Background: Most recently, there were 3 reports of prospective randomized clinical trials comparing the effects of primary tumor resection (PTR) for multiorgan metastatic colorectal cancer followed by chemotherapy with chemotherapy alone, but the results differed and unconvincing due to the prematurely study termination and research protocol changes. PTR was preferably performed for patients with asymptomatic synchronous unresectable colorectal liver-limited metastases (CRLMs) with conversion therapy purpose, including the CELIM, OLIVIA and our study (J Clin Oncol 2013;31:1931-8). This randomized phase III study investigated the superiority of preoperative chemotherapy prior to PTR for patients with asymptomatic synchronous unresectable CRLMs. Methods: Patients with asymptomatic synchronous unresectable CRLMs were randomly assigned to receive pre-PTR chemotherapy (arm A) or upfront PTR (arm B). Chemotherapy regimens of mFOLFOX6 plus cetuximab, mFOLFOX6 plus bevacizumab or mFOLFOX6 alone were decided according to the RAS genotype. The primary end point was progression-free survival (PFS); secondary end points included overall survival (OS), tumor response, disease control rate (DCR), liver metastases resection rate, surgical complications and chemotherapy toxicity. Results: Between June 2012 and June 2018, a total of 320 patients were randomly assigned to arm A (160 patients) or arm B (160 patients). The cutoff date for survival data was June 2021, the median follow-up time was 36.2 months. Patients were well balanced. For the intention-to-treat population, the median PFS, median OS, and 3-year OS rates were 9.9 months, 28.0 months, and 37.0%, respectively. The median PFS in arm A was significantly improved compared with arm B (10.5 v 9.1 months; hazard ratio [95% CI, 0.60 to 0.95], 0.76; P = 0.013). Patients in arm A also had a significantly better DCR (84.4% v 75.0%; P = 0.037). The median OS was not significantly different (29.4 v 27.2 months; hazard ratio [95% CI, 0.58 to 1.01], 0.77, P = 0.058), and the objective response rates were also not significantly different (53.1% v 45.0%; P = 0.146). The actual resection rate of liver metastases was not significantly different (21.9% v 18.1%; P = 0.402). There were mild morbidities and no 30-day postoperative mortalities in both arms. The rate of complications was not significantly different (37.7% v 30.8%, P = 0.201). The incidence of Clavien–Dindo 3-4 complications also did not reach statistical significance (4.5% v 3.8%, P = 0.759). Overall the observed toxicity was mostly mild. There was no significant difference in the overall incidence of predefined grade 3/4 events (42.2% v 40.4%, P = 0.744). There were no grade 5 events in either arm. Conclusions: For asymptomatic synchronous unresectable CRLMs, Pre-PTR chemotherapy improved the PFS compared with upfront PTR. Clinical trial information: NCT01307878.