TPS772 Background: Regorafenib (rego) is a novel multikinase inhibitor. A clinical, placebo-controlled phase III study (CORRECT) showed a 1.4-month improvement in Overall Survival (OS) as compared to the placebo group (HR 0.77, p = 0.0052) for metastatic colorectal cancer (mCRC). In the CORRECT trial, the Progression Free Survival (PFS) curve showed that half of the patients in the rego group showed a similar tendency to the placebo group. The finding suggested that patients might be divided into responders and non-responders, and there might be a biomarker to predict them. Although various studies including the CORRECT trial have searched for the biomarker of rego, it still has not been established. On the other hand, tumor metabolic analysis using FDG-PET has been reported to be more sensitive than the change of tumor size using CT to predict the response of molecular targeting therapies. We conducted a prospective multicenter phase II trial to identify the usefulness of FDG-PET as an imaging biomarker for rego. Methods: The main eligibility criteria are as follows; pathologically proven mCRC, sufficient organ functions, failure of standard therapies (fluorouracil, oxaliplatin, irinotecan and any biologic agents), with tumors detected by FDG-PET. The treatment schedule consists of 160mg once-daily oral administration of rego for 3 weeks, followed by 1 week of rest. On day 28, FDG-PET is re-taken and the change of the maximum standardized uptake value (SUVmax) from that of the pre-treatment examination is evaluated. The primary endpoint is the change of SUVmax in the region with the highest value in pretreatment FDG-PET. Secondary endpoints are changes of SUVmax in top 5 regions (maximum 2 regions in one organ), OS, PFS, response rate, disease control rate, adverse events, and correlation of their clinical parameters and changes of SUVmax. Because there has been no data about early change in SUVmax when rego was used, we set the threshold value and estimate value as 0% and 10%, respectively. Assuming a one-sided alpha of 2.5% and a power of ≥ 90%, 16 subjects are required, and a target sample size of 20 patients was determined. Ten of the planned 20 patients have been enrolled. Clinical trial information: UMIN000015563.