First-in-class fatty acid synthase inhibitor TVB-2640 reduces hepatic de novo lipogenesis in males with metabolic abnormalities
Majid M. Syed-Abdul1, Elizabeth J. Parks1,2, Ayman H. Gaballah3, Kimberlee Bingham1, Ghassan M. Hammoud2, George Kemble4, Douglas Buckley4, William McCulloch4, Camila M. Manrique5

1Department of Nutrition and Exercise Physiology, 2Department of Medicine-Division of Gastroenterology and Hepatology, 3Department of Radiology, and , and University of Missouri School of Medicine, Columbia, MO; 4Sagimet Biosciences (formerly 3-V Biosciences), Inc., Menlo Park, CA 5Department of Medicine-Division of Endocrinology, University of Missouri School of Medicine, Columbia, MO

Keywords: fatty acids, NAFLD, NASH, liver disease, drug

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/HEP.31000

Corresponding authors
Camila Manrique Acevedo, MD, Associate Professor, Division of Endocrinology, University of Missouri, Columbia, MO, 65212. Email: [email protected], Phone: (573) 882-2273

Elizabeth J. Parks, PhD, Professor, School of Medicine, Rm NW406, University of Missouri, Columbia, MO 65212. Email: [email protected], Phone: (573) 882-5864

List of Abbreviations: ACC, acetyl-CoA carboxylase; ALT, alanine aminotransferase; ANOVA, analysis of variance; AST, aspartate aminotransferase; AUC, area under the curve; BL, baseline; BP, blood pressure; CAP, controlled attenuation parameter; DBP, diastolic blood pressure; DGAT, diacylglycerol O-acyltransferase; DNL, de novo lipogenesis; F/GTT, fructose/glucose tolerance test; FAS, fatty acid synthase; FASi, fatty acid synthase inhibitor; GPAT, glycerol-3-phosphate acyltransferase; HDL, high-density lipoprotein cholesterol; IHTG, intrahepatic triacylglycerols; LDL,
low-density lipoprotein cholesterol; MRI, magnetic resonance imaging; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; NEFA, nonesterified fatty acids; PCSK9, proprotein convertase subtilisin/kexin type 9; PUFA, polyunsaturated fatty acids; SBP, systolic blood pressure; SREBP1c, sterol regulatory element-binding protein 1c; TG, triacylglycerols; VCTE, vibration controlled transient elastography; VLDL, very low-density lipoproteins

Financial support
This study was supported by 3V-Biosciences Inc., CA

Conflict of interest
EJP was a consultant for 3-V Biosciences Inc., Menlo Park, CA.
GK, DB, and WM were employees of 3-V Biosciences Inc., Menlo Park, CA.

Acknowledgements
The authors would like to express their appreciation to Nhan Le for his technical contributions and to the nursing staff at the University of Missouri Clinical Research Center for expert care of the subjects.

ABSTRACT
Elevated hepatic de novo lipogenesis (DNL) is a key distinguishing characteristic of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH). In rodent models of NAFLD, treatment with a surrogate of TVB-2640, a pharmacological inhibitor of FAS (FASi), has been shown to reduce hepatic fat and other biomarkers of DNL. The purpose of this Phase I clinical study was to test the effect of the TVB-2640 in obese men with certain metabolic abnormalities that put them at risk for NAFLD. Twelve subjects (mean±SE, 42±2y, BMI 37.4±1.2 kg/m2, glucose 103±2 mg/dL, TG 196±27 mg/dL, and elevated liver enzymes) underwent 10 days of treatment with TVB-2640 at doses ranging from 50-150 mg/d. Food intake was controlled throughout the study. Hepatic DNL was measured before and after an oral fructose/glucose (F/G) bolus using isotopic labeling with 1-13C1-acetate IV infusion, followed by measurement of labeled VLDL-palmitate via GC/MS. Substrate oxidation was measured by indirect calorimetry. Across the range of doses, fasting DNL was reduced by up to 90% (P=0.003). Increasing plasma concentrations of TVB-2640 were associated with progressive reductions in the percent of fructose-stimulated peak fractional DNL (R2= – 0.749, P=0.0003) and absolute DNL AUC 6h post F/G bolus (R2 = – 0.409, P=0.025). For all subjects combined, ALT was reduced by 15.8±8.4% (P=0.05). Substrate oxidation was unchanged and safety monitoring revealed that the drug was well tolerated, without an increase in plasma triglycerides. Alopecia occurred in two subjects (reversed after stopping the drug), but otherwise no changes were observed in fasting glucose, insulin, ketones, and renal function. These data support the therapeutic potential of FASi, TVB-2640 in particular, in patients with NAFLD and NASH.

1INTRODUCTION
2Nonalcoholic fatty liver disease (NAFLD) includes a spectrum of symptoms ranging from
3accumulation of lipids into the liver (steatosis) to inflammation (steatohepatitis, NASH), fibrosis,
4and cirrhosis (1). The epidemic of NASH and NAFLD continues to grow worldwide. As NASH
5is being recognized as a major cause of cirrhosis in the U.S. (2), NAFLD is predicted to become
6a leading cause (3) and replace viral hepatitis as the primary cause of end-stage liver disease
7(4). Several studies have identified different metabolic abnormalities associated with NAFLD (1,
85), and among those, increased hepatic de novo lipogenesis (DNL) has been recognized as a
9key characteristic of NAFLD patients (6, 7). Hepatic DNL is the process by which carbohydrates
10(glucose, fructose) are made into fat in the liver. We have previously shown that in obese
11subjects with NAFLD, ~26% of triacylglycerols accumulating in the liver are the product of DNL
12(7). To aid in the development of therapies targeted for the prevention and treatment of NAFLD
13some investigations have suggested a central role of transcription factors like the sterol
14regulatory element-binding (SREBP1c), the liver X receptor, and the carbohydrate receptor
15element-binding proteins (ChREBP) in the pathogenesis of NALFD (8-11), while others have
16pointed toward alterations in lipogenic enzymes as precipitating events (8, 11, 12). 17
18In this regard, studies utilizing rodent models with deletions of lipogenic enzymes such as
19acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS), elongases, stearoyl-CoA desaturase-
201, glycerol-3-phosphate acyltransferase (GPAT), and diacylglycerol O-acyltransferase (DGAT)
21have demonstrated successful reversal of metabolic abnormalities associated with NAFLD (13-
2218). Based on these preclinical findings, a number of these key enzymes have been targeted
23as promising areas of drug development. Early-stage clinical trials using ACC inhibitors have
24already been completed (19-21), while the clinical trials involving DGAT inhibitors are currently
25ongoing (PF-06865571; NCT03513588). Stiede et al conducted a randomized, controlled trial
26showing that increasing doses of ACC inhibitor (NDI-010976, GS-0976) progressively reduced
27fractional DNL (21). Further, in two different studies, Lawitz et al (20) and Loomba et al (19)
28tested the same ACC inhibitor (GS-0976, 5mg and 20mg respectively) for 12 weeks in patients
29with NASH. Lawitz reported significant reductions in fasting DNL and Loomba observed
30reductions in liver fat. However, one of the major concerns with ACC inhibitor treatment was the
31finding of increased concentrations of plasma triacylglycerols (TG) (19). FAS is another key
32enzyme in the DNL pathway and even though pre-clinical studies in knockout mouse models of
33FAS demonstrated reduced DNL and increased malonyl-CoA (15, 22), no previous clinical trials
34have been completed in humans examining the impact of FAS inhibition on DNL.

