Whether Detection of Gene Mutations Could Identify Low- or High-Risk Papillary Thyroid Microcarcinoma? Data from 393 Cases Using the Next-Generation Sequencing

Objective The objective of this study is to explore the utilization of next-generation sequencing (NGS) technology in evaluating the likelihood of identifying individuals with papillary thyroid microcarcinoma (PTMC ≤10 mm) who are at high or low risk. Design NGS was used to analyze 393 formalin-fixed, paraffin-embedded tissues of PTC tumors, all of which were smaller than 15 mm. Results The study found that bilateralism, multifocality, intrathyroidal spread, and extrathyroidal extension were present in 84 (21.4%), 153 (38.9%), 16 (4.1%), and 54 (13.7%) cases, respectively. Metastasis of cervical lymph nodes was identified in 226 (57.5%) cases and 96 (24.4%) cases with CLNM >5. Out of the total number of cases studied, 8 cases (2.3%) showed signs of tumor recurrence, all of which were localized and regional. Genetic alterations were detected in 342 cases (87.0%), with 336 cases revealing single mutations and 6 cases manifesting compound mutations. 332 cases (84.5%) had BRAFV600E mutation, 2 cases had KRASQ61K mutation, 2 cases had NRASQ61R mutation, 8 cases had RET/PTC1 rearrangement, 3 cases had RET/PTC3 rearrangement, and 1 case had TERT promoter mutation. Additionally, six individuals harbored concurrent mutations in two genes. These mutations were of various types and combinations: BRAFV600E and NRASQ61R (n = 2), BRAFV600E and RET/PTC3 (n = 2), BRAFV600E and RET/PTC1 (n = 1), and BRAFV600E and TERT promoter (n = 1). The subsequent analysis did not uncover a significant distinction in the incidence of gene mutation or fusion between the cN0 and cN1 patient cohorts. The presence of BRAFV600E mutation and CLNM incidence rates were found to be positively correlated with larger tumor size in PTMC. Our data showed that gene mutations did not appear to have much to do with high-risk papillary thyroid microcarcinoma (PTMC). However, when we looked at tumor size, we found that if the tumor was at least 5 millimeters in size, there was a higher chance of it being at high risk for PTM (P < 0.001, odds ratio (OR) = 2.55, 95% confidence interval (CI): 1.57–4.14). Identification of BRAFV600E mutation was not demonstrated to be significantly correlated with advanced clinicopathological characteristics, although it was strongly associated with a bigger tumor diameter (OR = 4.92, 95% CI: 2.40–10.07, P < 0.001). Conclusion In clinical practice, BRAFV600E mutation does not consistently serve as an effective biomarker to distinguish high-risk PTMC or predict tumor progression. The size of the tumor has a significant correlation with its aggressive characteristics. PTMC with a diameter of ≤5 mm should be distinguished and targeted as a unique subset for specialized treatment.


Introduction
Papillary thyroid carcinoma (PTC) is a pathological subtype of thyroid cancer that is the most common of all endocrine-related cancers.It is known to have low-grade, inert biological characteristics.Papillary thyroid microcarcinoma (PTMC) is defned as PTC measured ≤10 mm in its greatest diameter.In recent years, there has been a notable upward trajectory in the prevalence of PTC, with approximately half of cases diagnosed as PTMC [1][2][3].However, the majority of PTMC cases are actually occult and detected due to overscreening in healthy individuals.Based on the satisfactory results from Kuma Hospital [4,5], active surveillance (AS) has been proposed for patients with low-risk PTMC [6].However, how to reliably and safely identify low-risk PTMC patients has noticeably attracted thyroid surgeons' attention.
Over the last two decades, substantial molecular genetic alteration studies have provided better insights into the understanding of the progression of PTC.Specifc point mutations in BRAF or RAS, or the presence of RET chromosomal rearrangement, have the ability to activate the receptor tyrosine kinase (RTK)/mitogen-activated protein kinase (MAPK) pathway, a critical regulator of various cellular processes including proliferation, diferentiation, adhesion, migration, and apoptosis [7][8][9].PTC can progress to poorly or undiferentiated thyroid cancer by additional hits on TERT, tumor suppressors, or phosphoinositide 3-kinase (PI3K) pathway genes [8,[10][11][12].In this study, we retrospectively analyzed the gene mutation status in 393 PTC patients using patients' clinical and pathological characteristics by next-generation sequencing (NGS), in order to investigate the possibility of the identifcation of high-or low-risk PTMC patients.

