Unencrypted login | home

Program Information

Evaluation of Local Therapy Response for Locally Advanced Pancreatic Cancer (LAPC) Using PET/CT


Y Yue

Y Yue*, R Tuli, W Yang, A Antonuk, H Sandler, B Fraass, Cedars-Sinai Medical Center, Los Angeles, CA 90048

TU-A-WAB-7 Tuesday 8:00AM - 9:55AM Room: Wabash Ballroom

Purpose: To quantify local treatment response for pancreatic radiotherapy using PET/CT, and investigate prediction of residual high-uptake areas using Pre-radiotherapy scans.

Methods: 18FDG-PET/CT scans were obtained from ten LAPC patients, pre-RT, 8-12 weeks following treatment (Post-RT), and at 2 to 4month follow-up intervals. Four patients received 3 follow-up scans; others had at least one follow-up scan.

All PET/CT images were registered to the planning CT using object-specific deformable registration which preserves target volume rigidity while deforming surrounding tissue. Therapy response evaluation is performed by comparing changes in SUVmax and volume pre-RT and post-RT. Local response is further evaluated by dividing the volume of interest into 4.8x4.8x4.8mm^3 subvolumes. Each subvolume is characterized with mean uptakes (SUVpre, SUVpost), SUVdiff=SUVpre-SUVpost, and treatment dose. Local response is evaluated using 3 volumes (VOIs): (1) ITV, (2) volume between PTV and ITV, (3) 3mm expansion around the PTV. For each VOI, the quantified response index is calculated [QR = (#subvolumes with SUVdiff>0)/(#subvolumes)]. A large QR suggests better local control. Subvolumes with SUVdiff<0 are investigated by examining SUVpre and treatment dose. Local response results are correlated to global SUVmax changes, disease volume changes, and clinical radiologist readings.

Results: 10 follow-up scans of 5 patients show strong positive response to treatment, with QRs for all VOIs >0.75 [ITV: 0.97+/-0.07, PTV-ITV: 0.84+/-0.18, 3mm PTV expansion: 0.76+/-0.29]. Global disease volumes (identified by high-uptake values) are reduced >70%, global SUVmaxs are reduced more than 50%. Three partial response patients have QR=0.50-0.75 (0.62+/-0.11 for ITV volumes). Two cases (failures) have QR<0.4, with mean 0.26+/-0.06.

Conclusion: Quantification of local metabolic response provides a feasible approach for evaluating efficacy of local control of pancreatic cancer treatment. The results can be used to identify persistent and local failure of disease, toward implementing response-driven adaptive radiotherapy.

Contact Email: