1. NASH: State of the Science
COSIMO COLLETTA
LIVER MEETING
28 NOVEMBRE 2017

2. Subtypes of NAFLD NAFLD Caveats Steatosis in ≥ 5% hepatocytes
Minimal alcohol use
Biopsy consistent with NAFLD
No other etiology for liver disease
No secondary causes of NAFLD
Medications
HIV
Lipodystrophy
NAFL
NASH
Nonprogressive
Progressive
NAFL, nonalcoholic fatty liver; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis.
Borderline NASH

3. Steatohepatitis (NASH)
The Spectrum of NAFLD
Steatosis (NAFL)
Steatohepatitis (NASH)
Fibrosis
Cirrhosis
HCC

4. The Spectrum of NAFLD: Steatosis
Steatosis (NAFL)
Fat infiltration > 5%
± mild inflammation

5. The Spectrum of NAFLD: Steatohepatitis
Steatosis (NAFL)
Steatosis + necroinflammation (eg, ballooning, Mallory bodies, megamitochondria)
Steatohepatitis (NASH)

6. The Spectrum of NAFLD: Fibrosis/Cirrhosis to HCC
Steatosis (NAFL)
Increasing fibrosis, eventually leading to cirrhosis
Steatohepatitis (NASH)
Fibrosis
Cirrhosis
HCC

7. Obesity and Insulin Resistance as Pathogenic Drivers
Central Obesity
Hypertension
Hyperuricemia
Dyslipidemia
NAFLD
Hyperglycemia
Macrovascular Disease
Thrombophilia

8. Relationship Between Liver Fat and Metabolic Syndrome
Nondiabetic subjects (N = 271)
Presence of all components of metabolic syndrome correlated with liver fat content by 1H-MRS
Liver fat content significantly increased in pts with metabolic syndrome vs those without
This association is independent of age, sex, BMI
Kotronen A, et al. J Clin Endocrinol Metab. 2007;92:

9. NAFLD as a Complex Disease Trait: Genetic and Environmental Modifiers
Normal
Steatosis
NASH
Cirrhosis
Environment
Sedentary lifestyle
Snacking, fast food
Saturated fats
Trans fats
Processed red meat
Genes
PNPLA3
TM6SF2
GCKR
SOD2
MBOAT7

10. A “Right-Shift” in Population Body Weight
UK Distribution of BMI in Adults Aged ≥ 18 Yrs (Population Weighted)
Underweight
Healthy weight
Overweight
Obese
Severely obese
Men,
Men,
Women,
Women,
Proportion of population
Men 14.6% 24.8%
Women 16.4% 22.8%
Men 0.7% 0.9%
Women 1.3% 1.8%
Men 0.3% 1.7%
Women 1.4% 3.7%
Men 37.8% 28.4%
Women 47.6% 38.2%
Men 46.7% 44.3%
Women 33.3% 33.6% 12
18.5 25 30 40 50
BMI (kg/m2)
Public Health England. Patterns and trends in adult obesity. April 2016.

11. Prevalence of NAFLD in the General Population
In a population sample (n = 328)[1]:
NAFLD by ultrasound 46%
NASH 12.2% (29.9% of those with NAFLD)
Rising Obesity Prevalence Correlates With Rising Prevalence of NAFLD[4] 35 30
USA
Israel
In the Dionysos study, NAFLD present in[2]:
Obese pts (BMI > 30), 94%
Overweight pts (BMI > 25), 67%
Normal weight pts, 25% 25
Korea 20
Taiwan
India
Prevalence of NAFLD (%)
China
Japan 15
Mexico 10
Italy
In apparently healthy living liver donors, histological NASH in[3]:
Europe, 3% to 16%
USA, 6% to 15% 5 5 10 15 20 25 30 35
Prevalence of Obesity (%)
1. Williams CD, et al. Gastro. 2011;140:
2. Bellentani S, et al. E J Gastro Hep. 2004; Anstee et al. Nat Rev Gastroenterol Hepatol. 2013;10: Loomba R, et al. Nat Rev Gastroenterol Hepatol. 2013;10:

12. Is It Important to Determine if Your Patient Has NASH vs NAFL?
NAFL, nonalcoholic fatty liver; NASH, nonalcoholic steatohepatitis.

