The primary purpose of the amendment was to add/clarify several items in the original protocol. • Clarified eligibility criteria • Clarified that the most recent available result prior to randomization would be used to determine subject eligibility. • Clarified that the total daily capecitabine dose (2000 mg/m^2) was to be administered in approximately equal doses BID, in the morning and evening, approximately 12 hours apart, and that the dose was to be calculated according to BSA. • Added that subjects would continue receiving study treatment until they met withdrawal criteria. • Added that subjects who required the use of a potent cytochrome P450 3A4 inhibitor should have their ruxolitinib dose reduced from BID to once daily with frequent complete blood count monitoring during the period of coadministration. • Clarified that informed consent was to be obtained before performing any study-specific procedures that were not considered standard of care for the treatment of breast cancer and that screening assessments performed per standard of care did not need to be repeated after the ICF was signed if performed within 28 days of the Cycle 1 Day 1 visit. • Changed the frequency of radiological assessments to every 9 weeks until subjects had received 12 months of treatment. Thereafter, the frequency of scans was reduced to every 12 weeks. Screening CT scan or MRI of the brain was only to be performed if brain metastases or CNS symptoms were present. Bone scans were to be done at screening or if clinically indicated. Nontarget bone lesions (per RECIST v1.1) were followed every 9 weeks by CT scan with bone window or MRI. • Clarified that subjects who discontinued treatment because of disease progression did not need to repeat radiographic assessments at the EOT visit. • Clarified that subjects should bring all study-related medications with them to each study visit to assess study drug compliance.

The primary purpose of the amendment was to provide clarification regarding inclusion/exclusion criteria and acceptable forms of effective contraception. The clinically important changes included: • A new exclusion criterion was added to exclude pregnant and breastfeeding women. • Revised wording and updated the acceptable forms of effective contraception listed in the Protocol.

The primary purpose of the amendment was to update information regarding contraceptive methods in the Protocol, to be consistent with contraception recommendations from the 15 SEP 2014 Clinical Trial Facilitation Group meeting in Rome. Only highly effective methods of birth control (ie, failure rate < 1% per year when used consistently and correctly) were considered acceptable.