PFS was defined for each patient as the time in months from randomization to the date of progression. Patients who died without a reported prior progression were considered to have progressed on their date of death. Patients who did not progress or die were censored on the date of their last tumor assessment.

All randomized patients with measurable disease as stratified at the time of randomization; n=480 and n=480 for the 2 treatment groups, respectively. Analysis was conducted once 903 progressions or deaths were observed in 960 participants.

RR=number of patients in that group whose best response is "partial"(30% decrease in the sum of the longest diameter of target lesions) or "complete" (disappearance of all target lesions), according to the 4-item Response Evaluation Criteria in Solid Tumors (RECIST), divided by the total number of response-evaluable participants

Response-evaluable participants (all treated participants with the correct diagnosis of adenocarcinoma originating in the breast who had measurable disease as determined at baseline).

Measured from the time RECIST criteria (described in previous outcome measure) were first met for "complete" or "partial" response until first date of documented disease progression or death. Patients who neither relapsed nor died were censored on the date of last tumor assessment. Median w/ 95% CI estimated using Kaplan Meier Product Limit Method.

Response-evaluable participants (all treated patients with the correct diagnosis of adenocarcinoma originating in the breast who had measurable disease as determined at baseline)

Time to response was defined as the time from the first dose of study therapy until measurement criteria were first met for "partial" or "complete" (whichever status was recorded first) per RECIST criteria (a 4-item scale described in the previous outcome measure).

Response-evaluable participants (all treated patients with the correct diagnosis of adenocarcinoma originating in the breast who had measurable disease as determined at baseline)

Laboratory values, adverse events, and other symptoms were graded using the National Cancer Institute's Common Terminology Criteria for Adverse Events (CTC) Version 3.0

All patients who received at least 1 dose of ixabepilone and/or capecitabine. Participants with baseline hepatic impairment (combination arm, n = 50; capecitabine arm, n = 37), defined as Grade ≥2 AST, ALT or Grade ≥1 total bilirubin, were contraindicated due to disproportionate number of toxic deaths observed in study CA163-046.

Quality of life, as measured by the FBSI, an 8-item, participant-reported instrument to measure symptoms. Each item has 5 possible responses ranging from 0 (not at all) to 4 (very much). The scoring was conducted according to the Functional Assessment of Chronic Illness Therapy manual, Version 4; higher scores reflect fewer symptoms.

Analysis was conducted on all randomized participants on an intent to treat basis. (Note: while table only reports data up to 24 wks, which represents most results, statistical analysis includes ALL assessments through study and follow-up; a few participants were assessed after more than 100 weeks.)

Overall survival was defined as the time in months from randomization until the date of death. For those patients who had not died, survival duration was censored at the last date the patient was known to be alive. Median OS with 95% CI estimated using the Kaplan-Meier Product Limit Method.

Analysis was conducted on all randomized patients on an intent to treat basis. This study required at least 846 events (deaths) to ensure the 2-sided, α = 0.05 level, log-rank test to have 90% power to show a statistically significant difference in OS between treatment groups when the hazard ratio (HR) is 0.8.