WO2016205911A1 - In vitro method for detecting thyroid cancer in a patient, in vitro method for differentiating between thyroid cancer and normal thyroid tissue or benign thyroid lesions, use of a series of genes, method for obtaining data determining the treatment of thyroid cancer, laboratory kit and device for classifying a biological sample from the thyroid gland as malignant or benign - Google Patents

Abstract

The present invention relates, in general, to an in vitro method for detecting thyroid cancer in a patient, and to an in vitro method for distinguishing thyroid cancer from benign lesions or normal thyroid tissue, on the basis of the evaluation of the expression of the series of genes CLDN10, HMGA2 e LAMB3, using a quantitative molecular detection technique. The invention further relates to other aspects associated with these methods, namely, the use of a series of genes, a method for obtaining data for determining the treatment of thyroid cancer, a laboratory kit and a device for classifying a biological sample from the thyroid gland as malignant or benign.

Description

 IN VITRO THYROID CANCER DETECTION METHOD IN A

PATIENT, IN VITRO DIFFERENTIATION METHOD

THYROID AND NORMAL THYROID OR INJURY TISSUE

BENEFITS OF THYROID, USE OF A SET OF GENES, METHOD

DATA SHEET FOR DIRECTION OF TREATMENT

THYROID CANCER, LABORATORY KIT AND DEVICE FOR

RATE A BIOLOGICAL SAMPLE OF THYROID GLAND AS MALIGNA OR BENIGNA

The present invention relates generally to an in vitro method of detecting thyroid cancer in a patient as well as to an in vitro method of differentiating thyroid cancer from benign or normal tissue lesions. thyroid, as well as some other aspects related to such methods.

 In particular, the invention relates to in vitro methods which are based on the evaluation of expression of the CLDN 10, HMGA2 and LAMB3 gene set contained in a patient's tissue samples, with quantitative molecular detection technique, particularly from RT-qPCR (quantitative reverse polymerase chain reaction transcription) analysis data.

 Joint expression of the 3 specific transcripts has been shown to detect in vitro thyroid cancer and to discriminate malignant tumors, particularly papillary thyroid carcinomas, benign or normal thyroid tissue lesions, and tumors with larger or less risk of lymph node metastases, with high sensitivity and high specificity.

It is understood that, according to the context, the mention of "genes" can denote "transcripts", as the technician in the subject well knows.

BACKGROUND OF THE INVENTION

Conventionally the discrimination between malignant and benign thyroid nodules is made by scintigraphy, ultrasound and biopsy. fine-needle aspiration followed by istological analysis. Despite many advances in the diagnosis and therapy of thyroid nodules and thyroid cancer, such methods are known to lack specificity, particularly to discriminate between follicular thyroid adenoma (FTA) and follicular thyroid carcinoma (FTC). ). As a result, a large number of patients do not have malignant disease but are unnecessarily treated.

 [0 0 6] In relation to cancers other than thyroid cancer, the literature shows that the gene expression profile may allow the discrimination of different tumor entities. However, there are few or no transcripts in common among several studies that attempt to classify the different entities of thyroid carcinoma based on the gene expression profile. Applying a classifier from one study to data from another generally does not yield effective classification results.

[0 07] Well-differentiated thyroid carcinomas are generally associated with a slow course of disease and in 80-85% of cases are classified as papillary thyroid cancer (PTC). In addition to exposure to radiation, low or high iodine consumption, and family history of thyroid cancer, the risk factors contributing to the development of such tumors are poorly known.

[0 0 8] Increased expression of various growth factors and their receptors, changes in cell cycle regulators, and adhesion molecules are known events in the thyroid tumor process and its progression. However, to date, few studies have investigated in depth the biological functions related to the differentially expressed genes in PTC, which may influence tumorigenesis.

[0 0 9] Currently there has been a large increase in the incidence of PTC compared to the overall incidence of cancers. There are suggestions that this is a consequence of improved tumor detection. in particular by the extensive use of ultrasound-guided fine needle aspiration biopsy (or "FNAB").

 Even so, current strategies for diagnosing thyroid nodules often lead to unnecessary surgeries, because of the high percentages of FNAB found in the literature, between 10% and 26% reported as undetermined in nature, only about one year. third is later found to be malignant in nature, most of the follicular variant of PTC.