13. NAFLD: Mortality
Long-term follow-up of pts with biopsy-confirmed NAFLD
Survival
1.00
0.75
0.50
0.25 10 20 30
NAFL: fatty liver only (n = 8)
NASH: fatty liver with inflammation, ballooning, or fibrosis (n = 109)
NAFL, nonalcoholic fatty liver; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis.
Yrs
Slide credit: clinicaloptions.com
Söderberg C, et al. Hepatology. 2010;51:

14. A 50-Yr-Old Man With Fatty Liver on Ultrasound May be Considered for Biopsy if He Has…
Near normal ALT, AST > ALT suggests advanced fibrosis

15. Indications for Liver Biopsy in Pts With NAFLD
Perform liver biopsy
More features of metabolic syndrome
Obesity, hypertension, increased TG, low HDL, impaired glucose tolerance
Diabetes
Family history of diabetes
Older age
High AST/ALT
Low platelets/albumin
Consider liver biopsy
Cholecystectomy
Bariatric surgery
Low Cutoff (NPV)
High Cutoff (PPV)
LDL, low density lipoprotein; NAFLD, nonalcoholic fatty liver disease; NPV, negative predictive value; PPV, positive predictive value; TG, triglycerides.
Low Probability of F3/4
NAFLD Fibrosis Score <
FIB-4 Score < 1.3
Indeterminate
High Probability of F3/4
NAFLD Fibrosis Score > .676
FIB-4 Score > 3.25
Chalasani N, et al. Hepatology. 2012;55: Noureddin M, et al. Clin Liver Dis. 2012;1:

16. What Is the Significance of Fibrosis?

17. Key Histologic Predictors of Mortality in NAFLD
Fibrosis is the single most important predictor of mortality in NASH
Advanced fibrosis
Fibrosis
Fibrosis
Portal
inflammation
Increasing prognostic value
NASH
NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis.
NASH
Ballooning
Angulo P, et al. Gastroenterology. 2015;149: Loomba R, et al. Gastroenterology. 2015;149:

18. Quantitative Risk of Liver Mortality by Fibrosis Stage
Meta-analysis of 5 cohort studies with N = 1495 pts with NAFLD followed for 17,452 pt-yrs
Liver-related mortality exponentially increased with fibrosis stage
Liver-Related Mortality 25 20 15
Mortality Rate (per 1000 PYFU)
NAFLD, nonalcoholic fatty liver disease; PYFU, patient-years of follow-up. 10 5 1 2 3 4
Fibrosis Stage
Dulai PS, et al. Hepatology. 2017;65:

19. MR-Based Fibrosis Assessment in NASH: Innovations in Fibrosis Assessment
NASH, nonalcoholic steatohepatitis.

20. AUC for diagnosis of advanced fibrosis: 0.924
MRE: Advanced Fibrosis Diagnosis 7
Cutoff for Detecting Advanced Fibrosis
Sensitivity
Specificity
MRE stiffness
> 3.63 kPa
.86
.91 6
AUC, area under the curve; MRE, magnetic resonance elastography. 5
2D-MRE (Kpa) 4 3
AUC for diagnosis of advanced fibrosis: 0.924 2 1 2 3 4
Fibrosis stage
Slide credit: clinicaloptions.com
Dulai PS, et al. Hepatology. 2017;65:

21. Which Modality Is More Accurate? MRE vs VCTE/ARFI MRI-PDFF vs CAP
CAP, controlled attenuation parameter; MRE, magnetic resonance elastography; PDFF, proton density fat fraction; VCTE/ARFI, vibration-controlled transient elastography/acoustic radiation force imaging.

22. MRE Better Than ARFI for Detection of Fibrosis in NAFLD: Prospective Study
Head-to-head comparison in N = 125 consecutive pts with biopsy-proven NAFLD and contemporaneous MRE and ARFI
1.0
MRE
ARFI
P < .05
0.8
0.6
Sensitivity
Effect modifier for ARFI
BMI ≥ 30
0.4
ARFI, acoustic radiation force impulse; BMI, body mass index; MRE, magnetic resonance elastography; NAFLD, nonalcoholic fatty liver disease.
0.2
0.2
0.4
0.6
0.8
1.0
Specificity
Cui J, et al. Hepatology. 2016;63:

23. MRE Better Than TE for Detection of Fibrosis in NAFLD
Head-to-head prospective comparison
N = 104 pts with biopsy-proven NAFLD from UCSD NAFLD Research Center TE
MRE
Fibrosis Stage
Stage 1-4 vs 0
Stage 2-4 vs 0-1
Stage 3-4 vs 0-2
Stage 4 vs 0-3
AUROC
1.0
0.8
0.6
0.4
0.2
P = .01*
P = .46
P = .19
P = .05
AUROC, area under the receiving operator characteristic; MRE, magnetic resonance elastography; NAFLD, nonalcoholic fatty liver disease; TE, transient elastography; UCSD, University of California, San Diego.
*Delong test used to compare the AUROCs.
Park CC, et al. Gastroenterology. 2017;152:

24. MRI-PDFF Better Than CAP for Quantification of Steatosis Grade in NAFLD
CAP MRI
Steatosis Grade
Grade 1-3 vs 0
Grade 2-3 vs 0-1
Grade 3 vs 0-2
AUROC
1.0
0.8
0.6
0.4
0.2
P < .01*
P < .02*
AUROC, area under the receiving operator characteristic; CAP, controlled attenuation parameter; NAFLD, nonalcoholic fatty liver disease; PDFF, proton density fat fraction.
*Delong test used to compare the AUROCs.
Park CC, et al. Gastroenterology. 2017;152:

25. Innovations in Clinical Trial Design: How Will Future Clinical Trials Assess NASH?
NASH, nonalcoholic steatohepatitis.

