Clinical Thyroidology® for the Public

Summaries for the Public from recent articles in Clinical Thyroidology
Table of Contents | PDF File for Saving and Printing

THYROID CANCER
Predictors of survival in thyroid cancer with distant metastases

12

Instagram Youtube LinkedIn Facebook X

 

BACKGROUND
Thyroid cancer is common. Fortunately, patients with thyroid cancer have an excellent prognosis overall. Indeed, while <10% of patients have thyroid cancer that spreads outside outside of the neck (distant metastasis), some of these patients will have slow progression and remain stable for an extended time. However, there are others who have more aggressive cancers resulting in significant variability in survival among patients with thyroid cancer with distant metastases. This study was done to identify the role of specific clinical and pathologic factors that determine disease-specific survival and overall survival.

THE FULL ARTICLE TITLE
 Chen DW, et al. Survival prognostication in patients with differentiated thyroid cancer and distant metastases: a SEER population-based study. Thyroid 2024;34(7):837- 845; doi: 10.1089/thy.2023.0709. PMID: 38757633.

SUMMARY OF THE STUDY
Information on adult patients with thyroid and distant metastases was obtained from the Surveillance, Epidemiology, and End Results Program (SEER)-17 cancer registry (2010–2019). Variables included were age at the time of thyroid cancer diagnosis, sex, race and ethnicity, pathologic information (such as histology, cancer size, and site of distant metastases), and treatment information (such as surgery type, radioactive iodine [RAI] and chemotherapy use). Disease-specific survival (DSS) was defined as the time from thyroid cancer diagnosis to death from thyroid cancer, while overall survival (OS) was defined as the time from thyroid cancer diagnosis to death from any cause.

A total of 2411 patients were followed for an average of 62 months. The most common sites of distant metastases were the lungs (33.7%) and bone (18.9%). Overall 84.1% of patients underwent total thyroidectomy and 58.2% received RAI treatment. A total of 558 patients (23.1%) died from thyroid cancer, and 757 (31.4%) from all causes. Older age (>57), larger primary cancer size (>4 cm), and presence of lung metastases were associated with worse DSS and OS. DSS was worst in a group including patients ≥83 year, with cancers > 4 cm, or patients with cancers >4 cm and lung metastasis. This group had a 5-year survival rate of 41% (95% CI, 34–48%). OS was worst in group including patients ≥73 years of age with cancers >4 cm, or patients 58 to 72 years of age, with lung metastases, and cancers >4 cm. This group had a 5-year survival rate of 31% (95% CI, 26–37%).

WHAT ARE THE IMPLICATIONS OF THIS STUDY?
This study confirms the importance of patient age, presence of lung metastases at presentation and cancer size in predicting survival, although it isn’t perfect as there is much overlap between groups. It is mostly helpful in considering subsequent treatment after surgery and radioactive iodine. For example, in situations where initiation of chemotherapy is considered, it can help to make decisions about whether to use more aggressive treatment in older patients, while considering postponing lifelong systemic treatments in younger patients who fall into prognostic groups with better survival.

— Marjorie Safran, MD

ABBREVIATIONS & DEFINITIONS

Cancer Metastasis: spread of the cancer from the initial organ where it developed to other organs, such as the lungs and bone.

Disease-Specific Survival (DSS): The percentage of people in a study or treatment group who have not died from a specific disease in a defined period of time.

SEER: Surveillance, Epidemiology and End Results program, a nation-wide anonymous cancer registry generated by the National Cancer Institute that contains information on 26% of the United States population. Website: http://seer.cancer.gov/

Survival Trees: is a machine learning algorithm leverages the structure of a decision tree to make predictions about survival times.

Random Survival Forest: is a machine learning algorithm that creates an ensemble of multiple decision trees to reach a singular, more accurate prediction or result.

Table of Contents | PDF File for Saving and Printing