 It thus follows that there are patients who have unnecessarily undergone thyroid excision surgery, and for this reason are treated with thyroid hormone replacement for the rest of their lives. These surgeries represent increasing expenses for healthcare systems. Hormone replacement therapy changes the age of life and creates expenses for the patient. Both situations could otherwise be avoided with more practical and accurate diagnoses that have been sought.

 In this sense, it has been proposed to detect via FNAB the known high occurrences of genetic mutations of well-differentiated thyroid carcinomas, mainly BRAFV600E. However, there is low sensitivity of this test. Thyroid cancer diagnostic profiles were also investigated using m RNA, mi RNA and methylation assays in tumor samples obtained from surgery and FNAB - despite the high accuracy shown, most failed to demonstrate optimized sensitivity and specificity in the analyzes. validation by using independent sample groups.

Finally, there are prior art disclosures of molecular profile analysis with a large number of potential biomarkers. Such analyzes, however, are costly, so they are less accessible to the general population. In clinical practice small groups of markers would be more desirable and could be evaluated by RT-qPCR or immunohistochemistry.

 In view of these various circumstances, there is a critical need to identify molecular markers that can be applied accurately, more accurately, and more accurately to diagnose thyroid lesions.

 As known to the person skilled in the art, the following definitions apply to the present invention:

 Gene expression: The process by which information contained in a gene or DNA sequence is processed into a functional gene product, such as RNA, that can be translated into proteins.

 - reference gene: gene whose expression is maintained at a constant level in both normal and tumor cells and under different experimental conditions;

 - transcripts: messenger RNA molecules formed from the DNA template strand of a given gene;

 - Cq {cycle quantification) = representative number of the PCR cycle that the amplification exceeds an established fluorescence threshold, allowing a quantification of the initial target DNA / cDNA copy number

- ACq: difference between the Cq {cycle quantification) obtained for the reference sample and the sample under analysis;

 primers: nucleic acid segments, typically 15 to 30 nucleotides, with sequence complementary to the target DNA or RNA to be amplified by PCR (polymerase chain reaction).

 "quantitative molecular detection technique" means the use of RT-PCR, RT-qPCR, immunohistochemistry or any equivalent purpose-directed technique as known to one skilled in the art.

DESCRIPTION OF THE INVENTION The present invention provides for the reduction of the number of unnecessary surgeries that occur as a result of inaccurate diagnoses, thereby reducing healthcare costs and morbidity for patients.

 Unlike the prior art, the invention utilizes a set of only 3 molecular markers, sufficient not only to adequately detect but also to distinguish between thyroid cancer and normal or benign thyroid tissue, as well as carcinomas. with a higher risk of lymph node metastases.

Thus, within a first aspect, the invention relates to in vitro methods, namely detecting thyroid cancer in a patient and differentiating between thyroid cancer and benign lesions or normal thyroid tissue. These methods comprise subjecting the patient's suspicious tissue sample to the quantitative molecular detection technique employing, through gene expression analysis, the LAMB3, CLDN 10 and HMGA2 genes, as compared to at least one reference gene.

Without excluding any others, suitable reference genes are those chosen from, for example, EIF2B1, PUM 1, GAPDH, HPRT1, GUSB, ACTB, B2M, HM BS, IP08, PGK1, RPLPO, TBP, TFRC. , UBC, YWHAZ, PPIA, POLR2A, CASC3, CDKN IA, CDKN1B, GADD45A, PSMC4, PES1, ABL1, ELF1, ATP6, M RPL19, POP4, RPL37A, RPL30, RPS17. In particular, the EIF2B1 and PUM 1 genes are used for reference.

The aforementioned methods of the invention are particularly directed to follicular epithelium-derived thyroid carcinomas, in particular papillary thyroid carcinomas (PTC).

The patient tissue sample to be evaluated may be fresh, fixed in paraffin blocks or frozen. In particular the tissue sample is obtained via fine needle aspiration biopsy (FNAB). The basis of this invention was the profiling of gene expression using microarrays in 61 tissues with PTC and 13 adjacent normal (NT) tissues. A reliable list of transcripts was used for inter-study validation (138 PTC / NT cases in external public databases). The results were then collectively interpreted by in silico analysis. A panel of 28 transcripts were selected and evaluated using the RT-qPCR (quantitative reverse transcription polymerase chain reaction) technique, including benign thyroid lesions (BTL) and other follicular cell-derived thyroid carcinomas (OFDTC). The diagnostic algorithm of the invention was developed (training set: 23 NT, 8 BTL and 86 PTC), validated (independent set: 10 NT, 140 BTL, 120 PTC and 12 OFDTC) and associated with clinical features.

