Real-world evidence on non-invasive tests and associated cut-offs used to assess fibrosis in routine clinical practice

Background & Aims Non-invasive tests (NITs) offer a practical solution for advanced fibrosis identification in non-alcoholic fatty liver disease (NAFLD). Despite increasing implementation, their use is not standardised, which can lead to inconsistent interpretation and risk stratification. We aimed to assess the types of NITs and the corresponding cut-offs used in a range of healthcare settings. Methods A survey was distributed to a convenience sample of liver health experts who participated in a global NAFLD consensus statement. Respondents provided information on the NITs used in their clinic with the corresponding cut-offs and those used in established care pathways in their areas. Results There were 35 respondents from 24 countries, 89% of whom practised in tertiary level settings. A total of 14 different NITs were used, and each respondent reported using at least one (median = 3). Of the respondents, 80% reported using FIB-4 and liver stiffness by vibration-controlled transient elastography (Fibroscan®), followed by the NAFLD fibrosis score (49%). For FIB-4, 71% of respondents used a low cut-off of <1.3 (range <1.0 to <1.45) and 21% reported using age-specific cut-offs. For Fibroscan®, 21% of respondents used a single liver stiffness cut-off: 8 kPa in 50%, while the rest used 7.2 kPa, 7.8 kPa and 8.7 kPa. Among the 63% of respondents who used lower and upper liver stiffness cut-offs, there were variations in both values (<5 to <10 kPa and >7.5 to >20 kPa, respectively). Conclusions The cut-offs used for the same NITs for NAFLD risk stratification vary between clinicians. As cut-offs impact test performance, these findings underscore the heterogeneity in risk-assessment and support the importance of establishing consistent guidelines on the standardised use of NITs in NAFLD management. Lay summary Owing to the high prevalence of non-alcoholic fatty liver disease (NAFLD) in the general population it is important to identify those who have more advanced stages of liver fibrosis, so that they can be properly treated. Non-invasive tests (NITs) provide a practical way to assess fibrosis risk in patients. However, we found that the cut-offs used for the same NITs vary between clinicians. As cut-offs impact test performance, these findings highlight the importance of establishing consistent guidelines on the standardised use of NITs to optimise clinical management of NAFLD.

Background & Aims: Non-invasive tests (NITs) offer a practical solution for advanced fibrosis identification in non-alcoholic fatty liver disease (NAFLD). Despite increasing implementation, their use is not standardised, which can lead to inconsistent interpretation and risk stratification. We aimed to assess the types of NITs and the corresponding cut-offs used in a range of healthcare settings. Methods: A survey was distributed to a convenience sample of liver health experts who participated in a global NAFLD consensus statement. Respondents provided information on the NITs used in their clinic with the corresponding cut-offs and those used in established care pathways in their areas.
Results: There were 35 respondents from 24 countries, 89% of whom practised in tertiary level settings. A total of 14 different NITs were used, and each respondent reported using at least one (median = 3). Of the respondents, 80% reported using FIB-4 and liver stiffness by vibration-controlled transient elastography (Fibroscan ® ), followed by the NAFLD fibrosis score (49%). For FIB-4, 71% of respondents used a low cut-off of <1.3 (range <1.0 to <1.45) and 21% reported using age-specific cut-offs. For Fibroscan ® , 21% of respondents used a single liver stiffness cut-off: 8 kPa in 50%, while the rest used 7.2 kPa, 7.8 kPa and 8.7 kPa. Among the 63% of respondents who used lower and upper liver stiffness cut-offs, there were variations in both values (<5 to <10 kPa and >7.5 to >20 kPa, respectively).

