# Non-invasive detection of metastatic papillary thyroid carcinoma after radical surgery using salivary metabolomic biomarkers

**Authors:** Fei Yu, Jingya Pan, Liuting Zhang, Shaohua Li, Chuan Zhang, Jingjing Fu, Peng Zhou, Jun Wang, Xiaochen Yao, Yudan Ni, Ailing Zhang, Qingle Meng, Rui Yang, Lei Xu, Feng Wang, Jianhua Wang, Liang Shi

PMC · DOI: 10.20945/2359-4292-2026-0017 · 2026-02-16

## TL;DR

This study identifies salivary metabolites that can non-invasively detect metastatic thyroid cancer after surgery, helping determine proper radioactive iodine treatment.

## Contribution

The study introduces novel salivary metabolomic biomarkers for detecting metastatic papillary thyroid carcinoma post-surgery.

## Key findings

- 119 differential metabolites were identified, with 108 upregulated and 11 downregulated in metastatic PTC.
- Six metabolites (five lipids, one amine) showed strong diagnostic performance with area under the curve values exceeding 0.8.
- Key dysregulated pathways include necroptosis, choline metabolism, and sphingolipid signaling in metastatic PTC.

## Abstract

Accurate assessment of metastatic status is crucial for determining
radioactive iodine (RAI) dosing in postoperative papillary thyroid carcinoma
(PTC) patients. This study aimed to identify unbiased biomarkers in
metastatic PTC patients after surgery by applying a metabolomics workflow in
saliva samples.

Saliva samples from 70 postoperative PTC patients (35 metastatic PTC patients
in metastasis group and 35 non-metastatic PTC patients in control group)
were analyzed using liquid chromatography – mass spectrometry. Orthogonal
partial least-squares-discriminant analysis was applied to identify
differential metabolites and significant pathways were examined within these
metabolites. Receiver operating characteristic curve (ROC) analysis was
utilized to further evaluate the diagnostic performance of candidate
metabolites.

A total of 119 differential metabolites were identified, with 108 upregulated
and 11 downregulated. Pathway analysis revealed 13 significantly
dysregulated metabolic pathways in metastatic PTC, including necroptosis,
choline metabolism in cancer, sphingolipid signaling, valine, leucine and
isoleucine biosynthesis, linoleic acid metabolism and pantothenate and CoA
biosynthesis. ROC analysis demonstrated six discriminating biomarkers (5
lipids, 1 amine) that effectively distinguished metastatic from
non-metastatic PTC, with all area under the curve values exceeding 0.8.
Notably, these metabolites maintained diagnostic performance even in the
thyroglobulin antibody-positive subgroup (≥ 4.11 IU/mL) for
metastatic screening.

This study demonstrates the potential of salivary biomarkers as a
non-invasive diagnostic approach for metastatic PTC to aid the appropriate
dosing for RAI therapy. It also offers new insights into the mechanisms of
PTC metastasis and potential targets for adjuvant therapy.

## Linked entities

- **Diseases:** papillary thyroid carcinoma (MONDO:0005075)

## Full-text entities

- **Genes:** NR1H4 (nuclear receptor subfamily 1 group H member 4) [NCBI Gene 9971] {aka BAR, FXR, HRR-1, HRR1, PFIC5, RIP14}, TG (thyroglobulin) [NCBI Gene 7038] {aka AITD3, TGN}, VIP (vasoactive intestinal peptide) [NCBI Gene 7432] {aka PHM27}
- **Diseases:** benign (MESH:D009369), endocrine malignancy (MESH:D004700), pulmonary metastatic lesions (MESH:D008171), Metastatic lesions (MESH:D000092182), Metastasis (MESH:D009362), inflammatory (MESH:D007249), disease (MESH:D004194), DTC (MESH:D013964), necrosis (MESH:D009336), oral squamous cell carcinoma (MESH:D000077195), breast cancer (MESH:D001943), lymph node metastases (MESH:D008207), carcinogenesis (MESH:D063646), PTC (MESH:D000077273)
- **Chemicals:** isoleucine (MESH:D007532), amino acid (MESH:D000596), radioiodine (MESH:C000614965), choline (MESH:D002794), lipoic acid (MESH:D008063), fatty acid (MESH:D005227), alanine (MESH:D000409), phenylacetic acid (MESH:C025136), sphinganine (MESH:C005682), CDCA (MESH:D002635), amine (MESH:D000588), CoA (MESH:D003065), aldosterone (MESH:D000450), phenylalanine (MESH:D010649), acetonitrile (MESH:C032159), bile acid (MESH:D001647), proline (MESH:D011392), linoleic acid (MESH:D019787), methanol (MESH:D000432), adenosine (MESH:D000241), pyruvate (MESH:D019289), 1-palmitoyl-sn-glycero-3-phosphocholine (-), betaine (MESH:D001622), glycerophospholipid (MESH:D020404), ascorbic acid (MESH:D001205), sphingolipid (MESH:D013107), valine (MESH:D014633), phospholipid (MESH:D010743), water (MESH:D014867), leucine (MESH:D007930), ammonium hydroxide (MESH:D064753), iodine (MESH:D007455), guanidoacetic acid (MESH:C004946), endocannabinoid (MESH:D063388), lipid (MESH:D008055), arachidonic acid (MESH:D016718), ammonium acetate (MESH:C018824)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Figures

8 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12944297/full.md

---
Source: https://tomesphere.com/paper/PMC12944297