 The quantitative molecular detection technique employed used, in particular, RT-qPCR analysis to evaluate the target markers detected in the analysis of large scale gene expression data.

Although in the technical literature the transcripts CLDN10, HMGA2 and LAMB3 have already been associated with PTC, they are always mentioned within extensive gene lists as showing increased or decreased PTC expression. It is not known, by this technique, the indication or suggestion limited only to the 3 genes in a specific way, or that their association allowed adequate detection of PTC or discrimination between PCT and benign lesions or normal thyroid tissue.

Additionally, the use of such markers is also useful to better stratify the risk of lymph node metastases in patients diagnosed with malignant tumors. The results obtained by RT-qPCR (or equivalent quantitative molecular detection technique) can be associated with a risk scale which is calculated, for example, according to the PFAFFL (Pfaffl MW. A new math model at relative model) method. quantification in real-time RT-PCR Nucleic Acids Res. 2001 May 1.29 (9): e45). This method is based on dividing the ACq of the transcript of interest by the geometric mean ACq of the selected reference transcripts.

 [0026] This methodology does not exclude other known methods for the same purpose, for example that proposed by LIVAK & SCHMITTGEN (Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2 (- Delta Delta C (T)) Methods 2001 Dec 25 (4): 402-8).

 At the end of data normalization (Pfaffl or Livak & Schmittgen method) the following algorithm applies to determine the risk of malignancy in the studied sample, according to disease status:

Score = CDLNIO x 0.196 + HMGA2 x 0.343 + LAMB3 x 0.335 - 4.217

 The numerical value (score) obtained with this algorithm relates to the probability of malignancy as indicated below, and illustrated in fig. 1;

 - Between -50 and -1.6: benign range. The lower the value, the lower the likelihood of malignancy - within this range are nonmalignant lesions and normal tissue indistinctly;

 - Between -1.6 and 50: malignant interval. The higher the value, the greater the likelihood of malignancy.

 Within the malignant range, there is modulation of the likelihood of cervical lymph node involvement, namely:

 - low probability: score between -1.6 to 1.1;

 - intermediate probability: score between 1.1 and 2.2;

 - high probability (greater than 50%): score between 2.2 and 50.

 Without limiting the invention to any specific aspect, the invention has taken into account certain findings.

CLDN10, which encodes a membrane integral protein, shows increased expression in PTC as compared to tissue normal, but low expression in benign thyroid lesion (BTL) tissue, or in other follicular cell-derived thyroid carcinomas (OFDTC) such as those poorly differentiated, anaplastic, follicular, and Hurthle cell carcinoma. Thus, it was an important classifying member in the discrimination between BTL and PTC, despite the lower probability of FTC identification. However, increased expression of HMGA2 and LAMB3 and decreased expression of CLDN 10 may be indicative of FTC.

The HMGA2 product, which showed increased PTC expression, alters the regulatory structure of chromatin by affecting the transcription of various genes. Despite high sensitivity and specificity, as the only marker evaluated by RT-qPCR, HMGA2 expression was not sufficient to discriminate between benign and malignant lesions, even with very high PTC expression.

 The LAMB3 gene product belongs to a family of basement membrane proteins (laminins), with an important association between its increased expression and lymph node metastasis. This fact is confirmed in the higher risk of lymph node metastasis in patients with PTC who had higher scores.

DESCRIPTION OF THE INVENTION

 Necessary and sufficient diagnostic markers of the invention are the CLDN 10, HMGA2 and LAMB3 transcripts used in the thyroid cancer detection method and in vitro method of differentiating thyroid cancer from normal or thyroid tissue. benign lesions. For this purpose the quantitative molecular detection technique, particularly RT-qPCR, is used, where tissue samples suspected of the patient's tumor are used.

[0035] The method of the invention is performed in a particular embodiment by combining the results obtained in molecular detection. with a risk scale, for example with the methods of Pfaffl (2001) or Livak & Schmittgen (2001) mentioned above.

