Neuropharmacology 170 (2020) 107788

Contents lists available at ScienceDirect

Neuropharmacology
journal homepage: www.elsevier.com/locate/neuropharm

Endocrine effects of the novel ghrelin receptor inverse agonist PF-5190457: T

Results from a placebo-controlled human laboratory alcohol co-
administration study in heavy drinkers
Mary R. Leea, Mehdi Farokhniaa,b, Enoch Cobbinac, Anitha Saravanakumarc, Xiaobai Lid,
Jillian T. Battistaa, Lisa A. Farinellia, Fatemeh Akhlaghic, Lorenzo Leggioa,b,e,f,∗
a
Section on Clinical Psychoneuroendocrinology and Neuropsychopharmacology, National Institute on Alcohol Abuse and Alcoholism Division of Intramural Clinical and
Biological Research and National Institute on Drug Abuse Intramural Research Program, National Institutes of Health, Bethesda, MD, USA
b
Center on Compulsive Behaviors, National Institutes of Health, Bethesda, MD, USA
c
Clinical Pharmacokinetics Research Laboratory, Department of Biomedical & Pharmaceutical Sciences, College of Pharmacy, University of Rhode Island, Kingston, RI,
USA
d
Biostatistics and Clinical Epidemiology Service, Clinical Center, National Institutes of Health, Bethesda, MD, USA
e
Medication Development Program, National Institute on Drug Abuse Intramural Research Program, National Institutes of Health, Baltimore, MD, USA
f
Center for Alcohol and Addiction Studies, Department of Behavioral and Social Sciences, Brown University, Providence, RI, USA

HIGHLIGHTS

• We tested PF-5190457, a ghrelin receptor inverse agonist, in heavy drinkers.

• We assessed the effect of PF-5190457 on neuroendocrine hormones.
• Blood hormone levels were measured during dosing phase and alcohol administration.
• Neuroendocrine hormones remain largely unaffected by PF-5190457.

ARTICLE INFO ABSTRACT

Keywords: Both animal and human work suggests that the ghrelin system may be involved in the mechanisms that regulate
Ghrelin the development and maintenance of alcohol use disorder. Previously, in a Phase 1b study, we tested phar-
GHS-R1a macological blockade of the growth hormone secretagogue receptor 1a (GHS-R1a, also known as the ghrelin
PF-5190457 receptor), in heavy drinking individuals with PF-5190457, an orally bioavailable, potent and selective GHS-R1a
Alcohol
inverse agonist. We report here the effects of PF-5190457 on endocrine blood concentrations of amylin, gastric
Heavy drinking
inhibitory polypeptide, glucagon-like peptide 1, insulin, leptin, pancreatic polypeptide, peptide YY, thyroid
Alcohol use disorder
Hormones stimulating hormone, free triiodothyronine (T3), thyroxine (T4), cortisol, prolactin, and glucose during PF-
Neuroendocrinology 5190457 dosing, as compared to placebo, in absence of alcohol as well as during an alcohol challenge when PF-
5190457 was on steady-state. Blood hormone levels were largely unaffected by PF-5190457, both during dosing
and in the context of alcohol challenge. The safety-related relevance of these findings to further develop PF-
5190547 in alcohol use disorder is discussed.
Clinicaltrials.gov: NCT02039349.
This article is part of the special issue on ‘Neuropeptides’.

1. Introduction was originally discovered for its function in regulating growth hormone
(GH) secretion, subsequent work identified ghrelin as a hormone with
The peptide ghrelin is a gut hormone produced by endocrine cells pleiotropic functions that include (but are not limited to) regulation of
mainly localized in the stomach (Kojima et al., 1999). While ghrelin appetite, food intake, energy expenditure, and glucose homeostasis

∗
Corresponding author. Section on Clinical Psychoneuroendocrinology and Neuropsychopharmacology NIAAA & NIDA, NIH; 10 Center Drive (10CRC/15330) MSC
1108, Room 1-5429 Bethesda, MD, 20892-1108, USA.
E-mail address: lorenzo.leggio@nih.gov (L. Leggio).

