The purpose of this research study is to determine how well the combination of bevacizumab and carboplatin works in treating breast cancer that has spread to the brain. Bevacizumab is an antibody (a protein that attacks a foreign substance in the body) that is made in the laboratory. Bevacizumab works differently from the way chemotherapy drugs work. Usually chemotherapy drugs attack fast growing cancer cells in the body. Bevacizumab works to slow or stop the growth of cells in cancer tumors by decreasing the blood supply to the tumors. When the blood supply is decreased, the tumors don't get the oxygen and nutrients they need to grow. Carboplatin is in a class of drugs known as platinum-containing compounds and has been approved for use in the treatment of ovarian cancer. Information from other research studies suggests that the combination of bevacizumab with carboplatin may be effective in treating breast cancer.

This study used a two-stage design to evaluate efficacy bevacizumab and carboplatin based on Central Nervous System (CNS) response. The null and alternative therapy response rates are 5% versus 20%. If 1 or more participants assessable in the stage one cohort (n=12 assessable participants) achieve CNS response then accrual proceeds to stage two (n=25 additional assessable participants). If at least 4 participants in the final set of 37 assessable participants achieve CNS response then this regimen would be deemed worthy of further study. The probability of stopping early is 0.54 if the true CNS response rate is 5% and 0.07 if the true CNS response rate is 20%. The probability of deeming the treatment worthy of further study is 0.10 and 0.90 if the true CNS response rate is 5% and 20%, respectively.

Inclusion Criteria:

• Histologically or cytologically confirmed invasive breast cancer, with metastatic disease. patients without pathologic or cytologic confirmation of metastatic disease should have unequivocal evidence of metastasis by physical exam or radiologic study
• Measurable disease. Patients must have measurable CNS disease, defined as at least one parenchymal brain lesion that can be accurately measured in at least one dimension with longest dimension >/= 10mm by local radiology review
• New or progressive CNS lesions, as assessed by the patient's treating physician
• No increase in corticosteroid dose in the week prior to the baseline brain MRI
• 18 years of age or older
• Life expectancy of greater than 12 weeks
• Eastern Cooperative Oncology Group Performance Score (ECOG PS) performance status 0-2
• Normal organ and marrow function as outlined in the protocol
• Left ventricular ejection fraction >/= 50%, as determined by radionuclide ventriculography (RVG) or echocardiogram within 60 days prior to initiation of protocol therapy
• Prior carboplatin is allowed if it was not given in conjunction with bevacizumab
• Prior trastuzumab is allowed
• No prior bevacizumab since diagnosis of CNS metastases or within 6 months prior to diagnosis of CNS metastases
• Women of child-bearing potential and men must agree to use adequate contraception prior to study entry and for the duration of study participation

Exclusion Criteria:

• Patients who have had chemotherapy within 14 days prior to entering the study, or those who have not recovered adequately from adverse events due to agents administered earlier
• Patients may not receive any concurrent investigational agents while on study
• Patients may not receive any cancer-directed concurrent therapy , such as concurrent chemotherapy, radiotherapy, or hormonal therapy while on study. Concurrent treatment with bisphosphonates is allowed
• History of Grade 3 or 4 allergic reactions attributed to compounds of similar or identical biologic composition to bevacizumab, carboplatin, or trastuzumab
• Known contraindication to MRI with gadolinium contrast, such as cardiac pacemaker, shrapnel, or ocular foreign body
• Leptomeningeal carcinomatosis as the only site of CNS involvement
• More than 2 seizures over last 4 weeks prior to study entry
• Grade 1 or higher CNS hemorrhage on baseline brain MRI
• History of grade 2 or higher CNS hemorrhage within 12 months of study entry
• Inadequately controlled hypertension
• Prior history of hypertensive crisis or hypertensive encephalopathy
• New York Heart Association (NYHA) Grade II or greater congestive heart failure
• History of myocardial infraction or unstable angina within 6 months prior to day 1
• Significant vascular disease within 6 months prior to day 1
• History of hemoptysis within 1 month prior to day 1
• Evidence of bleeding diathesis or significant coagulopathy
• Current, ongoing treatment with full-dose warfarin or its equivalent
• Use of aspirin (>325 mg/day) within 10 days prior to day 1
• Major surgical procedure, open biopsy, or significant traumatic injury within 28 days prior to day 1 or anticipation of need for major surgical procedure during the course of the study.
• Core biopsy or other minor surgical procedure, excluding placement of a vascular access device, within 7 days prior to day 1
• History of abdominal fistula or gastrointestinal perforation within 6 months prior to day 1
• Serious, non-healing wound, active ulcer, or untreated bone fracture
• Proteinuria as demonstrated by a urine protein-creatinine ratio >/= 1.0 at screening
• Known hypersensitivity to any component of bevacizumab
• Pregnancy or lactation

Adverse events (AEs) were assessed day 1 of each cycle throughout the treatment period. Median (range) treatment duration for this study cohort was 8 cycles (1-20) which approximates months given the 4-week cycle duration.

Maximum grade toxicity by type was first calculated. Serious AEs were defined as events with treatment-attribution of possibly, probably or definitely and grade 3 or higher per NCI Common Toxicity Criteria for Adverse Events version 3 (CTCAEv3). All remaining events regardless of treatment attribution were classified as Other AEs. No further data is available to specify classification of other beyond the general term.