Clin Pharmacokinet

https://doi.org/10.1007/s40262-018-0668-z

REVIEW ARTICLE

Pharmacokinetics and Clinical Implications of Semaglutide:

A New Glucagon-Like Peptide (GLP)-1 Receptor Agonist
Sylvie Hall1 • Diana Isaacs2 • Jennifer N. Clements3

Ó Springer International Publishing AG, part of Springer Nature 2018

Abstract Glucagon-like peptide-1 receptor agonists application to clinical practice based on efficacy and safety
(GLP-1 RAs) came to market in the year 2005, as a new data.
therapeutic classification, for clinical use in the manage-
ment of type 2 diabetes mellitus (T2DM). Since 2005, there
have been six approved products on the market, with the
Key Points
newest product being semaglutide (Novo Nordisk). Several
studies have been conducted and completed evaluating its
Semaglutide is a new glucagon like peptide-1
pharmacokinetics as a once-weekly subcutaneous injection.
receptor agonist for the management of type 2
As a dose of 0.5 or 1 mg, semaglutide has a half-life of
diabetes.
7 days; therefore, it would reach steady state in 4–5 weeks.
There are few drug interactions and dose adjustments are Semaglutide is an effective option due to improved
not necessary. However, similar to other GLP-1 RAs, glycemic control, weight loss, and low risk of
semaglutide can delay gastric emptying and may impact hypoglycemia.
the absorption of oral medications. Based on clinical trials, Additional studies will determine its efficacy and
semaglutide has been compared with placebo, sitagliptin, safety in patients with obesity.
exenatide extended release, and insulin glargine as
monotherapy or add-on therapy. Semaglutide has resulted
in a 1.5–1.9% glycosylated hemoglobin A1c reduction after
30–56 weeks. It also produced 5–10% weight reduction 1 Introduction
from baseline in clinical efficacy studies. Semaglutide can
be another acceptable option for patients with T2DM, and In the USA, there are an estimated 30.3 million people
it has a potential role among patients who require weight diagnosed with diabetes mellitus; the incidence of new
loss with a low risk of hypoglycemia. This article evaluates diagnoses is 1.5 million people per year [1]. Compara-
the pharmacokinetics of semaglutide and summarizes its tively, there are approximately 422 million individuals
affected with diabetes worldwide [2]. Due to the growing
evidence regarding the pathophysiology of type 2 diabetes
& Jennifer N. Clements (T2DM), it is essential to individualize therapy for a patient
jclements@presby.edu based on several factors (i.e., patient preference, degree of
1 glycosylated hemoglobin A1c (HbA1c) lowering, risk of
Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH
44195, USA hypoglycemia, cost, effect on weight).
2 In the guidelines from the American Diabetes Associa-
Cleveland Clinic, Endocrine and Metabolism Institute
Diabetes Center, 9500 Euclid Avenue, Cleveland, OH 44195, tion (ADA) and American Association of Clinical
USA Endocrinologists (AACE), metformin is recommended as
3
Presbyterian College School of Pharmacy, 307 North Broad first-line therapy for the management of T2DM. Metformin
Street, Clinton, SC 29325, USA has been a mainstay medication as monotherapy and/or in
 S. Hall et al.

combination with other anti-hyperglycemic agents, as long It has a structure with 94% homology with human GLP-1.
as the patient does not have a contraindication to met- However, it has specific modifications at position 8, 26, and
formin [3, 4]. Glucagon-like peptide-1 (GLP-1) receptor 34 to extend its half-life, allowing for once-weekly
agonists (GLP-1 RAs) are recommended as a second-line administration [6].
medication in both published guidelines, with liraglutide As a GLP-1 RA, semaglutide has four different mech-
being the preferred GLP-1 RA, according to the ADA, due anisms of action to control blood glucose levels and pro-
to its cardiovascular evidence [3–5]. While the AACE mote weight loss. To improve glycemic control,
guidelines follow a hierarchy with recommendations, GLP- semaglutide promotes insulin secretion from b cells and
1 RAs are options for patients requiring glycemic control, suppresses glucagon secretion from a cells in the pancreas.
satiety control, and weight loss due to their mechanism of To promote weight loss, semaglutide slows gastric emp-
action—promoting insulin secretion from b cells while tying; in addition, semaglutide promotes satiety due to its
decreasing glucagon secretion in the a cells of the pan- low molecular weight [6].
creas, promoting satiety in the brain, and slowing gastric
emptying in the intestines [3, 4]. 2.1 Pharmacokinetics
Since 2005, there have been six GLP-1 RAs approved by
the US Food and Drug Administration (FDA). Table 1 The long half-life and low rate of total plasma clearance of
summarizes the clinical characteristics of these six medi- semaglutide support once-weekly administration.
cations. Semaglutide is the seventh GLP-1 RA approved in
the USA on 5 December 2017 (OzempicÒ, Novo Nordisk) 2.2 Absorption
and may be the second GLP-1 RA also indicated for obesity
management. Albiglutide (marketed as TanzeumÒ, Table 2 provides pharmacokinetic results from semaglutide
GlaxoSmithKline) will be discontinued from the US market studies. In healthy adults, subcutaneous administration of
in July 2018. This article reviews the pharmacokinetics, one-time semaglutide 0.5 mg reached maximum concen-
pharmacodynamics, clinical evidence, safety, dosage, and tration (Cmax) in 24–56 h [7, 8]. Dose-escalation studies
administration of semaglutide to predict its clinical impli- using weekly semaglutide 0.25 mg doses for four doses,
cations for practice. The purpose of this review is to sum- followed by 0.5 mg doses for four doses and then 1.0 mg
marize and highlight evidence regarding semaglutide for for five doses showed similar time to Cmax (tmax) of
the management of T2DM, as a newly approved GLP-1 RA, 33–36 h after the final 1.0 mg dose [9, 10]. Cmax and area
but also as a future option for the management of obesity. under the plasma concentration–time curve (AUC) are
similar following a single 0.5 mg dose and, likewise, are
also similar following the final 1.0 mg dose of a dose-
2 Chemistry and Pharmacology escalation strategy [7–10]. Subcutaneous bioavailability of
semaglutide is approximately 94%, the highest of all cur-
GLP-1 is a gastrointestinal hormone and rapidly degraded rently available GLP-1 RAs (Table 3) [11–14].
by dipeptidyl peptidase-4 (DPP-4) enzyme, resulting in a
short half-life. GLP-1 RAs have been developed to be 2.3 Distribution
similar in structure to endogenous GLP-1, but have dif-
ferent structural components in order to provide the bene- Semaglutide was developed with the intent to design a
ficial effects of GLP-1. Semaglutide is a human GLP-1 liraglutide analog with increased affinity for albumin
analog, studied for the management of T2DM and obesity. binding to provide once-weekly dosing of a GLP-1 RA. In

