- Email: [email protected]
Pharmacokinetics and pharmacodynamics of the glucagon-like peptide-1 analog liraglutide in healthy cats
Domestic Animal Endocrinology 51 (2015) 114–121
Contents lists available at ScienceDirect
Domestic Animal Endocrinology journal homepage: www.domesticanimalendo.com
Pharmacokinetics and pharmacodynamics of the glucagon-like peptide-1 analog liraglutide in healthy cats M.J. Hall a, C.A. Adin a, S. Borin-Crivellenti a, b, A.J. Rudinsky a, P. Rajala-Schultz c, J. Lakritz a, C. Gilor a, * a b c
Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, OH 43210, USA FCAV/Universidade Estadual Paulista (UNESP), Jaboticabal, SP, Brazil Department of Veterinary Preventative Medicine, The Ohio State University, Columbus, OH 43210, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 July 2014 Received in revised form 1 December 2014 Accepted 2 December 2014
Glucagon-like peptide-1 (GLP-1) is an intestinal hormone that induces glucose-dependent stimulation of insulin secretion while suppressing glucagon secretion. Glucagon-like peptide-1 also increases beta cell mass and satiation while decelerating gastric emptying. Liraglutide is a fatty-acid derivative of GLP-1 with a protracted pharmacokinetic profile that is used in people for treatment of type II diabetes mellitus and obesity. The aim of this study was to determine the pharmacokinetics and pharmacodynamics of liraglutide in healthy cats. Hyperglycemic clamps were performed on days 0 (HGC) and 14 (LgHGC) in 7 healthy cats. Liraglutide was administered subcutaneously (0.6 mg/cat) once daily on days 8 through 14. Compared with the HGC (mean standard deviation; 455.5 115.8 ng/L), insulin concentrations during LgHGC were increased (760.8 350.7 ng/L; P ¼ 0.0022), glucagon concentrations decreased (0.66 0.4 pmol/L during HGC vs 0.5 0.4 pmol/L during LgHGC; P ¼ 0.0089), and there was a trend toward an increased total glucose infused (median [range] ¼ 1.61 (1.11–2.54) g/kg and 2.25 (1.64–3.10) g/kg, respectively; P ¼ 0.087). Appetite reduction and decreased body weight (9% 3%; P ¼ 0.006) were observed in all cats. Liraglutide has similar effects and pharmacokinetics profile in cats to those reported in people. With a half-life of approximately 12 h, once daily dosing might be feasible; however, significant effects on appetite and weight loss may necessitate dosage or dosing frequency reductions. Further investigation of liraglutide in diabetic cats and overweight cats is warranted. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Feline Diabetes Insulin Glucagon Incretin
1. Introduction Feline diabetes mellitus (FDM) closely resembles type 2 diabetes mellitus in people, as both FDM and type II diabetes mellitus (T2DM) are characterized by insulin resistance, relative insulin deficiency, decreased beta cell mass, and islet-amyloid depositions [1,2]. Despite technological advances in insulin formulations and monitoring, insulin therapy across species is still
* Corresponding author. Tel.: þ1 6142923551; fax: þ1 6142921454. E-mail address: [email protected] (C. Gilor). 0739-7240/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.domaniend.2014.12.001
commonly associated with inadequate glycemic control leading to short- and long-term complications such as hypoglycemia, weight gain, and vascular diseases. Incretinbased therapy for T2DM was first introduced in 2005 and has quickly become a widely used adjunctive therapy. The incretin effect is the greater release of insulin occurring after oral glucose administration compared with intravenous administration of the same glucose dose. Two peptide hormones contribute to the incretin effect and are called incretin hormones: glucagon-like peptide-1 and glucosedependent insulinotropic peptide (GIP). They are secreted from specialized enteroendocrine cells in response to the presence of nutrients in the lumen of the gut (but not in
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
response to the presence of nutrients in the blood). By enhancing the response of pancreatic beta cells to glucose, GIP and GLP-1 enhance insulin secretion during hyperglycemia, thus leading to the incretin effect. In diabetics, the beta-cell response to GIP is diminished or absent. In contrast, GLP-1 retains its stimulatory effect in pancreatic beta-cells in diabetics, making it a more suitable candidate for diabetic therapy [3,4]. The American Diabetes Association recommends use of a glucagon-like peptide-1 receptor (GLP-1R) agonist as 1 of 5 additional therapeutic drugs to consider when lifestyle change and metformin therapy fail to achieve good glycemic control in patients with T2DM [5]. Beneficial effects of incretin therapy for T2DM in people include restoration of glucose sensitivity of pancreatic beta-cells, induction of glucose-dependent stimulation of insulin secretion while simultaneously suppressing glucagon secretion, inhibition of beta-cell apoptosis and increased beta-cell neogenesis, as well as slowing gastric emptying rate and enhancing satiation peripherally and centrally [3]. These effects result in decreased risk of treatment-associated hypoglycemia, improved beta-cell function, and weight loss [6–8]. Recently, nonglucosedependent effects in cardiovascular health in people have been recognized, and research continues into the possible use of incretin therapy for neuropathic disorders [9]. The active forms of endogenous GLP-1 in humans include GLP1(7–37) and GLP-1(7–36). Both forms are quickly degraded (within 1–2 min) in the body by the ubiquitous enzyme dipeptidyl peptidase-4 (DPP-4) to inactive metabolites [3]. Liraglutide is a synthetic GLP-1R agonist which differs from endogenous GLP-1 by the substitution of lysine for arginine at position 34, and the attachment of a C-16 fatty acid (palmitoyl acid) at position 26. These changes in molecular structure allow liraglutide to bind to interstitial albumin and avoid metabolic degradation, lending the drug a halflife of 13 h in humans after subcutaneous injection [10]. The purpose of this study was to investigate the pharmacokinetics and pharmacodynamics of liraglutide therapy in healthy laboratory cats. We hypothesized that liraglutide would result in a glucose-dependent stimulation of insulin secretion, inhibition of glucagon secretion, and improved glucose tolerance. 2. Materials and methods 2.1. Animals The study protocol was approved by The Ohio State University Institutional Animal Care and Use Committee. Eight healthy purpose-bred cats were used in this study, including 6 castrated males and 2 spayed females. All cats were 3 years old. In the beginning of the study, mean standard deviation (SD) (range) body weight was 5.5 1.0 kg (4.7–7.0 kg). Five of the cats were overweight (body condition score was 5/9 in 3 cats, 6/9 in 2 cats, 7/9 in 1 cat, and 8/9 in 2 cats) [11]. Cats were group-housed in AAALAC accredited facilities. All cats were acclimatized and socialized for 4 wk before the start of experiments with environmental enrichment provided. Cats were fed ad lib a dry commercial cat food (IAMS Proactive Health Original with Chicken) by twice-daily timedfeedings. Appetite was assessed indirectly by observing the
115
amount of food left in food bowls in individual cages during daily timed feedings. Daily physical examinations were performed, and body weight was monitored at least weekly. Body weight was stable in all cats during the acclimatization period. Routine laboratory tests (including complete blood counts, serum biochemistry, total thyroxine, coagulation profile, and urinalysis) were performed on days 1, 15, and 28 of the experiment. 2.2. Study design A repeated-measures study design was used for the pharmacodynamics aspect of the study. Cats were maintained in a fasting state for 14 h before each clamp procedure. Hyperglycemic clamps (see the following) were performed on days 0 and 14. Subcutaneous injections of liraglutide (Victoza 18 mg/3 mL multi-dose pen, Novo Nordisk, Copenhagen, Denmark) were administered once daily with the use of a 32-gauge hypodermic needle that was attached to the prefilled injection pen, as directed by the manufacturer. The injection was administered in a previously shaved area on the cranial dorsum. The pen was set to deliver a fixed dose of 0.6 mg (mean dose 0.112 0.019 mg/kg). Liraglutide was injected in each cat on days 1 and 8 through 14 of the study. The LgHGC was performed 2 h after the final liraglutide injection. 2.3. Hyperglycemic clamp procedure Blood glucose (BG) concentrations were measured 90 min before each clamp procedure to ensure that stressrelated hyperglycemia was not present. Blood glucose was measured at 15 and 0 min and averaged to obtain baseline fasting BG concentrations. Beginning at time zero, blood glucose was measured every 5 min for 90 min with a handheld point-of-care glucose meter (AlphaTRAK 2; Abbott Animal Health, Abbott Park, IL) [12]. Also beginning at time zero and for the next 90 min, 20% dextrose (an in-saline dilution of 50% Dextrose USP; VETONE, MWI, Boise, ID) was infused intravenously (through a cephalic catheter) at a changing rate. In the first 30 min (adjustment period), dextrose was administered at an increasing rate to achieve hyperglycemia at a target of 225 mg/dL (range 200– 250 mg/dL). The infusion rate was determined based on the 5-min BG measurements. In the final 60 min of the procedure (hyperglycemic-clamp period), the dextrose infusion was adjusted as necessary to maintain target BG. Samples were collected for measurement of insulin and glucagon concentrations at baseline (15 and 0 min) and 30, 45, 60, 75, and 90 min. Blood samples for glucose and hormones measurements were collected via jugular catheters. 2.4. GLP-1 pharmacokinetics study The pharmacokinetics of liraglutide was evaluated after a single subcutaneous injection on day 1. Cats were maintained in a fasting state for 14 h before the injection. Blood samples for measuring liraglutide concentrations were collected at 0, 0.25, 0.5, 1, 2, 4, 6, 8, 12, 16, 24, 36, 48, 60, 72, and 84 h.
