Value of pyruvate carboxylase in thyroid ne-needle aspiration wash-out uid for predicting PTC lymph node metastasis

Abstract


Abstract Background
The incidence of papillary thyroid carcinoma (PTC) is increasing yearly. Knowing whether there is lymph node metastasis is of great value in choosing surgery and surgical methods, but ultrasound cannot qualitatively diagnose and effectively detect lymph node metastases in the central area. Metabolic dysregulation is an important factor associated with malignancy and metastasis of tumors. Pyruvate carboxylase (PC) is a major anaplerotic enzyme that catalyzes the carboxylation of pyruvate to form oxaloacetate (OAA) and is involved in tumorigenesis in several cancers. This study aims to explore the relationship between PC mRNA expression in thyroid ne-needle aspiration (FNA) wash-out uid and lymph node metastasis of papillary thyroid carcinoma (PTC).

Methods
The expression levels of PC in PTCs and normal thyroid tissues were rst compared based on bioinformatics analysis of public databases, including the Gene Expression Pro ling (GEPIA), Oncomine and Gene Expression Omnibus (GEO) databases. Then, the PC mRNA levels were measured by RT-PCR in surgical tissues in a total of 42 patients with surgically con rmed PTC and compared to those of patients with and without central lymph nodal metastasis (CLNM). Further, to identify PC expression in diagnostic biopsies, a total of 71 thyroid nodule patients with ultrasound-guided ne-needle aspiration wash-out uid samples and cytological diagnosis were prospectively enrolled in the study.

Results
Through data mining, we found that PC is overexpressed in PTC and PTC with lymph node metastasis. In 42 surgically diagnosed PTC patient tumor tissues, PC was signi cantly higher in patients with CLNM (n=15) than in those without (n=27) nodal metastasis (median: 6.490 vs. 2.430, p = 0.014). Furthermore, in thyroid FNA wash-out uid, PC was signi cantly higher in patients with cytologically diagnosed PTC (n=52) than in those with benign nodules (n =19) (median: 2.456 vs. 0.495, p = 0.005). Thirty-four of the 52 patients with PTC received a total thyroidectomy, and PC was also signi cantly higher in patients with CLNM (n=17) than in those without nodal metastasis (n=17) (median: 3.665 vs. 1.621, p = 0.013). The area under the ROC curve for PC predicting PTC with lymph node metastasis in FNA wash out samples was 0.751 (p = 0.013), and the sensitivity and speci city were 70.6% and 76.5%, respectively.

Conclusions
The ndings of this study suggest that PC may predict malignancy in thyroid nodules and lymph nodal metastasis with thyroid FNA wash-out uid in PTC patients.

Page 3/18
Thyroid carcinoma is the most common endocrine tumor in the world, and its worldwide incidence has signi cantly increased yearly (1;2). Nearly 90% of thyroid cancers are differentiated, including papillary thyroid carcinoma (PTC) and follicular thyroid carcinoma (FTC) (3). PTC is the most common subtype, accounting for 80%-85% of all malignant thyroid tumors (2). PTCs exhibit an indolent biological behavior (4), and the 5-year survival rate reaches almost 95% after treatment (5), such as surgical operation and radioactive iodine therapy. There is a debate about whether PTC, especially papillary thyroid microcarcinoma (PTMC), is over-diagnosed and overtreated. Ito et al (6) reported that only a small portion of PTMCs showed progression and suggested that subclinical low-risk PTMC should be surveilled instead of receiving immediate surgery. Alternatively, some cases of PTMCs show nodule enlargement and early lymph node metastasis during the surveillance period. It is critical to develop a tool to identify those patients with early nodal metastasis to guide clinical management.
At present, the main clinical diagnostic methods of PTC are ultrasonographic and ne-needle aspiration biopsy (FNAB) (7). It is di cult to identify the aggressiveness of PTC based on these diagnostic methods given their inherent limitations (8). With the development of basic medical research, molecular genetic alteration has been increasingly studied, and substantial tumor biomarkers were found, improving the accuracy of PTC risk categories (4). The most widely studied and clinically used PTC tumor markers are BRAF and TERT promoter mutations (9), though the incidence of the two mutations are 83.7% and 7.5%-27%, respectively (10) (11), and neither has signi cant prognostic value for predicting the high risk clinicopathological features of PTC (12). Therefore, it is important to explore other tumor markers that affect the aggressive biological behavior of PTC including nodal metastasis.
Cancer cell proliferation requires energy, which is dependent on cellular metabolism (13). In proliferating cancer cells, the replenishment of intermediates for the tricarboxylic acid (TCA) cycle provides biosynthetic precursors to synthesize proteins, nucleic acids and lipids (14;15). The pyruvate carboxylase (PC)-mediated anaplerosis pathway for TCA intermediates plays an important role in improving the PTC cellular TCA cycle (16). By detecting the role of PC expression levels in thyroid FNA wash-out uids from PTC patients with LNM, we explored the feasibility of PC as a novel tumor biomarker for PTC tumor aggressiveness.

