Predictive Factors for RAI-Refractory Disease and Short Overall Survival in PDTC
David Kersting, Robert Seifert, Lukas Kessler, Ken Herrmann, Sarah Theurer, Tim Brandenburg, Henning Dralle, Frank Weber, Lale Umutlu, Dagmar Führer‐Sakel, Rainer Görges, Christoph Rischpler, Manuel Weber
Abstract
Background: The clinical phenotype of poorly differentiated thyroid cancer (PDTC) can vary substantially. We aim to evaluate risk factors for radioiodine refractory (RAI-R) disease and reduced overall survival (OS). Methods: We retrospectively screened our institutional database for PDTC patients. For the assessment of RAI-R disease, we included patients who underwent dual imaging with 18F-FDG-PET and 124I-PET/131I scintigraphy that met the internal standard of care. We tested primary size, extrathyroidal extension (ETE), and age >55 years as risk factors for RAI-R disease at initial diagnosis and during the disease course using uni- and multivariate analyses. We tested metabolic tumor volume (MTV), total lesion glycolysis (TLG) on 18F-FDG-PET, and the progression of stimulated thyroglobulin within 4–6 months of initial radioiodine therapy as prognostic markers for OS. Results: Size of primary >40 mm and ETE were significant predictors of RAI-R disease in the course of disease in univariate (81% vs. 27%, p = 0.001; 89% vs. 33%, p < 0.001) and multivariate analyses. Primary tumor size was an excellent predictor of RAI-R disease (AUC = 0.90). TLG/MTV > upper quartile and early thyroglobulin progression were significantly associated with shorter median OS (29.0 months vs. 56.9 months, p < 0.05; 57.8 months vs. not reached p < 0.005, respectively). Discussion: PDTC patients, especially those with additional risk factors, should be assessed for RAI-R disease at initial diagnosis and in the course of disease, allowing for early implementation of multimodal treatment. Primary tumor size >40 mm, ETE, and age >55 are significant risk factors for RAI-R disease. High MTV/TLG is a significant risk factor for premature death and can help identify patients requiring intervention.