Association between mutation profiles and clinicopathological features in Chinese patients with thyroid cancer

Recently, mutation profiles provided new insights into comprehensive understanding of TC biology by Next Generation Sequencing (NGS). We explored association between mutation profiles and clinicopathological features in Chinese patients with thyroid cancer (TC). Two hundred and twenty‐five formalin‐fixed, paraffin‐embedded tissue specimens from surgically removed thyroid samples were detected with 15 target genes by NGS. Mutation profiles and clinicopathological features were analyzed. Two hundred and seven mutations including two hundred mutations in 81.40% papillary thyroid carcinoma samples, three mutations in 50.00% MTC samples, and four mutations in 100% anaplastic thyroid carcinoma samples were detected. There were 19.56% samples without any mutations in target genes, 69.78% samples harbored mutations in single gene, 9.78% samples carried two gene mutations, and 0.89% samples had triple different gene mutations. For PTC, BRAF mutations were predominant, TERT mutations are more prevalent in advanced PTC and RET fusion was only observed among the PTC. For MTC, RET point mutations were predominant. For samples carried more than one gene mutations, the allelic frequency of mutants were almost similar. Multiple mutations in TC patients were significantly more frequent in cases of patients aged 55 and over (p <.001) and advanced American Joint Committee on Cancer (AJCC) cancer stage (p <.001). Gender (p = .309) and pathological subtype (p = .121) did not show significant correlation with mutations. Analysis between mutation profiles and clinicopathological features provides new insights into the biology of TC and is expected to increase the accuracy of diagnosis and prognostication in TC, leading to improved precision treatment for TC patients.

samples carried more than one gene mutations, the allelic frequency of mutants were almost similar. Multiple mutations in TC patients were significantly more frequent in cases of patients aged 55 and over (p <.001) and advanced American Joint Committee on Cancer (AJCC) cancer stage (p <.001). Gender (p = .309) and pathological subtype (p = .121) did not show significant correlation with mutations. Analysis between mutation profiles and clinicopathological features provides new insights into the biology of TC and is expected to increase the accuracy of diagnosis and prognostication in TC, leading to improved precision treatment for TC patients.

| Bioinformatics
The raw sequence data were mapped to the human genome (hg19)

| Clinical characteristics
The clinical characteristics of the patients are shown in Table 1 (Table 2). No mutation was detected in FTC case (Figure 1).

| Mutation allele frequency
All mutations were heterozygous mutations present with allelic frequency that ranged from 1.12 to 50.13% of alleles (which corresponds to 2.24-100.26% of cells with a heterozygous mutation). One RET point mutation (p.G691S) showed allelic frequency of more than 50% (50.13%) which was a germline variant. The fusion mutations showed the percent of reads ranged from 2.30 to 55.54%. For samples carried more than one gene mutations, the allelic frequency of mutants were almost similar. However, for sample harbored NRAS, TERT, and TP53 mutations simultaneously, the allelic frequency of NRAS mutation was similar to TERT mutation (33.62-39.67%), a lower allele frequency (4.13%) was detected in TP53 mutation.

| The relationship between clinical characteristics and mutations of TC patients
Multiple mutations in TC patients were significantly more frequent in cases of patients aged 55 and over (p <.001) and advanced AJCC cancer stage (p <.001). However, gender (p = .309) and pathological subtype (p = .121) did not show significant correlation with mutations (Table 1).

| DISCUSSION
In the past, single gene assays, BRAF point mutations in particular, have been commonly used for finding molecular alterations by Sanger sequence, Immunohistochemistry, and real time PCR in TC. In this study, we used NGS-based detection method for the analysis of multiple hotspot mutations concomitantly in a single experiment. This method offered a valuable tool for a comprehensive understanding of altered molecular events in TC.
Due to sample size limitations and patients from specific geographic locations, some mutations such as AKT1, TSHR, KRAS, PETN, and PAX8 mutations were negative among TC patients in the present study, which varied from previous studies. 5,8,[14][15][16] Our analysis showed the molecular profiles of the four TC subtypes were different. PTC samples were dominated by BRAF In conclusion, analysis between mutation profiles and clinicopathological features provides new insights into the biology of TC and is expected to increase the accuracy of diagnosis and prognostication in TC, leading to improved precision treatment for TC patients.

CONFLICT OF INTEREST
The authors declare that they have no competing interests.

AUTHOR CONTRIBUTIONS
ChangwenJing: conception, design, collection, and manuscript writing; Haixia Cao: data analysis; Jianzhong Wu, Rong Ma: conception and design; Zhuo Wang: conception and design. All authors: final approval of manuscript.

DATA AVAILABILITY STATEMENT
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

ETHICS STATEMENT
The research using human tissue was approved by the Nanjing medical university ethics committee. All patients participated in the study signed informed consent.