Non-invasive, quantitative assessment of liver fat by MRI-PDFF as an endpoint

in NASH trials

Cyrielle Caussy1,2, Scott B. Reeder3, Claude B. Sirlin4 and Rohit Loomba1,5,6

1
NAFLD Research Center, Department of Medicine, La Jolla, CA
2
Université Lyon 1, Hospices Civils de Lyon, Lyon, France
3
Department of Radiology, Medical Physics, Biomedical Engineering, Medicine, and Emergency Medicine
University of Wisconsin-Madison, Madison, WI
4
Liver Imaging Group, Department of Radiology, University of California at San Diego, La Jolla, CA
5
Division of Gastroenterology, Department of Medicine, La Jolla, CA
6
Division of Epidemiology, Department of Family and Preventive Medicine, University of California at San
Diego, La Jolla, CA

Short title: MRI-PDFF in NASH Trials

Word count: abstract 160 words, main text: 5988 words, tables: 0, figures: 3

Grant support: RL is supported in part by the American Gastroenterological Association (AGA)

Foundation – Sucampo – ASP Designated Research Award in Geriatric Gastroenterology and by a T.
Franklin Williams Scholarship Award; Funding provided by: Atlantic Philanthropies, Inc, the John A.
Hartford Foundation, OM, the Association of Specialty Professors, and the American Gastroenterological
Association and grant K23-DK090303. The authors also wish to acknowledge support from the National
Institute of Health (NIH) (UL1TR00427, R01DK106419, R01 DK083380, R01 DK088925, R01 DK100651
and K24 DK102595), as well GE Healthcare who provides research support to the UW-Madison and
UCSD. The content is solely the responsibility of the authors and does not necessarily represent the
official views of the NIH.
Please address correspondence to:
Rohit Loomba, MD, MHSc
ACTRI Building, 1W202
9452 Medical Center Drive
La Jolla, CA 92037
Ph: 858-246-2201
Fax: 858-246-2255
Email: [email protected]
Web: http://fattyliver.ucsd.edu

Abbreviations
AUROC: area under the receiver operator characteristic curve; BMI: body mass index; CAP: controlled
attenuation parameter; CI: confidence interval; CSE-MRI: chemical shift encoded magnetic resonance
imaging; CT: computerized tomography; DECT: dual-energy computerized tomography; FDA: U.S Food

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.29797

This article is protected by copyright. All rights reserved.

 Hepatology Page 2 of 24

Caussy et al.

and Drug administration; HCC: hepatocellular carcinoma; MRI: magnetic resonance imaging; MRI-PDFF:
magnetic resonance imaging proton density fat fraction; MRS: magnetic resonance spectroscopy;
NAFLD: nonalcoholic fatty liver disease; NASH: nonalcoholic steatohepatitis; TE: transient elastography;
VCTE: vibration-controlled transient elastography; ROI: region of interest

Role of study sponsor: The study sponsor(s) had no role in the study design, collection, analysis,
interpretation of the data, and/or drafting of the manuscript. All authors report that no conflicts of
interest exist.

Conflict of interests: Dr. Sirlin consults, advises, and is on the speakers’ bureau for Bayer. Drs. Loomba
and Sirlin received grants from GE Healthcare and Siemens Inc. Dr. Reeder reports that the University of
Wisconsin receives research support from GE Healthcare and Bracco Diagnostics. Dr. Reeder consults for
Parexel International and is a founder of Calimetrix, LLC. Dr. Caussy reports no other conflict of interests.

Author contributions:
Cyrielle Caussy: Drafting of the manuscript, approved final submission.
Scott B. Reeder: study concept and design, analysis and interpretation of data, drafting of the
manuscript, critical revision of the manuscript, obtained funding, study supervision, approved final
submission
Claude B. Sirlin: study concept and design, analysis and interpretation of data, drafting of the
manuscript, critical revision of the manuscript, obtained funding, study supervision, approved final
submission
Rohit Loomba: study concept and design, analysis and interpretation of data, drafting of the manuscript,
critical revision of the manuscript, obtained funding, study supervision, approved final submission
All authors approved the final version of this article.

2
Hepatology
This article is protected by copyright. All rights reserved.
Page 3 of 24 Hepatology

Caussy et al.

ABSTRACT

Nonalcoholic fatty liver disease (NAFLD) is now the most common cause of chronic liver disease

worldwide, and the progressive form of this condition, nonalcoholic steatohepatitis (NASH), has become

one of the leading indications for liver transplant. Despite intensive investigations, there are currently

no FDA approved therapies for treating NASH. A major barrier for drug development in NASH is that

treatment response assessment continues to require liver biopsy, which is invasive and interpreted

subjectively. Therefore, there is a major unmet need for developing non-invasive, objective and

quantitative biomarkers for diagnosis and assessment of treatment response. Emerging data support

the use of magnetic resonance imaging derived proton density fat fraction (MRI-PDFF) as a non-invasive,

quantitative, and accurate measure of liver fat content to assess treatment response in early-phase of

NASH trials. In this review, we will discuss the role and utility, including potential sample-size reduction,

of using MRI-PDFF as a quantitative and non-invasive imaging-based biomarker in early-phase NASH

trials.

3
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 4 of 24

Caussy et al.

INTRODUCTION

Nonalcoholic fatty liver disease (NAFLD) is now the most common cause of chronic liver disease

worldwide (1-3). NAFLD can be broadly classified into two categories: nonalcoholic fatty liver (NAFL),

which is thought to have a minimal risk of progression to cirrhosis, and nonalcoholic steatohepatitis

(NASH), the more progressive form of NAFLD, which is thought to have a significantly increased risk of

progression to cirrhosis (4). Over the past two decades, NASH-related cirrhosis has become the second

leading indication for liver transplants in the United States (5). For these reasons pharmacological

therapy for NASH is urgently needed. Despite intensive investigations, there are currently no FDA

approved therapies for treating NASH (6).

