor marker (see Patent Document 2) or the like.

JP-A-2010-110469 Japanese Patent No. 4608432

However, fine needle aspiration cytopathology and histology inevitably invade the body, and there is a problem that the burden on the subject is heavy.

In addition, mammography examinations often cause pain and pain in X-ray photography, and it is hard to say that the burden on the examinee is low, and it has been pointed out that the examination alone may overlook the lesion.

Further, the tumor marker can be carried out relatively easily because the presence or absence of a cancer antigen is detected by a blood test, but there is a problem that the detection sensitivity is generally insufficient.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a cancer discrimination method, a discrimination device, and a program, which are less burdensome to the subject and have high accuracy of positive / negative discrimination.

The present invention is based on a step of obtaining the amount of each substance (hereinafter, referred to as "amount of substance to be inspected") for a plurality of types of compounds volatilized from the urine of one subject, and the amount of each substance to be inspected. , A method for discriminating cancer, which comprises a step of discriminating whether the subject is positive or negative for cancer.

Further, the present invention is a step of obtaining a plurality of physical quantities (hereinafter, referred to as “index physical quantities”) as indicators for each of a plurality of types of compounds that are derived from a living body and have a correlation with positive and negative cancer. A step of classifying the plurality of index physical quantities into a positive group and a negative group of cancer, and obtaining a boundary formula that formulates the boundary between the positive group and the negative group. The step of acquiring the physical quantity of each of the plurality of types of compounds (hereinafter, referred to as "physical quantity to be inspected") and the value obtained by substituting the physical quantity to be inspected into the boundary formula are acquired, and the inspection target is obtained according to the value. A method for determining cancer, which comprises a step of classifying a person into the positive group or the negative group.

Further, the present invention comprises an inspection target acquisition means for acquiring the amount of each substance (hereinafter, referred to as “inspection target substance amount”) for a plurality of types of compounds volatilized from the urine of one subject, and each of the above-mentioned inspections. It is a cancer discriminating device characterized by having a discriminating means for discriminating whether the subject is positive or negative for cancer based on the amount of the target substance.

In addition, the present invention obtains a plurality of physical quantities as indicators for each of a plurality of types of compounds that are derived from a living body and have a correlation with positive and negative cancers, and classify them into a positive group and a negative group. , The boundary formula acquisition means for acquiring the boundary formula obtained by formulating the boundary between the positive group and the negative group, and the physical quantities of the plurality of types of compounds to be tested obtained from the test subject (hereinafter, "test". A device for discriminating cancer, which comprises a discriminating means for classifying the test subject into a positive group or a negative group according to the obtained value by substituting the "target physical quantity") into the boundary formula. Is.

Further, the present invention is a program that causes a computer to execute the above-mentioned determination method.

According to the present invention, it is possible to provide a cancer discrimination method, a discrimination device, and a program with less burden on the subject and high accuracy of positive / negative discrimination.

It is a flow figure which shows an example of the flow of the discrimination method which concerns on embodiment of this invention. It is a schematic diagram which shows the correlation of two kinds of compounds which concerns on embodiment of this invention. It is a graph which shows the outline of the discriminant analysis which concerns on embodiment of this invention. It is a block diagram which shows the outline of the discrimination apparatus which concerns on embodiment of this invention. It is a table which shows the empirical result of the discrimination method which concerns on embodiment of this invention. It is a table which shows the empirical result of the discrimination method which concerns on embodiment of this invention. It is a table which shows the empirical result of (A) comparative example and (B) the discrimination method which concerns on embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

<How to identify cancer>
FIG. 1 is a flow chart showing an example of a processing flow of a cancer discrimination method according to the present embodiment. The method for discriminating the present embodiment is particularly suitable for discriminating breast cancer, and the cancer in the following description refers to breast cancer as an example. However, the discrimination method of the present embodiment is applicable not only to breast cancer but also to cancers of other sites.

In the method for discriminating cancer of the present embodiment, the amount of each substance of a plurality of predetermined compounds is obtained from a sample of one subject whose positive / negative cancer is unknown, and the substance is subject to the substance based on the amount of each substance. It determines whether the examiner is positive or negative for cancer.

Here, the plurality of types of compounds are, for example, substances that can be identified and quantified by an analytical means (for example, chromatography). The plurality of types of compounds are preferably volatile organic compounds (VOCs) that can be identified and quantified by gas chromatography.

