One of the unmet need in patients with metabolic risk factors is the availability of noninvasive tests (NITs) for the prediction of nonalcoholic steatohepatitis (NASH) and metabolic liver disease (MLD) combining steatosis and the features of necroinflammatory activity 1–4. The nonalcoholic fatty liver disease (NAFLD) activity score, developed by the Nonalcoholic Steatohepatitis Clinical Research Network (NASH-CRN) 4 and used in the Fatty Liver Inhibition of Progression (FLIP) algorithm 5,6, is the standard reference for the diagnosis of NASH 4 (see Supplementary Table, Supplemental digital content 2, http://links.lww.com/EJGH/A259).
This histological reference has several limitations that have been discussed in the literature 7–10, including sampling error 7 and interobserver variations among pathologists 9. One limitation not fully analyzed, when NITs are validated using this highly specific reference, is the risk of false negatives. Recently, we have shown in 1081 cases at risk of MLD that a simpler definition of NASH using only a Steatosis–Activity–Fibrosis (SAF)-Activity-grade≥2 that does not require the presence of steatosis and the presence of both lobular inflammation and ballooning enabled the construction of NITs with fewer false-negatives cases 11. An example is patient with type 2 diabetes, no other cause of liver disease, with significant histological fibrosis but with steatosis less than 5%, and a grade 2 lobular inflammation without ballooning. Because of sampling error of biopsy and the temporal variability of MLD features, it seems appropriate to consider this case as a NASH.
The aim of this study was to construct a new quantitative test for the diagnosis of NASH using this simplified histological definition as the reference.
Patients and methods
All clinical investigations were performed according to the principles of the Declaration of Helsinki. All authors had access to the study data and reviewed and approved the final manuscript.
We followed FibroSTARD recommendations for the identification of new tests in presumed liver fibrosis that were also transposed for presumed necroinflammatory activity in patients at risk of MLD 12 (see Supplementary Table, Supplemental digital content 3, http://links.lww.com/EJGH/A259). The primary aim was to construct a quantitative test for NASH. The secondary aim was to use this test to assess the prevalence of NASH grades in two large populations: one with low risk and one with high risk of MLD.
For the primary aim, the population (Population-1) included patients from two prospective projects: the FibroFrance project (USA-NCT01927133) and the FLIP Consortium (http://www.flip-fp7.eu/) (Fig. 1). Inclusion criteria and the characteristics of these patients have been detailed elsewhere 13. In summary, all cases were at risk of MLD, without healthy volunteers, without primary selection according to previous activity or fibrosis tests results, including discordances. All patients were consecutively included if they had a biopsy scored according to the SAF scoring system and a contemporaneous blood test. Data collection was planned before the index sample, which was already used for the validation of FibroTest, ActiTest, and SteatoTest, but not for a quantitative NASH test. Patients included were not different from the nonincluded cases 13.
For the secondary aims, we included two large populations without biopsy but with contemporaneous NITs (SteatoTest, ActiTest, FibroTest): one (Population-2) including healthy volunteers at low risk of MLD representative of the French general population older than 40 years of age 14 and one (Population-3) including American high-risk patients with MLD who required the NASH-FibroSure.