Need for non-invasive assessment in treatment trials

Currently, therapeutic trials in NASH require liver biopsy to establish an initial diagnosis of NASH

and to document treatment response. However, it is an expensive and invasive procedure that carries

potential risks (abdominal pain, bleeding, death) and is consequently disfavored by many providers and

patients (7). Moreover, interpretation and scoring of biopsy are characterized by significant inter- and

intra-observer variability (7, 8) and biopsy assesses only a small liver sample, approximately 1/50,000th

of the liver. Given the known spatial heterogeneity of diffuse liver disease, limited sampling of the liver

can lead to meaningful errors in determining diagnosis, disease stage and longitudinal evolution (9, 10).

These limitations directly impact clinical trials design as the diagnostic accuracy, reliability, and response

to treatment end-points are key determinants of trial size requirements, feasibility, and costs.

Furthermore, extending clinical trial findings to clinical routine practice remains a major obstacle due to

barriers in obtaining repeated liver biopsies for treatment monitoring and patient follow-up. Therefore,

non-invasive, reliable, accurate, safe and quantitative biomarkers are needed as an alternative to liver

biopsy in clinical trials and to extend clinical trial practice to routine practice. Although histologic

4
Hepatology
This article is protected by copyright. All rights reserved.
Page 5 of 24 Hepatology

Caussy et al.

endpoint remain necessary in NASH clinical trials, emerging data support the use of magnetic

resonance imaging derived proton-density-fat-fraction (MRI-PDFF) for treatment response assessment

in early-phase trials in NASH for drugs which have an anti-steatotic mechanism of action.

MRI-PDFF

MRI-PDFF is a quantitative imaging biomarker that enables accurate, repeatable and

reproducible quantitative assessment of liver fat over the entire liver (9, 11-14). Thus, MRI-PDFF is

emerging as one of the leading non-invasive quantitative biomarkers suitable as a surrogate to liver

biopsy for assessing treatment response in a subset of NASH trials.

In this review article, we will summarize the currently available evidence regarding the benefits

of using MRI-PDFF in NASH patients and the advantages of this imaging technique compared to other

methods especially when this imaging is used as an endpoint in NASH trials.

MRI-PDFF Methodology: How does MRI-PDFF quantify liver fat?

Chemical-Shift Encoded MRI and Proton Density Fat Fraction

MRI is sensitive to the signal from protons in mobile, unbound molecules such as water and

triglycerides. Serendipitously, the signal from protons bound in structures such as the lipid bilayer of

cells (including cholesterols, sphingolipids, and phospholipids) are invisible using conventional MRI.

Further, differential electronic shielding of protons in water and triglycerides leads to differences in the

MR resonant frequency of different proton groups. This “chemical shift” between water and fat proton

signals can be exploited by emerging chemical shift encoded MRI (CSE-MRI) methods to quantify the

relative amount of water and fat signal arising from the tissue (15).

5
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 6 of 24

Caussy et al.

When all confounding factors such as T1 (16) and T2* (17-19), complex spectral characteristics of

triglycerides (18-20), the noise behavior of MRI (16) and MRI system instabilities (21, 22) have been

addressed, CSE-MRI methods can accurately measure the “proton density fat fraction” (PDFF) (23).

PDFF is defined as the ratio of the density of mobile protons from triglycerides and the total density of

protons from mobile triglycerides and mobile water. It is expressed as an absolute percentage (%) and

ranges from 0-100%. PDFF is a fundamental property of tissue that reflects the concentration of mobile

triglycerides within that tissue. Although PDFF correlates closely with chemically determined tissue

triglyceride concentration (24), PDFF and triglyceride concentration are not equivalent. Chemical-assay

measurements of triglyceride include MR invisible chemical species that do not contribute to PDFF

estimation. Similarly, PDFF is correlated with histological assessment of hepatic steatosis, which is

expressed as the percentage of cells containing intracellular droplets of fat (25). Although closely

correlated, these two metrics, both expressed as a percentage, are fundamentally different and not

equivalent metrics of tissue fat content Figure 1.

Types of MRI-PDFF Strategies

Two primary CSE-MRI strategies have emerged, known as “complex” CSE-MRI and “magnitude” CSE-

MRI. Complex-based methods utilize both the phase and the magnitude of the MRI signal and can fully

separate the water and fat proton signals. The advantages of complex CSE-MRI include a full PDFF range

from 0-100% fat concentration and improved signal to noise ratio (SNR) performance. Magnitude based

methods utilize only the magnitude of the MRI signal, and thus are limited to a dynamic range of 0-50%,

and lower SNR performance. However, magnitude based methods are less sensitive to system

instabilities. Both methods provide highly accurate and precise estimates of liver fat concentration,

which is almost always less than 50%, within the dynamic range of magnitude CSE-MRI. Both methods

can rapidly assess PDFF over the entire liver in a short breath-hold (~20s). PDFF maps are automatically

6
Hepatology
This article is protected by copyright. All rights reserved.
Page 7 of 24 Hepatology

Caussy et al.

reconstructed without user input or post-processing – a major advantage over MR spectroscopy based

methods Figure 2.

The data validating the use of advanced CSE-MRI methods to quantify PDFF is extensive. These include

phantom studies (26, 27), animal studies (28, 29), ex vivo human liver tissue (24), numerous studies

comparing CSE-MRI to MRS (12-14, 30-43), as well as liver biopsy as the reference standard (44, 45). A

comprehensive review of the technical details of CSE-MRI and data validating MRI-PDFF is beyond the

scope and purpose of this review.

When to use MRI-PDFF as an endpoint in NASH trials?

MRI-PDFF is accurate, precise, and reliable, with excellent inter and intra-rater agreement (11, 32, 37,

46, 47). Furthermore, it should be noted that complex MRI methods have been successfully

implemented, are FDA approved, and commercially available on the three major MRI vendors, GE

Healthcare, Siemens and Philips, ensuring eventual wide-spread availability. In addition, it has been

successfully applied in the setting of several clinical trials (32, 37, 48, 49).