Table 1 Characteristics of glucagon-like peptide-1 receptor agonists

Brand name Generic name FDA approval Homology (%) Half-life Frequency/administration

AdlyxinTM [32] Lixisenatide August 2016 50 3h Once daily/SC

Ò
Bydureon [33] Exenatide ER March 2014 53 2 weeks Once weekly/SC
ByettaÒ [34] Exenatide IR April 2005 53 2.4 h Twice daily/SC
OzempicÒ [35] Semaglutide December 2017 94 1 week Once weekly/SC
TanzeumÒ [36] Albiglutide April 2014 97 5 days Once weekly/SC
TrulicityÒ [13] Dulaglutide September 2014 90 5 days Once weekly/SC
Ò
Victoza [12] Liraglutide January 2010 97 13 h Once daily/SC
ER extended-release, FDA US Food and Drug Administration, IR immediate-release, SC subcutaneously
Semaglutide for Type 2 Diabetes

Table 2 Summary of findings from healthy individuals in pharmacokinetic trials [7–10]

Study Semaglutide dosing prior to study measurements tmax Cmax AUC t‘
(h) (nmol/L) (nmol h/L) (h)

Marbury et al. Single 0.5 mg dose 24 10.3 2600 183

[7]
Jensen et al. [8] Single 0.5 mg dose 56 10.9 3123 168
Blundell et al. Weekly 0.25 mg dose (4 weeks), weekly 0.5 mg dose (4 weeks), weekly 33 32 4467 N/A
[9] a 1.0 mg dose (5 weeks)
Kapitza et al. Weekly 0.25 mg dose (4 weeks), weekly 0.5 mg dose (4 weeks), weekly 36 33.8 4602 165
[10]a 1.0 mg dose (5 weeks)
AUC area under the plasma concentration–time curve, Cmax maximum (or peak) serum concentration, N/A not applicable, t‘ half-life, tmax time to
maximum serum concentration
a
Values represent steady state of final 1.0 mg dose after 12 weeks of treatment

Table 3 Subcutaneous bioavailability of available glucagon-like indicated that semaglutide was metabolized to six different
peptide-1 agonists [11–14] metabolites identified as P1–P3 and P5–P7. P4 represents
GLP-1 agonist Subcutaneous bioavailability (%) the parent compound [3H]-semaglutide, which was the
primary component detected in plasma (82.6%).
Semaglutide 94 Semaglutide is metabolized by proteolytic cleavage of the
Liraglutide 55–66 peptide backbone and b-oxidation of the fatty acid side
Dulaglutide 47–65 chain. Concentration of metabolites declined over time,
Exenatide 62 with only the parent compound detected in plasma at
Lixisenatide Unknowna 28 days post-dose. It is unknown what role the metabolites
GLP-1 glucagon-like peptide-1 have in efficacy or adverse effects.
a
The sponsor of lixisenatide did not perform absolute bioavailability
studies 2.5 Elimination

the foundation study that led to the creation of semaglutide, In the study mentioned in Sect. 2.4, excretion of radiola-
Lau et al. [11] showed that in in vivo characterization beled [3H]-semaglutide and its metabolites was also
studies in pigs, the half-life of semaglutide after intra- described [8]. Following 64 days of collection after a one-
venous administration was 46.1 h compared with the 12.4- time subcutaneous dose of semaglutide 0.5 mg, 75.1% of
h half-life of liraglutide [11]. Semaglutide showed a higher the dose was recovered, with 53% in urine, 18.6% in feces,
volume of distribution (0.102 L/kg) than liraglutide and 3.2% in expired air. Parent semaglutide and 21
(0.067 L/kg) and slower body clearance (0.0016 vs. metabolites were detected in urine, and 3.1% of the
0.0038 L/h/kg), indicating higher albumin binding and administered dose was found to be [3H]-semaglutide, with
time spent in the body. Subcutaneous mean residence time two metabolites (P6 and P7) each accounting for 14% of
following subcutaneous administration of semaglutide was the administered dose, and all other metabolites accounting
63.6 h compared with 23.0 h for liraglutide [11]. There for B 1.8% of the administered dose. The minimal amount
were no toxic effects noted in this study resulting from the of intact drug in the urine indicates that renal dosing may
higher mean residence time of semaglutide than of not be needed for semaglutide. No parent drug was
liraglutide. In the clinical trials discussed here, semaglutide detected in feces.
caused more gastrointestinal adverse effects than exe-
natide, possibly due to longer mean residence time, but had
a similar adverse effect profile to liraglutide [15, 16] (see 3 Special Populations
Table 4).
3.1 Renal Impairment
2.4 Metabolism
The effect of renal impairment (RI) on semaglutide phar-
The metabolism of semaglutide was studied in an absorp- macokinetics and tolerability was studied in 56 patients
tion, metabolism, and excretion study in seven male par- with varying degrees of renal function categorized as
ticipants who received a single subcutaneous dose of having normal renal function, mild RI, moderate RI, severe
radiolabeled [3H]-semaglutide [8]. Metabolite profiling RI, and end-stage renal disease (ESRD) requiring
 S. Hall et al.

Table 4 Summary of glycosylated hemoglobin-lowering ability of semaglutide from SUSTAIN trials [15, 16, 19–22]
Mean reduction in HbA1c Percentage achieving Percentage achieving
HbA1c \ 7% HbA1c B 6.5%
Semaglutide Comparator Semaglutide Comparator Semaglutide Comparator
0.5 mg 1 mg 0.5 mg 1 mg 0.5 mg 1 mg

SUSTAIN 1 - 1.45 - 1.55 - 0.02 74 72 25 59 60 13

Semaglutide vs. placebo
SUSTAIN 2 - 1.3 - 1.6 - 0.5 69 53 36 53 66 20
Semaglutide vs. sitagliptin
100 mg as an add-on to
metformin, TZDs, or both
SUSTAIN 3 N/A - 1.5 - 0.9 N/A 67 40 N/A 47 22
Semaglutide vs. exenatide
ER 2 mg as add-on to
1 or 2 oral agents
SUSTAIN 4 - 1.21 - 1.64 - 0.83 57 73 38 37 54 18
Semaglutide vs. insulin
glargine as add-on to
metformin ± sulfonylureas
SUSTAIN 5 - 1.4 –1.8 - 0.1 61 79 11 41 61 5
Semaglutide vs. placebo as
add-on to basal
insulin ± metformin
SUSTAIN 7 Low High Low High Low High Low High Low High Low High
dose dose dose dose dose dose dose dose dose dose dose dose
Semaglutide low (0.5 mg) and - 1.5 - 1.8 - 1.1 - 1.4 68 79 52 67 49 67 34 47
high dose (1.0 mg) vs.
dulaglutide low (0.75 mg)
and high dose (1.5 mg) as
add-on to metformin
ER extended release, HbA1c glycosylated hemoglobin, N/A not applicable, TZDs thiazolidinediones

hemodialysis who received one time subcutaneous 3.2 Hepatic Impairment

semaglutide 0.5 mg doses [7]. Cmax was 10–20% higher in
subjects with RI than in those with normal renal function; A phase I single-dose pharmacokinetic trial showed no
however, there was no trend in severity of RI and Cmax. In need for dose adjustment in patients with varying levels of
an analysis of covariance model adjusted for age, sex, and hepatic function [17]. Further, the effect of hepatic
body weight, there were no differences in AUC among the impairment on semaglutide pharmacokinetics was studied
different groups. Comparisons of point estimates of AUC in 44 patients with normal hepatic function or mild, mod-
pre- and post-dialysis in subjects with ESRD showed no erate, or severe hepatic impairment for 5 weeks following a
significant differences, indicating the hemodialysis has single 0.5 mg subcutaneous dose [18]. There was no sig-
little impact on semaglutide. There was also no difference nificant difference in AUC or Cmax among the four groups,
in protein binding across the groups. indicating that hepatic impairment does not affect
RI is a common complication of diabetes, making it semaglutide exposure. The tmax and the half-life of the drug
essential to understand renal dosing of anti-hyperglycemic were also similar across all groups. Despite reduced con-
agents. The study discussed in this section [7] provides centrations of albumin in hepatic impairment, plasma
evidence that the pharmacokinetics of semaglutide are not protein binding of semaglutide was [ 99% in all groups, so
affected by RI, and thus renal dose adjustments may not be a large amount of unbound drug should not be expected in
necessary. Patients with all stages of RI were included in patients with liver dysfunction. Additionally, the SUS-
the landmark SUSTAIN trials, which showed similar safety TAIN trials included patients with hepatic impairment and
and efficacy compared with patients with normal renal found no differences in the safety profile [15, 16, 19–23].
function.
Semaglutide for Type 2 Diabetes