https://doi.org/10.1016/j.neuropharm.2019.107788
Received 18 April 2019; Received in revised form 30 August 2019; Accepted 19 September 2019
Available online 23 September 2019
0028-3908/ Published by Elsevier Ltd.
M.R. Lee, et al.
(Leggio, 2010; Muller et al., 2015). Ghrelin is acylated by the ghrelin-O-
acyltransferase (GOAT) enzyme to acyl-ghrelin (hereafter mostly re-
ferred to as ‘ghrelin’), which then binds to its G-protein coupled GH
secretagogue receptor (GHS-R1a) to execute its physiological functions
(Howard et al., 1996).
Both animal and human work suggests that the ghrelin system may
be involved in the mechanisms that regulate the development and
maintenance of alcohol use disorder (AUD). As recently reviewed
(Farokhnia et al., 2019; Jerlhag, 2019; Morris et al., 2018; Zallar et al.,
2017), results from experiments conducted by independent laboratories
and using different rodent models indicate that ghrelin administration
leads to an increase in alcohol-related outcomes, such as alcohol con-
ditioned place preference (a proxy of reward), as well as alcohol intake,
preference, and self-administration. These effects are attenuated by
transgenic knock-out or pharmacological blockage (D-Lys3-GHRP-6,
BIM28163, or JMV2959) of the ghrelin receptor. In addition, human
studies have reported a relationship between endogenous ghrelin con-
centrations and alcohol craving, risk of relapse, subjective responses to
alcohol, and brain activity in response to alcohol cues, suggesting that
endogenous ghrelin signaling is positively associated with alcohol-re-
lated behaviors (for reviews, see (Farokhnia et al., 2019; Jerlhag, 2019;
Morris et al., 2018; Zallar et al., 2017)).
Further human studies support a role of the ghrelin system in AUD
by showing that manipulations of the ghrelin system lead to changes in
alcohol-related outcomes. Specifically, two double-blind, placebo-con-
trolled studies demonstrated that intravenous administration of ghrelin
increases cue-induced alcohol craving in a bar-like setting (Leggio et al.,
2014) and alcohol self-administration using a progressive ratio alcohol
intravenous paradigm (Farokhnia et al., 2018). In another experimental
human study, forced water intake led to reduction in endogenous
ghrelin concentrations (possibly due to the stomach distension), which
in turn was associated with reduced alcohol craving (Koopmann et al.,
2017).
Based on the rodent and human literature summarized above, we
tested the effects of GHS-R1a pharmacological blockade in heavy
drinking individuals. PF-5190457 is an orally bioavailable, potent, and
selective GHS-R1a inverse agonist (Bhattacharya et al., 2014) that in-
hibits the receptor's constitutive activity and competitively blocks its
activation by acyl-ghrelin (Kong et al., 2016). PF-5190457 is the first
GHS-R1a inverse agonist that progressed to a Phase 1a clinical trial,
conducted by the manufacturer, where PF-5190457 was well-tolerated
and safe in healthy volunteers (Denney et al., 2017). We also conducted
a translational study starting with a set of rat experiments to rule out
significant safety concerns due to potential interactions between PF-
5190457 and alcohol (Lee et al., 2018). Next, in heavy drinking in-
dividuals (Phase 1b human laboratory study), we examined the safety
of PF-5190457, alone and when administered with alcohol (Lee et al.,
2018).
In the Phase 1b human laboratory study, PF-5190457 or placebo
was administered up to steady-state (dosing phase), followed by oral
alcohol administration (alcohol challenge phase). We determined that
PF-5190457 (50 mg b.i.d. and 100 mg b.i.d.) alone and co-administered
with alcohol was safe and well tolerated. We also reported that there
were no pharmacokinetic (PK) changes in either PF-5190457 or alcohol
during their co-administration. Additionally, preliminary data in-
dicated that PF-5190457 may reduce cue-induced craving for alcohol
and food in a bar-like setting (Lee et al., 2018).
Furthermore, we investigated the potential effects of PF-5190457 on
endocrine pathways related to the ghrelin system, either directly (acyl-
ghrelin, total ghrelin, acyl-to-total ghrelin ratio) or indirectly (GH, in-
sulin-like growth factor 1 (IGF-1), the latter being a clinically used
surrogate measure of GH (Lee et al., 2018). We found an increase in
total ghrelin levels in the PF-5190457 100 mg condition, but limited to
the alcohol challenge phase, while during the dosing phase there was a
significant effect of PF-5190457 100 mg to reduce the acyl-to-total
ghrelin ratio. Additionally, during the alcohol challenge, there was a
2
Neuropharmacology 170 (2020) 107788
late increase in IGF-1 concentrations under both drug conditions (50 mg
b.i.d. and 100 mg b.i.d.) compared to placebo, suggesting that IGF-1
might represent a biomarker of the downstream effects of GHS-R1a
blockade with PF-5190457 in heavy drinkers (Lee et al., 2018). Overall,
these results indicated the safety of PF-5190457, alone and when co-
administered with alcohol. In summary, PF-5190457 affects ghrelin
signaling, but in a complex way that may become more intricate with
co-administration of alcohol.
We report here the effects of PF-5190457 on a wide range of other
endocrine outcomes investigated in this study (Lee et al., 2018). Spe-
cifically, given the effect of ghrelin to inhibit glucose-stimulated insulin
secretion and to impair glucose tolerance, we measured blood insulin
and glucose concentrations, as well as other hormones with known
effects on glucose homeostasis, appetite, and food intake. These hor-
mones include amylin, gastric inhibitory polypeptide (GIP), glucagon-
like peptide 1 (GLP-1), insulin, leptin, pancreatic polypeptide (PP),
peptide YY (PYY), thyroid stimulating hormone (TSH), free triio-
dothyronine (T3), and thyroxine (T4). Finally, given the role of ghrelin
in stress regulation, as well as the emerging complex interactions be-
tween ghrelin, stress pathways, and AUD (for review, see (Morris et al.,
2018)), we also measured blood concentrations of the stress hormones
cortisol and prolactin.
The endocrine-focused analyses presented here are important at
several levels, as they may contribute to a better understanding of: 1)
the safety of PF-5190457 administration, with respect to its effect on
these endocrine pathways, when administered alone or co-administered
with alcohol; 2) the potential indirect endocrine effects of GHS-R1a
blockade via PF-5190457 that may contribute to the downstream me-
chanisms of action of this novel compound; and 3) the potential for
using specific hormones as peripheral endocrine biomarkers of PF-
5190457's activity in heavy alcohol drinkers and in the context of al-
cohol use.
2. Materials and methods
2.1. Study design
This was a Phase 1b, within-subject, dose-escalating, single-blind,
placebo-controlled human laboratory study with three oral drug con-
ditions: placebo b.i.d., PF-5190457 50 mg b.i.d., and 100 mg b.i.d.,
conducted at the NIH Clinical Center, Bethesda, Maryland, USA.
2.2. Study approvals
This human study was approved by the NIH Addictions Institutional
Review Board and monitored by a Data Safety Monitoring Board.
Written informed consent was obtained from participants prior to in-
clusion in the study. The study was conducted under a federal
Certificate of Confidentiality. The use of PF-5190457 in this study was
under the Food and Drug Administration Investigational New Drug
119 365. Alcohol administration procedures were consistent with the
NIAAA Council Guidelines on Alcohol Administration: (https://www.
niaaa.nih.gov/Resources/ResearchResources/job22.htm).
2.3. Study sample
1. Eligible participants were male and female active, heavy alcohol
drinkers (≥21 drinks/week for men or ≥ 14 drinks/week for
women, based on the 90-day timeline follow-back at screening) and
were not seeking treatment for AUD. Recruitment took place via
word of mouth, as well as local and online advertisements. After a
preliminary phone pre-screening, potentially eligible participants
came to NIH and signed an informed consent to enroll in a screening
protocol for determination of suitability for enrollment in the clin-