Although MRI-PDFF is a useful tool, it is best suited for the following scenarios as a treatment endpoint

in NASH trials.

1. When the drug or intervention has a high likelihood of an anti-steatotic effect.

2. Typically in an early phase trial to see if there is a significant reduction in liver fat content along

with collateral improvement in an another biomarker (e.g. serum ALT or a plasma based

biomarker) before moving on to a larger study using a biopsy-based endpoint.

7
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 8 of 24

Caussy et al.

3. Interventions that are associated with weight loss would also benefit in quantifying the relative

reduction in liver fat content after intervention from baseline. This would help with sample-size

assessment for a larger phase 2B or phase 3 trial with liver histology as an endpoint.

4. When the drug or intervention has a strong likelihood of pro-steatotic effect, it would be useful

to include MRI-PDFF as an assessment of drug toxicity to quantify and assess the likelihood of

harm to the liver (e.g. basal insulin have been shown to increase liver fat content)

The role and utility of MRI-PDFF can be illustrated using the example of a trial that was designed to

assess the efficacy of colesevelam versus placebo in the treatment of NASH. In this trial, 50 biopsy-

proven NASH patients were randomized to either colesevelam or placebo for 24 weeks (37). Based upon

strong anti-steatotic from pre-clinical data in animal models, the primary hypothesis was that

colesevelam would significantly reduce liver fat compared to placebo. Therefore, MRI-PDFF was an

appropriate endpoint for assessment of treatment response in this clinical setting. Contrary to the study

hypothesis, colesevelam was found to increase hepatic MRI-PDFF. This small but real increase in liver

PDFF by colesevelam was only detectable with MRI-PDFF and not appreciable on liver histology

assessment, due to the lower sensitivity of liver biopsy compared to CSE-MRI to detect small but real

longitudinal changes. This example illustrates the utility of MRI-PDFF to detect small modification in liver

fat content in the setting of clinical trial and further studies are needed to determine how well an

improvement in hepatic steatosis correlates with resolution of NASH when a drug has different degree

of anti-steatotic activity. It is also important to understand the challenges and limitations of using MRI-

PDFF as a biomarker in early phase clinical trials.

1. When the drug or intervention has no or low likelihood of an anti-steatotic effect, MRI-PDFF is

unlikely to be useful for the assessment of treatment response in NASH. Under such

circumstances, alternative modalities and biomarkers should be considered.

8
Hepatology
This article is protected by copyright. All rights reserved.
Page 9 of 24 Hepatology

Caussy et al.

2. Although MRI-PDFF is suitable as a steatosis marker, it does not assess NASH, fibrosis,

inflammation, or other potential endpoints of interest

3. We anticipate that over the next several years, clinical trials will transition from magnitude-

based to complex-based MRI-PDFF sequences. This will likely alleviate some but not all the

quality control procedures currently needed. All of the commercially available MRI-PDFF

methods are complex based, and as they are disseminated will become increasingly available.

4. MRI-PDFF may not be feasible in a small minority of patients including claustrophobic patients,

patients too large to fit into the MRI scanner, and patients with contraindications to MRI such as

certain metallic implants. MRI-PDFF does not require the administration of gadolinium based

contrast agents.

5. MRI-PDFF sequences have relatively low accuracy for the detection of incidental but potentially

important abnormalities such as liver tumors. Patients should be counseled that the MRI-PDFF is

designed to measure liver fat, and that incidental abnormalities are not reliably excluded.

However, it should be noted that it is very straightforward to include MRI-PDFF as part of a

standard liver MRI protocol designed to detect and characterize liver tumors and other

abnormalities.

Benefits of using MRI-PDFF

MRI-PDFF versus MRS-PDFF

Advanced MRS is considered the most accurate method for measuring PDFF. However, MRS has limited

availability, is not fully supported by the system software on clinical MR scanners, usually needs to be

run on special research modes, requires technical expertise for its acquisition and analysis, and only

measures PDFF in one or a limited number of tissue voxels. The latter limitation introduces sampling

variability, especially in longitudinal studies, because the voxel placement is difficult to replicate exactly

9
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 10 of 24

Caussy et al.

with existing technology. Since MRI-PDFF and MRS-PDFF agree closely (13, 26, 32, 33, 37, 50), MRI-PDFF

is generally preferred in clinical trials due to its greater practicality and lower sampling variability.

MRI-PDFF versus liver tissue

Historically, liver biopsy with histology scoring was the reference standard for hepatic steatosis. Due to

its sampling variability and relatively broad grading categories, biopsy is insensitive to small but real

changes in liver fat content. A major problem with using liver tissue as an endpoint in clinical trials is

that true reductions (or progressions) in steatosis may be missed. Due to this problem as well as other

limitations mentioned earlier, liver biopsy is not recommended as an endpoint in clinical trials if the

primary outcome measure is steatosis reduction.

MRI-PDFF versus CT-attenuation; versus ultrasound-quantitative; versus ultrasound-qualitative; versus

CAP

The attenuation X-rays passing through fat is lower than water, therefore livers with hepatic steatosis

have lower attenuation on computed tomography (CT), an X-ray based imaging method (51, 52).

Recently, Kramer et al. directly examined the relationship between PDFF and CT in patients undergoing

same day CSE-MRI and non-contrast CT (53). Excellent linear correlation between PDFF and CT

attenuation was observed (r2=0.86), providing, for the first time, a direct calibration between PDFF and

CT attenuation. However, while the correlation of PDFF and CT attenuation was strong overall, there

was no meaningful correlation of PDFF and CT attenuation (r2=0.04) at low levels of liver fat (PDFF < 6%).

Thus, CT may be a useful biomarker to detect and quantify moderate to severe hepatic steatosis but its

utility is limited, particularly at low fat concentrations that are likely the most clinically relevant (41, 54).

Finally, the need for ionizing radiation makes CT less attractive, particularly in children, when alternative

non-invasive imaging methods such as MRI are readily available.