4 Drug Interactions participants (OR 15.99, 95% CI 7.82–32.68) and 60% of

participants (OR 18.34, 95% CI 8.96–37.54) compared
Semaglutide coadministered with metformin, warfarin, with 13% of placebo recipients. Semaglutide doses in
digoxin, atorvastatin, ethinyl estradiol, and levonorgestrel SUSTAIN 1 also showed a significant reduction in mean
has been studied to determine potential drug–drug inter- body weight compared with placebo, the secondary end-
actions. As semaglutide lacks a specific route for metabo- point of the study. While not significant, there were trends
lism, there is no drug–drug interaction between towards lower systolic and diastolic blood pressure similar
semaglutide and these medications; therefore, dose to effects seen in other studies [19].
adjustments are not necessary [24, 25]. However, caution is In SUSTAIN 2, investigators compared semaglutide 0.5
warranted regarding the absorption of oral medications, as and 1 mg weekly with 100 mg daily of the DPP-4 inhibitor
semaglutide delays gastric emptying [26]. sitagliptin [20]. Eligible patients were adults with T2DM
inadequately controlled on metformin, thiazolidinediones,
or both. Study sites were in 18 countries (not the USA)
5 Evidence of Efficacy in Type 2 Diabetes Mellitus with a total of 1231 enrolled participants. Baseline char-
acteristics were similar among the three groups, except
The efficacy of semaglutide was studied in the SUSTAIN HbA1c was slightly higher in the sitagliptin arm (8.2%)
trials [1–5, 7] and a series of phase IIIa randomized trials. than in each semaglutide arm (8.0% in both). The primary
SUSTAIN 1 compared weekly semaglutide 0.5 or 1 mg outcome, mean HbA1c change from baseline to 56 weeks,
monotherapy with placebo among treatment-naı̈ve adults showed significantly greater reduction in both semaglutide
with T2DM insufficiently controlled by lifestyle manage- arms compared with sitagliptin. Semaglutide 0.5 mg pro-
ment alone [19]. The study enrolled 387 participants in duced a –1.3% mean reduction in HbA1c (95% CI - 1.42 to
study sites located in eight countries, including the USA. - 1.21; P \ 0.0001) and semaglutide 1 mg produced a
Baseline characteristics were generally similar among the –1.6% mean reduction in HbA1c (95% CI - 1.71 to - 1.51;
treatment and control arms. The primary outcome was P \ 0.0001) compared with - 0.5% in patients randomized
change in mean HbA1c from baseline to 30 weeks. Both the to sitagliptin. Similar to SUSTAIN 1, the secondary out-
0.5 and 1 mg doses of semaglutide produced significantly comes were the proportion of participants achieving a
larger reductions in HbA1c than placebo: - 1.45% (95% reduction in HbA1c goals of \ 7 and B 6.5%. These goals
confidence interval [CI] - 1.65 to - 1.26; P \ 0.0001) and were achieved by 69 and 53% of participants in the
- 1.55% (95% CI - 1.74 to - 1.36; P \ 0.0001), respec- semaglutide 0.5 mg arm and 78 and 66% in the semaglu-
tively. This reduction in HbA1c is greater than seen in other tide 1 mg arm compared with 36 and 20% of participants in
GLP-1 RA clinical trials compared with placebo (Fig. 1) the sitagliptin arm. Both doses of semaglutide resulted in
[27]. Further analysis showed that 74% of participants in significantly higher proportions of participants achieving
the semaglutide 0.5 mg arm (odds ratio [OR] 16.92, 95% these goals compared with sitagliptin (P \ 0.0001).
CI 8.44–33.89) and 72% of participants in the 1 mg arm Additionally, both doses of semaglutide significantly
(OR 15.70, 95% CI 8.00–30.83) achieved a goal HbA1c of- reduced body weight and systolic blood pressure compared
\ 7% compared with 25% of participants in the placebo with sitagliptin [20].
arm. When using the more intense HbA1c goal of B 6.5%, SUSTAIN 3 compared semaglutide 1 mg with weekly
semaglutide 0.5 and 1 mg achieved target HbA1c in 59% of exenatide extended release (ER) 2 mg in adults with
T2DM not controlled on one or two oral antidiabetic agents
[15]. The study enrolled 819 subjects in 12 countries,
Mean Reducon in HbA1C (%) including the USA. The two study groups did not differ in
1.8
1.6 baseline characteristics, and 96.5, 48.1, and 2.3% of
1.4
1.2
patients were on biguanides, sulfonylureas, and thiazo-
1 lidinediones, respectively. In terms of primary outcome, at
0.8
0.6 56 weeks, semaglutide resulted in a significantly larger
0.4
0.2 reduction in HbA1c of 1.5%, compared with 0.9% by
0 exenatide (P \ 0.0001). Achievement of target HbA1c was
significantly higher among subjects taking semaglutide,
with 67% reaching HbA1c \ 7.0 and 47% reaching
HbA1c B 6.5%, while in the exenatide group 40 and 22%
achieved these goals, respectively (P \ 0.0001 for both
Fig. 1 Mean reduction in glycosylated hemoglobin (HbA1c) of HbA1c targets). Semaglutide also showed significantly
glucagon-like peptide-1 analogs compared with placebo [27] greater reductions of body weight, body mass index (BMI),
 S. Hall et al.

weight circumference, and systolic blood pressure than 0.75 mg, semaglutide 1.0 mg, or dulaglutide 1.5 mg
exenatide (P \ 0.05) for all measures [15]. weekly. The primary endpoint was change in HbA1c from
SUSTAIN 4 compared semaglutide 0.5 or 1 mg with baseline at week 40. Baseline characteristics were similar
insulin glargine as add-on therapy for insulin-naı̈ve patients between groups and included mean HbA1c of 8.2%, age
with T2DM not adequately controlled on metformin 55–56 years, and mean BMI of 33 kg/m2. In total, 301
monotherapy or in combination with a sulfonylurea [21]. patients were randomized to semaglutide 0.5 mg, 299 to
Participants randomized to receive insulin were initiated on dulaglutide 0.75 mg, 300 to semaglutide 1.0 mg, and 299
10 units daily; a titration protocol was followed to achieve to dulaglutide 1.5 mg weekly. The mean percentage HbA1c
target pre-breakfast plasma glucose levels of 72–99 mg/dL. was reduced by 1.5% with semaglutide 0.5 mg and 1.1%
The mean insulin dose at the end of the study was 29.2 with dulaglutide 0.75 mg (P \ 0.0001). Mean HbA1c was
units. This trial enrolled 1089 patients in 14 countries, reduced by 1.8% with semaglutide 1.0 mg versus 1.4%
including the USA. There were no significant baseline with dulaglutide 1.5 mg (P \ 0.00001). The authors con-
differences between treatment groups. Of the study popu- cluded that at low and high doses, semaglutide was supe-
lation, 48% received metformin monotherapy and 52% rior to dulaglutide for lowering HbA1c [16].
metformin plus a sulfonylurea. Primary and secondary
outcomes were the same as the previous SUSTAIN trials.
At 30 weeks, the mean HbA1c reduction in the semaglutide 6 Oral Semaglutide
0.5 mg group was - 1.21 (95% CI - 1.31 to - 1.10;
P \ 0.0001) and in the semaglutide 1 mg group was The SUSTAIN trials demonstrated positive effects of
- 1.64 (95% CI - 1.74 to - 1.54; P \ 0.0001) compared subcutaneous semaglutide on glycemic control; however,
with - 0.83% in the insulin glargine group. Target HbA1c injections can be a barrier to medication adherence. For
levels \ 7.0% (OR 2.39, 95% CI 1.73–3.28) were achieved this reason, an oral tablet containing semaglutide formu-
by 57% of participants in the semaglutide 0.5 mg group lated with an absorption enhancer is in clinical develop-
and 73% of the semaglutide 1 mg group (OR 5.78, 95% CI ment. A phase II trial compared various doses (2.5, 5, 10,
4.08–8.19) versus 38% of participants receiving insulin 20, 40 mg) and dosing regimens (daily 2.5 mg, standard
glargine. Similarly, semaglutide 0.5 and 1 mg resulted in dose escalation, and rapid dose escalation) of oral
greater proportions of patients achieving lower HbA1C semaglutide with once-weekly subcutaneous semaglutide
levels B 6.5%: 37% (OR 3.02, 95% CI 2.11–4.33) and 1 mg and oral placebo [28]. All doses and formulations of
54% (OR 6.86, 95% CI 4.76–9.89), respectively, compared semaglutide showed a significantly greater mean reduction
with 18% of participants in the insulin glargine group. in HbA1c than oral placebo at 26 weeks. Mean reduction in
Similar to the prior SUSTAIN trials, secondary endpoints HbA1c increased as oral semaglutide doses increased, with
demonstrated greater reduction in mean body weight and semaglutide 40 mg having comparable HbA1c reductions
systolic blood pressure than insulin glargine [21]. with the subcutaneous formulation, both resulting in 1.9
SUSTAIN 5 has not been fully published at the time of reductions. In addition to HbA1c reduction, secondary
writing, but results were presented as a poster at the outcome analysis revealed that oral doses of semaglutide
European Association for the Study for Diabetes in 2016 10 mg and higher resulted in a significantly greater
[22]. It was a double-blinded, placebo-controlled, parallel- reduction in body weight than placebo. It is not known
group, multicenter trial that compared semaglutide 0.5 and what the bioavailability of oral semaglutide is, but based on
1.0 mg with placebo in 397 patients with T2DM on these results in which 40 mg of oral drug produced effects
stable treatment with basal insulin alone or in combination similar to 1 mg subcutaneously, it is presumably much less
with metformin. The primary endpoint was change in than 100%.
HbA1c after 30 weeks. Baseline characteristics were simi-
lar between groups and included an average age of
58 years, HbA1c of 8.4%, and BMI of 32 kg/m2. The mean 7 Obesity
HbA1c reduction at week 30 was 1.4% and 1.8% with
semaglutide 0.5 and 1.0 mg, respectively, and 0.1% with The weight loss effects of semaglutide have been studied
placebo (P \ 0.0001 for both). Insulin doses also among small patient populations, as preliminary evidence
decreased by 10, 15, and 4% when comparing semaglutide [26, 29, 30]. In a double-blinded, crossover study,
0.5 mg, 1.0 mg, and placebo, respectively (P = 0.0046 and semaglutide was evaluated among 30 patients with obesity
P \ 0.0001 compared with placebo) [22]. [29]. Semaglutide 1 mg weekly was compared with pla-
SUSTAIN 7 was a randomized, open-label, active-con- cebo to determine the effect on energy intake and appetite
trolled trial in 1201 patients over 40 weeks [16]. Patients control. Overall, semaglutide reduced measures of appetite
were randomized to semaglutide 0.5 mg, dulaglutide on a visual analog scale (P = 0.0023) in comparison with
Semaglutide for Type 2 Diabetes

