System for determining whether a thyroid tissue lesion is a benign lesion or a malignant lesion and use thereof

Abstract

The application provides a system for determining whether a thyroid tissue lesion is a benign lesion or a malignant lesion and application thereof, the system is used for simultaneously detecting expression levels of MET gene, TIMP1 gene, TGFA gene, CITED1 gene and FN1 gene in thyroid tissue, and a model is constructed based on detection results of the five genes to determine whether the thyroid tissue lesion is a benign lesion or a malignant lesion. The application firstly constructs an arithmetic model for calculating mRNA expression of five genes based on high expression of 15 genes in thyroid cancer tissue, and obtains a final scoring model for distinguishing thyroid malignant canceration and benign lesion, and the scoring value can be more accurate than cytology and gene mutation methods.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of disease detection, and particularly relates to a system for determining whether a thyroid tissue lesion is a benign lesion or a malignant lesion and application thereof. BACKGROUND

[0002] Thyroid cancer is a malignant tumor originating from thyroid follicular epithelium or parafollicular epithelial cells, and is also the most common malignant tumor in the head and neck. According to the origin and differentiation of the tumor, thyroid cancer is divided into: papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC), medullary thyroid carcinoma (MTC), poorly differentiated thyroid carcinoma (PDTC), and anaplastic thyroid cancer (ATC), among which PTC is the most common, accounting for about 80-90% of all thyroid cancers, and PTC and FTC are collectively referred to as differentiated thyroid carcinoma (DTC). The incidence of DTC accounts for more than 95% of thyroid cancer, and the remaining MTC, PDTC and ATC are relatively rare.

[0003] Most thyroid nodule patients have no clinical symptoms. Usually, thyroid palpation and neck ultrasound examination are found during physical examination. When thyroid nodules are found by ultrasound and other methods, biopsy puncture is often needed to take the cells of the nodules for observation in order to confirm the nature of the lesion. However, about 30% of the specimens observed by cytology and pathology of the biopsy specimens still cannot be identified as malignant or benign lesions. Therefore, molecular biology methods are applied to thyroid biopsy specimens to determine whether they are benign or malignant lesions. The first molecular biology method used includes detecting gene mutations or gene rearrangements at the DNA level in the puncture specimens, such as BRAF, RAS gene mutations, and RET gene rearrangements. With the accumulation of data, studies have found that the detection of gene mutations and gene rearrangements has insufficient sensitivity, and about 1 / 3 of the samples cannot detect these common gene mutations and gene rearrangements even if they are malignant carcinomas, so recently high-throughput techniques have been used to detect a large number of gene mutations or gene rearrangements, which greatly increases the detection cost.

[0004] Then there are clinical applications that use gene expression in cells from biopsy puncture to determine whether it is cancerous, including the American commercial company Afirma in California using microarray chip technology to detect the expression of 167 genes to diagnose whether the biopsy puncture specimen is a benign or malignant tumor, see Alexander EK, Kennedy GC, Baloch ZW, Cibas ES, Chudova D, Diggans J, Friedman L, Kloos RT, LiVolsi VA, Mandel SJ et al: Preoperative diagnosis of benign thyroid nodules with indeterminate cytology. N Engl J Med 2012, 367(8):705-715; there is also a commercial product that detects the expression of 40 genes to distinguish between benign and malignant thyroid biopsy tissues, see Pluciennik A, Placzek A, Wilk A, Student S, Oczko-Wojciechowska M, Fujarewicz K: Data Integration-Possibilities of Molecular and Clinical Data Fusion on the Example of Thyroid Cancer Diagnostics. Int J Mol Sci 2022, 23(19). But the number of genes used is too large, the cost is too high, and it is difficult to meet the needs of clinical application. SUMMARY

[0005] Therefore, in order to solve the above technical problems, the present application provides a system for determining whether a thyroid tissue lesion is a benign lesion or a malignant lesion, which is used to simultaneously detect the expression levels of MET gene, TIMP1 gene, TGFA gene, CITED1 gene and FN1 gene in thyroid tissue, and a model constructed based on the detection results of the above five genes to determine whether the thyroid tissue lesion is a benign lesion or a malignant lesion.