116
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
2.5. Liraglutide side effects In all cats, monitoring for potential liraglutide-related side effects included daily physical examinations, as well as close observation of general attitude, level of activity, quantification of food intake during timed feedings in individual cages, and observation of urination and defecation (including color, consistency, volume and so forth ) and was continued throughout the experiment. Diarrhea was defined as an appreciable loss of form to the stool with an increase in fecal water content. On days 1, 15, and 28, a complete blood count, serum chemistry profile, coagulation profile, and urinalysis were repeated in all cats. 2.6. Catheter placement and maintenance, blood collection, and storage Blood was collected through Vascular Access Ports (VAPs) which were surgically implanted as previously described [13] 2 to 28 d before the beginning of the study in 6 cats. In the other 2 cats, central intravenous catheters (MILA International, Inc, Erlanger, KY) were placed 2 d before the beginning of the study as previously described [13] and were used for blood collection on days 0 to 5; these catheters were subsequently replaced with VAPs on day 5. Cephalic catheters (Terumo Surflo I.V. Catheter, 22 G X 100 ; Terumo Corporation, Tokyo, Japan) were placed the day before each clamp procedure (1 and 13) and removed at the end of the clamp. All catheters and VAPs were placed under sedation with intramuscular injections of dexmedetomidine (20 mg/kg). Sedation was reversed with a dexmedetomidine-equivalent concentration of intramuscularly administered atipamezole. Butorphanol (0.2 mg/kg IM) was used as an analgesic for all catheter placements. Buprenorphine (0.02 mg/kg IM) was administered as an analgesic for VAP placement. Samples were collected into chilled-ethylenediaminetetraacetic acid tubes, centrifuged within 2 h of collection (4 C, 2,271 g for 15 min), and then separated and placed in plain glass tubes. Plasma was stored at 20 C until analysis. Continuous glucose monitors (CGM; Medtronic MiniMed with SOF-SENSOR Glucose Sensor, Northridge, CA) were placed under sedation (see the previously mentioned section) on day 7 for monitoring of hypoglycemia during daily consecutive liraglutide dosing. The CGMs were maintained with twice-daily calibrations as previously described [14] until the sensor failed (range 3–7 d after placement).
(P < 0.0001), slope ¼ 0.896 0.02 (CI ¼ 0.848–0.944) and a Y intercept ¼ 0.36 0.57 (Fig. 1). At lower concentrations of the standard curve (range 1.5–3 pmol/L); the average coefficient of variation (CV) was 7.0%. In the range of 3 to 8 pmol/L, the average CV was 4.6%, and in the range of 8 to 15 the average CV was 5.7%. Overall the intra-assay CV was 6.7%, and the interassay CV was 8.1%. Recovery post spiking was at 86.8% to 102.0%. According to the manufacturer, the assay’s sensitivity is 1 pmol/L, and it does not cross react with glucagon-related peptides including oxyntomodulin, glicentin, mini-glucagon, GLP-1, GLP-2, and glicentinrelated pancreatic peptide. The concentration of liraglutide was measured with an active GLP-1 ELISA (High Sensitivity GLP-1 Active Chemiluminescent ELISA kit, Millipore Corporation, Billerica, MA) as previously described [16,17]. In brief, samples were incubated for 4 h at 37 C to completely degrade endogenous GLP-1, and then the manufacturer’s protocol was followed. This assay has a range of 0.14 pmol/L to 100 pmol/ L. It has 100% cross-reactivity with active GLP-1 (7–36), 72% cross-reactivity with active GLP-1 (7–37), and 0% crossreactivity with inactive GLP-1 (9–36), GLP-2, GIP, glucagon, and oxyntomodulin. The manufacturer of this assay reports an intra-assay CV of 3%–6% and an inter-assay CV of 10%–13%. We found the intra-assay CV to be 13.1% for samples <3.73 pmol/L, (standards 1 to 4), 3.4% between 3.74 and 33.12 pmol/L, (standards 5–6) and 1.7% for samples >33.125 (standard 7 and above). 2.8. Statistical analysis Statistical analysis was performed using commercially available computer software (GraphPad Prism; GraphPad Software Inc, CA and SPSS 14.0 for Mac; SPSS Inc 2005, Chicago, IL, SAS v. 9.3, SAS Institute Inc, Cary, NC). All data were assessed for normal distribution using the ShapiroWilk test before applying parametric and nonparametric analysis where appropriate. The Shapiro-Wilk test was used to assess deviance from normal distribution of data.
2.7. Glucose and hormone measurements Blood glucose concentrations were measured with a hand-held point-of-care glucose monitor that was previously described for use in cats [12], as well as CGMs as described previously. Samples were tested in duplicate for all hormone assays. Insulin concentrations were measured with a previously validated feline insulin ELISA (Mercodia AB, Uppsala, Sweden) [15]. Glucagon concentrations were measured with a highsensitivity, high-specificity glucagon ELISA (Glucagon ELISA; Mercodia AB) that was concurrently validated for use in cats. Linear regression results for expected vs observed results in serial dilutions were R2 ¼ 0.9958
Fig. 1. Glucagon dilution parallelism in feline plasma using a commercial glucagon ELISA assay. ELISA, enzyme-linked immunosorbent assay.
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
Normally distributed data are presented as the means ( standard error of the mean). Data that were not normally distributed are presented as median and range. Paired t tests were used for comparison of blood glucose concentrations during HGC vs LgHGC. A Wilcoxon matched-pairs signed rank test was used to compare total glucose infused (TGI) during HGC and LgHGC. The repeated measures of blood insulin and glucagon concentrations during the glucose clamp before (day 0) and after liraglutide injections (day 14) were used as the outcome in statistical modeling. Separate analyses were run for insulin and glucagon. Statistical analyses were performed using PROC MIXED in Statistical Analysis Systems (SAS, v. 9.3, SAS Institute Inc). The mean of the insulin and glucagon measurements at time 15 min and at time 0 min was calculated and considered as the baseline measurement (ie, as time 0 value). First order autoregressive correlation structure, AR(1), was used to account for the correlated data structure due to the repeated measurements on individual cats. Time, treatment effect (measurements taken before and after the injection of liraglutide), and blood-glucose levels measured at the same time as insulin and glucagon concentrations were considered as potential explanatory variables in the model. Initially, univariate analyses were run by including them in the model individually. If associated with the outcome with P < 0.20 in the univariate analysis, variables were included in a full model simultaneously. Nonsignificant variables were dropped from the model one at a time, if P > 0.05. Fit of the models was evaluated by visual assessment of the residual plots. Plasma liraglutide concentrations vs time curves were subjected to statistical moment analysis using a software program (Phoenix WinNonlin version 5.2; Certara, Cary NC). The maximum plasma concentrations (Cmax) were taken from the individual plasma drug concentration vs time curves, and the tmax was the sampling time at which the Cmax was first observed. The terminal elimination rate (lz) was calculated from at least 3 time points, where plasma liraglutide concentrations were greater than lower limit of quantitation. The plasma terminal half-life was calculated using the relationship:
T1=2 lz ¼
ln 2
lz
The area under the plasma drug concentration vs time curve from time ¼ 0 to time ¼ last plasma drug concentration, was calculated using the log-linear trapezoidal rule:
AUC ¼
X Ai
Ai
li
þ
Clast
lz
The area under the moment curve was defined as:
AUMC ¼
X
The mean residence time (time required for 63.2% of dose to be eliminated) was calculated as:
MRT ¼
AUMC AUC
3. Results 3.1. Pharmacodynamics All but 1 cat had at least 1 episode of vomiting or diarrhea during liraglutide treatment. These events occurred most commonly on days 9 to 12 of the study (Fig. 2). Appetite was decreased in all cats. In 1 cat, this progressed to complete anorexia on days 11 and 12, and the cat was withdrawn from the study on day 13. Weight loss was recorded in all 7 cats (5 males and 2 females, body weight [mean SD; range] ¼ 5.07 kg 0.87; 4.3–6.4 kg) who completed the study at day 14 (9% 3%; P ¼ 0.006). In the 7 cats that completed the study, body weight at 28 d (2 wk after cessation of liraglutide treatment) was 5.32 0.76 kg (range 4.5–6.6 kg) with an overall change of 4.9% 4.1% (P ¼ 0.02). The 1 cat that did not complete the study regained his normal appetite within 48 h of the last injection of liraglutide and regained body weight similarly to the others. No significant clinicopathologic changes were observed before and after liraglutide treatment. Baseline (the mean of times 15 and 0 min) BG, insulin, and glucagon concentrations did not differ significantly between HGC and LgHGC (Table 1). The mean of clamped BG concentrations (30–90 min) as well as the coefficient of variation of BG concentrations during that time did not differ significantly between HGC and LgHGC (Table 1 and Fig. 3). Mean insulin concentrations were significantly increased in the LgHGC compared with the HGC (P ¼ 0.0022). Higher insulin concentrations at baseline were associated with greater increases in insulin concentration during the LgHGC (but not during the HGC). Mean glucagon concentrations were decreased in the LgHGC compared with the HGC (P ¼ 0.0089; Fig. 4). In 6 of 7 cats, TGI was higher during the LgHGC compared with HGC (median [range] of 1.61 [1.11–2.54] g/kg vs 2.25 [1.64–3.10] g/k). In 1 cat the TGI was lower during the LgHGC compared with the HGC, and this cat was identified as an outlier (Grubb outlier test). After excluding this cat from TGI statistical analysis, there was a trend toward an increased TGI in the LgHGC vs the HGC (P ¼ 0.087; Fig. 5). 3.2. Pharmacokinetics
li
The AUC0/N was extrapolated to infinity using the relationship:
AUC ¼
117
1=2ðCn þ Cnþ1 Þ ðtnþ1 tn Þ ðtnþ1 Þ
Liraglutide concentrations were measured in 6 cats (4 males, 2 females, body weight [mean SD] ¼ 5.2 0.2 kg). The results are presented in Figure 6. Liraglutide was first detected at the first sampling point (15 min post injection) and reached a peak (Tmax) between 4 and 12 h after injection. Cmax ranged from 89.073 pmol/L to >104.987 pmol/ L (exceeding the upper limit of detection of the assay). By 48 h, the mean concentration was no longer significantly different from baseline. Liraglutide was still detectable in
118
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
Fig. 2. Adverse effects of liraglutide in 8 cats. Arrows indicate days of liraglutide injections (0.6 mg subcutaneous).