Introduction
One of the enduring challenges of addressing the burden of nonalcoholic fatty liver disease (NAFLD) is ensuring that individuals with advanced stages of liver fibrosis are identified and provided with appropriate care by liver health specialists. 1 Non-invasive tests (NITs) provide a practical way to assess fibrosis risk in patients. 2 NITs fall within two broad categories: 1) serum biomarkers; and 2) liver stiffness measured by ultrasound or magnetic resonance-based elastography techniques. 3 Currently available NITs are most reliable for ruling out advanced stages of fibrosis (i.e., stages 3-4 on the Non-Alcoholic Steatohepatitis Clinical Research Network and Steatosis-Activity-Fibrosis staging systems). 4,5 The cut-offs employed have important implications for the sensitivity and specificity of NITs and the size of the indeterminate range. Generally, a low cut-off will improve the sensitivity and negative predictive value, and is therefore suited for ruling out advanced fibrosis, while a high cut-off will improve the specificity, positive predictive value and ability to rule it in. The most commonly used diagnostic outcome for risk stratification of individuals with NAFLD is advanced fibrosis (stages 3-4) due to its prognostic value. Due to the typically low prevalence of advanced fibrosis among populations with NAFLD, the negative predictive value of NITs is generally high, meaning that individuals with results below the cut-off can temporarily be excluded from further investigations, with a high degree of confidence. This is the goal in primary care, where the aim is to select individuals at risk of progressive liver disease for referral to specialist care, while ensuring that no individuals with the disease are missed. However, the positive predictive value of the tests is typically lower, meaning that single NITs are unable to provide a definitive diagnosis. 6 In contrast, in secondary and tertiary care, where the focus is on diagnostic confirmation, the higher cut-offs used generate a high specificity and high positive predictive value. 7 NITsused as stand-alone tests or in combinations (simultaneous or sequential)are increasingly used in primary and secondary care to identify individuals for referral to a liver specialist. In some settings, NITs have facilitated the development of formal care pathways that aim to efficiently and effectively link patients to care, especially those with advanced liver disease who require intervention from a hepatologist/liver specialist or multidisciplinary team. 8 In developing these pathways, decisions need to be made about which cut-offs should be used based on the clinical scenario.
While NITs are becoming more widely utilised as a means of identifying individuals with NAFLD and advanced fibrosis, little is known about the cut-offs being employed in clinical practice. Many reports include cut-offs for a specific study populationoften transposing cut-offs identified in other aetiologies of chronic liver diseaseleading to a range of published cut-offs. We hypothesise that the NITs used, and the corresponding cutoffs applied in practice, are widely heterogeneous between different healthcare settings and practices. In this brief report, we aim to explore the different NITs and corresponding cut-offs being used in routine clinical practice in a range of healthcare settings.

Materials and methods
In March 2021, a short survey (see supplementary material) was distributed to a convenience sample of 215 liver health experts who participated in a NAFLD consensus statement process in early 2021; 9 completed surveys were returned by August 2021. Comprised of three parts, the survey collected information on: the respondents' clinical setting, including the level of care (i.e., primary, secondary, tertiary), and the predominant patient population seen in the clinic; the NITs used in the clinic and those used in formal care pathways in the respondents' setting; and the cut-offs employed and the existence of national and subnational risk stratification pathways. We provide a descriptive analysis of the findings, including the variations in NIT cut-offs reported by respondents.

Results
A total of 35 survey responses were received. Most respondents (31/35; 89%) described their clinic as being in a tertiary level hospital setting, while the rest were from secondary level settings that all managed individuals with confirmed or suspected liver disease. Respondents were from a total of 24 countries. Most respondents were based in Europe (21/35; 60%), followed by East Asia and the Pacific (6/35; 17%). Two clinics were based in each of Latin America and the Caribbean and the Middle East and North Africa (6%) and one each in North America, sub-Saharan Africa, Central Asia and South Asia (3%). Across the 35 settings, 14 different NITs were used, with each respondent reporting the use of at least one NIT (Fig.1) (median = 3; range 1-8). Fibrosis-4 (FIB-4) and transient elastography (Fibroscan ® ) were the most used, reported by 28 of the 35 respondents (80%), followed by the NAFLD fibrosis score (NFS) (17/35; 49%).
For FIB-4, 71% of respondents (20/28) reported a low cut-off of <1.3. The lowest low cut-off used was <1.0, while the highest low cut-off used was <1.45. Six respondents (21%) reported agespecific cut-offs for FIB-4, with five (83%) employing a low cutoff of <1.3 for patients < − 65 and of <2.0 for those >65, as has previously been proposed. 10 Five respondents (18%) reported the use of a single FIB-4 cut-off, while the rest employed an upper cut-off, with an intermediate range between the upper and lower thresholds. Of these 23, 11 (48%) used an upper cut-off of >2.67 while 9 (39%) used >3.25.
NFS, the third most used NIT, had the least variation in cutoffs, with all 17 respondents using <1.455 as the low cut-off threshold. One respondent reported a single low cut-off and the remainder used a high cut-off, which ranged from >0.672 to >0.676, with the latter being reported by 10 respondents (59%).

