This study aims to evaluate the prognostic impact of the detection of circulating tumor DNA in peripheral venous blood, portal venous blood, collected intraoperatively, as well as its detection in peritoneal fluid.

Pancreatic adenocarcinoma is the 6th most common cancer in France and is expected to be the 2nd leading cause of cancer mortality in Europe by 2030. Its overall 5-year survival for all stages combined is estimated at 7-8%.

Surgery is the only potentially curative treatment for pancreatic adenocarcinoma (PA), although only 10-20% of patients present with resectable disease. The oncological results of surgery alone are disappointing (median overall survival (OS) of 15-20 months and 5-year OS of 8-15%), due to the high frequency of recurrence. For this reason, surgery must be integrated into a comprehensive treatment sequence, also known as multimodal therapy.

Significant progress has been observed over the past 5 years, with a marked increase in median OS in recent clinical trials, and an intensification of perioperative chemotherapy (CT) protocols.

Liquid biopsy and circulating biomarkers show promise for improving the multidisciplinary approach to treatment of pancreatic adenocarcinoma.

Circulating tumor DNA (ctDNA) is the most extensively studied marker in blood liquid biopsies and can provide insight into the molecular profile and individual characteristics of the tumor. This could pave the way for the identification of new therapeutic targets and markers of tumor response to complement diagnosis and provide improved stratified treatment. Similarly, the measurement of tcDNA in peritoneal fluid could provide a better understanding of individual tumour characteristics and the risk of peritoneal metastasis.

It is important to assess whether portal venous ctDNA is likely to provide distinct information from peripheral venous ctDNA; if the two ctDNAs are highly correlated, then it seems unlikely that one can provide complementary information to the other.

In addition, our work would evaluate the prognostic impact of the detection of circulating tumor DNA in portal venous blood collected intraoperatively, as well as its detection in peritoneal fluid.

Inclusion Criteria:

• Age ≥ 18 years
• Patient with operable pancreatic adenocarcinoma (resectable or boderline with or without neoadjuvant therapy) with histological evidence or indirect diagnostic criteria (CA 19-9 Significant (≥ 200 U/ml), pancreatic hypodense lesion without differential diagnosis)
• Biological criteria: Neutrophils > 1,500 cmm, platelet count > 100,000 cmm, creatinine clearance (Cockroft and Gault equation) > 60 ml/min, hemoglobin > 10 g/dl.
• ECOG performance status 0 or 1
• Patient has read and understood the information letter and signed the non-opposition/consent form.
• For an adult unable to give consent: Representative of the adult who has read and understood the information letter and signed the consent form

Exclusion Criteria:

• Other active cancer or haematological malignancy
• Uncontrolled congestive heart failure - untreated angina pectoris; recent myocardial infarction (within the last year) - uncontrolled hypertension (SBP > 160 mm or DBP > 100 mm, despite optimal drug therapy), long QT.
• Uncontrolled major infections, chronic infectious diseases, immunodeficiency syndromes
• Premalignant hematological disorders, e.g. myelodysplastic syndrome
• Severe hepatic impairment
• Patients not registered with social security
• Pregnant, parturient or breast-feeding woman (a pregnancy test will be carried out unless there is a history of hysterectomy, tubal ligation or menopause)
• Person deprived of liberty by an administrative or judicial decision, or person placed under court protection, sub-guardianship or guardianship.
• History of illness or psychological or sensory abnormality likely to prevent the subject from fully understanding the conditions required for participation in the protocol, or to prevent him/her from giving informed consent.
• Treatment contraindicated for venous sampling