The SAF activity scoring system was considered the simplified histological reference for nonalcoholic steatohepatitis (H-NASHs) without the requirements used for NASH-CRN and FLIP algorithm definition (see Supplementary Table, Supplemental digital content 4, http://links.lww.com/EJGH/A259). The histological references for significant MLD were those defined by the SAF scoring system, Fibrosis stage≥2 and Activity grade≥2 5,6. The goal of the SAF score was to find a compromise between the development of a simple, easily applied system to make a firm diagnosis in individual patients, even when applied by nonspecialists, and of a more reliable and discriminating system for therapeutic trials or for the assessment of biomarker diagnostic performance. A FLIP Histopathology Consortium of eight members developed the FLIP algorithm, a diagnostic tool for the diagnosis and staging of severe forms of NAFLD according to the combination of each semi-quantification of the three elementary features of NAFLD using the SAF score for steatosis, inflammatory activity and fibrosis, respectively. The steatosis score (S) assesses the quantities of large-sized or medium-sized lipid droplets, with the exception of foamy microvesicles, and rates them from 0 to 3 (S0: <5%; S1: 5–33%, mild; S2: 34–66%, moderate; S3: >66%, marked). Activity grade (A, from 0 to 4) is the unweighted addition of hepatocyte ballooning (0–2) and lobular inflammation (0–2). Patients with A0 (A=0) had no activity; patients with A1 (A=1) had mild activity; patients with A2 (A=2) had moderate activity; patients with A3 (A=3) had severe activity; and patients with A4 (A=4) had very severe activity. Fibrosis stage (F) was assessed using the score described as follows: stage 0 (F0)=none; stage 1 (F1)=1a or 1b perisinusoidal zone 3 or 1c portal fibrosis; stage 2 (F2)=perisinusoidal and periportal fibrosis without bridging; stage 3 (F3)=bridging fibrosis; and stage 4 (F4)=cirrhosis (see Supplementary File, Supplemental digital content 1, http://links.lww.com/EJGH/A259). To reduce interobserver variability and homogenize the reading using the SAF-FLIP histological classification, we used only reports reviewed by members of the FLIP Pathology Consortium (D.T. and P.B. for the FLIP subpopulation and F.C. for the FibroFrance subpopulation).
The FibroTest, ActiTest, and SteatoTest are patented NITs (BioPredictive, Paris, France) that have been validated extensively to assess the stages of fibrosis, activity using the METAVIR 15–17 and the SAF scoring system including the grades of steatosis 13,18. These NITs are already used in patients at risk of MLD in 40 countries including USA branded as NASH-FibroSure (LabCorp., Burlington, North Carolina, USA). The FibroTest includes serum α2-macroglobulin, apolipoprotein-A1, haptoglobin, total bilirubin, and γ-glutamyl transpeptidase. The ActiTest includes the same components plus alanine aminotransferase. The SteatoTest includes the same six components of the FibroTest and ActiTest plus BMI, serum cholesterol, triglycerides, and fasting glucose. The preanalytical analytical procedures were those recommended by BioPredictive. Components were assessed on fresh samples for FibroFrance patients of Population-1, Population-2, and Population-3. For FLIP patients of Population-1 the serum stored at −80°C for non-French, were sent to the Paris reference center. Exclusion criteria were nonreliable results identified using security control algorithms 19.
The NIT-NASHs test was developed for the quantitative diagnosis of NASH using the 11 components of the SteatoTest without glucose and BMI. The H-NASHs was taken as the histological reference of the logistic regression analysis. We compared the performance of the new NIT-NASHs with ActiTest extensively validated as a NIT for necroinflammatory histological activity in chronic viral hepatitis C 17, and with three other nonpatented NITs often used in MLD for the prediction of NASH and fibrosis: the NAFLD score, the BARD, and the FIB4 index. We considered ActiTest as the main comparator as we found previously that it performed better than these three nonpatented tests for the diagnosis of activity in MLD 13. In the training population, we a-priori divided the NIT-NASHs test, which ranged from 0.00 to 1.00, into four grades, from 0 to 0.25 grade N0 (no NASH), from 0.25 to 0.50 grade N1 (minimal NASH), from 0.50 to 0.75 grade N2 (moderate NASH), and above 0.75 grade N3 (severe NASH).
Using both the FibroTest for fibrosis and NIT-NASHs for activity, it was possible to predict by these NITs the presence or the absence of a clinically significant MLD as defined by the histological SAF scoring system, fibrosis stage at least F2, or activity grade at least grade A2 5,6; A2 was defined as NIT-NASHs of at least 0.50 (the median of the logistic regression) and F2 as FibroTest more than 0.48, the validated standard cutoff for stage F2–F3–F4 13.