the serum at the final time point (84 h) in 5 of 6 cats. The terminal rate constant of elimination was 0.06 0.014 h1 (geometric mean) producing mean half-life of elimination of 12 2.8 h in these cats. The mean, maximum plasma concentrations of liraglutide were 102 5.9 pmol/L at 7.3 3.4 h. The AUC0/N (area under the curve [AUC] from time ¼ 0 to infinity) was 3,459 500 h pmol/L with the percentage of the AUC extrapolated from Tlast to infinity was 1.7% 1.2%. The mean residence time was 24 4 h. 4. Discussion This study investigated for the first time the effect of the DPP-4 resistant GLP-1 analog liraglutide in cats. Our study confirms that in healthy cats, 7 consecutive days of liraglutide therapy result in glucose-dependent augmentation of insulin secretion, and suppression of glucagon secretion with a trend toward increased TGI during a hyperglycemic clamp. As previously discussed, augmentation of insulin secretion through the incretin effect contributes at least 50%–70% of insulin secretion following a meal [3]. This outlines the importance of maintaining, regaining, or effectively replacing the incretin effect in T2DM [18]. The incretin effect is impaired in T2DM in people, and GLP-1 concentrations are lower in people with obesity, T2DM,
insulin resistance, and glucose intolerance [19–22]. GLP-1 is homologous among mammalian species [23] and GLP-1 mediates (at least in part) the incretin effect in cats as it does in people [15]. Exenatide is another DPP-4 resistant GLP-1R agonist that has been studied in cats [24–26]. In contrast to our findings with liraglutide, in all 3 studies on exenatide no side effects were observed despite using a dose of exenatide that is about 15 times higher than recommended in people. A similarly high dose in people typically results in nausea, vomiting, and diarrhea. Similar side effects are commonly reported with liraglutide therapy in people and as with exenatide, these side effects are dose dependent [27]. Although these side effects are observed in a substantial number of patients, they typically subside over the first few weeks of therapy. In fact, a stepwise escalation of dosing appears to have reduced the frequency that these side effects are noted in people [10]. Gastrointestinal side effects in people are likely related to the effect on the central and peripheral nervous system, as well as decreased gastric emptying [28]. Similarly, 7 of 8 cats in our study experienced at least 1 episode of vomiting or diarrhea. Decreased appetite was observed in all cats, and in 1 cat this progressed to complete anorexia. Although the duration of consecutive daily liraglutide doses was only 7 d, the
Table 1 Paired comparisons of baseline BG, insulin, and glucagon concentrations (mean SD [range]; time 15 and 0) and clamped BG, insulin, and glucagon concentrations (time 30–90 min) during 2 hyperglycemic clamp procedures performed before (HGC) and following 7 consecutive days of liraglutide administration (LgHGC).
BG (mg/dL), baseline Insulin (ng/L), baseline Glucagon (pmol/L), baseline BG (mg/dL), clamped Insulin (ng/L), clamped Glucagon (pmol/L), clamped
HGC
LgHGC
P-value
91 10.9 (80.5–106.0) 166.1 141.0 (14–412) 7.1 4.8 (2.1–14.8) 228.8 10.35 (214.3–247.7) 455.5 115.8 (411–660) 0.66 0.4 (0.0–1.0)
90.5 10.9 (84–128.7) 159.2 83.1 (79–325) 5.7 4.5 (3.1–12.6) 224.5 9.29 (212.1–239.4) 760.8 350.7 (449–1,493) 0.5 0.4 (0.0–1.1)
0.49 0.18 0.13 0.0581 0.0022a 0.0089a
Abbreviations: BG, blood-glucose concentrations; SD, standard deviation. a Denotes statistical significance P < 0.05.
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
119
Fig. 3. Line graph of median and range of blood glucose in the HGC (circles) and LgHGC (triangles) showing no significant difference in blood-glucose concentrations between clamps. The shaded gray area represents the target blood-glucose concentrations in both clamps.
frequency of gastrointestinal-related episodes was noted to be subsiding by the fifth day after initiation of consecutive daily dosing. Continuous exposure to a GLP-1 agonist downregulates the effects of that agonist on gastric emptying, which in turn reduces the effect on postprandial glucose excursions [29]. This downregulation does not occur with intermittent exposure even after many months (and years) of treatment. As a result, the overall effect of GLP-1 agonists on postprandial glucose excursion and gastric emptying are greater with intermittent exposure compared with continuous exposure. This is seen even
Fig. 4. Insulin and glucagon concentrations (mean SD) in 2 hyperglycemic clamps, before liraglutide administration (HGC; circles) and following 7 consecutive days of once daily liraglutide administration (0.6 mg subcutaneously; LgHGC; triangles). SD, standard deviation.
when peak drug concentrations are higher during continuous vs intermittent exposure. Conversely, the effect on fasting glucose is decreased with intermittent compared with continuous exposure [29]. These findings explain the clinical observation in people wherein short-acting GLP-1 analogs that do not reach steady-state concentrations (like exenatide) are not associated with a significant decrease in fasting blood glucose and lead to a high frequency of vomiting and nausea. In contrast, these side effects are mostly abated when drug concentrations stabilize, even at high concentrations, with long-acting GLP-1 agonists after steady state is achieved. At the same time, fasting blood glucose is reduced in people when steady state is achieved [30]. In this study, we did not observe any decline in fasting blood glucose; however, this might be related to the relatively short duration of treatment. It is also possible that an effect on fasting blood glucose would be observed if liraglutide is used in diabetic cats. In another study, on exenatide-extended release, we performed after this study, increased exenatide concentrations were maintained for many weeks and fasting blood glucose decreased significantly [31]. Liraglutide is packaged in a prefilled injection pen, and the dose administered to the cats in this pilot study was chosen as the lowest deliverable dose (0.6 mg/dose) for the pen to facilitate accurate dosing. For the cats in our study, this resulted in a mean dosage of 0.112 0.019 mg/kg. In comparison, when the same pen is used in an 80 kg person at the lowest dose, the person receives 0.0075 mg/kg. However, this dose comparison across species is of limited value. Indeed, our cats received a dose that was about 15 times higher than the starting dose in people but displayed relatively less side effects in terms of frequency and duration of gastrointestinal signs. In people, weight loss is
120
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
Fig. 5. A trend toward increased total amount of glucose infused was observed during the hyperglycemic clamp performed following 7 consecutive days of liraglutide therapy (0.6 mg subcutaneously once daily; LgHGC) as compared with the initial hyperglycemic clamp performed before any drug administration (HGC).
observed during GLP-1 analog treatment even in those that do not experience gastrointestinal side effects [10]. In this study, weight loss was significant even after excluding 1 cat that developed complete anorexia. This weight loss exceeded the current recommendations for safe weight loss in cats of 0.5%–2% of body weight per week [32]. Thus, if liraglutide is to be used as therapy for obesity in cats, its dose and/or frequency should be reduced. No evidence of clinical hypoglycemia was observed in any of the study cats. Similar findings have been obtained in human studies in which clinical hypoglycemia was not observed in people undergoing liraglutide therapy, unless another antidiabetic therapy was used concurrently (ie, insulin glargine or sulfonylureas) [10]. Liraglutide has also been studied in nondiabetic obese people as treatment for obesity [27,33,34]. Importantly, documented hypoglycemia in these nondiabetics is not observed (similar to our nondiabetic cats) emphasizing that the effects of liraglutide are indeed glucose dependent.
Fig. 6. Liraglutide concentrations (median and range) obtained over 84 h following a single subcutaneous injection of 0.6 mg/cat of liraglutide. By 48 h, the median concentrations were no longer significantly different from baseline values.