ical study with PF-5190457. A full list of the inclusion and exclusion
criteria is reported in the Supplementary Appendix 1.
M.R. Lee, et al.
2. Fourteen eligible participants signed the informed consent and
twelve completed the study. Demographics and baseline character-
istics are reported in the supplementary material of the main paper
(Lee et al., 2018) and here in the Supplementary Table S1. In ad-
dition to being heavy drinkers, all but one participant had a current
diagnosis of alcohol dependence based on the Diagnostic and Sta-
tistical Manual of Mental Disorders IV Edition – Text Revision (DSM-
IV-TR) criteria.
2.4. Study procedures
After signing the study-specific informed consent, research proce-
dures started. There were three identical inpatient visits (Visits 1–3)
separated by at least three days. The inpatient setting ensured full
compliance with the study medication, procedures, and alcohol ab-
stinence, and permitted strict safety monitoring. At each of the three
visits, PF-5190457/placebo administration took place twice on Day 1,
twice on Day 2, and on the morning of Day 3 (dosing phase). On Day 3,
approximately 30 min after the administration of PF-5190457, a drink
containing the same type of alcohol (40% Alcohol by Volume; 80%
Proof) for all subjects was administered (alcohol challenge phase). The
quantity of alcohol was calculated based on total body water for each
subject to achieve a blood alcohol concentration of approximately
0.06 mg% (Watson, 1989). At each of the three visits, participants were
discharged on Day 4 after a clinical assessment. Approximately 1–2
weeks after Visit 3, a brief outpatient clinical assessment took place,
during which participants’ health, safety, and alcohol consumption
were evaluated and counseling was delivered per NIAAA guidelines:
(https://pubs.niaaa.nih.gov/publications/Practitioner/
CliniciansGuide2005/clinicians_guide.htm). Participants received
monetary compensation for participation in this study.
The study timeline is outlined in the supplementary material of the
main paper (Lee et al., 2018) and here in the Supplementary Tables
S2–S3.
Additional details, procedures, and assessments, not relevant to the
analysis reported here, are described in the main paper (Lee et al.,
2018).
2.4.1. Timepoints for blood collection during the dosing phase
At each of the three visits, during the PF-5190457/placebo dosing
phase (Days 1–3), research blood samples were collected and processed,
and plasma samples were stored at −80 °C for future analyses.
Specifically, these research blood samples were collected at six time-
points in the morning of each day, Day 1–3: 90 min before the 1st dosing
(T1), 75 min after the 1st dosing (T2), 90 min before the 3rd dosing
(T3), 75 min after the 3rd dosing (T4), 60 min before the 5th dosing
(T5) and 15 min after the 5th dosing (T6; before the alcohol adminis-
tration began). The differences in the timing of blood samples collection
reflect the fact that the T6 blood draw (+15 min after the fifth dosing)
was drawn on the morning of the alcohol challenge. The alcohol chal-
lenge began 30 min after study drug dosing, therefore, required that the
T6 blood draw to be 15 min after dosing to acquire another morning
blood sample before alcohol administration.
Capillary glucose concentrations were also assessed with finger-
sticks at the four timepoints during Day 1 and Day 2, specifically at
−30, +30, +690, and +750. Furthermore, blood samples for clinical
labs (including hormones as detailed below) were transferred to the
NIH Clinical Center Department of Laboratory Medicine, immediately
after collection, and processed and analyzed there, as part of the pro-
tocol clinical assessments at three timepoints (T1, T3, and T5).
2.4.2. Timepoints for blood collection during the alcohol challenge phase
At each of the three visits, during the alcohol challenge phase (Day
3), research blood samples were collected from a peripheral vein,
processed and stored at −80 °C in our laboratory for future analyses.
Blood was drawn 60 min before and 15 min after study drug
3
Neuropharmacology 170 (2020) 107788
administration. Then, alcohol was administered 30 min after study drug
administration and blood samples were taken at: +30, +60, +120,
+180, +240, +360, +480, +1350 and + 1440. Day 3 glucose
monitoring was done at −30, +30, +60, +120, +180, +240, +360,
+480, +1350, +1440.
2.4.3. Standardization and timing of meals
During each inpatient stay, participants received meals standardized
for macronutrients and caloric content. For additional details, please
see (Lee et al., 2018). The timing of the meals relative to PF-5190457
dosing was as follows: breakfast was at −120, before daily PF-5190457
dosing; Lunch was eaten at +210, and dinner at +570.
2.5. Assays for blood hormone assessments
The research plasma samples were used to measure active amylin
(here referred to as amylin), GIP, active GLP-1 (here referred to as GLP-
1), insulin, leptin, PP and PYY. These plasma samples derived from
blood tubes centrifuged within 30 min post-collection (relative cen-
trifugal force: 1700×g, temperature: 4 °C, centrifugation time: 15 min),
and pre-treated with 4-(2-aminoethyl)benzenesulfonyl fluoride hydro-
chloride (Roche Diagnostics GmbH, Germany – Pefabloc® SC), di-
peptidyl peptidase IV inhibitor (EMD Millipore Corp., Billerica, MA –
Cat. #DPP4-010), and a protease inhibitor cocktail (Sigma-Aldrich Inc.,
Saint Louis, MO – Cat. #P8340) prior to blood collection. Samples were
inverted 10 times and kept on ice after collection. These hormones were
measured using a Millipore Human Metabolic Hormone Magnetic Bead
Panel 96-Well Plate MILLIPLEX® MAP kit (EMD Millipore Corp.,
Billerica, MA – Cat. #HMHEMAG-34K). The assay was performed on
fluorescence-coded magnetic beads coated with capture antibodies
specific for each marker. Introduction of biotinylated detection anti-
body and streptavidin-phycoerythrin permitted simultaneous detection
of all analytes on the MAGPIX® instrument (Luminex Corp., Austin, TX).
These multiplex data were pre-processed and analyzed in the MILLIP-
LEX® Analyst software (Version 3.5 – EMD Millipore Corp., Billerica,
MA) to calculate the concentration of each hormone. Capillary glucose
concentrations were assessed with a StatStrip Glucose Meter, in con-
junction with StatStrip Xpress Test Strips (Nova Biomedical), using the
Point of Care Testing nursing procedures at the NIH Clinical Center.
The thyroid panel (TSH, free T3, and thyroxine) and the stress
hormones, cortisol and prolactin, were assayed at the NIH Clinical
Center Department of Laboratory Medicine. Electrochemiluminescence
immunoassays were run on a cobas® 600 analyzer (Roche Diagnostics
International Ltd., Switzerland) to measure TSH, free T3, and thyroxine
concentrations. Chemiluminescence immunoassays were run on an
IMMULITE® 2000 XPi (Siemens Healthcare Diagnostics Inc., Tarrytown,
NY) to measure cortisol and prolactin.
2.6. Statistical analysis
The serial measures of blood hormone levels from the dosing phase
and the alcohol challenge phase were analyzed separately. Mean and
standard deviation for each hormone outcome was reported at each
assessment timepoint for each DOSE condition (placebo, PF-5190457
50 mg b.i.d., and 100 mg b.i.d.). To assess the treatment effect in these
hormones, data were analyzed using linear mixed effects models using
DOSE, TIME, and TIME × DOSE interaction as the fixed effects. Within-
subject correlation among the repeated measures were taken into ac-
count by the subject random effect. For the alcohol challenge phase, the
last two timepoints were collected the morning after (Day 4) and there
was a 14-h gap between the last Day 3 blood draw (7th) and the last
two (8th and 9th) timepoints (Day 4). Therefore, we did not include the
last two timepoints in the main analyses in order to maintain relatively
equal time intervals between the data points. That is, without including
the remote two timepoints, we avoided making extrapolations for that
14-h gap. However, the 8th and 9th timepoints are still represented in
M.R. Lee, et al. Neuropharmacology 170 (2020) 107788