35
36In the present investigation, TVB-2640, a highly-potent (IC50-0.05 μM), selective and reversible
37FAS inhibitor (FASi) was administered orally (23, 24). In a past study using a high-fat, high-
38fructose fed murine model, treatment with the drug’s analog prevented hepatic steatosis,
39inflammation, and fibrosis (25, 26). Further, in a cohort of cancer patients, the FASi reduced a
40systemic marker of DNL, serum tripalmitin (27), and also reduced skin DNL, as assessed by
41sapienic acid (16:1n10), a major component of human sebum (28). However, this drug had not
42been tested for its direct effect on the pathway of DNL. Therefore, the purpose of this present
43investigation was to identify a safe dose of TVB-2640 that reduced hepatic DNL in obese men
44with metabolic abnormalities.

45SUBJECTS AND METHODS
46The study drug, TVB-2640, was approved for use as an investigational drug by the Food and Drug
47Administration (IND 132646). All methods and procedures were approved by University of Missouri
48Institutional Review Board (MU-IRB# 2006432), and the study registered at ClinicalTrials.gov
49(NCT02948569). As shown in figure 1, 95 subjects who responded to an advertisement were
50contacted by phone to determine preliminary eligibility and seventeen subjects participated in the
51informed consent process and were screened in person. For screening purposes, fasted (10h) blood
52draws and anthropometrics were obtained. In addition, questionnaires on alcohol intake and physical
53activity were administered, as was urine drug screen (AlereTM i Cup® Dx 14, Catalog # I-DX-1147-022,
54Alere Technology Services, Portsmouth, VA). Two subjects were determined to be ineligible after the
55screening visit, two subjects did not respond further following the screening visit, and one subject
56dropped out immediately after the baseline (BL) visit before starting the drug treatment due to
57unwillingness to perform the magnetic resonance imaging (MRI). As planned, twelve subjects began
58and completed the study. 59
60Inclusion and Exclusion Criteria
61The inclusion criteria included male sex, age 35-60y, no use of tobacco products, sedentary to
62moderately active, BMI between the range of 27.1 to 45.0kg/m2, elevated alanine aminotransferase
63(ALT, 42-126 U/L) but below three-times the upper limit of the normal range, family history of
64cardiovascular disease or type 2 diabetes, willingness to consume the provided isocaloric diet during
65the study, and maintenance of physical activity. Because a significant concern exists for the exposure
66of an embryo/fetus to an investigational drug, women of childbearing potential were excluded from this
67study. Subjects were recruited to have at least one characteristic of the metabolic syndrome (29) or
68insulin resistance, as evidenced by fasting insulin ≥10mU/L and/or HbA1c within ≥4.0% to ≤5.6% (30).
69Subjects were excluded if they had a history of active psychiatric disease, clinically-significant
70abnormalities on laboratory results, creatinine clearance ≤80mL/min, possessed contraindications to
71MRI, use of medications for chronic diseases, consumed alcohol>56g/week, had a major surgery within
72the past year, or donated 500ml of blood 8-wks prior to starting the study. 73
74Study Design
75Subjects were asked to complete three-day food records to characterize their diet which was analyzed
76via NDSRTM software (University of Minnesota, Minneapolis, Minnesota). These data were used to
77assess food preferences and prepare a 13-day diet in order to maintain the subject’s body weight
78during the study (supplementary table S1). As shown in figure 2, the diet was provided 3 days prior

79to a baseline (BL) inpatient test and was continued through the 10-days of drug-treatment. Dietary
80intake (mean ± SE) was 3213±168kcal/d, 36±1% fat, 17±0% protein, 48±1% CHO (20±1% of total
81energy from sugars). Decisions regarding the dose escalation scheme were made by the research
82safety monitoring team. Subjects received either 50mg, 100mg, or 150mg of TVB-2640 to be taken
83orally once per day at bed time. The first six subjects received a 50mg/d dose and as planned a priori,
84the determinations of the next dosages were made based on the results of the first six subjects. Due to
85no observed adverse effects in the first six subjects and the effect on DNL, the next two subjects
86received 150mg/d of the TVB-2640. Subjects who received 150mg exhibited mild side effects, so the
87last four subjects received 100mg/d of the TVB-2640. 88
89Study Visits
90As shown in figure 2a, at BL and after 10d of treatment (post-treatment), subjects participated in a 24h
91inpatient study. As shown in figure 2b, subjects were admitted to the University of Missouri Clinical
92Research Center at 4:00 PM. Vital signs were recorded and a FibroScan® 530 Compact (EchosenTM
93North America, Waltham, MA) was performed to measure liver steatosis assessed using Controlled
94Attenuation Parameter (CAPTM). Liver stiffness was assessed by Fibroscan Vibration-Controlled
95Transient Elastography (VCTE) at baseline. Values after treatment were considered inaccurate due to
96the short 10-day duration of the study. Specifically, a reduction in liver fat can lead to a measured
97reduction in liver stiffness (independent of a change in fibrosis) due to shrinkage of the liver. MRI scan
98was performed by a radiologist, as previously published (31), to measure intrahepatic triacylglycerols
99(IHTG) using proton density fat fraction on a Siemens 3T Skyra (series #45839) analyzed by LiverLab
100software (Siemens Healthcare GmbH, Erlangen, Germany). Intravenous lines were placed in each
101antecubital vein – one for infusion of 1-13C1-sodium acetate and the other for blood drawing. Changes
102in skin surface lipid production (sebum) were assessed using Sebutape patch (S100, S121, CuDerm
103Corporation, Dallas, TX). A standardized dinner (pasta with grilled chicken, tomato sauce, and
104parmesan cheese) was provided at 7:00 PM on day 1 which contained 39% of the subject’s total daily
105energy needs, 25% as fat, 25% protein, and 50% CHO (sugars represented 12% of total daily energy
106needs). A low-fat, high-carbohydrate dinner was provided the night before to reduce intestinal lipid
107storage overnight, ensuring that TG-rich lipoproteins secreted the next morning would be predominantly
108from the liver (VLDL) and not from the intestine (chylomicron) (32). Subjects slept/rested through the
109night and the next day at 8:00 AM, an oral fructose/glucose tolerance test (F/GTT) was administered.
110This consisted of a mixture of 0.9g/kg body weight (BW) of fructose and 0.3g/kg BW dextrose (75:25
111ratio wt:wt) dissolved in 180 ml water. Hawkins et al describe how catalytic amounts of fructose may
112activate glucose phosphorylation (33) and we have shown previously the addition of fructose to glucose