Fasting glucagon concentrations in our study are lower than previously reported in cats [35]. In these previous reports, the glucagon assays were validated before the wide recognition of cross-reactivity between glucagon, GLP-1, and other glucagon gene products. Therefore, although these assays would be sensitive for detecting glucagon, they have decreased specificity and previous results likely represented cross-reaction with multiple glucagon gene products. During hyperglycemia, glucagon concentrations were often below the lower limit of detection of the glucagon assay. Therefore, the difference in mean glucagon concentrations between HGC and LgHGC, although statistically significant, should be cautiously interpreted. However, the frequency of samples below the limit of detection was 1.5 times greater in the LgHGC compared with the HGC which further supports a significant difference in glucagon concentrations between the clamps. Still, it remains possible that the difference detected in glucagon concentrations between clamps represent a type I error. Examination of the difference between the means of the blood-glucose concentrations during the hyperglycemic clamps before and after drug administration revealed a difference that approached statistical significance. The large number of data points in this calculation is responsible for a high power for detection of even small differences that are likely not biologically significant. However, considering the possibility of a biologically important difference, increased average blood glucose during HGC is expected to result in increased insulin secretion and a false decrease in the apparent effect of liraglutide. Thus, our results reflect a “worst case scenario” and most likely the effect of liraglutide in causing augmentation of insulin secretion could be even greater. The upper limit of the GLP-1 assay was frequently exceeded when measuring liraglutide plasma concentrations. This precluded our ability to determine a more accurate time to peak concentration and may have been avoided with sample dilution. Further studies are needed to evaluate the efficacy and tolerability of liraglutide in FDM patients. Given the similarities between FDM and T2DM in people and the results of our study in healthy cats, liraglutide has the potential to improve glycemic control (through increased insulin and decreased glucagon secretion) and promote weight loss in diabetic cats. Given the relatively long half-life and evidence in people that prolonged administration leads to higher sustained plasma concentrations, evaluation of decreasing frequency of drug administration to improve tolerability could be considered [17]. Our study has shown that liraglutide potentiates glucose-dependent insulin secretion and glucagon inhibition in healthy cats without affecting fasting blood glucose. Liraglutide administration was associated with decreased appetite and significant weight loss, although these effects appeared to be worse at the initiation of therapy. The maximum concentration of liraglutide in cats was observed between 4 and 12 h, with an elimination half-life of approximately 12 h making once daily dosing feasible. However, a decrease in dosing frequency may result in less severe weight loss and potential-associated complications. Further investigation into the use of liraglutide in cats with
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
feline diabetes mellitus, and as a weight loss drug in overweight cats, is warranted. References [1] Henson MS, O’Brien TD. Feline models of type 2 diabetes mellitus. ILAR J 2006;47:234–42. [2] Nelson RW, Griffey SM, Feldman EC, Ford SL. Transient clinical diabetes mellitus in cats: 10 cases (1989-1991). J Vet Intern Med 1999;13:28–35. [3] Baggio LL, Drucker DJ. Biology of incretins: GLP-1 and GIP. Gastroenterology 2007;132:2131–57. [4] Nauck MA, Heimesaat MM, Orskov C, Holst JJ, Ebert R, Creutzfeldt W. Preserved incretin activity of glucagon-like peptide 1 [7-36 amide] but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. J Clin Invest 1993; 91:301–7. [5] Bailey T. Options for combination therapy in type 2 diabetes: comparison of the ADA/EASD position statement and AACE/ACE algorithm. Am J Med 2013;126(9 Suppl 1):S10–20. [6] Jendle J, Nauck MA, Matthews DR, Frid A, Hermansen K, During M, Zdravkovic M, Strauss BJ, Garber AJ. Weight loss with liraglutide, a once-daily human glucagon-like peptide-1 analogue for type 2 diabetes treatment as monotherapy or added to metformin, is primarily as a result of a reduction in fat tissue. Diabetes Obes Metab 2009;11:1163–72. [7] Nauck M, Frid A, Hermansen K, Thomsen AB, During M, Shah N, Tankova T, Mitha I, Matthews DR. Long-term efficacy and safety comparison of liraglutide, glimepiride and placebo, all in combination with metformin in type 2 diabetes: 2-year results from the LEAD-2 study. Diabetes Obes Metab 2013;15:204–12. [8] Macconell L, Pencek R, Li Y, Maggs D, Porter L. Exenatide once weekly: sustained improvement in glycemic control and cardiometabolic measures through 3 years. Diabetes Metab Syndr Obes 2013;6:31–41. [9] Holscher C. The incretin hormones glucagonlike peptide 1 and glucose-dependent insulinotropic polypeptide are neuroprotective in mouse models of Alzheimer’s disease. Alzheimers Dement 2014; 10(1 Suppl):S47–54. [10] Bode B. An overview of the pharmacokinetics, efficacy and safety of liraglutide. Diabetes Res Clin Pract 2012;97:27–42. [11] Baldwin K, Bartges J, Buffington T, Freeman LM, Grabow M, Legred J, Ostwald D. AAHA nutritional assessment guidelines for dogs and cats. J Am Anim Hosp Assoc 2010;46:285–96. [12] Zini E, Moretti S, Tschuor F, Reusch CE. Evaluation of a new portable glucose meter designed for the use in cats. Schweiz Arch Tierheilkd 2009;151:448–51. [13] Aubert I, Abrams-Ogg AC, Sylvestre AM, Dyson DH, Allen DG, Johnstone IB. The use of vascular access ports for blood collection in feline blood donors. Can J Vet Res 2011;75:25–34. [14] Moretti S, Tschuor F, Osto M, Franchini M, Wichert B, Ackermann M, Lutz TA, Reusch CE, Zini E. Evaluation of a novel real-time continuous glucose-monitoring system for use in cats. J Vet Intern Med 2010;24:120–6. [15] Gilor C, Graves TK, Gilor S, Ridge TK, Weng HY, Dossin O. The incretin effect in cats: comparison between oral glucose, lipids, and amino acids. Domest Anim Endocrinol 2011;40:205–12. [16] Hoenig M, Jordan ET, Ferguson DC, de Vries F. Oral glucose leads to a differential response in glucose, insulin, and GLP-1 in lean versus obese cats. Domest Anim Endocrinol 2010;38:95–102.
121
[17] Agerso H, Jensen LB, Elbrond B, Rolan P, Zdravkovic M. The pharmacokinetics, pharmacodynamics, safety and tolerability of NN2211, a new long-acting GLP-1 derivative, in healthy men. Diabetologia 2002;45:195–202. [18] Kim W, Egan JM. The role of incretins in glucose homeostasis and diabetes treatment. Pharmacol Rev 2008;60:470–512. [19] Vilsboll T, Krarup T, Deacon CF, Madsbad S, Holst JJ. Reduced postprandial concentrations of intact biologically active glucagon-like peptide 1 in type 2 diabetic patients. Diabetes 2001;50:609–13. [20] Toft-Nielsen MB, Damholt MB, Madsbad S, Hilsted LM, Hughes TE, Michelsen BK, Holst JJ. Determinants of the impaired secretion of glucagon-like peptide-1 in type 2 diabetic patients. J Clin Endocrinol Metab 2001;86:3717–23. [21] Rask E, Olsson T, Soderberg S, Johnson O, Seckl J, Holst JJ, Ahren B. Impaired incretin response after a mixed meal is associated with insulin resistance in nondiabetic men. Diabetes Care 2001;24: 1640–5. [22] Muscelli E, Mari A, Casolaro A, Camastra S, Seghieri G, Gastaldelli A, Holst JJ, Ferrannini E. Separate impact of obesity and glucose tolerance on the incretin effect in normal subjects and type 2 diabetic patients. Diabetes 2008;57:1340–8. [23] Irwin DM. Molecular evolution of mammalian incretin hormone genes. Regul Pept 2009;155:121–30. [24] Gilor C, Graves TK, Gilor S, Ridge TK, Rick M. The GLP-1 mimetic exenatide potentiates insulin secretion in healthy cats. Domest Anim Endocrinol 2011;41:42–9. [25] Seyfert T, Brunker JD, Maxwell LK, Payton ME, McFarlane D. Effects of a glucagon-like Peptide-1 mimetic (exenatide) in healthy cats. Intern J Appl Res Vet Med 2012;10:147–56. [26] Padrutt I, Zini E, Kaufmann K, et al. Comparison of the GLP-1 analogues exenatide (short-acting), exenatide (long-acting) and the DPP-4 inhibitor sitagliptin to increase insulin secretion in healthy cats. J Vet Intern Med 2012;26:1520. [27] Lean ME, Carraro R, Finer N, Hartvig H, Lindegaard ML, Rossner S, Van Gaal L, Astrup A. Tolerability of nausea and vomiting and associations with weight loss in a randomized trial of liraglutide in obese, non-diabetic adults. Int J Obes (Lond) 2014;38:689–97. [28] Drucker DJ. The biology of incretin hormones. Cell Metab 2006;3: 153–65. [29] Fineman MS, Cirincione BB, Maggs D, Diamant M. GLP-1 based therapies: differential effects on fasting and postprandial glucose. Diabetes Obes Metab 2012;14:675–88. [30] Flint A, Kapitza C, Hindsberger C, Zdravkovic M. The once-daily human glucagon-like peptide-1 (GLP-1) analog liraglutide improves postprandial glucose levels in type 2 diabetes patients. Adv Ther 2011;28:213–26. [31] Rudinsky AJ, Adin CA, Borin-Crivellenti S, Hall MJ, Gilor C. The pharmacology of the GLP-1 analog exenatide extended-release in healthy cats. J Vet Intern Med 2014;28:1026. [32] Brooks D, Churchill J, Fein K, Linder D, Michel KE, Tudor K, Ward E, Witzel A. AAHA weight management guidelines for dogs and cats. J Am Anim Hosp Assoc 2014;50:1–11. [33] van Can J, Sloth B, Jensen CB, Flint A, Blaak EE, Saris WH. Effects of the once-daily GLP-1 analog liraglutide on gastric emptying, glycemic parameters, appetite and energy metabolism in obese, nondiabetic adults. Int J Obes (lond) 2014;38:784–93. [34] Madsbad S. The role of glucagon-like peptide-1 impairment in obesity and potential therapeutic implications. Diabetes Obes Metab 2014;16:9–21. [35] Furrer D, Kaufmann K, Tschuor F, Reusch CE, Lutz TA. The dipeptidyl peptidase IV inhibitor NVP-DPP728 reduces plasma glucagon concentration in cats. Vet J 2010;183:355–7.