Table 1
Main effect of TIME, DOSE, and TIME × DOSE interaction on blood hormones
and glucose concentrations during the dosing phase.
Variable Main effect INTERACTION Transformation
TIME X DOSE
TIME DOSE

Amylin <0.001 0.008 0.866 Log-transformed

GIP <0.001 0.970 0.820 Square root transformed
GLP-1 0.013 0.152 0.460 Square root transformed
Insulin <0.001 <0.001 0.220 Square root transformed
Leptin <0.001 0.780 0.970 Log-transformed
PP 0.16 0.270 0.400 Square root transformed
PYY 0.048 0.122 0.140 Square root transformed
TSH <0.015 0.011 0.051 Square root transformed
free T3 0.990 0.930 0.630 None
Thyroxine 0.040 0.860 0.320 None
Cortisol <0.001 0.640 0.450 None
Prolactin 0.040 0.490 0.850 None
Glucose <0.001 0.130 0.990 Square root transformed
Fig. 1. Blood concentration [Mean (±SD)] of thyroid stimulating hormone
(TSH) at 90 min before (timepoints 1,3,5) drug dosing on Day 1,2,3. Drug doses:
Placebo, PF-5190457, 50 mg and 100 mg.
Table 2
Main effect of TIME, DOSE, and TIME × DOSE interaction on blood hormones
and glucose concentrations during the alcohol challenge phase. doses of PF-5190457, TSH levels remained similar to baseline (Fig. 1).
Variable Main Effect INTERACTION Transformation This finding was observed in the absence of a main effect of DOSE or
TIME X DOSE TIME × DOSE interaction for free T3 or thyroxine. There was a sig-
TIME DOSE nificant main effect of DOSE, as well for TSH (p = 0.011). There were
also significant main effects of DOSE for amylin (p = 0.008) and insulin
Amylin 0.034 0.030 0.950 Log-transformed
GIP <0.001 0.040 0.300 Square root transformed (p < 0.001), so that there were lower concentrations of insulin (Fig. 2)
GLP-1 <0.001 0.560 0.710 Square root transformed and amylin (Fig. 3) under PF-5190457, as compared to placebo. There
Insulin <0.001 <0.001 0.073 Square root transformed was a significant main effect of TIME for all but 2 of the hormones, as
Leptin <0.001 <0.001 0.720 Log-transformed detailed in Table 1.
PP <0.001 0.220 0.460 Square root transformed
PYY 0.330 0.160 0.590 Square root transformed
The supplementary figures outline the non-significant results related
Glucose <0.001 0.170 0.100 Square root transformed to the lack of a main effect of DOSE for blood concentrations of GIP
(Fig. S1A), GLP-1 (Fig. S1B), leptin (Fig. S1C), PP (Fig. S1D), PYY (Fig.
S1E), free T3 (Fig. S1F), thyroxine (Fig. S1G), cortisol (Fig. S1H), pro-
the figures for descriptive purposes and to allow for a visual inspection. lactin (Fig. S1I), and glucose (Fig. S1J).
The logarithm or square root transformation was applied to the Finally, the heatmap (amylin, GIP, GLP-1, insulin, leptin, PP, and
outcome variables when the model normality assumption was violated PYY only) and the related dendogram in Fig. 4 showed a similar effect
(see also Tables 1–2). Values below the lower limit of quantitation for for leptin, amylin, and insulin, where AUC decreased in the PF-
each assay were set to 1/2 of the lower limit of quantitation (Keizer 5190457 100 mg b.i.d. condition compared to the placebo b.i.d. con-
et al., 2015). A p value of 0.05 or lower was considered statistically dition. Conversely, AUC for PP, GLP, PYY, and GIP in the PF-
significant. All statistical analyses were conducted in SAS (version 9.4, 5190457 100 mg b.i.d. condition was greater than the placebo condi-
SAS Institute Inc., Cary, NC). tion.
Finally, we conducted a heatmap analysis which was limited to the
hormones for which six timepoints were available during the dosing
phase (amylin, GIP, GLP-1, insulin, leptin, PP, and PYY only). We did
not include hormone measurements during the alcohol challenge on
Day 3 (due to the potential confounding effect of alcohol on these
measures) and we examined only the placebo and the highest dose of
PF-5190457 (100 mg b.i.d.). Specifically, the aggregate drug effect was
summarized by the area under the hormone-concentration-time curve
(AUC) for the PF-5190457 100 mg b.i.d. and placebo b.i.d. conditions.
The difference in the log-transformed AUCs between the two conditions
was calculated for each subject. We used a heatmap analysis to examine
whether the effect of PF-5190457 dosing compared to placebo was si-
milar amongst any of the hormones, with respect to the direction and
magnitude of the effect. The heatmap clustering analysis was conducted
using the heatmap package with RStudio (http://www.rstudio.org/,
integrated development environment (IDE) for R, Boston, MA, USA).

3. Results

3.1. Dosing phase

Results (p values) are outlined in Table 1. TSH was the only hor- Fig. 2. Blood concentration [Mean (±SD)] of insulin at 90 min before (time-
mone with a nominal trend TIME × DOSE interaction (p = 0.051). In points 1,3,5) and 75 min after (timepoints 2,4,6) drug dosing on Day 1,2,3.
the placebo condition, TSH decreased over time, while under the 2 Drug doses: Placebo, PF-5190457, 50 mg and 100 mg.

