Laboratory Section - Combination of Serum Markers for Liver Fibrosis in the Evaluation and Monitoring of Patients with Chronic Liver Disease

IMPORTANT REMINDER

This Medical Policy has been developed through consideration of medical necessity, generally accepted standards of medical practice, and review of medical literature and government approval status.

Benefit determinations should be based in all cases on the applicable contract language. To the extent there are any conflicts between these guidelines and the contract language, the contract language will control.

The purpose of medical policy is to provide a guide to coverage. Medical Policy is not intended to dictate to providers how to practice medicine. Providers are expected to exercise their medical judgment in providing the most appropriate care.

Description

The stage of fibrosis is the most important single predictor of significant morbidity and mortality in patients with hepatitis C. Liver biopsy is typically recommended prior to the initiation of antiviral therapy, and repeat biopsies may be performed to monitor fibrosis progression. Liver biopsies are analyzed according to a histologic scoring system; the most commonly used one is the METAVIR scoring system, which scores fibrosis from F0-F4. A METAVIR score of F2 to F4 signifies significant fibrosis, while a score of F3 and F4 signifies advances fibrosis. Biopsies can also be evaluated according to the degree of inflammation presented, referred to as the grade or activity level. For example, the METAVIR systems includes scores for necroinflammatory activity ranging from A0 to A3 (A0= no activity, A1 = minimal activity, A2 = moderate activity, A3 = severe activity.) However, several limitations to liver biopsy are noted, including its invasive nature, small sample size, and subjective grading system. Regarding small sample size, liver fibrosis can be patchy and thus missed on a biopsy sample, which includes only 1:50,000 of the liver tissue.

A noninvasive alternative to liver biopsy would be particularly helpful, both to initially assess patients and then as a monitoring tool to assess response to therapy. A variety of laboratory tests have been proposed as an alternative to liver biopsy. Laboratory tests can be broadly categorized into indirect and direct markers of liver fibrosis. Indirect markers include liver function tests such as ALT (alanine aminotransferase), AST (aspartate aminotransferase), the ALT/AST ratio (also referred to as the AAR), platelet count and prothrombin index. In recent years there has been growing understanding of the underlying pathophysiology of fibrosis, leading to direct measurement of the factors involved. For example, the central event in the pathophysiology of fibrosis is activation of the hepatic stellate cell. Normally, the stellate cells are quiescent, but are activated in the setting of liver injury, producing a variety of extracellular matrix (ECM) proteins. In normal livers, the rate of ECM production equals its degradation, but in the setting of fibrosis, production exceeds degradation. Metalloproteinases are involved in intracellular degradation of ECM, and a profibrogenic state exists when there is either a down regulation of metalloproteinases or an increase in tissue inhibitors of metalloproteinases (TIMP). Both metalloproteinases and TIMP can be measured in the serum, which directly reflect fibrotic activity. Other direct measures of ECM deposition include hyaluronic acid or alpha-2 macroglobulin.

While there have been many studies of these individual markers or groups of markers in different populations of patients with liver disease, recently, there has been interest in analyzing multiple markers using proprietary algorithms to generate a score that categorizes patients according to the METAVIR score. It is proposed that using these algorithms can be used as an alternative to liver biopsy in patients with liver disease, particularly hepatitis C.

HCV FIBROSURE™ uses a combination of six serum biochemical indirect markers of liver function plus age and gender in a patented algorithm to generate a measure of fibrosis and necroinflammatory activity in the liver that corresponds to the METAVIR scoring system for stage (fibrosis) and grade (necroinflammatory activity). The biochemical markers include the readily available measurements of alpha-2 macroglobulin, haptoglobin, gamma glutamyl transpeptidase (GGT), ALT, and apolipoprotein A1. Developed in France, the test has been clinically available in Europe for the past two years and in this country is exclusively offered by LabCorp®.

FIBROSpect II® uses a combination of three markers that directly measure fibrogenesis of the liver, analyzed with a patented algorithm. The markers include hyaluronic acid, TIMP-1 and alpha-2 macroglobulin. FIBROSpect II® is offered exclusively by Prometheus™ laboratories.

Policy/Criteria

Combined serum markers of hepatic fibrosis, evaluated with algorithms to produce a predictive score, are considered investigational in the diagnosis and monitoring of patients with chronic liver disease.

Scientific Background

Validation of the clinical use of any diagnostic test focuses on three main principles:

1. the technical feasibility of the test;
2. the diagnostic performance of the test, such as sensitivity, specificity, and positive and negative predictive value in different populations of patients and compared to the gold standard; and
3. the clinical utility of the test, i.e., how the results of the diagnostic test will be used to improve the management of the patient.