The new NIT was constructed and validated according to FibroSTARD standards, adapted to necroin-flammatory activity NITs 12,13. NCSS-2013 statistical software was used 20. All the NITs components were assessed without knowing biopsy results. The study was carried out on a per-protocol basis after exclusion of less than 1% of NITs not interpretable according to the components outliers as described previously 19.
A total of 1081 patients with biopsies and NITs were included: 1081 with metabolic risk factors (Population-1), including 1009 with MLD, and 72 controls without inflammation and steatosis (A0–S0). A total of 94 843 patients with NITs and without biopsy were included: 7416 volunteers at low risk of MLD (Population-2) and 87 427 (Population-3) with presumed metabolic factors at high risk of MLD who required NITs (Fig. 1). Characteristics of these populations are shown in Table 1 for Population-1, without significant differences between the 541 patients randomized in the training group 1 and the 540 patients of the validation group. Among the participants, there were both men and women, median age around 48 years, one-third with diabetes or fasting glucose more than 6.1, 70% with BMI of at least 30 kg/m2, one-third with significant fibrosis (stages 2–4), and 50% with significant activity.
Prevalence of histological metabolic liver diseases in Population-1
The prevalence of NASH using the standard CRN-definition was 50.8% (47.8–53.8) (549/1081), that is, 39 less cases than that with the simplified definition, 54.4% (51.4–57.4) (588/1081). These 39 NASH cases (3.6%; 2.6–4.9) that were missed by the FLIP algorithm (which required both ballooning and lobular inflammation) included 15 cases with significant fibrosis (six F2, five F3, and four cirrhosis).
Construction of noninvasive tests in the training group
The logistic regression analysis using the simplified NASHs definition as a histological reference enabled the construction of the NIT-NASHs, which combined 11 components including α2-macroglobulin, haptoglobin, apolipoprotein-A1, alanine aminotransferase, aspartate aminotransferase, γ-glutamyl transpeptidase, total bilirubin, cholesterol, triglycerides, age, and sex (patent pending).
For the diagnosis of NASHs, the NIT-NASHs had a significant area under the receiver operating characteristic curve (AUROC) of 0.773 (0.730–0.810; P<0.0001vs. 0.500). This new test had a significantly higher AUROC than that of the ActiTest (0.718; 0.689–0.773; P=0.01) (Fig. 2a).
For the diagnosis of significant MLD, NIT-A2orF2, the combination of NASHs and FibroTest by regression analysis (patent pending) had an AUROC of 0.800 (0.759–0.835; P<0.0001 vs. 0.500), which was higher (P<0.0001) than that of the ActiTest (0.735; 0.689–0.774; P<0.0001) and the FibroTest (0.746; 0.701–0.785; P<0.0001), but not different from that of NASHs (P=0.36) (Fig. 3a).
Internal validation of noninvasive tests
The AUROC of NIT-NASHs in the control group was 0.814 (0.774–0.847) (Fig. 2b), not different from the AUROC observed in the training group, 0.773 (0.730–0.810; P=0.60). As observed in the training group, the AUROC of NIT-NASHs was also higher than the ActiTest AUROC (0.778; 0.735–0.814; P=0.01).
The AUROC of NIT-A2orF2 was 0.833 (0.794–0.865), not different from the AUROC observed in the training group (P=0.60). As observed in the training group, The AUROC of NIT-A2orF2 was also higher than that of the ActiTest AUROC (0.789; 0.747–0.825; P=0.003) and the FibroTest AUROC (0.776; 0.734–0.813; P=0.0001), but not different from that of NASHs (P=0.49) (Fig. 3b).
Comparison with nonpatented noninvasive tests in Population-1
A total of 574 patients had simultaneous measurements of NIT-NASHs, NIT-A2orF2, and nonpatented NAFLD score, BARD, and FIB4 index.
NIT-NASHs had a significantly higher AUROC (0.671; 0.614–0.721), than NAFLD score AUROC (0.570; 0.510–0.626; P=0.006), FIB4 (0.528; 0.467–0.584; P=0.0003), and the BARD index (0.541; 0.476–0.599; P=0.003) (see Supplementary Fig., Supplemental digital content 5, http://links.lww.com/EJGH/A259).
