Pancreatic ductal adenocarcinoma (PDAC) is the fifth leading cause of cancer-related death in Western countries, and its incidence has increased over the last 40 years. Curative surgery to manage PDAC is possible in only a fraction of patients; indeed, a vast majority (85%) of patients is diagnosed with locally advanced tumors and/or metastases because they lack specific symptoms and early markers for this disease. For these patients, palliative armamentarium consists of conventional chemotherapeutic agents such as Gemcitabine and, more recently, FOLFIRINOX, which offer marginal survival benefits. Consequently, the prognosis for PDAC is still very poor and there is great need for new treatments that can change this poor outcome. In this context, the investigators have devised, in the past few years, a highly innovative approach based on therapeutic gene transfer, which does not rely on a specific genetic and/or cellular background to inhibit PDAC tumor growth. the investigators found that SSTR2 and DCK::UMK gene transfer demonstrated complementary therapeutic effects to inhibit tumor progression and dissemination, and reduced tumor burden, respectively. On the basis of these promising preclinical data, the investigators conducted past three years the first clinical study of non-viral vector-mediated therapeutic gene delivery, guided by endoscopy (EUS), and combined with standard Gemcitabine therapy in patients with locally advanced and metastatic PDAC. The phase 1 demonstrated that the gene-therapy product CYL-02 is expressed in PDAC tumors (with long-lasting expression within tumor tissues), is distributed within the bloodstream in some extent, when combined with Gemcitabine it can inhibit primary-tumor progression and dissemination. Our results tend to demonstrate therapeutic efficacy, especially in patients with locally advanced tumors. Based on these encouraging results, the investigators propose that patients with locally advanced PDAC at the time of diagnosis may clinically benefit from this approach.

This phase II study is designed to compare the efficacy of intra-tumoral gene delivery of CYL-02 plus Gemcitabine treatment or Gemcitabine alone in patient with locally advanced PDAC.

Pancreatic adenocarcinoma is the fifth leading cause of cancer related deaths in Western countries. The sole curative treatment is surgical resection. Unfortunately, curative surgery is possible in only 10 to 15 % of cases. In a palliative way, the remaining patients can be treated with chemotherapy in locally advanced (Gemcitabine) or metastatic (FOLFIRINOX) tumors. These treatments ameliorate some clinical parameters but modestly the median survival. There is an urgent need in novel treatment and gene therapy is one of these innovative approaches. The investigators demonstrated that SST2 gene (that codes for somatostatin receptor subtype 2) when transferred in pancreatic experimental tumours significantly decreases tumor progression and exerts an anti-metastatic effect. SST2 gene transfer also induced a local bystander effect due to antiproliferative, pro-apoptotic and anti-angiogenic actions. SST2 gene transfer also sensitizes pancreatic tumors to the therapeutic effect of Gemcitabine. In order to improve this chemosensitization the investigators added the DCK::UMK gene, which is a fusion between two cDNAs each coding enzymes that are critical for Gemcitabine intracellular metabolism. Effect of DCK::UMK gene transfer in combination of Gemcitabine treatment dramatically reduced the size of primary tumours in experimental models of pancreatic cancer. Finally, combining SST2 and DCK::UMK gene transfer using a single expression plasmid intratumorally injected, the investigators obtained a significant "regression" of experimental primary pancreatic tumors (Patent Cayla-INSERM 2010). In collaboration with Invivogen company, this proof of concept led us to set up a pilot Phase I clinical trial of gene therapy in unresectable pancreatic cancer patients.

For this phase I gene therapy trial (THERGAP-01NCT01274455 Dec. 2010 - Sept. 2012), the investigators produced a gene therapy product (CYL-02) in GMP grade (plasmid encoding for SST2 and DCK::UMK genes complexed to a non-viral synthetic vector PolyEthyleneImine 22 kDa). The protocol was based on two Endoscopic Ultrasound (EUS)-guided direct intratumoral injections (at one month interval) of increasing doses of DNA (125 to 1000 µg) followed by Gemcitabine infusions within 2 months. After inclusion of 22 unresectable pancreatic cancer patients the investigators demonstrated the excellent feasibility of this protocol whatever the localization of the tumor and previous treatment. No serious adverse events directly imputable to the experimental product occurred. There was no major diffusion of CYL-02 in blood (none in urine) while transgene was detected and expressed in tumour tissues in a dose-dependent manner at one month post-injection (maximum at dose 1000µg - DNA + mRNA). Primary tumor was reduced in size at one month and remained stable at 2 months with no metastatic progression in the subgroup of locally advanced patients in whom median progression-free survival and overall survival were respectively of 6.4 and 12.6 months.