The patient sample to be evaluated according to the methods of the invention may be freshly woven, paraffin blocks or frozen. In particular, the tissue sample is obtained via fine needle aspiration biopsy (FNAB).

 Relative expression of markers HMGA2, LAMB3 and CLDN10 was analyzed by RT-qPCR using comparison with expression of reference genes EIF2B1 (synonyms: EIF2B, EIF2BA) and PUM1 (synonyms: HSPUM, PUMH, PUMH1, PUML1) .

 The performance of this classifier (composed of the three marker transcripts) was tested in a validation trial, showing 98.1% accuracy against papillary thyroid carcinoma and 80.5% accuracy for other types of thyroid derived carcinomas. follicular epithelium (OFDTC).

 In another aspect, the invention is used in thyroid nodule biopsies to guide surgery and clinical follow-up, while alerting which malignant cases are most at risk for lymph node involvement, particularly when there is increased expression of LAMB3.

 Within another aspect, the invention also relates to the use of a set of transcripts characterized in that it is in an in vitro diagnostic method of thyroid cancer, said set comprising the transcripts LAMB3, CLDN10 and HMGA2 compared to at least one reference transcript.

Within another aspect, the invention further relates to the use of a set of transcripts characterized in that it is in a method of analysis to differentiate between thyroid cancer and normal thyroid tissue. or of benign lesions, said set comprising the transcripts LAMB3, CLDNIO and HMGA2 compared to at least one reference transcript. In another aspect, the invention relates to a method of obtaining data for directing the treatment of Thyroid cancer characterized by the fact that it comprises subjecting a patient's suspicious tissue sample to the quantitative molecular detection technique that uses, for gene expression analysis, the set of LAMB3, CLDNIO and HMGA2 transcripts, compared to the expression of at least a reference gene, optionally associated with a risk scale.

In another aspect, the invention relates to a laboratory kit comprising reagents for detecting the expression of the LAMB3, CLDNIO, and HMGA2 genes related to the diagnosis of PTC, as well as for evaluating or assisting in the evaluation. of its expression with respect to at least one reference gene, optionally containing an instruction manual for use.

 Within a particular embodiment the laboratory kit of the invention comprises the components:

a- suitable component for RNA extraction from a patient's tissue sample;

b- component suitable for the conversion of RNA into complementary DNA;

triphosphate c-deoxyribonucleotides (dNTPs)

d- primers for target transcripts CLDNIO, HMGA2,

 LAMB3 and for one or more reference transcripts;

e-buffer for reverse transcriptase reaction;

f - 5 mM diethyltreitol (DTT) solution;

g- reverse transcriptase enzyme;

h- solution containing PCR buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 0.25% DMSO, 1 unit Taq DNA polymerase, and 0.01X reporter molecule Sybr Green I; i- means to calculate the scores according to a risk table that indicates the results by applying the diagnostic algorithm:

Score = CDLN 10 x 0.196 + H MGA2 x 0.343 + LAM B3 x 0.335 - 4.217.

In yet another aspect, the invention relates to a device for classifying a biological sample of the thyroid gland as malignant or benign, comprising:

 means for measuring the level of expression of the LAM B3, CLDN 10 and H MGA2 genes;

- means for correlating expression level with a thyroid disease status classification;

 - means to express such status.

 [004 6] Said means are known to the person skilled in the art:

means for measuring expression level are, for example, RT-PCR, RT-qPCR or any other equivalent;

 Means for correlating expression level with thyroid disease status are, for example, the methodologies of Pfaffl (2001) or Livak & Schmittgen (2001), already mentioned above;

 - "Means for expressing" disease status are any, for example a display where digits, codes, words, colors or images appear without excluding any other suitable means or devices.

 For the quantitative molecular detection of the transcript set of the invention, including one or more reference transcript, commercial primers or specific primers may be used as described below.

EXAMPLES

For a better understanding of the invention and to clarify its applicability, examples are given below which show particular embodiments of the invention, which in no way impose limits on the scope of the invention. Example 1 - Methods of the invention using commercial primers in the quantitative molecular detection technique.

