Results:  Recruitment was slow, resulting in early termination of the study. Eight patients participated, 4 in group 1

and 4 in group 2. Five (62.5%) subjects completed the 6 months of treatment, 4/4 subjects in group 1 and 1/4 in

group 2. Three subjects stopped KD early: 2 (25%) because of GBM progression and one (12.5%) because of diet

restrictiveness. Four subjects, all group 1, continued KD on their own, three until shortly before death, for total of

26, 19.3, and 7 months, one ongoing. The diet was well tolerated. TEAEs, all mild and transient, included weight loss

and hunger (n = 6) which resolved with caloric increase, nausea (n = 2), dizziness (n = 2), fatigue, and constipation

(n = 1 each). No one discontinued KD because of TEAEs. Seven patients died. For these, mean (range) survival time

from diet initiation was 20 months for group 1 (9.5–27) and 12.8 months for group 2 (6.3–19.9). Mean survival time

from diagnosis was 21.8 months for group 1 (11–29.2) and 25.4 months for group 2 ( 13.9–38.7). One patient with

recurrent GBM and progression on bevacizumab experienced a remarkable symptom reversal, tumor shrinkage, and

edema resolution 6–8 weeks after KD initiation and survival for 20 months after starting KD.

Conclusion:  Treatment of GBM patients with 4:1 KD using total meal replacement program with standardized

recipes was well tolerated. The small sample size limits efficacy conclusions.

Study design

This was an open label phase 1 study treatment of adults

with GBM for 6 months with 4:1 [fat]:[protein + carbohydrate] ratio, 1600 kcal/day diet. Treatment was initiated either early in the disease, with initiation of

radiation and temozolomide therapy (group 1) or following recurrence (group 2). Inclusion criteria, other than

disease stage, evaluations, and treatment protocols were

identical for both groups. Primary outcomes were feasibility, safety, and tolerability; secondary outcome was efficacy. Primary outcome measures included, for

feasibility, (1) retention in the study; for safety, (2) treatment emergent adverse events (TEAEs); and (3) treatment discontinuation because of TEAEs. Secondary

outcome measures were, for efficacy, (4) overall survival

time from treatment initiation and (5) time to MRI progression. Other outcome measures included treatment

compliance, hunger scale scores, fasting serum glucose,

and beta-hydroxy butyrate (BHB) levels and urine ketone

levels.

Subjects

Subjects were men and women aged 18–65 with histologically confirmed GBM of either early stage (after initial surgery/biopsy) or late stage (recurrence or

progression after radiation and temozolomide treatment). Patients in group 2 had to have measurable

contrast-enhancing progressive or recurrent GBM by

MRI imaging. Exclusion criteria included Karnofsky Performance Score < 70, anticoagulation treatment with

coumadin ≥ 1 mg/day, history of non-glioma malignancy

within 2 years, history of uncontrolled hyperlipidemia,

renal calculi, hyperuricemia, mitochondrial disease, disorders of fatty acid metabolism, porphyria, carnitine deficiency, pancreatitis, and presence of any other unstable illness.

Evaluations

Screening

Pathology of the GBM was confirmed by neuropathology

review. Patients with recurrent GBM had to have MRIdocumented tumor progression or recurrence. Baseline

laboratory studies included serum electrolytes, renal and

liver functions, CBC, PT/PTT, fasting blood glucose

(FPG), and serum lipid profile (cholesterol, triglycerides,

high-, low-density lipoprotein, [HDL, LDL]), uric acid

levels, and serum BHB.

The subject’s known food allergies and special (e.g., religious) dietary requirements were reviewed. Participants

were taught to measure urine ketone body (KB) levels

using Ketostix (Bayer AG, Germany) which measures

acetoacetate, and blood for glucose and ketone levels

using self-administered Precision Xtra® Meter (Abbot

Diabetes Care, Alameda, CA, USA) which measures

BHB. These were done fasted in the morning and 2 h

post-prandially in the evening. Subjects were instructed

to keep urine ketone/blood glucose and ketone diary.

Subsequent evaluations

Subsequent evaluations included face-to-face visits on

treatment days 7, 14, and 28 to review possible early

AEs, and for further education about the diet, then

monthly for the 6 months of treatment and posttreatment months 6–12, and quarterly afterward, the latter either face-to-face or by telephone. With 2 subjects,

some visits occurred via Skype because of the subjects’

long distance from the site. During each visit AEs, Karnofsky Performance Score, treatment compliance, issues

with KD, urine ketone body, and blood ketone and glucose diaries were reviewed. Hunger was evaluated using

7-point Likert scale (no hunger–extremely hungry). For

patients with > 5% BMI loss, caloric restriction was

stopped. Caloric supplementation required to remain weight-neutral was calculated, with instructions to

add extra calories using 100% fat-containing calories

such as olive oil or 100% fat dairy produce.

Baseline laboratory evaluations were repeated at treatment months 1, 2, 3, 4, and 6. Blood was drawn at 8 am,

following an 8 h fast. MRI of the brain was performed

before starting treatment (baseline) and at treatment

months 2, 4, and 6, and every 2–3 months after that.

Treatment diet

KD consisted of 4:1 [fat]:[protein + carbohydrate] ratio

by weight, with 10 g CH/day, and with 1600 kcal restriction. We chose 4:1 KD because the animal KD study

with the greatest treatment effect to date used 4:1 KD

[22]. Patients who did not tolerate the 4:1 ratio could choose 3:1 ratio with 20 g CH/day. The diet was supplemented with vitamins, calcium, and phosphorus supplements to meet the requirements of US Dietary

Reference Intakes (DRI) standard. The program consisted of 5 meals/day (breakfast, morning snack, lunch,

afternoon snack, dinner), different for each day of a 2-

week cycle, with repeating cycles. All meals were prepared using designed recipes (Anemone LLC, Bethesda,

MD). All participants received the same meal plan but

with recipe adaptation to allow personal or religious

dietary restrictions (vegetarian, n = 1, no pork, n = 1)

with the same caloric and macronutrient composition.

Meals were prepared uniformly by one catering facility

and were delivered frozen once a week. Participants were

counseled not to eat any other food or and drink only 0

calorie beverages. One patient administered the diet on

his own after the first 2 months, using the same KD parameters. The food was provided free by Anemone LLC.

Medication adjustment

For subjects on steroids (n = 5), attempts were made to

taper off steroids as quickly as clinically feasible.

Compliance

Compliance with the diet was evaluated at each visit by

reviewing patients’ consumption of food supplied by the

study, extra food consumed instead of or in addition to

study food, subjects’ urine and blood ketone diaries, and

monthly serum β-hydroxybutyrate levels. It was scored as

a composite of these factors on a 0–3 scale as 3 =

complete compliance, 2 = partial, substantial compliance, 1 = partial, slight compliance, and 0 = complete noncompliance.

Statistical analysis

Only descriptive statistics were used because of the small

sample size.