[0006] In an embodiment, the primers and probes for detecting MET gene are SEQ ID NO. 1, SEQ ID NO. 2 and SEQ ID NO. 3, respectively, the primers and probes for detecting TIMP1 gene are SEQ ID NO. 10, SEQ ID NO. 11 and SEQ ID NO. 12, respectively, the primers and probes for detecting TGFA gene are SEQ ID NO. 13, SEQ ID NO. 14 and SEQ ID NO. 15, respectively, the primers and probes for detecting CITED1 gene are SEQ ID NO. 19, SEQ ID NO. 20 and SEQ ID NO. 21, respectively, and the primers and probes for detecting FN1 gene are SEQ ID NO. 22, SEQ ID NO. 23 and SEQ ID NO. 24, respectively.

[0007] In an embodiment, the system is further used for detecting the expression level of the internal reference gene GAPDH, and the primers and probes for detecting the human internal reference gene GAPDH are SEQ ID NO. 46, SEQ ID NO. 47 and SEQ ID NO. 48, respectively.

[0008] In an embodiment, the system is used for preparing a product for determining whether a thyroid tissue lesion is a benign lesion or a malignant lesion.

[0009] In an embodiment, a primer, a probe or a combination thereof is provided, which is used for simultaneously detecting the expression levels of MET gene, TIMP1 gene, TGFA gene, CITED1 gene and FN1 gene in a thyroid tissue, and a model is constructed based on the detection results of the five genes to determine whether a thyroid tissue lesion is a benign lesion or a malignant lesion.

[0010] The present application is the first time to construct a scoring model for differentiating thyroid malignant cancer and benign lesion based on the high expression of 15 genes in thyroid cancer tissues, the mRNA expression of 5 genes, the construction of an arithmetic model for the 5 mRNAs, and the scoring value. The method is more accurate than cytology and gene mutation method, and the combination of the 5 genes (MET+TIMP1+TGFA+CITED1+FN1) achieves the best sensitivity, specificity and AUC area. DETAILED DESCRIPTION

[0011] Genes highly expressed in thyroid cancer tissues were retrieved from public databases and published literature, identifying 15 genes highly expressed in thyroid cancer tissues: MET, SERPINA1, KRT19, TIMP1, TGFA, QPCT, CITED1, FN1, EPS8, PROS1, LRP4, PSD3, SDC4, ETV5, and CD44. Primers and probes were designed for the mRNA expressed by each gene, requiring that at least one primer or probe contain a transgenomic intron to ensure that mRNA detection is not affected by residual genomic DNA in the extracted nucleic acid. Then, thyroid biopsy specimens confirmed by cytology or postoperative pathology to be malignant or benign lesions were tested. The mRNA of each gene and the mRNA of the internal reference gene GAPDH were detected simultaneously. Finally, the mRNA expression of 5 genes was screened out, and an arithmetic model for calculating these 5 mRNAs was constructed to obtain a scoring model that can ultimately distinguish between malignant and benign thyroid lesions. Through the scoring value, the malignant and benign lesions can be distinguished from thyroid biopsy tissue more accurately than cytological and gene mutation methods. The primers and probes designed for detecting the mRNA expression levels of the 15 genes and the internal reference gene are shown in Tables 1 and 2.

[0012] Table 1

[0013]

[0014] Table 2

[0015]

[0016] The qPCR reaction system for the above primers and probes used a premixed qRT-PCR reaction system produced by Novizan. The specific reaction system is shown in Table 3 below.

[0017] Table 3

[0018]

[0019] Add water to 20μl.

[0020] The reaction conditions are shown in Table 4 below.

[0021] Table 4

[0022]

[0023] Fluorescence was collected at 61℃. The Ct values ​​of the internal reference gene and the target gene were recorded.

