ECE2022 Poster Presentations Thyroid (136 abstracts)
1ASST Sette Laghi, Endocrine Unit, Varese, Italy; 1 ASST Sette Laghi, Endocrine Unit, Varese, Italy; 3ASST Sette Laghi, Health Physics Service, Varese, Italy; 4University of Insubria, EPIMED Centre, Varese, Italy; 5University of Insubria, Medicine and Surgery, Varese, Italy
Aim of the study: recurrence or persistence of hyperthyroidism occur in 15-25% of cases after radioiodine therapy (RAI) in Graves disease (GD). Our study aimed to establish prognostic factors affecting RAI outcome and to investigate if a tailored dosimetric approach based on the application of the effective biological dose (BED) offered a higher chance of success.
Materials and methods: our cohort comprised 365 GD patients (280 women and 85 men; age 49 ±14 years) treated with RAI in the period 2001-2021. Patients were allocated into two groups: failure group in case of persistence of hyperthyroidism for more than six months post-RAI or hyperthyroidism relapse; success group in case persistent euthyroidism or hypothyroidism were achieved after treatment. A multivariate analysis was performed to construct a predictive model of success; a BED threshold value was derived using a ROC curve.
Results: success was achieved in 80% of cases. No significant differences emerged between the two groups when analyzing demographic data (age, sex), smoking habit, GD duration and ongoing medical therapies, presence/severity of orbitopathy, RAI uptake and the applied formula for dosimetry calculation. Negative predictive factors for success were higher thyroid volume, nodules (presence and volume) and disease severity at diagnosis, and a higher effective half-life of radioiodine. The success group had higher FT4 levels immediately before the caption curve (within/slightly above the upper limit of the normal range), a higher difference between the administered activity and the calculated one, a higher BED. Thyroid volume was the most predictive variable in the multivariate analysis, but no single factor was able to predict the outcome. A BED of 369 Gy emerged in the ROC curve as a cut-off value (AUC 0.60, P=0.01).
Conclusions: Our study suggests that the outcome of RAI is influenced by clinical, biochemical, ultrasound and dosimetric factors, which interact with each other in a complex way. It is a delicate balance, in which no factor, considering individually, can predict the treatment approach. A multidisciplinary approach is necessary, to better understand the different interactions and confounding factors.