The type (number) of the compound is not limited, but is, for example, two different compounds (first compound and second compound).

Further, the plurality of types (here, two types) of compounds are preferably compounds having a correlation such that they tend to be different from each other with respect to positive and negative cancer. Here, "having a correlation that tends to be different from that of positive and negative cancer" means, for example, that in the case of positive cancer, the first compound tends to increase more than the second compound, and the negative of cancer. In the case of, it means that the first compound has a correlation that tends to decrease more than the second compound.

In addition, a plurality of types of compounds are any of a compound contained in a biological secretion or excrement, a compound volatile from a biological secretion or excrement, a compound contained in the human body, and a compound contained in a sample possessed by the human body. However, in order to reduce the inspection burden on the subject, it is desirable that the sample can be easily obtained, and compounds that volatilize from excrement or body fluid, which are non-invasive samples, or are contained therein are more preferable. Is. Here, a compound that volatilizes from the urine of the human body will be described as an example.

In the present embodiment, as a plurality of types of compounds, the urine of a patient diagnosed with cancer (positive) (cancer sample) and the urine of a non-cancer diagnosed as non-cancer (negative for cancer) (non-cancer). The headspace gas of the sample) was measured with a gas chromatograph mass analyzer, and a peak having a significant difference between the cancer sample and the non-cancer sample was selected from 100 or more peaks.

Specifically, the first compound is, for example, 2-butanone (2-Butanone, ethyl methyl ketone, MEK), and the second compound is, for example, 2-propanol (2-propanol, propane-). 2-ol (propan-2-ol), isopropanol (isopropanol), isopropyl alcohol (isopropanol alcohol, IPA)).

As shown in FIG. 1, the method for discriminating cancer of the present embodiment is, more specifically, a step of acquiring a plurality of index physical quantities for each of a plurality of types of compounds (step S01) and cancer from the plurality of index physical quantities. A step of acquiring a boundary formula for classifying positive and negative of (step S03), a step of acquiring a physical quantity to be inspected (step S05), and a positive test subject (subject) according to the physical quantity to be inspected (step S03). It has a step (step S07) of classifying into a group) or a negative (group).

First, in step S01, a plurality of index physical quantities are acquired for each of the plurality of types of compounds. Here, the index physical quantity is a compound (first compound and second compound) of a plurality of types (here, the above two types) that serve as an index (sample) for determining whether cancer is positive or negative. It is a physical quantity.

That is, in advance, 2-butanone, which is the first compound (positive index first compound), which is a positive index, and the second compound (positive index second compound), which is a positive index, are obtained from a plurality of cancer samples, respectively. To obtain 2-propanol.

In addition, the physical quantity of 2-butanone, which is the first compound (negative index first compound), which is a negative index, and the second compound, which is a negative index (negative index first compound), are obtained from a plurality of non-cancer samples in advance. Obtain the physical quantity of 2-propanol (two compounds).

The individual physical quantities of the positive index first compound, the positive index second compound, the negative index first compound, and the negative index second compound are all index physical quantities, and these plurality of index physical quantities are collectively referred to as an index physical quantity group. In addition, a group of a plurality of index physical quantities serving as a positive index (a group of a plurality of index physical quantities of a positive index first compound and a group of a plurality of index physical quantities of a positive index second compound) is called a positive index physical quantity group, and a group of a plurality of index physical quantities serving as a negative index. (A group of index physical quantities of the negative index first compound and the negative index second compound) is called a negative index physical quantity group.

That is, both the index physical quantity of the first compound (positive first index physical quantity) and the index physical quantity of the second compound (positive second index physical quantity) obtained from a plurality of cancer samples are included in the positive index physical quantity group. included. In addition, the index physical quantity of the first compound (negative first index physical quantity) and the index physical quantity of the second compound (negative second index physical quantity) obtained from a plurality of non-cancer samples are both negative index physical quantities. include.

Here, the physical quantity of the present embodiment is the area value of each peak of the compound by mass spectrometry using, for example, a gas chromatography method in both positive and negative cases. Specifically, for example, urine (cancer sample and non-cancer sample) collected by a plurality of patients and non-patients is sterilized and filtered, and the amount is adjusted to an arbitrary amount in a measurement vial or the like, and then the headspace gas is subjected to gas chromatograph mass spectrometry. Measure with a meter. Then, 2-butanone and 2-propanol are identified for each individual sample (index sample) from the measurement results of the gas chromatograph mass spectrometer, and the area value of each peak is obtained.