placebo. Nausea was the most common adverse event, but 1.0 mg had a 3.7 and 6.4 kg weight loss, respectively,
was similar between the two groups. Patients receiving compared with - 1.4 kg with placebo [22]. These effects
semaglutide had less hunger and fewer food cravings; due were observed over 30 weeks. In addition, semaglutide had
to reduced intake and improved control of eating, greater weight loss effects than dulaglutide over 40 weeks,
semaglutide resulted in a 5.0 kg weight loss from baseline as reported in the SUSTAIN 7 trial [16]. Those receiving
(101.3 kg), compared with a weight increase with placebo semaglutide 0.5 mg had an average weight loss of 4.6 kg
(1.0 kg). In another study, semaglutide 1 mg weekly was compared with - 2.3 kg with dulaglutide 0.75 mg
compared with placebo among 30 patients with obesity (P \ 0.0001). With a higher dose, semaglutide 1 mg pro-
[26]. Due to semaglutide and its mechanism of action, duced a 6.5 kg weight loss, compared with - 3.0 kg with
improvements in metabolic marks, such as fasting insulin, dulaglutide 1.5 mg (P \ 0.0001) [16].
C-peptide, and lipoprotein concentrations, are observed.
While there was no difference between semaglutide and
placebo in post-prandial gastric emptying, semaglutide did 8 Cardiovascular Disease
reduce gastric emptying within the first hour after a stan-
dardized high-fat breakfast. This preliminary evidence The SUSTAIN 6 trial was completed in accordance with
indicates the potential use of semaglutide in patients with FDA regulations requiring that all new medications for
obesity with and without diabetes. T2DM not be associated with increased cardiovascular risk
In a phase II study, approximately 1000 patients were [23]. This randomized, double-blind, placebo-controlled
randomized to various doses of semaglutide; this study was trial enrolled 3297 patients with T2DM aged 50 years and
double-blinded and lasted for 52 weeks in order to deter- older to receive semaglutide (dose escalation to 0.5 or 1 mg
mine the dose–response of semaglutide for weight loss subcutaneously weekly) or placebo (0.5 or 1 mg subcuta-
[30]. There were approximately nine groups with roughly neously weekly). Background therapy was provided in the
100 patients, and the dose of semaglutide varied from 0.05 form of current best practices for glycemic control (treat-
to 0.4 mg/day. One of the nine groups was the placebo ment with a DPP-4 was an exclusion criteria), blood pres-
comparator. All patients received counseling on lifestyle sure control, and lipid control. Drug therapy and other
modifications with caloric reduction and physical activity. patient characteristics were evenly distributed at baseline.
Baseline weight was 111 kg with a BMI of 39 kg/m2. After The primary outcome was a composite of major adverse
52 weeks of treatment, semaglutide produced a 17.8 kg cardiac event (MACE) comprised of death from cardio-
weight loss (approximately 16% from baseline), whereas vascular causes, non-fatal myocardial infarction, and non-
the placebo arm only produced a 2.3% weight loss [27]. As fatal stroke. Semaglutide was associated with lower rates of
the results have not been fully released, it is not known the primary outcome (6.6% vs. 8.9% [hazard ratio {HR}
what the proportion of patients with and without diabetes in 0.74, 95% CI 0.58–0.95; P \ 0.001 for non-inferiority,
the trial was. Semaglutide will be investigated among P = 0.02 for superiority) than placebo. Of the composite
patients who are overweight and those with obesity, in endpoint, non-fatal stroke was most significantly reduced,
phase III trials starting in 2018; it will be several years occurring in 1.6% of patients in the semaglutide arm versus
until the final datasets are reported and published. 2.7% in the placebo arm (HR 0.61, 95% CI 0.38–0.99;
Currently, there is evidence of weight loss among P = 0.04). Cardiovascular death was similar between the
patients with T2DM who have obesity from the SUSTAIN two arms (2.7 vs. 2.8%; P = 0.92). The secondary outcome,
trials. As mentioned earlier, there were significant differ- an expanded composite outcome including death from
ences in weight loss with semaglutide and the comparator cardiovascular causes, non-fatal myocardial infarction, non-
(P \ 0.0001) in the SUSTAIN 2, 3, and 4 trials [15, 20, 21]. fatal stroke, coronary or peripheral revascularization, and
Specifically, from these three trials, semaglutide 0.5 mg hospitalization for unstable angina or heart failure was also
produced a 3.47 kg (SUSTAIN 4) and 4.6 kg (SUS- reduced, occurring in 12.1% of participants in the
TAIN 2) weight loss in comparison with –6.1 kg (SUS- semaglutide arm and 16.0% of participants receiving pla-
TAIN 2) and –5.17 kg (SUSTAIN 4) with semaglutide cebo (HR 0.74, 95% CI 0.62–0.89; P = 0.002). This result
1 mg [20, 21]. It is important to note that the sitagliptin was driven by reduced revascularization among participants
group lost weight (1.9 kg) during the SUSTAIN 2 trial, in the semaglutide arm (5.0 vs. 7.6%) [HR 0.65, 95% CI
whereas patients in the insulin glargine group gained weight 0.50–0.86; P = 0.003]. In addition to primary and sec-
(1.15 kg; SUSTAIN 4) [20, 21]. In the SUSTAIN 3 trial, ondary outcomes, semaglutide showed statistically signifi-
semaglutide was compared with another once-weekly GLP- cant lowering of systolic blood pressure by an average of
1 RA (exenatide) and had greater reductions in weight of - 2.59 mmHg (P \ 0.001) and lipids by an average of
5.6 kg compared with –1.9 kg with exenatide [15]. In the - 0.92 mg/dL (P \ 0.001). While these values may not be
SUSTAIN 5 trial, patients receiving semaglutide 0.5 and clinically significant, they may contribute to the
 S. Hall et al.

