NEW ORLEANS – The 2015 American Thyroid Association risk stratification system for patients with differentiated thyroid cancer performed well in a real-world cohort with a high proportion of high-risk patients, according to a study presented at the annual meeting of the Endocrine Society.

“The 2015 ATA Risk Stratification System is an excellent predictor of both persisting disease and survival,” wrote Evert F.S. van Velsen, MD, and his colleagues at Erasmus Medical Center, Rotterdam, the Netherlands, in a poster accompanying the presentation.

Among a group of 236 patients with differentiated thyroid cancer (DTC), Dr. van Velsen and his coauthors looked at how the ATA high-risk criteria influenced patient response to therapy. By the end of the 14-year study period, initial gross extrathyroidal disease extension meant patients were much less likely to have an excellent response (odds ratio, 0.26; P less than .001), and much more likely to have persistent disease (OR, 2.57; P = .001).

Odds of having an excellent response were reduced by having high postoperative thyroglobulin levels (OR, 0.21; P less than .001), and persistent disease was more likely (OR, 2.39; P = .002).

Other high-risk criteria associated with significantly lower odds of excellent response included distant metastases (OR, 0.36), incomplete resection (OR, 0.51), and having follicular thyroid carcinoma (FTC) with extensive vascular invasion (OR, 0.27). All these risk factors also were associated with higher odds of persistent disease.

“Recurrence after no evidence of disease occurred in 14%” of the study population, said Dr. van Velsen and his coauthors, adding, “Clinicians should be aware of the relatively high recurrence risk, even after an excellent response to therapy.”

The study aimed to evaluate the 2015 ATA risk stratification system’s prognostic value in a population that included a relatively large proportion of high-risk DTC patients, to include many FTC patients. This work, they noted, augments previous assessments of the risk stratification system in lower-risk populations.

The authors noted that, in addition to predicting disease recurrence, the risk stratification system also worked as a predictor of disease-specific survival. Patients with structural incomplete response fared the worst, with a survival probability below 0.5 at 200 months on a Kaplan-Meier curve of disease-specific survival. Survival probability remained at 1.0 for patients with excellent response after first therapy and was intermediate for those with indeterminate response and biochemical incomplete response.

Overall mortality was higher in FTC patients. Over the study period, 31 of the 76 FTC patients (41%) died, compared with 39 of the PTC patients (24%; P = .010). In all, 28% of the FTC patients and 18% of the PTC patients died of thyroid cancer, but this difference didn’t reach statistical significance.

The retrospective study included adults with DTC meeting the 2015 ATA high-risk criteria who were diagnosed and/or treated at Erasmus Medical Center over a 13-year span ending in December 2015.

Overall, the investigators found 236 patients meeting inclusion criteria; 160 had papillary thyroid cancer (PTC), and the remaining 76 had FTC. The latter group were significantly older at baseline than PTC patients (64 versus 53 years), and were significantly less likely to undergo neck dissection (22% versus 55%).

In the full cohort, 96 patients (41%) had one high-risk factor, and an additional 74 (31%) had two risk factors. The remaining patients had three or more risk factors.

There was no between-group difference in the likelihood of receiving radioactive iodine treatment, but those with FTC had a lower cumulative radiation dose (195 versus 298 mCi; P less than .001).

More than half of patients (58%) had persistent disease after completing their first therapy. Of these, 51% had structural incomplete response and 7% had biochemical incomplete response. The response was indeterminate for about a quarter of the cohort, and the remaining 17% had an excellent initial response.

By the end of the study period, 55% of patients had persistent disease, and 51% had structural incomplete response (a more likely result for those with FTC than PTC). Just 4% had a biochemical incomplete response, and the response was indeterminate for 16%. Response was judged excellent for 29% of patients.

Dr. van Velsen and his coauthors reported that they had no relevant disclosures.

koakes@mdedge.com

SOURCE: van Velsen EFS et al. ENDO 2019, Abstract MON-549.

ATA risk stratification for DTC performs well in real-world cohort