Materials And Methods
Gene expression data acquisition and bioinformatic analysis Gene Expression Pro ling Interactive Analysis (GEPIA)-(http://gepia.cancer-pku.cn/) (17) is an online server for pro ling and interactive analyses of cancer and normal gene expression. Oncomine (https://www.oncomine.org/) (18;19) is a cancer microarray database and integrated data-mining platform online tool. We searched PC in the two databases to analyze PC expression in thyroid cancer and other types of cancer. In addition, we used the expression pro ling arrays (GSE60542) from Gene Expression Omnibus databases to analyze the different expression genes between thyroid cancer tissue Page 4/18 with or without lymph metastasis. GSE60542 dataset was analyzed by Affymetrix Human Genome Array with annotation platform GPL570 [HG-U133_plus_2]. The GSE60542 dataset contained 14 PTC samples with no lymph nodes metastasis, 19 samples with lymph node invasion and 10 control samples from normal thyroid. Using R packages, we assessed GSE60542 RAW datasets by background correction, normalization, expression calculation, and probe integration. Robust multi-array average (RMA) and mismatch probe (PM) were created for processing the datasets. P-values was adjusted by the Benjamini-Hochberg method and fold-changes (FC) were calculated using the false discovery rate (FDR) procedure. PTC-DEGs with a |log2-fold change| > 1 and P < 0.05 were selected. The basic features of the GEO database are shown in Table 1. Surgically-con rmed PTC patient populations The study enrolled 42 surgically-con rmed PTC patients with surgical tissues from July 2018 to November 2018. Clinical characteristics of the patients are shown in Table 2. Thyroid nodule patient populations A total of 71 thyroid nodule patients with FNA wash-out uid samples from November 2019 to January 2020 were also included. Fifty-two of the 71 patients were diagnosed as PTC, and 19 had benign nodules based on cytological ndings. Thirty-four of the 52 cytologically diagnosed PTC patients underwent total thyroidectomy and central nodal dissection with nal tissue sample con rmation. Clinical characteristics of the patients are shown in Table 3.

