Comparative Cyto-histological Genetic Profile in a Series of Differentiated Thyroid Carcinomas

1. Introduction

Thyroid nodular disease is very common and distinguishing benign from malignant nodules is still a major challenge in clinical practice. Thyroid nodules are diagnosed by palpation in 5% of women and 1% of men in iodine-sufficient parts of the world [1]. By thyroid ultrasound, nodules can be identified in 19% to 68% of the individuals [2]. Thyroid cancers are diagnosed in 1% to 5% of nodules [3], with more than 90% being differentiated thyroid carcinoma (DTC) [4].

The incidence of DTC, from which papillary thyroid carcinoma (PTC) is the most prevalent (over 95%) [5], has increased in the last years, mainly attributed to enhanced sensitivity of complementary diagnostic techniques and environmental factors, but not accompanied by a corresponding change in mortality rate [6,7,8]. Despite its favorable prognosis, it was advanced that clinicopathological features and the presence of some genetic alterations can influence their progression [9]. Pre-surgical accurate diagnosis is crucial to decide the optimal medical or surgical treatment.

Ultrasound-guided fine needle aspiration cytology (US-FNAC) is central in the diagnosis of nodular thyroid disease, but up to 30% of nodules remain without a definitive diagnosis [10]. Bethesda II and VI cytological results have excellent performance to assess the benign or malignant nature of a thyroid nodule. Diagnostic limitations are particularly significant in the non diagnosis (ND), and in indeterminate Bethesda categories III- atypia of undetermined significance (AUS), and IV: follicular neoplasm (FN) [11]. Each category harbors a different risk of malignancy and a different option for treatment and follow-up [12,13,14].

Numerous studies have been conducted to identify genetic or molecular biomarkers that facilitate the pre-surgical diagnosis (in particular in categories ND, III and IV) and prognosis of thyroid cancer, but current results remain inconclusive and without definitive evidence [15,16].

Thyroid surgery remains the only diagnostic (and usually curative) approach for suspicious and malignant nodules. However, the inability to differentiate, preoperatively, between benign and malignant nodules among those classified as Bethesda III or IV hampers the ability to prevent unnecessary surgeries and the reduction of its possible complications [17].

In 2017, the European Thyroid Association (ETA) presented guidelines for molecular diagnoses in US-FNAC of thyroid nodules after reviewing methodological aspects, and limitations of molecular diagnoses in thyroid cytology. Molecular tests have the potential in clinical practice for diagnosis (pre-surgical markers) and follow-up of thyroid nodules (post-surgical markers), if they are performed in specialized laboratories and with adequate calibration and analytical validation before being implemented in clinical practice [18].

Although American and European Thyroid Associations [19,20] consider the inclusion of molecular tests in US-FNAC, no consensus is yet achieved on the best molecular panel to use in the diagnosis of thyroid nodules. Available molecular tests have been developed in United States of America, and it is important to assess whether such tests can be used in the European population [21,22]. The most significant disadvantage of commercially available tests is related to their cost, not supported by National Health Systems in Europe where, in many countries, its price is equal to or greater than a thyroidectomy; on the other hand, to perform these tests, highly specialized reference laboratories are required [23,24].

Studies have emerged on the use of artificial intelligence (AI) in the diagnosis and classification of thyroid nodules, constituting a new area of interest in scientific research, but still with no practical applicability [25,26].

The high prevalence of thyroid nodules [27,28], coupled with the rising incidence of thyroid neoplasm, urges the exploration of strategies to prevent unnecessary surgeries while effectively treating patients. This study aims to compare the cyto-histological genetic profile in matching cytology and histology samples, and to evaluate the reliability of molecular tests in preoperative diagnosis in order to improve the diagnostic accuracy of US-FNAC in indeterminate nodules.

2. Materials and Methods

2.1. Study design

This retrospective study was conducted in a series of patients with thyroid nodular disease that underwent surgery based in clinic, imageology and cytological criteria for suspicion of malignancy, in a single non oncologic hospital between 2013 and 2020.

Inclusion criteria: patients diagnosed, after surgery, with differentiated thyroid carcinoma (209 cases) or benign thyroid nodules (50 controls), of both gender, and over 18 years old with available representative histological and cytological sample.

