Exenatide dosing in alpacas

Glucagon like-peptide-1 (GLP-1) agonists are seeing increasing use in human and veterinary medicine (Smith & Cebra, 2009; Gilor et al., 2011; Neumiller, 2011). Actions include enhanced insulin production and glucose clearance in the face of hyperglycemia, slowed gastric emptying, and suppression of appetite. Their major indications for use are diabetes mellitus and diabetes-like conditions. Because llamas and alpacas appear to have natural diabetes-like traits which may contribute to or worsen during certain diseases processes (Cebra, 2009), and because insulin, the primary treatment for some of these disorders, may induce hypoglycemia, exenatide, a commercially available GLP-1 agonist, was tested as an alternative to insulin treatment in alpacas (Smith & Cebra, 2009). It augmented the insulin response and enhanced glucose clearance during hyperglycemia, but unfortunately induced moderate colic signs in most treated alpacas. The current study was undertaken to see if the beneficial effects could be retained without the deleterious effects at lower doses of the medication. Additionally, whether the enhanced insulin response had any effect on blood triglyceride, beta-hydroxybutyrate (BHB), and nonesterified fatty acids (NEFA) was investigated. These fat fractions often increase in camelids with disorders of energy metabolism, and hence, suppressing them would be an important goal of this medication.

Six healthy adult alpacas from the University research herd were paired up and acclimated to stalls for 24 h, during which time catheters were placed in their right jugular veins. Catheter patency was maintained throughout the trial by 3-mL injections of heparinized saline (2 USP units of heparin per milli liter) every 6 h and after each dextrose injection or sample collection. The alpacas were between 3 and 6 years of age and weighed between 68 and 91 kg. They were vaccinated annually with toxoids against the major clostridial disease and dewormed twice annually. They had not received any medications or been used in any other trials for at least 6 months prior to the current trial.

Alpacas were provided with approximately 2% of body weight in grass hay daily, split over a morning and evening feeding. The morning feeding occurred approximately 2 h before the start of each trial. Over the 4 days following catheter placement, each alpaca received an 8 AM subcutaneous injection of saline, or 0.02, 0.05, or 0.1 mcg/kg of exenatide (Byetta; Bristol-Myers Squibb, New York, NY, USA), in that order. A rapid IV bolus of 0.5 gm/kg of dextrose in a 50% solution was given 3 h later. Blood samples were collected through the catheter after discarding the first 6 mL of withdrawn fluid and collected into glass tubes containing lithium heparin. Collection times were before the morning injection, 180 min later before the glucose bolus, and 15, 30, 45, and 60 min after the glucose bolus. Plasma was separated within 15 min of collection in a refrigerated centrifuge, and samples were either analyzed immediately for glucose, BHB, and NEFA concentrations using an automated chemistry analyzer or frozen at −80 °C until analyzed 1 month later by radioimmunoassay for insulin (Coat-a-Count Insulin; Siemens, Los Angeles, CA, USA). All assays have been used for camelid studies in the past after standard validation. Alpacas were monitored during and for 48 h after the trial for signs of adverse reactions. If colic signs were noted, alpacas were treated and removed from the remaining days of the trial. At the conclusion, jugular catheters were removed, and the alpacas returned to the herd. All animal procedures were conducted with the approval of the Institutional Animal Care and Use Committee.

Plasma glucose, triglyceride, BHB, NEFA, and insulin concentrations were each analyzed for changes over time and the effects of the exenatide dose using two-way anova for repeated measures. Values of P < 0.05 were considered significant.

None of the alpacas displayed any evidence of an adverse reaction at the lowest dose, and four of them also tolerated the higher doses well. Two developed mild colic signs including lack of hay intake or cud chewing, preferentially remaining in sternal recumbency, and mild muscle fasciculation within 4 h of receiving the 0.05 mcg/kg dose. They were treated with intravenous flunixin meglumine (1.1 mg/kg) and signs resolved within 3 h. These two alpacas did not receive the highest dose, so all comparisons involving the 0.1 mcg/kg doses relate only to the remaining four alpacas.

Plasma glucose concentrations remained significantly higher than baseline for all time points after the glucose bolus, but were significantly lower than control values for the last two time points on all treatment days (Table 1). There were no significant differences between treatment days.

Plasma triglyceride samples did not change significantly over the course of the study. Plasma NEFA and BHB values had a slight, insignificant dose-dependent increase or lack of decrease, followed by a significant decrease in all groups. There were no significant differences between treatment days.

