7642 A Mixed Model of Clinical Characteristics and Ultrasound ACR-TIRADS Predicts Malignancy in Thyroid Micro Lesions

Abstract Disclosure: N. Angelopoulos: None. I. Iakovou: None. D. Goulis: None. R. Paparodis: None. S. Livadas: None. A. Boniakos: None. J.C. Jaume: None. Background: The frequency of papillary thyroid carcinoma (PTC) has increased, mainly due to the detection of small tumors incidentally found during unrelated imaging. While most of these small cancers have a non-aggressive course, some can be more biologically hostile, with potential for metastasis. Current guidelines discourage routine biopsy for nodules ≤10 mm, but the challenge is identifying higher-risk microcarcinomas. This study aims to identify ultrasound and clinical characteristics indicating malignancy in small thyroid lesions. Patients and methods: 151 patients with accidentally discovered solid nodules in thyroid ultrasound were enrolled in the study. All patients were stratified according to both ACR and EU-TIRADS systems. Two qualified endocrinologists performed ACR-TIRADS scoring and accordingly, all patients were stratified to the five ACR categories TR1-5 Strain elastography was also performed in all nodules and the elastography ratio (E2/E1) was calculated (E2 elastography of the nodule, E1 elastography of healthy thyroid gland). FNA was performed in nodules classified as EU-TIRADS 5 Results: 41 patients were classified as EU-TIRADS 5. The rest of the patients were classified as 69 EU-TIRADS 4, 40 EU-TIRADS 3, and one EU-TIRADS 2. According to FNA results, 11 patients had Bethesda II cytology (Group A), 2 Bethesda III, one Bethesda IV, 10 Bethesda V, and 17 Bethesda VI (V and VI comprised Group B). FNAs indicative of malignancy (Group B) were more frequent in patients with autoimmune thyroiditis (Chi-squared 5,833, df(1), p=0.015). In Group B, both ACR-scoring points and ACR-TIRADS categories were significantly higher than those detected in patients in Group A. Logistic regression analysis revealed a significant prognostic role of ACR-scoring points in discriminating suspicious FNA results for thyroid cancer. ROC analysis (AUC:0.993 with Youden Index:0.926, associated criterion >6, Sensitivity 92.59 % and Specificity 100 %). To analyze the predictive power of the clinical parameters in the presence of thyroid malignancy, age, BMI, gender, autoimmune disease, therapy, nodule diameter, and elastography ratio were examined in a “clinical model”. The overall prognostic value of the model was significant (p=0.0029, AUC:0.744) with a sensitivity of 77.78% and specificity of 72.73%, (Youden Index: 0.505, associated criterion >0.6). Finally, a “mixed model”, combining the clinical characteristics and the ACR scoring points, was generated for methodological reasons. The model was overall significant (p<0.001) with AUC:1 (95% Confidence interval: 0.907 to 1), Youden Index:1, sensitivity, and specificity 100%. Conclusion: ACR- TIRADS points > 6 may be a strong indicator of thyroid malignancy and the “mixed model” is an excellent predictor of thyroid malignancy, but its usefulness in everyday clinical practice may be cumbersome. Presentation: 6/1/2024

Authors

Institutions

Publication Details

Journal
Journal of the Endocrine Society
Published
2024-10-01
DOI
https://doi.org/10.1210/jendso/bvae163.1854
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

7642 A Mixed Model of Clinical Characteristics and Ultrasound ACR-TIRADS Predicts Malignancy in Thyroid Micro Lesions

Juan Carlos Jaume, Dimitrios G. Goulis, Rodis D. Paparodis, Ιoannis Iakovou et al.
Journal of the Endocrine Society
Radiomics and Machine Learning in Medical Imaging
article

7642 A Mixed Model of Clinical Characteristics and Ultrasound ACR-TIRADS Predicts Malignancy in Thyroid Micro Lesions

Juan Carlos Jaume, Dimitrios G. Goulis, Rodis D. Paparodis, Ιoannis Iakovou, Anastasios Boniakos, Sarantis Livadas, Nikolaos Angelopoulos
article en

Abstract

Abstract Disclosure: N. Angelopoulos: None. I. Iakovou: None. D. Goulis: None. R. Paparodis: None. S. Livadas: None. A. Boniakos: None. J.C. Jaume: None. Background: The frequency of papillary thyroid carcinoma (PTC) has increased, mainly due to the detection of small tumors incidentally found during unrelated imaging. While most of these small cancers have a non-aggressive course, some can be more biologically hostile, with potential for metastasis. Current guidelines discourage routine biopsy for nodules ≤10 mm, but the challenge is identifying higher-risk microcarcinomas. This study aims to identify ultrasound and clinical characteristics indicating malignancy in small thyroid lesions. Patients and methods: 151 patients with accidentally discovered solid nodules in thyroid ultrasound were enrolled in the study. All patients were stratified according to both ACR and EU-TIRADS systems. Two qualified endocrinologists performed ACR-TIRADS scoring and accordingly, all patients were stratified to the five ACR categories TR1-5 Strain elastography was also performed in all nodules and the elastography ratio (E2/E1) was calculated (E2 elastography of the nodule, E1 elastography of healthy thyroid gland). FNA was performed in nodules classified as EU-TIRADS 5 Results: 41 patients were classified as EU-TIRADS 5. The rest of the patients were classified as 69 EU-TIRADS 4, 40 EU-TIRADS 3, and one EU-TIRADS 2. According to FNA results, 11 patients had Bethesda II cytology (Group A), 2 Bethesda III, one Bethesda IV, 10 Bethesda V, and 17 Bethesda VI (V and VI comprised Group B). FNAs indicative of malignancy (Group B) were more frequent in patients with autoimmune thyroiditis (Chi-squared 5,833, df(1), p=0.015). In Group B, both ACR-scoring points and ACR-TIRADS categories were significantly higher than those detected in patients in Group A. Logistic regression analysis revealed a significant prognostic role of ACR-scoring points in discriminating suspicious FNA results for thyroid cancer. ROC analysis (AUC:0.993 with Youden Index:0.926, associated criterion >6, Sensitivity 92.59 % and Specificity 100 %). To analyze the predictive power of the clinical parameters in the presence of thyroid malignancy, age, BMI, gender, autoimmune disease, therapy, nodule diameter, and elastography ratio were examined in a “clinical model”. The overall prognostic value of the model was significant (p=0.0029, AUC:0.744) with a sensitivity of 77.78% and specificity of 72.73%, (Youden Index: 0.505, associated criterion >0.6). Finally, a “mixed model”, combining the clinical characteristics and the ACR scoring points, was generated for methodological reasons. The model was overall significant (p<0.001) with AUC:1 (95% Confidence interval: 0.907 to 1), Youden Index:1, sensitivity, and specificity 100%. Conclusion: ACR- TIRADS points > 6 may be a strong indicator of thyroid malignancy and the “mixed model” is an excellent predictor of thyroid malignancy, but its usefulness in everyday clinical practice may be cumbersome. Presentation: 6/1/2024

Journal of the Endocrine SocietyVol. 8(Supplement_1)
Aristotle University of Thessaloniki (GR), Edward Hines, Jr. VA Hospital (US), Loyola Medicine (US)
Good health and well-being
Openalex Percentile: Top 71%
Radiomics and Machine Learning in Medical Imaging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.