Exploratory Statistical Analyses of Clinical and Biochemical Factors for Differentiated Thyroid Cancer from a...

Thyroid cancer (TC) is among the most common endocrine malignancies, with incidence rates increasing worldwide. However, careful inferential analysis based on refined data is needed to provide a sharper clinical and epidemiological description of this serious condition in a biologically and technologically evolving society. This study presents an exploratory statistical analysis of data from 1470 patients who underwent thyroid surgery for differentiated TC.

The analysis combines bivariate exploration of associations between variables with univariate and multivariate analyses stratified by histological subtype. We examined pathological characteristics (tumor location, nodal metastases, distant metastases, margin involvement, lymphovascular invasion, vascular invasion, and perineural invasion), clinical characteristics (clinical stage), biochemical markers (thyroglobulin, anti-thyroglobulin antibodies, and thyroid-stimulating hormone), and demographic variables (sex assigned at birth and age). In addition, exploratory multivariable models were used to investigate factors associated with lymph node metastasis and margin involvement in papillary microcarcinoma, the diffuse sclerosing variant, and the classical variant of papillary thyroid carcinoma.

Notably, moderate to high effect size correlations highlight the interdependence of invasive histopathological features in thyroid cancer and their collective link to adverse surgical outcomes and prognosis.

This study provides an analysis of associations between the variables and subtype-specific descriptive estimates, serving as a foundation for future work in tailoring personalized medicine.