The ORR was defined as the percentage of participants with best overall response (BOR) of confirmed complete response (CR) or partial response (PR), based on Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Targeted lesions were assessed by computed tomography (CT) and magnetic resonance imaging (MRI) which were then assessed by the investigator based on RECIST. CR was defined as the disappearance of all target lesions. PR was defined as at least 30% decrease in the sum of diameters of target lesions, taking as reference baseline sum diameters. Possible CR and PR had to be confirmed no fewer than 4 weeks after the initial response assessment. A brain and bone scan was performed by CT/MRI within 1 week after confirmation of a response to ensure no new metastases. To be assigned a status of CR or PR, changes in tumor measurements had to be confirmed by repeat evaluations, to be performed not fewer than 4 weeks after the response criteria were first met. ORR = CR + PR

Full analysis set included all participants who received at least one dose of eribulin mesylate.

Time to first response was defined for participants whose BOR was a CR or PR. Analysis was based on the Kaplan-Meier estimated number of months to CR or PR. This statistical analysis method measures the effect of study drug on CR or PR.

Full analysis set included all participants who received at least one dose of eribulin mesylate and were determined to be responders (CR or PR) to study treatment.

Duration of response was measured for participants who were responders only, had attained a BOR that was CR or PR. The duration of response was measured from time that response criteria for CR or PR (whichever was recorded first) were first met until the date that progressive disease (PD) or death from any cause was first objectively documented. Participants who did not have PD were censored on the day of their last tumor assessment. Duration of response was summarized for the responders using Kaplan-Meier estimation method. This statistical analysis method measures the effect of study drug on the length of response time.

Full analysis set included all participants who received at least one dose of eribulin mesylate and were determined to be responders (CR or PR) to study treatment.

PFS was defined as the time from the date of the first dose of study drug until the date of first documentation of PD or date of death from any cause, whichever occurred first. Participants who died without reported PD were considered to have progressed on the day of their death. Participants who were lost to follow-up or alive and without reported PD at the end of study were censored on the date of their last tumor assessment. Participants without evidence of PD upon discontinuation of study drug during the Extension Phase returned to the clinic for disease evaluation and PFS calculation every 12 weeks until PD was documented. PFS was analyzed using Kaplan-Meier product-limit estimates. This statistical analysis method measures the effect of study drug on PFS.

Full analysis set included all participants who received at least one dose of eribulin mesylate.

Treatment-emergent adverse events (TEAEs) were defined as AEs that emerged during treatment, having been absent at pretreatment, and occurring within 30 days of the last dose of study treatment, or if they were present prior to the first dose administration and increased in severity during the study. For each AE a participant with two or more TEAEs in that category were counted only once. TEAEs were considered related if the relationship of the event to study drug was possibly or probably related. Serious adverse events (SAEs) were defined as any untoward medical experience that resulted in death, was life-threatening, required hospitalization or prolongation of hospitalization, persistent or significant disability/incapacity, or was a congenital anomaly/birth defect. Safety information will be summarized with adverse events. AEs were graded on a five-point scale according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.0.

Safety Analysis Set population included all participants who received at least one dose of study drug and had at least one postbaseline safety assessment.