113increases DNL significantly above glucose alone (34). One gram of Kool-Aid powder (non-caloric
114lemonade flavor©, 2018 Kraft Foods, Inc., Chicago, IL) was added to the solution to increase
115palatability. The prepared drink was served with 100g of ice and subjects asked to consume it within
1165min. Blood draws were taken at 10 min prior to the drink and afterward at 15min, 30min, 45min, 1h 15
117min, 2h 45 min, 3h 20 min, 3h 40 min, 4h 10 min, and 5h. Measurements of energy expenditure and
118substrate oxidation were performed at 6:00 AM (fasted) and 10:00 AM (post-sugars) using a Parvo
119Medics metabolic cart (MMS-2400, Parvo Medics, Murray, Utah, USA). At 2:00 PM body composition
120was measured using dual x-ray emission absorptiometry (Hologic A version 13.5.2, Marlborough, MA).
121A standardized lunch was then provided and the subject was discharged from the unit. 122
123After the BL visit, the subject was given 9 doses of the drug to be taken at 10:00 PM each night. For
124the post-treatment visit after 10 days of drug administration, the subject took their last dose of the drug
125at 10:00 PM on day 10 of the follow-up inpatient study. Subsequently, blood draws were taken
126overnight until the next day to measure the drug’s pharmacokinetics. TVB-2640 was precipitated from
127human plasma with acetonitrile and measured using high pressure liquid chromatography and mass
128spectrometry (HPLC/MS/MS). An aliquot of the extract was injected onto an HHPLC/MS/MS triple
129quadrupole mass spectrometer. The peak area of the product ion of the TVB-2640 was measured
130against the peak area of the product ion of the internal standard. A calibration curve spanning the
131curve range and containing at least six concentrations in duplicate were used to quantify the analyte
132concentration. The retention times for the TVB-2640 and the internal standard were approximately 1.6
133minutes. This quantitative method has a lower limit of quantitation of 2.0ng/mL. Plasma samples were
134collected over 16h and a 24h AUC concentration/time was extrapolated. In addition to the inpatient
135studies, four safety visits were performed at 2 days, 4 days, and 7 days during treatment (figure 2b).
136During these visits, subjects were queried for potential adverse effects of the drug using a
137questionnaire. Blood was drawn for the measurement of liver enzymes (aspartate aminotransferase,
138AST and ALT) and glucose concentrations. During treatment at day 7, the skin sebum collection was
139also performed. Within one week after completion of drug treatment, the final safety visit (post 6d) was
140conducted. Two subjects who reported transient alopecia during treatment, were also seen for
141monitoring at 2, 4, and 6-weeks after the end of treatment. 142
143Primary and secondary outcomes
144The putative effects of TVB-2640 is presented in figure 2c. The primary outcomes of this study were
145the assessment of drug safety and changes in DNL (hypothesized to be significantly reduced). A
146secondary outcome was a change in liver fat mediated by the metabolic pathways shown in figure 2c.

147Other secondary outcomes were sebum production and changes in blood pressure (BP). Peripheral
148BP was measured using a Philips, SureSigns S53 Sphygmomanometer (Andover, MA) and data
149represent the average of three measurements taken one minute apart. Other measurements included
150concentrations of HbA1c, plasma lipids, glucose, insulin, ketones, lactate, apolipoprotein B100
151(apoB100) in very low-density lipoproteins (VLDL), and nonesterified fatty acids (NEFA). 152
153Analytical methods
1541-13C1-Sodium acetate was purchased from Cambridge Isotope Laboratory, Inc. (99%,
155microbiological/pyrogen tested, Tewksbury, MA). The isotope infusion was prepared in the University
156of Missouri Hospital’s Investigational Pharmacy using Good Clinical Practice requirements under a
157laminar flow bench and was made up of 10g of 1-13C1-sodium acetate was dissolved in 1L of 0.5%
158saline as described previously (34, 35). During the inpatient visit, the 1-13C1-sodium acetate infusion
159was started at 6:00 PM at a continuous rate of 0.8mL/min (48mL/hr) for 14h (figure 2b). From plasma,
160VLDL-TG particles were immediately isolated by ultracentrifugation at 40,000 rpm in a 50.3Ti rotor
161(Beckman Instruments, Palo Alto, CA) for 20h at 150C (36), TG isolated and fatty acid methyl esters
162prepared as described previously (36). Labeled palmitate was measured using a 6890N gas
163chromatography coupled to 5975 mass spectrophotometry detector (Agilent Technologies, Palo Alto,
164CA) and DNL calculated by mass isotopomer distribution analysis (37). Fatty acid composition was
165measured by 6890N gas chromatography system (Agilent Technologies, Palo Alto, CA). Fractional
166DNL was calculated using the MIDA method (6). Absolute VLDL DNL-palmitate (16:0) was calculated
167by multiplying the fractional DNL by the quantity of VLDL-TG that was palmitate, as measured via
168GCMS. Here, palmitate is used as a surrogate for new fatty acids in TG (6, 7, 34, 35, 38), as a
169significant proportion of saturated (39) and monounsaturated fatty acids are produced through this
170pathway (40), the percentage of palmitate increases 2-3 fold with sugar consumption, and the
171percentage of palmitate is directly correlated with DNL (41). Sebum production was assessed via
172Sebutape analysis (S100, S121, CuDerm Corporation, Dallas, TX) according to the manufacturer’s
173directions. Accordingly, skin oils were collected by placing the Sebutape on the forehead for 30 min,
174lipids were extracted and analyzed by Metabolon, Inc. (Morrisville, NC). 175
176Plasma concentrations of total cholesterol, TG, low-density lipoprotein cholesterol (LDLc), HDLc, AST,
177and ALT were measured by a CLIA-standardized laboratory (Boyce and Bynum Pathology Laboratory,
178Columbia, MO, 26D0652373). The measurements of lipids were performed via auto-analyzer (Roche
179Cobas 8000 System, CV 0.6-0.9%, Indianapolis, IN) using electrochemiluminescent immunoassay.
180Liver enzymes were measured using UV Absorbance (Roche Cobas 8000 System, CV 0.5-3.2% for

181AST and 0.5-3.1% for ALT, Indianapolis, IN). Assay kits were used to measure the concentrations of
182plasma glucose (Wako #439-90901, CV 6.6%, Mountain View, CA) and NEFA (Wako #991-34891, CV
1836.9%, Mountain View, CA). Plasma insulin was measured using an enzyme-linked immunosorbent
184assay (Human Insulin, EMD Millipore #EZHI-14K, CV 7.2%, Billerica, MA) and plasma ketones, using a
185cyclic enzymatic assay (Wako #415-73301 R1, 411-73401 R2, CV 1.34-1.92%, Mountain View, CA).
186Plasma lactate was measured using YSI 2300 Stat Plus Glucose & Lactate Analyzer (Yellow Springs,
187Ohio). VLDL-apo B100 was measured by ELISAPRO kit (Human apoB, Mabtech, Inc # 3715-1HP-2,
188CV 2%, Cincinnati, OH). The amount of VLDL-TG per particle was calculated as the molar ratio of
189VLDL-TG per VLDL-apoB100 (mol/mol). 190
191Statistical analysis and calculations
192StatView®, 5.0.1 software (v2008) was used when a paired sample t-test was performed. Regression
193analysis, one-factor analysis of variance (ANOVA), two-factor ANOVA, and Holm-Sidak post hoc
194analyses were performed using the statistical package for the social sciences (SPSS®, v24, 2016).
195Pearson correlation analysis was performed using SPSS® (v24, 2016). HOMA-IR was calculated as
196[(glucose in mg/dL*insulin in U/mL)/405].

