Thyroid cancer (TC) is the most common endocrine malignancy, with well-differentiated thyroid carcinomas (DTCs)-papillary (PTC) and follicular (FTC)-comprising the majority of cases. While DTCs generally have favorable prognoses, a subset progresses to poorly differentiated or anaplastic thyroid carcinoma (ATC), which is highly aggressive. Tumor classification is based on histopathology, invasiveness, and molecular characteristics, with new entities like thyroid tumors of uncertain malignant potential (TT-UMP) and non-invasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) refining diagnostic criteria.
Current standard treatments include surgical resection, radioactive iodine therapy, and thyroid hormone replacement. However, some patients develop radioiodine-refractory disease with an increased risk of recurrence and progression. Molecular alterations in the MAPK and PI3K pathways play critical roles in thyroid tumorigenesis, influencing therapeutic response and prognosis. Identifying novel biomarkers for early detection and risk stratification is crucial. Emerging evidence highlights the role of microRNAs (miRNAs) in thyroid cancer progression, functioning as oncogenes or tumor suppressors.
This retrospective case-control study aims to identify novel molecular markers linked to thyroid cancer aggressiveness. Archived formalin-fixed paraffin-embedded (FFPE) tissue and blood samples will be analyzed from patients with varying degrees of PTC and FTC invasiveness. Control samples will be histologically normal thyroid tissue from the same patients.
Next Generation Sequencing (NGS), including RNA-seq and miRNA-seq, will be employed to detect differentially expressed RNA molecules. Validation will be performed using Real-Time PCR in an independent cohort. High-throughput genomic sequencing (Illumina TruSight Oncology 500) will assess mutations, copy number variations, and tumor mutation burden to correlate genetic alterations with malignancy. Variants will be prioritized based on frequency differences in tumor vs. non-tumor populations and functional relevance.
The study will enroll patients with follicular cell-derived thyroid carcinoma. A power analysis indicates that 80 subjects provide \>80% statistical power for biomarker identification. Descriptive statistics, parametric/non-parametric tests, and machine learning approaches will analyze transcriptomic and genomic data. Receiver operating characteristic (ROC) curves will assess diagnostic biomarker accuracy, while logistic regression will model associations between molecular alterations and disease severity.
This study aims to uncover molecular mechanisms driving thyroid cancer progression and identify biomarkers for improved risk stratification, early diagnosis, and potential therapeutic targeting. Findings may enhance personalized treatment approaches in thyroid oncology.
Patients of either sex aged > 18 years with thyroid cancer of follicular origin.
Patients who do not fit the inclusion criteria.
The study population consists of patients with thyroid carcinoma originating from the follicular thyroid cell at varying degrees of malignancy. For such a retrospective population, considering the study's primary objective of identifying potential tissue biomarkers of pathology, an a posteriori power analysis was performed, considering a paired statistical design of the case-control type, where the control group is afferent to the healthy counterpart of the tissues available in the archives of the same thyroid carcinoma patients. For the subgroup analyses associated with the different degrees of invasiveness/aggressiveness, the statistical power associated with the possible clinical stratifications will be assessed before proceeding with the statistical analyses
Cohort
Cross-sectional
Not listed
Identification of novel molecular biomarkers associated with the progression and aggressiveness of follicular-derived thyroid carcinomas
RNA-seq and miRNA-seq on serum samples
Determine genetic alterations hat may contribute to disease progression
Identification of mutations, copy number variations, and tumor mutation burden)
Develop of predictive models for improved risk stratification and prognosis
Application of machine learning approach to integrate transcriptomic and genomic data
Click on the lead sponsor to view all of their trials.
IRCCS SYNLAB SDN
Lead sponsor
University Federico II of Naples, Department of Clinical and Surgical Medicine
Collaborator