FIG. 2 is a schematic diagram showing the distribution of the area values of the peaks of 2-butanone and 2-propanol thus obtained. The figure (A) is a boxplot showing the distribution of 2-butanone (the index physical quantity of the first compound), and the figure (B) shows the distribution of 2-propanol (the index physical quantity of the second compound). It is a box plot. In both the figures (A) and (B), the vertical axis is the peak area value, and the horizontal axis is the positive sample (cancer sample: P) on the left, that is, the data of the positive index physical quantity group is plotted. Then, the data of the negative sample (non-cancer sample: N), that is, the negative index physical quantity group is plotted on the right. At the bottom of each boxplot, the corresponding five-number summary (minimum value, first quartile, median, third quartile, maximum value) is shown.

As described above, the first compound (2-butanone) and the second compound (2-propanol) have a correlation with each other for positive / negative cancer, and specifically, in the case of positive, the first compound (2-propanol). One compound (2-butanone) is more than the second compound (2-propanol), and if cancer is negative, the first compound (2-butanone) is more than the second compound (2-propanol). There is a decreasing relationship. Alternatively, when the cancer changes from negative to positive, the physical quantity of the second compound (2-propanol) decreases and the physical quantity of the first compound (2-butanone) increases.

In this way, by selecting two compounds having a correlation that tends to be different from each other for positive and negative cancer, the same tendency can be obtained for positive and negative cancer. Compared with the case where there is such a correlation (for example, a correlation in which both physical quantities increase when the cancer changes from negative to positive), the variation in physical quantities due to the difference in concentration (for example, the concentration of the sample (urine)) It can be suppressed, and the limitation of the sample collection time can be eliminated. Therefore, it is possible to discriminate not only the first urine collection in the morning, which is desired in general health examinations, but also any time zone, and it is possible to reduce the troublesomeness of the subject.

Returning to FIG. 1 again, in step S03, a boundary formula for classifying the positive and negative of cancer from a plurality of index physical quantities (index physical quantity group) is acquired.

Specifically, for a plurality of patient samples (cancer sample) and a plurality of non-patient samples (non-cancer sample), 2-butanone (positive index first compound and negative index first compound) and 2-propanol (positive). The substance amounts of the index second compound and the negative index second compound) are shown in a scatter diagram, and the boundary of separation between the positive group (positive index physical quantity group) and the negative group (negative index physical quantity group) is formulated.

As an example, FIG. 3 is a scatter diagram of the index physical quantities of 2-butanone and 2-propanol, in which the vertical axis is the index physical quantity of 2-propanol and the horizontal axis is the index physical quantity of 2-butanone. In addition, the positive index physical quantity is indicated by a circle, and the negative index physical quantity is indicated by a square.

Then, these index physical quantity groups (positive index physical quantity group and negative index physical quantity group) are discriminated (classified and analyzed) by a predetermined discriminant method. Here, as an example, a linear discriminant analysis is performed, classified into a positive index physical quantity group and a negative index physical quantity group, and a straight line (linear (D) in the figure) that is the boundary between the two is mathematically expressed as a linear discriminant function, and a boundary formula (discriminant) is used. Expression) is obtained.

Equation 1 below is a general equation of a linear discriminant function with x and y as variables (when the variable is "2").
z = ax + by + c (Equation 1)

The discriminant analysis, factor formula 1 a, b, the proportion of intergroup variation S _B to the total variation S _T (correlation ratio eta ²⁾ is determined so as to maximize.

ここで、
ｚｉ：式１で示されるｚのｉ番目の個体（ここでは癌検体又は非癌検体）の値（判別得点）
である。 The total variation ST is _calculated by the following equation 2.

here,
zi: Value (discrimination score) of the i-th individual (here, a cancer sample or a non-cancer sample) represented by the formula 1.
Is.

である。 Further, the intergroup variation SB is _calculated by the following equation 3.

Is.

また、Ｓ_Ｔ＝ｎ、すなわちｚの分散は、解の任意性を排するため１とする。 The correlation ratio η ² is calculated by the following equation 4.

