A randomized, double-blind, placebo-controlled, multiple-dose, parallel-group clinical trial to assess the effects of teduglutide on gastric emptying of liquids in healthy subjects

The proglucagon gene yields a single mRNA transcript expressed in the intestines, pancreas, and central nervous system (CNS) [1‐3]. Tissue-specific cleavage of the resulting proglucagon precursor protein generates several peptides with distinct biological activity [3]. In the intestines, the proglucagon-derived proteins glucagon-like peptide (GLP)-1 and GLP-2 are expressed in the enteroendocrine L cells and are secreted in response to food intake [4‐6]. The primary effects of GLP-1 are to increase postprandial insulin levels, inhibit glucagon secretion, and slow gastric emptying [7‐15], whereas the primary effects of GLP-2 are to increase growth of intestinal epithelium, maintain intestinal mucosal morphology and function, and regulate energy intake [16‐19]. Unlike GLP-1, GLP-2 has limited effect on insulin secretion or glucose homeostasis [20‐22]. Furthermore, in contrast to GLP-1, GLP-2 does not decrease pancreatic glucagon secretion [21, 23].

Multiple animal and human studies have demonstrated a delay in gastric emptying with GLP-1 or glucagon, another protein encoded by the proglucagon gene [7‐12, 24, 25]. Because GLP-1 and GLP-2 are both derived from proglucagon and secreted following nutrient stimulation, it has been hypothesized that they coregulate gastric motility [26]. Indeed, in preclinical studies, GLP-2 significantly decreased the amplitude and frequency of postprandial gastric contractions and promoted gastric muscle relaxation [26, 27]. However, conflicting results have emerged from studies on human subjects, with some reporting no effect of GLP-2 and others demonstrating a GLP-2–dependent inhibition of gastric emptying [21, 28‐30]. The divergent outcomes may have been influenced by differences between the studies, including variations in the amounts of GLP-2 delivered, the methods used to evaluate gastric emptying, and the content of test meals administered [21, 28‐30].

Teduglutide, a recombinant GLP-2 analog, is a novel therapy recently approved for the treatment of adult patients with short bowel syndrome (SBS) who are dependent on parenteral support [31, 32]. SBS is defined as a clinically significant reduction in intestinal absorptive capacity resulting from surgical resection of the intestine due to disease, trauma, congenital defects, or complications of surgery [33, 34]. Teduglutide promotes expansion of the remaining normal intestinal epithelium by increasing villus height and crypt depth in the small bowel mucosa, leading to increased absorptive area [31, 32, 35]. In a phase III, placebo-controlled trial, teduglutide significantly reduced parenteral nutrition and/or intravenous fluid (PN/IV) volume requirements and the number of infusion days required in patients with SBS [36].

Teduglutide is a synthetic protein that differs from native GLP-2 by the substitution of glycine for alanine at the second position from the N-terminus [37]. This single amino acid substitution renders it resistant to degradation by dipeptidyl peptidase-4 [37, 38]. As a result, the half-life (t_½) of teduglutide is increased compared with native GLP-2 following subcutaneous injection (t_½ of 180–330 minutes vs 60–90 minutes, respectively) [39, 40]. Because GLP-2 reduces gastric motility and delays gastric emptying, at least in animal and some human studies [26, 27, 29, 30], it is possible that teduglutide, a similar but distinct molecular entity, may also inhibit gastric emptying. Modification of gastric emptying rate may alter the bioavailability of concomitantly administered drugs and modulate drug-drug interactions [41]. Thus, given the recent approval of teduglutide, clarification of the physiological effects of this novel peptide would provide important information to the medical community.

The primary objective of this study was to assess the effect of teduglutide on gastric emptying of liquids in healthy subjects as gauged by acetaminophen pharmacokinetics (PK), an accepted measure of gastric emptying kinetics [42].

