ACTR-44. FEASIBILITY, PHARMACODYNAMICS, AND BIOLOGIC ACTIVITY OF THE GLIOMA ATKINS-BASED DIET (GLAD) FOR PREVENTING TUMOR RECURRENCE IN GLIOMA PATIENTS

• Published in: Neuro-oncology (Charlottesville, Va.) Va.), 2019-11, Vol.21 (Supplement_6), p.vi23-vi23
• Main Authors: Schreck, Karisa, Hsu, Fang-Chi, Berrington, Adam, Henry-Barron, Bobbie J, Blair, Lindsay, Hartman, Adam, Kossoff, Eric, Easter, Linda, Self, Mary, Whitlow, Chris, Barker, Peter, Cervenka, Mackenzie, O Blakeley, Jaishri, Strowd, Roy
• Format: Article
• Language: English
• Subjects: Adult Clinical Trials - Non-Immunologic
• ISSN: 1522-8517; 1523-5866
• DOI: 10.1093/neuonc/noz175.086

Abstract INTRODUCTION Exploiting metabolic vulnerabilities via ketosis is a promising approach for gliomas. The modified Atkins diet is a ketogenic diet therapy efficacious in adults with refractory epilepsy. We evaluated the feasibility, pharmacokinetics/pharmacodynamics(PK/PD), and cerebral activity of this dietary intervention intended to induce ketosis. METHODS 25 patients with biopsy-confirmed WHO Grade 2–4 astrocytoma with stable disease after adjuvant chemotherapy were enrolled in an 8-week GLioma Atkins-based Diet (GLAD). GLAD consisted of 2 ‘intermittent fasting’ days(IF; calories < 20% of recommended daily allowance) interleaved between 5 modified-Atkins diet days(MAD; carbohydrates < 20 gm/day) each week. The primary outcome was dietary compliance. Secondary outcomes were PK assessed by urine ketones post-FAST and post-MAD, PD assessed by serum insulin and IGF-1, and cerebral activity measured by MR spectroscopy at baseline and week 8. RESULTS Grade 2(n=2;8%), Grade 3(n=11;44%) and GBMs(n=12;48%) were enrolled. While only 48% of participants satisfied pre-defined compliance criteria, overall compliance with MAD(80%) was higher than IF(72%). Weight loss was observed (-4.8 + 2.2kg,p< 0.0001) consisting primarily of decreased fat mass (-2.5 + 3.1%,p< 0.0001), with increased lean body mass (2.4 + 3.2%,p< 0.0001), stable nutritional status (phase angle, -0.26 + 0.94%,p=0.22), and normalization of BMI. Urine acetoacetate significantly increased with 55% achieving moderate ketosis at week 8 (p=0.0005). Serum insulin and IGF-1 significantly decreased indicating systemic dietary response and were associated with higher ketones post-IF, but not post-MAD. MRS demonstrated cerebral activity with increased ketones in lesional (0.06±0.03- >0.27±0.06i.u,p=0.005) and contralateral brain at week 8 (0.041±0.01- >0.16±0.04i.u.,p=0.004). Higher cerebral acetone correlated with higher urine ketones (r >0.75,p< 0.02) and lower fasting glucose (r >-0.74,p< 0.05). CONCLUSIONS The GLAD diet was challenging to maintain but demonstrated quantifiable biologic effects systemically and intratumorally. MAD was more feasible than IF, but changes in PD markers correlated most strongly with IF. The role of ketogenic diet therapy for preventing glioma growth remains uncertain.