197RESULTS
198Subject anthropometrics and the pharmacokinetics of TVB-2640
199No differences were found in subject characteristics among the dosing groups (table 1). The age
200(mean±SE) of all subjects combined was 42±2y. The subjects were overweight (121±5 kg and BMI
20137.4±1.2 kg/m2) and had elevated plasma glucose (103±2mg/dL) and (HbA1c 5.7±0.1%). Liver
202enzymes were elevated and within three-times the upper limit of the normal range. Baseline liver fat
203was 10.0±2.4% by MRI, 317±20 dB/m by Fibroscan, and liver stiffness (E) was 11.6±3.1 kPa. The
204subjects’ body weights were maintained throughout treatment, as expected (supplementary figures
205S1a-c). With regard to the steady-state (10d) pharmacokinetics of TVB-2640, figure 3a presents the
206plasma concentrations of the drug for each dose and figure 3b represents the calculated 24-h area-
207under-the-curve (AUC0-24) for each dose. The AUC0-24 was dose dependent and was different for
208100mg (P=0.071) and 150mg (P=0.014) compare to AUC0-24 for 50mg dose. The half-life (t1/2) of the
209drug, determined from a larger dataset of results from other studies of TVB-2640 in humans, has been
210found to be between 10-14 h which is in line with other studies of this drug (42). 211
212De novo lipogenesis, plasma lipids, and liver fat
213As shown in figure 4a, the fractional level of DNL in the fasting state, measured before the F/G bolus
214was given, was significantly inhibited with the TVB-2640 100 mg (P≤0.001) and 150 mg dose (P≤0.001)
215but not with 50mg dose (P=0.220). For absolute DNL (figure 4b), TVB-2640 significantly inhibited
216fasting DNL with the 150mg dose (P<0.001) and 100mg (P=0.036), and no change was found with the
21750mg dose (P=0.544). Interestingly, inhibition of DNL-16:0 occurred without changes in VLDL-16:0
218(figure 4c) or plasma TG (figure 4d) for all three groups. These data, along with the number of TG
219molecules per particle and the lack of change in VLDL-apoB100 (either fasting or the peak after the
220sugars bolus, table 2) suggest that another source of fatty acids may have been used to support
221lipoprotein assembly. With regard to the acute effect of sugars, BL fractional DNL and absolute DNL-
22216:0 were significantly elevated after the F/G bolus (figure 4a & b). Following 10d of treatment, the
223F/G bolus-induction was inhibited; DNL stimulation was 23% lower after 50mg/d (figure 5a, P=0.033),
22465% lower after 100mg/d (P=0.008), and 77% lower after 150mg/d (P<0.001). The IHTG analyzed by
225MRI using proton density fat fraction was significantly reduced on average from 10.0% to 8.4% (figure
2265b, t-test, P=0.016, ANOVA=0.048) and this reduction in IHTG primarily appeared to be driven by the
227subjects who were treated with 100mg/d (P=0.06). Liver fat, as assessed via FibroScan CAP score,
228was significantly reduced in the 100mg/d (P=0.025) and 150mg/d dose groups (P=0.004, figure 5c).
229As shown in figures 5d-f, reductions in peak absolute stimulation of DNL (figure 5d) and IHTG (figure
2305e) were inversely associated with the drug's AUC0-24h (P=0.0003 and P=0.029, respectively). Further,

231the reduction in fractional DNL was significantly associated with reduction in the liver fat percent (figure 232 5f).
233
234Blood cholesterols and ketosis
235Prior studies of ACC inhibitors have shown increased TG concentration with a decrease in PUFA
236composition of TG. As shown in figure 6a, neither plasma TG, nor VLDL-TG changed although
237concentrations were highly variable after 10-days of TVB-2640. No reductions in PUFA were found in
238the present study (supplementary figures S3a-c). All forms of cholesterols i.e., total cholesterol
239(P=0.010) and LDLc (P=0.003) were significantly reduced in subjects who took 100mg/d, whereas no
240significant changes were observed in the other groups. HDLc was significantly reduced in subjects who
241took 50mg (P=0.001) and 100mg (P=0.036) but not in the two subjects who took 150mg (P=0.139,
242figure 6a). We found no differences in ketone concentrations which is in contrast to past observations
243of ketosis with ACC inhibitor (13). As shown in table 2, no changes were observed in respiratory
244quotient (fasting or fed), glucose oxidation, or fatty acid oxidation for all three groups. Further, no
245changes were observed for plasma lactate and blood CO2 in all three groups. 246
247
248Drug safety and adverse drug reactions
249With regard to drug safety, during and after completion of the study, ALT concentrations were
250significantly reduced, whereas no changes were observed in plasma AST levels (figures 6b-c). As
251shown in supplementary table S2, systolic BP did not change (ANOVA, P=0.413) but diastolic BP
252reduced significantly (ANOVA, P=0.045). No changes were observed in fasting and fed glucose and
253insulin concentrations with all three doses (supplementary figure S2a-b). Similarly, fasting NEFA was
254not different for all three groups but the rebound effect on postprandial NEFA was significantly lower
255with the 100mg/d dose at 3.5h, 5h, and 6h (supplementary figure S2c). When AUC values were
256calculated for these biochemical measurements in response to F/G bolus, NEFA AUC values were
257significantly lower after treatment with 150mg/d dose. Few adverse drug reactions were reported with
258each dose of drug (supplementary table S3). One subject in the 50mg/d group reported dry skin and
259one subject in the 100mg/d group reported dry mouth; both symptoms disappeared upon completion of
260drug treatment. Two subjects (one treated with 100mg and one treated with 150mg) reported mild and
261transient alopecia. Following completion of treatment, these subjects were seen for 6wks, at 2wk
262intervals and hair loss was documented to have recovered. Supplementary figures S4a-b present the
263sebum analysis for these two subjects. TG48:0 and TG48:1 make up approximately 50% of all TG in
264sebum. The 16:1 fatty acid in TG48:1 is most likely sapenic acid (16:1, n-10) which is unique to sebum

265(43, 44). Concentrations of TG species 48:0 and TG 48:1 fell at 6d and 10d of treatment and began to
266recover 2 weeks after completion of treatment. Total TG in sebum and the top 50 individual TG for
267these two subjects are presented in the supplementary figures S4c-e. Lastly, in addition to the
268symptoms that are listed in the supplementary table S3, subjects were also asked to report any other
269reactions they may have experienced during the treatment. None of the subjects reported other
270complications, including ophthalmological complications.

