In the United States in 2017, most of the 135,000 people diagnosed with colorectal cancer received oxaliplatin to treat stage II-IV disease. About 70% of patients develop oxaliplatin-induced peripheral neuropathy (OIPN) that is characterized by upper and lower extremity numbness (N) and tingling (T), which can persist for years. Painful OIPN develops after N and T in 30% of patients. OIPN (N, T, and pain) poses a major health risk because it is associated with impaired function, falls, depression, impaired sleep, poor quality of life, and is a common reason for chemotherapy dose reductions. A critical gap in our scientific knowledge is that no known preventive interventions for OIPN exist. To address this gap, our overall objective is to test whether duloxetine prevents oxaliplatin-induced N, T, and pain, using a sequential Phase II to Phase III design that will be conducted via the National Cancer Institute (NCI) Community Oncology Research Program (NCORP), a large, multisite research network with access to diverse patient populations. Duloxetine will be tested in this study based on evidence of its efficacy for established OIPN from two clinical trials (Yang et al, 2012; Smith et al., 2013), and our pre-clinical data showing that duloxetine prevents painful OIPN in rats. We will first conduct a randomized, 3-arm, double-blind, placebo-controlled, non-comparative, multi-center study (N = 171) to screen two daily doses of duloxetine—30 mg and 60 mg—to prevent OIPN (N, T, and pain). If duloxetine is shown to be clinically active in the Phase II study, we will proceed to a randomized, double-blind, placebo-controlled, multi-center Phase III study to compare what appears to be the most promising duloxetine dose to placebo. To maximize the use of patient resources, the Phase II data from patients who either completed treatment with placebo (n = 54) or the most promising duloxetine dose (n = 54) will be pooled with data obtained from new Phase III trial accruals to the placebo (n = 70) and most promising duloxetine dose arms (n = 70), respectively (N = 248). We will use pre-established stopping rules to determine the optimal dose based on the proportions of patients who do not develop N, T, and pain, and adverse event severity. The two primary hypotheses in the Phase III study are that the most promising duloxetine dose will be more effective than placebo to prevent 1) N, T, & pain during oxaliplatin treatment and 2) chronic neuropathic pain one month after treatment. The temporal patterns of OIPN and functional impairment will be assessed for 18 months after oxaliplatin treatment. This study addresses the NCI Cancer Moonshot goal to minimize cancer treatment-associated debilitating side effects, and the priority recommendation outlined in the Institute of Medicine's Relieving Pain in America report regarding the need for non-opioid treatments for chronic pain. By addressing these priorities, we expect to make a major advancement in the field of symptom prevention, including pain, by identifying a well-tolerated, widely available, non-opioid, preventive intervention for a distressing and debilitating chemotherapy side effect experienced by millions of cancer survivors, for which no good treatment exists.