Differentiated thyroid cancer (DTC) is the most common endocrine malignancy, with a high 5-year survival rate of approximately 98%. Despite advances in diagnosis and treatment, up to 20% of patients experience recurrence, adversely affecting their quality of life. Predictive models have been developed to assess recurrence risk and guide clinical decision-making, but these models often face limitations such as retrospective design, lack of diversity in study populations and absence of external validation. The primary aim is to externally validate existing predictive models for DTC recurrence using prospective data from a diverse Latin American cohort. The secondary aim is to explore opportunities for model recalibration to improve their performance in our population.

The CaTaLiNA study is a multicentre prospective observational study conducted across 10 hospitals in five Latin American countries, including Ecuador, Peru, Uruguay and Mexico. Patients aged 18 years or older receiving treatment for DTC, such as the first thyroid surgery, active surveillance or radiofrequency ablation will be included. Recruitment will occur from November 2023 to June 2025, with follow-up extending until June 2028. Data collection will include baseline clinical, surgical and histological characteristics, treatment details and follow-up outcomes. Statistical analysis will follow the Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis guidelines, using imputation strategies for missing data and evaluating calibration and discrimination of the prediction models. Calibration measures include the ratio of expected and observed events, calibration slope and calibration plot, while discrimination will be assessed using the C-index and area under the receiver operating characteristic curve.

This study protocol was approved by Comité de Ética de Investigación en Seres Humanos de la Universidad San Francisco de Quito USFQ ‘CEISH-USFQ’ APO-010–2023-CEIHS-USFQ Oficio No. 161-2023-CA-23030M-CEISH-USFQ. Results will be disseminated via peer-reviewed publications.