Results
Dataset online analysis of PC expression in thyroid cancer PC mRNA expression through the GEPIA datasets differs in different types of tumors (Fig. 1A, B) but is signi cantly overexpressed in thyroid cancer (THCA) (Fig. 1C, D). In addition, expression of PC is also signi cantly higher in PTC than in normal tissues in 2 separate public datasets in Oncomine. In Vasko's dataset, PC is overexpressed in thyroid cancer with a fold change of 2.516 ( Fig. 2A). In He's dataset, PC is also highly expressed with a fold change of 2.061 (Fig. 2B). According to GSE60542, PC expression level is signi cantly higher in PTC with or without lymph metastasis tissues compared to normal thyroid samples, and the fold change is 1.10 and 1.22, respectively (Fig. 2C).  Fig. 3B). Univariable logistic regression analysis revealed that PC expression and ages < 55 were associated with CLNM (all p < 0.05). Furthermore, multivariable analysis showed that PC and younger ages were independent predictors for CLNM (Table 4). To validate the sensitivity and speci city of PC expression as a biomarker for PTC with CLNM, we applied ROC curve analysis. The area under the curve (AUC) was 0.751 (P = 0.013, Fig. 4). The sensitivity and speci city were 70.6% and 76.5%, respectively. The results suggest that PC expression is potentially a novel biomarker for LNM of PTC. A review of a surveillance, epidemiology, and end results (SEER) database study found that cervical lymph node metastases represent a high risk of locoregional recurrence and survival (23)(24)(25). Bake et al (26) reported that the PTC with LNM group has a shorter disease-free survival period than the LNMnegative group. The ratio of the number of LNMs being greater than 30% of the monitored lymph nodes is a signi cant independent prognostic factor in PTC (27). Central lymph node metastasis, which is the sentinel lymph node of cervical lymph node metastasis, is also a risk factor of recurrence (28). However, it is sometimes di cult to recognize central lymph nodes with ultrasonography (29).
Molecular tests have improved the accuracy of US-FNA cytology (30)(31)(32), but there is no effective molecular marker for predicting lymph node metastasis. Although BRAF V600E and TERT C228T mutations are widely tested in the clinic, 40% of DTCs with distant metastases show negative BRAF or TERT genetic mutation (33). Liu et al. (34) found that the TERT C228T mutation and the BRAF V600E mutation led to dedifferentiation and aggressive biological behavior of thyroid cancer (35). However, the probability of the BRAF V600E and TERT C228T mutations occurring simultaneously in PTC is 13% (36). In addition, Ren et al (37) reported that coexistence of BRAF V600E and TERT C228T mutations has no obvious correlation with PTC lymph node metastasis. Hence, nding a new biomarker is important in the management of PTC.
Overexpression of PC is found in many human cancers (38-40). Using bioinformatics analysis of online databases, we found that PC mRNA expression level was signi cantly higher in the PTC group than in the normal group. Additionally, PC expression is also higher in the CLNM positive group compared to the negative group. Then, we con rmed the above results in surgical PTC tumor tissues, suggesting that PC plays an important role in predicting PTC tumor aggressiveness, consistent with a previous nding (16).
To our knowledge, this study is the rst to study the diagnostic role of PC in thyroid nodule patients after FNA. Our results suggested that PC has a certain signi cance for suggesting malignant thyroid nodules. Additionally, younger age is another independent predictor for CLNM of PTC, consistent with previous ndings.
The mechanism of high PC expression in PTC with LNM may be related to its involvement in cancer metabolic pathways (41). The aerobic glycolysis response restrains pyruvate oxidation in mitochondria, inducing more TCA cycle intermediates (41;42), and the PC-regulated anaplerotic reaction is important for cancer cells to replenish TCA cycle intermediates, which has been identi ed in many cancers (43).
Christen et al (38) used 13 C tracer analysis and found that the PC mRNA expression level and PC enzyme activity were increased in breast cancer with lung metastasis compared to primary breast cancer. Lee et al (45) suggested that PC may be associated with tumor cell invasion by activating the Wnt/Snail signaling pathway and EMT (epithelial-mesenchymal transition) in breast cancer cells. However, the role of PC regulating LNM in PTC needs further research.
Because PC may be an independent predictor for CLNM and PC mRNA expression in thyroid FNA washout uid can be easily obtained, quantitatively determined, and speci cally re ected lymph node metastasis of PTC, surgical treatment should be actively performed on PTC patients whose thyroid FNA wash-out uid shows high expression of PC, which may improve the patient's prognosis and survival rate. Alternatively, active surveillance should be suitable for patients with low or no PC expression, which may improve the quality of life of PTC patients.
The speci city of PC for predicting CLNM is higher than the sensitivity, which may be because it is di cult to distinguish lower PC expression of PTC from benign tissue. Although it is not ideal to predict the sensitivity and speci city of CLNM by PC expression level in thyroid FNA wash-out uid, it is possible to further improve their diagnostic sensitivity and speci city by combining other molecular markers, such as BRAF gene detection.

Conclusion
Our study is the rst to report that PC is an independent predictor for CLNM and re ects malignant thyroid nodules after FNA treatment. These results may provide clinical guidance for active surgical treatment of thyroid nodules with high PC expression and present a good application prospect.

Declarations
Ethics approval and consent to participate The study design was approved by the Ethical Review Board of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. All patients had signed informed consents.