This article is more than 5 years old. Information may no longer be current.
Nomogram predicted mortality in patients with thyroid cancer
A nomogram based on a competing risks model predicted the probability of mortality in patients with thyroid cancer, according to results of a population-based study.
Thyroid cancer often is associated with excellent prognosis. The OS rate for patients with thyroid cancer consistently approaches 90% to 95%, according to SEER data. Because of long-term survival, patients frequently die from other causes. Therefore, clinicians physicians should consider additional causes of death when thyroid cancer prognosis, according to background information in the study.
Researchers from the National Research Institute for Child Health and Development in Tokyo used the SEER database to evaluate data on 29,255 patients with thyroid cancer between 1988 and 2003. Investigators analyzed the associations of patient and tumor characteristics with patient death, and they used a proportional subdistribution hazard competing risks model to develop a nomogram to predict probability of death.
The median duration of follow-up until censoring or death was 85 months, and the median age of patients at follow-up was 44 years.
Study results showed the 5-year probabilities of death were 1.9% for thyroid cancer, 0.8% for other cancers and 1.7% for noncancer-related causes.
Ten-year probability of death rates were 3% for thyroid cancer, 2% for other cancers and 3.9% for noncancer-related causes.
“Patients showed a nearly twofold higher risk of dying from causes other than thyroid cancer,” the researchers wrote.
Five and 10-year probability of death increased with age (P<.001 for all outcomes). Tumor size, male sex, poorly differentiated carcinoma, regional and metastatic disease, and lympho node involvement all were associated with increased cumulative incidence of death from thyroid cancer, the researchers wrote.
“Radiotherapy was associated with a significantly higher cumulative incidence of death among patients who died of thyroid cancer but was associated with a significantly lower cumulative incidence of death among patients who died of other noncancer causes,” the researchers wrote.
The nomogram may be useful in determining targeted treatment options for patients, according to the investigators.
“Our nomogram may help clinicians identify individuals at a higher risk of thyroid cancer death and provides more individualized treatment planning,” they concluded. “Performance of the model was excellent. Thus, the nomogram should be considered as an accurate tool for clinical prognosis prediction.”
Perspective
Back to TopWhile the majority of thyroid cancers are associated with an excellent prognosis, there is a minority of patients with progressive and sometimes fatal disease. As there are more systemic agents available for the treatment of thyroid cancer, a common clinical conundrum pertains to when it is appropriate to initiate systemic therapy for an asymptomatic patient with slow-growing disease, especially in light of the toxicities associated with the currently available tyrosine kinase inhibitors. Thus, the ability to better define those patients at high risk for death from thyroid cancer, or at high risk from noncancer-related diagnoses, can be of utility when counseling patients with metastatic disease.
Yang and colleagues did a thorough analysis of a large cohort of patients from the SEER database. Reflecting the overall good prognosis of this disease, the 5-year probability of death from thyroid cancer was 1.9%. It is not surprising that known clinical risk factors for more aggressive disease — such as increasing age, tumor size, male sex, metastatic disease and poorly differentiated histology — also were associated with increased incidence of death in the authors’ analysis. Clinicians likely have already been considering many of these factors in their current discussions with patients regarding prognosis and treatment options.
There are some limitations with this study. First, the majority of patients in this series (>88%) had papillary thyroid cancer, while only 0.7% had medullary thyroid cancer. Thus, the applicability of this nomogram to the nonpapillary histologies is less clear. In addition, a majority of patients (98.6%) had locoregional disease, whereas only 1.4% had distant metastases. Finally, the significance of non-iodine avid disease, a known risk factor for more aggressive differentiated thyroid cancers, was not incorporated into the analysis. Consequently, it is not surprising that the overall 5- and 10-year risk for death in this series is very low.
In the future, it may be more appropriate to evaluate medullary and papillary thyroid cancers separately, given the differing overall biology of these two malignancies. In addition, having more data on the use and reliability of this nomogram in a higher-risk population with metastatic, non-iodine avid papillary thyroid cancer would greatly aid clinicians when deciding how to use this tool in their practice. Overall, the effort undertaken in this analysis provides a comprehensive, quantitative tool that is worthy of further discussion and study to aid us in the management of thyroid cancer patients.