4
M.R. Lee, et al.
Fig. 3. Blood concentration [Mean (±SD)] of amylin at 90 min before (time-
points 1,3,5) and 75 min after (timepoints 2,4,6) drug dosing on Day 1,2,3.
Drug doses: Placebo, PF-5190457, 50 mg and 100 mg.
Fig. 4. Heatmap representation of the difference in the hormone-concentration-
time curve (AUC) encompassing Days 1,2,3 for the PF-5190457 100 mg minus
the placebo conditions for each of the 12 subjects who completed the study
(N101-N114). The heatmap analysis was limited to the hormones for which six
timepoints were available during the dosing phase: amylin, gastric inhibitory
polypeptide (GIP), glucagon-like peptide 1 (GLP-1), insulin, leptin, pancreatic
polypeptide (PP), and peptide YY.
3.2. Alcohol challenge phase
Results (p values) are outlined in Table 2. Insulin was the only
hormone with a trend significant TIME × DOSE interaction
(p = 0.073), as well as a significant main effect (p < 0.001), where PF-
5190457, co-administered with alcohol, reduced blood insulin levels
(Fig. 5a). There was a significant main effect of TIME for all variables
except PYY, as detailed in Table 2.
There was a significant main effect of DOSE for amylin (p = 0.03),
GIP (p = 0.04), and leptin (p < 0.001), with higher concentrations of
amylin (Fig. 5b) and lower concentrations of GIP (Fig. 5c) and leptin
(Fig. 5d) under PF-5190457, co-administered with alcohol. The sup-
plementary material includes the figures of the non-significant results
related to the lack of a main effect of DOSE for blood concentrations of
GLP-1 (Fig. S1B), PP (Fig. S1D), PYY (Fig. S1E) and glucose (Fig. S1J).
5
Neuropharmacology 170 (2020) 107788
4. Discussion
This report provides additional information on the effect of a GHS-
R1a inverse agonist (PF-5190457) (Lee et al., 2018) on blood con-
centrations of endocrine hormones when PF-5190457 is dosed to steady
state and administered with alcohol. The study population comprised of
individuals who were heavy drinkers and mostly (11 out of 12) with a
DSM-IV-TR diagnosis of alcohol dependence. Notwithstanding the ex-
ploratory and non-comprehensive nature of this work, the present
secondary analysis represented an opportunity to expand our knowl-
edge on the safety and endocrine effects of GHS-R1a blockade via PF-
5190457, in heavy drinkers and during an acute alcohol administration.
There was evidence, as we previously reported (Lee et al., 2018),
that PF-5190457 modulated endogenous ghrelin levels, as there was a
dose-dependent reduction in the acyl-to-total ghrelin ratio. Aside from
this finding, the majority of the hormones reported here were not af-
fected by PF-5190457, either during the dosing phase or during the
alcohol challenge phase. The Phase 1a first-in-human study in healthy
controls also reported no significant effects of PF-5190457 on blood
levels of a partially overlapping list of neurotransmitters and hormones:
dopamine, noradrenaline, adrenaline, glucagon, C-peptide, insulin, and
PP (Denney et al., 2017). A comprehensive, side-by-side comparison of
the two studies is not possible, given differences in study design, times
of blood sample collection, and fasting conditions.
Most importantly, unlike the previous Phase 1a study in healthy
volunteers (Denney et al., 2017), the present Phase 1b study was con-
ducted in heavy drinkers, most of whom (11 out of 12) were alcohol
dependent (DSM-IV-TR) and included an alcohol administration ex-
periment. AUD may be associated with several changes in a variety of
endocrine systems (for a review, see (Kenna et al., 2012)). As such,
assessing the safety of PF-5190457 administration in heavy drinkers in
terms of endocrine endpoints, including when co-administered with
alcohol, offers highly relevant information given the role of PF-
5190457 as a new compound currently under study to investigate GHS-
R1a inverse agonism as a novel pharmacotherapy for AUD patients.
Indeed, the lack of clinically-relevant safety concerns when PF-5190457
was co-administered with alcohol is critical, given the importance of
ruling out clinically-relevant drug-alcohol interactions early on, from a
medication development perspective (Haass-Koffler et al., 2017).
The significant TIME × DOSE interaction indicates that PF-5190457
affected the time course of blood TSH levels over the 3 days of study.
Specifically, in the placebo condition, TSH levels fell over the 3 days.
This decrease would be expected in this largely alcohol dependent
population, who were current heavy drinkers. Previous studies have
reported that individuals with AUD have elevated TSH levels compared
to controls, and that during acute abstinence, TSH levels fall. Under PF-
5190457, this decrease was attenuated. Human studies administering
ghrelin reported a decrease in TSH and an increase in T4 (Kluge et al.,
2010). Accordingly, an inverse agonist theoretically could increase
TSH, however we found that this biochemical increase in TSH was not
accompanied by any effect of PF-5190457 on either free T3 or T4. As
such, while it is unlikely that PF-5190457 may affect the thyroid
function, these results suggest that TSH might represent a potential
marker of PF-5190457's activity in this population of heavy drinkers.
Furthermore, as outlined in Fig. 1, it is possible that the effect of PF-
5190457 on TSH here described is apparent only at the highest dose
(100 mg).
It has been shown that central administration of TSH reduces food,
but not water, intake in rats (Lin et al., 1983), suggesting that TSH may
act through a central mechanism to influence physiological or beha-
vioral functions like appetite and food intake (Amin et al., 2011; Lin
et al., 1983). In the parent paper (Lee et al., 2018), we reported that, on
Day 2 (dosing phase), PF-5190457 versus placebo reduced both al-
cohol- and food cue-induced cravings. Albeit speculative, it is possible
to hypothesize that TSH signaling may centrally contribute to the ef-
fects of PF-5190457 on alcohol- and food-seeking behaviors, without
M.R. Lee, et al.
Fig. 5. Blood concentrations [Mean (±SD)] of A) Insulin, B) Amylin, C) Gastrointestinal Peptide (GIP), and D) Leptin during the alcohol challenge, which was
