Background: The C-X-C chemokine receptor type 4 (CXCR4)–stromal cell-derived factor-1α (SDF-1α) axis regulates function and trafficking of immune cells and the tumour microenvironment. CXCR4 antagonists have been shown to enhance the activity of different anticancer treatments in preclinical models. We assessed the safety, tolerability, pharmacokinetics, and preliminary phase 1 activity of the CXCR4 antagonist, balixafortide, in combination with eribulin chemotherapy in patients with heavily pretreated, relapsed metastatic breast cancer. Methods: This single-arm, dose-escalation, phase 1 trial enrolled patients at 11 sites in Spain and the USA. Eligible patients were women aged 18 years or older who had histologically confirmed HER2-negative metastatic breast cancer, evidence of tumour cell CXCR4 expression, an Eastern Cooperative Oncology Group performance status of 0 or 1, and who had previously received between one and three chemotherapy regimens for metastatic breast cancer, and at least one endocrine therapy if they had hormone receptor-positive disease, unless they were considered unsuitable for endocrine therapy. A standard 3+3 dose-escalation design was used, followed by an expanded cohort at the established maximum tolerated dose or highest dose if no dose-limiting toxicity was observed for the combination. After a treatment-related fatal adverse event in the first cohort who received 21-day cycles of treatment with eribulin and balixafortide, a protocol amendment modified the study design to be done in two parts. Patients enrolled to part 1 received an initial 28-day run-in cycle, with some cohorts receiving de-escalated doses of eribulin plus balixafortide to assess the safety and pharmacokinetics of the combination. The evaluation of part 1 did not confirm any dose-limiting toxicities or eribulin–balixafortide interactions, and therefore part 2 started enrolling patients to receive eribulin at the originally planned dose of 1·4 mg/m 2 on days 2 and 9 of a 21-day cycle and balixafortide from a starting dose of 2 mg/kg with dose increments of 0·5 or 1 mg/kg on days 1–3 and 8–10 of the 21-day cycle. Both drugs were administered as intravenous infusions. All patients were to receive treatment until disease progression or unacceptable toxicity. The primary endpoints were dose-limiting toxicities and adverse events, and the establishment of a maximum tolerated dose or recommended phase 2 dose, and pharmacokinetic parameters. Safety analysis was done in all patients who received at least one dose of study treatment. Analysis of antitumour activity was done in all patients who received at least one full cycle of study treatment. The trial is registered at ClinicalTrials.gov, number NCT01837095, and is closed to accrual. Findings: Between Jan 28, 2014, and Oct 4, 2016, 56 patients were enrolled into the trial. No dose-limiting toxicities were confirmed and the maximum tolerated dose was not reached. The highest dose was established as eribulin 1·4 mg/m 2 on days 2 and 9, and balixafortide 5·5 mg/kg on days 1–3 and 8–10 of the 21-day cycle. Objective responses (all partial responses) were observed in 16 (30%; 95% CI 18–44) of 54 patients who were evaluable for antitumour activity. The most common treatment-emergent adverse events of any grade were fatigue (44 [79%] of 56 patients), neutropenia (32 [57%]), infusion-related reactions (27 [48%]), alopecia (26 [46%]), constipation (26 [46%]), and nausea (25 [45%]). Serious adverse events occurred in 21 (38%) of 56 patients, including febrile neutropenia in five (9%) of 56 patients, neutrophil count decrease in two (4%) patients, constipation in two (4%) patients, pneumonia in two (4%) patients, and urinary tract infection in three (5%) patients. Two (4%) of 56 patients died while receiving study treatment; one from septic shock and one from pneumonia. Interpretation: The safety and tolerability of balixafortide plus eribulin seems to be similar to that of eribulin or balixafortide monotherapy, and the preliminary activity of the combination seems promising in patients with HER-negative metastatic breast cancer. The results suggest that balixafortide plus eribulin has potential to provide a new therapeutic option in heavily pretreated patients with metastatic breast cancer and warrants further investigation in randomised trials. Funding: Polyphor.