cardiovascular benefits of semaglutide shown in this trial especially in light of the overall demonstrated cardiovas-
and are consistent with findings in the previous SUSTAIN cular benefits.
trials. In the SUSTAIN 6 trial, rates of retinopathy complica-
tions, including vitreous hemorrhage, blindness, or condi-
tions requiring treatment with an intravitreal agent or
9 Safety and Tolerability photocoagulation, were significantly higher in the
semaglutide group (3.0%, n = 50) than with placebo (1.8%,
The adverse events of semaglutide are similar to other n = 29) (HR 1.76, 95% CI 1.11–2.78; P = 0.02). Most had
GLP-1 RAs, with gastrointestinal adverse events occurring pre-existing retinopathy at baseline (84% in semaglutide
most frequently. In clinical trials, nausea peaked after group and 82.85% in placebo group). This is a unique
treatment initiation and diminished over time; it is finding that was not seen with other GLP-1 RAs and the
observed with both the oral and subcutaneous formulations. cause is not known; however, it is thought that large, rapid
Slow dose titration reduces the incidence of gastrointestinal improvements in HbA1c may cause transient worsening of
adverse effects [19–21, 23]. All GLP-1 RAs are con- diabetic retinopathy [23, 31].
traindicated in patients with a personal or family history of
medullary thyroid carcinoma (MTC) and in patients with
multiple endocrine neoplasia syndrome type 2 (MEN 2) 10 Clinical Implications
because of dose- and treatment duration-dependent thyroid
C cell tumors discovered in rats and mice. This finding has As summarized, semaglutide is a new long-acting GLP-
not been observed in humans related to any GLP-1 RA use. 1 RA with once-weekly administration made possible by
In the SUSTAIN 1 trial, nausea was reported in 26 its pharmacokinetic profile. Additional head-to-head stud-
(20%) patients randomized to semaglutide 0.5 mg, 31 ies should be conducted; more evidence will become
(24%) patients who received semaglutide 1.0 mg, and ten available as subcutaneous semaglutide 0.5 and 1 mg per
(8%) patients who received placebo, and diarrhea was week is being compared with dulaglutide 0.75 and 1.5
reported in 16 (13%) patients who received semaglutide weekly by subcutaneous injection in SUSTAIN 7. In this
0.5 mg, 14 (11%) patients who received semaglutide trial, semaglutide has demonstrated superior efficacy with a
1.0 mg, and three (2%) patients who received placebo. 1.8% HbA1c reduction at 1 mg per week when compared
There were no cases of pancreatitis and four cases of with both doses of dulaglutide [16]. This finding was sta-
cholelithiasis in the semaglutide groups compared with tistically significant, even though the results have not yet
none in the placebo group [19]. SUSTAIN 6 had a similar been published. In addition, this dose of semaglutide also
incidence of gallbladder disorder between groups (58 produced a 6.5 kg weight loss from baseline, after 40
patients in semaglutide vs. 61 in placebo), although slightly weeks of therapy. Additional trials are underway to assess
more cases of cholelithiasis (4.6 vs. 3.8%) [23]. More the role of semaglutide among patients with obesity, with
malignant neoplasms were reported in the semaglutide or without T2DM.
groups (four in total) than in the placebo group (none) in The change in HbA1c levels was the primary outcome of
SUSTAIN 1, while the much larger SUSTAIN 6 trial had the SUSTAIN 1, 2, 3, and 4 trials [15, 19–21]. In addition,
an equal distribution of malignant neoplasms. Similar rates there were other endpoints, such as the rate of hypo-
of hypoglycemia were observed as compared with placebo. glycemia, nausea, and change in body weight. Based on the
There were lipase and amylase increases observed with SUSTAIN trials, semaglutide is a superior option when
use of semaglutide, although no increases in the incidence used as monotherapy or as add-on therapy to oral anti-
of pancreatitis. Initial case reports of acute pancreatitis hyperglycemic agents or insulin. There is a predicted 1.6%
with exenatide, the first GLP-1 RA to come to market, HbA1c reduction and estimated 4.5 kg weight loss with
have been published so it is recommended that patients be semaglutide over 6–12 months in patients with T2DM.
monitored for signs of pancreatitis with all other GLP-1 While the efficacy and safety of semaglutide are positive,
RAs and discontinue use if suspected. A minor increase in long-term evidence is needed to determine the clinical
pulse rate and decrease in systolic blood pressure was effect on glycemic control beyond 52 weeks. In addition,
observed in clinical trials, comparable to findings with the role of semaglutide is not known among individuals
other GLP-1 RAs [19–21, 23]. For example, in the SUS- with HbA1c levels above 10% or with known proliferative
TAIN 6 trial the mean pulse rate in the semaglutide group retinopathy or maculopathy; these individuals were exclu-
was 2.0 beats per min (bpm) higher in the 0.5 mg group ded from the SUSTAIN trials [15, 16, 19–23].
and 2.5 bpm higher in the 1.0 mg group than with placebo As a disposable, pre-filled pen, semaglutide can be
(P \ 0.001 for both comparisons) [23]. It is not known if injected in the subcutaneous portion of the abdomen, thigh,
there is any clinical significance to the increased pulse rate, or upper arm for once-weekly administration [25]. Similar
Semaglutide for Type 2 Diabetes