The dispersibility of S T _{= n,} i.e. z is 1 for Haisuru any of the solution.

Further, in the constant term c of the equation 1, the value of z is calculated by the equation 1 for all the individuals in the positive index physical quantity group and all the individuals in the negative index physical quantity group, and the midpoint of their average values is the origin. To decide.

Based on the above method, an example of a linear discriminant function whose variables are the physical quantities of 2-butanone and 2-propanol, which is obtained from a scatter plot as shown in FIG. 3, is shown in Equation 5 below. That is, the formula 5 is an example of the boundary formula (discriminant for classifying positive and negative) of the present embodiment.

Z = (-7.61 x ^10-5 ) X + (1.22 x 10 ^-4 ) Y + (-2.38 x 10 ^-1 ) (Equation 5)
here,
X: Physical quantity of 2-propanol (peak area value)
Y: 2-Physical quantity of butanone (peak area value)
Is.

The values of the coefficients (a, b, c) in Equation 5 are examples, and are values that vary depending on the individual values of the index physical quantities and the number of index physical quantities.

As described above, in the present embodiment, it is derived from the index physical quantities of the first compound (2-butanone) and the second compound (2-propanol) obtained from the cancer sample and the non-cancer sample as the index (model). Using the cancer positive and negative boundary formula (Equation 5), a subject whose positive / negative is unknown is discriminated by the following steps (steps S05 to S07).

In step S05, the physical quantity to be inspected is acquired. That is, each of the first compound (2-butanone) and the second compound (2-propanol) contained in the urine of one subject whose positive / negative cancer is unknown (for each subject). To get the amount of substance. Hereinafter, the physical quantity of the first compound obtained from the urine of the subject (specimen to be tested) is the physical quantity to be tested, the physical quantity of the second compound is the physical quantity to be tested, and both are collectively the physical quantity to be tested. That is.

Specifically, for example, the physical quantity to be inspected is obtained by sterilizing and filtering urine (specimen to be inspected) collected for each subject, making an arbitrary amount in a vial for measurement, and then gas chromatograph mass analysis of headspace gas. Measure with a meter. Then, 2-butanone and 2-propanol are identified for each individual sample from the measurement results of the gas chromatograph mass spectrometer, and the area value of each peak is obtained.

In step S07, the test subject (subject) is classified as positive or negative according to the physical quantity to be tested.

For each test target sample, the first test target physical quantity and the second test target physical quantity acquired in step S05 are substituted into the boundary equation represented by the formula 5, and the value of z (discrimination value) is obtained, whereby the test target sample is obtained. Determine whether to belong to the positive group or the negative group. As an example, here, when the value of z is 0 or positive (+), it is determined that the sample to be tested belongs to the negative group, and the subject to be tested (the subject to be tested) is determined to be negative for cancer. On the other hand, when the value of z is negative (−), it is determined that the sample to be tested belongs to the positive group, and the subject to be tested (the subject to be tested) is determined to be positive for cancer.

<Cancer discrimination device>
FIG. 4 is a schematic block diagram showing an example of the configuration of the cancer discrimination device 100 of the present embodiment. The cancer discrimination device 100 includes a boundary type acquisition means 101, an inspection target acquisition means 103, a discrimination means 105, an input means 107, an output means 109, a control means 111, and the like, and the discrimination method shown in FIG. Each step of is possible. Since the determination method is as described above, detailed description thereof will be omitted below.

The input means 107 is a means such as a keyboard or a touch panel display that receives information input from the outside, and the output means 109 is a means such as a monitor or a printer that outputs a discrimination result or the like.

The control means 111 is composed of a CPU, a RAM, a ROM, and the like, and controls each configuration of the discriminating device 100 to execute various controls. The CPU is a so-called central processing unit, and various programs are executed to realize various functions. The RAM is used as a work area of the CPU. The ROM stores the basic OS and programs executed by the CPU.

The boundary formula acquisition means 101 is derived from the amount of each indicator substance of a plurality of types of compounds (here, 2-butanone and 2-propanol) contained in the human body and having a correlation with positive and negative cancer. Obtain positive and negative boundary formulas. More specifically, the index substance amount group for each of the plurality of types of compounds is obtained in advance from a plurality of cancer samples (urine of a cancer patient) and a plurality of non-cancer samples (urine of a non-cancer patient) as indicators. , The index physical quantity group is classified into a positive group and a negative group. Then, a boundary formula (formula 5 described above) that formulates the boundary between the positive group and the negative group is obtained by discriminant analysis.

