Brian Jin MD, Presenter: Nothing to Disclose

Elizabeth M. Hecht MD, Abstract Co-Author: Nothing to Disclose

David K. Leung MD, PhD, Abstract Co-Author: Nothing to Disclose

Kyung Chu NP, Abstract Co-Author: Nothing to Disclose

Ranjit Singh Sandhu MD, Abstract Co-Author: Nothing to Disclose

William H. Sherman MD, Abstract Co-Author: Nothing to Disclose

To investigate whether baseline FDG PET-CT parameters can be used to predict tumor RECIST 1.1 response following neoadjuvant chemotherapy of locally advanced pancreatic adenocarcinoma. 

Patients with locally advanced pancreatic adenocarcinoma enrolled in a neo-adjuvant GTX chemotherapy trial. All patients underwent pre-treatment FDG PET-CT and MRI. Baseline PET-CT parameters were measured: Tumor SUVmax, SUVmean, 3-D metabolic tumor volume (MTV, obtained using an automated half-maximum-SUV threshold), and total lesion glycolysis (TLG; SUVmean*MTV). Tumor size (maximum axial diameter) was measured on MRI (pre-treatment, post 3 and 6 cycles of GTX). Tumor response of the target lesion after 3 and 6 cycles was recorded using RECIST 1.1 criteria: complete response (CR); partial response (PR); progressive disease (PD); stable disease (SD). Pre-treatment PET-CT measurements in patients with CR or PR after 3 and 6 cycles of GTX were compared to patients with SD or PD using independent t-test,  P-value < 0.05 was considered statistically significant. Analysis was performed using a commercially available statistical software package (SPSS version 16.0, Chicago, Illinois). 

34 subjects (17M, 17F; mean 66.9 y) were enrolled. Mean tumor size: 3.4 cm. Quantitative results are as follows: tumor SUVmax 7.0 ± 3.1, tumor SUVmean 4.5 ± 2.0, MTV 19 ± 12cm3, TLG 80 ± 59 SUV-cm3 . After 3 cycles of GTX, 8 subjects exhibited PR, 26-SD. There was no difference between tumor SUVmax or SUVmean in the PR and SD cohorts but there was a statistically significant difference in MTV (33 ± 17 cm3 - PR group, 15 ± 6.7cm3 -SD group, P<0.001) and TLG (138 ± 88 SUV- cm3  versus 62 ± 31 SUV- cm3, P=0.001). After 6 cycles, 15 patients had PR, 2-CR, and 16-SD (1 death occurred between cycles 3-6). Given the small number of CR patients, they were grouped together with PR patients (PR+CR group). No baseline PET-CT parameter demonstrated statistically significant differences between the PR+CR and SD groups. 

In patients with locally advanced PDA, baseline PET-CT quantitiative metrics including tumor SUVmax, SUVmean, MTV and TLG fail to predict RESIST response after 6 cycles, suggesting that the prognostic ability of PET-CT to predict anatomic response is limited.

Pre-treatment PET-CT fails to predict anatomic tumor response in patients with locally advanced pancreatic cancer undergoing neoadjuvant chemotherapy.

Jin, B, Hecht, E, Leung, D, Chu NP, K, Sandhu, R, Sherman, W, Can Pretreatment PET-CT Predict RECIST 1.1 Response in Subjects with Locally Advanced Pancreatic Adenocarcinoma Undergoing Neoadjuvant GTX Chemotherapy?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045472.html