Effectiveness of lifestyle interventions in NAFLD (nonalcoholic fatty liver disease) – how are clinical trials affected?

KEYWORDS:

• NAFLD
• NASH
• lifestyle intervention
• pharmacotherapy
• clinical trials

1. Introduction

Nonalcoholic fatty liver disease (NAFLD) is the leading cause of liver disease on a global scale. The distinction of two histologically defined subtypes – nonalcoholic fatty liver (NAFL) and nonalcoholic steatohepatitis (NASH) is of prognostic and therapeutic relevance. The latter is considered an inflammatory form with hepatocyte injury that can progress to liver cirrhosis, and hepatocellular carcinoma (HCC). The increasing prevalence of NAFLD is linked to a sedentary lifestyle and dietary habits. Although current guidelines recommend a lifestyle aiming at weight loss to be achieved through increasing physical activity and dietary changes, not all patients respond to these measures and pharmacotherapy is eagerly awaited for subgroups. As a matter of fact, despite the broad consensus that lifestyle is beneficial to patients with metabolic disease, only few well performed and controlled trials are available in NAFLD and thus the interpretation of their usefulness is limited. A large and prospectively performed lifestyle intervention in cardiovascular outcomes – the Look AHEAD study – was negative at the time point of 8 years [1], and highlighted the need to define a feasible and effective non-pharmacological treatment for patients with NAFLD. When comparing the available trials (see Table 1), the extent and intensity of the lifestyle interventions differ largely and it will be difficult to transfer findings from a dedicated lifestyle intervention trial to pivotal trials. At current, several drugs are in late stage clinical development for fibrotic NASH and all of these include advice on lifestyle. This summary aims to dissect recently published clinical trials that specifically focus on lifestyle intervention, and compare these to effects observed in phase 2 and phase 3 pharmacological trials.

Table 1. Lifestyle intervention trials.

Name of study	Study population, criteria of inclusion	What kind of lifestyle intervention?	Frequency of consultations during the study period	Primary endpoints	Effect, response rates	Reference
Lifestyle intervention for morbid obesity: effects on liver steatosis, inflammation, and fibrosis.	152 patients with obesity class I–III (BMI >30 kg/m^2);	800 kcal/day (96 g of carbohydrates, 70 g of protein, 15 g of fat; 10 g of fibers/day) for 12 weeks in total; replacement by balanced, normal meals at week 13 til week 26; until the end of the study patients had normal, balanced, protein-rich meals (recommendations of the German Nutrition Society) with a caloric intake balanced to hold weight from week 26 onwards.	Total of 52 weeks with weekly group and individual counseling sessions (medical doctors, psychologists, nutrition scientists, and physical education instructors); Clinical data and serum were collected on day 0 as well as after weeks 13, 26, and 52.	Weight loss of >5% for patients with a BMI of < 35 kg/m^2 and weight loss of > 10% for patients with a BMI of > 35 kg/m^2;	In the intention-to-treat-analysis, 67.8% of patients achieved therapy response, whereas per protocol response was 85.1% at a follow up of 1 year.	Hohenester et al. 2018 [3]
Low Glycemic Index Dietary Intervention Program in Nonalcoholic Fatty Liver Disease (NCT 00868933)	154 NAFLD patients. Screening of NAFLD with proton-magnetic resonance spectroscopy (H-MRS) and transient elastography. Inclusion criteria: an intrahepatic triglyceride content (IHTG) of 5% or above. Age 18 to 70 years and elevated serum alanine aminotransferase (ALT) level (≥30 U/L in men and ≥19 U/L in women).	Intervention group (n = 77) participated in a lifestyle intervention program led by dietitians and exercise trainers for 12 months. Placebo group (n = 77) received routine care at the medical clinic with the advice to reduce weight and to exercise.	Intervention group: Individual education and consultation sessions weekly in the first four months and monthly for the remaining eight months Follow up meetings were scheduled at baseline, 3, 6, 9 and 12 months during the trial period, and yearly thereafter.	Remission of NAFLD at month 12, defined as an IHTG below 5.0% by H-MRS	BMI ≥25 kg/m^2: 61% in the intervention group and 21% in the control group had remission of NAFLD (p < 0.001). BMI <25 kg/m^2: 67% in the intervention group had remission of NAFLD, compared with 18% in the control group (p < 0.001) In the intervention group, a similar proportion of patients with baseline BMI <25 and ≥25 kg/m^2 achieved remission of NAFLD (67% vs. 61%; p = 0.58).	Wong et al. 2018 [4]
Hepatic Inflammation and Physical Performance in Patients With NASH (HELP‐Study) (NCT 02526732)	44 patients with histologically confirmed NAFLD. NASH was present in 70.7% (n = 29) with 58.5% (n = 24) exhibiting F2/F3 fibrosis.	8‐week individualized training program consisting of combined endurance and strength exercise three to five times per week.	The individual counseling on exercise was stopped at week 8 (end of treatment, EoT), while the recommendation to continue was given to all participants with free, non‐supervised access to the online platform. The end of follow‐up (EoF) visit was conducted at week 20, 12 weeks after EoT.	The primary endpoint of the study was defined as a change in peak oxygen uptake (VO2peak) from the baseline (SoT)	The increase in VO2peak during the 8‐week intervention was comparable to the increase in 16‐weeks.	Huber, et al. 2019 [9]
Diet and Body Composition (NCT01530724)	278 patients in total. Abdominal obesity (waist circumference [WC] >40 inches [102 cm] for men and >35 inches [88 cm] for women), or participants with triglycerides ≥150 mg/dl and high-density lipoprotein (HDL) <40 mg/dl for men and <50 mg/dl for women.	First randomization: participants were randomly assigned to 1 of 2 calorie-restricted diets for the entire study period: an LF diet (n = 139) or a Mediterranean/low-carbohydrate (MED/LC) diet (n = 139). Second randomization: 6 months after initiating the dietary intervention, each diet group was further randomized into added physical activity (PA) groups (LFPA+, MED/LCPA+) or no added PA groups (LFPA, MED/LCPA) for another 12 months of intervention (diets were continued throughout, according to the first randomized dietary assignment).	The 18-month dietary intervention included a 90-min nutritional session in the workplace with clinical dietitians every week during the first month of the intervention, and every month thereafter. Participants were counseled to consume whole grains, vegetables, fruits, and legumes and to limit their consumption of additional fats, sweets, and high-fat snacks. Evaluation of adherence to the study was conducted through a food-frequency and activity questionnaire (FFQ) at baseline and after 6 and 18 months of intervention.	Change in body fat composition.	MED/LC induced a significantly greater decrease in HFC than the LF diet, also after accounting the differences in VAT loss.	Gepner et al. 2019 [7]

