Background: A randomized controlled trial was conducted to assess the efficacy and safety of dose-dense, weekly paclitaxel plus carboplatin (ddTC) with or without bevacizumab (Bmab) compared to conventional, tri-weekly paclitaxel plus carboplatin (cTC) with or without Bmab, in metastatic or recurrent cervical carcinoma not amenable to curative treatments with local therapy. Methods: Patients were randomly assigned to either a cTC or a ddTC regimen. After Bmab was approved in Japan (in May 2016) the protocol was amended, and patients on both arms received Bmab if not contraindicated. The cTC was paclitaxel 175 mg/m² intravenously (IV) for 3 h on day 1 followed by carboplatin at an area under the curve of five IV for 1 h on day 1. The ddTC was paclitaxel 80 mg/m² IV for 1 h on day 1 followed by carboplatin at an area under the curve of five IV for 1 h on day 1 and paclitaxel 80 mg/m² IV for 1 h on day 8 and day 15. Both cTC and ddTC treatments were repeated every three weeks, for up to nine cycles. Bmab 15 mg/kg IV was repeated until progression or unacceptable toxicity. The primary endpoint of phase II was the response rate (RR) in patients with measurable lesion, who had received Bmab. If the RR of the ddTC + Bmab arm was greater than that of the cTC + Bmab arm for more than 5%, the study would proceed to phase III, which had overall survival (OS) as its primary endpoint. The planned sample size in the phase II part was 56 to select the ddTC arm with a probability of at least 75% if the difference of RR was 15% or more (45% vs. 60%). Results: Patient accrual started in October 2015. It was suspended in May 2019 because the number of Bmab-treated patients with measurable lesions reached 56. In total, 122 patients were enrolled and randomly assigned to either the cTC arm (cTC: 29 patients; cTC + Bmab: 32 patients) or the ddTC arm (ddTC: 30 patients; ddTC + Bmab: 31 patients). The primary analysis of the phase II part was conducted in November 2019. The RRs of each regimen were 67.9% [95% CI, 47.7-84.1] (19/28, cTC + Bmab), 60.7% [40.6-78.5] (17/28, ddTC + Bmab), 55.2% [35.7-73.6] (16/29, cTC), and 50.0% [29.9-70.1] (13/26, ddTC). Conclusions: The study did not meet the primary endpoint of phase II. Dose-dense, weekly paclitaxel plus carboplatin is not promising for metastatic or recurrent cervical carcinoma. Clinical trial information: jRCTs031180007.