Highlights
– In the randomized, open‑label, phase 3 PACT‑21 CASSANDRA trial, preoperative PAXG (cisplatin, nab‑paclitaxel, capecitabine, gemcitabine) prolonged median event‑free survival (EFS) to 16.0 months versus 10.2 months with mFOLFIRINOX (HR 0.63; 95% CI 0.47–0.84; p=0.0018).
– Grade 3 or higher adverse events were frequent in both arms (66% PAXG vs 61% mFOLFIRINOX); one treatment‑related death occurred. All 260 randomized patients received at least one cycle.
– Investigators conclude that preoperative PAXG could be considered a standard option for resectable and borderline resectable pancreatic ductal adenocarcinoma (PDAC) pending overall survival and longer follow‑up data.
Background and disease burden
Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal solid malignancies. Even among patients with anatomically resectable or borderline resectable disease, recurrence rates after surgery are high and long‑term survival is limited. Perioperative and adjuvant systemic therapies aim to treat micrometastatic disease, improve margin‑negative (R0) resection rates, and prolong disease‑free and overall survival.
Adjuvant modified FOLFIRINOX (mFOLFIRINOX) established an overall survival benefit compared with gemcitabine in the PRODIGE‑24 trial and has been widely adopted in appropriate patients with good performance status. Neoadjuvant strategies in resectable or borderline resectable PDAC have gained traction for patient selection, early systemic control, and increasing R0 resection rates (e.g., PREOPANC and other trials). Whether regimens other than mFOLFIRINOX provide superior perioperative outcomes is an open question.
Study design
The PACT‑21 CASSANDRA trial is a randomized, open‑label, 2 × 2 factorial phase 3 study conducted across 17 academic centres in Italy. Eligible patients were 18–75 years old with pathologically confirmed resectable or borderline resectable PDAC (clinical stage I–III).
Initial randomisation was 1:1 to one of two preoperative chemotherapy regimens for four months:
- PAXG: capecitabine total daily dose 1250 mg/m2 (625 mg/m2 twice daily) combined with intravenous cisplatin 30 mg/m2, nab‑paclitaxel 150 mg/m2, and gemcitabine 800 mg/m2 every 14 days;
- mFOLFIRINOX: intravenous fluorouracil 2400 mg/m2 (continuous infusion), leucovorin 400 mg/m2, irinotecan 150 mg/m2, and oxaliplatin 85 mg/m2 every 14 days.
After the initial four months, participants underwent a second randomisation (reported separately) to 2 months of additional chemotherapy delivered either before or after surgery. The primary endpoint for the first randomisation was event‑free survival (EFS) in the intention‑to‑treat population. Safety analyses included all patients receiving at least one therapy administration.
Key findings
Between November 3, 2020 and April 24, 2024, 260 eligible patients were randomized: 132 to PAXG and 128 to mFOLFIRINOX. Baseline characteristics were well balanced; median ages were 65 years (IQR 60–70) and 63 years (IQR 57–69), respectively, and sex distribution was similar.
Primary outcome: event‑free survival
PAXG significantly improved median EFS compared with mFOLFIRINOX:
– Median EFS: 16.0 months (95% CI 12.4–19.8) with PAXG vs 10.2 months (95% CI 8.6–13.5) with mFOLFIRINOX.
– Hazard ratio for an event (PAXG vs mFOLFIRINOX): 0.63 (95% CI 0.47–0.84); p = 0.0018.
This magnitude of reduction in the risk of progression, failure to reach resection, or death in the preoperative period is clinically meaningful in a curative‑intent population and suggests superior early disease control with PAXG.
Safety and tolerability
Toxicity was substantial with both regimens, reflecting the intensity of neoadjuvant therapy and patient population. At least one grade 3 or worse adverse event occurred in 87 of 132 patients (66%) in the PAXG group and 78 of 128 patients (61%) in the mFOLFIRINOX group. There was one fatal adverse event reported (the report does not specify attribution in the provided summary).
Key practical considerations include the need for proactive supportive care (antiemetics, growth factor support when indicated, hydration and renal monitoring for cisplatin, neuropathy surveillance with both platinum and taxane exposure, and close management of cytopenias).
Other outcomes and procedural data
The published summary focused on the first randomisation EFS analysis and safety. Details on secondary endpoints that are important for translation—such as R0 resection rate, pathological response, postoperative morbidity and mortality, time to surgery, number of patients completing surgery, and quality‑of‑life measures—were not included in the provided extract and will be important to evaluate when full results are available. Overall survival follow‑up is ongoing.
Expert commentary and contextual interpretation
These randomized phase 3 data are notable because they directly compare two multi‑agent, intensive neoadjuvant regimens in a perioperative PDAC population. Strengths include multicentre enrolment, randomisation with stratification by centre and CA19‑9, intention‑to‑treat analysis, event‑driven primary endpoint achievement, and completion of accrual.
