Neoadjuvant PAXG regimen doubles 3-year EFS over mFOLFIRINOX in resectable pancreatic cancer

Compared with modified FOLFIRINOX (mFOLFIRINOX), the neoadjuvant PAXG regimen, comprising cisplatin, nab-paclitaxel, gemcitabine, and capecitabine, significantly improved event-free survival (EFS) in participants with resectable or borderline resectable pancreatic ductal adenocarcinoma (PDAC).

The final results from the phase 3 CASSANDRA trial (NCT04793932) were presented by Prof. Michele Reni (Vita-Salute San Raffaele University, Italy) [1]. The trial employed a 2x2 factorial design and enrolled participants aged ≤75 years with treatment-naïve, histologically confirmed PDAC, and a Karnofsky performance status of ≥70. Participants were randomised to receive either PAXG (n=133) or mFOLFIRINOX (n=128), with chemotherapy administered either as a 6-month preoperative course or split between pre- and postoperative settings. The primary endpoint of EFS included a broad definition of failure (radiologic or biochemical progression, intraoperative discovery of metastases, unresectability, recurrence, or death), aligning with regulatory standards for neoadjuvant trials.

Among 261 eligible participants, PAXG promoted a 3-year EFS rate in 31% of participants versus 13% with mFOLFIRINOX (HR 0.64; 95% CI 0.48–0.86; P=0.003). The median EFS of 16.0 months (95% CI 12.4–19.8) for PAXG-treated versus 10.2 months (95% CI 8.7–13.0) for mFOLFIRINOX-treated participants. “PAXG significantly improved the median EFS by nearly 6 months, doubling the 3-year EFS rate,” summarised Prof. Reni.

The secondary endpoints also favoured PAXG, with a disease control rate of 98% versus 91% (P=0.009, PAXG vs mFOLFIRINOX). Additionally, 88% of PAXG-treated participants experienced a CA19-9 reduction greater than 50%, compared with 64% in the mFOLFIRINOX arm (P<0.001). While resection rates did not differ significantly (75% vs 67%, P=0.165, PAXG vs mFOLFIRINOX), PAXG yielded superior pathological outcomes: 35% of participants had stage <2 disease (vs 23%, P=0.03), and node-negative resections (R0/N0) were more common.  Although overall survival (OS) data are still maturing, trends mirror the EFS curves. The median OS was 37.3 months in the PAXG arm (95% CI 26.9 to not reached) versus 26.0 months with mFOLFIRINOX (HR 0.70; 95% CI 0.47–1.04; P=0.07).

The safety profile of PAXG was manageable, with higher rates of grade 3/4 neutropenia (42% vs 29%) in the PAXG arm, but similar incidences of other high-grade toxicities across arms. Participants receiving mFOLFIRINOX experienced more treatment discontinuations (47% vs 28%), mostly due to toxicity or disease progression. Quality-of-life outcomes were similar overall, although mFOLFIRINOX-treated participants reported greater deterioration by the 4th month in nausea, vomiting, and symptom burden.

Prof. Reni noted that the trial was initially designed as a phase 2 study but was upgraded to phase 3 following the adoption of neoadjuvant mFOLFIRINOX as standard care after the PREOPANC-03 trials (NCT04927780). This positions CASSANDRA as a timely head-to-head comparison of 2 leading neoadjuvant regimens.

“The improved tumour response, lower perioperative metastasis rate, and enhanced pathological outcomes make PAXG an appealing option for patients with resectable PDAC,” Prof. Reni concluded. “Upon confirmation from overall survival data, PAXG could redefine the standard-of-care in resectable PDAC.”

1. Reni M, et al. Results of a randomized phase III trial of pre-operative chemotherapy with mFOLFIRINOX or PAXG regimen for stage I-III pancreatic ductal adenocarcinoma. Abstract LBA4004, ASCO Annual Meeting 2025, 30 May–3 June, Chicago, IL, USA.

Posted on 30 July 2025