This article is more than 5 years old. Information may no longer be current.
Risk model may predict thyroid cancer in childhood cancer survivors
Click Here to Manage Email Alerts
A model that accounts for self-reported risk factors, as well as basic radiation and chemotherapy information, predicted thyroid cancer risk in a cohort of childhood cancer survivors.
Researchers examined data from the Childhood Cancer Survivor Study and two nested case-control studies to develop three absolute risk models for second primary thyroid cancer to assist with long-term clinical monitoring of childhood cancer survivors. The studies included data on cancer survivors who were diagnosed between 1936 and 1986.
The first model (M1) included self-reported risk factors along with birth year, initial cancer type, age at diagnosis, sex and past thyroid nodule diagnosis. The second model (M2) added yes/no radiation information, particularly about radiation to the neck, and yes/no data for an alkylating agent. These data were taken from medical records. The third model (M3) refined the second by incorporating reconstructed radiation absorbed dose to the thyroid.
The investigators used an independent cohort of childhood cancer survivors in France to validate the models.
The strongest risk factor was past thyroid nodule (M1 RR=10.8; M2 RR=6.8; M3 RR=8.2).
The 20-year absolute risk predictions for a second primary thyroid cancer ranged from 0.04% to 7.4% for M2 in the validation cohort.
Expected events agreed with observed events for each model, an indicator of good calibration, according to the researchers.
The investigators also reported good discriminatory ability in each model. The area under the receiver operating characteristics curve (AUC) for the first model was 0.71 (95% CI, 0.64-0.77). The AUC for the second model was 0.8 (95% CI, 0.73-0.86), and the AUC for the third model was 0.75 (95% CI, 0.69-0.82).
The researchers concluded that the second model may be useful for monitoring thyroid cancer risk in childhood cancer survivors.
“These models will need to be updated as more outcome data become available for survivors in older adulthood,” the researchers wrote. “It will also be important to determine how changes in treatment modalities after the 1990s might have an impact on the applicability of these models to later cohorts of childhood cancer survivors.”
Disclosure: The researchers report no relevant financial disclosures.
Click Here to Manage Email Alerts