Oral Cavity Cancer Surgical Pathologic Factors Influence Accuracy of 3-Month Postradiation Therapy FDG Positron Emission Tomography/Computed Tomography

Abstract

Recurrence monitoring for oral cavity squamous cell carcinoma (SCC) following definitive surgery and adjuvant (chemo) radiation therapy (RT) with F-18 Fluorodeoxyglucose (FDG) PET/CT is not well established. This study evaluates the accuracy of 3 month post-RT PET/CT in oral cavity cancer patients treated with surgery and adjuvant (chemo)RT, with an additional objective evaluating patient tumor characteristics which influence PET/CT accuracy. Oral cavity SCC patients receiving definitive surgical resection, adjuvant (chemo)RT, and 3 month post-RT PET/CT were retrospectively examined between 2005 and 2015. PET/CTs were classified as positive or negative using the nuclear medicine interpretation, with equivocal scans assigned positive if the ordering clinician performed biopsy, planned for closer follow-up, or ordered additional imaging. True or false results were based on clinical follow-up within one year of PET/CT. Imaging accuracy was calculated separately for locoregional and distant disease. Further analysis evaluated associations between 3 month post-RT PET/CT accuracy, and disease sub-site and pathologic risk factors including positive lymph nodes, extracapsular extension, perineural invasion (PNI), lymphovascular invasion, bone invasion, positive margin, and pathologic grade. Fifty-six patients met inclusion criteria; with 14 (25%) receiving concurrent chemoRT. Median time from RT end to PET/CT was 101 days. Twenty-three (41%) had a positive scheduled surveillance 3 month-post RT PET/CT: 14 (25%) had PET/CTs that were positive in the head and neck, 3 of which also had distant positive findings, and 9 were positive distantly alone. PET/CT revealed 14 (25%) patients with true recurrences: 6 locoregional and 9 distant, with 1 patient having both locoregional and distant recurrence uncovered by PET/CT. For locoregional recurrence, PET/CT sensitivity, specificity, positive predictive value (PPV), and negative predictive values (NPV) were 67%, 82%, 43%, and 92%, respectively. For distant recurrence, the respective values were 100%, 93%, 75%, and 100%. PNI was associated with greater 3 month post-RT PET/CT specificity (P = 0.0182). Cancer within the tongue versus other oral cavity sites was associated with greater PET/CT PPV (P = 0.0047). True recurrence was revealed by PET/CT in 25% of oral cavity cancer patients. PET/CT accuracy varied based on several tumor characteristics, including PNI and tumor location in the tongue. This study presents post-therapy PET/CT accuracy data on a similarly treated patient population, and may help to guide further treatment management decisions when using PET/CT in the post-therapy setting.