HCV FIBROSURE™

Technical Feasibility

Measurement of the serum levels of liver function tests (i.e., alpha-2 macroglobulin, haptoglobin, GGT, total bilirubin, and apolipoprotein A1) are readily available biochemical tests. However, measurement of serum factors that directly measure fibrogenesis are relatively novel, and not readily available.

Diagnostic Performance

Initial research into the HCV FIBROSURE™ algorithm involved testing an initial panel of 11 serum markers in 339 patients with liver fibrosis who had undergone liver biopsy. From the original group of 11 markers, 5 were selected as the most informative, based on logistic regression, neural connection, and receiver operating curves. Markers included alpha-2 macroglobulin, haptoglobin, gamma globulin, apolipoprotein A1, gamma glutamyl transpeptidase and total bilirubin. (2) Using an algorithm-derived scoring system ranging from 0–1.0, the authors reported that a score of less than 0.10 was associated with a negative predictive value of 100% (i.e., absence of fibrosis, as judged by liver biopsy scores of less than METAVIR F2). A score greater than 0.60 was associated with a 90% positive predictive value of fibrosis (i.e., METAVIR F2-F4). The authors concluded that liver biopsy might be deferred in patients with a score less than 0.10.

The next step in the development of this test was the further evaluation of the algorithm in a cross section of patients, including patients with hepatitis C participating in large clinical trials before and after the initiation of antiviral therapy. One study focused on patients with hepatitis C who were participating in a randomized study of peginterferon and ribavirin. (3) From the 1,530 participants, 352 patients with stored serum samples and liver biopsies at study entry and at 24-week follow-up were selected. The FibroSure score was calculated and then compared to the METAVIR liver biopsy score. At a cutoff point of 0.30, the FibroSure score had 90% sensitivity and 88% positive predictive value for the diagnosis of METAVIR F2-F4. The specificity was 36%, and the negative predictive value was 40%. There was a large overlap in scores for patients in the METAVIR F2-F4 categories, and thus the scoring system has been primarily used to subdivide patients with and without fibrosis (i.e., METAVIR F0-F1 vs. F2-F4). When used as a monitoring test, patients can serve as their own baseline. Patients with a sustained virological response to interferon also experienced reductions in the FibroTest and ActiTest scores.

Further studies were done to formally validate the parameters used to calculate the FibroSure scores. Acceptable levels of intra-laboratory and intra-patient variability were reported. (4,5) Poynard and colleagues also evaluated discordant results in 537 patients who underwent liver biopsy and the Fibro and Actitest on the same day, with the discordance attributed to either the limitations in the biopsy or serum markers.(6) In this study, cutoff values were used for the individual METAVIR scores (F0-F4) and also for combinations of METAVIR scores (i.e., F0-F1, F1-F2, etc.) The definition of a significant discordance between FibroTest and ActiTest and biopsy scores was a discordance of at least two stages or grades in the METAVIR system. Discordance was observed in 29% of patients. Risk factors for biopsy failure included the biopsy size, number of fragments, and the number of portal tracts represented in the biopsy sample. Risk factors for failure of the FibroSure scoring system were presence of hemolysis, inflammation, possible Gilbert syndrome, acute hepatitis, drugs inducing cholestasis, or an increase in transaminases. Discordance was attributable to markers in 2.4% of patients and to the biopsy in 18% and nonattributed in 8.2% of patients. The authors suggested that biopsy failure, frequently due to the small size of the biopsy sample, is a common problem. As noted in two reviews, the bulk of the research regarding FibroSure was conducted by researchers with an interest in the commercialization of the algorithm. (7,8) Only one Australian study has attempted to independently duplicate the results of FibroSure in 125 patients with hepatitis C, using the cutoff point of less than 0.1 to identify lack of bridging fibrosis (i.e., METAVIR stages F0-F1) and greater than 0.6 to identify fibrosis (i.e., METAVIR stages F2-F4). (9) The negative predictive value for a score <0.1 was 89%, compared to the 100% originally reported by Imbert-Bismut, and the positive predictive value of a score greater than 0.6 was 78% compared to 90%. The reasons for the inferior results in this study are unclear, but the authors concluded that the FibroSure score did not accurately predict the presence or absence of fibrosis and could not reliably be used to reduce the need for liver biopsy.

