Enrollment was going slowly and needed to be closed.

The investigators' hypothesis is that valproic acid given before surgery for newly diagnosed breast cancer will increase breast tumor histone acetylation at tolerable doses and that the increase in breast tumor histone acetylation will correlate with valproic acid blood levels and changes in peripheral blood white blood cell histone acetylation. Published in vitro studies have shown sensitivity of breast cancer cells to histone deacetylase inhibitors (Fortunati et al., 2008; Fuino et al., 2003; Hodges-Gallagher et al., 2007; Olsen et al., 2004). The investigators' gene array data predict sensitivity to valproic acid in over half of breast cancers [Bild, unpublished]. The investigators hypothesize that in women with newly diagnosed breast cancers valproic acid will have an unacceptable toxicity rate less than 15% at doses that increase tumor histone acetylation and that valproic acid will decrease the Ki-67 in at least half of breast tumors by over 20%. The investigators also hypothesize that their genomically-derived signature for sensitivity to valproic acid (GDSS-VPA) can be used to predict which tumors will have a decrease proliferation as measured by Ki-67 by at least 20%. The investigators hypothesize that valproic acid levels and histone acetylation levels in peripheral leukocytes will correlate with a decrease in the Ki-67 proliferation index by 20%. The investigators hypothesize that DCE-MRI imaging studies will provide an accurate and quantitative means of assessing tumor response to valproic acid. Finally, the investigators hypothesize that response to valproic acid will not be affected by intrinsic breast cancer subtype.

Before treatment each woman needs a DCE-MRI and a biopsy of the breast mass. For women who have had diagnoses of breast cancer outside our institution, the study procedures are illustrated in figure 1. Following screening labs and enrollment, a DCE-MRI will be performed followed by a biopsy to obtain two cores for study immunohistochemistry studies and one core for the GDSS-VPA. The biopsy will be examined by the pathologist to confirm diagnosis and stained for markers of histone acetylation, proliferation, and apoptosis.

For women who are coming to the HCI for a breast biopsy for a suspicious mass that is greater than 1.5 cm by exam, mammogram, ultrasound, CT, or MRI, study procedures are illustrated in figure 2. A DCE-MRI will be done on the same day prior to the biopsy under existing research MRI consents, if possible. Biopsy material is already obtained for research purposes using a tissue collection consent. One of these cores will be placed in liquid nitrogen for storage. If no breast cancer is seen on other cores, then the frozen core will be used by the clinical pathologist and the woman will not be eligible for the study. If the diagnosis of cancer is confirmed, the woman will then be consented for this trial. If she consents, then the frozen core may have one section taken to confirm the presence of tumor, and the rest will be sent for GDSS-VPA. Once clinical examination of the non-frozen parts of core biopsy is completed, then study immunohistochemistry will be done on the remaining tissue in the paraffin block. If the DCE-MRI was not done prior to the biopsy, then it should be done as soon as possible after the biopsy.

Once the biopsy and DCE-MRI have been obtained, all women will receive therapy according to the following schedule.

Valproic acid 30mg/kg/day divided BID starting AM of day 1. We will assess toxicity after 2.5 days. If grade 2 side effects are present, continue at the same dose, reassessing every three days. When all side effects are grade 1 or less, then increase dose by 10 mg/kg/day every three days to a maximum of 50mg/kg/day. If a grade 3 side effect is encountered, then hold medicine until side effect resolves and restart at previous dose level for remainder of time. The highest tolerated dose will be continued until the day of surgery, which will not be before 7 days of valproic acid therapy or after 12 days of valproic acid therapy. If a dose-limiting toxicity is encountered, then the patient will be removed from the study. See appendix 18.2 for schedule based on what day of the week treatment is started.

On day 3 and 6, a valproic acid level and PWBC histone acetylation will be drawn prior to the dose increase. These labs are used for secondary endpoints and as such are optional for women for whom transportation here to obtain blood may be difficult.

The proportion of patients that experience a dose limiting toxicity is expected to be low. Dose limiting toxicity would be grade 4 confusion, grade 3 encephalopathy, grade 3 cognitive dysfunction, grade 3 somnolence, grade 3 dizziness, grade 3 tremor that does not improve with beta-blocker therapy, or any other grade 4 non-hematologic adverse event. A 2% rate of dose limiting toxicity would be considered acceptable, while a 15% rate of dose limiting toxicity would be considered unacceptable. The stopping rule will allow a maximum of one (1) patient to experience a dose limiting toxicity per eight (8) patients accrued. Patient accrual will be stopped for excessive toxicity if this level of toxicity is exceeded. More detail is provided in section 11.1. With this stopping rule, the probability of stopping the trial for excessive toxicity is 0.16 if the true rate of DLT is 2%, and the probability of stopping the trial for excessive toxicity is 0.88 if the true rate of DLT is 15%.

Between day 7 and day 12, once the subject has been on the highest tolerated dose of valproic acid treatment for at least two days, a DCE-MRI will be performed followed by surgical excision of the primary tumor per standard of care. If surgery is to be delayed for neoadjuvant therapy or other non-study related reasons, then a repeat biopsy will be performed. The last dose of valproic acid will be taken on the morning of surgery or biopsy. Following the surgery or biopsy, there will be one end of study visit with the subject to assess for any lingering toxicity. Data on subjects will then be taken from clinical appointments for 6 months after surgery to assess for relapse rates.

Inclusion Criteria:

1. Biopsy-proven invasive adenocarcinoma of the breast 1.5cm or larger by clinical exam or imaging including ultrasound, mammogram, CT, or MRI
2. Females at least 18 years-old,
3. Not pregnant, as demonstrated by a negative serum or urine pregnancy test in women of child bearing potential, and not planning on becoming pregnant
4. Willing to have a biopsy at the start of study if adequate sample for gene array is not available.
5. Willing to have a biopsy at the end of the trial if breast surgery is not planned.
6. ECOG Performance status 0-2
7. Able to provide informed consent and have signed an approved consent form that conforms to federal and institutional guidelines

Exclusion Criteria:

1. Need for immediate chemotherapy as determined by the patients' physicians, e.g., present or imminent compromise of vital organs or unacceptable symptoms from the tumor.
2. Known hypersensitivity to valproic acid or its components or peanut allergy
3. Inadequate bone marrow, kidney, and liver function (greater than grade 1 by CTCAE version 4) as defined by the protocol.
4. Immunocompromised due to medications or HIV as documented in medical history
5. Use of other antiepileptics or medications with known interactions with valproic acid (See protocol for full list)
6. Inborn errors of metabolism (valproic acid is contraindicated in patients with known urea cycle disorders)
7. History of pancreatitis
8. Use of a ketogenic diet
9. Inability to have an MRI due to extreme claustrophobia, possible metal fragments in the eye, cardiac pacemaker, implanted cardiac defibrillator, aneurysm clips, carotid artery vascular clamp, neurostimulator, insulin or infusion pump, implanted drug infusion device, bone growth/fusion stimulator, or cochlear, otologic, or ear implant
10. Tumor that is unlikely to yield adequate tissue for genomic studies in the opinion of the principle investigator