NIT-A2orF2 had a significantly higher AUROC (0.671; 0.613–0.721) than the NAFLD score (0.570; 0.510–0.626; P=0.006), FIB4 (0.528; 0.467–0.584; P=0.0003), and BARD (0.541; 0.476–0.599; P=0.003) (see Supplementary Fig., Supplemental digital content 6, http://links.lww.com/EJGH/A259). Patients with and without nonpatented tests were different because most of the missing data were from cases of the obese cohort with less severe MLD, including more patients who were women, younger, and with biopsies performed during bariatric surgery (see Supplementary Table, Supplemental digital content 7, http://links.lww.com/EJGH/A259).
Application of the new noninvasive tests for assessing the prevalence of metabolic liver disease in larger populations
In the French population, the presumed prevalence of H-NASH, grade moderate or severe (NIT-NASHs>0.50 grade N2 and N3), was 31.8% (30.8–32.9; 2358/7416) for H-NASHs and 32.1% (31.0–33.1; 2378/7416) for A2orF2. In the US high-risk population, the presumed prevalences were 64.2% (63.9–64.5; 57 416/89 427) for H-NASHs and 65.1% (64.8–65.4; 58 202/89 427) for A2orF2 (Table 2).
In both populations, the presumed prevalences of H-NASHs and A2orF2 were associated strongly with diabetes, defined by fasting glucose of at least 7 mmol/l. Irrespective of the NITs, the prevalence of MLD in metabolic risk patients was always much higher in the US population selected as high-risk cases than in the French population representative of the general population. However, these differences were markedly reduced when both BMI of at least 30 kg/m2 and fasting glucose of at least 7.0 mmol/l were taken into account, with 78.1% (77.4–78.8; 9784/12 528) of H-NASHs in the USA and 71.4% (61.4–80.1; 70/98) in the French population, respectively (Table 3).
In our sample of the French general population, the prevalence of N2–N3 (31.8%; 30.8–32.9) (2358/7416) was much higher in comparison with what we observed using ActiTest and the METAVIR A2–A3 grades defining inflammation (1.1%; 0.8–1.3) (78/7416). However, in this population (at least 40 years of age), there was also a high prevalence of alcohol, metabolic, or both risk factors. Only 34.5% (2562/7416) had no ‘overt’ alcohol or metabolic risks factors, and in this group, the prevalence of N2–N3 decreased to 18.9% (17.4–20.4) (Table 3).
In this subgroup of 2562 apparently healthy individuals, and without overt type 2 diabetes, dyslipidemia, morbid obesity, or arterial hypertension, we observed that the 483 patients with presumed N3–N4 had, both in the univariate (median; 95% confidence interval) and in the multivariate logistic regression analysis [odds ratio (95% confidence interval)], higher BMI and higher fasting glucose than the 2071 patients presumed N0–N1 (all P=0.0001): 24.5 (24.1–24.8) versus 23.4 (23.2–23.6), 5.27 (5.22–5.33) versus 5.16 (5.11–5.16) with odds ratio of 1.10 (1.06–1.14) and 1.59 (1.25–2.02), respectively. The %carbohydrate deficient transferrin was available in 59 patients with N3–N2 versus 221 patients with N0–N1, without a significant difference, 1.33 (1.26–1.38) versus 1.34 (1.29–1.37; P=0.99).
Our results show that a simpler definition of NASH that does not require the presence of steatosis and the presence of both lobular inflammation and ballooning enabled the construction of a new quantitative NIT with better performances in comparison with standard NITs already in use in MLD for the diagnosis of NASH.