performed on Day 3. Alcohol was administered at +30, lunch at +210, and dinner at +570. Timepoints +1350 and + 1440 were not included in the analysis and
their inclusion in the figures is only for illustrative purposes.
affecting free T3 and thyroxine, and their peripheral effects on energy
expenditure and/or metabolism. Notably, while the role of the hy-
pothalamic-pituitary-thyroid axis on appetite is well-known, previous
work from our group indicated a negative relationship between blood
TSH levels and alcohol craving (Aoun et al., 2015; Leggio et al., 2008),
and higher TSH levels in AUD patients compared to controls (Lee et al.,
2015). Although preliminary and speculative, the latter previous find-
ings fit well with the present data, where PF-5190457 increased TSH
levels (present results) while it reduced alcohol craving (reported pre-
viously in (Lee et al., 2018)).
PF-5190457 significantly reduced blood concentrations of glucose-
regulating hormones, chiefly insulin and amylin, during the dosing
phase. This result was also apparent in the heatmap analysis, where the
AUC variable, effectively a time average for these two hormones, was in
the same cluster. During the alcohol challenge phase, insulin levels
were higher in the placebo condition both before and after lunch (lunch
was at timepoint 210). The reason for the late (timepoint 480) rebound
in insulin levels, which drives the trend-level DOSE × TIME interaction,
is unclear. This rebound was also seen in amylin, although at an earlier
timepoint, but there was no significant interaction. It is also important
to keep in mind the expected variability in hormone concentrations
across individuals, which may limit the interpretation of the results. For
example, although the small sample does not allow for additional
analysis of sub-groups, it is worth noticing that a visual inspection of
the heatmap dendrogram suggests that the decreasing trend of leptin,
amylin and insulin concentrations observed in the last 7 individuals is
not exhibited by the first 5 individuals.
The results presented here represent biochemical changes in insulin
6
Neuropharmacology 170 (2020) 107788
and amylin levels, as they were not accompanied by changes in glucose
levels, therefore it is unlikely that PF-5190457 affects peripheral glu-
cose homeostasis. In other words, while the present findings and pre-
vious work (Broglio et al., 2001; Haass-Koffler et al., 2016; Tong et al.,
2010) support a link between the ghrelin system and insulin signaling
in humans, the effect of the inverse agonist on insulin levels is the
opposite from what would be expected given the effect of ghrelin, itself,
to reduce insulin levels. This inconsistency most likely reflects: 1) the
complexity of pharmacological blockade of GHS-R1a, whose down-
stream effects on insulin signaling may considerably vary and 2) the
difficulty of comparing the dosing phase and the alcohol challenge
phase within our own present study, given the different experimental
conditions (e.g., timepoints and caloric control) applied.
The results discussed above are important and potentially inter-
esting, considering they are drawn from humans and from a clinically-
relevant sample. However, the considerations above remain speculative
and bed-to-bench approaches are needed to examine mechanisms un-
derlying the effect of PF-5190457 on hormones like TSH, insulin,
amylin, and probably others.
We speculate that the increase in TSH, in the absence of changes in
free T3 and thyroxine levels, as well as the reduction in insulin and
amylin levels, in the absence of effects on blood glucose levels, suggest
that PF-5190457 may not result in total or continuous blockade at GHS-
R1a. The previous Phase 1a first-in-human study was terminated due to
tachyphylaxis, which was limited to peripheral, but not central effects
(Denney et al., 2017). Therefore, there is the possibility that the effects
reported here on TSH and insulin, which do not result in downstream
metabolic effects, are a consequence of central blockade of GHS-R1a,
M.R. Lee, et al.
which may be incomplete; that is, PF-5190457 may cause continuous
blockade at the peripheral GHS-R1a, whereas central blockade is not
continuous, allowing for enduring efficacy (Denney et al., 2017). In-
deed, our rat experiments indicate some PF-5190457 brain penetrance;
however, this work was limited to measuring PF-5190457 concentra-
tions in brain homogenates and we did not use any isotopically labeled
PF-5190457. Finally, both the present results and the previous Phase 1a
study (Denney et al., 2017) did not find an effect of PF-5190457 on
blood PP levels, which is largely under vagal control (Schwartz, 1983)
and is considered a marker of peripheral vagus nerve activation
(Tamboli et al., 2017).
PF-5190457 is a novel compound and the information provided
here is the first of its kind, given the specific population we studied and
the experimental design that included alcohol administration.
Furthermore, the study was conducted in a well-controlled research
inpatient setting, which strengthened compliance, consistency across
visits, and safety monitoring. However, this work also has important
limitations. First, the sample size was small (N = 12 completers), al-
though the within-subject design attenuated this limitation. Some of the
hormone results (TSH, free T3, thyroxine, cortisol, and prolactin) were
available for the dosing phase, but not for the alcohol challenge phase;
this is a weakness that limits the interpretation of the present findings
but was necessary due to the safety-related need to limit the amount of
blood drawn in a clinical research protocol. We did not correct for
multiple testing, an approach consistent with the exploratory nature of
these analyses. Therefore, caution needs to be exercised given the risk
for false positive results. Furthermore, the lack of an alcohol-matched
placebo (to make the design fully factorial) calls for future larger,
randomized studies that consider these limitations. Finally, it is im-
