ABSTRACT: Objectives: To evaluate the feasibility of a very low carbohydrate, ketogenic, diet in women with endometrial cancer and obesity, and to assess the effects of this dietary pattern on metabolism and tumor biology. Design: A multicentre, parallel, prospective, randomized, controlled feasibility study in treatment naïve women with endometrial cancer. Participants were randomized in a 2:1 fashion to a very-low carbohydrate (VLCD) diet, or to continue with their standard diet (SD) by the method of random permuted block. Setting: Eight locations from two teaching hospitals in the New York and New Jersey areas. Participants: 19 treatment naïve women with endometrial cancer (stages IA-IIIC2) and obesity. 15 completed the study. Intervention: After initial biopsy and pathology confirmation, participants received 21-28 days of VLCD and weekly nutritional counseling, followed by surgical staging. Main outcome measures: Dietary adherence, rates of attrition, and adverse effects. Changes in circulating concentration of metabolic biomarkers. Changes in immunohistochemical staining and RNA-Sequencing for insulin signaling pathway, cell apoptosis, proliferation, and immunity. Results: Of the 19 participants randomized, 15 finished the trial with an overall attrition rate of 21%. Participants in the VLCD arm reported consuming 91 ± 4% of the meals, leading to a rise in ketone bodies for all participants by day 3 or 4 and an overall weight loss of 5.5 ± 0.8%. The VLCD was well-tolerated with no grade 3/4 adverse events. Fasting plasma glucose and insulin levels were reduced by 22 ± 5.9% and 60 ± 3.8%, respectively. Total cholesterol and low-density lipoprotein (LDL) were increased by 6 ± 2.7% and 17.8 ± 8.9%, respectively. Gene Set Enrichment Analysis using RNA-Seq revealed enrichment in pathways related to fatty acid oxidation, oxidative phosphorylation, estrogen response, interferon-alpha response and CD8+ T cells (p < 0.05 and NES > 1.60). CD8+ T cells were found to infiltrate the tumor margins under VLCD compared to SD (p = 0.019). Conclusions: A VLCD is feasible and well-tolerated in obese and overweight women with endometrial cancer and leads to significant host and tumor metabolic changes that suggest potential therapeutic benefits that warrant further investigation in larger clinical trials.