271DISCUSSION
272The primary finding of the current investigation is that 10d of treatment with TVB-2640, an inhibitor of
273the β-ketoacyl reductase domain of the fatty acid synthase enzyme (FAS) complex, reduced hepatic de
274novo lipogenesis (DNL) and decreased intrahepatic lipid content (IHTG) in male subjects with
275characteristics of metabolic syndrome. Specifically, increasing drug doses suppressed DNL in a
276stepwise fashion such that at 150mg/d, DNL appeared to be fully suppressed when stimulated by
277consumption of a bolus of liquid sugars. The maximum absolute reduction in IHTG at 10 days was
2784.2% and this effect was found predominantly in those subjects who received 100mg/d dose. However,
279it is uncertain whether the 50mg/d dose was less effective because of the amount of drug or because
280the subjects who received this dose started with lower levels of liver fat at the beginning of the study,
281compare to 100mg/d group. Further, in 100mg/d dose group, concentrations of total cholesterol, LDLc,
282and HDLc were reduced. Among the three sources of fatty acids that can contribute to IHTG (adipose,
283diet, and DNL), increased DNL has been shown as a key contributor to the pathogenesis of NAFLD in
284subjects with insulin resistance (6, 7). In this study, treatment decreased IHTG but VLDL-TG and apo-
285B100 did not change. Thus, the primary mechanisms for the decrease in IHTG found here were likely a
286direct effect of inhibiting the lipogenesis pathway (28). With these changes, ALT and AST were
287significantly reduced. These data highlight the potential role of the FAS enzyme to contribute to liver
288lipid accrual. Treatment did not change fasting NEFA or ketone body concentrations nor whole body
289fatty acid oxidation. Thus, continued flux of NEFA to the liver also may have supported VLDL-TG
290production.
291
292The lack of change in both VLDL- and plasma-TG in the present human study are consistent with a
293preclinical study in mice in which TG levels did not change with complete knockout of the FAS enzyme
294(45). A recent study in a diet-induced, obese murine model of NASH showed a decrease of plasma-TG
295with a drug similar to TVB-2640 (25). By contrast, targeting hepatic lipid synthesis through acetyl-CoA
296carboxylase inhibition (ACCi) in humans and rodents consistently raises VLDL- and plasma-TG
297concentrations (13, 19, 20, 46). For example, in an open-label prospective study, Lawitz et al
298administered an ACC inhibitor (ACCi GS 0976) to patients with NASH and reported lower DNL and
299IHTG contents accompanied by non-significant increases in plasma-TG (20). In a study of healthy
300subjects with hepatic steatosis, Kim et al found that inhibition of ACC with the compound MK-4074
301reduced liver fat but increased fasting VLDL-TG concentrations (13). Goedeke et al treated Sprague-
302Dawley rats with an ACCi drug termed 'compound 1' and found significant reductions in liver-TG
303content, while plasma-TG more than doubled (46). These changes were attributed a 15% increase in
304VLDL production and a 20% reduction LPL-mediated TG clearance. Lastly, Kim et al investigated

305hepatic lipid synthesis in an ACC knockout mouse and concluded that the observed increase in VLDL-
306TG concentration resulted from a reduction in liver PUFA content, a consequence of lower malonyl-CoA
307levels, and increased expression of SREBP1c and GPAT leading to increased VLDL production (13). A
308role for PUFA signaling in these mechanisms is unclear. In the present study, FASi did not elevate
309plasma-TG nor reduce PUFA levels in fasting VLDL-TG. 310
311With regard to safety, as described above, although plasma- and VLDL-TG responses vary depending
312on whether ACC or FAS is targeted, inhibition of lipid synthesis causes reductions in liver-TG content
313that are consistently associated with reduced plasma ALT concentrations. We were surprised to
314observe this effect as early as four days and ALT remained lower even after 6 days post-treatment with
315TVB-2640. Reductions in ALT are consistent with previous studies conducted in both animal models of
316FASi (15) and in human subjects who were treated with different ACCi (19, 20). The fact that ALT
317concentrations did not increase above baseline supports the observation that there was no toxic impact
318on the liver; indeed a significant reduction in ALT suggests liver function improved. Further indications
319of safety included no change in systolic or diastolic BP, consistent with past literature (47). The
320transient alopecia observed in two subjects was likely due to a decline in fatty acid production in
321sebocytes (44). As shown in supplementary figures S2a & b, this decline in sebum TG began to
322resolve 2wks after cessation of drug and evidence of hair regrowth was observed. With regard to
323LDLc, data from preclinical and primary cell culture demonstrate that FASi decreases proprotein
324convertase subtilisin/kexin type 9 (PCSK9) and increases expression of the insulin-induced gene 1
325(Insig-1), both effects that would lower LDLc (15, 48, 49). 326
327Limitations of this investigation included a relatively small size, short duration and inclusion of only men,
328common characteristics of Phase 1 studies. Future studies should include subjects with documented
329NASH treated with TVB-2640 for longer periods of time. The choice of 1-13C1-sodium acetate to
330measure lipogenesis was made due to the isotope's fast decay rate which accommodates studies of
331short duration. To be utilized in lipogenesis, acetate requires activation by acetyl coenzyme-A synthase
332(ACS) and, although highly unlikely, it is possible that the observed reduction in DNL could have been
333due to an off-target effect of the drug on ACS. To test this possibility, the acetyl-CoA precursor
334enrichment (p) was used as an indicator of the efficiency of ACS to activate the isotope. Drug
335treatment did not change the enrichment of the acetylCoA precursor pool during the F/GTT (data not
336shown) and thus, the drug-induced reductions in DNL likely reflect a direct inhibition of FAS. Further
337support for this was found in the significant correlation between drug concentration and the reduction in
338liver fat (figures 5e-f).

339
340In summary, ten days of treatment with TVB-2640 significantly reduced DNL in subjects with
341characteristics of metabolic syndrome without raising concentrations of blood lipids. These data are
342consistent with previous studies conducted in animal models with inhibition of FAS (15, 22, 25, 28, 45).
343Reductions in DNL acutely decreased liver fat, biochemical markers of liver injury, and cholesterol
344levels. These effects were observed primarily in subjects who were administered the 100mg/d of dose.
345Future studies are needed to determine the appropriate dose and effects of longer-term use.