Patients and Samples.
A total of 393 formalin-fxed, parafn-embedded (FFPE) tumor tissues from surgically removed thyroid samples were retrospectively analyzed by the NGS between August 2018 and July 2020.All of the tumors with a maximum diameter of less than 15 mm were confrmed to be PTC by pathological examination.Te postoperative follow-up periods for a total of 393 patients spanned a range from 6 to 103 months, with a median duration of 20 months.Clinical records of these individuals who underwent surgical procedures between the timeframe of February 2013 and July 2020 were meticulously reviewed utilizing a comprehensive database retrospectively compiled by the Department of Head and Neck Surgery, the Afliated Sir Run Run Shaw Hospital.Te indication of thyroidectomy was strictly in accordance with the 2009 or 2015 American Tyroid Association (ATA) guideline [13,14].Te indications of central neck dissection (CND) and modifed lateral neck dissection (LND) were previously described [15].Tis research project was endorsed by the Ethical Committee of the Sir Run Run Shaw Hospital Afliated to Zhejiang University School of Medicine.
Tere were 300 cases with cN0, including 103 cases with a diameter of ≤5 mm, 151 cases with a diameter of 5∼10 mm, and 46 cases with a diameter of 10∼15 mm, as well as 93 cases with cN1 among 393 PTC patients.Tere were 320 patients with PTMC, including 218 high-risk and 102 low-risk PTMC patients (Figure 1).

Sample Preparation and Nucleic Acid Extraction.
Following vendor manual instructions, genomic DNA and mRNA from FFPE samples were extracted using an AllPrep DNA/RNA FFPE tissue kit.Tis extraction kit is manufactured by Qiagen in Germany.Concentrations of DNA and cDNA were estimated using Qubit 3.0 from Termo Fisher Scientifc in the USA.In addition, DNA fragment distribution was assessed by a Qsep100 system from Bioptic, Taiwan, China.

NGS Library Preparation.
Te sequencing library should be prepared by the specifc guidelines outlined in the U.S. Paragon Genomic DNA related to thyroid carcinoma was enriched by multiplex PCR, which was then connected with barcode and sequencing adapters.Te amplifed DNA libraries were purifed using AMPure XP beads from Beckman Company of the United States.Te Termo Qubit 1 × dsDNA Assay Kit was a specifc assay for quantifying next-generation sequencing libraries.Te Qsep100 system developed by Biotic was used to analyze fragment size distribution.Te well-known oncogenic drivers, including BRAF, RAS, TERT, and TP53, as well as gene rearrangements, such as RET and PAX-PPARG, were detected and are described in detail in Table 1.

Sequencing and Bioinformatics
Analysis.Te sequencing process was carried out on the NextSeq 500 sequencing platform from Illumina, using the PE150 sequencing strategy.Tumor samples demonstrated a mean coverage depth of approximately 1000X.Te threshold for the initiation of allele mutations was set at 5%.Sequencing data were frst converted to FASTQ format, and then, the human genome was mapped to the human reference sequence (hg19) using the Burrows-Wheeler aligner.Strelka was used to detect somatic insertions and deletions (InDel), while muTect was used to identify somatic single nucleotide variants (SNV).Genomic variant annotation was performed via the ANNOVAR 21 software.

Statistical
Analysis.SPSS 20.0 software (IBM, Armonk, NY, USA) was employed for the execution of the statistical analysis.All continuous variables were reported using the mean value (standard deviation, SD) or median value (range).Categorical variables were represented as counts and corresponding percentages (%) to indicate the frequency of each category.In the event of data consistent with normal distribution and variance homogeneity, statistical analysis of diferences between groups was conducted through the application of ANOVA.Mann-Whitney-Wilcoxon tests were used to evaluate diferences among various groups, 2 International Journal of Endocrinology which were composed of data with skewed distributions.Categorical variables were assessed using either Pearson's chi-square test for large samples or Fisher's exact test for small sample sizes.To determine statistical signifcance, we adopted a two-sided signifcance threshold of P < 0.05 across the distribution.No mutations in TP53 and HRAS genes were found, and PAX8/PPARG translocation was not detected.Besides, there were 6 cases simultaneously carrying 2 gene mutations (fusions), including BRAF V600E and NRAS Q61R (n � 2), BRAF V600E and RET/PTC3 (n � 2), BRAF V600E and RET/PTC1 (n � 1), and BRAF V600E and TERT promoter mutation (n � 1).No one had three or more gene mutations (fusions).However, it is noteworthy that 5 patients had older ages (>60 years old), and all of their PTCs had low-risk manifestations.Te clinicopathological data of these 6 patients are listed in Table 3.

Comparison of Gene Mutation Status and Clinical Pathological Characteristics between cN0 and cN1 Patients
with Tumor Diameter Less than 15 mm.Among 393 PTC patients, 93 (23.7%) patients showed clinical CLNM (cN1) before surgery.Te 393 PTC patients whose tumor size was less than 15 mm were dispensed into two groups: cN0 group (n � 300) and cN1 (n � 93) group.Group cN1 was regarded as simulated clinical progression of observable PTC with cN0 (≤15 mm in size).Table 2 compares the clinicopathological features and genetic alterations between the cN0 and cN1 groups.

Correlation Analysis of Tumor Growth with Clinicopathological Features and Gene Mutation Status in cN0 PTC
Patients with Tumor Size Less than 15 mm.In order to gain insight into the genetic mutation status during the progression of PTC, a total of 300 cN0 PTC cases were categorized into three parts based on primary tumor diameter: the frst group (group A) consisted of 103 patients with tumor diameters of 5 mm or less, while the second group (group B) had a total of 151 patients with tumor diameters ranging from 5 mm to 10 mm.Te third group (group C) contained 46 patients with tumor diameters ranging from 10 mm to 15 mm.Clinical progression of observable PTMCs was simulated by evaluating group C. Te data are shown in Table 4.
Terefore, the incidence rates of CLNM and BRAF V600E mutation could increase during the PTMC growth.

Perform Univariate and Multivariate Logistic Regression
Analyses on PTMC High-Risk Factors.All PTMC patients were divided into two groups: low-risk group (n � 102) and high-risk group (n � 218).It was found that BRAF V600E , NRAS Q61R , KRAS Q61K , RET/PTC1, and RET/PTC3 showed no correlation with high-risk PTMC (Table 5).Te multivariate logistic regression analysis has been further confrmed that gene mutations were not risk factors for highrisk PTMC in the present study; however, a signifcant correlation was identifed between bigger tumor diameter and high-risk PTMC (P < 0.001, OR � 2.55, 95% confdence interval (CI): 1.57-4.14).