The test target acquisition means 103 indicates the amount of each substance (test target substance) for a plurality of types of compounds (here, 2-butanone and 2-propanol) contained in the urine (test target sample) of one subject. Amount of substance).

The discriminant means (classification means) 105 discriminates (classifies) whether the test subject (test subject) is positive or negative for cancer based on the physical quantity to be tested obtained from the test target sample. That is, the discriminating means 105 calculates the value (the value of z in the formula 5 and the discriminant value) by substituting (each of) the physical quantity to be inspected into the boundary formula (Equation 5) acquired by the boundary formula acquiring means 101. , The test target is classified into a positive group or a negative group according to the obtained discriminant value. As an example, when the value of z is 0 or positive (+), it is determined that the test target sample belongs to the negative group, and the test subject (test subject) is determined to be negative for cancer. On the other hand, when the value of z is negative (−), it is determined that the sample to be tested belongs to the positive group, and the subject to be tested (the subject to be tested) is determined to be positive for cancer.

The discriminating device 100 may include, for example, the analytical means 115. The analysis means 115 is, for example, a chromatography means for performing mass spectrometry (for example, a gas chromatograph mass spectrometer), and the chromatography means may be included in the boundary type acquisition means 101 and the inspection target acquisition means 103. In this case, the index substance amount group is obtained by measuring the headspace gas of each of a plurality of cancer samples and non-cancer samples by an analytical means (chromatography means) 115. Further, the amount of the substance to be inspected is obtained by measuring the headspace gas of the sample to be inspected by the analysis means 115.

Alternatively, the analysis means 115 may be separately provided outside the discrimination device 100. In that case, each data of the index substance amount group may be input to the discriminating device 100 via the input means 107, and the boundary type acquisition means 101 may acquire each data of the index substance amount group. Further, the (each) data of the physical quantity to be inspected may be input to the discrimination device 100 via the input means 107, and the inspection target acquisition means 103 may acquire the (each) data of the physical quantity to be inspected.

In addition, although FIG. 4 is shown as a block diagram in which each configuration is integrally included inside the discrimination device 100, one of the configurations (for example, inspection target acquisition means 103, etc.) is separately provided outside. It may be a configuration.

Further, the program of the present embodiment is a program that causes a computer to execute the determination method shown in FIG.

That is, the discrimination device 100 may be a device (computer) equipped with a program for causing the computer to execute the discrimination method of the present embodiment.

In the discrimination method / discrimination device of the present embodiment described above, both sensitivity and specificity are 80% or more, which is extremely high reliability. Furthermore, since the collection of the sample is non-invasive, the burden on the subject can be significantly reduced.

The results of demonstrating the discrimination method of the present embodiment will be described with reference to FIGS. 5 to 7. First, FIG. 5 is a table showing the determination results of the boundary equation of Equation 5. The amount of substance (peak area value) of 2-butanone and 2-propanol for a plurality of cancer samples and a plurality of non-cancer samples for which the figure (A) serves as an index (model) for creating the boundary formula (formula 5). Is substituted into the boundary equation (Equation 5) to obtain a value (discrimination value) and verified. Further, in FIG. 3B, the amount of substance (peak area value) of 2-butanone and 2-propanol was obtained for the sample to be tested whose positive / negative was unknown, and the value was substituted into the boundary equation (Equation 5). It is the result of discrimination by (discrimination value).

The "true condition" in the table refers to the condition of each person determined by the result of a detailed test such as a biopsy, and the "test result" is the positive / negative discrimination result by the discrimination method of the present embodiment. ..

In addition, "sensitivity" is the rate at which a certain test correctly determines a person with cancer as "positive", and high sensitivity means that there are few oversights of the test method. In addition, "specificity" is the rate at which a certain test correctly determines a person without cancer as "negative", and high specificity is a false positive (despite the absence of cancer). It means that there are few (those that are judged to be "positive" in the test), and it is the most important index in the case of screening for cancer.

With reference to FIG. 3A, a plurality of cancer specimens and a plurality of non-cancer specimens, which serve as an index (model) for creating the boundary formula (Equation 5), are positive as true states (here, the breast cancer stage). The number of samples (1,2) is 60 (57 + 3), and the number of negative samples is 59 (9 + 50).