The published literature returned additional validation studies of biochemical markers for the prediction of liver fibrosis in patients with non-alcoholic fatty liver disease (13), in patients with hepatitis C (14, 15), and in patients with HIV and hepatitis C coinfections (16).  None of these studies directly address the impact of the tests on management decisions and patient health outcomes.  Studies continue to report conflicting results concerning the positive and negative predictive values of the tests compared to liver biopsy.  For example, Ratziu and colleagues conducted a study to validate the diagnostic utility of the FibroSure algorithm for the detection of advanced fibrosis in patients with non-alcoholic fatty liver disease in two prospective groups, one in a single center study and one in a multicenter study. (13) A fibrosis test score of 0.03 had 77 % sensitivity and 90% negative predictive value for advanced fibrosis.  A fibrosis test score of 0.70 had 98% specificity and 76% positive predictive value.  A fibrosis test score equal to or higher than 0.30 had 92% sensitivity and 98% negative predictive value for bridging fibrosis or cirrhosis.  Halfon and colleagues found that in patients with hepatitis C the fibrosis test threshold giving the highest sensitivity and specificity was 0.36: sensitivity 73%, specificity 72%, negative predictive value 76% and positive predictive value 69%. (14) For the diagnosis of severe fibrosis, the fibrosis test threshold giving the highest sensitivity and specificity was 0.44: sensitivity of 76%. Specificity of 70%, negative predictive value of 90% and positive predictive value of 44%.  On the other hand Colletta and colleagues found that in patients with hepatitis C and normal aminotransferase levels that the sensitivity and specificity of the fibrosis test score compared to liver biopsy was64% and 31% respectively. (15) The positive predictive value of the fibrosis test score was 33% and the negative predictive value was 62%.

Bourliere and colleagues reported validation of the FibroTest (available in Europe and similar to FibroSure) and reported that based on ROC (Receiver Operator Curve) analysis that FibroTest was superior to APRI (AST to platelet ratio index) for identifying significant fibrosis with areas under the ROC curve of 0.81 and 0.71, respectively. (17) These studies continue to report on a number of assays used to evaluate fibrosis in patients with liver disease.

Clinical Utility
The clinical utility of a test depends on the demonstration that the test can be used to improve patient management. The primary impact of the FibroSure test is its ability to direct liver biopsy decisions, and potentially to follow response to therapy. Although the FibroSure test is reported to be widely disseminated and accepted in France, a literature search of English language publications did not identify any clinical article in which the HCV FIBROSURE™ was actively used in the management of the patient. It is not clear whether the HCV FIBROSURE™ could be used in lieu of an initial liver biopsy, or whether it could be used as an interval test in patients receiving therapy to determine whether an additional liver biopsy was necessary.

FIBROSpect II®

Technical Feasibility
As noted above, the FIBROSpect II® test consists of measurements of hyaluronic acid, TIMP-1, and alpha 2 macroglobulin. In a 2004 review, Lichtinghagen and Bahr noted that the lack of standardization of assays of matrix metalloproteinases and tissue inhibitors of metalloproteinase (TIMP) limited the interpretation of studies. (7)

Diagnostic Performance
In contrast to the FibroSure test, there are minimal published data regarding FibroSpect. Patel and colleagues investigated the use of these serum markers in an initial training set of 294 patients with hepatitis C and further validated the resulting algorithm in a set of 402 patients. (10) The algorithm was designed to distinguish between no/mild fibrosis (F0-F1) and moderate to severe fibrosis (F2-F4). With the prevalence of F2-F4 disease of 52% and a cutoff value of 0.36, the positive and negative predictive values were 74.3% and 75.8%, respectively.

Using a FibroSpect II cutoff score of 0.42 (Patel (10) reported a cutoff value of 0.36), Christensen reported a sensitivity of 93%, specificity of 66%, overall accuracy of 76% and a negative predictive value of 94. (19)

Clinical Utility
No studies were identified in the published literature in which results of the FibroSpect test were actively used in the management of the patient.

An updated search of literature failed to identify any studies of clinical utility, thus the policy statement is unchanged. The published studies for these “combinations of markers” continue to focus on test characteristics such as sensitivity, specificity, and accuracy. (26-29)

Other Scoring Systems
Other scoring systems have also been developed. For example, the APRI scoring system (aspartate aminotransferase [AST] to platelet ratio) requires only the serum level of AST and the number of platelets; this system uses a simple non-proprietary formula that can be calculated at the bedside to produce a score for the prediction fibrosis. (11) Using an optimized cutoff value derived from a training set and validation set of patients with hepatitis C, the authors reported that the negative predictive value for fibrosis was 86% and that the positive predictive value was 88%. Rosenberg and colleagues developed a scoring system based on an algorithm combining hyaluronic acid, amino terminal propeptide of type III collagen, and TIMP-1. (12) The algorithm was developed in a test set of 400 patients with a wide variety of chronic liver diseases and then validated in another 521 patients. The algorithm was designed to discriminate between no or mild fibrosis and moderate to severe fibrosis. The negative predictive value for fibrosis was 92%.