Following RNA extraction, performed by RNeasy mini kit (from the US company Qiagen), the RNA is converted to complementary DNA or cDNA (day 1). The protocol for converting RNA to cDNA began with the addition of oligo dT ^{12 18} primers (from US Healthcare Life Sciences) and / or random primers (from Invitrogen), triphosphate deoxyribonucleotides (dNTPs) ) (from US company Invitrogen), Super Script III buffer First Strand Buffer (from US company Invitrogen), 5 mM diethyltreitol (DTT) (from US company Invitrogen) and 200 units of the enzyme Super Script III Reverse Transcriptase (from the US company Invitrogen), following a series of incubations at varying temperatures. It is important to note that any alternative methods of converting RNA to cDNA can be used. PCR amplification was carried out by using 20 ng of cDNA, IX TaqMan Gene Expression assays (probes and primers, the North American company Applied Biosystems), IX TaqMan Universal Master Mix ^® II (North American company Applied Biosystems) supplementing with ultrapure water (Ambion®, from the US company Life Technologies) to 15 final volume (day 2). Three commercial primers from the US company Applied Biosystems were used for amplification of the target transcripts CLDN10 (TaqMan ^® Hs00734479_m1 assay), HMGA2 (TaqMan ^® Hs00971724_m1 assay) and LAMB3 (TaqMan ^® Hs00165078_m l assay) and for two reference transcripts, EIF2B1 (TaqMan ^® Assay Hs00426752_m l) and PUM1 (TaqMan ^® Assay Hs00472881_m l). The reactions were mounted in duplicates in the presence of negative controls (No Template Control) in a PCR system. To calculate the quantification of the transcripts of interest, the proposed method was used. by PFAFFL (2001), which is based on the division of ACq of the transcript of interest by the geometric mean ACq of the selected reference transcripts. Example 2 - Methods of the invention using specific primers developed for use in the quantitative molecular detection technique.

After RNA extraction and subsequent conversion to cDNA

(day 1), PCR is performed using the following sequences for the target transcript primers (day 2):

 - CLDN10 - Seq. ID 1 and Seq. ID 2

 - HMGA2 - Seq. ID 3 and Seq. ID 4

 - LAMB3 - Seq. ID 5 and Seq. ID 6

 - EIF2B1 - Seq. ID 7 and Seq. ID 8

 - PU M1 - Seq. ID 9 and Seq. ID 10

 PCR was performed containing 20 ng cDNA, 1x Sybr

Green I Master Mix (from US Applied Biosystems), 200 μΜ from each primer for a volume of 12.5 μΙ_. Reactions were mounted in duplicates in the presence of negative controls ("/ Vo Template Control") in a PCR system at 40 amplification cycles. To calculate the quantification of the transcripts of interest, the method proposed by PFAFFL (2001) is used, which is based on the division of ACq of the transcript of interest by the geometric mean of the ACq of the selected reference transcripts. At the end of normalization, the algorithm of the invention was applied

 Score = CDLN10 x 0.196 + HMGA2 x 0.343 + LAMB3 x 0.335 - 4.217

Example 3 - Methods of the invention applied to real case.

MCS patient had a suspected thyroid nodule and, according to clinical, laboratory and imaging tests, fine-needle aspiration biopsy was indicated;

The remaining biopsy material was subjected to RNA extraction followed by conversion to cDNA (estimated time 5 hours); PCR was performed with the primers according to example 1 (40 amplification cycles: 1 minute at 95 ° C, 30 seconds at 60 ° C) (estimated time 2 hours);

 Applied the diagnostic algorithm obtained the score of 3.72, indicating high risk of malignancy;

 The case underwent total thyroidectomy, which was confirmed by histopathological analysis if it was a classic variant papillary thyroid carcinoma;

 As the value of the score was higher than 2.2, the case was also classified as high risk of lymph node involvement, and neck dissection was also indicated, where histopathological analysis confirmed the presence of lymph node injury.

Example 4 - Method of the invention applied to real case.

Patient TBB had a suspected thyroid nodule and, according to clinical, laboratory and imaging tests, fine-needle aspiration biopsy was indicated;

 The remaining biopsy material was subjected to RNA extraction followed by conversion to cDNA (estimated time 5 hours).

PCR was performed as per example 2 (40 amplification cycles: 1 minute at 95 ° C, 30 seconds at 60 ° C) (estimated time 2 hours);

 When applied the diagnostic algorithm obtained a score of -1.5, presenting high risk of malignancy.

 The case underwent total thyroidectomy and was diagnosed as minimally invasive follicular carcinoma.