10
Hepatology
This article is protected by copyright. All rights reserved.
Page 11 of 24 Hepatology

Caussy et al.

Sonographic image brightness relates to the backscatter and attenuation of the ultrasound wave.

Compared to lean liver tissue, fatty tissue scatters and attenuates sound waves. Hence, mildly fatty liver

appears bright due to backscattered signals returning to the transducer. As the amount of liver fat

increases, the ultrasound wave becomes attenuated. Radiologists assess these changes (brightening of

the liver in the near field, darkening of the liver in the far field, blurring of vessels) qualitatively to

determine the presence of steatosis. Although qualitative assessment for liver fat may be useful

clinically, subjectivity and imprecision render this approach unsuitable for measuring steatosis and its

longitudinal change in clinical trials. To provide an objective sonographic assessment of hepatic

steatosis, new methods are being developed to quantify the degree of backscatter (55, 56) as potential

biomarkers of hepatic steatosis. Among them, the controlled attenuation parameter (CAP), measured by

Fibroscan (Echosens) allows a rapid assessment and is reasonably accurate for diagnosing the presence

of steatosis (57, 58). However CAP is limited by high failure rates in obesity, lack of exact anatomic

localization, and low accuracy for quantifying the amount of steatosis (59). The latter two factors make

CAP unsuitable for use as an endpoint in clinical trials due to measurement imprecision and inability to

monitor treatment response (60). Other quantitative ultrasound methods are investigational and not

ready for use in clinical trials.

Correlation between change in steatosis vs change MRI-PDFF and between change in MRS-PDFF vs

change in MRI-PDFF

Two recent studies, one in adults (32) and one in children (42) with known or suspected NAFLD, have

shown that longitudinal change in MRI-PDFF agrees closely with longitudinal change in MRS-PDFF (with

correlation coefficients of 0.96 to 0.99 and 0.986, respectively), when the MRI and MRS measurements

at each time points are meticulously co-localized. Longitudinal change in MRI-PDFF after weight loss

surgery is shown in Figure 3. These data provide further validation for the use of MRI-PDFF in

11
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 12 of 24

Caussy et al.

longitudinal clinical studies that will help to determine the prognostic significance of the severity and

change of steatosis.

How to use MRI-PDFF as an endpoint in NASH trials?

Site selection, qualification, training, and technical support

Although now commercialized by the three major vendors, complex-based MRI-PDFF sequences are

generally available only on the latest-generation scanners. Even if available, the sequences may require

the purchase of a license to enable their use. Therefore, a substantial proportion of sites participating in

clinical trials currently or in the next few years are unlikely to have access to these advanced sequences.

Until now, therefore, most clinical trials have used magnitude-based MRI-PDFF sequences. These usually

can be implemented by using commercial sequences developed for other purposes and then modifying

the acquisition parameters as needed, often under the guidance of a central radiology coordinating

center.

Site selection for clinical trials is usually based on hepatology expertise and enrollment capacity, not

imaging capability. Hence, the radiology coordinating center must determine the relevant technical

capabilities of each site, including field strength, scanner manufacturer, and scanner software. Based on

this information, a standardized protocol is developed that is both feasible at every site and adequate

for reliable PDFF estimation. A detailed MRI procedure manual is given to the sites, which usually

provides all the necessary training. If needed, additional questions can be answered by email or

teleconference. Site qualification is done in parallel with training and begins with a review of the site’s

technical qualifications. The ability to perform the exact protocol is then confirmed by scanning a

phantom (an inanimate object such as a bottle of water) or a human volunteer. Images are sent to

coordinating center, which verifies that all parameters are within the allowable range. Additionally, the

12
Hepatology
This article is protected by copyright. All rights reserved.
Page 13 of 24 Hepatology

Caussy et al.

coordinating center provides as-needed technical support if sites are unable to acquire images using the

allowed parameters or diverge from the protocol at any point in the study.

Quality control: acquisition, intake, analysis, reporting

Quality control (QC) is essential. The technologist performing the study at each site is responsible for

verifying MRI safety for each patient, positioning the patient correctly, adhering to the research

protocol, checking images as they are collected, and repeating any images that are degraded by a

correctable error such as sudden motion from a cough. The coordinating center is responsible for intake

QC (verifying all parameters are within range, the appropriate anatomy was covered, images are of

adequate quality for analysis), analysis QC (verifying that the analysis described below follows a

standard operating procedure, that the exact locations of the regions of interest are recorded, and that

the values taken from the images are within expected range or rechecked), and reporting QC (ensuring

that the values recorded form the images match the values inputted into the database and that the final

imaging database is complete)

Analysis: co-localization, number of ROIs, size of ROIs

Once PDFF maps have been acquired, analysis of these maps must be performed to derive a single PDFF

estimate for that MRI exam. There are many possible approaches and currently there is no official

consensus on how to analyze PDFF maps. Region of interest (ROI)-based methods that measure the

average PDFF value in a region of the liver are generally used, and standardized approaches are

emerging. Standardized approaches that have excellent intra- and inter-reader variability are preferred,

although they may be labor intensive. Whole-liver segmentation that avoids large blood vessels, bile

ducts, liver lesions and image artifacts would likely provide the best intra- and inter-reader variability,

but are not practical and probably not necessary. The use of multiple large ROI’s is usually sufficient to

13
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 14 of 24

Caussy et al.

provide adequate sampling of the liver. Hines et al first proposed a sampling strategy that placed one

ROI per Couinaud segment (61), while other groups have described the use of four ROI paradigms, with

one ROI in the anterior, posterior, medial and lateral segments (62). More recently, Campo et al

rigorously evaluated the effects of the number of ROI’s used, ROI size, and ROI location on the intra- and

inter-reader variability of PDFF measurements, and evaluated the time burden required for different ROI

strategies. Based on the results of this study, we conclude that the use of large ROI’s (≥4cm2) with either

a 4-ROI paradigm (anterior, posterior, medial, and lateral) or 9-ROI paradigm (Couinaud segments) are

preferred to provide the best intra- and inter-reader variability with an analysis time of approximately 1-

2.5 minutes by an experienced user (63)

Longitudinal changes: what is the quantitative change in liver fat that is associated with improvement in

liver histology?

In order for MRI-PDFF to be more widely acceptable for the assessment of treatment response in NASH,

one of the key data that are needed are the amount of liver fat decline that is clinical meaningful. Long-

term studies are needed to assess whether a sustained and significant reduction in liver fat will lead to

improvement in fibrosis, reduction in the risk to progression to cirrhosis and reduction in the risk of

death from liver disease. This may be a task that is not likely to be achieved in the near future.

Therefore, investigators have initiated seminal studies that aim to solve this puzzle by answering

questions that can be answered in the near future. One such attempt was done by Patel et al using

paired MRI-PDFF and liver histology data from two high quality randomized trials. Utilizing paired MRI-

PDFF and liver histology data, Patel et al. demonstrated that a relative reduction of 29% in liver fat on

MRI-PDFF is associated with a histologic response in NASH (defined as a 2-point improvement in NAFLD

activity Score) (64). Although these preliminary data need to be confirmed in larger cohort, they are now

being used to design future NASH clinical trials using the change in hepatic fat quantified by MRI-PDFF as

a treatment endpoint. Future studies are needed to assess whether there is a tipping point for liver fat

14
Hepatology
This article is protected by copyright. All rights reserved.
Page 15 of 24 Hepatology

Caussy et al.

reduction that if sustained over a 6 month period would be associated with either resolution of NASH or

improvement in one stage of fibrosis.

CONCLUSIONS

Non-invasive, quantitative, precise and reproducible, MRI-PDFF is emerging as a useful

biomarker to assess treatment response in the setting of early phase clinical trials in NASH. It is suitable

for quantifying liver fat content, however, it does not assess NASH, fibrosis, inflammation, or other

potential endpoints of interest. Therefore, multi-modality assessment of treatment response is needed

to examine the treatment response as MRI-PDFF is restricted to the liver fat domain alone. However, if

there is a large enough quantitative decline in liver fat content it may also be associated with other

collateral benefits such as improvement in inflammation. Emerging data suggests that a relative decline

of 30% or more may be clinically meaningful but remains to be validated in larger studies. We anticipate

that over the next several years, clinical trials will transition from magnitude-based to complex-based

MRI-PDFF sequences. This will likely alleviate some but not all the quality control procedures that are

currently needed, and help standardize liver fat quantification. Further research is being conducted to

develop an MRI-based package including MRI-PDFF, two-dimensional magnetic resonance elastography

(2D MRE) and three-dimensional (3D) MRE and other MR based biomarkers to have a comprehensive

liver disease assessment. The future is extremely bright for non-invasive assessment of treatment

response using imaging modalities with an exponential increase in innovative applications of MR-based

modalities. Such novel quantitative imaging modalities will likely continue to transform clinical trial

design in the years to come.

15
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 16 of 24

Caussy et al.

REFERENCES

1. Loomba R, Sanyal AJ. The global NAFLD epidemic. Nat Rev Gastroenterol Hepatol 2013;10:686-
690.
2. Chalasani N, Younossi Z, Lavine JE, Diehl AM, Brunt EM, Cusi K, Charlton M, et al. The diagnosis
and management of non-alcoholic fatty liver disease: practice Guideline by the American Association for
the Study of Liver Diseases, American College of Gastroenterology, and the American
Gastroenterological Association. Hepatology 2012;55:2005-2023.
3. 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.
4. Singh S, Allen AM, Wang Z, Prokop LJ, Murad MH, Loomba R. Fibrosis progression in
nonalcoholic fatty liver vs nonalcoholic steatohepatitis: a systematic review and meta-analysis of paired-
biopsy studies. Clin Gastroenterol Hepatol 2015;13:643-654 e641-649; quiz e639-640.
5. Wong RJ, Aguilar M, Cheung R, Perumpail RB, Harrison SA, Younossi ZM, Ahmed A. Nonalcoholic
steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation
in the United States. Gastroenterology 2015;148:547-555.
6. Rotman Y, Sanyal AJ. Current and upcoming pharmacotherapy for non-alcoholic fatty liver
disease. Gut 2016.
7. Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl J Med 2001;344:495-500.
8. Rockey DC, Caldwell SH, Goodman ZD, Nelson RC, Smith AD, American Association for the Study
of Liver D. Liver biopsy. Hepatology 2009;49:1017-1044.
9. Bonekamp S, Tang A, Mashhood A, Wolfson T, Changchien C, Middleton MS, Clark L, et al.
Spatial distribution of MRI-Determined hepatic proton density fat fraction in adults with nonalcoholic
fatty liver disease. J Magn Reson Imaging 2014;39:1525-1532.
10. Ratziu V, Charlotte F, Heurtier A, Gombert S, Giral P, Bruckert E, Grimaldi A, et al. Sampling
variability of liver biopsy in nonalcoholic fatty liver disease. Gastroenterology 2005;128:1898-1906.
11. Noureddin M, Lam J, Peterson MR, Middleton M, Hamilton G, Le TA, Bettencourt R, et al. Utility
of magnetic resonance imaging versus histology for quantifying changes in liver fat in nonalcoholic fatty
liver disease trials. Hepatology 2013;58:1930-1940.
12. Negrete LM, Middleton MS, Clark L, Wolfson T, Gamst AC, Lam J, Changchien C, et al. Inter-
examination precision of magnitude-based MRI for estimation of segmental hepatic proton density fat
fraction in obese subjects. J Magn Reson Imaging 2014;39:1265-1271.
13. Yokoo T, Shiehmorteza M, Hamilton G, Wolfson T, Schroeder ME, Middleton MS, Bydder M, et
al. Estimation of hepatic proton-density fat fraction by using MR imaging at 3.0 T. Radiology
2011;258:749-759.
14. Kang GH, Cruite I, Shiehmorteza M, Wolfson T, Gamst AC, Hamilton G, Bydder M, et al.
Reproducibility of MRI-determined proton density fat fraction across two different MR scanner
platforms. J Magn Reson Imaging 2011;34:928-934.
15. Reeder SB, Cruite I, Hamilton G, Sirlin CB. Quantitative assessment of liver fat with magnetic
resonance imaging and spectroscopy. J Magn Reson Imaging 2011;34:729-749.
16. Liu CY, McKenzie CA, Yu H, Brittain JH, Reeder SB. Fat quantification with IDEAL gradient echo
imaging: correction of bias from T(1) and noise. Magn Reson Med 2007;58:354-364.
17. Yu H, McKenzie CA, Shimakawa A, Vu AT, Brau AC, Beatty PJ, Pineda AR, et al. Multiecho
reconstruction for simultaneous water-fat decomposition and T2* estimation. J Magn Reson Imaging
2007;26:1153-1161.