Exclusion criteria: patients with toxic goiter, papillary carcinoma smaller than 1 cm unless metastatic, and patients with other than thyroid neoplasm at the time of thyroid surgery.

A total of 518 samples (259 histology samples and corresponding 259 US-FNAC), representing 259 patients, along with clinical information, were included in the study. Epidemiological and clinicopathological features of patients and tumors were gathered based on the information available in the clinical and histopathological reports from the reference hospital.

All the samples were reviewed by two independent pathologists based on the 4th edition of the World Health Organization (WHO) classification of tumors of endocrine organs [29]. Two subgroups were separately analyzed, the indeterminate nodules corresponding to Bethesda categories III and IV, and the cases diagnosed as PTCs, to obtain homogeneous group.

2.2. Methodology

2.2.1. DNA extraction

To extract genomic DNA, 10 µm sections of formalin-fixed-embedded (FFPE) tissues were manually micro dissected using H&E guidance, and the GRS Genomic DNA BroadRange Kit (GRiSP Research Solutions, Portugal) was used. The smear of US-FNAC slides were scraped to a tube and the QIAmp® DNA Investigator Kit (Qiagen, USA) was used for DNA extraction of the US-FNACs according to the manufacturer’s instructions. NanoDropTM One UV185 Vis Spectrophotometer (Thermo Fisher Scientific Inc) was used for quantification of isolated DNA.

2.2.2. Mutational screening

PCR and Sanger sequencing were used for genetic characterization of tumors regarding TERTp, BRAF and RAS (NRAS, HRAS and KRAS) mutations, with primers design accounting for the most frequent regions mutated on thyroid carcinoma, namely TERTp -124 and -146 promoter regions, BRAF exon 15 (codon 600), NRAS exon 2 (codon 61), HRAS and KRAS exons 1 and 2) (codons 12, 13 and 61) [30,31].

Multiplex PCR kit (QIAGEN, USA) and Bioline PCR kit (MyTaq HS Mix 2X, USA) were used for DNA amplification following manufacturer’s instructions. Sanger sequencing using the ABI Prism Big Dye Terminator kit v3.1 Cycle Sequencing and capillary electrophoresis using the Applied Biosystems 3130/3130xl Genetic Analyzers was performed. All detected mutations were validated by performing a new independent analysis.

2.2.3. Clinicopathological features

All primary lesions were evaluated, both concerning patient’s characteristics (age and gender), as well as tumor’s clinicopathological features: tumor size, extra thyroidal invasion, capsule invasion, vascular and lymphatic invasion, presence of fibrosis, inflammatory infiltrate, tall cell, oncocytic component, psammoma bodies, calcification, necrosis, and presence of lymph nodes metastases, focality and laterality. Due to the reduced number of some histological subtypes of DTC, only PTCs were considered in the statistical analysis of the clinicopathological features of the patients.

2.2.4. Statistical analysis

The analysis was performed with categorical variables presented with frequencies (percentages), and quantitative variables with mean, and standard deviation (SD) and range (minimum-maximum). Regarding quantitative variables, Student’s t and non-parametric Mann Whitney tests were used to compare groups, as appropriate. In the case of categorical variables, the Chi-square or Fisher’s exact tests were used, as required.

To assess the strength of cyto-histological genetic profile agreement, Cohen’s Kappa was estimated and interpreted according to Altman (1999) [32]. The discriminative ability of genetic biomarkers regarding malignancy was performed by estimating sensitivity, specificity, positive and negative predictive values, with corresponding 95% confidence intervals. A level of significance α=0.05 was considered. Data analysis was performed using the SPSS software version 27.0 (IBM Corp. Released 2020. IBM SPSS Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp).

3. Results

3.1. Series description

3.1.1. Epidemiologic data

From the 259 patients, 209 (80.7%) were female, with mean age of 53 (15.8) years (range 18-84), and 50 (19.3%) were male, with mean age of 54 (15.9) years (range 23-81). The distribution of age at diagnosis was the similar across categories of gender (p =0.890), and across categories of histological diagnosis (p=0.186).

Mean tumor size was 33.5 mm (10.79) for benign lesions and 27.6 mm (15.57) for malignant lesions; the distribution of tumor size was not the same across categories of histologic diagnosis, being malignant tumors smaller than benign lesions (p<0.001).

