Background: Pancreatic cancer (PC) is well-known as a strongly immunosuppressive tumor. The tumor microenvironment of PC play a fundamental part in maintaining a non-immunogenic and immuno-suppressive environment through the production of inhibitory cytokines and complex interaction between tumor cells, regulatory T cells, and myeloid-derived suppressor cells (Li KY et al. 2020). However, through integrated genomics and immunophenotyping, the immune-rich type had a relatively good prognosis in advanced PC (Wartenberg M et al. 2018). Based on this, various attempts have been made to apply immune checkpoint inhibitor (ICI) to PC. Disappointingly, ICI monotherapy is ineffective in most advanced PCs, and a strategy of combining ICI with ICI, chemotherapy or targeted therapy has been recently tried. However, the mechanism of immune modulation induced by ICI in combination with chemotherapy in PC is still not well-known, so biomarker studies are needed to understand the role of immunotherapy in advanced PC. Also based on this, it is expected to develop an effective ICI treatment strategy in PC. This study is a non-randomized, parallel assignment phase 2 trial aimed to explore dynamic modulation of the immune system caused by co-administration of pembrolizumab and cytotoxic chemotherapy in patients with advanced PC. Methods: Eligible patients have recurred or metastatic pancreatic cancer. Patients who have received chemotherapy for advanced PC are excluded, except for previous adjuvant chemotherapy. Patients who have received ICI are excluded. There are 2 chemotherapy cohorts (FOLFIRINOX, n=41; Gemcitabine/Nab-paclitaxel, n=36). Patients enrolled in the FOLFIRINOX cohort will receive pembrolizumab 200mg intravenously once every 3 weeks, plus FOLFIRINOX chemotherapy every 2 weeks. In Gemcitabine/Nab-paclitaxel cohort, patients will receive gemcitabine 1000mg/m2 plus nab-paclitaxel 125mg/m2 day 1, 8, 15 of a 4-week cycle, plus pembrolizumab 200mg intravenously once every 3 weeks. The primary endpoint is objective response rate, with key secondary endpoints including progression-free survival, duration of response, disease control rate, overall survival, safety, and patient-reported outcome. For evaluating metabolic response, ¹⁸F-FDG PET-CT scan will be conducted before treatment and at the first response evaluation. Tumor tissue biopsies are performed three times in total: screening, the first response evaluation, and disease progression. Blood samples are being collected every cycles for translational biomarker studies. Clinical trial information: NCT04447092.