2. Dietary, physical and pharmacological interventions in NAFLD

2.1. Studies on dietary interventions

Up to date, the largest and histologically controlled trial performed in NASH was the study performed in Cuba by Vilar Gomez et al. Herein a 52-week long intensive intervention that combined physical activity, dietary counseling, and psychological support in the treatment arm examined 293 patients with histologically proven NASH. With this elaborate lifestyle intervention program, 30% of patients lost more than 5% of their body weight and 25% achieved resolution of steatohepatitis [2]. Thus, this study has set the standard and comparator of what can be reached in NASH at 52 weeks from intensive counseling. Importantly, it also highlighted that not all patients achieved resolution of steatohepatitis – even when weight loss exceeded 5% from baseline [2]. Thus, further studies, that use less resources could be important in the management of NASH. In a monocentric, single arm, interventional study by Hohenester et al., the effects of a formula diet on hepatic steatosis and fibrosis were determined in patients with morbid obesity in Germany [3]. The intervention consisted of calorie restriction to 800 kcal/day for 12 weeks, with a follow-up of 13 weeks on a normal, balanced meal. From week 26 until the end of the study (52nd week) patients remained on a protein-rich diet with a balanced caloric intake to hold body weight. Treatment response was predefined as weight loss of >5% for patients with a BMI of <35 kg/m² and weight loss of >10% for patients with a BMI of >35 kg/m². Liver histology was not available, but the authors provide noninvasive biomarkers including ALT, AST, fatty liver index (FLI), NAFLD fibrosis score (NFS), and liver ultrasound. This trial can be set into perspective of the Cuban weight loss study – which demonstrated a resolution of steatohepatitis from 5% to 7% of weight loss in 26% and fibrosis resolution in 81% that achieved >10% weight loss [2]. In the trial by Hohenester et al., a weight loss of more than 10% showed a significant decline of ALT serum levels and an improvement of hepatic steatosis. In contrast, weight loss <10% was overall less significantly efficient in order to achieve a ‘complete hepatic response’ (normal serum ALT values, improved FLI and NFS) at week 52 [3].

