473 Background: Cancer of the exocrine pancreas is a highly lethal malignancy. Based on a phase III study, FOLFIRINOX regimen became the standard first-line treatment for patients with good performance status. However, the optimal management strategy for patients who fail initial FOLFIRINOX remains undefined. We aim at reporting our experience with single-agent gemcitabine as a second-line treatment for advanced pancreatic cancer patients who progressed on FOLFIRINOX. Methods: Patients with advanced exocrine pancreatic adenocarcinoma who received gemcitabine (1.000 mg/m² on days 1, 8 and 15 every 4 weeks) until disease progression, as second-line therapy after FOLFIRINOX failure at our institution were retrospectively evaluated. Progression-free survival (PFS) and overall survival (OS) were estimated by the Kaplan-Meier method. Results: A total of 20 patients were reviewed. Most of them (60%) had metastatic disease while 40% had locally advanced tumors. Median age was 60 years (range 43–74) and 80% were male. Eastern Cooperative Oncology Group (ECOG) performance status was 0 or 1 in 65% and 2 or 3 in 35% of the patients. Median time on prior FOLFIRINOX therapy was 5 months. Median PFS and OS with gemcitabine were 2,0 (95% CI 1,2-2,8) and 5,7 months (95% CI 3,9-7,4), respectively. There were no deaths due to the treatment. Conclusions: In this study, gemcitabine was a reasonable second-line treatment option for patients with advanced pancreatic adenocarcinoma. Phase III trials are urgently needed exploring the role of gemcitabine in the second-line setting.

468 Background: There is only limited data regarding efficacy and safety of FOLFIRINOX in elderly patients with advanced pancreatic adenocarcinoma. We aim to evaluate our experience with dose-reduced FOLFIRINOX in patients > 65 years-old. Methods: All patients > 65 years-old with biopsy-proven advanced pancreatic adenocarcinoma who received at least one cycle of modified dose-attenuatedFOLFIRINOX (no bolus FU and reduced dose of oxaliplatin and/or irinotecan) at our institution were identified. Results: Twenty-one consecutive patients (62% males) were reviewed. Median age was 67 (range 65-79). Grade 3/4 toxicities were reported in 7 (33%) patients ; the most common toxicities were anemia (62%), and nausea/vomiting (45%). Elevations in AST and/or ALT above the upper limit of normality were identified in 38%. No deaths due to toxicities were reported. Prophylactic granulocyte colony stimulator factor (G-CSF) was given in 16 (70 %) patients. Dose reductions were substantial with median reductions of 27,8 % (range 15 - 50%) for 5-FU, 27,1% (range 6-50%) for oxaliplatin and 25% (range 12-75%) for irinotecan. Median overall survival (OS) was 11,8 months (95% CI 10,2 - 13,3). Median progression free survival (PFS) was 6,9 months (95% CI 5,3 - 8,5). Conclusions: Modified dose-attenuated FOLFIRINOX is a reasonable option for elderly patients with advanced pancreatic adenocarcinoma. Adverse events were manageable and no deaths due to toxicity were reported. Median OS and PFS were similar to the pivotal phase III trial, demonstrating that dose-attenuated FOLFIRINOX may be beneficial to elderly patients.

216 Background: Despite a large number of randomized trials, there is no consensus as to the best regimen for advanced gastroesophageal cancer. Combination therapy with docetaxel, cisplatin, and 5-FU has shown increased response rates, progression-free survival (PFS) and overall survival (OS), but at the cost of significant toxicity. Based on a small phase II study, FLOT has been adopted by some groups given its better toxicity profile. We aim at reporting our experience with this regimen Methods: Patients with unresectable advanced gastroesophageal adenocarcinoma who received FLOT (oxaliplatin 85 mg/m2, leucovorin 200 mg/m2, and fluorouracil 2600 mg/m2 as a 24h infusion in combination with docetaxel 50 mg/m2 on day 1 every 2 weeks) as first-line therapy at our institution were retrospectively evaluated. PFS and OS were estimated by the Kaplan-Meier method. Results: A total of 23 patients were reviewed, most of them with metastatic disease (60%). Median age was 56 years (range 26–76) and 74% were male. Median PFS and OS were 5,2 (95% CI 4,0-6,3) and 8,5 months (95% CI 4,4-12,6), respectively. Only 13%of the patients experienced prolonged PFS (>12 months). There were no deaths due to the treatment. Conclusions: Before FLOT could be adopted as a first-line regimen for metastatic gastroesophageal cancer, phase III trials are urgently needed comparing FLOT with more traditional regimens.

72 Background: A considerable number of patients with mGIC progress after exhausting all approved standard therapies but maintain an adequate performance status and could be candidates for further treatment. We aim at reviewing our experience with the use of a NGS platform in refractory mGIC and its clinical utility. Methods: We retrospectively reviewed demographics, NGS results, and the suggested therapies received by patients undergoing NGS (Foundation Medicine, Cambridge, MA, USA): exonic sequencing of 236 genes and selective intronic sequencing from 19 genes for refractory mGIC. Co-primary endpoints were the percentage of patients with targeted therapy options uncovered by mutational profiling and the percentage of them who received genotype-directed therapy. Results: Samples from 32 patients were tested. Primary tumors consisted of colorectal adenocarcinoma (37,5%), pancreatic adenocarcinoma (31,2%), gastric adenocarcinoma (12,5%), cholangiocarcinoma (12,5%) and hepatocellular carcinoma (6%). Most patients (87,5%) were found to harbor potentially actionable genetic alterations involving mitogen-activated protein kinase (93,7%), phosphatidylinositol 3-kinase-AKT (18,7%), p53 (50%) and cell-cycle regulation (9%) pathways. Of the 6 patients who received the suggested targeted therapy, 4 achieved an objective response. Conclusions: Mutational profiling using a targeted NGS panel identified potentially actionable alterations in the majority of advanced gastrointestinal cancer patients. The assay provided clinical benefit in 12% of the patients.

315 Background: Metastatic PA is a lethal malignancy with poor prognosis and limited therapeutic options. We hypothesized that a comprehensive next-generation sequencing (NGS) assay could identify novel therapy targets not routinely explored in PA patients. Methods: We included patients with advanced pancreatic adenocarcinoma (PA) with accessible lesion for biopsy. Paraffin-embedded tumor tissues were evaluated by the method of NGS (Foundation Medicine, Cambridge, MA, USA). Hybridization capture of 3,769 exons from 236 cancer-related genes and 47 introns of 19 genes commonly rearranged in cancer was applied to ≥ 50 ng of DNA extracted from 10 PA specimens and sequenced to high, uniform coverage. Genomic alterations (GAs), were characterized and reported for each patient sample. Actionable GAs were defined as those identifying anti-cancer targeted therapies either available on the market or in registered clinical trials. Results: A total of 10 PA patients were included. Median age was 56 years (range 31-79). All cases harbored at least one GA with a mean of 2.7 actionable GAs per tumor. The most common actionable GAs were KRAS (90%), PTCH1 (20%), STK11 (10%) and GNAS (10%). KRAS mutations potentially treatable with MEK-inhibitor were identified in 9 patients. However, none of them received a MEK-inhibitor. Activating STK11 point mutations not detectable by IHC/FISH, potentially targetable with m-Tor inhibitor therapies were identified in 1 patient, who received everolimus and had a long-lasting (> 6 months) partial response. Conclusions: NGS of PA was able to reveal genomic alterations that may be potentially responsive to available targeted therapy. We are just at the dawn of exploring this strategy and NGS can effectively contribute for increasing the understanding of the disease.