Contents lists available at ScienceDirect
Domestic Animal Endocrinology journal homepage: www.domesticanimalendo.com
Pharmacokinetics and pharmacodynamics of the glucagon-like peptide-1 analog liraglutide in healthy cats M.J. Hall a, C.A. Adin a, S. Borin-Crivellenti a, b, A.J. Rudinsky a, P. Rajala-Schultz c, J. Lakritz a, C. Gilor a, * a b c
Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University, Columbus, OH 43210, USA FCAV/Universidade Estadual Paulista (UNESP), Jaboticabal, SP, Brazil Department of Veterinary Preventative Medicine, The Ohio State University, Columbus, OH 43210, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 July 2014 Received in revised form 1 December 2014 Accepted 2 December 2014
Glucagon-like peptide-1 (GLP-1) is an intestinal hormone that induces glucose-dependent stimulation of insulin secretion while suppressing glucagon secretion. Glucagon-like peptide-1 also increases beta cell mass and satiation while decelerating gastric emptying. Liraglutide is a fatty-acid derivative of GLP-1 with a protracted pharmacokinetic profile that is used in people for treatment of type II diabetes mellitus and obesity. The aim of this study was to determine the pharmacokinetics and pharmacodynamics of liraglutide in healthy cats. Hyperglycemic clamps were performed on days 0 (HGC) and 14 (LgHGC) in 7 healthy cats. Liraglutide was administered subcutaneously (0.6 mg/cat) once daily on days 8 through 14. Compared with the HGC (mean standard deviation; 455.5 115.8 ng/L), insulin concentrations during LgHGC were increased (760.8 350.7 ng/L; P ¼ 0.0022), glucagon concentrations decreased (0.66 0.4 pmol/L during HGC vs 0.5 0.4 pmol/L during LgHGC; P ¼ 0.0089), and there was a trend toward an increased total glucose infused (median [range] ¼ 1.61 (1.11–2.54) g/kg and 2.25 (1.64–3.10) g/kg, respectively; P ¼ 0.087). Appetite reduction and decreased body weight (9% 3%; P ¼ 0.006) were observed in all cats. Liraglutide has similar effects and pharmacokinetics profile in cats to those reported in people. With a half-life of approximately 12 h, once daily dosing might be feasible; however, significant effects on appetite and weight loss may necessitate dosage or dosing frequency reductions. Further investigation of liraglutide in diabetic cats and overweight cats is warranted. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Feline Diabetes Insulin Glucagon Incretin
1. Introduction Feline diabetes mellitus (FDM) closely resembles type 2 diabetes mellitus in people, as both FDM and type II diabetes mellitus (T2DM) are characterized by insulin resistance, relative insulin deficiency, decreased beta cell mass, and islet-amyloid depositions [1,2]. Despite technological advances in insulin formulations and monitoring, insulin therapy across species is still
* Corresponding author. Tel.: þ1 6142923551; fax: þ1 6142921454. E-mail address: [email protected] (C. Gilor). 0739-7240/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.domaniend.2014.12.001
commonly associated with inadequate glycemic control leading to short- and long-term complications such as hypoglycemia, weight gain, and vascular diseases. Incretinbased therapy for T2DM was first introduced in 2005 and has quickly become a widely used adjunctive therapy. The incretin effect is the greater release of insulin occurring after oral glucose administration compared with intravenous administration of the same glucose dose. Two peptide hormones contribute to the incretin effect and are called incretin hormones: glucagon-like peptide-1 and glucosedependent insulinotropic peptide (GIP). They are secreted from specialized enteroendocrine cells in response to the presence of nutrients in the lumen of the gut (but not in
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
response to the presence of nutrients in the blood). By enhancing the response of pancreatic beta cells to glucose, GIP and GLP-1 enhance insulin secretion during hyperglycemia, thus leading to the incretin effect. In diabetics, the beta-cell response to GIP is diminished or absent. In contrast, GLP-1 retains its stimulatory effect in pancreatic beta-cells in diabetics, making it a more suitable candidate for diabetic therapy [3,4]. The American Diabetes Association recommends use of a glucagon-like peptide-1 receptor (GLP-1R) agonist as 1 of 5 additional therapeutic drugs to consider when lifestyle change and metformin therapy fail to achieve good glycemic control in patients with T2DM [5]. Beneficial effects of incretin therapy for T2DM in people include restoration of glucose sensitivity of pancreatic beta-cells, induction of glucose-dependent stimulation of insulin secretion while simultaneously suppressing glucagon secretion, inhibition of beta-cell apoptosis and increased beta-cell neogenesis, as well as slowing gastric emptying rate and enhancing satiation peripherally and centrally [3]. These effects result in decreased risk of treatment-associated hypoglycemia, improved beta-cell function, and weight loss [6–8]. Recently, nonglucosedependent effects in cardiovascular health in people have been recognized, and research continues into the possible use of incretin therapy for neuropathic disorders [9]. The active forms of endogenous GLP-1 in humans include GLP1(7–37) and GLP-1(7–36). Both forms are quickly degraded (within 1–2 min) in the body by the ubiquitous enzyme dipeptidyl peptidase-4 (DPP-4) to inactive metabolites [3]. Liraglutide is a synthetic GLP-1R agonist which differs from endogenous GLP-1 by the substitution of lysine for arginine at position 34, and the attachment of a C-16 fatty acid (palmitoyl acid) at position 26. These changes in molecular structure allow liraglutide to bind to interstitial albumin and avoid metabolic degradation, lending the drug a halflife of 13 h in humans after subcutaneous injection [10]. The purpose of this study was to investigate the pharmacokinetics and pharmacodynamics of liraglutide therapy in healthy laboratory cats. We hypothesized that liraglutide would result in a glucose-dependent stimulation of insulin secretion, inhibition of glucagon secretion, and improved glucose tolerance. 2. Materials and methods 2.1. Animals The study protocol was approved by The Ohio State University Institutional Animal Care and Use Committee. Eight healthy purpose-bred cats were used in this study, including 6 castrated males and 2 spayed females. All cats were 3 years old. In the beginning of the study, mean standard deviation (SD) (range) body weight was 5.5 1.0 kg (4.7–7.0 kg). Five of the cats were overweight (body condition score was 5/9 in 3 cats, 6/9 in 2 cats, 7/9 in 1 cat, and 8/9 in 2 cats) [11]. Cats were group-housed in AAALAC accredited facilities. All cats were acclimatized and socialized for 4 wk before the start of experiments with environmental enrichment provided. Cats were fed ad lib a dry commercial cat food (IAMS Proactive Health Original with Chicken) by twice-daily timedfeedings. Appetite was assessed indirectly by observing the
115
amount of food left in food bowls in individual cages during daily timed feedings. Daily physical examinations were performed, and body weight was monitored at least weekly. Body weight was stable in all cats during the acclimatization period. Routine laboratory tests (including complete blood counts, serum biochemistry, total thyroxine, coagulation profile, and urinalysis) were performed on days 1, 15, and 28 of the experiment. 2.2. Study design A repeated-measures study design was used for the pharmacodynamics aspect of the study. Cats were maintained in a fasting state for 14 h before each clamp procedure. Hyperglycemic clamps (see the following) were performed on days 0 and 14. Subcutaneous injections of liraglutide (Victoza 18 mg/3 mL multi-dose pen, Novo Nordisk, Copenhagen, Denmark) were administered once daily with the use of a 32-gauge hypodermic needle that was attached to the prefilled injection pen, as directed by the manufacturer. The injection was administered in a previously shaved area on the cranial dorsum. The pen was set to deliver a fixed dose of 0.6 mg (mean dose 0.112 0.019 mg/kg). Liraglutide was injected in each cat on days 1 and 8 through 14 of the study. The LgHGC was performed 2 h after the final liraglutide injection. 2.3. Hyperglycemic clamp procedure Blood glucose (BG) concentrations were measured 90 min before each clamp procedure to ensure that stressrelated hyperglycemia was not present. Blood glucose was measured at 15 and 0 min and averaged to obtain baseline fasting BG concentrations. Beginning at time zero, blood glucose was measured every 5 min for 90 min with a handheld point-of-care glucose meter (AlphaTRAK 2; Abbott Animal Health, Abbott Park, IL) [12]. Also beginning at time zero and for the next 90 min, 20% dextrose (an in-saline dilution of 50% Dextrose USP; VETONE, MWI, Boise, ID) was infused intravenously (through a cephalic catheter) at a changing rate. In the first 30 min (adjustment period), dextrose was administered at an increasing rate to achieve hyperglycemia at a target of 225 mg/dL (range 200– 250 mg/dL). The infusion rate was determined based on the 5-min BG measurements. In the final 60 min of the procedure (hyperglycemic-clamp period), the dextrose infusion was adjusted as necessary to maintain target BG. Samples were collected for measurement of insulin and glucagon concentrations at baseline (15 and 0 min) and 30, 45, 60, 75, and 90 min. Blood samples for glucose and hormones measurements were collected via jugular catheters. 2.4. GLP-1 pharmacokinetics study The pharmacokinetics of liraglutide was evaluated after a single subcutaneous injection on day 1. Cats were maintained in a fasting state for 14 h before the injection. Blood samples for measuring liraglutide concentrations were collected at 0, 0.25, 0.5, 1, 2, 4, 6, 8, 12, 16, 24, 36, 48, 60, 72, and 84 h.