26. Phase II MOZART Trial: Fat and Stiffness Mapping Before and After Treatment
Randomized, double-blind, allocation-concealed, placebo-controlled phase II study
First study to assess 2D and 3D MRE in NASH
Why do we need to colocalize?
Heterogeneity in distribution
More comprehensive assessment
Whole-liver fat mapping
Liver stiffness
Higher precision and accuracy
MRE, magnetic resonance elastography; NASH, nonalcoholic steatohepatitis.
Enhanced responsiveness
Efficiency in clinical trial
Loomba R, et al. Hepatology. 2015;61:

27. Summary: Caveats Associated With Available Modalities for Diagnosing and Staging NAFLD
Transient elastography or ARFI or other ultrasound-based test have following limitations:
Obesity
Ascites
Acute inflammation
Cirrhosis
MRE improves upon all except
Iron overload
Acute inflammation
MRE
Accuracy
ARFI, acoustic radiation force imaging; MRE, magnetic resonance elastography; NAFLD, nonalcoholic fatty liver disease; VCTE/ARFI, vibration-controlled transient elastography.
ARFI/Shear wave elastography
Accessibility
Ease of use
VCTE/FibroScan
Park CC, et al. Gastroenterology. 2017;152: Cui J, et al. Hepatology. 2016;63:

28. Approach to Initial Assessment and Consideration for a Liver Biopsy
Elevated liver enzymes or evidence of hepatic steatosis on imaging
Baseline workup:
CBC, platelet count, ALT, AST, ALP, GGT, INR, total bilirubin, albumin
Rule out other causes of chronic liver disease (eg, viral hepatitis, autoimmune)
Fasting glucose and lipid levels, A1C
Trial of 3-6 mos of diet and exercise for weight loss
Unsuccessful
ALP, alkaline phosphatase; CBC, complete blood count; GGT, gamma-glutamyl transferase; INR, international normalized ratio.
Presence of:
Diabetes
Metabolic syndrome
Older age
High AST:ALT
High AST:platelet
Decreased albumin or platelet count
Consider biopsy if undergoing cholecystectomy or bariatric surgery
Yes No
Liver biopsy
Reassess every 6-12 mos
Noureddin M, et al. Clin Liver Dis. 2012;1:

29. Population of Interest for Treatment
NASH
NASH with fibrosis
Advanced fibrosis
NASH-related cirrhosis
Do Not Treat
NAFL
Pt without biopsy-confirmed NASH
Steatosis alone
Focus on CVD risk factor modification in primary care; no need for liver clinic
CVD, cardiovascular disease; NAFL, nonalcoholic fatty live; NASH, nonalcoholic steatohepatitis.

30. How Much Weight Loss Is Needed for Improvement in NASH?
Improvement in fibrosis stage
(45% of pts)
NASH resolution (90% of pts)
≥ 10% Weight Loss[3]
7% to 10% Weight Loss[2]
Improvement in NASH Activity Score
NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis.
5% Weight Loss[1]
Improvement in liver fat and liver stiffness
1. Patel NS, et al. Clin Gastroenterol Hepatol. 2017;15: Promrat K, et al. Hepatology. 2010;51: Vilar-Gomez E, et al. Gastronterol. 2015;149:

31. Summary: NASH and Weight Loss
Weight loss in obese pts leads to improvement in liver histology
Lifestyle modification leads to improvement in NASH
NASH, nonalcoholic steatohepatitis.

32. PIVENS Trial: Pioglitazone vs Vitamin E vs Placebo in NASH
Randomized, double-masked, double-dummy, placebo-controlled study
Pioglitazone 30 mg QD
Vitamin E placebo QD
Pts with NASH and no diabetes (N = 247)
Treatment until Wk 96, then follow-up for additional 24 wks
Vitamin E 800 IU QD
Pioglitazone placebo QD
Pioglitazone placebo QD Vitamin E placebo QD
NASH, nonalcoholic steatohepatitis.
Sanyal AJ, et al. N Engl J Med. 2010;362: Chalasani NP, et al. Contemp Clin Trials. 2009;30:88-96.