The result of discriminating (re-verifying) the samples that serve as these indexes by the boundary formula (Equation 5) is the number of positive samples that are judged to be "positive" (that is, the number of true positives) in the true state (the number of true positives). (A)) is 57, the number of positive samples in the true state as "negative" (that is, false judgment) (the number of false negatives (c)) is 3, and the sensitivity (a / (a + c)) is 95. It was 0.0% (= 57/60 × 100).

In addition, the number of negative samples in the true state is determined as "positive" (that is, false judgment) (the number of false positives (b)) is 9, and the number of negative samples in the true state is determined as "negative" (that is, positive). The number (determined) (the number of true negatives (d)) was 50, and the specificity (d / (b + d)) was 84.7% (= 50/59 × 100).

In addition, among those judged as "positive" by the boundary formula (formula 5), the ratio of true positives (positive predictive value (a / (a + b))) was 86.4% (= 57 / (57 + 9)). ) × 100), and among those judged as “negative” by the boundary formula (Equation 5), the proportion of those who were truly negative (negative predictive value (d / (c + d))) was 94.3% ( = 50 / (50 + 3) × 100), and the accuracy ((a + d) / (a + b + c + d)) was 89.9%.

With reference to FIG. 3B, the number of positive (Breast stages 0-4, unknown) samples is 59 (52 + 7) and the number of negative samples is 117 (13 + 104) as true states. The case was verified.

As a result of discriminating these test target samples by the boundary formula (Equation 5), the number of positive samples in the true state was judged to be "positive" (that is, positive judgment) was 52, and the number of positive samples in the true state was 52. The number of "negative" determinations (that is, erroneous determinations) was 7, and the sensitivity was 88.1%.

In addition, the number of negative samples in the true state as "positive" (that is, erroneous judgment) is 13, and the number of negative samples in the true state as "negative" (that is, positive judgment) is 104, which is specific. The degree was 88.9%. In this case, the positive predictive value was 80.0%, the negative predictive value was 93.7%, and the accuracy was 88.6%.

That is, as shown in Fig. (B), an accuracy of nearly 90% can be obtained even for a sample to be tested other than the sample (model) used as an index when acquiring the boundary formula (Equation 5). I understand.

Next, FIG. 6 is a table showing the results of verifying the reproducibility of the discrimination of the boundary equation (Equation 5). A plurality of cancer samples and a plurality of non-cancer samples, which serve as an index (model) for preparing the boundary formula (formula 5), were measured three times in total on different measurement dates, and discriminated by the boundary formula (formula 5).

N1 to n10 in the upper row (column P) are 10 cancer samples, and the true state (positive: P) and (stages 1 and 2) are described for each cancer sample. In addition, the first to third discrimination by the boundary formula (Equation 5) and the correctness (correct: ○, incorrect: ×) are described for each cancer sample.

Further, n1 to n10 in the lower row (column N) are 10 non-cancer samples, and the true state (negative: N) is described for each non-cancer sample. In addition, for each non-cancer sample, the first to third discrimination by the boundary formula (Equation 5) and the correctness (correct: ○, incorrect: ×) are described. As a result, as shown in the figure, the concordance rate (judgment recall rate) was 100% for both sensitivity and specificity.

FIG. 7 is a table comparing the accuracy of screening results by age group, and FIG. 7 (A) is a list of sensitivity and specificity by age group by mammography examination (Source: Age-specific interval breast cancers in Japan: estimation of the). proper sensitivity of screening using a population-based cancer registry. Suzuki A1, Kuriyama S, Kawai M, Amari M, Takeda M, Ishida T, Ohnuki K, Nishino Y, Tsuji I, Shibuya D, Ohuchi N. Cancer Sci. 2008 Nov ; 99 (11): 2264-7.). In addition, FIG. 3B is a list of results (sensitivity and specificity) of discrimination by age according to the discrimination method of the present embodiment. All of the samples of the present embodiment are test target samples prepared again for the test.

In general, it has been pointed out that Japanese people in their 40s have a high rate of high-concentration breast cancer, and that there is a possibility that breast cancer may be overlooked even if a mammography test is performed. It remains at about 70%.

