Durvalumab(MEDI4736)/Tremelimumab in Combination With Gemcitabine/Cisplatin in Chemotherapy-naïve Biliary Tract Cancer

Overview

\<Research Hypothesis\> The dynamics of immune systems by cytotoxic chemotherapy and its changes by combination with immuno-oncology agents will be uncovered. The combination of Durvalumab/Tremelimumab with gemcitabine/cisplatin chemotherapy is feasible and efficacious in chemo-naïve biliary tract cancer. \<Purpose of the study\> To assess the effect of Durvalumab/Tremelimumab in combination with gemcitabine/cisplatin on response rate (RR) in chemo-naïve advanced biliary tract cancer patients.

\<Rationale for conducting this study\> 1. Rational #1. The incidence of biliary tract cancer (BTC) is higher in Korea than the West. (Korea: 10 new cases/100,000 population every year, the West:1-2 cases/100,100 population every year). Therefore, to conduct clinical study of BTC in Korea is very feasible and efficient. 2. Rational #2 The Gemcitabine/cisplatin is the current standard of care in 1st-line treatment for advanced BTC ((N Engl J Med 2010; 362 (14): 1273-81). No one-targeted therapy has been approved in BTC, yet. The overall survival of advanced BTC with cytotoxic chemotherapy is only 8-10 months, in general. Therefore, there is a huge unmet medical need. 3. Rational #3 In recent sequencing data of BTC showed the BTC patients with the worse prognosis had significant enrichment of hypermutated tumors and a characteristic elevation in the expression of immune checkpoint molecules. According, immune-modulating therapies also be potentially promising options for these patients (Nat Genet. 2015 Sep;47(9):1003-10.) 4. Rational #4 In PDL1 (+) BTC, anti-PD1 Ab shows promising activity as a monotherapy (Bang YJ, et al. ECC/ESMO 2015) In one clinical study of pembrolizumab, 37 out of 89 BTC patients (41.6%) showed the PDL1 (+) tumor. Among 24 PDL1 (+) patients who were enrolled and treated with pembrolizumab, 50% were Asian, 62.5% had ECOG 1, 16.7% had gallbladder cancer, 80% were at the 3rd-line or later setting. Four patients showed PR (3 from Seoul National University Hospital), 4 patients SD, which led the overall response rate of 17.4%. A total 40% of patients showed tumor shrinkage. The decreases in tumor size were generally maintained over time. This study gives us the evidence that immune checkpoint inhibitor is working on BTC likewise other solid tumors. 5. Rational #5 In recent studies have shown the combination of CTLA4 inhibitor with anti-PD1/PDL1 agents shows the enhanced clinical activities, especially, regardless of PDL1 status. This combination strategy is being actively under test in many solid tumors. 6. Rational #6 Certain chemotherapeutic drugs stimulate cancer-specific immune responses by inducing immunogenic cell death and other effector mechanisms (Immunity 2013, Annu Rev Immunol 2013) With the cytotoxic chemotherapy, the PDL-1 is induced and this increased expression of PDL1 contributes the resistance to cytotoxic chemotherapy. Successful eradication of tumors by immunogenic chemotherapy requires removal of immunosuppressive IgA+, PDL1+plasmocytes (Nature 2015). More importantly, still vast majorities of dynamic changes of immune system by cytotoxic chemotherapy are unanswered. These support that the combination of cytotoxic chemotherapy with immuno-oncology agents including immune checkpoint inhibitors might be efficacious and needed. 7. Rational #7 The advantages of "Immunotherapy and cytotoxic chemotherapy" combination are; 1) can explore science on the dynamic immunologic changes by cytotoxic chemotherapy and its overcome by immunotherapy 2) easy way to be incorporated in the current clinical practice 3) can be applied to variety of tumor types such as NSCLC, urothelial cancer. 4) relatively affordable than "immunotherapy and targeted agent" combination. Durvalumab/Tremelimumab in combination of cytotoxic chemotherapy has not been tested, especially in BTC. \<Sample Size Determination\> Primary efficacy endpoint is response rate. In BTC, the response rate of 1st-line gemcitabine/cisplatin chemotherapy is about 20% (20% in BT22 clinical trial, 25% in ABC-02 clinical trial). Therefore, we set H0 as 20%, and H1 as 40%. Using 75% of power and a-error of 0.05, a total 28 patients will be needed. When we assume the drop-out rate of 10%, a total of 31 patients will be enrolled.