The 259 histology samples were composed by 50 (19.3%) benign lesions (control), and 209 (80.7%) malignant lesions (cases). PTC represented the larger group in malignant histology, 182 (87.1%) cases.

The matched 259 cytology samples were distributed according to the Bethesda classification, as follows: I. non diagnostic (ND), 15 samples (5.8%); II. benign (B), 47 samples (18.1%); III. atypia of undetermined significance (AUS) 43 samples (16.6%); IV. Follicular Neoplasm (FN) 58 samples (22.4%); V. suspicious for malignancy (SM) 42 (16.2%); VI. malignant 54 samples (20.9%).

The distribution of the cytology samples within each histological subtype is presented in Table 1.

The 39% of nodules whose cytological result was indeterminate corresponded on histology to 23 benign (22.8%) and to 78 malignant lesions (77.2%).

3.1.2. Cyto-histological genetic profile

The molecular status of TERTp, BRAF, and RAS (NRAS, HRAS and KRAS) genes of the cytologies and histologies in all series is summarized in Table 2. Mutations were present in 85 cytologies (32.8%), and in 130 histologies (50.2%).

The mutation frequencies observed in cytology and histology specimens within our series were, respectively, TERTp: 3.7% vs. 7.9%; BRAF: 19.5% vs. 25.1%; NRAS: 4.4% vs. 7.9%; HRAS: 4.8% vs. 7.6%; KRAS: 1.6% vs. 2.8%.

Mutation frequencies in cytology and histology samples by histology subtypes is show in Table 3, being detected in PTCs in 97.6% (83/85) of the cytology’s and in 94.6% (122/129) of the histology specimens.

TERTp and BRAF mutation was present only in PTCs, both in cytology and in histology. RAS mutation was present in cytology and histology, respectively in 25 and 38 PTCs, in one and three benign cases, in two and one WDT-UMD cases, and only in histology samples of one Follicular thyroid carcinomas (FTC), and one Non-invasive follicular thyroid neoplasm with papillary like nuclear features (NIFTP). In Hürthle Cell carcinoma (HCC) no mutations were identified.

Genetic mutations in cytology samples through Bethesda categories were presented in Supplementary Table 1.

BRAF mutations represented the majority of mutations, 43 cases, being five cases in Bethesda III and IV, eight cases in Bethesda V and 30 cases in Bethesda VI. RAS mutations were present in 26 cases, divided into the different Bethesda categories. Concomitant mutation of TERTp/BRAF (6 cases) or TERT/RAS (1 case) were identified always in cases with malignant histology, but with variable cytology: one Bethesda II, two Bethesda III and four Bethesda VI. No mutations were found in Bethesda ND category.

3.2. Genetic alterations in Indeterminate nodules

From the 259 cases 39% presented cytology of indeterminate nodules (101/259), with 43 AUS (42.6%) cases and 58 FN (57.4%). Among the indeterminate nodules, 23 (22.8%) were classified as benign lesions in histology, while 78 (77.2%) were classified as malignant in histology.

From the 78 malignant cases with indeterminate cytology, 66 were PTCs in histology. The remaining 12 indeterminate nodules were distributed among the other types of DTC cases (Table 1).

Twenty five (24.8%) indeterminate nodules presented mutations in cytology samples, one RAS mutation in a benign lesion and 24 in malignant tumors, being 13 (17.3%) for RAS mutation, seven (9.3%) for BRAF and four (5.6%) for TERTp mutation.

In the histology samples of the indeterminate nodules, 48 (47.5%) were mutated, being two benign lesions with RAS mutation and in 46 malignant tumors, of which 11 (14.5%) for TERTp mutation, 13 (16.9%) for BRAF and 22 (29.3%) for RAS mutation. Concomitant TERTp and BRAF mutations were identified in three cases.

Mutation frequencies in cytology and histology samples of indeterminate nodules were, respectively: TERTp: 4.3% vs. 11.1%; BRAF: 7.2% vs. 13%; RAS: 14.4% vs. 24.5% (Table 4).

The molecular status of TERTp, BRAF, and RAS (NRAS, HRAS and KRAS) genes, including mutations type in cytologies and histologies for indeterminate nodules is summarized in Supplementary Table 2.