Strengths of the study
First, this study suggests that although the existing and arbitrary definitions of MLD may be appropriate for a pathologist’s ‘context of use’, they are not appropriate for the clinician. The CRN-definition of NASH is based on two studies including 2.3 and 0.3%, respectively, of cases without steatosis or inflammation, who are the appropriate controls for the valid evaluation of the specificity of the definition of NASH (see Supplementary Table, Supplemental digital content 8, http://links.lww.com/EJGH/A259) 11. The actual prevalence of MLD among patients with metabolic risk factors was unknown in these studies. We therefore suggest that ‘chronic carriers of metabolic risk factors in the absence of other known liver disease’ would be a more appropriate definition for use by clinicians, which does not exclude ‘a priori’ any cases without MLD, defined as steatosis (S), necroinflammatory activity, and fibrosis (SAF), as ranked by severity. This should be the appropriate population of interest for the construction of NITs.
Second, we have used the NASHs definition as the histological reference for NASH, which had the lowest risk of false positive/negatives out of 96 possible combinations between the histological definition, the MLD definition, and NITs cutoffs, with the highest concordance rate with validated NITs 11. Here, the simplified definition of NASH (H-NASHs) identified 39 (3.6%) high-risk cases that were missed by the FLIP algorithm including 15 cases with significant fibrosis (six F2, five F3 and four cirrhosis). Moreover, we showed that the H-NASHs definition covered 18 (67%) out of 27 combinations of possible features versus four (15%) for current CRN-FLIP-definitions, which reduced the risk of a false-negative diagnosis (see Supplementary Table, Supplemental digital content 4, http://links.lww.com/EJGH/A259).
Our results also suggest that a highly sensitive cutoff should be chosen to predict a grade of significant activity, such as more than 0.17 for the ActiTest for SAF-Activity-grade≥A2, compared with more than 0.52 for METAVIR A2 in CHC (see Supplementary Table, Supplemental digital content 2, http://links.lww.com/EJGH/A259). This seems rational as grade-1 ballooning and lobular inflammation are less severe than the METAVIR-grade-2 requiring moderate necrosis and inflammation. In larger populations with a risk of MLD, the prevalence of SAF-Activity-grade≥A2 presumed by the ActiTest varied from 0.8 to 3.3% and from 28.7 to 71.4% with the NIT-NASHs (Table 3). We constructed the NIT-NASHs as a sensitive NIT for H-NASHs providing a high negative predictive value. This should reassure patients with NIT-NASHs less than 0.50 in whom the risk of H-NASHs is very low. Furthermore, the NIT-NASHs test did not include fasting glucose and BMI among its components, which simplifies assessments and enables its use for assessing the impact of NASH on fibrosis independent of diabetes and obesity.
Finally, this study shows that the performance of the new NITs constructed using this simpler histological definition was better for the prediction of NASH than validated NITs such as the ActiTest or the NAFLD score. Although these are preliminary results that must be validated independently in an external population, the diagnostic value of these new NITs constructed with H-NASHs seemed better than standard NITs, such as the NAFLD score or FIB4. Our two preliminary applications of these tests suggest that the new NITs improved screening of A2orF2 in large populations with a metabolic risk or even in the general population. As a proof of concept, we successfully assessed the increase of presumed H-NASHs prevalence and significant activity and fibrosis in relation to fasting glucose and obesity. This was possible because the new NITs did not include fasting glucose or BMI in their components.
Further studies are needed to confirm whether the high sensitivity of NIT-NASHs is useful for the prediction of the progression of fibrosis to assess the efficacy of future NASH treatment independent of fibrosis biomarkers. Although this study was limited by its retrospective design, we included a large number of biopsies that were read using the same scoring system in prospectively included participants, and all biopsy specimens were read by experienced pathologists familiar with the SAF scoring system. Despite the absence of an external validation population, both the training and the control groups included more than 500 cases, without significant differences between their main characteristics.
Another obvious limitation of the population in our study was the small sample size of S0–A0 controls (51 with S0% and 39 with S1–S4%) from the FLIP population with a metabolic risk. However, these cases represent the highest number of A0–S0 controls ever analyzed. In a separate study 11, we additionally included 113 supplementary S0–A0 controls who were mostly patients with virologically cured CHC, which showed the significant impact of adding controls, adjusted for the other factors of variability, for the construction of more specific NITs.
