Pancreatic cancer is one of the deadliest cancers, largely because most therapies are poorly active in patients or are too toxic when administered. Indeed, pancreatic cancer patients become ill very quickly, and cannot withstand the side effects of chemotherapy that patients with other types of cancer can tolerate. Therefore, we need to identify new therapies that kill pancreatic cancer cells effectively and are well tolerated by patients. To accomplish this goal, we have developed a new model system from pancreatic tumors, called organoids. Organoids are 3D cultures grown in an extracellular material rich matrix, called Matrigel, and can faithfully mimic the patient’s tumor, from which it was derived. Organoids can be used to sequence for mutations in the cancer cells and to test for therapies that could kill the cells. Using organoids, we have identified a number of compounds that can surprisingly kill the different cell types present in pancreatic tumors, including several drugs that are given to millions of people daily and are well tolerated but not previously considered to be cancer medicines. Importantly, we also find that certain combinations of these new drugs can shrink human pancreatic tumors engrafted in mice. Here, we propose to extend these exciting preliminary findings to a broader collection of drugs and a larger collection of organoids, and develop the most promising candidates as new strategies for an early phase clinical trial. Our goal is to test at least one novel combination of non-traditional drugs in pancreatic cancer patients within three years.