3.3. Papillary thyroid carcinomas

3.3.1. Clinicopathological features of PTCs

For statistical analysis of the clinicopathological features in our series, only PTCs were considered due to the limited number of the other malignant subtypes of DTC. From the 209 patients with malignant histology of DTCs, 182 (87.1%) were PTCs, and 27 (12.9%) were other subtypes of DTC (Table 3).

Of the 182 PTC tumors, 150 (82.4%) were females; the mean age at diagnosis was 52 (15.9) years (range 18 - 84); mean tumor size was 26.7 mm (15.38) (p< 0.001).

The histological cases of PTCs encompassed different cytological diagnoses. The distribution by Bethesda categories was as follows: I. (ND) 12 (6.6%); II. (B) 16 (8.8%); III. (AUS) 30 (16.5%); IV. (FN) 36 (19.8%); V. (SM) 35 (19.2%); VI. (M) 53 (29.1%) samples.

The histological distribution of PTC variants and the corresponding clinicopathological features is presented in Supplementary Table 3.

3.3.2. Cyto-histological genetic profile in PTCs

In PTCs, mutation frequencies in cytology and histology were, respectively: TERTp: 5.3% (9) vs. 10.8% (19); BRAF: 27.4% (48) vs. 35.6% (63); RAS: 14.3% (25) vs. 22.1% (38).

Concomitant TERTp and BRAF mutations were identified in 10 PTCs, in cytology and/or histology samples.

The mutations in cytology and histology samples through histology subtypes of PTCs are showed in Supplementary Table 4.

3.3.3. Relationship between the clinicopathological features and the genetic profile in PTCs

The associations between the clinicopathological features and the genetic alterations of PTCs are presented in Supplementary Table 5 (A, B and C). No statistical significant diference was achieved for gender or age. The distribution of tumor size was different across types of mutations at diagnosis: tumors with BRAF mutations were smaller (p<0.001), than tumors without mutation.

The presence of TERTp mutations was significantly associated with vascular invasion (p=0.004) and with oncocytic component (p=0.017). BRAF mutations showed a significant association with the presence of extra thyroidal invasion, capsule invasion, vascular invasion, lymphatic invasion, oncocytic component, fibrosis, inflammatory infiltrate, tall cell component, psammoma bodies, calcification, focality, and LNM (p<0.001 for all). Concomitant TERTp and BRAF mutations were significantly associated with extra thyroidal invasion, capsular invasion, vascular and lymphatic invasion, tall cells, and oncocytic component (p<0.001 for all).

Concerning RAS mutations, cases exhibiting NRAS mutations demonstrated a significant association with the presence of capsule (p=0.030) and unifocality (p=0.030). No significant associations were observed between the presence of HRAS or KRAS mutations and any clinicopathological features.

3.4. Statistical analysis of cyto-histological profile

The cyto-histological agreement achieved for molecular alterations in our series is summarized in Table 5. The comparison between cytology and histology for molecular alterations was possible in 244 cases. The cyto-histological concordance was observed in 94.9% of cases, with a Cohen’s k=0.67, which is considered substantial agreement (0.60-0.80). The concordance obtained in PTCs was 94.6% with a substantial cyto-histological agreement (k=0.659).

The agreement of the molecular testing in cytological and histological samples was evaluated in 95 indeterminate nodules, with concordance in 95.6% of cases, and a k=0.643 also considered as substantial agreement.

When analyzing the genes independently, the genes showing cyto-histological molecular agreement considered substantial were BRAF and HRAS in DTC, in PTCs, and in indeterminate nodules. NRAS and TERTp genes presented a cyto-histological molecular agreement considered moderate (0.40-0.59). KRAS mutation was excluded from this analysis due the small number of cases.

The mutations’ discriminative ability for the diagnosis of malignancy in DTCs, PTCs and indeterminate nodules is shown in Table 6 for all series and in Supplementary Table 6 for each group.

The TERTp and BRAF mutations exhibited a specificity of 100% both in cytology and in histology, with a Positive Predictive Value (PPV) of 100% in all the series. RAS genes mutations demonstrated a specificity of 98% in cytology and 94% in histology for DTCs and PTCs, with a PPV in cytology and histology, respectively of 96.3% and 93.3% in DTCs, and 96.2% and 92.7% in PTCs; in indeterminate nodules, RAS presented a specificity of 95.5% in cytology (PPV 92.9%) and 91.3% in histology (PPV 91.7%). However, the sensitivities and the Negative Predictive Value (NPV) of all mutations were very low in our series for DTCs, PTCs and indeterminate nodules.