116
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
2.5. Liraglutide side effects In all cats, monitoring for potential liraglutide-related side effects included daily physical examinations, as well as close observation of general attitude, level of activity, quantification of food intake during timed feedings in individual cages, and observation of urination and defecation (including color, consistency, volume and so forth ) and was continued throughout the experiment. Diarrhea was defined as an appreciable loss of form to the stool with an increase in fecal water content. On days 1, 15, and 28, a complete blood count, serum chemistry profile, coagulation profile, and urinalysis were repeated in all cats. 2.6. Catheter placement and maintenance, blood collection, and storage Blood was collected through Vascular Access Ports (VAPs) which were surgically implanted as previously described [13] 2 to 28 d before the beginning of the study in 6 cats. In the other 2 cats, central intravenous catheters (MILA International, Inc, Erlanger, KY) were placed 2 d before the beginning of the study as previously described [13] and were used for blood collection on days 0 to 5; these catheters were subsequently replaced with VAPs on day 5. Cephalic catheters (Terumo Surflo I.V. Catheter, 22 G X 100 ; Terumo Corporation, Tokyo, Japan) were placed the day before each clamp procedure (1 and 13) and removed at the end of the clamp. All catheters and VAPs were placed under sedation with intramuscular injections of dexmedetomidine (20 mg/kg). Sedation was reversed with a dexmedetomidine-equivalent concentration of intramuscularly administered atipamezole. Butorphanol (0.2 mg/kg IM) was used as an analgesic for all catheter placements. Buprenorphine (0.02 mg/kg IM) was administered as an analgesic for VAP placement. Samples were collected into chilled-ethylenediaminetetraacetic acid tubes, centrifuged within 2 h of collection (4 C, 2,271 g for 15 min), and then separated and placed in plain glass tubes. Plasma was stored at 20 C until analysis. Continuous glucose monitors (CGM; Medtronic MiniMed with SOF-SENSOR Glucose Sensor, Northridge, CA) were placed under sedation (see the previously mentioned section) on day 7 for monitoring of hypoglycemia during daily consecutive liraglutide dosing. The CGMs were maintained with twice-daily calibrations as previously described [14] until the sensor failed (range 3–7 d after placement).
(P < 0.0001), slope ¼ 0.896 0.02 (CI ¼ 0.848–0.944) and a Y intercept ¼ 0.36 0.57 (Fig. 1). At lower concentrations of the standard curve (range 1.5–3 pmol/L); the average coefficient of variation (CV) was 7.0%. In the range of 3 to 8 pmol/L, the average CV was 4.6%, and in the range of 8 to 15 the average CV was 5.7%. Overall the intra-assay CV was 6.7%, and the interassay CV was 8.1%. Recovery post spiking was at 86.8% to 102.0%. According to the manufacturer, the assay’s sensitivity is 1 pmol/L, and it does not cross react with glucagon-related peptides including oxyntomodulin, glicentin, mini-glucagon, GLP-1, GLP-2, and glicentinrelated pancreatic peptide. The concentration of liraglutide was measured with an active GLP-1 ELISA (High Sensitivity GLP-1 Active Chemiluminescent ELISA kit, Millipore Corporation, Billerica, MA) as previously described [16,17]. In brief, samples were incubated for 4 h at 37 C to completely degrade endogenous GLP-1, and then the manufacturer’s protocol was followed. This assay has a range of 0.14 pmol/L to 100 pmol/ L. It has 100% cross-reactivity with active GLP-1 (7–36), 72% cross-reactivity with active GLP-1 (7–37), and 0% crossreactivity with inactive GLP-1 (9–36), GLP-2, GIP, glucagon, and oxyntomodulin. The manufacturer of this assay reports an intra-assay CV of 3%–6% and an inter-assay CV of 10%–13%. We found the intra-assay CV to be 13.1% for samples <3.73 pmol/L, (standards 1 to 4), 3.4% between 3.74 and 33.12 pmol/L, (standards 5–6) and 1.7% for samples >33.125 (standard 7 and above). 2.8. Statistical analysis Statistical analysis was performed using commercially available computer software (GraphPad Prism; GraphPad Software Inc, CA and SPSS 14.0 for Mac; SPSS Inc 2005, Chicago, IL, SAS v. 9.3, SAS Institute Inc, Cary, NC). All data were assessed for normal distribution using the ShapiroWilk test before applying parametric and nonparametric analysis where appropriate. The Shapiro-Wilk test was used to assess deviance from normal distribution of data.
2.7. Glucose and hormone measurements Blood glucose concentrations were measured with a hand-held point-of-care glucose monitor that was previously described for use in cats [12], as well as CGMs as described previously. Samples were tested in duplicate for all hormone assays. Insulin concentrations were measured with a previously validated feline insulin ELISA (Mercodia AB, Uppsala, Sweden) [15]. Glucagon concentrations were measured with a highsensitivity, high-specificity glucagon ELISA (Glucagon ELISA; Mercodia AB) that was concurrently validated for use in cats. Linear regression results for expected vs observed results in serial dilutions were R2 ¼ 0.9958
Fig. 1. Glucagon dilution parallelism in feline plasma using a commercial glucagon ELISA assay. ELISA, enzyme-linked immunosorbent assay.
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
Normally distributed data are presented as the means ( standard error of the mean). Data that were not normally distributed are presented as median and range. Paired t tests were used for comparison of blood glucose concentrations during HGC vs LgHGC. A Wilcoxon matched-pairs signed rank test was used to compare total glucose infused (TGI) during HGC and LgHGC. The repeated measures of blood insulin and glucagon concentrations during the glucose clamp before (day 0) and after liraglutide injections (day 14) were used as the outcome in statistical modeling. Separate analyses were run for insulin and glucagon. Statistical analyses were performed using PROC MIXED in Statistical Analysis Systems (SAS, v. 9.3, SAS Institute Inc). The mean of the insulin and glucagon measurements at time 15 min and at time 0 min was calculated and considered as the baseline measurement (ie, as time 0 value). First order autoregressive correlation structure, AR(1), was used to account for the correlated data structure due to the repeated measurements on individual cats. Time, treatment effect (measurements taken before and after the injection of liraglutide), and blood-glucose levels measured at the same time as insulin and glucagon concentrations were considered as potential explanatory variables in the model. Initially, univariate analyses were run by including them in the model individually. If associated with the outcome with P < 0.20 in the univariate analysis, variables were included in a full model simultaneously. Nonsignificant variables were dropped from the model one at a time, if P > 0.05. Fit of the models was evaluated by visual assessment of the residual plots. Plasma liraglutide concentrations vs time curves were subjected to statistical moment analysis using a software program (Phoenix WinNonlin version 5.2; Certara, Cary NC). The maximum plasma concentrations (Cmax) were taken from the individual plasma drug concentration vs time curves, and the tmax was the sampling time at which the Cmax was first observed. The terminal elimination rate (lz) was calculated from at least 3 time points, where plasma liraglutide concentrations were greater than lower limit of quantitation. The plasma terminal half-life was calculated using the relationship:
T1=2 lz ¼
ln 2
lz
The area under the plasma drug concentration vs time curve from time ¼ 0 to time ¼ last plasma drug concentration, was calculated using the log-linear trapezoidal rule:
AUC ¼
X Ai
Ai
li
þ
Clast
lz
The area under the moment curve was defined as:
AUMC ¼
X
The mean residence time (time required for 63.2% of dose to be eliminated) was calculated as:
MRT ¼
AUMC AUC
3. Results 3.1. Pharmacodynamics All but 1 cat had at least 1 episode of vomiting or diarrhea during liraglutide treatment. These events occurred most commonly on days 9 to 12 of the study (Fig. 2). Appetite was decreased in all cats. In 1 cat, this progressed to complete anorexia on days 11 and 12, and the cat was withdrawn from the study on day 13. Weight loss was recorded in all 7 cats (5 males and 2 females, body weight [mean SD; range] ¼ 5.07 kg 0.87; 4.3–6.4 kg) who completed the study at day 14 (9% 3%; P ¼ 0.006). In the 7 cats that completed the study, body weight at 28 d (2 wk after cessation of liraglutide treatment) was 5.32 0.76 kg (range 4.5–6.6 kg) with an overall change of 4.9% 4.1% (P ¼ 0.02). The 1 cat that did not complete the study regained his normal appetite within 48 h of the last injection of liraglutide and regained body weight similarly to the others. No significant clinicopathologic changes were observed before and after liraglutide treatment. Baseline (the mean of times 15 and 0 min) BG, insulin, and glucagon concentrations did not differ significantly between HGC and LgHGC (Table 1). The mean of clamped BG concentrations (30–90 min) as well as the coefficient of variation of BG concentrations during that time did not differ significantly between HGC and LgHGC (Table 1 and Fig. 3). Mean insulin concentrations were significantly increased in the LgHGC compared with the HGC (P ¼ 0.0022). Higher insulin concentrations at baseline were associated with greater increases in insulin concentration during the LgHGC (but not during the HGC). Mean glucagon concentrations were decreased in the LgHGC compared with the HGC (P ¼ 0.0089; Fig. 4). In 6 of 7 cats, TGI was higher during the LgHGC compared with HGC (median [range] of 1.61 [1.11–2.54] g/kg vs 2.25 [1.64–3.10] g/k). In 1 cat the TGI was lower during the LgHGC compared with the HGC, and this cat was identified as an outlier (Grubb outlier test). After excluding this cat from TGI statistical analysis, there was a trend toward an increased TGI in the LgHGC vs the HGC (P ¼ 0.087; Fig. 5). 3.2. Pharmacokinetics
li
The AUC0/N was extrapolated to infinity using the relationship:
AUC ¼
117
1=2ðCn þ Cnþ1 Þ ðtnþ1 tn Þ ðtnþ1 Þ
Liraglutide concentrations were measured in 6 cats (4 males, 2 females, body weight [mean SD] ¼ 5.2 0.2 kg). The results are presented in Figure 6. Liraglutide was first detected at the first sampling point (15 min post injection) and reached a peak (Tmax) between 4 and 12 h after injection. Cmax ranged from 89.073 pmol/L to >104.987 pmol/ L (exceeding the upper limit of detection of the assay). By 48 h, the mean concentration was no longer significantly different from baseline. Liraglutide was still detectable in
118
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
Fig. 2. Adverse effects of liraglutide in 8 cats. Arrows indicate days of liraglutide injections (0.6 mg subcutaneous).