Al-Mohri and colleagues conducted a validation study of the APRI scoring system in 46 patients coinfected with HIV and hepatitis C. (16) Compared to liver biopsy the APRI scores of 1.5 or greater (the higher cut-off) were 100% specific and 52% sensitive.  The positive predictive value was 100% and negative predictive value was 45%.  For APRI scores less than 0.5 (the lower cut-off) the sensitivity was 82% and specificity was 46% in ruling out significant fibrosis (positive predictive value was 79% and negative predictive value was 50% for lower cut-off).  None of the studies provide sufficient information to be able to reach conclusions concerning how the results of the fibrosis biomarker tests impact treatment decisions and health outcomes.

Several review articles have reached similar conclusions. (20, 21) These articles have also noted concerns about test accuracy, cost-effectiveness, and the need for independent validation. In addition, some studies are reporting on other approaches in the non-invasive assessment of fibrosis such as elastography (FibroScan) using ultrasonography. (22)

While most of the studies to identify fibrosis have been in patients with hepatitis C, studies are also being conducted in patients with chronic hepatitis B. (23, 24) There also are no studies of the clinical utility for these patients. Of note, some researchers have noted that different markers may be needed for this assessment in patients with hepatitis B.

Giannini reported that use of the AST/ALT ration and platelet counts in a diagnostic algorithm would have avoided liver biopsy in 69% of their patients and would have correctly identified the absence/presence of significant fibrosis in 80.5% of these cases. (25)

Conclusion
The FibroSure test has been developed and extensively tested by the same group of investigators. However, in the only study that attempted independent validation, the diagnostic performance of the FibroSure test was inferior to that reported by the original investigators. (9) There are less published data regarding the FibroSpect test. With respect to clinical utility, i.e., how the results of either test can be used to improve patient management, it is suggested that biopsy can be deferred, and presumably treatment as well, when there is a score with a high negative predictive value for fibrosis. Although the negative predictive value for the FibroSure was reported as 100% by the authors who developed the test (2), another group of investigators reported a 89% negative predictive value (9), suggesting that 11% of patients would potentially forego initial antiviral therapy. The negative predictive value of FibroSpect was reported as 75.8%. (10) There were no studies identified that actually used the results of either the FibroSure or FibroSpect in the management of patients to direct liver biopsy decisions. Therefore, there is inadequate scientific data to permit conclusions regarding the impact of either FibroSure or FibroSpect on health outcomes.

