• D P Noij, R M Martens, T Koopman, O S Hoekstra, E F I Comans, B Zwezerijnen, R de Bree, P de Graaf, and J A Castelijns.
• Department of Radiology and Nuclear Medicine, VU University Medical Center, Amsterdam, the Netherlands. Electronic address: d.noij@vumc.nl.
• Clin Oncol (R Coll Radiol). 2018 Dec 1; 30 (12): 780-792.

AimsOur purpose was to assess the diagnostic accuracy and prognostic value of diffusion-weighted imaging (DWI) and 18F-fluorodeoxyglucose positron emission tomography combined with computed tomography (18F-FDG-PET/CT) carried out 3-6 months after (chemo)radiotherapy in head and neck squamous cell carcinoma.Materials And MethodsFor this retrospective cohort study we included 82 patients with advanced-stage head and neck squamous cell carcinoma treated between 2012 and 2015. Primary tumours and lymph nodes were assessed separately. DWI was analysed qualitatively and quantitatively. 18F-FDG-PET/CT was evaluated using the Hopkins criteria. Dichotomous qualitative analysis was carried out for both modalities. Cox regression analysis was used for univariate analysis of recurrence-free survival (RFS). Significant univariate parameters were included in multivariate analysis.ResultsIn 12 patients, locoregional recurrence occurred. With all imaging strategies, either single-modality or multi-modality, a high negative predictive value (NPV) was achieved (94.3-100%). In response evaluation of the primary site, the preferred strategy is 18F-FDG-PET/CT only, which resulted in a sensitivity of 85.7%, specificity of 86.5%, positive predictive value (PPV) of 37.5% and NPV of 98.5%. For response evaluation of the neck, the best results were obtained with a sequential approach only including the second modality in positive reads of the first modality. It did not matter which modality was assessed first. This strategy for lymph node assessment resulted in a sensitivity, specificity, PPV and NPV of 83.3%, 95.6%, 62.5%, and 98.5%, respectively. After correction for received treatment and human papillomavirus status, primary tumour (P = 0.009) or lymph node (P < 0.001) Hopkins score ≥4 on 18F-FDG-PET/CT remained significant predictors of RFS.ConclusionFor response evaluation of the primary tumour 18F-FDG-PET/CT only is the preferred strategy, whereas for the neck a sequential approach including both DWI and 18F-FDG-PET/CT resulted in the best diagnostic accuracy for follow-up after (chemo)radiotherapy. Qualitative analysis of 18F-FDG-PET/CT is a stronger predictor of RFS than DWI analysis.Copyright © 2018 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

31 keywords...

hide…