On the other hand, as shown in FIG. 3B, in the discrimination method of the present embodiment, the sensitivity in the 40s or younger is 89.3% and the specificity is 82.9%, both of which have high accuracy of more than 80%. rice field.
It can be said.

In addition, excellent results were obtained even when compared with conventional tumor markers and the like. Specifically, the above-mentioned Patent Document 2 (Patent No. 4608432) describes the positive rates of urinary DiAcSpm, serum CEA, and serum CA15-3 as tumor markers by breast cancer stage, and the average value of each is The values were 60.2%, 37.3%, and 37.3%, and the results of comparing these with the accuracy of FIG. 5 also showed that the accuracy of the discrimination method of the present embodiment was good.

As described above, 2-butanone and 2-propanol have been exemplified as two kinds of compounds in the present embodiment, but they have a correlation that tends to be different from each other for positive and negative cancers, and an analytical means (for example, gas chromatography). It is not limited to this as long as it is a compound that can be identified and quantified by the above. For example, compounds such as alcohols, ketones, and aromatics can be applied.

In addition, a plurality of types of compounds are any of a compound contained in a biological secretion or excrement, a compound volatile from a biological secretion or excrement, a compound contained in the human body, and a compound contained in a sample possessed by the human body. It should be. For example, it may be a compound contained in a volatile component from tears, saliva, or sweat, or a compound contained in exhaled breath, skin gas, or the like.

In addition, the discrimination method and the discrimination device of the present embodiment classify into a positive group and a negative group according to the amount of each substance (index substance amount) of a plurality of types of compounds as indicators, and demarcate the boundary between the positive group and the negative group. If a formulated boundary formula is obtained, the physical quantity to be tested is applied to the boundary formula, and the test target (subject) is divided into a positive group and a negative group according to the obtained value (discrimination value). It is not limited to the above example.

That is, the plurality of types of compounds may be two or more types. In the above example, since the discrimination is performed by two kinds of compounds, the boundary (boundary formula) is represented by a straight line, but when the discrimination is performed by three or more kinds of compounds, the boundary (boundary formula) is not a straight line but a surface. It is represented by.

The classification (discriminant) of the positive group and the negative group is not limited to the linear discriminant analysis, for example, multiple logistic regression analysis, [0] discriminant analysis based on the distance of Maharanobis, or the decision tree method / MT (Maharanobis-Taguchi) method. -It may be performed by an analysis method such as a support vector machine.

Further, the amount of substance of each compound is not limited to the peak area value of the compound, and may be a waveform of gas chromatography or a signal intensity.

In addition, the identification of each compound is not limited to gas chromatography mass spectrometry, but also liquid chromatography mass spectrometry, various chromatographys, analysis by FT-IR (Fourier transform infrared spectrophotometer), and analysis by ultraviolet-visible spectrophotometer. , Other various mass spectrometers and various measuring devices may be used. That is, the analysis means 115 may be liquid chromatography-mass spectrometry, various chromatographys, Fourier transform infrared spectrophotometer, ultraviolet-visible spectrophotometer, other various mass spectrometers, various measuring devices, and the like.

It should be noted that the present invention is not limited to the above-described embodiment, and it goes without saying that various modifications can be made without departing from the gist of the present invention.

100 Discrimination device 101 Boundary type acquisition means 103 Inspection target acquisition means 105 Discrimination means 107 Input means 109 Output means 111 Control means 115 Analytical means

Claims (21)