4. Discussion

Molecular tests have been proposed to be used in clinical practice for diagnosis (pre-surgical markers) and follow-up of thyroid nodules. The 2017 ETA guidelines review methodological aspects and limitations of molecular diagnoses of US-FNAC of thyroid nodules (18). In recent years, panels of somatic genetic alterations have been proposed as a potential approach. However, due to the current limitations in accurately predicting the malignancy of thyroid nodules, most authors consider that further studies with larger sample numbers, rigorous normalization techniques and results validation, are required before their widespread adoption in clinical practice [33,34].

The present study was designed to contribute to clarify some of these challenges. A series of 259 surgical resected thyroid nodules, 209 with a histologic diagnosis of malignancy and 50 with benign diagnosis (as control), was select and the matched pre surgical cytologic diagnostic slides collected. Then we evaluated the genetic alterations in the 518 matched histological/cytological samples. Strength from our study is the relatively large number of matched cytology/histology samples included. This less usual design of the study also implies that our series will have a significant number of malignant tumors in the Bethesda category III and IV (please see bellow limitations of the study), resulting from the series selection criteria. But that design has also advantages by allowing more accurately evaluate at what extent genetic markers can improve cytology pre-surgical results.

BRAF mutations were predominant in our series, either in cytology as well as in histology samples, followed by RAS mutations, reflecting the composition of the series that was enriched in PTC and in low-grade tumors. In accordance with other published studies [36,37], including a study from Whitney S. Goldner (2019) [38], the majority of BRAF mutations were p.V600E. TERTp mutations were assessed in two different hotspots and most mutations in cytologies and histologies were found in the -124 position, as reported previous in thyroid tumors [39]. Tumors in the malignant category (Bethesda VI) presented a high frequency of BRAF mutations, followed by concomitant TERT and BRAF, and TERT and RAS mutations.

Our first approach was to evaluate cyto-histological genetic profile agreement for molecular alterations with a special attention to the nodules that have been diagnosed in indeterminate cytological diagnostic, category that raise important diagnostic doubts.

An acceptable cyto-histological genetic profile agreement for molecular alterations was attained in the overall series, indicating a good level of consistency between the two types of samples in the genetic analysis. BRAF and HRAS were the genes showing a substantial cyto-histological molecular agreement, whereas a moderated agreement was found for NRAS and TERTp genes. In each case the divergent molecular results between cytology and histology can reflect tumor heterogeneity, multifocality and/or low allelic frequencies, being the repercussion of these factors more marked on cytology samples that present a lower percentage of (neoplastic) cells that could increase the number of false negative results.

Then we went to evaluate if the mutational status in cytology, particularly for lesions categorized as indeterminate (AUS and FN), can enhance the diagnostic ability, and improve patient management [35]. In cytology the frequency of the mutations increased from lower Bethesda category (B and AUS) to higher Bethesda category (SM and M). Of note, in the control cases (50 selected histological benign lesions), only one case in cytology, and 3 in histology, present mutations and all in the RAS genes. If we consider TERTp, BRAF and RAS mutations in indeterminate categories (n=101 cases), the cytologic result taken together with clinical, imageology and molecular information, could be improved in 25% of the cases (24 cases with mutations in any gene). Being more cautious, since RAS mutations are relatively frequent, and were found in a few, benign nodules of our series, and considering only the predictive value for malignancy of TERTp and BRAF mutations (both with a 100% specificity, 100% PPV), then 10% of the indeterminate nodules of our series (n=10) could be diagnosed as malignant in a pre-surgical phase, without performing more FNACs and allowing a better surgical option and health costs reduction.

Indeterminate nodules represented a heterogeneous group. The high number of malignant cases in our series result, as referred before, from two reasons: in one hand because the case selection was done based in a histologic diagnosis of malignancy (or benignity in the 50 control cases), and in the other hand these patients underwent surgery due to clinical reasons besides FNACs results ( US features p. ex.). A good cyto-histological genetic profile agreement for molecular alterations was achieved between the two types of samples in indeterminate nodules, reinforcing the role of molecular analyses before surgery in those cases. RAS mutations were the most frequent mutated gene in indeterminate nodules, in accordance with Censi S. (2017), who evaluated a large cohort of indeterminate thyroid nodules and detected RAS mutations in 18% of all series.