the serum at the final time point (84 h) in 5 of 6 cats. The terminal rate constant of elimination was 0.06 0.014 h1 (geometric mean) producing mean half-life of elimination of 12 2.8 h in these cats. The mean, maximum plasma concentrations of liraglutide were 102 5.9 pmol/L at 7.3 3.4 h. The AUC0/N (area under the curve [AUC] from time ¼ 0 to infinity) was 3,459 500 h pmol/L with the percentage of the AUC extrapolated from Tlast to infinity was 1.7% 1.2%. The mean residence time was 24 4 h. 4. Discussion This study investigated for the first time the effect of the DPP-4 resistant GLP-1 analog liraglutide in cats. Our study confirms that in healthy cats, 7 consecutive days of liraglutide therapy result in glucose-dependent augmentation of insulin secretion, and suppression of glucagon secretion with a trend toward increased TGI during a hyperglycemic clamp. As previously discussed, augmentation of insulin secretion through the incretin effect contributes at least 50%–70% of insulin secretion following a meal [3]. This outlines the importance of maintaining, regaining, or effectively replacing the incretin effect in T2DM [18]. The incretin effect is impaired in T2DM in people, and GLP-1 concentrations are lower in people with obesity, T2DM,
insulin resistance, and glucose intolerance [19–22]. GLP-1 is homologous among mammalian species [23] and GLP-1 mediates (at least in part) the incretin effect in cats as it does in people [15]. Exenatide is another DPP-4 resistant GLP-1R agonist that has been studied in cats [24–26]. In contrast to our findings with liraglutide, in all 3 studies on exenatide no side effects were observed despite using a dose of exenatide that is about 15 times higher than recommended in people. A similarly high dose in people typically results in nausea, vomiting, and diarrhea. Similar side effects are commonly reported with liraglutide therapy in people and as with exenatide, these side effects are dose dependent [27]. Although these side effects are observed in a substantial number of patients, they typically subside over the first few weeks of therapy. In fact, a stepwise escalation of dosing appears to have reduced the frequency that these side effects are noted in people [10]. Gastrointestinal side effects in people are likely related to the effect on the central and peripheral nervous system, as well as decreased gastric emptying [28]. Similarly, 7 of 8 cats in our study experienced at least 1 episode of vomiting or diarrhea. Decreased appetite was observed in all cats, and in 1 cat this progressed to complete anorexia. Although the duration of consecutive daily liraglutide doses was only 7 d, the
Table 1 Paired comparisons of baseline BG, insulin, and glucagon concentrations (mean SD [range]; time 15 and 0) and clamped BG, insulin, and glucagon concentrations (time 30–90 min) during 2 hyperglycemic clamp procedures performed before (HGC) and following 7 consecutive days of liraglutide administration (LgHGC).
BG (mg/dL), baseline Insulin (ng/L), baseline Glucagon (pmol/L), baseline BG (mg/dL), clamped Insulin (ng/L), clamped Glucagon (pmol/L), clamped
HGC
LgHGC
P-value
91 10.9 (80.5–106.0) 166.1 141.0 (14–412) 7.1 4.8 (2.1–14.8) 228.8 10.35 (214.3–247.7) 455.5 115.8 (411–660) 0.66 0.4 (0.0–1.0)
90.5 10.9 (84–128.7) 159.2 83.1 (79–325) 5.7 4.5 (3.1–12.6) 224.5 9.29 (212.1–239.4) 760.8 350.7 (449–1,493) 0.5 0.4 (0.0–1.1)
0.49 0.18 0.13 0.0581 0.0022a 0.0089a
Abbreviations: BG, blood-glucose concentrations; SD, standard deviation. a Denotes statistical significance P < 0.05.
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
119
Fig. 3. Line graph of median and range of blood glucose in the HGC (circles) and LgHGC (triangles) showing no significant difference in blood-glucose concentrations between clamps. The shaded gray area represents the target blood-glucose concentrations in both clamps.
frequency of gastrointestinal-related episodes was noted to be subsiding by the fifth day after initiation of consecutive daily dosing. Continuous exposure to a GLP-1 agonist downregulates the effects of that agonist on gastric emptying, which in turn reduces the effect on postprandial glucose excursions [29]. This downregulation does not occur with intermittent exposure even after many months (and years) of treatment. As a result, the overall effect of GLP-1 agonists on postprandial glucose excursion and gastric emptying are greater with intermittent exposure compared with continuous exposure. This is seen even
Fig. 4. Insulin and glucagon concentrations (mean SD) in 2 hyperglycemic clamps, before liraglutide administration (HGC; circles) and following 7 consecutive days of once daily liraglutide administration (0.6 mg subcutaneously; LgHGC; triangles). SD, standard deviation.
when peak drug concentrations are higher during continuous vs intermittent exposure. Conversely, the effect on fasting glucose is decreased with intermittent compared with continuous exposure [29]. These findings explain the clinical observation in people wherein short-acting GLP-1 analogs that do not reach steady-state concentrations (like exenatide) are not associated with a significant decrease in fasting blood glucose and lead to a high frequency of vomiting and nausea. In contrast, these side effects are mostly abated when drug concentrations stabilize, even at high concentrations, with long-acting GLP-1 agonists after steady state is achieved. At the same time, fasting blood glucose is reduced in people when steady state is achieved [30]. In this study, we did not observe any decline in fasting blood glucose; however, this might be related to the relatively short duration of treatment. It is also possible that an effect on fasting blood glucose would be observed if liraglutide is used in diabetic cats. In another study, on exenatide-extended release, we performed after this study, increased exenatide concentrations were maintained for many weeks and fasting blood glucose decreased significantly [31]. Liraglutide is packaged in a prefilled injection pen, and the dose administered to the cats in this pilot study was chosen as the lowest deliverable dose (0.6 mg/dose) for the pen to facilitate accurate dosing. For the cats in our study, this resulted in a mean dosage of 0.112 0.019 mg/kg. In comparison, when the same pen is used in an 80 kg person at the lowest dose, the person receives 0.0075 mg/kg. However, this dose comparison across species is of limited value. Indeed, our cats received a dose that was about 15 times higher than the starting dose in people but displayed relatively less side effects in terms of frequency and duration of gastrointestinal signs. In people, weight loss is
120
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
Fig. 5. A trend toward increased total amount of glucose infused was observed during the hyperglycemic clamp performed following 7 consecutive days of liraglutide therapy (0.6 mg subcutaneously once daily; LgHGC) as compared with the initial hyperglycemic clamp performed before any drug administration (HGC).
observed during GLP-1 analog treatment even in those that do not experience gastrointestinal side effects [10]. In this study, weight loss was significant even after excluding 1 cat that developed complete anorexia. This weight loss exceeded the current recommendations for safe weight loss in cats of 0.5%–2% of body weight per week [32]. Thus, if liraglutide is to be used as therapy for obesity in cats, its dose and/or frequency should be reduced. No evidence of clinical hypoglycemia was observed in any of the study cats. Similar findings have been obtained in human studies in which clinical hypoglycemia was not observed in people undergoing liraglutide therapy, unless another antidiabetic therapy was used concurrently (ie, insulin glargine or sulfonylureas) [10]. Liraglutide has also been studied in nondiabetic obese people as treatment for obesity [27,33,34]. Importantly, documented hypoglycemia in these nondiabetics is not observed (similar to our nondiabetic cats) emphasizing that the effects of liraglutide are indeed glucose dependent.
Fig. 6. Liraglutide concentrations (median and range) obtained over 84 h following a single subcutaneous injection of 0.6 mg/cat of liraglutide. By 48 h, the median concentrations were no longer significantly different from baseline values.