References

1. BlueCross BlueShield Association Medical Policy Reference Manual, Policy No. 2.04.41
2. Imbert-Bismut F, Ratziu V, Pieroni L et al. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001;357(9262):1069-75
3. Poynard T, McHutchison J, Manns M et al. Biochemical surrogate markers of liver fibrosis and activity in a randomized trial of peginterferon alfa-2b and ribavirin. Hepatology 2003;38(2):481-92
4. Imbert-Bismut F, Messous D, Thibaut V et al. Intra-laboratory analytical variability of biochemical markers of fibrosis (Fibrotest) and activity (Actitest) and reference ranges in healthy blood donors. Clin Chem Lab Med 2004;42(3):323-33
5. Halfon P, Imbert-Bismut F, Messous D et al. A prospective assessment of the interlaboratory variability of biochemical markers of fibrosis (FibroTest) and activity test (ActiTest) in patients with chronic liver disease. Comp Hepatol 2002;I(1):3
6. Poynard T, Munteanu M, Imbert-Bismut F et al. Prospective analysis of discordant results between biochemical markers and biopsy in patients with chronic hepatitis C. Clin Chemistry 2004;50(8):1344-55
7. Lichtinghagen R, Bahr MJ. Noninvasive diagnosis of fibrosis in chronic liver disease. Expert Rev Mol Diagn 2004;4(5):715-26
8. Afdhal NH, Nunes D. Evaluation of liver fibrosis: a concise review. Am J Gastroenterol 2004;99(6):1160-74
9. Rossi E, Adams L, Prins A et al. Validation of the FibroTest biochemical markers score in assessing liver fibrosis in hepatitis C patients. Clin Chem 2003;49(3):450-4
10. Patel K, Gordon SC, Jacobson I et al. Evaluation of a panel of non-invasive serum markers to differentiate mild from moderate-to-advanced liver fibrosis in chronic hepatitis C patients. J Hepatol 2004;41(6):935-42
11. Wai CT, Greenson JK, Fontana RJ et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38(2):518-26
12. Rosenberg WM, Voelker M, Thiel R et al. Serum markers detect the presence of liver fibrosis: a cohort study. Gastroenterology 2004;127(6):1704-13
13. Ratziu V, Massard J, Charlotte F et al. Diagnostic value of biochemical markers (FibroTest-FibroSURE) for the prediction of liver fibrosis in patients with non-alcoholic fatty liver disease. BMC Gastroenterol 2006;6(6)
14. Halfon P, Bourliere M, Deydier R et al. Independent prospective multicenter validation of biochemical markers (Fibrotest –Actitest) for the prediction of liver fibrosis and activity in patients with chronic hepatitis C: The Fibropaca Study. Am J Gastroenterol 2006;101:547-55
15. Colletta C, Smirne C, Fabris C et al. Value of two noninvasive methods to detect progression of fibrosis among HCV carriers with normal aminotransferases. Hepatology 2005;42:838-45
16. Al-Mohri H, Cooper C, Murphy T et al. Validation of a simple model for predicting liver fibrosis in HIV/hepatitis C virus-coinfected patients. HIV Medicine 2005; 6: 375-78
17. Bourliere M, Penaranda G, Renou C et al. Validation and comparison of indexes for fibrosis and cirrhosis prediction in chronic hepatitis C patients: proposal for a pragmatic approach classification without liver biopsies. J Viral Hepat 2006;13(10):659-70
18. Christensen C, Bruden D, Livingston S et al. Diagnostic accuracy of a fibrosis serum panel (FIBROSpect II) compared with Knodell and Ishak liver biopsy scores in chronic hepatitis C patients. J Viral Hepat 2006;13(10):652-8
19. Thuluvath PJ, Krok KL. Noninvasive markers of fibrosis for longitudinal assessment of fibrosis in chronic liver disease: are they ready for prime time? Am J Gastroenterol 2005;100(9):1981-3
20. Rockey DC, Bissell DM. Noninvasive measures of liver fibrosis. Hepatology 2006; 43(2 suppl 1):S113-20
21. Foucher J, Chanteloup E, Vergniol J et al. Diagnosis of cirrhosis by transient elastography (FibroScan): a prospective study. Gut 2006; 55(3):403-8
22. Zeng MD, Lu LG, Mao YM et al. Prediction of significant fibrosis in HBeAG-positive patients with chronic hepatitis B by a noninvasive model. Hepatology 2005;42(6):1437-45
23. Mohamadnejad M, Montazeri G, Fazlollahi A et al. Noninvasive markers of liver fibrosis and inflammation in chronic hepatitis B-virus related liver disease. Am J Gastroenterol 2006;101(11):2537-45
24. Wai CT, Cheng CL, Wee A et al. Non-invasive models for predicting histology in patients with chronic hepatitis B. Liver Int 2006; 26(6):666-72
25. Giannini EG, Zaman A, Ceppa P et al. A simple approach to noninvasively identifying significant fibrosis in chronic hepatitis C patients in clinical practice. J Clin Gastroenterol 2006;40(6):521-7
26. Mehta P, Ploutz-Snyder R, Nandi J et al. Diagnostic accuracy of serum hyaluronic acid, FIBROSpect-II, and YKL-40 for discriminating fibrosis stages in chronic hepatitis C. Am J Gastroenterol 2008; 103(4):928-36
27. Patel K, Nelson DR, Rockey DC et al. Correlation of FIBROSpect II with histologic and morphometric evaluation of liver fibrosis in chronic hepatitis C. Clin Gastroenterol Hepatol 2008; 6(2):242-7
28. Snyder N, Nguyen A, Gajula L et al. The APRI may be enhanced by the use of the FIBROSpect II in the estimation of fibrosis in chronic hepatitis C. Clin Chim Acta 2007; 381(2):119-23
29. Sebastiani G, Vario A, Guido M et al. Performance of noninvasive markers for liver fibrosis is reduced in chronic hepatitis C with normal transaminases. J Viral Hepat 2008; 15(3):212-8

Cross References

None

Laboratory Section Table of Contents