Encrypted login | home

Program Information

Interim Esophageal Cancer Response Assessment Via 18FDG-PET Scanning During Radiation Therapy

no image available
K Higgins

K Higgins1*, Q Wu2 , B Perez2 , B Czito2 , M Palta2 , C Willett2 , S Das2 (1) Duke University Medical Physics Graduate Program, Durham, NC, (2) Duke University Medical Center, Durham, NC


TH-E-BRF-10 Thursday 1:00PM - 2:50PM Room: Ballroom F

Purpose: Local failure occurs in a large proportion of esophageal cancer patients treated with chemoradiation. The treatment strategy for non-responders could potentially be modified if they are identified during therapy. This work investigates the utility of an interim 18FDG-PET scan acquired during the course of therapy as a predictor of pathological response post-therapy.

Methods: Fifteen patients underwent 18FDG-PET scanning prior to radiation therapy (RT) and once during RT, after delivery of ~32 Gy. The physician-contoured GTV on the planning CT scan was used to automatically segment a PET-based GTV on the pre-RT PET (GTV-pre-PET) as the volume with >40% of the maximum GTV PET SUV value. The pre- and intra-RT CTs were deformably registered to each other to transfer the GTV-pre-PET to the intra-RT PET (GTV-intra-PET). The fractional decrease in the maximum SUV, mean SUV and the SUV to the highest intensity 10% - 90% volumes from GTV-pre-PET to GTV-intra-PET were compared to pathological response assessed at the time of post-RT surgery.

Results: Based on post-treatment pathology of 15 patients, 7 were classified as achieving favorable response (treatment effect grade ≤ 1) and 8 as unfavorable response (treatment effect grade > 1). Neither fractional decrease in maximum SUV nor mean SUV were significant between the favorable and unfavorable groups. However, the fractional decrease in SUV20% (SUV to the highest 20% volume) was significant (p = 0.02), with an area under the Receiver Operating Characteristics (ROC) curve of 0.84. An optimal cutoff value of 0.46 for this metric was able to distinguish between the two groups with 71% sensitivity (favorable) and 88% specificity (unfavorable).

Conclusion: The fractional decrease in SUV to the volume with highest 20% intensity from pre- to intra-RT 18FDG-PET imaging may be used to distinguish between favorable and unfavorable responders with high sensitivity and specificity.

Contact Email: