GABA induces a hormonal counter-regulatory response in subjects with long-standing type 1 diabetes

Ethics and study design

GABA is considered a pharmaceutical drug within the European Union. All participants were provided oral and written information and signed a written consent prior to inclusion in the study.

The study was an open-label, investigator-driven phase I clinical trial, primarily assessing the safety of a controlled-release formulation of GABA, Remygen (Diamyd Medical, Stockholm, Sweden), in adult male subjects with long-standing T1D. GABA was dosed in three steps: low (200 mg), medium (600 mg) and high (1200 mg) dose, with subsequent pharmacokinetic evaluation. Remygen was administered orally as a single daily dose to be taken under fasting conditions in the morning and each dose was administered for three consecutive days. The study period consisted of visits at the hospital for 11 consecutive days. Sample size was decided after dialogue with the Swedish Medical Products Agency. The study included two mixed meal tolerance tests (MMTT), four hypoglycemic clamps and three pharmacokinetic evaluations. All study visits were conducted at Uppsala University Hospital. A detailed study design and all inclusion/exclusion criteria can be found at ClinicalTrials.gov.

Safety monitoring

All subjects underwent general and neurological examinations, as well as blood sampling for laboratory testing, at multiple time points during the safety trial. The study was monitored by Uppsala Clinical Research Center. An external Data Safety Monitoring Board reviewed the data.

Laboratory testing

Laboratory tests included metabolic parameters (glucose, hemoglobin A1c (HbA1c) and C peptide), hematology status, renal function and markers of liver injury (aspartate transaminase (AST), alanine aminotransferase (ALT), alkaline phosphate, bilirubin). During the hypoglycemic clamp, the counter-regulatory hormones glucagon, growth hormone (GH), cortisol, epinephrine and norepinephrine were analyzed. Routine blood samples were analyzed at the central laboratory of Uppsala University Hospital. Epinephrine and norepinephrine concentrations were analyzed at Karolinska University Hospital laboratory. C peptide concentrations were also analyzed with an ultrasensitive C peptide ELISA (Mercodia, Uppsala, Sweden). Glycemic control was monitored by flash glucose monitoring (FGM) (FreeStyle Libre, Abbott Laboratories, Chicago, Illinois, USA).

Pharmacokinetics of GABA

GABA concentrations in plasma were determined the first day of every dose increase at 0, 30, 60, 120, 180, 300 min and 24 hours (nadir value) after GABA administration. GABA levels were analyzed by mass spectrometry at the Mass Spectrometry Based Metabolomics Facility at Uppsala University with a method approved by the US Food and Drug Administration.

Mixed meal tolerance test

An MMTT was performed after overnight fasting at baseline and at day 11, that is, last day of treatment with 1200 mg GABA. C peptide and glucose levels were measured at 0, 15, 30, 60, 90 and 120 min after ingestion of 6 mL/kg (maximum 360 mL) Resource Protein (Nestlé Health Science, Switzerland).

Hypoglycemic clamp

A hypoglycemic clamp was performed after overnight fasting at baseline and at each GABA dosage step. An individualized insulin infusion dose was calculated based on body weight (2 mIE/kg/min) and a separate glucose infusion was titrated to achieve and maintain euglycemia (5.5 mmol/L) for 30 min, followed by hypoglycemia (2.5 mmol/L) for 30 min. Blood samples were collected at the end of each plateau. GABA was administrated 30 min before start.

Endpoints

The primary outcomes of the trial were (1) number of serious and adverse events (SAE/AE) and (2) changes in laboratory parameters, physical examinations and vital signs. The secondary outcomes were (1) C peptide response during MMTT determined as area under the curve (AUC), mean and peak values; and (2) hormonal counter-regulatory response. Additional variables such as fasting plasma C peptide, HbA1c, autoantibodies and number of self-reported hypoglycemic events were included in the general assessment.

Statistical analysis

Safety laboratory tests were compared with baseline with repeated measurements one-way analysis of variance (ANOVA) with Dunnett’s post-hoc test. For FGM data a non-parametric one-way ANOVA with Dunn’s multiple comparisons test was applied. Hormonal counter-regulatory response under normoglycemic and hypoglycemic conditions was computed by multiple t-tests and corrected for multiple testing with the Holm-Sidak method. Data are presented as mean±SEM. P values <0.05 were considered statistically significant.

Characteristics of the subjects

Ten male subjects with T1D were screened, of whom four were excluded from participation in the study (C peptide >0.12 nmol/L (n=2), elevated AST/ALT (n=1) and difficult venous access (n=1)). Six subjects were included and all completed the 11-day trial. All subjects had long-standing T1D (age 24.8±1.5 years, disease duration 14.7±2.2 years). Full descriptive data are presented in table 1.

Safety evaluation

None of the patients had SAE and there were no AEs reported during treatment with the low dose, while four patients reported mild and transient AEs (mostly neurological sensations) at the medium and high dose of GABA (online supplemental table 1). Hemoglobin levels decreased when compared with baseline during the medium-dose and high-dose treatment, which was deemed to be related to the extensive blood sampling during the study (baseline 142.8±3.1 g/L, low-dose 137.5±2.6 g/L, medium-dose 131.3±2.3 and high-dose 124.8±2.5, p<0.001). Blood platelet and leukocyte counts were unaffected by GABA treatment. Also, ALT, AST, bilirubin and creatinine levels were unaffected, whereas alkaline phosphate levels slightly decreased during the high-dose treatment (data not shown). As expected, the GABA pharmacokinetics displayed increasing AUC levels at higher doses (online supplemental table 2).

Metabolic parameters

Five of the patients had undetectable C peptide levels when analyzed with the standard clinical assay (ie, <0.01 nmol/L), and both the fasting and stimulated levels were unaffected by short-term GABA treatment. When assessed by the ultrasensitive C peptide ELISA, three patients were found to have detectable C peptide under fasting conditions and one additional patient during stimulated conditions. The levels were, however, unaffected by the short-term GABA treatment (data not shown). Given the small sample size and limited study period, we could not detect any significant differences with regard to time in range, or time below range, when analyzing the FGM data (figure 1A).

Request permission

Response to hypoglycemia

The patient with remaining C peptide displayed a normal hormonal counter-regulatory response to hypoglycemia at baseline, which was maintained under the GABA treatment (data not shown). In contrast, the five patients without detectable C peptide (<0.01 nmol/L) displayed absence of counter-regulatory response at baseline (figure 1B–F). During GABA treatment (600 mg) a counter-regulatory response of glucagon, epinephrine, GH, as well as cortisol occurred in response to hypoglycemia (figure 1B–F).