Multicenter external validation of FIB-6: a novel, machine-learning, simple bedside score to rule out severe liver fibrosis and cirrhosis in patients with MAFLD

Background: We developed and validated a noninvasive diagnostic index based on routine laboratory parameters for predicting the stage of hepatic fibrosis in patients with chronic hepatitis C (HCV). Patients and Methods: Machine learning with random forests algorithm was used to develop a noninvasive index using retrospective data of 7238 biopsy-proven chronic hepatitis C (CHC) patients from two centers; derivation dataset (n=1821) and validation set in the second center (n=5417). ROC curve analysis was used to define cutoffs for different stages of fibrosis. Performance of the new score was externally validated in cohorts from two other sites in Egypt (n=560) and seven different countries (n=1317). Results were also compared with three established tools (FIB-4, APRI, and AAR). Results: Age, aspartate, and alanine aminotransferases, alkaline phosphatase, albumin (g/dL), and platelet count (/cm3) correlated with the biopsy-derived stage of liver fibrosis in the derivation cohort and were used to construct the model for predicting the fibrosis stage, resulting in a FIB-6 index. Applying the cutoffs from the ROC curve analysis of the derivation group in the validation groups indicated very good performance in ruling out cirrhosis (negative predictive value [NPV] = 97.7%), severe fibrosis (NPV = 90.2%), and significant fibrosis (NPV = 65.7%). In the external validation groups, FIB-6 demonstrated higher sensitivity and NPV than FIB-4, APRI, and AAR. Conclusion: FIB-6 score is an accurate, simple, non-invasive test for ruling out advanced fibrosis and liver cirrhosis in chronic hepatitis C patients and performs better than APRI, FIB-4, and AAR. Funding Statement: Public donations for ELRIAH. Declaration of Interests: The authors declare no conflict of interest. Ethics Approval Statement: The study protocol was approved by the Research and Ethics Committee of ELRIAH (OHRP IRB #8819). The protocol and conduct of the study complied with the International Ethical Guidelines for Biomedical Research Involving Human Subjects [28] and its amendments in 2008. The requirement for written informed consent was waived by the ethics committee as this was a retrospective analysis of anonymized patient data collected during routine clinical care.