Fasting glucagon concentrations in our study are lower than previously reported in cats [35]. In these previous reports, the glucagon assays were validated before the wide recognition of cross-reactivity between glucagon, GLP-1, and other glucagon gene products. Therefore, although these assays would be sensitive for detecting glucagon, they have decreased specificity and previous results likely represented cross-reaction with multiple glucagon gene products. During hyperglycemia, glucagon concentrations were often below the lower limit of detection of the glucagon assay. Therefore, the difference in mean glucagon concentrations between HGC and LgHGC, although statistically significant, should be cautiously interpreted. However, the frequency of samples below the limit of detection was 1.5 times greater in the LgHGC compared with the HGC which further supports a significant difference in glucagon concentrations between the clamps. Still, it remains possible that the difference detected in glucagon concentrations between clamps represent a type I error. Examination of the difference between the means of the blood-glucose concentrations during the hyperglycemic clamps before and after drug administration revealed a difference that approached statistical significance. The large number of data points in this calculation is responsible for a high power for detection of even small differences that are likely not biologically significant. However, considering the possibility of a biologically important difference, increased average blood glucose during HGC is expected to result in increased insulin secretion and a false decrease in the apparent effect of liraglutide. Thus, our results reflect a “worst case scenario” and most likely the effect of liraglutide in causing augmentation of insulin secretion could be even greater. The upper limit of the GLP-1 assay was frequently exceeded when measuring liraglutide plasma concentrations. This precluded our ability to determine a more accurate time to peak concentration and may have been avoided with sample dilution. Further studies are needed to evaluate the efficacy and tolerability of liraglutide in FDM patients. Given the similarities between FDM and T2DM in people and the results of our study in healthy cats, liraglutide has the potential to improve glycemic control (through increased insulin and decreased glucagon secretion) and promote weight loss in diabetic cats. Given the relatively long half-life and evidence in people that prolonged administration leads to higher sustained plasma concentrations, evaluation of decreasing frequency of drug administration to improve tolerability could be considered [17]. Our study has shown that liraglutide potentiates glucose-dependent insulin secretion and glucagon inhibition in healthy cats without affecting fasting blood glucose. Liraglutide administration was associated with decreased appetite and significant weight loss, although these effects appeared to be worse at the initiation of therapy. The maximum concentration of liraglutide in cats was observed between 4 and 12 h, with an elimination half-life of approximately 12 h making once daily dosing feasible. However, a decrease in dosing frequency may result in less severe weight loss and potential-associated complications. Further investigation into the use of liraglutide in cats with
M.J. Hall et al. / Domestic Animal Endocrinology 51 (2015) 114–121
feline diabetes mellitus, and as a weight loss drug in overweight cats, is warranted. References [1] Henson MS, O’Brien TD. Feline models of type 2 diabetes mellitus. ILAR J 2006;47:234–42. [2] Nelson RW, Griffey SM, Feldman EC, Ford SL. Transient clinical diabetes mellitus in cats: 10 cases (1989-1991). J Vet Intern Med 1999;13:28–35. [3] Baggio LL, Drucker DJ. Biology of incretins: GLP-1 and GIP. Gastroenterology 2007;132:2131–57. [4] Nauck MA, Heimesaat MM, Orskov C, Holst JJ, Ebert R, Creutzfeldt W. Preserved incretin activity of glucagon-like peptide 1 [7-36 amide] but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. J Clin Invest 1993; 91:301–7. [5] Bailey T. Options for combination therapy in type 2 diabetes: comparison of the ADA/EASD position statement and AACE/ACE algorithm. Am J Med 2013;126(9 Suppl 1):S10–20. [6] Jendle J, Nauck MA, Matthews DR, Frid A, Hermansen K, During M, Zdravkovic M, Strauss BJ, Garber AJ. Weight loss with liraglutide, a once-daily human glucagon-like peptide-1 analogue for type 2 diabetes treatment as monotherapy or added to metformin, is primarily as a result of a reduction in fat tissue. Diabetes Obes Metab 2009;11:1163–72. [7] Nauck M, Frid A, Hermansen K, Thomsen AB, During M, Shah N, Tankova T, Mitha I, Matthews DR. Long-term efficacy and safety comparison of liraglutide, glimepiride and placebo, all in combination with metformin in type 2 diabetes: 2-year results from the LEAD-2 study. Diabetes Obes Metab 2013;15:204–12. [8] Macconell L, Pencek R, Li Y, Maggs D, Porter L. Exenatide once weekly: sustained improvement in glycemic control and cardiometabolic measures through 3 years. Diabetes Metab Syndr Obes 2013;6:31–41. [9] Holscher C. The incretin hormones glucagonlike peptide 1 and glucose-dependent insulinotropic polypeptide are neuroprotective in mouse models of Alzheimer’s disease. Alzheimers Dement 2014; 10(1 Suppl):S47–54. [10] Bode B. An overview of the pharmacokinetics, efficacy and safety of liraglutide. Diabetes Res Clin Pract 2012;97:27–42. [11] Baldwin K, Bartges J, Buffington T, Freeman LM, Grabow M, Legred J, Ostwald D. AAHA nutritional assessment guidelines for dogs and cats. J Am Anim Hosp Assoc 2010;46:285–96. [12] Zini E, Moretti S, Tschuor F, Reusch CE. Evaluation of a new portable glucose meter designed for the use in cats. Schweiz Arch Tierheilkd 2009;151:448–51. [13] Aubert I, Abrams-Ogg AC, Sylvestre AM, Dyson DH, Allen DG, Johnstone IB. The use of vascular access ports for blood collection in feline blood donors. Can J Vet Res 2011;75:25–34. [14] Moretti S, Tschuor F, Osto M, Franchini M, Wichert B, Ackermann M, Lutz TA, Reusch CE, Zini E. Evaluation of a novel real-time continuous glucose-monitoring system for use in cats. J Vet Intern Med 2010;24:120–6. [15] Gilor C, Graves TK, Gilor S, Ridge TK, Weng HY, Dossin O. The incretin effect in cats: comparison between oral glucose, lipids, and amino acids. Domest Anim Endocrinol 2011;40:205–12. [16] Hoenig M, Jordan ET, Ferguson DC, de Vries F. Oral glucose leads to a differential response in glucose, insulin, and GLP-1 in lean versus obese cats. Domest Anim Endocrinol 2010;38:95–102.
121
[17] Agerso H, Jensen LB, Elbrond B, Rolan P, Zdravkovic M. The pharmacokinetics, pharmacodynamics, safety and tolerability of NN2211, a new long-acting GLP-1 derivative, in healthy men. Diabetologia 2002;45:195–202. [18] Kim W, Egan JM. The role of incretins in glucose homeostasis and diabetes treatment. Pharmacol Rev 2008;60:470–512. [19] Vilsboll T, Krarup T, Deacon CF, Madsbad S, Holst JJ. Reduced postprandial concentrations of intact biologically active glucagon-like peptide 1 in type 2 diabetic patients. Diabetes 2001;50:609–13. [20] Toft-Nielsen MB, Damholt MB, Madsbad S, Hilsted LM, Hughes TE, Michelsen BK, Holst JJ. Determinants of the impaired secretion of glucagon-like peptide-1 in type 2 diabetic patients. J Clin Endocrinol Metab 2001;86:3717–23. [21] Rask E, Olsson T, Soderberg S, Johnson O, Seckl J, Holst JJ, Ahren B. Impaired incretin response after a mixed meal is associated with insulin resistance in nondiabetic men. Diabetes Care 2001;24: 1640–5. [22] Muscelli E, Mari A, Casolaro A, Camastra S, Seghieri G, Gastaldelli A, Holst JJ, Ferrannini E. Separate impact of obesity and glucose tolerance on the incretin effect in normal subjects and type 2 diabetic patients. Diabetes 2008;57:1340–8. [23] Irwin DM. Molecular evolution of mammalian incretin hormone genes. Regul Pept 2009;155:121–30. [24] Gilor C, Graves TK, Gilor S, Ridge TK, Rick M. The GLP-1 mimetic exenatide potentiates insulin secretion in healthy cats. Domest Anim Endocrinol 2011;41:42–9. [25] Seyfert T, Brunker JD, Maxwell LK, Payton ME, McFarlane D. Effects of a glucagon-like Peptide-1 mimetic (exenatide) in healthy cats. Intern J Appl Res Vet Med 2012;10:147–56. [26] Padrutt I, Zini E, Kaufmann K, et al. Comparison of the GLP-1 analogues exenatide (short-acting), exenatide (long-acting) and the DPP-4 inhibitor sitagliptin to increase insulin secretion in healthy cats. J Vet Intern Med 2012;26:1520. [27] Lean ME, Carraro R, Finer N, Hartvig H, Lindegaard ML, Rossner S, Van Gaal L, Astrup A. Tolerability of nausea and vomiting and associations with weight loss in a randomized trial of liraglutide in obese, non-diabetic adults. Int J Obes (Lond) 2014;38:689–97. [28] Drucker DJ. The biology of incretin hormones. Cell Metab 2006;3: 153–65. [29] Fineman MS, Cirincione BB, Maggs D, Diamant M. GLP-1 based therapies: differential effects on fasting and postprandial glucose. Diabetes Obes Metab 2012;14:675–88. [30] Flint A, Kapitza C, Hindsberger C, Zdravkovic M. The once-daily human glucagon-like peptide-1 (GLP-1) analog liraglutide improves postprandial glucose levels in type 2 diabetes patients. Adv Ther 2011;28:213–26. [31] Rudinsky AJ, Adin CA, Borin-Crivellenti S, Hall MJ, Gilor C. The pharmacology of the GLP-1 analog exenatide extended-release in healthy cats. J Vet Intern Med 2014;28:1026. [32] Brooks D, Churchill J, Fein K, Linder D, Michel KE, Tudor K, Ward E, Witzel A. AAHA weight management guidelines for dogs and cats. J Am Anim Hosp Assoc 2014;50:1–11. [33] van Can J, Sloth B, Jensen CB, Flint A, Blaak EE, Saris WH. Effects of the once-daily GLP-1 analog liraglutide on gastric emptying, glycemic parameters, appetite and energy metabolism in obese, nondiabetic adults. Int J Obes (lond) 2014;38:784–93. [34] Madsbad S. The role of glucagon-like peptide-1 impairment in obesity and potential therapeutic implications. Diabetes Obes Metab 2014;16:9–21. [35] Furrer D, Kaufmann K, Tschuor F, Reusch CE, Lutz TA. The dipeptidyl peptidase IV inhibitor NVP-DPP728 reduces plasma glucagon concentration in cats. Vet J 2010;183:355–7.