Discussion
The findings reveal that the cut-offs employed for the same NITs vary between individual practitioners, especially for the high cut-offs, which aim at ruling in advanced fibrosis. The level of variation in cut-offs differs by NIT, with cut-offs varying much less for some tests, such as NFS, than others, including FIB-4 and Fibroscan ® . As lower and upper cut-offs have important implications for the sensitivity and specificity of the test, these findings can inform ongoing discussions around the benefits of using standardised cut-offs for specific settings and population groups.
Some of the variation identified may result from the different approaches employed across clinical settings. In some settings, clinicians are using the high negative predictive value of NITs as a means of ruling out advanced disease and, where necessary, these tests are followed by further investigations that can include a liver biopsy. Given the lower positive predictive value of NITs, in settings where clinicians use these to make a definitive diagnosis, they may be required to use a higher upper threshold to increase the certainty of the result. Most respondents in this study employed more than one NIT in their setting, and it is increasingly common that care pathways employ NITs simultaneously or sequentially. 11,12 Where care pathways use multiple sequential NITS, a lower cut-off to ruleout advanced fibrosis, used prior to a higher cut-off to rule it in, is justified. 13 Age can also influence the accuracy of NITs, with low specificity for advanced fibrosis in those <35 and low sensitivity in those >65, leading to calls for age-adjusted cut-offs, including for NFS and FIB-4. 10 In our small study sample, just over one in five clinics used age-adjusted cut-offs for FIB-4.
The liver health field needs to consider the reasons for the variation in NIT cut-offs and the clinical and public health implications that this entails. Focus should be on understanding where the variation arises because of new data and evidence, such as specific cut-offs in different population groups, and where it is the result of a lack of clear, uniform guidance and care pathways, as well as heterogeneity in available biomarkers. Qualitative research approaches would help to elucidate some of these reasons, including interviews with clinicians to understand the rationale behind their use of a particular NIT and cut-off.
NITs are a valuable tool to identify those with NAFLD who require specialist care. These findings show that the cut-offs being used in routine practice vary widely between the 35 respondents. There is a gap in the current literature on the implications of this variation and an urgent need for research to help guide efforts to better understand the implications of different cut-offs on patient outcomes and on health system resourcing. These discussions need to be integrated into broader discussions on advancing more efficient, patient-centred models of care for people living with NAFLD. 8 Regional and national liver disease associations and other norm setting bodies have a critical role to play in collating and analysing the latest data and incorporating these into clinical practice guidelines. 14,15 Respondents from 11 countries indicated that a national risk stratification pathway exists, yet for five of these counties respondents indicated that a pathway did not exist. This points to a definitional issue around what constitutes a national pathway, and the need for a clearer definition of this.
Our study has several strengths and limitations. To our knowledge, this is the first study to present the variation of cutoffs among NITs used in routine clinical practice. These heterogeneous pilot data highlight the need for larger and more detailed studies of this kind, which should also include the corresponding actions based on the test results. Assessment of the current practice landscape is the first step toward standardisation of cut-offs. Our data come from an opportunistic sample of clinicians and researchers engaged in a previous study, 9 and while there were primary care clinicians in this sample, we received no response from this setting. While the small number of respondents is a limitation, this study does establish that there is a lack of uniform use of NIT cut-off values. Future studies of this kind should be larger and aim to include respondents from primary care, as that is a critical first-line setting for identifying advanced liver disease. 16 The variability in NIT type and cut-offs used in primary care may be even higher than the one seen among hepatology specialists, given lesser familiarity with the intricacies of biomarkers and the lack of standardisation by guidance documents. 17 These findings demonstrate that cut-offs used for the same NITs vary between clinicians. As lower and upper cutoffs have important implications for the sensitivity and specificity of the test, i.e. ruling advanced fibrosis in or out, these findings can inform ongoing discussions on the benefits of implementing standardised setting-and population-specific cut-offs, and the revision of current testing guidelines.

Financial support
This study did not receive any funding.