16
Hepatology
This article is protected by copyright. All rights reserved.
Page 17 of 24 Hepatology

Caussy et al.

18. Yu H, Shimakawa A, McKenzie CA, Brodsky E, Brittain JH, Reeder SB. Multiecho water-fat
separation and simultaneous R2* estimation with multifrequency fat spectrum modeling. Magn Reson
Med 2008;60:1122-1134.
19. Bydder M, Yokoo T, Hamilton G, Middleton MS, Chavez AD, Schwimmer JB, Lavine JE, et al.
Relaxation effects in the quantification of fat using gradient echo imaging. Magn Reson Imaging
2008;26:347-359.
20. Hamilton G, Middleton MS, Hooker JC, Haufe WM, Forbang NI, Allison MA, Loomba R, et al. In
vivo breath-hold (1) H MRS simultaneous estimation of liver proton density fat fraction, and T1 and T2 of
water and fat, with a multi-TR, multi-TE sequence. J Magn Reson Imaging 2015;42:1538-1543.
21. Yu H, Shimakawa A, Hines CD, McKenzie CA, Hamilton G, Sirlin CB, Brittain JH, et al. Combination
of complex-based and magnitude-based multiecho water-fat separation for accurate quantification of
fat-fraction. Magn Reson Med 2011;66:199-206.
22. Hernando D, Hines CD, Yu H, Reeder SB. Addressing phase errors in fat-water imaging using a
mixed magnitude/complex fitting method. Magn Reson Med 2012;67:638-644.
23. Reeder SB, Hu HH, Sirlin CB. Proton density fat-fraction: a standardized MR-based biomarker of
tissue fat concentration. J Magn Reson Imaging 2012;36:1011-1014.
24. Bannas P, Kramer H, Hernando D, Agni R, Cunningham AM, Mandal R, Motosugi U, et al.
Quantitative magnetic resonance imaging of hepatic steatosis: Validation in ex vivo human livers.
Hepatology 2015;62:1444-1455.
25. Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW, Ferrell LD, et al.
Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology
2005;41:1313-1321.
26. Hines CD, Yu H, Shimakawa A, McKenzie CA, Brittain JH, Reeder SB. T1 independent, T2*
corrected MRI with accurate spectral modeling for quantification of fat: validation in a fat-water-SPIO
phantom. J Magn Reson Imaging 2009;30:1215-1222.
27. Hernando D, Sharma SD, Aliyari Ghasabeh M, Alvis BD, Arora SS, Hamilton G, Pan L, et al.
Multisite, multivendor validation of the accuracy and reproducibility of proton-density fat-fraction
quantification at 1.5T and 3T using a fat-water phantom. Magn Reson Med 2016.
28. Hines CD, Yu H, Shimakawa A, McKenzie CA, Warner TF, Brittain JH, Reeder SB. Quantification of
hepatic steatosis with 3-T MR imaging: validation in ob/ob mice. Radiology 2010;254:119-128.
29. Hines CD, Agni R, Roen C, Rowland I, Hernando D, Bultman E, Horng D, et al. Validation of MRI
biomarkers of hepatic steatosis in the presence of iron overload in the ob/ob mouse. J Magn Reson
Imaging 2012;35:844-851.
30. Artz NS, Haufe WM, Hooker CA, Hamilton G, Wolfson T, Campos GM, Gamst AC, et al.
Reproducibility of MR-based liver fat quantification across field strength: Same-day comparison
between 1.5T and 3T in obese subjects. J Magn Reson Imaging 2015;42:811-817.
31. Bashir MR, Zhong X, Nickel MD, Fananapazir G, Kannengiesser SA, Kiefer B, Dale BM.
Quantification of hepatic steatosis with a multistep adaptive fitting MRI approach: prospective
validation against MR spectroscopy. AJR Am J Roentgenol 2015;204:297-306.
32. Cui J, Philo L, Nguyen P, Hofflich H, Hernandez C, Bettencourt R, Richards L, et al. Sitagliptin vs.
placebo for non-alcoholic fatty liver disease: A randomized controlled trial. J Hepatol 2016;65:369-376.
33. Meisamy S, Hines CD, Hamilton G, Sirlin CB, McKenzie CA, Yu H, Brittain JH, et al. Quantification
of hepatic steatosis with T1-independent, T2-corrected MR imaging with spectral modeling of fat:
blinded comparison with MR spectroscopy. Radiology 2011;258:767-775.
34. Johnson BL, Schroeder ME, Wolfson T, Gamst AC, Hamilton G, Shiehmorteza M, Loomba R, et al.
Effect of flip angle on the accuracy and repeatability of hepatic proton density fat fraction estimation by
complex data-based, T1-independent, T2*-corrected, spectrum-modeled MRI. J Magn Reson Imaging
2014;39:440-447.

17
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 18 of 24

Caussy et al.