Comparisons with patented tests FibroTest and ActiTest were performed as scheduled on all cases, in the training and the control population, but the comparison of the new NITs with standard NITs was limited because only half of the cases had available nonpatented tests. The patients with missing-NITS were significantly different, with more obese women and younger patients with less significant liver disease than diabetics or hyperlipidemic cases, which mathematically induced a spectrum effect and reduced the AUROCs as for fibrosis-NITs 21,22, but the head to head comparisons were still highly significant in favor of NIT-NASHs (see Supplementary Fig., Supplemental digital content 6, http://links.lww.com/EJGH/A259).
Our integrated database was limited by median biopsy lengths ranging from 15 to 22 mm and because most of the biopsies in obese cases were performed during bariatric surgery, which may be less appropriate than intercostal liver biopsy if not using needle biopsy 23.
We acknowledge that despite the highly significant AUROCs, the performances of these new NITs (around 0.800) were poorer than that of the standard almost perfect reference, which is between 0.900 and 1.000. However, both the limitations of the biopsy and the spectrum effect could explain these results. In the context of liver disease, even for fibrosis, which is easier to define than inflammatory activity, there is no perfect reference, the main reason being the sampling error of the biopsy, with intra patient variability of the SAF features 7. Using large surgical biopsies in chronic hepatitis C (CHC), the AUROC of 20 mm biopsy was around 0.80 for the diagnostic between F1 and F2 stages 24,25. Therefore, the maximum of a perfect test cannot be above 0.90 26. Furthermore, the most important comparison is between the new test versus the reference test in the same patients and controls. The crude value of an AUROC is not interpretable because of the risk of the spectrum effect. This was clear from the AUROC range of 0.65–0.89 of the same test (FibroTest) in the same disease (hepatitis C) according to the prevalence of fibrosis stages 27. The AUROC of NASH tests must be performed in the real context of use and not in populations with only 10% of non-NASH, the less severespectrum reducing automatically the AUROC 11.
This study focused on NITs developed by several coauthors of the article who have an obvious conflict of interest. However, the other coauthors were totally independent, and they recruited the patients and performed the assay independent of the company and with full access to all data and analyses.
Despite the sample size of our training and internal validation populations, an external validation of these new NITs is necessary.
The requirement of steatosis and both ballooning and lobular inflammation may be appropriate for a pathologist to specifically define the paradigm of NASH, and there is a rational to use them as a possible surrogate marker of severe MLD. However, there is no evidence that activity is a better surrogate marker of clinical severity than the progression of fibrosis, as observed in chronic viral hepatitis, independent of steatosis and activity. This notion was supported recently by studies showing that ‘steatosis-only’ can progress to NASH and to significant fibrosis 28,29. Histologically, steatosis progressed by grade≥1 in 19% of patients and regressed by grade≥1 in 33% of patients, and ballooning progressed by grade≥1 in 37% of patients and regressed by the same degree in 16% of patients. The sample population in these studies was small and also raised the question of the natural time-dependent variability of steatosis or inflammation in the same patient. This is another reason to simplify the definitions of MLD and exclude a dependence between SAF features.
The rational of 5% of hepatocytes, excluding minimal steatosis (1–4%) for the definition of steatosis, seems arbitrary 10 as minimal activity grade and minimal fibrosis stage are already accepted as features of MLD in the current definitions. In our study, only 39 (3.6%) of presumed NAFLD patients had 1–4% steatosis including only one case with significant activity (NASH), one case of cirrhosis, and 10 F1 (Supplementary Table, Supplemental digital content 4, http://links.lww.com/EJGH/A259). Therefore, it is possible that in larger populations, the true prevalence of NASH cases with 1–4% steatosis would be much higher as the NASH prevalence predicted by NITs was 64.2% in the US population requiring Nash-FibroSure (Table 3).