FIGURE LEGENDS
Figure 1. CONSORT 2010 flow diagram

Figure 2. Overall study design and in-patient protocol. 2a. Subjects were treated with TVB- 2640 for 10d during which time they consumed a standardized diet and were monitored for side effects (as described in the methods section).
2b. Inpatient metabolic studies were performed before and after 10d of treatment.
2c. Schematic representation of pathways effected by inhibition of FAS enzyme. Abbreviations: MRI, magnetic resonance imaging; DEXA, dual energy x-ray absorptiometry; ALT, alanine aminotransferase; AST, aspartate aminotransferase; DNL, de novo lipogenesis; FAS, fatty acid synthase; IHTG, intrahepatic triacylglycerols; LDL, low-density lipoprotein cholesterol; NEFA, nonesterified fatty acids; TG, triacylglycerol; VLDL, very low-density lipoproteins.

Figure 3. Pharmacokinetics and plasma concentration AUC after 10d of dosing. Data are mean ± SE. 3a. Dose response curve for three doses of TVB-2640. Repeated-measures ANOVA was performed to test the dose response curve for three doses over time. Dose was used as a between-group factor and ‘time post administration’ was used as a within-group factor. Concentrations of drug tended to be different between three groups (P=0.066). Post hoc testing was conducted using Sidak-Holm analysis. The dose response curve for the 50 mg group was significantly lower than for the 150 mg group (P=0.011) and tended to be lower than 100 mg group (P=0.083). No differences were observed between the 100 mg and 150 mg groups (P=0.328). *P<0.05, compared to baseline value for 100 mg group.
3b. Area-under-the-curve (AUC) for each dose response curve. One-way ANOVA was conducted to test the differences between three doses. A significant difference was found between the three doses (P=0.009). Post hoc testing was conducted using Sidak-Holm analysis. A difference was found between the 50 mg and 150 mg groups (P=0.014).
Concentrations obtained in the 100 mg group tended to be different compared to 50 mg (P=0.071) and were not different compared to 150 mg group (P=0.418).

Figure 4. Changes in DNL, VLDL- and plasma-TG during consumption of the fructose/glucose bolus. Data are mean ± SE. Repeated-measures ANOVA was conducted using two within factors and one between factor. Time post-consumption of fructose/glucose bolus was used as a first factor, 10 days of treatment was used as a second factor for within factors, and group (dose) was used as a between factor. Open circles (○, baseline), closed boxes (■, post- treatment). Abbreviations: TG, triacylglycerols; VLDL, very low density lipoproteins; DNL, de novo lipogenesis.

4a. Fractional DNL in response to the F/GTT. Between-factor analysis revealed that fractional DNL was significantly reduced with drug treatment in all subjects combined (P=0.001) and the values were lower during the follow-up visit compare to baseline values (P<0.001). This reduction in fractional DNL was also different between groups (P=0.007). Within-factor analysis revealed that before drug treatment, fractional DNL was significantly stimulated with fructose/glucose bolus for all three groups (50 mg, P=0.005; 100 mg, P=0.015; 150 mg, P=0.021). *P≤0.05 compare to fasting value. After the drug treatment, for all subjects combined, this stimulation of DNL was significantly reduced (P=0.001) and therefore, no changes were observed in response to fructose/glucose bolus in all three groups (50 mg, P=0.256; 100 mg, P=0.549; 150 mg, P=0.619).
† represents a significant difference between baseline and post-treatment at each point.
P≤0.05 for the 50 mg group, and P≤0.001 for the 100 mg and 150 mg groups.
4b. Absolute DNL-16:0 in response to F/GTT. Between factor analysis revealed that absolute DNL-16:0 was significantly lower at follow-up compare to baseline values (P<0.001). This reduction in absolute DNL-16:0 was tended to be different between groups (P=0.072).
*P≤0.05 compare to fasting value. † represents significant difference between baseline and post-treatment at each point. P≤0.05 for 50 mg and 100 mg group, and P≤0.001 for 150 mg group.
4c. VLDL-TG 16:0 concentration in response to F/GTT. For between factor analysis, VLDL- TG 16:0 did not change for all subjects (P=0.500) or between groups (P=0.685). Within factor analysis revealed that VLDL-TG 16:0 did not change in response to fructose/glucose bolus for all three groups both before (50 mg, P=0.260; 100 mg, P=0.664; 150 mg, P=0.630) and for

one group after the drug treatment (50 mg, P=0.017; 100 mg P=0.262; 150 mg, P=0.404). † represents significant difference between baseline and post-treatment at each point (P≤0.05). 4d. Plasma-TG concentration in response to the F/GTT. For between-factor analysis, plasma- TG did not change for all subjects (P=0.129) or between groups (P=0.968). Within-factor analysis revealed that plasma-TG did not change in response to fructose/glucose bolus for all three groups both before (50 mg, P=0.162; 100 mg, P=0.608; 150 mg, P=0.131) and after the drug treatment (50 mg, P=0.579; 100 mg P=0.338; 150 mg, P=0.992).

Figure 5. Changes in DNL, liver fat, and correlation analysis between TVB-2640 AUC, DNL, and liver fat.
5a. AUC of the absolute DNL-16:0 before (□) and post-treatment (■). Repeated-measures ANOVA was performed to test the effect of drug on DNL-16:0 before and after the treatment between the three groups. During the analysis, dose was used as a between-group factor and time was used as within-group factor. Within-group analyses revealed that for all subjects combined, DNL-16:0 was reduced significantly (P<0.001). The reduction in DNL-16:0 was significantly different within each group (P<0.001). Post hoc analysis revealed that the reduction in DNL-16:0 was significant in all three groups (50 mg, P=0.027, 100 mg, P=0.007 and 150 mg group, P<0.001).
5b. Liver fat measured by MRI scan (PDFF). Liver fat was measured for a selected region of interest (ROI). Repeated-measures ANOVA was performed to test the effect of drug on liver fat measured via MRI scan (IHTG) over time. During this analysis, dose was used as a between-group factor and time was used as a within-group factor. Within-group analysis revealed that for all subjects combined, IHTG was reduced significantly (P=0.048). The trend of a reduction in IHTG observed within each group was not different i.e., different doses of the drug reduced IHTG in a similar pattern (P=0.732). Post hoc analysis revealed that this effect was primarily driven by the 100 mg group (P=0.06). IHTG tended to be reduced in the 50 mg group (P=0.137) but not in the 100 mg group (P=0.545).
5c. Liver fat measured by FibroscanTM (CAP, dB/m). Repeated-measures ANOVA was performed to test the effect of drug on liver fat measured via FibroScan (CAP) over time. Dose was used as a between-group factor and time was used as a within-group factor. Within-group analysis revealed that for all subjects combined, the CAP score was reduced significantly

(P=0.002). The reduction in CAP score within each group was significantly different (P=0.020). Post hoc analysis revealed that this effect was primarily driven by 100 mg group (P=0.025) and 150 mg group (P=0.004). The CAP score did not change in the 50 mg group (P=0.968).
5d-f. Relationship between the AUC of the drug in plasma and the percent change in peak absolute DNL and change in total IHTG measured by MRI. Relationship between change in fractional DNL and IHTG. A Pearson correlation analysis was performed to test the relationship between DNL, concentration of drug, and IHTG.
Abbreviations: AUC, area under the curve; DNL, de novo lipogenesis; AUC, area-under-the- curve; IHTG, intrahepatic triacylglycerol; CAP, liver fat score measured by FibroScan (db/m); BL, baseline; and FL, follow-up (post-treatment).