portant to note that, ~90% of study participants in our study were
African-Americans and, given the strict inclusion and exclusion criteria,
recruitment of females was very difficult (only one out of 12 com-
pleters). Therefore, these results may not be generalizable, and future
studies are needed to further investigate PF-5190457 across individuals
with different race, as well as potential sex differences.
Author contributions
Study neuroscientific basis, rationale, concept and design: LL; Provided
funding: LL, FA; Statistical analysis: XL; Acquisition and management of
data: MRL, MF, EC, AK, JTB, LAF, FA, LL; Clinical and Safety monitoring:
MRL, LL; Administrative, technical, or material support: MRL, MF, EC, AK,
XL, JTB, LAF, FA, LL; Analysis and interpretation of data: MRL, MF, EC,
AK, XL, JTB, LAF, FA, LL; Drafting the manuscript: MRL, LL. All authors
have critically reviewed the manuscript for important intellectual
content and approved the final version of the manuscript.
Funding source
This work was supported by NIH intramural funding ZIA-AA000218
(Section on Clinical Psychoneuroendocrinology and
Neuropsychopharmacology; PI: Dr. Lorenzo Leggio), jointly supported by
the Division of Intramural Clinical and Biological Research of the
National Institute on Alcohol Abuse and Alcoholism (NIAAA) and the
Intramural Research Program of the National Institute on Drug Abuse
(NIDA); and by the National Center for Advancing Translational
Sciences (NCATS) grant UH2/UH3-TR000963 (PIs: Drs Lorenzo Leggio
and Fatemeh Akhlaghi). The content of this article is solely the re-
sponsibility of the authors and does not necessarily represent the offi-
cial views of the National Institutes of Health.
Conflicts of interest
The authors declare that they have no conflict of interest.
7
Neuropharmacology 170 (2020) 107788
Acknowledgments
The authors would like to thank the clinical and research staff in-
volved in data collection and support at the National Institute on
Alcohol Abuse and Alcoholism (NIAAA) Division of Intramural Clinical
and Biological Research, i.e. in the NIAAA/NIDA Section on Clinical
Psychoneuroendocrinology and Neuropsychopharmacology and in the
NIAAA clinical intramural program. The authors would also like to
thank the research staff involved in samples and data analysis in the
Clinical Pharmacokinetics Research Laboratory at the University of
Rhode Island. The authors would like to thank the clinical and research
staff involved in data collection and patient care at the NIH Clinical
Center, i.e. in the Department of Nursing (the nurses of the 1SE
Inpatient Unit and of the 1-HALC 1SE Outpatient Clinic), in the
Department of Nutrition, and in the Department of Pharmacy.
Furthermore, the authors would like to express their gratitude to the
participants who took part in this study. Finally, the authors would like
to thank the Steering Committee of the UH2/UH3-TR000963 grant (PIs:
Drs Lorenzo Leggio and Fatemeh Akhlaghi) whose members included
members from the NIAAA Division of Medication Development (in
particular Dr. Joanne Fertig), the Drug Development Partnership
Programs of the National Center for Advancing Translational Sciences
(NCATS) and Pfizer, which kindly provided the study drug under the
NCATS grant UH2/UH3-TR000963. Pfizer did not have any role in the
study design, execution, or interpretation of the results, and this pub-
lication does not necessarily represent the official views of Pfizer.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.neuropharm.2019.107788.
References
Amin, A., Dhillo, W.S., Murphy, K.G., 2011. The central effects of thyroid hormones on
appetite. J. Thyroid Res. 2011, 306510.
Aoun, E.G., Lee, M.R., Haass-Koffler, C.L., Swift, R.M., Addolorato, G., Kenna, G.A.,
Leggio, L., 2015. Relationship between the thyroid axis and alcohol craving. Alcohol
Alcohol 50, 24–29.
Bhattacharya, S.K., Andrews, K., Beveridge, R., Cameron, K.O., Chen, C., Dunn, M.,
Fernando, D., Gao, H., Hepworth, D., Jackson, V.M., Khot, V., Kong, J., Kosa, R.E.,
Lapham, K., Loria, P.M., Londregan, A.T., McClure, K.F., Orr, S.T., Patel, J., Rose, C.,
Saenz, J., Stock, I.A., Storer, G., VanVolkenburg, M., Vrieze, D., Wang, G., Xiao, J.,
Zhang, Y., 2014. Discovery of PF-5190457, a potent, selective, and orally bioavail-
able ghrelin receptor inverse agonist clinical candidate. ACS Med. Chem. Lett. 5,
474–479.
Broglio, F., Arvat, E., Benso, A., Gottero, C., Muccioli, G., Papotti, M., van der Lely, A.J.,
Deghenghi, R., Ghigo, E., 2001. Ghrelin, a natural GH secretagogue produced by the
stomach, induces hyperglycemia and reduces insulin secretion in humans. J. Clin.
Endocrinol. Metab. 86, 5083–5086.
Denney, W.S., Sonnenberg, G.E., Carvajal-Gonzalez, S., Tuthill, T., Jackson, V.M., 2017.
Pharmacokinetics and pharmacodynamics of PF-05190457: the first oral ghrelin re-
ceptor inverse agonist to be profiled in healthy subjects. Br. J. Clin. Pharmacol. 83,
326–338.
Farokhnia, M., Faulkner, M.L., Piacentino, D., Lee, M.R., Leggio, L., 2019. Ghrelin: from a
gut hormone to a potential therapeutic target for alcohol use disorder. Physiol.
Behav. 204, 49–57.
Farokhnia, M., Grodin, E.N., Lee, M.R., Oot, E.N., Blackburn, A.N., Stangl, B.L., Schwandt,
M.L., Farinelli, L.A., Momenan, R., Ramchandani, V.A., Leggio, L., 2018. Exogenous
ghrelin administration increases alcohol self-administration and modulates brain
functional activity in heavy-drinking alcohol-dependent individuals. Mol. Psychiatry
23, 2029–2038.
Haass-Koffler, C.L., Akhlaghi, F., Swift, R.M., Leggio, L., 2017. Altering ethanol phar-
macokinetics to treat alcohol use disorder: can you teach an old dog new tricks? J.
Psychopharmacol. 31, 812–818.
Haass-Koffler, C.L., Giovenco, D.E., Lee, M.R., Zywiak, W.H., de la Monte, S.M., Kenna,
G.A., Swift, R.M., Leggio, L., 2016. Serum insulin levels are reduced by intravenous
ghrelin administration but do not correlate with alcohol craving in alcohol-dependent
individuals. Int. J. Neuropsychopharmacol. 19, pyw048.
Howard, A.D., Feighner, S.D., Cully, D.F., Arena, J.P., Liberator, P.A., Rosenblum, C.I.,