35. Kim KY, Song JS, Kannengiesser S, Han YM. Hepatic fat quantification using the proton density
fat fraction (PDFF): utility of free-drawn-PDFF with a large coverage area. Radiol Med 2015;120:1083-
1093.
36. Kuhn JP, Hernando D, Mensel B, Kruger PC, Ittermann T, Mayerle J, Hosten N, et al. Quantitative
chemical shift-encoded MRI is an accurate method to quantify hepatic steatosis. J Magn Reson Imaging
2014;39:1494-1501.
37. Le TA, Chen J, Changchien C, Peterson MR, Kono Y, Patton H, Cohen BL, et al. Effect of
colesevelam on liver fat quantified by magnetic resonance in nonalcoholic steatohepatitis: a randomized
controlled trial. Hepatology 2012;56:922-932.
38. Levin YS, Yokoo T, Wolfson T, Gamst AC, Collins J, Achmad EA, Hamilton G, et al. Effect of echo-
sampling strategy on the accuracy of out-of-phase and in-phase multiecho gradient-echo MRI hepatic fat
fraction estimation. J Magn Reson Imaging 2014;39:567-575.
39. Mashhood A, Railkar R, Yokoo T, Levin Y, Clark L, Fox-Bosetti S, Middleton MS, et al.
Reproducibility of hepatic fat fraction measurement by magnetic resonance imaging. J Magn Reson
Imaging 2013;37:1359-1370.
40. Motosugi U, Hernando D, Bannas P, Holmes JH, Wang K, Shimakawa A, Iwadate Y, et al.
Quantification of liver fat with respiratory-gated quantitative chemical shift encoded MRI. J Magn Reson
Imaging 2015;42:1241-1248.
41. Rehm JL, Wolfgram PM, Hernando D, Eickhoff JC, Allen DB, Reeder SB. Proton density fat-
fraction is an accurate biomarker of hepatic steatosis in adolescent girls and young women. Eur Radiol
2015;25:2921-2930.
42. Tyagi A, Yeganeh O, Levin Y, Hooker JC, Hamilton GC, Wolfson T, Gamst A, et al. Intra- and inter-
examination repeatability of magnetic resonance spectroscopy, magnitude-based MRI, and complex-
based MRI for estimation of hepatic proton density fat fraction in overweight and obese children and
adults. Abdom Imaging 2015;40:3070-3077.
43. Zand KA, Shah A, Heba E, Wolfson T, Hamilton G, Lam J, Chen J, et al. Accuracy of multiecho
magnitude-based MRI (M-MRI) for estimation of hepatic proton density fat fraction (PDFF) in children. J
Magn Reson Imaging 2015;42:1223-1232.
44. Tang A, Tan J, Sun M, Hamilton G, Bydder M, Wolfson T, Gamst AC, et al. Nonalcoholic fatty liver
disease: MR imaging of liver proton density fat fraction to assess hepatic steatosis. Radiology
2013;267:422-431.
45. Idilman IS, Aniktar H, Idilman R, Kabacam G, Savas B, Elhan A, Celik A, et al. Hepatic steatosis:
quantification by proton density fat fraction with MR imaging versus liver biopsy. Radiology
2013;267:767-775.
46. Permutt Z, Le TA, Peterson MR, Seki E, Brenner DA, Sirlin C, Loomba R. Correlation between liver
histology and novel magnetic resonance imaging in adult patients with non-alcoholic fatty liver disease -
MRI accurately quantifies hepatic steatosis in NAFLD. Aliment Pharmacol Ther 2012;36:22-29.
47. Dulai PS, Sirlin CB, Loomba R. MRI and MRE for non-invasive quantitative assessment of hepatic
steatosis and fibrosis in NAFLD and NASH: Clinical trials to clinical practice. J Hepatol 2016;65:1006-
1016.
48. Middleton MS, Heba ER, Hooker CA, Bashir MR, Fowler KJ, Sandrasegaran K, Brunt EM, et al.
Agreement Between Magnetic Resonance Imaging Proton Density Fat Fraction Measurements and
Pathologist-assigned Steatosis Grades of Liver Biopsies from Adults with Nonalcoholic Steatohepatitis.
Gastroenterology 2017.
49. Loomba R, Sirlin CB, Ang B, Bettencourt R, Jain R, Salotti J, Soaft L, et al. Ezetimibe for the
treatment of nonalcoholic steatohepatitis: assessment by novel magnetic resonance imaging and
magnetic resonance elastography in a randomized trial (MOZART trial). Hepatology 2015;61:1239-1250.

18
Hepatology
This article is protected by copyright. All rights reserved.
Page 19 of 24 Hepatology

Caussy et al.

