Trial Information

A Low-Carbohydrate Diet for Advanced or Metastatic Cancer

In their natural state within the body, cells normally depend on cellular respiration as the
source of energy and ATP. During times of complete or partial food deprivation these cells
can also rely on ketosis as an additional energy source. Tumor cells in contrast are
defective in this respect, lacking in the ability to utilize ketone bodies for energy and
instead are heavily dependent on glycolysis for their survival. Malignant cells depend
heavily on glycolysis for energy production even in high oxygen states. If the
investigators limit the supply of glucose, glycolysis would decrease and tumor cells would
lose their main source of energy supply.

In addition, increased carbohydrate intake is sensed by the pancreas and results in
increased insulin secretion. Insulin is important for normal growth in the body and is a
potent growth factor stimulating mitosis via signal transduction and DNA synthesis. It is a
potent growth factor which acts by binding to cell membrane insulin receptors and stimulates
mitosis via protein kinase B/Akt- mediated signal transduction and DNA synthesis.

Insulin also has anti-apoptotic properties, meaning less tumor cell death. Increased
circulating insulin levels have been associated with increasing risk of certain cancers
(i.e. prostate and colon cancer), but no association with serum glucose levels (8). By
limiting carbohydrate intake, the investigators might expect decreased tumor growth and
increased tumor cell destruction.

Most human and mouse tumors take up more glucose than their surrounding normal tissue. By
using positron emission tomography with fluoro-deoxy-glucose, FDG-PET scans can be used to
detect and monitor tumor response to chemotherapy and radiation.

In this study, PET/CT scanning would be utilized to measure the amount of (at the screening
visit) glucose activity within certain tumors, before and after the diet.

Primary Aim: To determine the safety and tolerability of a modified low carbohydrate diet in
people with advanced cancer across different tumor types.

Hypothesi(e)s: The long term tolerability of the KD (Ketogenic Diet) has been established
(Groesbeck et al., 2006). Recent studies involving human patients with brain cancer showed
tolerability of the Ketogenic diet over a period as long as 19 months with minimal side
effects. However the MAD (Modified Atkins Diet) long term side effects are as of yet
unclear. Given the similarity of the MAD compared to the KD, and being less restrictive, the
long term side effects (kidney stones, dyslipidemia, gastroesophageal reflux) of the MAD is
projected to be similar, if not reduced, in comparison to the KD. Modified carbohydrate
diets as well as ketogenic diets have been tolerated by people with epilepsy. It is
hypothesized that the effect this diet will have on overall weight loss, hyperlipidemia, and
blood glucose levels will be minimal and tolerable even by cancer patients over a prolonged
period of time, up to 12 months or possibly longer. Given the tolerability of the diet in
brain cancer patients, it is hypothesized that patients with other types of tumors will find
this diet tolerable and safe.

Secondary Aims: To determine if quality of life is improved with the use of this diet. To
determine if the diet has any effect on tumor growth (size or spread) or to determine if
there is any effect on progression free survival (PFS) and overall survival (OS) of subjects
enrolled in the study.

Hypothesis(e): The diet showed good tolerance in human adult patient studies wherein the
diet was tolerated well up to 12 weeks in a group of diverse tumor types and up to more than
10 years in a glioma patient. It is hypothesized that cancer patients will be able to
tolerate this diet at 16 weeks or more. Studies in animals as well as in tumor tissue models
show slowing of tumor growth, and in some instances also showed tumor shrinkage. Human case
reports and a small study of 16 patients show tumor shrinkage and stable disease while on
the diet.