Background/Objectives: Anaplastic thyroid cancer (ATC) is an aggressive thyroid cancer subtype with a poor prognosis. The Controlling Nutritional Status (CONUT) score, reflecting both immune and nutritional status, is a prognostic marker in several malignancies; however, its utility in ATC has not been established. We aimed to evaluate the predictive value of the pretreatment CONUT score in ATC and compare its prognostic utility with that of other nutritional indices, including the Prognostic Nutritional Index (PNI) and Geriatric Nutritional Risk Index (GNRI). Methods: We retrospectively reviewed clinical characteristics, laboratory parameters, and survival outcomes of 156 patients with ATC at our institution between January 2004 and May 2024. Based on survival analysis, patients were categorized into low- and high-risk groups based on each nutritional index (CONUT score, PNI, GNRI) using optimal cut-off values. One-year survival differences were evaluated using Kaplan–Meier curves and log-rank test. Independent predictors of 1-year mortality were identified using multivariable Cox proportional hazards regression. Results: Optimal thresholds were 3, 42, and 102 for the CONUT score, PNI, and GNRI, respectively. Patients with CONUT scores ≥ 3 exhibited significantly higher 1-year mortality, compared with those with scores < 3. Multivariable analysis revealed that CONUT score ≥ 3, PNI ≤ 42, and GNRI ≤ 102 were independently associated with increased 1-year mortality risk. Incorporation of CONUT score ≥ 3 into the baseline prediction model significantly enhanced its discriminatory performance. Conclusions: These findings underscore the prognostic value of pretreatment immuno-nutritional assessment and support the integration of the CONUT score into early risk stratification strategies for patients with ATC.

Background/Objectives: The TOETVA and TORT have been successfully refined in recent years, with few complications and excellent results. In this study, we compared TORT and TOETVA in patients with papillary thyroid cancer. Methods: From September 2016 to January 2022, we retrospectively compared patient characteristics and clinical features from 100 TORT cases and 300 TOETVA cases, as well as their respective learning curves. Results: Median operation time of TORT was 80.4 min, similar to TOETVA (81.4 min, p = 0.719). Median docking time of TORT was 4.51 min and median console time was 27.72 min. The length of stay of TORT and TOETVA was 2.29 and 2.54 (p = 0.002). The median number of retrieved central lymph nodes for TOETVA and TORT were 2.89 and 2.99 (p = 0.746). Discussion: There was no significant difference with clinical features between TORT and TOETVA.

Kim NK, et al. J Endocr Surg. 2025 Jun;25(2):66-77. https://doi.org/10.16956/jes.2025.25.2.66

Background/Objectives: Anaplastic thyroid cancer (ATC) is an aggressive thyroid cancer subtype with a poor prognosis. The Controlling Nutritional Status (CONUT) score, reflecting both immune and nutritional status, is a prognostic marker in several malignancies; however, its utility in ATC has not been established. We aimed to evaluate the predictive value of the pretreatment CONUT score in ATC and compare its prognostic utility with that of other nutritional indices, including the Prognostic Nutritional Index (PNI) and Geriatric Nutritional Risk Index (GNRI). Methods: We retrospectively reviewed clinical characteristics, laboratory parameters, and survival outcomes of 156 patients with ATC at our institution between January 2004 and May 2024. Based on survival analysis, patients were categorized into low- and high-risk groups based on each nutritional index (CONUT score, PNI, GNRI) using optimal cut-off values. One-year survival differences were evaluated using Kaplan–Meier curves and log-rank test. Independent predictors of 1-year mortality were identified using multivariable Cox proportional hazards regression. Results: Optimal thresholds were 3, 42, and 102 for the CONUT score, PNI, and GNRI, respectively. Patients with CONUT scores ≥3 exhibited significantly higher 1-year mortality, compared with those with scores <3. Multivariable analysis revealed that CONUT score ≥3, PNI ≤42, and GNRI ≤102 were independently associated with increased 1-year mortality risk. Incorporation of CONUT score ≥3 into the baseline prediction model significantly enhanced its discriminatory performance. Conclusions: These findings underscore the prognostic value of pretreatment immuno-nutritional assessment and support the integration of the CONUT score into early risk stratification strategies for patients with ATC.

Introduction: Three‐dimensional (3D) organoids derived from patient tissues have emerged as promising preclinical models that preserve tumor architecture and phenotypic heterogeneity better than conventional two‐dimensional cultures. This study aimed to establish patient‐derived 3D organoid models for anaplastic thyroid carcinoma (ATC) and clinically radioiodine‐refractory papillary thyroid carcinoma (RAIR‐PTC) and evaluate whether they retained the histopathologic and immunophenotypic features of their source tumors. Methods: Tumor tissues from patients with ATC and RAIR‐PTC were collected, dissociated, and embedded in Matrigel for 3D culture. Organoid growth was monitored during serial passaging. Histopathologic, immunophenotypic, and targeted molecular analyses were performed. Results: The TCO001 sample, derived from ATC, showed substantial growth for more than five months, up to Passage 11, whereas the TCO002 sample, derived from RAIR‐PTC, exhibited limited proliferation and could not be maintained beyond Passage 3 after four months. These differences in culture performance were further supported by Ki‐67 and p53 staining, which demonstrated markedly higher proliferative activity and a more aberrant p53 expression pattern in TCO001 than in TCO002. Histopathologic and immunophenotypic analyses showed a dedifferentiated, ATC‐like phenotype in TCO001 and a more differentiated thyroid carcinoma‐like phenotype in TCO002. Targeted molecular analysis further supported the biological differences between the two lines: TCO001 harbored a BRAF mutation and a TERT promoter alteration, whereas TCO002 was wild type for both genes, consistent with the matched primary tumor tissue. Discussion: This study provides preliminary proof of concept that patient‐derived organoid technology can be extended to aggressive thyroid cancer phenotypes and preserve subtype‐relevant differences in differentiation state. Although preliminary, these findings support the value of thyroid cancer organoids as patient‐relevant preclinical models and provide a basis for future molecular, functional, and therapeutic studies in aggressive thyroid cancer.