A different approach was taken by Wong et al. investigating lifestyle changes in an Asian population that was considered non-obese and exhibited NAFLD (so-called lean NAFLD) [4]. Half of these patients exhibited a BMI below 25 kg/m². Hepatic steatosis was assessed noninvasively using proton-magnetic resonance spectroscopy (H-MRS) and hepatic fibrosis by transient elastography. Patients with intrahepatic triglyceride content (IHTG) exceeding 5.0% detected on H-MRS were included, and the primary outcome analysis focused on the reduction of IHTG below 5.0% at month 12. The experimental arm received individualized education and physical exercise training by specialists focused on increasing energy expenditure and reducing caloric intake. The control group received more general advice on a healthy diet with recommendations to reduce calorie intake and increase exercise. This intensity of lifestyle counseling likely reflects the real-world clinical scenario. Remission of hepatic steatosis on H-MRS was achieved in all patients with >10% weight reduction almost equally distributed among the two BMI baseline groups of <25 kg/m² and >25 kg/m², respectively. In 70% of patients a remission of NAFLD was seen after a weight reduction of 5–7%. Therefore, a weight loss of 5–10% from lifestyle changes appear to effectively lead to the resolution of hepatic steatosis – even in (Asian) patients with a BMI below 25 kg/m² [4].

The Mediterranean diet, which is rich in monounsaturated fats, fruits, vegetables, nuts, and low in processed foods, as well as red meat, has been studied extensively in cardiovascular disease, and also in obesity, as well as NAFLD [5]. This type of diet appears to be promising, as the amount of red and processed meat consumption correlates with NAFLD and insulin resistance [6]. In a randomized-controlled trial by Gepner et al., 278 participants were randomly assigned to either a low-fat or Mediterranean/low carbohydrate diet with or without physical activity, respectively, for 18 months. The primary read out was hepatic fat content and visceral adipose tissue determined by magnetic resonance imaging. Overall, a Mediterranean diet was more effective in reducing hepatic fat content and led to an improvement in surrogates of cardiometabolic risk as compared to a simple low-fat diet. During the study duration of 18 months, the dietary intervention caused a predominant reduction in hepatic fat and only to a lesser extent to visceral fat. Although significant changes were seen in hepatic fat content, a loss in body weight was only moderately affected from a baseline BMI of 30.8 kg/m² (- 5.8% and – 3.1% after 6 and 18 months, respectively) [7]. In this regard, Bjermo H et al. investigated the effects of a change in dietary fats (polyunsaturated fatty acids vs saturated fatty acids) on liver fat. The participants in the study received counseling related to the fat intake without other lifestyle modifications. A change to plant-based polyunsaturated fatty acids – foods that were rich in linoleic acid from margarine, sunflower oil, and sunflower seeds – was associated with a significant decrease of hepatic fat as compared to a diet rich in saturated fats – even in the absence of weight loss [8].

2.2. Physical exercise

Although physical activity is improving metabolic control and is recommended in patients with NAFLD, most patients fail to adhere to exercise recommendations. Thus, supported exercise programs are needed as they give a guiding framework based on ability and exhaustion. Huber et al. investigated the effects of a short, web-based, individualized exercise program of only 8 weeks on noninvasive markers of steatosis, inflammation, and fibrosis. In this study, physical activity was assessed in the absence of specific dietary recommendations. While patients did not reduce weight, ALT decreased after 8 weeks and Pro-collagen 3 (PRO-C3) – a surrogate of fibrogenesis – decreased after 20 weeks [9]. Importantly, the web-based study exhibited a relatively low dropout rate. Secondary outcomes included markers of hepatic inflammation and fibrosis and quality of life, all of which showed improvements even beyond the exercise period. Interestingly, patients with NASH exhibited a larger benefit compared to plain hepatic steatosis without inflammation. This study highlights the role that supervised exercise programs could have a potential in providing care to patients with NASH [9].

2.3. Lifestyle in clinical trials

A number of phase 2 and phase 3 clinical trials are currently running to evaluate the efficacy and safety of pharmacological agents in NASH and are summarized in this issue. While lifestyle could be a way to achieve NASH resolution in a subset of patients, a significant number will not improve their liver phenotype from lifestyle recommendations alone. Considering that all clinical trials included counseling on lifestyle and exercise as part of the treatment arms, an estimation on the effect size can be derived from these clinical trials, in particular, the placebo arm. Importantly, counseling on lifestyle change is not the main focus in those studies, but the extent could be more comparable to an out-patients hepatologist visit in the clinical-care setting. Yet, no drug has been approved and more double-blind randomized controlled studies may be necessary to obtain pharmacological interventions with a lasting effect.