 As the value was lower than 2.2, the case was also classified as low risk of lymph node involvement, with no neck dissection being indicated.

Example 5 - Method of the invention applied to real case. Patient WS had a suspicious thyroid nodule and fine-needle aspiration biopsy was indicated on imaging studies;

The remaining biopsy material was subjected to RNA extraction followed by conversion to cDNA as per example 2 (estimated time 5 hours);

 PCR was performed according to example 2 (40 amplification cycles: 1 minute at 95 ° C, 30 seconds at 60 ° C) (estimated time 2 hours);

 Applying the diagnostic algorithm obtained a score of -7.1, indicating low risk of malignancy;

 Although only follow-up was recommended, the patient opted for total thyroidectomy, being diagnosed by histological analysis of the surgical specimen only follicular adenoma.

 It is known that from the information provided, with the aid of the examples described above, one skilled in the art can make use of the invention using different forms as claimed herein, but performing functions and obtaining the same or similar results as those presented herein. which indicates that it does not depart from the protected invention.

Claims

CLAIMS TIPS

IN VITRO THYROID CANCER DETECTION METHOD IN A PATIENT, which comprises subjecting a patient's suspect tissue sample to the quantitative molecular detection technique, characterized by the use of the LAMB3, CLDN transcript set for gene expression analysis. 10 and HMGA2, compared to the expression of at least one reference gene.

 IN VITRO METHOD OF DIFFERENTIATION BETWEEN NORMAL THYROID TISSUE THYROID CANCER OR BENIGN INJURIES comprising subjecting a sample of suspect patient tissue to the quantitative molecular detection technique characterized by the use of LAMB3 gene set for gene expression analysis, CLDN 10 and HMGA2, compared to the expression of at least one reference gene.

Method according to either claim 1 or claim 2, characterized in that the reference gene is chosen from EIF2B1, PUM 1, GAPDH, HPRT1, GUSB, ACTB, B2M, HMBS, IP08, PGK1, RPLPO, TBP, TFRC, UBC, YWHAZ, PPIA, POLR2A, CASC3, CDKN 1A, CDKN 1B, GADD45A, PSMC4, PES1, ABL1, ELF1, ATP6, MRPL19, POP4, RPL37A, RPL30, RPS17. METHOD according to claim 3, characterized in that the reference genes EIF2B1 and PUM 1 are used.

 Method according to either claim 1 or claim 2, characterized in that thyroid cancer is one of the thyroid carcinomas derived from the follicular epithelium, particularly papillary carcinoma.

Method according to either claim 1 or claim 2, characterized in that said patient sample is fresh tissue, paraffin blocks or frozen. Method according to either claim 1 or claim 2, characterized in that the tissue sample is obtained via a fine needle aspiration biopsy.

 Method according to either of Claims 1 and 2, characterized in that the quantitative molecular detection technique is RT-qPCR.

 Method according to either of Claims 1 and 2, characterized in that the amplification reaction primers for quantitative molecular detection are the sequences Seq. ID 1 and Seq. ID 2 for CLDN 10, Seq. ID 3 and Seq. ID 4 for HMGA2, Seq. ID 5 and Seq. ID 6 for LAMB3, Seq. ID 7 and Seq. ID 8 for the reference genes EIF2B1, and Seq. ID 9 and Seq. ID 10 for reference gene PUM1.

 Method according to either claim 1 or claim 2, which comprises associating the results obtained by the quantitative molecular detection method with a risk scale.

 Method according to claim 10, characterized in that the risk scale is calculated according to the method of PFAFFL 2001 or LIVAK & SCHMITTGEN 2001.

 Method according to claim 10, characterized in that the risk scale is calculated according to the PFAFFL 2001 method by dividing the ACq of the transcripts of interest LAMB3, CLDN 10 and HMGA2 by the geometric mean of the ACq of the reference transcripts EIF2B1 and PUM 1 and applying the diagnostic algorithm to normalized relative expression data.

 Score = CDLN10 x 0.196 + HMGA2 x 0.343 + LAMB3 x 0.335 - 4.217 with a score value between -50 and -1.6 relating to benign lesions, and the range between -1.6 and 50 relating to lesions. malignant.

13) Method according to claim 12, characterized in that the scores obtained by the diagnostic algorithm are related to probabilities. of involvement of cervical lymph nodes, that is, low probability with a score between -1.6 to 1.1; intermediate probability with a score between 1.1 and 2.2; high probability, greater than 50%, with a score between 2.2 and 50.