Discussion
In a 30-year longitudinal study of 140 cases, Woolner et al. established the concept of occult papillary carcinoma in 1960, referring to PTCs less than or equal to 15 mm in size [16].Back in 1989, the World Health Organization created a new terminology to replace occult papillary carcinoma.Tey named it papillary microcarcinoma and decided it should be defned as papillary thyroid cancer that is less than or equal to 10 mm in diameter [17].PTMC has been recognized as an essential variation of PTC due to exhibiting minimal malignant potential and rarely undergoing distant metastasis.Tese characteristics are typically found in International Journal of Endocrinology  International Journal of Endocrinology than or equal to 15 mm.We divided these patients into three groups based on their tumor size.Te tumors that measured from 10 mm to 15 mm were presumed to represent a clinical progression of the visually noted PTMC.
BRAF, RAS, and RET/PTC mutations are commonly recognized as mutually exclusive in PTC.Tis suggests that more than one mutation in these genes would likely not Furthermore, they found that the concomitant mutations were associated with the advanced stage of the disease [24].
In the present study, there were 6 PTC cases simultaneously carrying 2 gene mutations, including BRAF V600E and NRAS Q61R (n � 2), BRAF V600E and RET/PTC3 (n � 2), BRAF V600E and RET/PTC1 (n � 1), and BRAF V600E and TERT promoter mutation (n � 1).However, such cases showed no noticeable aggressive tumor behaviors.Among these 6 patients, it was found that 5 patients had older age (> 60 years old); thus, we hypothesized that with the increase of age, the efects of carcinogenic factors on the body were enhanced, which might lead to multiple mutations or fusion of somatic genes.
It is recommended at Kuma Hospital that low-risk patients with PTMC undergo active surveillance (AS) as the primary treatment [6].In the event that a tumor exhibits an increase in size of 3 mm or more, a tumor measuring 12 mm in size, or the presence of lymph node metastases during the follow-up period, prompt consideration should be given to the implementation of "rescue surgery."What role does gene mutation play in the course of disease progression?Whether gene mutation could predict the progression of PTMC?In light of these observations, we employed our dataset to model the progression of clinical symptoms and designed two distinct segments.In the frst part, it was attempted to simulate LNM using two groups of cN0 and cN1, according to whether LNM was confrmed preoperatively.Expectedly, more aggressive features of PTMC were found in the cN1 group, while gene mutation showed no signifcant diference between the two groups.In the second part, cN0 patients were divided into three groups: ≤5 mm, 5∼10 mm, and 10∼15 mm according to the tumor size, in order to simulate tumor growth.It was found that the incidence of BRAF V600E mutation signifcantly increased during the growth of PTMC, suggesting that BRAF V600E mutation might promote tumor growth.Te mechanism may include three aspects: (1) the BRAF V600E mutation leads to sustained activation of the RAS-RAF-MEK-ERK/MAPK pathway, enhancing cell mitotic capacity, ultimately leading to abnormal cell proliferation, and inducing tumor occurrence [10].(2) Te BRAF V600E mutation promotes methylation of the thyrotropin receptor (TSHR) gene promoter, leading to TSHR silencing or a signifcant decrease in expression, increasing the feedback of thyrotropin (TSH), and promoting tumor cell growth [25].(3) Te BRAF V600E mutation upregulates TERT expression in a non-TERT promoter-dependent manner through the oncogene MYC, which also promotes the occurrence of PTC [26].However, in clinical practice, BRAFV600E mutation could be used as a biomarker for identifying thyroid nodule malignancy, rather than predicting tumor growth.
Multiple studies have documented a favorable prognosis among patients with PTMC [27][28][29][30], with results indicating a clinical course that is generally favorable.Tis favorable prognosis has been corroborated by prospective clinical studies, in which patients with PTMC who chose not to undergo surgical intervention received AS primarily [31,32].Zaid et al. reported a series of 30,180 adult patients with PTMC.Notably, 5,628 patients (18.7%) showed advanced features, including central lymph node metastasis (8.0%), lateral lymph node metastasis (4.4%), microscopic extrathyroidal extension (6.7%), macroscopic extrathyroidal extension (0.3%), lymphovascular invasion (4.4%), and distant metastasis (0.4%).Tey described PTMC as a wolf in sheep's clothing and suggested that lobectomy seemed to be a more appropriate treatment for low-risk PTMC patients rather than AS [33].Whether gene detection can identify low-or high-risk PTMC patients preoperatively?PTMC was dispensed into two groups in our study: the high-risk group, in which PTMC had some advanced features, including ETE, LNM, multifocality, and intrathyroidal spreading, and the low-risk group (without any aggressive features).As a result, it was revealed that the larger the tumor size, the more aggressive the characteristics, and only tumor size could afect the advanced features of PTMC, rather than gene mutation, especially BRAF V600E .
Although the primary objective of this study was to explore the correlation between gene mutations and the aggressiveness of the PTMC, it was clearly found that the tumor size, rather than gene mutation, was signifcantly associated with the progression of high-risk PTMC.For instance, when the tumor grew up from 5 to 10 mm, the incidence of LNM was doubled, and the nonmutation rate of BRAF V600E was tripled.Te fndings of the present study were also similar to the previously reported results.After analysis of 977 PTMC cases, Zheng et al. found that tumor size (>5 mm) was an independent risk factor associated with LNM [34].In a study looking at two screening criteria for AS, Qian et al. found that low-risk patients with less than 5 mm of PTMC had a lower chance of having multifocal lesions, ETE, central LNM, or disease progression.Tis means they were also less likely to relapse and had longer disease-free survival times [18].Terefore, a tumor with a size of less than 5 mm should be regarded as a special subgroup of PTMC that showed to be less aggressive than the other PTMCs and might be treated specially with local ablation, such as radiofrequency, in order to block the tumor progression during the follow-up of PTMC.
In conclusion, our fndings indicate that BRAFV600E mutation may not be a reliable biomarker to clinically distinguish high-risk PTMC patients or predict tumor progression.Tumor size was found to be highly correlated with the disease's aggressiveness.PTMC tumors measuring ≤5 mm should be identifed as a distinct subgroup and treated with specifc treatment strategies.