A step of obtaining the amount of each substance (hereinafter referred to as "the amount of substance to be inspected") for a plurality of types of compounds volatilized from the urine of one subject, and
A step of determining whether the subject is positive or negative for cancer based on the amount of each substance to be tested, and
A method for discriminating cancer, which is characterized by having. Judgment is made using a positive and negative boundary formula derived from the amount of substance that serves as an index for each of the plurality of types of compounds (hereinafter referred to as "amount of substance").
The method for determining cancer according to claim 1. A step of obtaining a plurality of the index substance amounts (hereinafter, referred to as "index substance amount group") for each of the plurality of types of compounds volatilized from the urine of a cancer patient and a non-patient in advance.
The step of acquiring the boundary formula by the discriminant analysis of the index substance amount group, and
A claim characterized in that each subject has a step of substituting the amount of the substance to be inspected for each of the plurality of types of compounds into the boundary formula and determining positive or negative based on the obtained value. Item 2. The method for determining cancer according to item 2. The plurality of compounds are two compounds having a correlation such that they tend to be different from each other for positive and negative cancer.
The method for determining cancer according to any one of claims 1 to 3, wherein the method is characterized by the above. The two compounds have a correlation that when positive, one tends to increase more than the other, and when negative, one tends to decrease more than the other.
The method for discriminating cancer according to claim 4. The amount of substance is the area value of each peak of the plurality of compounds by mass spectrometry.
The method for determining cancer according to any one of claims 1 to 5, wherein the method is characterized by the above. The mass spectrometry is an analysis by a gas chromatography method.
The method for discriminating cancer according to claim 6. The cancer is breast cancer,
The method for determining cancer according to any one of claims 1 to 7, wherein the method is characterized by the above. The plurality of compounds are 2-propanol and 2-butanone.
The method for determining cancer according to any one of claims 1 to 8, wherein the method is characterized by the above. A step of acquiring a plurality of physical quantities (hereinafter referred to as "index physical quantities") that are indicators for each of a plurality of types of compounds that are derived from a living body and have a correlation with positive and negative cancers.
A step of classifying the plurality of index physical quantities into a positive group and a negative group of cancer, and obtaining a boundary formula that formulates the boundary between the positive group and the negative group.
A step of obtaining the physical quantity of each of the plurality of types of compounds to be inspected (hereinafter referred to as "physical quantity to be inspected") from the person to be inspected, and
It has a step of acquiring a value obtained by substituting the physical quantity to be tested into the boundary formula, and classifying the test subject into the positive group or the negative group according to the value.
A method for discriminating cancer, which is characterized by the fact that. An inspection target acquisition means for acquiring the amount of each substance (hereinafter referred to as "inspection target substance amount") for a plurality of types of compounds volatilized from the urine of one subject.
A discriminating means for determining whether the subject is positive or negative for cancer based on the amount of each substance to be tested, and
A cancer discriminating device characterized by having. The discriminating means discriminates using a positive and negative boundary formula derived from a plurality of substance amounts (hereinafter referred to as "index substance amounts") that serve as indicators for each of the plurality of types of compounds.
The cancer discrimination apparatus according to claim 11. A plurality of the index substance amounts (hereinafter, referred to as “index substance amount group”) are obtained in advance for each of the plurality of types of compounds volatilized from the urine of cancer patients and non-patients, and the index substance amount group is discriminated and analyzed. Has a boundary formula acquisition means for acquiring the boundary formula by
The discriminating means calculates a discriminant value by substituting the amount of the substance to be tested for each of the plurality of types of compounds into the boundary formula for each subject, and discriminates between positive and negative based on the discriminant value. ,
The cancer discrimination apparatus according to claim 12. The plurality of compounds are two compounds having a correlation such that they tend to be different from each other for positive and negative cancer.
The cancer discrimination apparatus according to claim 11 to 13. The two compounds have a correlation that when positive, one tends to increase more than the other, and when negative, one tends to decrease more than the other.
The cancer discrimination apparatus according to claim 14. The amount of substance is the area value of each peak of the plurality of compounds by mass spectrometry.
The cancer discrimination apparatus according to any one of claims 11 to 15. The inspection target acquisition means includes an analysis means for performing the mass spectrometry.
The cancer discrimination apparatus according to claim 16. The cancer is breast cancer,
The cancer discriminating device according to any one of claims 11 to 17, wherein the cancer discriminating device is characterized. The plurality of compounds are 2-propanol and 2-butanone.
The cancer discrimination apparatus according to any one of claims 11 to 18. For each of a plurality of types of compounds that are derived from a living body and have a correlation with positive and negative cancers, a plurality of physical quantities as indicators are obtained and classified into a positive group and a negative group, and the positive group and the above are described. Boundary expression acquisition means for acquiring a boundary expression that formulates the boundary of a negative group,
The physical quantity of each of the plurality of types of compounds to be inspected (hereinafter referred to as "physical quantity to be inspected") obtained from the inspected person is substituted into the boundary formula, and the inspected person is according to the obtained value. Has a discriminating means for classifying the group into a positive group or a negative group.
A cancer discrimination device characterized by the fact that. A program that causes a computer to execute the determination method according to any one of claims 1 to 10.

Images