Hamelin, M., Hreniuk, D.L., Palyha, O.C., Anderson, J., Paress, P.S., Diaz, C., Chou,
M., Liu, K.K., McKee, K.K., Pong, S.S., Chaung, L.Y., Elbrecht, A., Dashkevicz, M.,
Heavens, R., Rigby, M., Sirinathsinghji, D.J., Dean, D.C., Melillo, D.G., Patchett, A.A.,
Nargund, R., Griffin, P.R., DeMartino, J.A., Gupta, S.K., Schaeffer, J.M., Smith, R.G.,
Van der Ploeg, L.H., 1996. A receptor in pituitary and hypothalamus that functions in
M.R. Lee, et al.
growth hormone release. Science 273, 974–977.
Jerlhag, E., 2019. Gut-brain axis and addictive disorders: a review with focus on alcohol
and drugs of abuse. Pharmacol. Ther. 196, 1–14.
Keizer, R.J., Jansen, R.S., Rosing, H., Thijssen, B., Beijnen, J.H., Schellens, J.H., Huitema,
A.D., 2015. Incorporation of concentration data below the limit of quantification in
population pharmacokinetic analyses. Pharmacol. Res. Perspect. 3, e00131.
Kenna, G.A., Swift, R.M., Hillemacher, T., Leggio, L., 2012. The relationship of appetitive,
reproductive and posterior pituitary hormones to alcoholism and craving in humans.
Neuropsychol. Rev. 22, 211–228.
Kluge, M., Riedl, S., Uhr, M., Schmidt, D., Zhang, X., Yassouridis, A., Steiger, A., 2010.
Ghrelin affects the hypothalamus-pituitary-thyroid axis in humans by increasing free
thyroxine and decreasing TSH in plasma. Eur. J. Endocrinol. 162, 1059–1065.
Kojima, M., Hosoda, H., Date, Y., Nakazato, M., Matsuo, H., Kangawa, K., 1999. Ghrelin is
a growth-hormone-releasing acylated peptide from stomach. Nature 402, 656–660.
Kong, J., Chuddy, J., Stock, I.A., Loria, P.M., Straub, S.V., Vage, C., Cameron, K.O.,
Bhattacharya, S.K., Lapham, K., McClure, K.F., Zhang, Y., Jackson, V.M., 2016.
Pharmacological characterization of the first in class clinical candidate PF-05190457:
a selective ghrelin receptor competitive antagonist with inverse agonism that in-
creases vagal afferent firing and glucose-dependent insulin secretion ex vivo. Br. J.
Pharmacol. 173, 1452–1464.
Koopmann, A., Lippmann, K., Schuster, R., Reinhard, I., Bach, P., Weil, G., Rietschel, M.,
Witt, S.H., Wiedemann, K., Kiefer, F., 2017. Drinking water to reduce alcohol
craving? A randomized controlled study on the impact of ghrelin in mediating the
effects of forced water intake in alcohol addiction. Psychoneuroendocrinology 85,
56–62.
Lee, M.R., Schwandt, M.L., Bollinger, J.W., Dias, A.A., Oot, E.N., Goldman, D.,
Hodgkinson, C.A., Leggio, L., 2015. Effect of functionally significant deiodinase single
nucleotide polymorphisms on drinking behavior in alcohol dependence: an ex-
ploratory investigation. Alcohol Clin. Exp. Res. 39, 1665–1670.
Lee, M.R., Tapocik, J.D., Ghareeb, M., Schwandt, M.L., Dias, A.A., Le, A.N., Cobbina, E.,
Farinelli, L.A., Bouhlal, S., Farokhnia, M., Heilig, M., Akhlaghi, F., Leggio, L., 2018.
The novel ghrelin receptor inverse agonist PF-5190457 administered with alcohol:
preclinical safety experiments and a phase 1b human laboratory study. Mol.
Psychiatry (in press).
Leggio, L., 2010. Role of the ghrelin system in alcoholism: acting on the growth hormone
secretagogue receptor to treat alcohol-related diseases. Drug News Perspect. 23,
157–166.
Leggio, L., Ferrulli, A., Cardone, S., Malandrino, N., Mirijello, A., D'Angelo, C., Vonghia,
L., Miceli, A., Capristo, E., Kenna, G.A., Gasbarrini, G., Swift, R.M., Addolorato, G.,
Neuropharmacology 170 (2020) 107788
2008. Relationship between the hypothalamic-pituitary-thyroid axis and alcohol
craving in alcohol-dependent patients: a longitudinal study. Alcohol Clin. Exp. Res.
32, 2047–2053.
Leggio, L., Zywiak, W.H., Fricchione, S.R., Edwards, S.M., de la Monte, S.M., Swift, R.M.,
Kenna, G.A., 2014. Intravenous ghrelin administration increases alcohol craving in
alcohol-dependent heavy drinkers: a preliminary investigation. Biol. Psychiatry 76,
734–741.
Lin, M.T., Chu, P.C., Leu, S.Y., 1983. Effects of TSH, TRH, LH and LHRH on thermo-
regulation and food and water intake in the rat. Neuroendocrinology 37, 206–211.
Morris, L.S., Voon, V., Leggio, L., 2018. Stress, motivation, and the gut-brain Axis: a focus
on the ghrelin system and alcohol use disorder. Alcohol Clin. Exp. Res. 42,
1378–1389.
Muller, T.D., Nogueiras, R., Andermann, M.L., Andrews, Z.B., Anker, S.D., Argente, J.,
Batterham, R.L., Benoit, S.C., Bowers, C.Y., Broglio, F., Casanueva, F.F., D'Alessio, D.,
Depoortere, I., Geliebter, A., Ghigo, E., Cole, P.A., Cowley, M., Cummings, D.E.,
Dagher, A., Diano, S., Dickson, S.L., Dieguez, C., Granata, R., Grill, H.J., Grove, K.,
Habegger, K.M., Heppner, K., Heiman, M.L., Holsen, L., Holst, B., Inui, A., Jansson,
J.O., Kirchner, H., Korbonits, M., Laferrere, B., LeRoux, C.W., Lopez, M., Morin, S.,
Nakazato, M., Nass, R., Perez-Tilve, D., Pfluger, P.T., Schwartz, T.W., Seeley, R.J.,
Sleeman, M., Sun, Y., Sussel, L., Tong, J., Thorner, M.O., van der Lely, A.J., van der
Ploeg, L.H., Zigman, J.M., Kojima, M., Kangawa, K., Smith, R.G., Horvath, T., Tschop,
M.H., 2015. Ghrelin. Mol. Metab. 4, 437–460.
Schwartz, T.W., 1983. Pancreatic polypeptide: a hormone under vagal control.
Gastroenterology 85, 1411–1425.
Tamboli, R.A., Antoun, J., Sidani, R.M., Clements, A., Harmata, E.E., Marks-Shulman, P.,
Gaylinn, B.D., Williams, B., Clements, R.H., Albaugh, V.L., Abumrad, N.N., 2017.
Metabolic responses to exogenous ghrelin in obesity and early after Roux-en-Y gastric
bypass in humans. Diabetes Obes. Metab. 19, 1267–1275.
Tong, J., Prigeon, R.L., Davis, H.W., Bidlingmaier, M., Kahn, S.E., Cummings, D.E.,
Tschop, M.H., D'Alessio, D., 2010. Ghrelin suppresses glucose-stimulated insulin se-
cretion and deteriorates glucose tolerance in healthy humans. Diabetes 59,
2145–2151.
Watson, P.E., 1989. Total body water and blood alcohol levels: updating the funda-
mentals. In: Crow, K., Barr, R. (Eds.), Human Metabolism of Alcohol.
Pharmcokinetics, Medicolegal Aspects, and General Interest, vol. 1. CRC Press., Boca
Raton, FL, pp. 41–58.
Zallar, L.J., Farokhnia, M., Tunstall, B.J., Vendruscolo, L.F., Leggio, L., 2017. The role of
the ghrelin system in drug addiction. Int. Rev. Neurobiol. 136, 89–119.