Early Prediction of Response to Radiation Therapy Using Inter-Fractional 18F-FDG PET/CT in Esophageal Cancer

Abstract

To identify whether early metabolic responses (based on 18F-FDG-PET/CT during radiotherapy [RT]) could predict treatment outcomes in esophageal cancer for implementation of FDG-PET-based adaptive treatment algorithm. A retrospective review of 21 patients with esophageal cancer who had a pre-treatment PET (PET1) and a mid-RT PET (PET2) after the 11 fractions of radiotherapy (Median 23.1 Gy, 2.1 Gy per fraction) between November 2015 and June 2017 was undertaken. Region of interest for each SUV calculation was delineated on the PET1 and PET2 scans using PET Edge (semi-automatic gradient-based delineation method). We calculated PET parameters including maximum and mean standardized uptake value (SUVmax and SUVmean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) between PET1 and PET2. At a median follow-up of 13 months (range, 1.6‒23.6), the 1-year overall survival and progression free survival was 79.0% and 34.4%. Among all patients, 4 patients developed loco-regional recurrences and 8 patients showed distant metastases. Overall 1-year loco-regional recurrence free rate was 76.9% and distant metastasis free rate was 60.6%. The relative changes of MTV (i.e., ΔMTV) were significantly associated with loco-regional recurrence (HR 0.98, 95% CI 0.96–1.00; p = 0.03). Conversely, relative changes of SUVmean (i.e., ΔSUVmean) were associated with risk of distant recurrence (HR 0.90, 95% CI 0.82–0.99; p = 0.02). The ideal cut-off value to distinguish responders from non-responders for ΔMTV and ΔSUVmean was statistically determined at ratio of 1.14 (MTV PET2/MTV PET1) and -35% (i.e., a 35% decrease). At the threshold 1.14, the MTV yielded a sensitivity of 60%, specificity of 94%, and accuracy of 86% for predicting a loco-regional recurrence. In addition, the ΔSUVmean yielded a sensitivity of 67%, specificity of 83%, and accuracy of 76% for prediction a distant metastasis with the threshold of 35% decrease. This study demonstrates that changes in tumor metabolic activity during radiotherapy based on FDG-PET could be used as a surrogate marker to predict RT response, recurrence, and prognosis in esophageal cancer. Furthermore, PET-based adaptive treatment algorithm to potentially allow a change of treatment plan in non-responding patients might even be applicable.