Ketogenic diet propels cancer therapy forward

Naveed Saleh, MD, MS, for MDLinx | August 16, 2018

A ketogenic diet may be the key to success for PI3K inhibitors, according to a new study published in Nature. Mutations in PI3K, or phosphatidylinositol-3 kinase, are linked to many types of cancer.


The authors posited that a ketogenic diet could improve clinical tumor response to PI3K inhibitors by lowering serum insulin levels and reducing the ability of insulin to trigger receptors in tumors.

“Pharmacological inhibition of PI3K has resulted in variable clinical responses, raising the possibility of an inherent mechanism of resistance to treatment,” wrote authors, led by Benjamin D. Hopkins, PhD, Meyer Cancer Center, Weill Cornell Medicine, New York.

PI3K inhibitors suppress the PI3K pathway, a cell-signaling pathway that goes into overdrive in many types of cancer. Importantly, this pathway is initiated by rising insulin levels. Although 20-plus PI3K inhibitors have been tested in clinical trials, only two have been approved.

When administered, PI3K inhibitors usually cause only a transient hyperglycemia, which is followed by systemic hyperinsulinemia as the pancreas rushes to maintain homeostasis. Some patients, however, have to discontinue taking PI3K inhibitors because their blood glucose levels do not return to normal.

The authors hypothesized that PI3K-induced insulin feedback reactivates the PI3K-mTOR signaling axis in tumors, thus making the drugs less effective.

In this study, the team administered PI3K inhibitors to wild-type mice. As expected, this intervention resulted in a transient hyperglycemia followed by a spike in insulin.

The researchers then examined whether these PI3K-induced spikes in glucose and insulin influenced tumor growth by examining mouse models with Kras-Tp53-Pdx-Cre (KPC) tumor allografts at the level of the pancreas. They observed via fluorodeoxyglucose positron emission tomography (FDG-PET) that KPC tumors started taking in glucose after 90 minutes of PI3K inhibitor treatment.

The addition of insulin to KPC cells in vitro could rescue PI3K signaling, which leads to tumor cell proliferation.

“These observations support the conclusions that insulin is a potent activator of PI3K signaling in certain tumors,” wrote the authors, “and that elevation of serum insulin following administration of a PI3K inhibitor can reactivate PI3K signaling and potentially other PI3K-independent responses to insulin in both normal tissues and tumors.”

Dr. Cantley and colleagues then set out to find ways of improving PI3K inhibitor treatments by avoiding acute glucose-insulin feedback. For 10 days, the researchers pretreated KPC tumor mice with either drugs (metformin/SGLT2 inhibitor) or a ketogenic diet. The mice were then treated with the PI3K inhibitor BKM120 after 10 days.

“The rationale for using a ketogenic diet,” the authors wrote, “was to deplete hepatic glycogen stores and thereby limit the acute release of glucose from the liver that occurs following PI3K inhibition.”

The investigators found that pretreatment with metformin minimally affected serum increases in glucose and insulin, as well as mTORC1 growth signaling. However, pretreatment with SGLT2 inhibitor and ketogenic diet both decreased the release of insulin and reduced mTORC1 signaling, or tumor shrinkage, in mice. Moreover, the researchers noted similar effects in mouse models treated with the p110α-specific inhibitor BYL719.

Many hormones and metabolites can retrigger growth in the cellular milieu of PI3K inhibition. Hence, the researchers examined whether the enhancement in signaling, or cell proliferation, was mediated by insulin.

They created a doxycycline-inducible short hairpin RNA (shRNA) that targeted the insulin receptor located in KPC tumors. Hairpin induction without PI3K inhibitor displayed virtually no effect, whereas hairpin induction after starting BYL719 treatment yielded tumor shrinkage akin to the ketogenic diet.

“This result supports a model in which the insulin receptor does not have a major role in tumour growth until supra-physiological amounts of insulin are released following treatment with a PI3K inhibitor,” the authors wrote.

The authors posited that a ketogenic diet could improve clinical tumor response to PI3K inhibitors by lowering serum insulin levels and reducing the ability of insulin to trigger insulin receptors in tumors.

The researchers also found that ketogenic diet improved the efficacy of a variety of PI3K inhibitors. Furthermore, insulin feedback saps the efficacy PI3K inhibitors in a number of tumor models.

“Any drug that targets PI3K may not be effective unless patients can maintain low blood sugar levels through diet or medication,” said Lewis C. Cantley, PhD, the Meyer Director of the Sandra and Edward Meyer Cancer Center at Weill Cornell Medicine, and one of the authors of the study. “We demonstrated that if we keep insulin down with the ketogenic diet, it dramatically improves the effectiveness of these cancer drugs.”

Looking forward, the investigators plan to test whether adding a ketogenic diet to treatment with FDA-approved PI3K inhibitors betters outcomes in patients with breast cancers, endometrial cancers, leukemias, and lymphomas.

“In light of these results,” the authors concluded, “it may also be important to think about how common clinical practices such as intravenous glucose administration, glucocorticoid use, or providing patients with glucose-laden nutritional supplements may impact therapeutic responses. As therapeutic agents that target this critical oncogenic pathway are brought through clinical trials, they should be paired with strategies such as SGLT2 inhibition or the ketogenic diet to limit this self-defeating systemic feedback.”