Plasma insulin concentrations did not change significantly on the control day. On the three treatment days, there was a significant increase that lasted 1 time point after the glucose bolus on the low-dose day and the duration of the study in the medium- and high-dose days. Insulin concentrations were not significantly different between the high- and medium-dose days, but these were both significantly greater than on the low-dose day for all postglucose time points.

This study was designed to evaluate lower doses of exenatide as a possible treatment for disorders of energy metabolism in alpacas. In general, goals of that treatment are to increase blood insulin and lower blood glucose and fat fractions, preferably without adverse effects. Alpacas are similar in size to adult humans, and the highest dose approximated that recommended for human adults (5–10 mcg). Dosing based upon peak serum concentrations (Cp) is supported by interspecies modeling data which support some consistency and predictability of the weight-normalized volume of distribution of exenatide between species (Chen et al., 2013).

Unfortunately, two of the six alpacas developed colic signs at the medium dose, although these were much milder and more responsive to treatment than those seen in alpacas at a higher dose (0.2 mcg/kg; Smith and Cebra, 2009). Such complications are uncommon in healthy humans receiving the lower dosage (5 mcg), but common in those receiving the higher one (10 mcg; Shi et al., 2012). From a clinical standpoint, the signs seen were not severe enough to preclude any use of this medication, but were common and noticeable enough to require that any initial use be carefully monitored.

The different dosages were given in an ascending order in case there was any accumulation of medication and also to more closely mimic a clinical situation of daily medication. As such, it was possible, but thought to be unlikely, that adverse events and performance of the medication were influenced by drug accumulation. Nothing is known of the pharmacokinetic properties of exenatide in alpacas. In monkeys and humans, the half-life after subcutaneous administration is under 3 h (Ai et al., 2008; Neumiller, 2011), and standard treatment protocols include twice-a-day administration with the 2 doses coming as little as 6 h apart. With daily dosing at 10 mcg, monkeys have no detectable accumulation. In cats receiving 1 mcg/kg subcutaneously, exenatide is undetectable in serum within 12 h (Gilor et al., 2011). Thus, although some small amount may have been carried over from day to day, the effect of that residual medication was unlikely to have affected the results more than would be expected if following a daily dosing schedule.

Reactions aside, the 0.05 mcg/kg dose was clearly superior to the lower dose in insulin stimulation and also appeared to offer most of the advantages of the higher dose. None of the dosages used here led to as high insulin concentrations as were seen with the 0.2 mcg/kg dose, but all resulted in similar glucose tolerance. It is likely that camelids, having partial insulin resistance, are able to upregulate glucose uptake only to a certain extent, and it appears that is maximized even at the 0.05 mcg/kg dose. Suppression of fat mobilization is thought to occur at even lower insulin concentrations and, based on the current study, could be achieved to some degree even without exenatide.

Although the changes were not significant, it was interesting to note that these alpacas had a slight increase in fat mobilization concurrent with the slight decrease in glucose and insulin concentrations after exenatide and prior to the glucose bolus. Exenatide is known to reduce fasting glucose and insulin in diabetic humans (Cersosimo et al., 2011), but reports of effects on the fasting lipid profile are scarce. The increases in fat fractions may represent the precarious state of insulin sufficiency thought to exist in healthy camelids (Cebra, 2009). Blood insulin is normally just high enough to suppress fat mobilization and ketogenesis, but even small reductions can relax this suppression. Although the changes were small, they might have been larger in camelids already experiencing fat mobilization and thus highlight the need to monitor the lipid profile or at least to induce hyperglycemia in a timely fashion after exenatide administration.

It is also possible that the slight initial increase in NEFA with the higher doses of exenatide relates to the cumulative effects of heparinized saline administration over the course of the study. The heparin dose used to maximize this effect in human patients in some studies is 250 U/h for 4–48 h, often with concurrent lipid administration (Carpentier et al., 2001, 2005). This dose is far higher than the small amount of heparin used in the current study to maintain catheter patency, but an effect cannot be completely excluded.

This model was not ideal for studying the effects of exenatide on fat mobilization. The alpacas were not fasted and thus had fairly low resting levels of all measured lipid compound. As most of the changes in fat substrates could be attributed to changes in plasma insulin, and concurrent exenatide and glucose clearly increased plasma insulin, it is likely that the fat-reducing effects of exenatide would resemble the fat-reducing effects of insulin in camelids with lipemia, hyperketonemia, or high NEFA concentrations (Waitt & Cebra, 2008), although the extent and duration of those effects has yet to be determined. Additionally, because exenatide relies on the endogenous pancreatic insulin response to complete most of its actions, it may be less therapeutic in camelids with pancreatic dysfunction.