Liver Fibrosis Testing in Patients With Type 2 Diabetes: The Time Is Now Free

In a study in this issue of Diabetes Care, Caussy et al. (1) have looked at the outcomes of noninvasive fibrosis testing in patients with type 2 diabetes and metabolic dysfunction–associated steatotic liver disease (MASLD). MASLD is the most prevalent liver disease in Western countries, affecting up to 30% of the general adult population (2). The prevalence is even higher among people with metabolic comorbidities, such as obesity or type 2 diabetes, estimated to be >50% (2). People with MASLD are either asymptomatic or have subtle nonspecific symptoms, such as tiredness or low-grade right upper quadrant pain; therefore, the diagnosis is often incidental, following abnormal liver blood tests or an abdominal ultrasound performed for different indications (3).

Recent guidelines from learned societies recommend case finding of liver fibrosis in patients with risk factors for MASLD, particularly those with type 2 diabetes or obesity with at least one cardiometabolic risk factor (4,5). This recommendation has not been widely applied in clinical practice to date, due to a combination of factors including limited awareness, limited noninvasive testing availability in nonhepatology settings, a perceived lack of interventions for MASLD, and concerns regarding testing capacity (6).

In an interim analysis of a prospective study in France, Caussy et al. (1) present the results of noninvasive fibrosis testing in 654 patients with MASLD on a background of either type 2 diabetes or obesity. Noninvasive testing included indirect fibrosis markers (fibrosis-4 index [FIB-4] score, nonalcoholic fatty liver disease fibrosis [NFS] score, Metabolic Dysfunction–Associated Fibrosis 5 [MAF-5] score), patented serum tests (Enhanced Liver Fibrosis [ELF] test, FibroTest, FibroMeter), and liver stiffness measurement with FibroScan, shear wave elastography, and magnetic resonance elastography in a subset of patients. The authors used a pragmatic and well-constructed hierarchical risk stratification for assessing the diagnostic performance of the various noninvasive tests, mirroring clinical practice where liver biopsy is indicated or available for only a minority of patients.

Caussy et al. subsequently examined the performance of the recommended two-step risk stratification algorithms (FIB-4 followed by ELF or FibroScan), to understand the burden of referrals to hepatology clinics and the prevalence of advanced fibrosis in at-risk populations. Using the two-step algorithm would result in a referral rate of 14%–18% and a positive predictive value of 39%–62% depending on the combination used, with a negative predictive value of 88%–91%. Overall, 17.6% and 9.3% of patients had an intermediate/high risk and a high risk of advanced fibrosis, respectively. There are several important messages from this article that are worth discussing in more detail.

Firstly, 98.4% of patients at high risk of advanced fibrosis had type 2 diabetes, while high BMI in isolation was not an independent risk factor for advanced fibrosis. This observation validates the recent recommendation of the European Association for the Study of the Liver/European Association for the Study of Diabetes/European Association for the Study of Obesity to target for fibrosis testing those with obesity and an additional cardiometabolic risk factor (and not those with obesity in isolation) (4). As case finding does not currently occur in routine clinical practice, it is important to start from those at higher risk of significant liver disease. The population with diabetes, with a risk of advanced fibrosis of >9% (also confirmed in other studies [7,8]), is the logical cohort to start with. A liver health check can be incorporated in the annual diabetes review as recently highlighted (9).

Secondly, FIB-4 performed significantly better than the other available indirect serum tests as the first step of risk stratification. MAF-5, designed for testing the general population, cannot be used in this context of relatively high prevalence of advanced liver disease. The age-adapted FIB-4 cutoff of >2.0 in those aged >65 years is probably not fit for purpose, as it would result in suboptimal diagnostic accuracy. The Camden and Islington pathway previously showed suboptimal sensitivity of the 2.0 cutoff (10), and this study confirms that until better tests become available for this age-group, we should continue to use the 1.3 cutoff and accept a higher number of referrals.

Thirdly, area under the receiver operating character curve for the ELF test was numerically higher than that for the FibroMeter and the FibroTest, and until proven otherwise the ELF test should be considered the patented serum test of choice for the MASLD population. The study confirmed that the 7.7 ELF cutoff is too low and not fit for clinical use. The study also confirmed that a cutoff of 9.6–9.8 is optimal in deciding which patients should be referred to secondary care and therefore externally validated the findings of the Camden and Islington pathway (10).

The results of this study indicate that there is a significant burden of advanced fibrosis in patients with type 2 diabetes. The licensing of medications with an indication for fibrotic MASLD (11) invalidates previous arguments on the lack of interventions for identified patients. Therefore, establishing frameworks for noninvasive liver fibrosis testing as part of the routine clinical care of patients with type 2 diabetes is important going forward. A two-step algorithm (FIB-4 followed by either ELF or transient elastography, based on local availability) seems to work well based on the results of this study (Fig. 1). The next steps are to increase awareness in nonhepatology settings and create the capacity required for testing these patients. Multimorbidity care models will need to be codesigned by hepatologists and diabetologists with metrics of effectiveness and cost-effectiveness incorporated into them (12). What is abundantly clear is that we cannot ignore the presence of significant liver disease in patients with type 2 diabetes any longer.