A phase II clinical and biological study of lithium carbonate (Li) in patients with low-grade neuroendocrine tumors

Abstract

e15662 Background: Low-grade neuroendocrine tumors (NETs), such as carcinoid, islet cell tumors, and medullary thyroid carcinomas, respond poorly to chemotherapy, and effective, less toxic therapies are needed. Glycogen Synthase Kinase (GSK)-3β, a multifunctional protein kinase, has been shown to regulate growth and hormone production in NETs. Use of lithium carbonate (Li) in murine models suppressed carcinoid cell growth, reduced GSK-3β levels and reduced expression of chromogranin A. This study assessed the efficacy of Li in patients with NETs. Methods: Eligible pts had pathologically-proven, measurable low-grade NETs. Prior treatment was allowed if completed >4wks prior to registration. Standard eligibility criteria were used, and use of medications affecting Li metabolism or levels were prohibited. A single-arm, open-label, two-stage Phase II design was used. Li was dosed at 300mg orally TID with meals, titrated to a target serum level of 0.8–1.0mmol/L. The primary endpoint was objective tumor response by RECIST. Secondary endpoints included overall survival, progression-free survival, decrease of serum tumor markers, toxicity, and quality of life. Results: 15 pts were enrolled between 10/3/07and 7/17/08; 6 men, 9 women. The median age was 58 (range 47–74). Patients’ diagnoses were carcinoid tumor for 8 subjects, islet cell tumor for 5 subjects, and 2 unknown primary sites. ECOG PS was 0 (6 patients) or 1 (9 patients). Two pts came off study due to side effects (tremor, dizziness/abdominal pain). There were no radiographic responses. Due to an early stopping rule requiring at least 1 objective response in the first 13 evaluable pts, the study was closed to further accrual. 13 patients had pre- and post-therapy biopsies. Evaluation of quality of life and GSK-3β levels in tumor tissue is ongoing. Conclusions: Li was ineffective at obtaining a radiographic response in our 13 evaluable patients who were treated as part of this study. We will determine from tumor biopsies whether Li was effective at phosphorylating GSK-3β in order to make conclusions about GSK-3β as a therapeutic target for future NET treatment strategies. Funded through NIH grant R21CA117117- 01A2 and CTRC grant 1ULRR025011. No significant financial relationships to disclose.