These results suggested that this new test enables a quantitative assessment of NASH, and when associated with the FibroTest, identifies cases with clinically significant MLD, either inflammation or fibrosis. Despite the large sample size, an independent external validation is needed.
The members of the FLIP Partners’ Consortium, the FLIP Pathology Consortium, the FibroFrance-CPAM Group, and the FibroFrance-Obese Group (see Supplementary File, Supplemental digital content 1, http://links.lww.com/EJGH/A259).
Financial support was provided by European Community’s Seventh Framework Program (FP7/2007-2013) grant agreement HEALTH-F2-2009-241762 for the project FLIP.
Conflicts of interest
T.P. is the inventor of FibroTest/SteatoTest and the founder of BioPredictive, the company that markets these tests. Patents belong to the French Public Organization Assistance Publique-Hôpitaux de Paris. M.M., Y.N., and O.D. are BioPredictive employees. BioPredictive did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript, and only provided financial support in the form of authors’ salaries. For the remaining authors there are no conflicts of interest.
1. Sanyal AJ, Brunt EM, Kleiner DE, Kowdley KV, Chalasani N, Lavine JE, et al. Endpoints and clinical trial design for nonalcoholic steatohepatitis
. Hepatology 2011; 54:344–353.
2. Castera L, Vilgrain V, Angulo P. Noninvasive evaluation of NAFLD. Nat Rev Gastroenterol Hepatol 2013; 10:666–675.
3. European Association for the Study of the Liver (EASL); European Association for the Study of Diabetes (EASD); European Association for the Study of Obesity (EASO). EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease. J Hepatol 2016; 64:1388–1402.
4. Kleiner DE, Brunt EM, van Natta M, Behling C, Contos MJ, Cummings OW, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease
. Hepatology 2005; 41:1313–1321.
5. Bedossa P, Poitou C, Veyrie N, Bouillot JL, Basdevant A, Paradis V, et al. Histopathological algorithm and scoring system for evaluation of liver lesions in morbidly obese patients. Hepatology 2012; 56:1751–1759.
6. Bedossa P. FLIP Pathology Consortium. Utility and appropriateness of the fatty liver inhibition of progression (FLIP) algorithm and steatosis, activity, and fibrosis (SAF) score in the evaluation of biopsies of nonalcoholic fatty liver disease
. Hepatology 2014; 60:565–575.
7. Ratziu V, Charlotte F, Heurtier A, Gombert S, Giral P, Bruckert E, et al. Sampling variability of liver biopsy in nonalcoholic fatty liver disease
. Gastroenterology 2005; 128:1898–1906.
8. Younossi ZM, Gramlich T, Liu YC, Matteoni C, Petrelli M, Goldblum J, et al. Nonalcoholic fatty liver disease
: assessment of variability in pathologic interpretations. Mod Pathol 1998; 11:560–565.
9. Brunt EM, Kleiner DE, Wilson LA, Belt P. Neuschwander-Tetri BA; NASH Clinical Research Network
(CRN). Nonalcoholic fatty liver disease
(NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology 2011; 53:810–820.
10. Brunt EM. Nonalcoholic fatty liver disease
: pros and cons of histologic systems of evaluation. Int J Mol Sci 2016; 17:97.
11. Poynard T, Munteanu M, Charlotte F, Perazzo H, Ngo Y, Deckmyn O, et al. Impact of metabolic liver disease
definitions on the performance of noninvasive biomarkers of NASH
. Eur J Gastroenterol Hepatol 2017. [in press].
12. Boursier J, de Ledinghen V, Poynard T, Guéchot J, Carrat F, Leroy V, et al. An extension of STARD statements for reporting diagnostic accuracy studies on liver fibrosis tests: The Liver-FibroSTARD standards. J Hepatol 2015; 62:807–815.
13. Munteanu M, Tiniakos D, Anstee Q, Charlotte F, Marchesini G, Bugianesi E, et al. Diagnostic performance of FibroTest
, SteatoTest, and ActiTest
in patients with NAFLD using the SAF-score as histological reference. Aliment Pharmacol Ther 2016; 44:877–889.