These results tend to demonstrate a therapeutic effect and lead to this randomized open multicenter phase 2 trial comparing the therapeutic effect of the association "intratumoral delivery of the gene therapy product CYL-02 plus gemcitabine" versus "gemcitabine alone" in patients with locally advanced non metastatic unresectable/untreated patients.

The primary objective is to compare the effect of each modality of treatment on the progression free survival. The secondary objectives are to compare the effect on overall survival, to evaluate the antitumor response, Quality of Life local and general tolerance.

For the 6 french centers the investigators plan to include 100 patients (18 months for inclusion, 12 months for duration of the trial for each patient) with proven locally advanced non metastatic untreated pancreatic adenocarcinoma, 50 patients in each Arm: Arm A: two EUS-guided injections of CYL-02 (dose 1000 µg) at one month interval in combination of Gemcitabine infusion 1000 mg/m2 every week within two months followed by 4 months (or until progression) of Gemcitabine (same dose and rhythm); Arm B: Gemcitabine alone 1000 mg/m2 within 6 months (or until progression). In each Arm patients will be subjected to clinical and imaging follow up after 6 months including or not other treatments. CYL-02 compound will be produced and supply by DBI/Eurofins and financed by Invivogen company.

This clinical trial will be accompanied by an independent monitoring committee, which will review safety data and provide recommendations to the Sponsor regarding the safety of subjects, the conduct of the study and potential premature termination.

Evaluation will be made upon clinical symptoms, signs of tumor progression, EUS in Arm A (at one month), CT-Scan, at 2, 4, 6, 9 and 12 months (ARM A and B, RECIST 1.1 criteria). Quality of Life every month (EORTC QLC-C30 questionnaire), local and general tolerance and Pharmacokinetics of CYL-02 will be also assessed.

The dose of 1000 µg will be used and corresponds to the maximal dose tested during the phase 1b (Thergap-1). This dose demonstrated a long lasting expression within tumor, a good antitumor effect and a good tolerance (buscail et al, mol her 2015).

Inclusion Criteria:

• Histologically or cytologically proven local advanced pancreatic adenocarcinoma patients,
• Non metastatic locally advanced non resectable CP assessed after multidisciplinary staff* and/or surgery,
• Patient without metastasis,
• No previous antitumor treatment or pancreatic resection,
• OMS status ≤ 2,
• Measurable tumor according RECIST criteria v 1.1,
• Patient that give their informed consent,
• Patient older than 18 years of age,
• Patients no contraindication of general anaesthesia,
• Patient with primary pancreatic tumour accessible to EUS (no digestive stenosis or stomach resection).

(*: at least oncologist, radiologist, digestive surgeon and gastroenterologist)

Exclusion Criteria:

• Patients with metastatic pancreatic tumors disease,
• Contraindication of Gemcitabine infusion,
• Non-measurable primary tumour (less than 2 cm in size),
• Borderline tumour according,
• Tumour eligible to a possible neo-adjuvant treatment by radio-chemotherapy or chemotherapy (after multidisciplinary staff*),
• Contraindication to EUS-guided fine needle aspiration biopsy (coagulation disorders),
• Patient in exclusion period or participating in another clinical research protocol,
• Patient that cannot understand or read the information form / consent or is not being able to take the decision to participate to the study,
• Pregnant woman, or of childbearing potential not using contraception,
• Patient under judicial protection, guardianship or curatorship,
• Patient with cystic tumor or pancreatic pseudocyst,
• Patient bearing solid tumors other than adenocarcinoma of the pancreas (endocrine tumor, metastasis),
• Granulocytopenia: granulocytes <1000/mm3,
• Thrombocytopenia: platelet count <100 000/mm3,
• Patient not effectively treated for malignant jaundice (biliary stent or bypass) if present at diagnosis.

(*: at least oncologist, radiologist, digestive surgeon and gastroenterologist)