Background: No standard treatment exists for triple negative breast cancer (TNBC) with failure of multi-line therapies. Apatinib is a small-molecule tyrosine kinase inhibitor that has promising anti-angiogenesis and antitumor activity for TNBC. We aimed to evaluate the safety and efficacy of adding apatinib to chemotherapy in patients with metastatic TNBC with failed first/second-line treatment. Methods: This randomized, open-label, phase 2 trial recruited patients with advanced TNBC who failed to receive first or second-line treatment. A total of 66 patients were randomly assigned, in a 1:1 ratio, to receive vinorelbine 25 mg/m² (days 1, 8, 15) or vinorelbine 20 mg/m² (days 7, 14, 21) with apatinib (250 mg once daily, days 1-5, 8-12, 15-19, if tolerable, the second cycle started with 500 mg per day) in 28-day cycles. The efficacy was evaluated every two treatment cycles (8 weeks ± 3 days). According to the RECIST criterion, patients with CR, PR and SD continued treatment until disease progression or unacceptable toxicity or withdrawal of consent. The primary end point was progression-free survival (PFS). Secondary end points included overall survival (OS), overall response rate (ORR) and safety. Results: Between Sep 14, 2017 and Dec 08, 2020, 66 patients underwent randomization. Median follow-up was 21.3 months. 33 received apatinib plus vinorelbine and 32 received vinorelbine (1 was withdrawal of consent). Median PFS was significantly longer in the apatinib plus vinorelbine group than in the vinorelbine group (3.8 months vs. 1.9 months; hazard ratio for disease progression or death, 1.76; 95% confidence interval [CI], 1.02 to 3.05; P= 0.039). Median OS was 14.6 months with apatinib plus vinorelbine and 14.1 months with vinorelbine (HR,1.34; 95% CI, 0.60 to 3.00; P= 0.469). The ORR was 48.5% in the apatinib plus vinorelbine group and 31.3% in the vinorelbine group (P= 0.156). The most common treatment-related hematologic grade 3–4 adverse events in those treated with apatinib plus vinorelbine versus vinorelbine, respectively, were leukopenia (42.4% vs. 34.4%), granulocytopenia (57.6% vs. 28.1%), anemia (9.1% vs. 12.5%) and thrombocytopenia (3.1% vs. 3.0%). The most frequent grade 3 nonhematologic toxicities were hand–foot syndrome (21%), proteinuria (9%), hypertension (9%) and increased ALT (9%) and which only occurred in apatinib plus vinorelbine group. No treatment-related nonhematologic grade 4 adverse events or treatment-related deaths were observed. Conclusions: Collectively, among patients with advanced TNBC with failed first/second-line treatment, apatinib plus vinorelbine show a promising benefit in PFS compared to vinorelbine monotherapy. Apatinib plus vinorelbine regimen shows promising efficacy and manageable toxicity, which might be a previously unappreciated therapeutic option for advanced TNBC. Clinical trial information: NCT03254654.