The first weekly KD meal package will be given to the patient at the study site. The patient will be instructed in how to process the week-long meal plan package content. Participants will measure urine ketone bodies with Ketostix 2x day and blood for glucose and ketone levels using self-administered Precision Xtra® Meter (Abbot Diabetes Care, Alameda, CA, USA) starting with the first day of the diet. Self-administered blood checks for glucose and ketone bodies will be done in fasted state in the morning and 2 hours post-prandially in the evening. Participants will be seen on day 7 of treatment by the study nutritionist and a study investigator-physician (separately) to review possible AEs, and for further education about the diet. Study staff will review and evaluate the participant's method of urine ketone and blood glucose and ketone levels testing. Subsequently, participants will be seen at one, 2 weeks, and 4 weeks after KG diet initiation, and then monthly. KD treatment will last until exit criteria are met or for 6 months, whichever comes first. Exit criteria are the primary outcome measures, the first of either (a) cerebral edema requiring steroid rescue therapy or (b) death.

Treatment will occur in outpatient office setting at the Mid-Atlantic Epilepsy and Sleep Center, Bethesda, MD. The location of subsequent treatment administration may change to hospital setting at Holy Cross Hospital should a patient need hospitalization during the study for any reason, as determined by the patient's clinical care needs.

Malignant gliomas are the most common type of brain tumor in adults. They are the second leading cause of cancer mortality in people under the age of 35 and the fourth leading cause in those under the age of 54. Standard therapy for glioblastoma multiforme (GBM) includes surgery followed by radiation and chemotherapy. Despite optimal treatment the prognosis remains poor. Patients with GBM have a median survival of approximately 10-15 months. Essentially all patients suffer recurrent disease, usually within 8 months of diagnosis. For patients with malignant glioma whose tumors recur, the median time to tumor progression is 9-13 weeks with the current standard of treatment for recurrent glioblastoma, bevacizumab (AVASTIN®). There is an urgent need for more effective therapies.

GBM cancer cells depend on glucose for energy supply and survival. When glucose levels are reduced, normal brain cells metabolize ketone bodies for energy. Brain tumor cells cannot do so. This makes the tumor cells vulnerable to death using therapies that target glucose metabolism. In brain cancer patients, high blood glucose levels are associated with reduced survival. In mice, reduction in circulating glucose levels through ketogenic diet (KD) reduces tumor growth

Ketogenic diet (KD) reduces blood glucose levels while elevating levels of ketone bodies, and may thus be beneficial in the treatment of GBM as a non-toxic metabolic therapy. KD is a high fat, low carbohydrate diet used for treatment of refractory seizures. There have been two case reports of KD use in patients with GBM showing slowing of tumor progression.

Ketogenic diet (KD) is a high fat, low protein, low carbohydrate diet that is an effective treatment of refractory epilepsy in children. The diet consists of long chain saturated triglycerides with a 3:1 or 4:1 [fat] : [protein + carbohydrate] ratio by weight, with 87-90% of calories derived from fat. The investigators have recently evaluated KD with caloric restriction of 1600 kcal /day in adults with refractory epilepsy. The diet was well tolerated. Side effects included mild nausea (n=2), mild diarrhea (n= 2), and mild constipation (n=2). No subjects stopped treatment because of side effects. Mild transient hunger was experienced by 3 subjects but there was no clinically significant hunger.

The goal of the present study is to evaluate efficacy, safety and tolerability of 4:1 KD in a small number of patients with GBM progressing or recurring after standard treatment of surgery, radiation and chemotherapy with temozolamide and, after further rescue threrapy with bevacizumab (AVASTIN®), as the first step in evaluation of therapeutic potential of KD in the treatment of GBM. This will be a small, open label pilot study. 6 patients with GBM recurring or progressing after surgery, radiation and chemotherapy (temazolamide) and after further rescue treatment with bevacizumab will be treated with 4:1 [fat]:[protein+carbohydrate] ratio, 1600 kcal/day diet. Survival, time to steroid rescue treatment and progression free survival will be compared to historical controls with life expectancy of ≤ 3 months.

The diet (KD) will consist of pre-made meals, including breakfast, lunch, dinner and two snacks, one each between breakfast and lunch and lunch and dinner, with a 2 week-long meal plan consisting of different recipes for each day of the month, with repeating 2-week-long cycles. The diet will be supplemented with vitamins, calcium and phosphorus supplements to meet the requirements of US Dietary Reference Intakes (DRI) standard. Treatment will last 6 months. Primary aims of the study will be to (1) obtain pilot data on efficacy of ketogenic diet as adjunctive treatment of treatment-refractory glioblastoma multiforme, (2) evaluate the safety of ketogenic diet as adjunctive treatment of treatment-refractory glioblastoma multiforme, and (3) evaluate tolerability of ketogenic diet as adjunctive treatment of treatment-refractory glioblastoma multiforme.

Participants will be evaluated in face-to face visit during weeks 1, 2 and 4 of the study, then monthly. Primary outcome measures will include (a) survival time and (b) time to dexamethasone rescue therapy for cerebral edema, determined by treating physician, and (c) adverse events. Secondary outcome measures will include MRI progression, treatment compliance, hunger scale scores, fasting serum glucose and BOH levels.

Inclusion Criteria:

1. Age 18-65
2. Ability and willingness to signed informed consent form.
3. Astrocytoma grade 4 (glioblastoma multiforme, GBM, histologically confirmed, WHO criteria)
4. Documented recurrence or progression after surgical resection/debulking, radiation and temozolamide chemotherapy.
5. Measurable contrast-enhancing progressive or recurrent GBM by MRI imaging ≤ two weeks before screening.
6. (a) ≥ 3months after completion of radiation; (b) 6 weeks from a nitrosourea chemotherapy; (c) ≥ weeks from a non-nitrosourea chemotherapy (all [a-c] in order to allow recovery from the potential of severe toxicity related to these treatments)
7. Karnovsky Performance Score of 70 or more.

Exclusion Criteria:

1. Acute intracranial or intratumoral hemorrhage > Grade 1 either by MRI or CT scan ≤2 weeks of screening. (Subjects with resolving hemorrhage changes, punctate hemorrhage, or hemosiderin may enter the study)

2. Prior treatment with any of the following: (a) small-molecule kinase inhibitor; (b) non-cytotoxic hormonal agent; (c) KD ≤6 months of enrollment

3. Planned continued use of glucocorticoids

4. Anticoagulation treatment with ≥ 1 mg/day coumadin ≤ 7 days prior to screening (low-dose [≤ 1 mg/day] coumadin, heparin, and low-molecular-weight heparin are permitted

5. Any systemic illness or unstable medical condition that might pose additional risk, including: cardiac, unstable metabolic or endocrine disturbances, renal or liver disease, past history of renal calculi, hyperuricemia, hypercalcemia, mitochondrial disease, known disorder of fatty acid metabolism, porphyria, carnitine deficiency and pancreatitis

6. History of non-glioma malignancy other than:

   • Surgically excised non-melanoma skin cancer or in situ carcinoma of the cervix.
   • A malignancy diagnosed ≥2 years ago if the subject has had no evidence of disease for 2 years prior to screening.

7. History of uncontrolled hyperlipidemia.

8. Active drug or alcohol dependence or any other factors that, in the opinion of the site investigators would interfere with adherence to study requirements

9. History of human immunodeficiency virus, or hepatitis C

10. Failure to recover from \