[0024] The probes for each gene and the internal reference gene GAPDH were combined for detection. The probes for the target gene were labeled with FAM fluorescent dye, and the probes for the GAPDH gene were labeled with VIC fluorescent dye. Based on the Ct values ​​of the target gene and the GAPDH gene, the relative expression value of the target gene and GAPDH was obtained. （GAPDH的Ct-靶基因的Ct） *100, and then an expression score is obtained based on the relative abundance of each target gene.

[0025] Nucleic acid extracted from 159 biopsy specimens (including thyroid cancer and 58 benign lesions) was used as training data. The expression abundance of the above 15 genes was detected using reverse transcription quantitative PCR, and the relative expression value of each gene was obtained by comparing it with the internal reference gene GAPDH. This training data was then input into a logistic regression model to combine different genes and assign a judgment parameter to each gene to distinguish between cancerous and benign lesion specimens in the training data with maximum accuracy. The probability of cancer in the sample was calculated based on the score: p = 1 - 1 / (1 + EXP(-score)). A p-value < 0.50 was considered benign, and a p-value >= 0.50 was considered malignant.

[0026] Then, using the gene combinations and interpretation parameters obtained from the training data, we interpreted a new set of test specimens (134 cases of thyroid cancer and 38 cases of benign lesions). Among the test specimens, we listed the sensitivity and specificity of the 6 gene combinations with the highest accuracy as follows.

[0027] Here, we list the six gene combinations with the highest accuracy ranking to obtain the Score scoring formula. The number preceding each target gene in the calculation formula is calculated by the logistic regression model based on known clinical specimen groupings. For example, the FN1 gene has a relatively high expression abundance based on the Ct value. During the logistic regression model calculation, it was found that the coefficient preceding the FN1 gene should be relatively small. Only then can the score distinguish between malignant cancer and benign lesions. Furthermore, the p-value is calculated from the Score, and then the sensitivity and specificity of thyroid cancer are determined based on the p-value.

[0028] Combination 1: MET+TIMP1+TGFA+CITED1+FN1 combination;

[0029] Score=2.96-0.056*MET-0.0012*TIMP1+0.02*TGFA-0.062*CITED1-0.0002*FN1

[0030] Combination 2: KRT19+TIMP1+TGFA+QPCT+CITED1+FN1+EPS8 combination:

[0031] Score= 2.96-0.002*KRT19-0.0018*TIMP1+0.02*TGFA+0.06*QPCT-0.05*CITED1-0.001*FN1+0.01*EPS8

[0032] Combination 3: SERPINA1+TIMP1+TGFA+CD44+FN1 combination:

[0033] Score=2.96-0.02* SERPINA1-0.0012*TIMP1+0.018*TGFA+0.03*CD44-0.00018*FN1

[0034] Combination 4: CITED1+EPS8+PSD3+SDC4+ETV5

[0035] Score=2.96-0.08* CITED1+0.02*EPS8+0.011*PSD3+0.01*SDC4-0.04*ETV5

[0036] Combination 5: MET+CITED1+FN1+EPS8+PROS1+LRP4

[0037] Score=2.96-0.04* MET-0.05*CITED1-0.001*FN1+0.034*EPS8+0.02*PROS1+0.03*LRP4

[0038] Combination 6: SERPINA1+KRT19+QPCT+TGFA+LRP4+SDC4

[0039] Score=2.96-0.02* SERPINA1-0.001*KRT19+0.05*QPCT+0.02*TGFA+0.028*LRP4+0.01*SDC4

[0040] The score values, p-values, and pathological concordance of 134 clinically diagnosed thyroid cancer specimens and 38 benign lesion specimens in combination 1 are shown in Tables 5, 6, 7, 8, 9, and 10 below.

[0041] Table 5

[0042]

[0043] Table 6

[0044]

[0045] Table 7

[0046]

[0047] Table 8

[0048]

[0049] Table 9

[0050]

[0051] Table 10

[0052]

[0053] The sensitivity and specificity of combination 1 are calculated based on the score and p-value data in the table above, as shown in the table below. Similarly, for combination 2, combination 3, combination 4, combination 5 and combination 6, the score and p-value are calculated (the original data for calculating the score and p-value are not shown in this application), thereby obtaining the corresponding sensitivity and specificity data, as shown in Table 11.

