I am actually not comfortable discussing this analysis,” said formal discussant Vincent Grégoire, MD, PhD, of the Department of Radiation Oncology, St-Luc University Hospital, Brussels, Belgium. “Instead of me you need a biostatistician. Nevertheless, I will try to address the study.”

“The study raises the question about what it means to use models to try to analyze data. On the positive side, the rank-preserving structural failure time analysis was prespecified. But the issue is the power of the analysis. We need to have confidence about the model used. There are other models that could have been used,” Dr. Grégoire continued.

Radioactive iodine–refractory differentiated thyroid cancer is a complex disease with strange behavior, and treatment may not be needed upfront, he noted.

“Not to decrease the interest in these findings, for me, I need to know more about the power and adequacy of the model used to estimate a survival benefit,” he concluded. ■

Disclosure: Dr. Grégoire reported no potential conflicts of interest.