Diagnostic approach to thyroid cancer based on amino acid metabolomics in saliva by ultra-performance liquid chromatography with high resolution mass spectrometry

Highlights

• A simple and economical method based on UPLC-HRMS was developed for analyzing salivary amino acids as PTC biomarkers.

• Salivary amino acid levels were significantly reduced in PTC patients.

• 10 biomarkers for PTC diagnosis were successfully verified and a combined index showed high sensitivity/specificity.

• Salivary amino acids provide an auxiliary mean for non-invasive diagnosis of PTC.

Abstract

Thyroid cancer is a malignant disease with dramatically low advanced-stage 10-year survival. Meanwhile, the metabolites in saliva are becoming a wealthy source of disease biomarkers. However, there is a lack of non-invasive analytical methods for the identification of biomarkers in saliva for the preoperative diagnosis of thyroid cancer. Therefore, we developed an ultra-high performance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS) method to simultaneously determine the metabolic levels of 10 amino acids in saliva, aiming to study the amino acid metabolism profile to promote early diagnosis of thyroid cancer. We tested unstimulated whole saliva from patients with papillary thyroid carcinoma (PTC; n = 61) and healthy controls (HC; n = 61), and used receiver operating characteristic (ROC) curves to establish the diagnostic value of potential markers. The method validation results showed good precision, linearity (R² > 0.99), recovery (92.2 %–110.3 %), intra- and inter-day precision (RSD < 7 % and RSD < 9 %, respectively). The concentration of 10 amino acids was significantly different between PTC and HC in human salivary analysis (P < 0.05), the area under the curve (AUC) values of a single marker for the diagnosis of PTC were ranging from 0.678 to 0.833. A panel of alanine, valine, proline, phenylalanine was selected in combination yielded the AUC of 0.936, which will improve the accuracy of early diagnosis of thyroid cancer (sensitivity: 91.2 %; specificity: 85.2 %). This study proved the possibility of salivary amino acid biomarkers for PTC early diagnosis, providing a simple auxiliary way for the non-invasive diagnosis of thyroid cancer.

Introduction

Thyroid carcinoma is one of the most common endocrine cancer in head and neck, which responsible for 586,000 cases worldwide in 2020. Among them, the global incidence rate in women (10.2 per 100,000) is three times higher than that in men [1]. According to the origin and differentiation of tumors, Thyroid carcinoma can be divided into four categories: papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC), medullary thyroid carcinoma (MTC), and anaplastic thyroid cancer (ATC). PTC accounts for about 85–90 % of all thyroid tumors with an early stage of 10-year survival rates are as high as 90 %, while the 10-year survival rates are dramatically lower for later stage PTC [2]. Therefore, early diagnosis of PTC is an efficient strategy for improving the prognosis in PTC patients.

At present, Ultrasound-guided fine needle aspiration biopsy (FNAB) is the gold standard for diagnosis of thyroid cancer [3]. However, there are still shortcomings, such as FNAB cannot cytologically distinguish between benignity and malignancy, resulting in 10–25 % diagnostic uncertainty [4]. Some molecular biomarkers, including BRAF (A B-Raf proto-oncogene which from the RAF family of serine/threonine kinases), TERT (Telomerase reverse transcriptase) or RAS (Ras proto-oncogenes, which are the members of the superfamily of guanosine triphosphatase-binding proteins) mutations, RET/PTC (Rearranged in translation/PTC, RET proto-oncogene receptor) or PAX8/PPAR (Paired box 8/peroxisome proliferator-activated receptor) translocations have been reported for the diagnosis of PTC [5]. However, these results are still unsatisfactory, especially for invasiveness, low sensitivity and high cost. Therefore, it is imminent to develop new preoperative methods, especially methods for biomarkers to effectively diagnose PTC that improve the accuracy and sensitivity of diagnosis. In recent years, non-invasive matrices such as saliva, consensual exhaled breath, exhaled breath, among others have presented great and good advances in the discovery of biomarkers for various diseases, such as breast cancer [[6], [7], [8]], lung cancer [9], gastrointestinal cancer [10], oral cancer [11], Sjogren's syndrome [12], diabetes [13], Alzheimer's disease [14], etc. As one of the main sources of disease-related molecular biomarkers in the circulatory system of body fluids, saliva possesses practical superiorities of being non-invasive collection and cost-effective compared to blood and urine, which can significantly reduce anxiety and discomfort in matrix collection. Furthermore, there is no need for the medical personnel with professional training and easy to store [15,16]. On account of this, salivary biomarker analysis is appropriate for the detection of overall diseases and health [13,17]. Metabolomics is another important field of systems biology following genomics, proteomics and transcriptomics, it is aims to systematically study the small molecular metabolites that exist in biological systems after cellular responses to endogenous or exogenous disturbances such as disease or drugs [18]. As a mature “omics” technology, metabolomics has been widely applied in disease diagnosis and treatment, environmental detection, and finding biomarkers [[19], [20], [21], [22]]. The currently two major analytical techniques used in metabolomics profiling are nuclear magnetic resonance (NMR) [23,24] and mass spectrometry (MS) [25,26]. ¹H NMR analysis was performed to differentiate PTC patients from the healthy controls (HC) by Wojtowicz et al., and a panel of nine metabolites including Valine, Alanine and etc. were selected in combination yielded the area under the curve (AUC) of 0.83 [27]. Xu et al. use ultraperformance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-QTOFMS) and gas chromatography-time-of-flight mass spectrometry to discriminate individuals with PTC and benign thyroid adenoma from HC and found increased amino acid metabolism in tumor tissues [28]. According to literature reports, to a certain extent, UPLC-MS has the core advantages of higher sensitivity and mass accuracy compared with NMR [29]. Therefore, in order to obtain a satisfactory chromatogram with good resolution and peak shape in high efficiency, an UPLC coupling with high-resolution mass spectrometry (HRMS) method was used for the analysis of this study. UPLC is a further development of the method on the basis of LC, by using a smaller particle size (sub-2 μm) to achieve higher resolution, analysis speed and sensitivity. In addition, HRMS has higher mass accuracy and separation compared to ordinary MS [30].

Amino acids are the source of protein synthesis and the precursors of many important biological molecules in the body which have been proved to be bioactive and directly influence the health status of the body, serving as an important indicator of the body metabolic disorders and clinical symptoms [31]. According to the results of previous studies, it can be seen that some amino acids generally have significant differences between PTC and HC [32,33]. However, because amino acids are highly polar metabolite, it is difficult to retain and separate well in reversed-phase liquid chromatography (RPLC) without derivatization. In this case, it is a wise choice to analyze amino acids by hydrophilic interaction chromatography (HILIC). HILIC is orthogonal to RPLC and has been widely used due to its robustness in recent years. By coupling with MS instrument, which can improve the sensitivity of MS and obtain faster scanning speed without the need for complicated sample derivatization process and ion pairing reagents that are incompatible with MS [34,35]. In addition, the analysis of thyroid cancer biomarkers based on metabolomics is currently mainly limited to studies using blood or tissue samples [36,37]. Therefore, it is imperative to conduct research on amino acid markers in saliva to diagnose thyroid cancer.

In this study, HILIC-UPLC-HRMS was utilized to conduct metabolomic analysis of saliva samples from PTC patients, and saliva samples from healthy volunteers were used as a negative control. The objective is to develop and validate a direct, reliable and sensitive method for the analysis of amino acids in saliva, establish criteria for the diagnosis of PTC, and provide supporting evidence for clinical diagnosis. The workflow for quantitative detection of saliva samples used in this work is shown in Fig. 1.