50. Yokoo T, Bydder M, Hamilton G, Middleton MS, Gamst AC, Wolfson T, Hassanein T, et al.
Nonalcoholic fatty liver disease: diagnostic and fat-grading accuracy of low-flip-angle multiecho
gradient-recalled-echo MR imaging at 1.5 T. Radiology 2009;251:67-76.
51. Pickhardt PJ, Hahn L, Munoz del Rio A, Park SH, Reeder SB, Said A. Natural history of hepatic
steatosis: observed outcomes for subsequent liver and cardiovascular complications. AJR Am J
Roentgenol 2014;202:752-758.
52. Jehangir M, Nazir R, Jang A, Rana A, Rafique S, Dar FS. Macrovesicular steatosis in living related
liver donors: correlation of biopsy findings with CT liver attenuation index and body mass index. Clin
Transplant 2016;30:1016-1020.
53. Kramer H, Pickhardt PJ, Kliewer MA, Hernando D, Chen GH, Zagzebski JA, Reeder SB. Accuracy of
Liver Fat Quantification With Advanced CT, MRI, and Ultrasound Techniques: Prospective Comparison
With MR Spectroscopy. AJR Am J Roentgenol 2017;208:92-100.
54. Tang A, Desai A, Hamilton G, Wolfson T, Gamst A, Lam J, Clark L, et al. Accuracy of MR imaging-
estimated proton density fat fraction for classification of dichotomized histologic steatosis grades in
nonalcoholic fatty liver disease. Radiology 2015;274:416-425.
55. Lin SC, Heba E, Wolfson T, Ang B, Gamst A, Han A, Erdman JW, Jr., et al. Noninvasive Diagnosis
of Nonalcoholic Fatty Liver Disease and Quantification of Liver Fat Using a New Quantitative Ultrasound
Technique. Clin Gastroenterol Hepatol 2015;13:1337-1345 e1336.
56. Sasso M, Beaugrand M, de Ledinghen V, Douvin C, Marcellin P, Poupon R, Sandrin L, et al.
Controlled attenuation parameter (CAP): a novel VCTE guided ultrasonic attenuation measurement for
the evaluation of hepatic steatosis: preliminary study and validation in a cohort of patients with chronic
liver disease from various causes. Ultrasound Med Biol 2010;36:1825-1835.
57. Karlas T, Petroff D, Garnov N, Bohm S, Tenckhoff H, Wittekind C, Wiese M, et al. Non-invasive
assessment of hepatic steatosis in patients with NAFLD using controlled attenuation parameter and 1H-
MR spectroscopy. PLoS One 2014;9:e91987.
58. Caussy C, Alquiraish MH, Nguyen P, Hernandez C, Cepin S, Fortney LE, Ajmera V, et al. Optimal
threshold of controlled attenuation parameter with MRI-PDFF as the gold standard for the detection of
hepatic steatosis. Hepatology 2017.
59. de Ledinghen V, Vergniol J, Capdepont M, Chermak F, Hiriart JB, Cassinotto C, Merrouche W, et
al. Controlled attenuation parameter (CAP) for the diagnosis of steatosis: a prospective study of 5323
examinations. J Hepatol 2014;60:1026-1031.
60. Park CC, Nguyen P, Hernandez C, Bettencourt R, Ramirez K, Fortney L, Hooker J, et al. Magnetic
Resonance Elastography vs Transient Elastography in Detection of Fibrosis and Noninvasive
Measurement of Steatosis in Patients With Biopsy-Proven Nonalcoholic Fatty Liver Disease.
Gastroenterology 2017;152:598-607.e592.
61. Hines CD, Frydrychowicz A, Hamilton G, Tudorascu DL, Vigen KK, Yu H, McKenzie CA, et al. T(1)
independent, T(2) (*) corrected chemical shift based fat-water separation with multi-peak fat spectral
modeling is an accurate and precise measure of hepatic steatosis. J Magn Reson Imaging 2011;33:873-
881.
62. Sofue K, Mileto A, Dale BM, Zhong X, Bashir MR. Interexamination repeatability and spatial
heterogeneity of liver iron and fat quantification using MRI-based multistep adaptive fitting algorithm. J
Magn Reson Imaging 2015;42:1281-1290.
63. Campo CA, Hernando D, Schubert T, Bookwalter CA, Pay AJV, Reeder SB. Standardized Approach
for ROI-Based Measurements of Proton Density Fat Fraction and R2* in the Liver. AJR Am J Roentgenol
2017:1-12.
64. Patel J, Bettencourt R, Cui J, Salotti J, Hooker J, Bhatt A, Hernandez C, et al. Association of
noninvasive quantitative decline in liver fat content on MRI with histologic response in nonalcoholic
steatohepatitis. Therap Adv Gastroenterol 2016;9:692-701.

19
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 20 of 24

Caussy et al.

Author names in bold designate shared co-first authorship

20
Hepatology
This article is protected by copyright. All rights reserved.
Page 21 of 24 Hepatology

Caussy et al.

FIGURE LEGENDS

Figure 1 Correlation between MRI-PDFF and percentage of hepatocytes with steatosis by histology

Correlation between MRI-PDFF and histologic steatosis grade classified by the percentage of hepatocyte

with steatosis using (25) in individuals with biopsy-proven NAFLD (60).

Figure 2 MRI-PDFF assessment and quantification of hepatic steatosis

Example PDFF maps using complex MRI (C-MRI, left) and magnitude MRI (M-MRI, right) both show

elevated PDFF in the liver (~32%). M-MRI is limited to a dynamic range of 0-50% unlike C-MRI which has

a full dynamic range from 0-100%.

Figure 3 Longitudinal changes in MRI-PDFF after weight loss surgery

PDFF is shown in a patient before (pre-op) and after (post-op) weight loss surgery. Images from left to

right represent PDFF before a very low caloric diet (pre-VLCD) pre-op and the longitudinal follow-up

showing a decrease in PDFF.

21
Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 22 of 24

Figure 1 Correlation between MRI-PDFF and percentage of hepatocytes with steatosis by histology
Correlation between MRI-PDFF and histologic steatosis grade classified by the percentage of hepatocyte with
steatosis using (25) in individuals with biopsy-proven NAFLD (60).

80x60mm (300 x 300 DPI)

Hepatology
This article is protected by copyright. All rights reserved.
Page 23 of 24 Hepatology

Figure 2 MRI-PDFF assessment and quantification of hepatic steatosis

Example PDFF maps using complex MRI (C-MRI, left) and magnitude MRI (M-MRI, right) both show elevated
PDFF in the liver (~32%). M-MRI is limited to a dynamic range of 0-50% unlike C-MRI which has a full
dynamic range from 0-100%.

184x89mm (300 x 300 DPI)

Hepatology
This article is protected by copyright. All rights reserved.
 Hepatology Page 24 of 24

Figure 3 Longitudinal changes in MRI-PDFF after weight loss surgery

PDFF is shown in a patient before (pre-op) and after (post-op) weight loss surgery. Images from left to right
represent PDFF before a very low caloric diet (pre-VLCD) pre-op and the longitudinal follow-up showing a
decrease in PDFF.

189x69mm (300 x 300 DPI)

Hepatology
This article is protected by copyright. All rights reserved.