Background: Chemotherapy in combination with targeted agents are standard-of-care options for patients for mCRC with response rates > 50% in first line. In the second line setting, efficacy of chemotherapy and targeted agents are much lower with response rates of 5% for FOLFIRI (5-fluorouracil, leucovorin, irinotecan) + bevacizumab (anti-VEGF). New treatment options are urgently needed in particular for the 50 % of patients harboring a KRAS mutation. PLK1 is a serine/threonine kinase, master regulator of the mitotic checkpoint and cell division. PLK1 is overexpressed in CRC and its overexpression is associated with poor prognostic. A genome wide RNAi screen identified PLK1 as a synthetic lethal target in KRAS mutant CRC cells, inducing cell cycle arrest and apoptosis upon inhibition. Onvansertib is an oral, highly selective PLK1 inhibitor that demonstrates single agent and synergistic activity with irinotecan and 5-FU in preclinical CRC models. Additionally, KRAS mutated versus wild-type cells showed higher sensitivity to onvansertib. PLK1 inhibition is a potential target in KRAS-mutated mCRC, and the combination of onvansertib + FOLFIRI + bevacizumab may provide a new second-line treatment option. Methods: The primary objective of this single-arm Phase 1b/2 study is to assess the safety and preliminary efficacy of onvansertib in combination with FOLFIRI and bevacizumab in the second line setting for KRAS-mutated mCRC patients. For the Phase 1b segment, a standard 3 + 3 dose-escalation design is used to determine the maximum tolerated dose or recommended phase 2 dose (RP2D) of onvansertib. As of September 15^th 2020, the first two dose levels have been cleared for safety and the third is enrolling. Efficacy will be determined by objective response rate (ORR) according to RECIST v1.1 (primary endpoint), progression-free survival and reduction in KRAS allelic burden in liquid biopsies (secondary endpoints). Biomarker analyses include genomic profiling of circulating tumor DNA genomic to evaluate altered pathways that correlate with patient clinical response. Based on a one-sided one sample log-rank test with 10% Type I error, there will be at least 90% power to detect an improvement in ORR from 5% to 20% with 32 patients. Clinical trial information: NCT03829410