In the FLINT trial, a multicenter double-blind placebo-controlled phase 2b trial, the farnesoid X nuclear receptor ligand – obeticholic acid – was evaluated in non-cirrhotic NASH patients [10]. A total of 283 patients were treated in two groups receiving obeticholic acid or placebo, respectively. Resolution of steatohepatitis was observed in 45% of the patients in the treatment group as compared to 21% in the placebo group. Overall, patients assigned to treatment had an improvement of the NAFLD activity score (NAS; steatosis, hepatocellular ballooning, and lobular inflammation). However, no difference between the groups regarding a complete resolution of NASH was observed [10]. The effect size of 21% in the placebo group is largely related to the effects of lifestyle changes that are implemented by the patients during clinical trials. Unfortunately, none of the current trials capture the extent of these changes, and weight loss is the only measure to assess the degree.

In another double-blind placebo-controlled trial, the efficacy and safety of elafibranor – an agonist of the peroxisome proliferator-activated receptor-α and -δ – a more stringent primary endpoint, resolution of NASH without fibrosis worsening, was chosen [11]. Patients were treated for 1 year with two doses of elafibranor (80 and 120 mg) or placebo, respectively, in patients with a NAFLD activity score (NAS) ranging from 3 to 8. This study highlighted that patients with lower NAS can achieve resolution of steatohepatitis more easily in the placebo group, talking to the susceptibility of NASH to lifestyle changes – in particular when the histological activity was low. It was only on post-hoc analysis and after exclusion of 15% of patients with mild steatohepatitis (NAS = 3, n = 40) that 120 mg elafibranor was superior in achieving the primary endpoint compared to placebo [11]. This study highlights that patients with an intermediated NAS Score might be good candidates to undergo lifestyle changes, whereas pharmacotherapy is warranted in patients with advanced fibrosis, inflammation, and NAS of higher than 4.

In a randomized, double-blind, placebo-controlled phase 2a study with the administration of pegbelfermin, a PEGylated human fibroblast growth factor 21 (FGF21) analogue, the primary outcomes were chosen to be safety and the absolute change in hepatic fat fraction measured by MRI after 16 weeks of treatment in patients with NASH (fibrosis stages 1–3) [12]. Inclusion criteria were based on a BMI of 25 kg/m², biopsy-confirmed NASH (fibrosis stage 1–3), and a hepatic fat fraction of at least 10%. While the treatment groups reduced hepatic fat fraction by −6.8% and −5.2% on MRI-PDFF, the placebo group exited only a small decrease of −1.3%. Likewise, the effect of pegbelfermin on low-density lipoprotein cholesterol (LDL-C), triglycerides, adiponectin, and PRO-C3 were beneficial and significant compared to the placebo group [12]. Interestingly, in this study that included patients with precirrhotic fibrosis stages, an unsupervised recommendation to change lifestyle did not translate into a lasting benefit with improving metabolic parameters or surrogates of liver disease severity.

3. Conclusion

The future of NAFLD and NASH treatment needs to be considered from several perspectives. Counseling on lifestyle changes – including dietary adjustments and physical activity – will be important but not sufficient for patients in most of the regular clinical settings. While weight loss will be a good predictor of response to lifestyle – in particular when exceeding 10% from baseline – liver histology might still be required to assess the disease activity and stage until better performing noninvasive scores have been developed. A temporary alternative to immediate pharmacotherapy could be supervised and counseled lifestyle programs; however, considering the failure rates of these, drugs will still be required to prevent disease progression in most patients. Future pivotal trials could benefit from a differential recording of activity and diet. In the end, it has to be expected that these measures will not be sufficient to reduce the large-disease burden currently observed.

4. Expert opinion

Lifestyle interventions can successfully achieve NASH and fibrosis resolution. In most cases, this is associated with weight loss of more than 5%. But even in the absence of weight loss, the dietary composition can influence the hepatic phenotype and in particular fat content – and despite the lack of histology in most lifestyle trials – a benefit can be assumed based on the available surrogate markers. The need for pharmacological treatment becomes obvious when considering that approximately two-thirds of patients do not respond with NASH resolution. Additionally, the complexity of achieving a permeant change in lifestyle is high. One way to overcome this hurdle is to analyze subgroups that could benefit from a specific intervention or combination approach. In order to provide care to all patients, pharmacotherapy in combination with a lifestyle intervention will likely be required. The aim of future research in this indication is to identify patients that benefit most from a specific intervention and to implement this before starting drug therapy.

Declaration of interest

J Schattenberg has received consultancy fees from BMS, Boehringer Ingelheim, Echosens, Galmed, Genfit, Gilead Sciences, Intercept Pharmaceuticals, IQVIA, Madrigal, Novartis, Pfizer, research funding from Gilead Sciences and lecturing fees from AbbVie, Falk Foundation, Takeda, and MSD. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.