to the SUSTAIN trials, a patient can be initiated on a 2. World Health Organization. Diabetes fact sheet. 2017. http://
0.25 mg dose for 4 weeks. After 4 weeks, the dose can be www.who.int/mediacentre/factsheets/fs312/en/. Accessed 18 Dec
2017.
increased to 0.5 mg and then 1 mg after an additional 3. American Diabetes Association. Standards of medical care in
4 weeks. If a single dose is missed, the patient should be diabetes—2017. Diabetes Care. 2017;40(Suppl 1):S1–142.
instructed and counseled to administer the medication 4. Abrahamson MJ, Barzilay JI, Blonde L, et al. AACE/ACE
within 5 days of the missed dose. Semaglutide will only be comprehensive type 2 diabetes management algorithm—2017.
Endocr Pract. 2017. https://doi.org/10.4158/EP16182.CS.
available in a pre-filled device in cartons of one to two 5. Marso SP, Daniels GH, Brown-Frandsen K, et al. Liraglutide and
pens. If dispensed as a carton of one pen, the carton should cardiovascular outcomes in type 2 diabetes. N Engl J Med.
be used as initial treatment of 0.25 mg to a maintenance 2016;375(4):311–22.
dose of 0.5 mg. If dispensed as a carton of two pens, the 6. Lau J, Bloch P, Schaffer L, et al. Discovery of the once-weekly
glucagon-like peptide-1 (GLP-1) analogue semaglutide. J Med
carton should be used as the maintenance dose of 1 mg per Chem. 2015;58(18):7370–80.
week. The pen can be stored in the refrigerator until use; 7. Marbury TC, Flint A, Jacobsen JB, Derving Karsbøl J, Lasseter
after the first use, the pen can be stored in the refrigerator K. Pharmacokinetics and tolerability of a single dose of
or at room temperature for 56 days [25]. The cost or semaglutide, a human glucagon-like peptide-1 analog, in subjects
with and without renal impairment. Clin Pharmacokinet.
insurance coverage of semaglutide is unknown as the 2017;56(11):1381–90.
medication will not be available until early 2018. 8. Jensen L, Helleberg H, Roffel A, et al. Absorption, metabolism
and excretion of the GLP-1 analogue semaglutide in humans and
nonclinical species. Eur J Pharm Sci. 2017;104:31–41.
9. Blundell J, Finlayson G, Axelsen M, et al. Effects of once-weekly
11 Conclusion semaglutide on appetite, energy intake, control of eating, food
preference and body weight in subjects with obesity. Diabetes
Semaglutide produced greater HbA1c reductions than active Obes Metab. 2017;19(9):1242–51.
groups (i.e., sitagliptin, exenatide ER, and insulin glargine) in 10. Kapitza C, Nosek L, Jensen L, Hartvig H, Jensen CB, Flint A.
Semaglutide, a once-weekly human GLP-1 analog, does not
patients with T2DM and thus far has shown consistent weight reduce the bioavailability of the combined oral contraceptive,
loss among patients with and without diabetes. In clinical ethinylestradiol/levonorgestrel. J Clin Pharmacol.
practice, there may be several advantages to semaglutide, 2015;55(5):497–504.
such as weight loss and low risk of hypoglycemia. 11. Lau J, Bloch P, Schaffer L, et al. Discovery of the once-weekly
glucagon-like peptide-1 (GLP-1) analogue semaglutide. J Med
Semaglutide has a predictable pharmacokinetic profile, Chem. 2015;58(18):7370–80.
allowing for a longer duration of action and once-weekly 12. Victoza (liraglutide) [prescribing information]. Bagsvaerd: Novo
subcutaneous administration. There are limited drug inter- Nordisk. 2017. http://www.novo-pi.com/victoza.pdf. Accessed 8
actions and no dosing adjustments in patients with renal or Mar 2018.
13. Trulicity (dulaglutide) [prescribing information]. Indianapolis:
liver impairment. Therefore, semaglutide is an option, most Eli Lily and Company. 2017. http://pi.lilly.com/us/trulicity-uspi.
likely as add-on therapy, for improved glycemic control, pdf. Accessed 8 Mar 2018.
weight loss, and reduced risk of hypoglycemia. Similarly, to 14. Gedulin BR, Smith PA, Jodka CM, et al. Pharmacokinetics and
other GLP-1 RAs, the most common adverse event associ- pharmacodynamics of exenatide following alternate routes of
administration. Int J Pharm. 2008;356(1–2):231–8.
ated with semaglutide is nausea. It may have more advantages 15. Ahmann AJ, Capehorn M, Charpentier G, et al. Efficacy and
for clinical use once additional data are published regarding safety of once-weekly semaglutide versus exenatide ER in sub-
the oral formulation and potential indication for obesity jects with type 2 diabetes (SUSTAIN 3): a 56-week, open-label,
management. Overall, semaglutide is an option, with diet and randomized clinical trial. Diabetes Care. 2018;41(2):258–66.
16. Pratley RE, Aroda VR, Lingvay I, et al. Semaglutide versus
exercise, for the management of T2DM. dulaglutide once weekly in patients with type 2 diabetes (SUS-
TAIN 7): a randomised, open-label, phase 3b trial. Lancet Dia-
Compliance with Ethical Standards betes Endocrinol. 2018;6(4):275–86. https://doi.org/10.1016/
S2213-8587(18)30024-X.
Funding No external funding was used in the preparation of this 17. Novo Nordisk. Semaglutide subcutaneous once-weekly treatment
manuscript. to improve glycemic control in adults with type 2 diabetes mel-
litus. Endocrinologic and Metabolic Drug Advisory Committee.
Conflict of interest Sylvie Hall, Diana Isaacs, and Jennifer N. Cle- 2017.
ments declare they have no conflicts of interest that might be relevant 18. Jensen L, Kupcova V, Arold G, Pettersson J, Hjerpsted JB.
to the contents of this manuscript. Pharmacokinetics and tolerability of semaglutide in people with
hepatic impairment. Diabetes Obes Metab. 2018;20(4):998–1005.
https://doi.org/10.1111/dom.13186.
References 19. Sorli C, Harashima SI, Tsoukas GM, et al. Efficacy and safety of
once-weekly semaglutide monotherapy versus placebo in patients
with type 2 diabetes (SUSTAIN 1): a double-blind, randomised,
1. American Diabetes Association. Statistics about diabetes. 2017.
placebo-controlled, parallel-group, multinational, multicentre
http://www.diabetes.org/diabetes-basics/statistics/. Accessed 18
phase 3a trial. Lancet Diabetes Endocrinol. 2017;5(4):251–60.
Dec 2017.
 S. Hall et al.