Figure 6. Changes in plasma lipids and liver enzymes. For all graphs, data are mean ± SE. Abbreviations: ALT, alanine transaminase; AST, aspartate aminotransferase; PDFF, proton density fat fraction; IHTG, intrahepatic triacylglycerols; CAP, units for liver fat measured by FibroScanTM; Sc, screening value.
6a. Students t-test was performed to test the effect of drug on blood lipids. * P<0.05 represents significant difference from baseline and after completion of the treatment. Changes in lipids for each group is shown in the figure. One-way ANOVA was performed to test if the changes observed in lipids were different between the three doses. No differences were observed for total cholesterol (P=0.487), high density lipoprotein cholesterol (P=0.551), low density lipoprotein cholesterol (P=0.361), nor plasma-TG (P=0.740).
6b. Repeated-measures ANOVA was performed to test the effect of drug on ALT concentrations over time. During the analysis, dose was used as a between-group factor and time was used as within-group factor. Overall, ALT values reduced significantly (P=0.038).
However, the reduction in ALT was not different between three doses (P=0.223). Between- group analysis suggested that ALT values were significantly different between three groups (P=0.017). Pairwise comparisons (Sidak-Holm post hoc) revealed that only the 100 mg group’s ALT values were significantly higher compared to the 50 mg group (P=0.018) but were not different compare to 150 mg group (P=0.170). No differences were observed between 50 mg and 100 mg group (P=0.880).

6c. Repeated-measures ANOVA was performed to test the effect of the drug on AST concentrations over time. During the analysis, dose was used as a between-group factor and time was used as within-group factor. Within-group analysis revealed that AST values tended to decrease (P=0.075). The reduction in AST observed within each group was not different i.e., different doses of drug reduced AST in a similar pattern (P=0.617). Between-group analysis suggested that AST values tended to be different between three groups (P=0.087).
The pairwise comparisons (Sidak-Holm post hoc) revealed that only 100 mg group’s AST values were higher compare to 50 mg group (P=0.095) but were similar to 150 mg group (P=0.501). No differences were observed between the 50 mg and 100 mg groups (P=0.901).

REFERENCES
1.Browning JD, Horton JD. Molecular mediators of hepatic steatosis and liver injury. J Clin Invest 2004;114:147-152.
2.Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global epidemiology of nonalcoholic fatty liver disease: Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology 2016;64:73-84.
3.Calzadilla Bertot L, Adams LA. The natural course of non-alcoholic fatty liver disease. Int J Mol Sci 2016;17.
4.Ray K. NAFLD-the next global epidemic. Nat Rev Gastroenterol Hepatol 2013;10:621.
5.Eslam M, Valenti L, Romeo S. Genetics and epigenetics of NAFLD and NASH: Clinical impact. J Hepatol 2018;68:268-279.
6.Lambert JE, Ramos-Roman MA, Browning JD, Parks EJ. Increased de novo lipogenesis is a distinct characteristic of individuals with nonalcoholic fatty liver disease. Gastroenterology 2014;146:726-735.
7.Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ. Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest 2005;115:1343-1351.
8.Postic C, Girard J. Contribution of de novo fatty acid synthesis to hepatic steatosis and insulin resistance: lessons from genetically engineered mice. J Clin Invest 2008;118:829-838.
9.Ahmed MH, Byrne CD. Modulation of sterol regulatory element binding proteins (SREBPs) as potential treatments for non-alcoholic fatty liver disease (NAFLD). Drug Discov Today 2007;12:740-747.
10.Byrne CD, Olufadi R, Bruce KD, Cagampang FR, Ahmed MH. Metabolic disturbances in non-alcoholic fatty liver disease. Clin Sci (Lond) 2009;116:539-564.
11.Finck BN. Targeting metabolism, insulin resistance, and diabetes to treat nonalcoholic steatohepatitis. Diabetes 2018;67:2485-2493.
12.Imai N, Cohen DE. Trimming the fat: Acetyl-CoA Carboxylase inhibition for the management of NAFLD. Hepatology 2018;68:2062-2065.
13.Kim CW, Addy C, Kusunoki J, Anderson NN, Deja S, Fu X, Burgess SC, et al. Acetyl CoA Carboxylase inhibition reduces hepatic steatosis but elevates plasma triglycerides

in mice and humans: A bedside to bench investigation. Cell Metab 2017;26:394-406 e396.
14.Savage DB, Choi CS, Samuel VT, Liu ZX, Zhang D, Wang A, Zhang XM, et al. Reversal of diet-induced hepatic steatosis and hepatic insulin resistance by antisense oligonucleotide inhibitors of acetyl-CoA carboxylases 1 and 2. J Clin Invest 2006;116:817-824.
15.Chakravarthy MV, Pan Z, Zhu Y, Tordjman K, Schneider JG, Coleman T, Turk J, et al. "New" hepatic fat activates PPARalpha to maintain glucose, lipid, and cholesterol homeostasis. Cell Metab 2005;1:309-322.
16.Ntambi JM, Miyazaki M, Stoehr JP, Lan H, Kendziorski CM, Yandell BS, Song Y, et al. Loss of stearoyl-CoA desaturase-1 function protects mice against adiposity. Proc Natl Acad Sci U S A 2002;99:11482-11486.
17.Neschen S, Morino K, Hammond LE, Zhang D, Liu ZX, Romanelli AJ, Cline GW, et al. Prevention of hepatic steatosis and hepatic insulin resistance in mitochondrial acyl- CoA:glycerol-sn-3-phosphate acyltransferase 1 knockout mice. Cell Metab 2005;2:55- 65.
18.Yamaguchi K, Yang L, McCall S, Huang J, Yu XX, Pandey SK, Bhanot S, et al. Inhibiting triglyceride synthesis improves hepatic steatosis but exacerbates liver damage and fibrosis in obese mice with nonalcoholic steatohepatitis. Hepatology 2007;45:1366-1374.
19.Loomba R, Kayali Z, Noureddin M, Ruane P, Lawitz EJ, Bennett M, Wang L, et al. GS- 0976 reduces hepatic steatosis and fibrosis markers in patients with nonalcoholic fatty liver disease. Gastroenterology 2018;155:1463-1473 e1466.
20.Lawitz EJ, Coste A, Poordad F, Alkhouri N, Loo N, McColgan BJ, Tarrant JM, et al. Acetyl-CoA Carboxylase inhibitor GS-0976 for 12 weeks reduces hepatic de novo lipogenesis and steatosis in patients with nonalcoholic steatohepatitis. Clin Gastroenterol Hepatol 2018;16:1983-1991 e1983.
21.Stiede K, Miao W, Blanchette HS, Beysen C, Harriman G, Harwood HJ, Jr., Kelley H, et al. Acetyl-coenzyme A carboxylase inhibition reduces de novo lipogenesis in overweight male subjects: A randomized, double-blind, crossover study. Hepatology 2017;66:324- 334.

