Preoperative gemcitabine-nab-paclitaxel (G-NP) for (borderline) resectable (BLR) or locally advanced (LA) pancreatic ductal adenocarcinoma (PDAC): Feasibility results and early response monitoring by Diffusion-Weighted (DW) MR.

Abstract

4116 Background: G-NP has recently emerged as an effective combination treatment for metastatic PDAC. We aimed to evaluate tolerability and activity of induction/neoadjuvant G-NP before surgery. In addition, we evaluated DW-MR imaging to monitor tumoral changes and to predict early response to chemotherapy compared with RECIST evaluation. Methods: Patients (pts) with ECOG PS 0/1, cytologically proven PDAC, normal bilirubin level and BLR or LA disease were included and received at least one cycle of G 1000 mg/m2 and NP 125 mg/m2, weekly 3week/4. Additional cycles were indicated according to sequential tumor restaging for resectability evaluation; complementary chemoradiation (CRT) (50.4 Gy/28 fr) with capecitabine (825 mg/m2 BID) was administered in case of persisting non resectability. Sequential DW-MR was performed at baseline and weekly during cycle 1 and monthly for subsequent cycles. DW-derived quantitative parameters (DW tumour volume and histogram based apparent diffusion coefficient (H-ADC) and pure diffusivity (H-D)) were compared between responders and non-responders, based on clinical/biochemical (CA 19.9) morphological and pathological response (pR) Results: 23 pts received therapy (M/F: 13/10; median age: 63, ECOG 0/1: 12/23), 13 BLR and 10 LA PDAC; median number of cycles: 3. Eight pts received CRT. Related grade III and IV toxicity occurred in 15 and 1 pts. No death occurred during the whole sequence. Best RECIST responses were 8 PR, 12 SD and 3 PD. Ten/16 pts (62%) with an elevated CA19.9 level at baseline decreased CA 19.9 levels by at least 50 %. R0 (n = 7)/R1(n = 6) resection was performed in 13/23 pts (56%).There were 0 pCR, 3 major pR and 10 minor/no pR. In 16 analyzed pts for DW-MR images, DW tumor volume, H-ADC and H-D variations were significantly different between responders and non-responders at the end of cycle 1. Conclusions: Induction chemotherapy with G-NP is feasible and active before surgery; our data suggest a potential interest of monitoring response to preop G-NP chemotherapy by using DW-MR for the early discrimination of responders/non-responders. Clinical trial information: NCT01715142.