20. Ahren B, Masmiquel L, Kumar H, et al. Efficacy and safety of receptor agonists in type 2 diabetes: a systematic review and
once-weekly semaglutide versus once-daily sitagliptin as an add- mixed-treatment comparison analysis. Diabetes Obes Metab.
on to metformin, thiazolidinediones, or both, in patients with type 2017;19(4):524–36.
2 diabetes (SUSTAIN 2): a 56-week, double-blind, phase 3a, 28. Davies M, Pieber TR, Hartoft-Nielsen ML, Hansen OKH, Jab-
randomised trial. Lancet Diabetes Endocrinol. 2017;5(5):341–54. bour S, Rosenstock J. Effect of oral semaglutide compared with
21. Aroda VR, Bain SC, Cariou B, et al. Efficacy and safety of once- placebo and subcutaneous semaglutide on glycemic control in
weekly semaglutide versus once-daily insulin glargine as add-on patients with type 2 diabetes: a randomized clinical trial. JAMA.
to metformin (with or without sulfonylureas) in insulin-naive 2017;318(15):1460–70.
patients with type 2 diabetes (SUSTAIN 4): a randomised, open- 29. Blendell J, Finlayson G, Buhl M, et al. Semaglutide reduced
label, parallel-group, multicentre, multinational, phase 3a trial. appetite and energy intake, improves control of eating, and pro-
Lancet Diabetes Endocrinol. 2017;5(5):355–66. vides weight loss in subjects with obesity [poster no. 23-OR]. In:
22. Rodbar HW, Lingvay I, Reed J. Semaglutide once-weekly vs American Diabetes Association Annual Meeting; 9–13 Jun 2017;
placebo as add-on to basal insulin alone or in combination with San Diego.
metformin in subjects with type 2 diabetes (SUSTAIN 5) [poster 30. Company announcement. Novo Nordisk no. 50/2017. Bagsvaerd:
no. 766]. In: 52nd European Association for the Study of Dia- Novo Nordisk A/S Investor Relation. 2017.
betes Annual Meeting; 13–16 Sep 2016; Munich. 31. Vilsbøll T, Bain SC, Leiter LA, Lingvay I, Matthews D, Simó R,
23. Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardio- et al. Semaglutide, reduction in glycated haemoglobin and the
vascular outcomes in patients with type 2 diabetes. N Engl J Med. risk of diabetic retinopathy. Diabetes Obes Metab.
2016;375(19):1834–44. 2018;20(4):889–97. https://doi.org/10.1111/dom.13172.
24. Hausner H, Derving KJ, Holst AG, et al. Effect of semaglutide on 32. Adlyxin (lixisenatide) [prescribing information]. Bridgewater:
the pharmacokinetics of metformin, warfarain, atorvastatin, and Sanofi-Aventis US LLC. 2016.
digoxin in healthy subjects. Clin Pharmacokinet. 33. Bydureon (exenatide) [prescribing information]. Wilmington:
2017;56(11):1391–401. AstraZeneca Pharmaceuticals LP. 2017.
25. Ozempic [package insert]. Plainsboro: Novo Nordisk. 2017. 34. Byetta (exenatide) [prescribing information]. Wilmington:
26. Hjerpsted JB, Flint A, Brooks A, Axelsen MB, Kvist T, Blundell AstraZeneca Pharmaceuticals. 2015.
J. Semaglutide improves postprandial glucose and lipid metabo- 35. Ozempic (semaglutide) [prescribing information]. Bagsvaerd:
lism, and delays first-hour gastric emptying in subjects with Novo Nordisk. 2017.
obesity. Diabetes Obes Metab. 2018;20(3):610–9. 36. Tanzeum (albiglutide) [prescribing information]. Wilmington:
27. Htike ZZ, Zaccardi F, Papamargaritis D, Webb DR, Khunti K, GlaxoSmithKline. 2017.
Davies MJ. Efficacy and safety of glucagon-like peptide-1