33. Pts With Improvement* (%)
PIVENS Primary Endpoint: Histologic Improvement With Vitamin E or Pioglitazone
P = .04
P = .001
NNT = 6.9
NNT = 4.2
Vitamin E
800 IU/day
Placebo
Pioglitazone
30 mg/day 83 80 84
100 60 40 20
n =
Pts With Improvement* (%) 43 19 34
NAS, nonalcoholic fatty liver disease activity score; NNT, number needed to treat.
*Histologic improvement: ≥ 1-point improvement in hepatocellular ballooning score, no increase in fibrosis score, and either a decrease in NAS to ≤ 3 or a ≤ 2-point decrease in NAS plus ≥ 1-point decrease in either the lobular inflammation or steatosis score.
Sanyal AJ, et al. N Engl J Med. 2010;362:

34. Pts With Resolved NASH (%)
PIVENS: Histologic Resolution of NASH at Wk 96 With Vitamin E or Pioglitazone 47
33/70 20 40 60 80
100 36 21
P = .05
n/N =
15/72
29/80
P = .001
Pts With Resolved NASH (%)
NASH, nonalcoholic steatohepatitis.
Vitamin E
800 IU/day
Placebo
Pioglitazone
30 mg/day
Sanyal AJ, et al. N Engl J Med. 2010;362:

35. Vitamin E: The Glass Is Half Full
Does vitamin E:
Improve NASH? Yes
Reverse NASH? Yes
Improve fibrosis? No (based on RCTs)
Improve long-term outcomes? No data
NASH, nonalcoholic steatohepatitis; RCT, randomized controlled trial.

36. Vitamin E: Risks Increases risk of bleeding in a dose-dependent manner
Especially ≥ 400 units daily[1]
Increases risk of prostate cancer in older men [2]
Increases risk of hemorrhagic stroke[3]
May be preventive in reducing the risk of ischemic stroke[3]
1. Miller ER III, et al. Ann Intern Med. 2005;142: Klein EA, et al. JAMA. 2011;306: Schürks M, et al. BMJ. 2010;341:c5702.

37. Pragmatic Approach for Using Vitamin E: Balancing Risks vs Benefits

38. Vitamin E: Treatment Considerations
Consider in nondiabetic pts with biopsy-proven NASH[1]
Not recommended in elevated ALT/AST with suspected NAFLD without a liver biopsy
Not recommended in mild NAFL with no evidence of NASH on biopsy
No efficacy data in:
Diabetes
Cirrhosis
Post liver transplantation
Risks may outweigh benefits in:
Older men
Uncontrolled hypertension (risk factors for hemorrhagic stroke)
Family history of prostate cancer
Personal history of stroke or prostate cancer
NAFL, nonalcoholic fatty liver; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis.
Chalasani N, et al. Hepatology. 2017;[Epub ahead of print].

39. Pioglitazone: Balancing Risks vs Benefits

40. Pioglitazone in Diabetes: Improvement or Resolution of NASH at 18 Mos
Randomized, placebo-controlled, double-blind clinical trial of pts with NASH and prediabetes or type 2 diabetes mellitus (N = 101)
Other pioglitazone outcomes:
Improved fibrosis score (treatment difference: -0.5; CI: -0.9 to 0; P = .039)
Greater weight gain (2.5 kg vs placebo)
Placebo (n = 51)
Pioglitazone (n = 50)
100
P < .001 80
P < .001 58 60 51
Pts With Improvement (%) 40
NAS, nonalcoholic fatty liver disease activity score; NASH, nonalcoholic steatohepatitis. 17 19 20
≥ 2-Point Reduction in NAS
(No Worsening of Fibrosis)
Resolution of NASH
Cusi K, et al. Ann Intern Med. 2016;165:

41. PIVENS: Change in Weight by Treatment
5.0
Pioglitazone
Vitamin E
Placebo
2.5
Change From Baseline (kg)
-2.5 24 48 72 96
120
Wks
Sanyal AJ, et al. N Engl J Med. 2010;362:

42. Pioglitazone in NASH: When and How
Biopsy-proven NASH with diabetes or prediabetes[1]
Monitor:
Body weight (and address with lifestyle interventions such as exercise and diet)
ALT and AST response
DEXA scan
DXA, dual-energy x-ray absorptiometry, NASH, nonalcoholic steatohepatitis.
1. Chalasani N, et al. Hepatology. 2017;[Epub ahead of print].

43. Conclusions: Current Management of NASH
NASH can lead to cirrhosis and HCC
Initial assessment
Natural history
Indications for liver biopsy
Metabolic traits
Current and future status of elastography-based noninvasive assessment
Innovations in clinical trial endpoints: MRI/MRE
Treatment of NASH
Lifestyle, vitamin E, pioglitazone
HCC, hepatocellular carcinoma; MRE, magnetic resonance elastography; NASH, nonalcoholic steatohepatitis.