Palatine tonsil SUVmax on FDG PET-CT as a discriminator between benign and malignant tonsils in patients with and without head and neck squamous cell carcinoma of unknown primary

TY - JOUR

T1 - Palatine tonsil SUVmax on FDG PET-CT as a discriminator between benign and malignant tonsils in patients with and without head and neck squamous cell carcinoma of unknown primary

AU - Pencharz, D

AU - Dunn, J

AU - Connor, S

AU - Siddiqui, A

AU - Sriskandan, N

AU - Thavaraj, S

AU - Jeannon, J-P

AU - Oakley, R

AU - Lei, M

AU - Guerrero-Urbano, T

AU - Cook, G J

AU - Szyszko, T A

N1 - Crown Copyright © 2018. Published by Elsevier Ltd. All rights reserved.

PY - 2018/11/16

Y1 - 2018/11/16

N2 - AIM: To analyse the maximum standardised uptake value (SUVmax) ratio between tonsils in patients with and without tonsillar carcinoma to determine useful diagnostic thresholds.MATERIALS AND METHODS: Positron-emission tomography (PET)/computed tomography (CT) examinations of patients with suspected head and neck squamous cell carcinoma (SCC) and controls from April 2013 to September 2016 were reviewed retrospectively. Tonsillar SUVmax ratios (ipsilateral/contralateral for malignant tonsils, maximum/minimum for patients without [controls]) were calculated and used to construct a receiver operating characteristic (ROC) curve.RESULTS: Twenty-five patients had tonsillar carcinoma (mean SUVmax ratio of 2, range 0.89-5.4) and 86 patients acted as controls (mean SUVmax ratio of 1.1, range 1-1.5). Using the ROC, the most accurate SUVmax ratio for identifying malignancy was >1.2 (77% sensitivity, 86% specificity). A potentially more clinically useful SUVmax ratio is ≥1.6 with 62% sensitivity and 100% specificity.CONCLUSION: An SUVmax ratio between tonsils of ≥1.6 is highly suspicious for SCC and could be used to direct site of biopsy. Some malignant tonsils had normal FDG uptake; therefore, PET/CT should not be used to exclude tonsillar cancer. Minor asymmetrical uptake is frequently seen in non-malignant tonsils and does not necessarily require further investigation. Due to the single centre nature of this study and the recognised variation in SUV measurements between PET scans, other centres may need to develop their own cut-offs.

AB - AIM: To analyse the maximum standardised uptake value (SUVmax) ratio between tonsils in patients with and without tonsillar carcinoma to determine useful diagnostic thresholds.MATERIALS AND METHODS: Positron-emission tomography (PET)/computed tomography (CT) examinations of patients with suspected head and neck squamous cell carcinoma (SCC) and controls from April 2013 to September 2016 were reviewed retrospectively. Tonsillar SUVmax ratios (ipsilateral/contralateral for malignant tonsils, maximum/minimum for patients without [controls]) were calculated and used to construct a receiver operating characteristic (ROC) curve.RESULTS: Twenty-five patients had tonsillar carcinoma (mean SUVmax ratio of 2, range 0.89-5.4) and 86 patients acted as controls (mean SUVmax ratio of 1.1, range 1-1.5). Using the ROC, the most accurate SUVmax ratio for identifying malignancy was >1.2 (77% sensitivity, 86% specificity). A potentially more clinically useful SUVmax ratio is ≥1.6 with 62% sensitivity and 100% specificity.CONCLUSION: An SUVmax ratio between tonsils of ≥1.6 is highly suspicious for SCC and could be used to direct site of biopsy. Some malignant tonsils had normal FDG uptake; therefore, PET/CT should not be used to exclude tonsillar cancer. Minor asymmetrical uptake is frequently seen in non-malignant tonsils and does not necessarily require further investigation. Due to the single centre nature of this study and the recognised variation in SUV measurements between PET scans, other centres may need to develop their own cut-offs.

U2 - 10.1016/j.crad.2018.10.007

DO - 10.1016/j.crad.2018.10.007

M3 - Article

C2 - 30454841

SN - 0009-9260

JO - Clinical Radiology

JF - Clinical Radiology

ER -