 14) USE OF A TRANSCRIPT SET characterized in that it is in an in vitro diagnostic method of thyroid cancer, said set comprising the transcripts LAMB3, CLDN 10 and HMGA2 in comparison with the expression of at least one reference gene.

 15) USE OF A TRANSCRIPT SET characterized in that it is in an in vitro method of differentiating thyroid cancer from normal thyroid tissue or benign lesions, said set comprising the transcripts LAMB3, CLDN 10 and HMGA2 in comparison with the expression of at least one reference gene.

 Use of a TRANSCRIPT SET according to any one of claims 14 or 15 wherein the reference genes are preferably EIF2B1 and PUM 1.

 17) METHOD FOR OBTAINING DATA FOR RECYCLING THE THYROID CANCER TREATMENT characterized by submitting a sample of the patient's suspicious tissue to the quantitative molecular detection technique using, for gene expression analysis, the LAMB3 gene set, CLDN 10 and HMGA2, optionally associated with a risk scale.

 Method according to Claim 17, characterized in that the risk scale is calculated according to the method of PFAFFL 2001 or LIVAK & SCHMITTGEN 2001.

19) DATA-METHOD METHOD FOR DI RECION OF THYROID CANCER TREATMENT according to claim 17 characterized in that the risk scale is calculated according to the PFAFFL 2001 method by dividing the ACq of the transcripts. of interest LAMB3, CLDN 10 and HMGA2 by the geometric mean of the ACq of reference transcripts EIF2B1 and PUM 1 and applying to normalized relative expression data and the diagnostic algorithm

 Score = CDLN10 x 0.196 + HMGA2 x 0.343 + LAMB3 x 0.335 - 4.217 and a score between -50 and -1.6 is related to benign lesions and a score between -1.6 and 50 is related to malignant lesions.

 20) METHOD OF OBTAINING DATA FOR RECONDITION OF THYROID CANCER TREATMENT ACCORDING TO Claim 19, characterized by the fact that the score obtained by the algorithm that exceeds 2.2 on the risk scale predicts, besides high risk malignancy, a greater than 50% probability of cervical lymph node

21) LABORATORY KIT characterized by the fact that it comprises reagents to detect the expression of the LAMB3, CLDN 10 and HMGA2 genes related to the diagnosis of PTC, as well as to evaluate or assist the evaluation of its expression, optionally containing an instruction manual for use.

LABORATORY KIT according to claim 19 comprising the components:

 a- suitable component for RNA extraction from a patient's tissue sample;

 b- component suitable for the conversion of RNA into complementary DNA;

 triphosphate c-deoxyribonucleotides;

 d- primers for the target transcripts CLDN10, HMGA2,

 LAMB3 and for one or more reference transcripts;

 e-buffer for reverse transcriptase reaction;

 f- 5 mM diethyltreitol solution;

 g- reverse transcriptase enzyme;

h- solution containing PCR buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 0.25% DMSO, 1 unit Taq DNA polymerase, and 0.01X reporter molecule Sybr Green I i- means for calculating scores according to a risk table indicating the results by applying the diagnostic algorithm

 Score = CDLN10 x 0.196 + HMGA2 x 0.343 + LAMB3 x 0.335 - 4.217.

Laboratory kit according to claim 20, characterized in that the primers for the CLDN 10 target transcript have the sequences Seq. ID 1 and Seq. ID 2; primers for the HMGA2 target transcript have the sequences Seq. ID 3 and Seq. ID 4; primers for the target transcript LAMB3 have the sequences Seq. ID 5 and Seq. ID 6; primers for the reference gene EIF2B1 have the sequences Seq. ID 7 and Seq. ID 8; primers for the reference gene PUM 1 have the sequences Seq. ID 9 and Seq. ID 10.

 24) DEVICE FOR CLASSIFYING A BIOLOGICAL SAMPLE OF THE THYROID GLAND AS MALIGNA OR BENIGNA in relation to PTC, characterized in that it comprises:

 means for measuring the level of expression of the LAMB3, CLDN 10 and HMGA2 genes compared to the expression of at least one reference gene;

- means for correlating expression level with a thyroid disease status classification;

 - means to express such status.