In this study, teduglutide administered at 4 mg/day for 10 days did not affect gastric emptying of liquids in healthy subjects as measured by acetaminophen PK. Previous studies evaluating gastric emptying following administration of GLP-2, a similar but distinct molecular entity, have yielded conflicting results [21, 28‐30]. Comparisons with and among these studies are hampered by heterogeneities in trial design, study populations, GLP-2 dosages, routes of GLP-2 administration, and timing of treatments. Nonetheless, results with teduglutide presented here are in agreement with 2 prior studies demonstrating no effect of infused native GLP-2 (0.75–2.25 pmol ∙ kg^–1 ∙ min^–1 for 180–390 minutes) on gastric emptying rate following a 250-to 310-kcal solid meal, as determined by either scintigraphic measurement or ¹³C-sodium octanoate breath test [21, 28]. In contrast, 2 other studies have reported delayed gastric emptying with GLP-2 [29, 30]. In one of these studies, conducted in healthy adults following a 7.5-kcal liquid meal, antral emptying time was increased by 9.5 minutes following GLP-2 (initial bolus of 4.5 pmol/kg followed by infusion at 1.0 pmol ∙ kg^–1 ∙ min^–1 for 60 minutes) compared with placebo (P = 0.049), as determined by ultrasound scanning. GLP-2 infused at a lower rate (0.5 pmol ∙ kg^–1 ∙ min^–1) after the initial bolus did not significantly affect antral emptying time [29]. Contrary to the modest results reported by Nagell et al., Jeppesen et al. described a 30-minute increase in time to 50% gastric emptying of solids following 35 days of treatment with subcutaneous GLP-2 (400 μg twice daily; P = 0.002) using a scintigraphic technique [30]. In this case, however, the study enrolled patients with SBS who may have had baseline disturbances in gastric emptying. For example, half of the patients in the Jeppesen study had an end-jejunostomy, which in a previous study was correlated with accelerated emptying of gastric liquids for patients with SBS [45]. Furthermore, end-jejunostomy patients have diminished intestinal secretion of GLP-2 [30, 46]. For these reasons, GLP-2 may have a greater effect on gastric emptying in the end-jejunostomy population.

The liquid meal provided to study subjects contained fats in addition to carbohydrates and proteins. Previous studies suggest that the lipid phase empties more slowly than the aqueous phase in the context of a mixed solid-liquid meal [47, 48]. However, when fats are homogenized with liquids before ingestion, as is the case with a liquid-only meal, both fats and the aqueous phase empty at the same rate [49]. Therefore, it is likely that the data from this study reflect simultaneous gastric emptying of both fats and the aqueous phase of the liquid meal.

Recently, rapid tachyphylaxis has been documented for GLP-1–dependent inhibition of gastric emptying in humans following prolonged infusion [50], prompting the question of whether the physiological effects of GLP-2 or teduglutide could also be affected by tachyphylaxis. Presently, however, no data are available concerning the potential for tachyphylaxis with GLP-2 or teduglutide with regard to gastric emptying. Of the 4 published studies evaluating the effects of GLP-2 on gastric emptying in humans, none provided multiple measurements of gastric emptying following GLP-2 administration [21, 28‐30]. The results presented in this study show no effect of teduglutide on gastric emptying following 10 days of subcutaneous administration; however, the study was not designed to evaluate gastric emptying rates earlier in the treatment period, or following multiple meals consumed within a period of less than 24 hours. However, several studies have demonstrated sustained biological activity of GLP-2 or teduglutide over extended treatment periods. For example, mice who received subcutaneous GLP-2 for 12 weeks showed a progressive increase in small bowel weight over the course of the study [17]. Patients with SBS treated with subcutaneous teduglutide (0.05 mg/kg/day) experienced continued reductions in PN/IV volume requirements throughout a 52-week study. Furthermore, 11 of 19 patients in the same study who were nonresponders following 24 weeks of teduglutide achieved a clinically significant response, defined as a ≥20% reduction in PN/IV volume from baseline, by Week 52 [51]. These data indirectly suggest continued improvement with long-term teduglutide treatment, with no dampening of response.

Teduglutide was well tolerated in this study. No unexpected safety signals were observed. AEs were generally mild, with no severe AEs reported. The most common AEs observed in this study were gastrointestinal related, which is consistent with the known mechanism of action of teduglutide. These results are also in agreement with clinical trials for teduglutide conducted in patients with SBS dependent on parenteral support. In two 6-month, placebo-controlled, phase III studies, gastrointestinal-related AEs were the most frequently reported class of AEs among patients receiving 0.05 mg/kg/day teduglutide [31, 36].

This study was limited by the small sample size and use of healthy volunteers. Furthermore, the methods applied in this study assess gastric emptying of liquids only. However, the appearance of ingested acetaminophen in the plasma is an established and validated measure of gastric emptying kinetics [42] and has been used extensively in the analysis of GLP-1 receptor agonists [52‐57].

The results presented here suggest that teduglutide does not act to delay liquid-phase gastric emptying in healthy subjects. Therefore, teduglutide is unlikely to modulate the bioavailability of orally administered concomitant medications through inhibition of gastric motility in this population. However, the effects of teduglutide on gastric emptying in patients with SBS remain to be investigated.