4.1.
Limitation.Te present study had several limitations.First, the 393 patients in this cohort were analyzed inconsecutive from 2013 to 2020 because we were unable to obtain consent from all patients for NGS of their samples.Te other reason was that not all patient samples were of suitable quality for NGS.Terefore, a selective bias might exist in the study.Second, the relatively short follow-up time may have an impact on the long-term recurrence rate.In our study, tumor recurrence was detected in 9 (2.3%) cases, and it may be better than other studies, but we believe the data are true.From January 2013 to December 2015, our medical center treated a total of 562 patients with papillary thyroid cancer (PTC) with initial surgical therapy during our prior research endeavor, which consisted of either unilateral lobectomy or total thyroidectomy, and central neck dissection.Some patients also underwent lateral neck dissection.Five recurrence cases were identifed between a minimum followup period of 40 months and a maximum follow-up period of 75 months, resulting in a median follow-up period of 57 months [35].We speculate that the recurrence rate is lower than others because all the patients underwent prophylactic or therapeutic cervical lymph node dissection, and the surgical team is very experienced and well-practiced.Tird, samples for the NGS were from formalin-fxed, parafn-embedded tumor tissues and histologic subtypes of PTC were not further classifed; therefore, the map of gene mutation might be slightly diferent from other relative studies.Lastly, this NGS panel was small, but it included 9 hot spot gene mutations and rearrangements related to thyroid cancer and is a cost-efective auxiliary diagnostic test for thyroid nodules.Te other PTC-associated drivers, such as EIF1AX, AKT1, and PIK3CA, will be considered and measured in future research.

Table 1 :
Genetic testing site in the study.
history of undergoing radiation therapy for head and neck cancer before surgery and distant metastasis.In addition, 3 patients were lost to follow-up.As presented in Table2, 332 (84.5%) cases showed BRAF V600E mutation, and RAS mutation was detected in 4 (1%) cases, including KRAS Q61K and NRAS Q61R in 2 cases

Table 2 :
Clinicopathological characteristics and genetic alterations of 393 PTCs subdivided into cN0 and cN1 groups.All of the point mutations were missense mutations.RET/PTC rearrangement was found in 11 (2.8%) cases, including RET/PTC1 in 8 cases and RET/PTC3 in 3 cases.TERTpromoter gene mutation was detected in 1 (0.3%) case.

Table 3 :
Te clinicopathological characteristics of PTC patients simultaneously carried two gene mutations (fusions).
[18]ction rate of this type of cancer in the living.Te best diameter of low-risk PTMCs for observation is usually between 10 mm and 15 mm.However, the Chinese Association of Tyroid Oncology (CATO) recommends that the dimensions of low-risk papillary thyroid microcarcinomas identifed should not exceed 5 mm[18].Te perspective of CATO is derived from a comprehensive examination of two discriminating markers for AS in a sample of 1,001 low-risk patients diagnosed with PTMC.Compared to the fndings in the Kuma low-risk PTMC group (lesions ≤10 mm), Qian et al. observed a signifcantly lower incidence of multifocal lesions, extrathyroidal extensions (ETEs), central LN metastasis, and progression rates in the CATO low-risk PTMC group (lesions ≤5 mm).Moreover, there was an observed improvement in the duration of disease-free survival.So we selected PTC patients with primary tumors that were less

Table 4 :
Patient characteristics and genetic alterations of cN0 PTCs (n � 300).: wild type, na: not available, and PTC: papillary thyroid carcinoma.Data are expressed as mean ± SD or absolute number (%).P value means statistical analysis among the three groups by ANOVA or chi-square test.Data are expressed as mean ± SD or absolute number (%).Signifcant diferences are shown in bold characters (P < 0.05). wt

Table 5 :
Patient characteristics of 320 PTMC divided into low-and high-risk groups.
wt: wild type, na: not available, and PTMC: papillary thyroid microcarcinoma.Te high-risk PTMC was defned as having one or more advanced characteristics including multifocality, extrathyroidal extension (ETE), intrathyroid spread, and LN metastasis, whereas the low-risk PTMC has none of these features.Data are expressed as mean ± SD or absolute number (%).Comparison between low-risk and high-risk groups was performed by ANOVA or chi-square test when appropriate.Signifcant diferences are shown in bold characters (P < 0.05).

Table 6 :
Multivariate logistic regression analyses of more than 5 cervical lymph node metastases in 320 PTMC.