PAXG Opens a New Chapter in Preoperative Treatment for Pancreatic Cancer

Is the PAXG Regimen Superior to mFOLFIRINOX?

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most aggressive cancers. Long-term survival rates are limited even in patients diagnosed with locally advanced disease. It is therefore important to determine whether preoperative (neoadjuvant) treatment strategies can prolong survival in stage I–III patients who have a chance of surgery.

Addressing this question, and presented at ASCO 2025, the CASSANDRA Phase 3 studycompared the current standard, mFOLFIRINOX , with a new combination, the PAXG regimen . It is noteworthy as the first large randomized trial directly comparing the effectiveness of these regimens in patients with resectable and borderline resectable pancreatic cancer.

Study Design: Which Patients, When, and Which Regimen?

CASSANDRA included patients with stage I–III, resectable (R) or borderline resectable (BR) PDAC: 261 patients aged 75 or younger who had not received systemic treatment. Patients were randomized in a 2-by-2 factorial design:

• PAXG regimen: Capecitabine + cisplatin + nab-paclitaxel + gemcitabine
• mFOLFIRINOX: 5-FU + irinotecan + oxaliplatin

The first randomization determined whether patients received 6 months of preoperative PAXG or mFOLFIRINOX. In the second randomization, some patients received the same chemotherapy for an additional 2 months after surgery.

The primary endpointwas event-free survival (EFS), with events defined as:

• Radiologic progression,
• Loss of resectability,
• An increase in CA19-9,
• Metastasis detected during surgery,
• Local or distant recurrence,
• Death.

Main Findings: PAXG Clearly Ahead

🔹 Median follow-up: ~24 months

🔹 Median event-free survival (EFS):

• PAXG: 16 months
• mFOLFIRINOX: 10 months
  HR: 0.64; p = 0.003

🔹 3-year event-free survival rate:

• PAXG: 31%
• mFOLFIRINOX: 13%

🔹 Similar benefits were seen in subgroup analyses:

• Resectable patients: 41% vs 22%
• Borderline resectable patients: 19% vs 9%

Secondary Outcomes: Pathologic and Biomarker Responses

PAXG was superior not only in EFS but also in several secondary endpoints:

Safety: An Acceptable Toxicity Profile

Both regimens were well tolerated. However:

• Grade 3–4 neutropenia: 42% with PAXG, 29% with mFOLFIRINOX
• Other adverse effects (fatigue, nausea, neuropathy) were similar.
• By the end of month 4, both groups experienced quality-of-life problems such as fatigue, concerns about weight loss, and taste disturbances.

What Do the Experts Say? Is the Standard Changing?

The study presenter, Dr. Michele Reni, described PAXG as “the most suitable neoadjuvant treatment candidate” for this patient population, while discussant Dr. Brian Wolpin was more cautious:

“This study is very promising, but OS data are not yet mature. There is a trend toward a survival benefit, but the evidence is not yet sufficient to change the standard.”

🔍 Current data:

• Median overall survival: PAXG 37 months vs mFOLFIRINOX 26 months
• 3-year overall survival rate: 51% vs 40% (HR 0.70; p = 0.07)

Conclusion: New Hope, but Cautious Optimism Is Essential

In CASSANDRA, PAXG provided significantly better event-free survival and several secondary clinical benefits compared with mFOLFIRINOX. However, overall survival data should mature before it is accepted as standard treatment.

This study has the potential to shape the future of neoadjuvant treatment in pancreatic cancer. Alongside other large ongoing studies, such as PREOPANC-3 and ALLIANCE A021806 , it appears likely to further clarify personalized treatment strategies.

Ultimately, this study illustrates the upper limit of what chemotherapy combinations can achieve. The time has come for entirely new strategies in pancreatic cancer.