Interpretation should be balanced by several caveats. First, the trial is open‑label, which is common in chemotherapy comparisons but can introduce post‑treatment management biases. Second, EFS is an important proximal endpoint but not synonymous with overall survival; OS results are awaited. Third, toxicity was high in both arms—translating trial tolerability into real‑world practice (especially outside academic centres or in older patients beyond the 75‑year cutoff) may be challenging.
Biological plausibility: PAXG combines a platinum agent (cisplatin), a taxane (nab‑paclitaxel), gemcitabine (a backbone used in PDAC), and capecitabine, a fluoropyrimidine prodrug. This combination exerts multi‑mechanistic cytotoxic pressure that may increase early systemic control and tumor shrinkage, potentially explaining higher EFS. mFOLFIRINOX is itself a powerful multi‑agent regimen (fluorouracil, irinotecan, oxaliplatin) and is established in the adjuvant setting by PRODIGE‑24 (Conroy et al., NEJM 2018). Whether superior EFS will translate into OS advantage depends on multiple factors including perioperative mortality, adjuvant therapy selection, and patterns of recurrence.
Limitations and questions remaining
- Overall survival data are immature; OS will determine whether EFS gains translate into survival benefit.
- Detailed surgical outcomes (resection rates, R0 margin status, postoperative complications) and patient‑reported outcomes are necessary to assess net clinical benefit.
- The trial population was limited to age ≤75 and conducted in 17 Italian academic centres; generalisability to older patients and non‑academic settings requires confirmation.
- Open‑label design and absence of central radiology/surgical review in the provided summary may introduce assessment variability.
- Biomarker‑driven patient selection (e.g., homologous recombination deficiency, BRCA status, baseline CA19‑9 kinetics) was not reported here and might identify subgroups deriving most benefit.
Clinical implications and practice integration
If subsequent OS and surgical outcomes corroborate the EFS findings, PAXG could be adopted as a standard neoadjuvant option for fit patients with resectable or borderline resectable PDAC. For now, clinicians should weigh the potential EFS advantage against toxicity profiles, centre experience in delivering complex multi‑agent chemotherapy, and patient preferences.
Key practical recommendations while awaiting full dataset:
- Discuss both regimens in multidisciplinary boards, including surgeons, medical oncologists, radiation oncologists, and supportive‑care teams.
- Screen patients carefully for comorbidities (renal function for cisplatin, neuropathy risk, performance status) and provide aggressive supportive care measures to maintain dose intensity when possible.
- Enroll eligible patients into ongoing trials or registries to gather real‑world evidence and biomarker data.
Research priorities
Important next steps include the completion of OS analysis, reporting of surgical and pathologic outcomes, and quality‑of‑life data. Prospective biomarker studies (BRCA/HRD status, circulating tumor DNA dynamics, and CA19‑9 kinetics) could refine selection for PAXG versus mFOLFIRINOX. Comparative effectiveness in older or frailer populations, and combinations with radiotherapy or novel targeted/immunotherapeutic agents, are logical future directions.
Conclusion
The first randomisation analysis of PACT‑21 CASSANDRA demonstrates a statistically and clinically meaningful improvement in event‑free survival with preoperative PAXG versus mFOLFIRINOX in resectable and borderline resectable PDAC. Toxicity is substantial but broadly comparable between arms. Pending overall survival, surgical outcome, and patient‑reported outcome data, preoperative PAXG represents a promising neoadjuvant option and may serve as an appropriate comparator regimen in future trials.
Funding and trial registration
The study was funded by MyEverest and Codice Viola. Trial registrations: ClinicalTrials.gov NCT04793932; EudraCT 2020‑003080‑26 and 2024‑519031‑42‑00.
Selected references
1. Reni M, Macchini M, Orsi G, et al. Preoperative mFOLFIRINOX versus PAXG for stage I–III resectable and borderline resectable pancreatic ductal adenocarcinoma (PACT‑21 CASSANDRA): results of the first randomisation analysis of a randomized, open‑label, 2 × 2 factorial phase 3 trial. Lancet. 2025 Nov 20: S0140‑6736(25)01685‑X. doi:10.1016/S0140‑6736(25)01685‑X. PMID: 41275879.
2. Conroy T, Hammel P, Hebbar M, et al. FOLFIRINOX or gemcitabine as adjuvant therapy for pancreatic cancer. N Engl J Med. 2018;379:2395–2406. (PRODIGE‑24/CCTG PA.6)
3. van Tienhoven G, Versteijne E, Suker M, et al. Preoperative chemoradiotherapy versus immediate surgery for resectable and borderline resectable pancreatic cancer (PREOPANC): long‑term results of a randomized, controlled, multicentre trial. Lancet Oncol. 2020;21:1090–1101.