14. Poynard T, Lebray P, Ingiliz P, Varaut A, Varsat B, Ngo Y, et al. Prevalence of liver fibrosis and risk factors in a general population using non-invasive biomarkers. BMC Gastroenterol 2010; 10:40.
15. Houot M, Ngo Y, Munteanu M, Marque S, Poynard T. Systematic review with meta-analysis: direct comparisons of biomarkers for the diagnosis of fibrosis in chronic hepatitis C and B. Aliment Pharmacol Ther 2016; 43:16–29.
16. Chou R, Wasson N. Blood tests to diagnose fibrosis or cirrhosis in patients with chronic hepatitis C virus infection: a systematic review. Ann Intern Med 2013; 158:807–820.
17. Poynard T, Munteanu M, Ngo Y, Castera L, Halfon P, Ratziu V, et al. ActiTest
accuracy for the assessment of histological activity grades in patients with chronic hepatitis C, an overview using Obuchowski measure. Gastroenterol Clin Biol 2010; 34:388–396.
18. Poynard T, Ratziu V, Naveau S, Thabut D, Charlotte F, Messous D, et al. The diagnostic value of biomarkers (SteatoTest) for the prediction of liver steatosis. Comp Hepatol 2005; 4:10.
19. Poynard T, Munteanu M, Deckmyn O, Ngo Y, Drane F, Messous D, et al. Applicability and precautions of use of liver injury biomarker FT. A reappraisal at 7 years of age. BMC Gastroenterol 2011; 11:39.
20. Hintze J. NCSS 12. Kaysville, UT: NCSS LLC. Available at: http://www.ncss.com
. [Accessed 15 December 2017].
21. Poynard T, Halfon P, Castera L, Munteanu M, Imbert-Bismut F, Ratziu V, et al. Standardization of ROC curve areas for diagnostic evaluation of liver fibrosis markers based on prevalences of fibrosis stages. Clin Chem 2007; 53:1615–1622.
22. Poynard T, Lassailly G, Diaz E, Clement K, Caïazzo R, Tordjman J, et al. Performance of biomarkers FibroTest
, SteatoTest, and NASHTest in patients with severe obesity: meta analysis of individual patient data. PLoS One 2012; 7:e30325.
23. Padoin AV, Mottin CC, Moretto M, Berleze D, Kupski C, Glock L, et al. A comparison of wedge and needle hepatic biopsy in open bariatric surgery. Obes Surg 2006; 16:178–182.
24. Bedossa P, Dargère D, Paradis V. Sampling variability of liver fibrosis in chronic hepatitis C. Hepatology 2003; 38:1449–1457.
25. Poynard T, Lenaour G, Vaillant JC, Capron F, Munteanu M, Eyraud D, et al. Liver biopsy analysis has a low level of performance for diagnosis of intermediate stages of fibrosis. Clin Gastroenterol Hepatol 2012; 10:657–663.
26. Mehta SH, Lau B, Afdhal NH, Thomas DL. Exceeding the limits of liver histology markers. J Hepatol 2009; 50:36–41.
27. Poynard T, Halfon P, Castera L, Munteanu M, Imbert-Bismut F, Ratziu V, et al. Standardization of ROC curve areas for diagnostic evaluation of liver fibrosis markers based on prevalences of fibrosis stages. Clin Chem 2007; 53:1615–1622.
28. Pais R, Charlotte F, Fedchuk L, Bedossa P, Lebray P, Poynard T, et al. A systematic review of follow-up biopsies reveals disease progression in patients with non-alcoholic fatty liver. J Hepatol 2013; 59:550–556.
29. McPherson S, Hardy T, Henderson E, Burt AD, Day CP, Anstee QM. Evidence of NAFLD progression from steatosis to fibrosing-steatohepatitis using paired biopsies: implications for prognosis and clinical management. J Hepatol 2015; 62:1148–1155.