Our series was particularly enriched in PTCs, which allow us to evaluate the clinicopathological implications of the genetic background of the tumors, as evaluated by the molecular analysis of the histologies. Regarding the molecular results of the PTC’s, we found a prevalence of mutations in BRAF and TERT genes similar to those reported by Liu R. (2014) [41] and Insilla A. (2017) [42], but lower than in some studies (DOI: 10.3390/cancers13092048).

Several statistically significant associations between the clinicopathological and molecular features of the PTCs were found in our series. BRAF mutations were significantly associated with several clinicopathological features, in accordance with Liu, X. (2018) [43]. TERTp mutations were significantly associated with features of worst prognosis, as mentioned by Bournaud et all [44]. On the contrary, RAS mutations were associated with the presence of capsule and a better outcome, as reported in previous study (Povoa et al) [45].

Xing, M. et al. (2014) reported that 6.9% of all PTCs have concomitant mutations in TERTp and BRAF genes, which were significantly associated with clinicopathological features of worst prognosis and tumor progression, namely LNM, extra thyroidal, and vascular invasion. Estrada Flórez A. P. et al. (2019) published similar results, with concomitant TERTp and BRAF mutations in 10% of their series and the same significant associations. In accordance with the above mentioned studies, we observed concomitant TERTp and BRAF mutations in 5.6% of PTCs, significantly associated with clinicopathological features related to tumor aggressiveness. On the other hand, Melo M. et al. (2014) have not found this association in their series. Ren H. et al. (2018) have found no associations between presences of both mutations and LNM, but they found a significant association with extra thyroidal invasion, large tumors, and older patients. No concomitant mutations were observed for BRAF and RAS as expected since these mutations are described as mutually exclusive events [46].

The BRAF and/or TERTp mutations were only present in malignant cases, in both cytology and histology samples, which we could consider as “rule in” tests, in accordance with American Thyroid Association (ATA) guidelines (PPV >95%) [47,48]. RAS mutations presented slightly lower specificity in our series due to its presence in some benign cases and low grade tumors. However, RAS mutation showed a higher specificity when compared with previous reports, resulting from the selection of cases in our series. The sensitivities and NPV of all mutations were much lower in both cytology and histology samples, not excluding malignancy within our series, as reported in other studies [49,50].

This study, conducted in a European population, presented limitations namely due to its retrospective nature, the only way to obtain a large series of operated thyroid nodules, and to perform a matching cyto-histological comparative analysis of the genetic profile in DTCs. The small number of genetic mutations analyzed, and the detection limit of Sanger sequencing used in this study, could be considered a potential limitation, especially when compared to the extensive mutation panels available [51,52] and other sequencing methods. However, our aim was to assess the reliability of molecular diagnoses obtained by US-FNAC in thyroid nodules using a laboratory method that offers practical replicability, is widely available, and at low costs. By focusing on a smaller set of mutations, we assessed the confidence level of our findings and provided valuable insights within the constraints of our study design.

Our results obtained by cyto-histological comparative analysis of the genetic profile in thyroid nodules reinforce the potential clinical utility of molecular testing in cytological assessments by US-FNAC, contributing to anticipate the genetic profile of the tumors and its biological behavior. By incorporating genetic information, clinicians achieve better informed decisions, optimize treatment strategies, including active surveillance and surgical options, and provide better care for patients with ambiguous cytological findings.

5. Conclusions

Our study yielded compelling results, demonstrating a good agreement between cytological and histological findings for molecular alterations in both DTCs and in indeterminate nodules. This minimally invasive and cost-effective approach has the potential to enhance diagnostic accuracy and streamline patient management by integrating molecular testing into routine clinical practice. While the genetic profile’s ability to exclude malignancy is limited, it effectively confirms malignancy in cytology samples, reducing repeated US-FNAC, and contributing to refer patients towards diagnostic, including surgical option. Moreover, the numerous associations identified between clinicopathological features and genetic profile of PTCs, suggest that molecular analysis in US-FNAC can provide early insights into tumor genetic profile and behavior. To gain a comprehensive understanding of tumor evolution and assess the applicability of molecular tests in the European population, further studies are needed that will shed new light on the dynamics of these tumors and their genetic characteristics.