[0054] Table 11

[0055]

[0056] The sensitivity and specificity of interpreting thyroid cancer based on p-values ​​are as follows: only combination 1 (MET+TIMP1+TGFA+CITED1+FN1 combination) of 5 genes achieved the highest sensitivity and specificity.

[0057] In fields such as medical diagnosis and machine learning model evaluation, sensitivity, specificity, and AUC area are important indicators for measuring the performance of models or diagnostic methods. Sensitivity refers to the proportion of individuals who are correctly identified as having the disease (or being positive) among all actual infected (or positive) individuals. It reflects the model's ability to "not miss diagnoses"; higher sensitivity means it can identify more individuals who are actually infected. Specificity refers to the proportion of individuals who are correctly identified as not having the disease (or being negative) among all actual uninfected (or negative) individuals. It reflects the model's ability to "not misdiagnose"; higher specificity means it can exclude individuals who are not actually infected (e.g., in scenarios where overtreatment needs to be avoided, high specificity can reduce unnecessary intervention). AUC (Area Under the ROC Curve) is the area under the ROC curve (Receiver Operating Characteristic Curve), which is a curve plotted with the false positive rate (FPR) on the horizontal axis and the true positive rate (TPR, i.e., sensitivity) on the vertical axis. The AUC value ranges from 0 to 1, representing the model's overall ability to distinguish between "positive" and "negative" results. The closer the AUC is to 1, the stronger the model's ability to distinguish between positive and negative results. AUC is unaffected by the diagnostic threshold (the cutoff value for judging "positive / negative") and comprehensively reflects the model's overall performance at different thresholds, making it suitable for comparing the merits of different models. In summary, sensitivity and specificity are indicators for specific thresholds, focusing on "no missed diagnoses" and "no false diagnoses," respectively. AUC is a comprehensive indicator that reflects the model's overall discriminative ability across all possible thresholds.

[0058] The sensitivity and specificity of interpreting thyroid cancer based on p-values ​​are as follows: Only combination 1 (MET+TIMP1+TGFA+CITED1+FN1 combination) of these 5 genes achieved the best sensitivity, specificity and AUC area.

[0059] It will be readily understood by those skilled in the art that the aforementioned advantageous methods can be freely combined and superimposed without conflict.

[0060] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention. The above are merely preferred embodiments of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the protection scope of the present invention.

Claims

1. Use of a system in the manufacture of a product for determining whether a thyroid tissue lesion is a benign lesion or a malignant lesion, characterized in that, The system is used for detecting the expression levels of MET gene, TIMP1 gene, TGFA gene, CITED1 gene and FN1 gene in thyroid tissue simultaneously, so as to determine whether the thyroid tissue lesion is a benign lesion or a malignant lesion; The sequences of the upstream and downstream primers and the probe for detecting the MET gene are respectively shown as SEQ ID NO. 1, SEQ ID NO. 2 and SEQ ID NO. 3, the sequences of the upstream and downstream primers and the probe for detecting the TIMP1 gene are respectively shown as SEQ ID NO. 10, SEQ ID NO. 11 and SEQ ID NO. 12, the sequences of the upstream and downstream primers and the probe for detecting the TGFA gene are respectively shown as SEQ ID NO. 13, SEQ ID NO. 14 and SEQ ID NO. 15, the sequences of the upstream and downstream primers and the probe for detecting the CITED1 gene are respectively shown as SEQ ID NO. 19, SEQ ID NO. 20 and SEQ ID NO. 21, and the sequences of the upstream and downstream primers and the probe for detecting the FN1 gene are respectively shown as SEQ ID NO. 22, SEQ ID NO. 23 and SEQ ID NO.

24.

2. Use according to claim 1, characterized in that, The system is also used for detecting the expression level of the reference gene GAPDH, and the sequences of the upstream and downstream primers and the probe for detecting the human reference gene GAPDH are respectively shown as SEQ ID NO. 46, SEQ ID NO. 47 and SEQ ID NO. 48.

Images