22.Wu M, Singh SB, Wang J, Chung CC, Salituro G, Karanam BV, Lee SH, et al. Antidiabetic and antisteatotic effects of the selective fatty acid synthase (FAS) inhibitor platensimycin in mouse models of diabetes. Proc Natl Acad Sci U S A 2011;108:5378- 5383.
23.Jones SF, Infante JR. Molecular Pathways: Fatty Acid Synthase. Clin Cancer Res 2015;21:5434-5438.
24.Buckley D, Duke G, Heuer TS, O'Farrell M, Wagman AS, McCulloch W, Kemble G. Fatty acid synthase - Modern tumor cell biology insights into a classical oncology target. Pharmacol Ther 2017;177:23-31.
25.Duke G, Wagman A, Buckley D, McCulloch W, Feigh M, Veidal SS, Kemble G. Fatty acid synthase inhibitor TVB-3664 reduces collagen accumulation in bleomycin-induced murine skin fibrosis and reverses multiple components of diet-induced and biopsy- confirmed nonalcoholic steatohepatitis in mice treated with or without co-administered pirfenidone. Hepatology 2017;66:1056A.
26.Parks EJ, Manrique CM, Syed-Abdul MM, Gaballah AH, Hammoud GM, Buckley D, Duke G, et al. Pharmacologic inhibition of FASN reverses diet-induced steatohepatitis in mice and inhibits lipogenesis in humans. Hepatology 2017;66:1045A.
27.Brenner A, Infante J, Patel M, Arkenau H-T, Voskoboynik M, Borazanci E, Falchook G, et al. Abstract A54: First-in-human study of the first-in-class fatty acid synthase (FASN) inhibitor, TVB-2640 as monotherapy or in combination - final results of dose escalation. Molecular Cancer Therapeutics 2015;14:A54-A54.
28.Duke G, Wagman AS, Buckley D, McCulloch W, Kemble G. Establishing the foundation for a novel, first-in-class, fatty acid synthase inhibitor, TVB-2640, for the treatment of NASH. Journal of Hepatology 2017;66:S99-S100.
29.Grundy SM, Brewer HB, Jr., Cleeman JI, Smith SC, Jr., Lenfant C, American Heart A, National Heart L, et al. Definition of metabolic syndrome: Report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Circulation 2004;109:433-438.
30.Ahren B, Larsson H. Quantification of insulin secretion in relation to insulin sensitivity in nondiabetic postmenopausal women. Diabetes 2002;51 Suppl 1:S202-211.

31.Kannengiesser S, Neji R, Zhong X. Case Study LiverLab. MAGNETOM Flash 2014;3:18-19.
32.Chavez-Jauregui RN, Mattes RD, Parks EJ. Dynamics of fat absorption and effect of sham feeding on postprandial lipema. Gastroenterology 2010;139:1538-1548.
33.Hawkins M, Gabriely I, Wozniak R, Vilcu C, Shamoon H, Rossetti L. Fructose improves the ability of hyperglycemia per se to regulate glucose production in type 2 diabetes. Diabetes 2002;51:606-614.
34.Parks EJ, Skokan LE, Timlin MT, Dingfelder CS. Dietary sugars stimulate fatty acid synthesis in adults. J Nutr 2008;138:1039-1046.
35.Parks EJ, Krauss RM, Christiansen MP, Neese RA, Hellerstein MK. Effects of a low-fat, high-carbohydrate diet on VLDL-triglyceride assembly, production, and clearance. J Clin Invest 1999;104:1087-1096.
36.Barrows BR, Parks EJ. Contributions of different fatty acid sources to very low-density lipoprotein-triacylglycerol in the fasted and fed states. J Clin Endocrinol Metab 2006;91:1446-1452.
37.Hellerstein MK, Neese RA. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations. Am J Physiol 1999;276:E1146- 1170.
38.Santoro N, Caprio S, Pierpont B, Van Name M, Savoye M, Parks EJ. Hepatic de novo lipogenesis in obese youth is modulated by a common variant in the GCKR gene. J Clin Endocrinol Metab 2015;100:E1125-1132.
39.Chong MF, Fielding BA, Frayn KN. Mechanisms for the acute effect of fructose on postprandial lipemia. Am J Clin Nutr 2007;85:1511-1520.
40.Aarsland A, Chinkes D, Wolfe RR. Hepatic and whole-body fat synthesis in humans during carbohydrate overfeeding. Am J Clin Nutr 1997;65:1774-1782.
41.Hudgins LC, Parker TS, Levine DM, Hellerstein MK. A dual sugar challenge test for lipogenic sensitivity to dietary fructose. J Clin Endocrinol Metab 2011;96:861-868.
42.O'Farrell M, Crowley R, Heuer T, Fridlib M, Buckley D, McCulloch W, Kemble G. Biomarker analyses from dose escalation phase of FASN inhibitor TVB-2640 phase 1 study shows target engagement in solid tumor patients. AACR; Molecular Cancer Therapeutics 2015;14.

43.Pappas A, Johnsen S, Liu JC, Eisinger M. Sebum analysis of individuals with and without acne. Dermatoendocrinol 2009;1:157-161.
44.Esler WP, Tesz GJ, Hellerstein MK, Beysen C, Sivamani R, Turner SM, Watkins SM, et al. Human sebum requires de novo lipogenesis, which is increased in acne vulgaris and suppressed by acetyl-CoA carboxylase inhibition. Sci Transl Med 2019;11.
45.Funai K, Song H, Yin L, Lodhi IJ, Wei X, Yoshino J, Coleman T, et al. Muscle lipogenesis balances insulin sensitivity and strength through calcium signaling. J Clin Invest 2013;123:1229-1240.
46.Goedeke L, Bates J, Vatner DF, Perry RJ, Wang T, Ramirez R, Li L, et al. Acetyl-CoA Carboxylase inhibition reverses NAFLD and hepatic insulin resistance but promotes hypertriglyceridemia in rodents. Hepatology 2018;68:2197-2211.
47.Razani B, Zhang H, Schulze PC, Schilling JD, Verbsky J, Lodhi IJ, Topkara VK, et al. Fatty acid synthase modulates homeostatic responses to myocardial stress. J Biol Chem 2011;286:30949-30961.
48.Lee S, Zhang C, Liu Z, Klevstig M, Mukhopadhyay B, Bergentall M, Cinar R, et al. Network analyses identify liver-specific targets for treating liver diseases. Mol Syst Biol 2017;13:938.
49.Roth EM, Davidson MH. PCSK9 Inhibitors: Mechanism of action, efficacy, and safety. Rev Cardiovasc Med 2018;19:S31-S46.