Objective: To evaluate the accuracy of PET/CT using (18)F-fluorodeoxyglucose (FDG) with IV contrast for suspected recurrent head and neck squamous cell carcinoma (HNSCC).
Methods: One hundred and seventy patients previously treated for HNSCC underwent PET/CT, consisting of non-contrast-enhanced and contrast-enhanced CT, to investigate suspected recurrence. Diagnostic performance of PET/contrast-enhanced CT (PET/ceCT), PET/non-contrast-enhanced CT (PET/ncCT) and contrast-enhanced CT (ceCT) for local or regional recurrence, distant metastasis, overall recurrence and second primary cancer was evaluated. The reference standard included histopathology, treatment change and imaging follow-up.
Results: The patient-based areas under the receiver operating characteristic curves (AUC) for ceCT, PET/ncCT and PET/ceCT were 0.82, 0.96 and 0.98 for local recurrence, 0.73, 0.86 and 0.86 for regional recurrence, 0.86, 0.91 and 0.92 for distant metastasis, 0.72, 0.86 and 0.87 for overall recurrence, and 0.86, 0.89 and 0.91 for a second primary cancer. Both PET/ceCT and PET/ncCT statistically showed larger AUC than ceCT for recurrence, and the difference between PET/ceCT and PET/ncCT for local recurrence reached a significant level (p = 0.039). The accuracy of PET/ceCT for diagnosing overall recurrence was high, irrespective of the time interval after the last treatment (83.3-94.1 %).
Conclusion: FDG-PET/CT was a more accurate HNSCC restaging tool than ceCT. The added value of ceCT at FDG-PET/CT is minimal.
Key points: • FDG-PET/CT is a more accurate post-treatment surveillance tool than ceCT for HNSCC. • FDG-PET/ceCT was useful, irrespective of the time interval after the last treatment. • FDG-PET/ceCT showed high negative predictive value and limited positive predictive value. • The added value of ceCT at FDG-PET/CT is minimal and likely not clinically relevant.
Keywords: 18F-FDG; Enhanced CT; Head and neck squamous cell carcinoma (HNSCC); Positron emission tomography/computed tomography (PET/CT); Recurrence.