FibroScan-based assessment tool: best predictor of bad liver outcomes if NAFLD/MASLD

 A recent article found that the Agile assessment using vibration-controlled transient elastography (eg, FibroScan) was highly predictive of liver-related events in patients with metabolic dysfunction-associated steatotic  liver disease (MASLD, otherwise known as MAFLD, as well as NAFLD of yore), see NAFLD fibroscan JAMA2024 in dropbox, or doi:10.1001/jama.2024.1447)

 

Details:

-- 16,603 patients from 16 tertiary referral centers in the US, Europe, and Asia, in a study from February 2004 to January 2023 with data collected prospectively

-- eligible patients were adults aged at least 18yo with hepatic steatosis (steatosis in at least 5% of hepatocytes found on histologic exam), or imaging studies (ultrasound, CT, MRI, or vibration-controlled transient elastography (VCTE) finding controlled attenuation parameter of at least 248 dB/m)

-- patients were excluded if they had other liver diseases such as viral hepatitis, HIV, excessive alcohol consumption, secondary causes of hepatic steatosis (e.g. use of steroids), or history of hepatocellular carcinoma (HCC), hepatic decompensation, liver resection, liver transplant, or other malignant carcinomas

-- mean age 53, 58% male

-- 18% were from the US or Europe, 82% from Asia

-- BMI 27, diabetes 35%, hypertension 35%

-- ALT 37 IU/L, AST 31 IU/L, GGT 44 IU/L, albumin 44 g/L, bilirubin 12 µmol/L, platelet count 237, creatinine 72 µmol/L

-- liver stiffness measured by FibroScan 6.0 kPa, controlled attenuation parameter 303 dB/m

-- fibrosis score: F0 16%/F1 34%/F2 17%/F3 21%/F4 12%

-- 35% had liver by biopsy: 34% had F3 or F4 fibrosis

 

-- Primary outcome: liver-related events (LREs), defined as HCC or hepatic decompensation (ascites, variceal hemorrhage, hepatic encephalopathy, or hepatorenal syndrome), liver transplant, and liver-related deaths.
    -- The Agile scores were compared with histologic assessment and 8 other noninvasive tests

        -- the Agile scores seem to be a good predictor of liver-related events: https://www.journal-of-hepatology.eu/article/S0168-8278(23)00161-7/fulltext

        -- these scores involve a computation including liver stiffness measurement by VCTE, age, sex, diabetes status, platelet count, and AST and ALT levels

            -- those with Agile 3+ and  Agile 4 scores have been found to reliably  identify advanced fibrosis and cirrhosis, respectively: https://pubmed.ncbi.nlm.nih.gov/36807720/

        -- other fibrosis scores tested in this analysis include noninvasive and histologic ones: histologic fibrosis stage, FIB-4 (Fibrosis-4 index), LSM (liver stiffness measurement), NFS (NAFLD fibrosis score), APRI (aspartate aminotransferase to platelet ratio) , FAST score (FibroScan-aspartate aminotransferase), AST:ALT ratio, BARD (BMI, AST:ALT ratio and diabetes)

        -- all fibrosis scores classified patients into low-risk, intermediate risk, and higher-risk groups based on their published  low and high cutoffs

-- secondary outcome: HCC and hepatic decompensation, analyzed separately

 

-- Follow-up median of 52 months

  

Results:

 

Baseline model (assessing scores of the different tests done at baseline)

-- 316 patients (1.9%) developed LREs over the 52 month followup

    -- 139 cases of HCC and 209 cases of hepatic decompensation

-- both the Agile 3+ and Agile 4 scores were the best predictors of LREs per the AUROC (area under the receiver-operating characteristic curve) and AUPRC (area under the time-dependent precision-recall curve):

    -- Agile 3+ and Agile 4 scores correctly reclassified fewer patients (10.2% [1314 of 12948] for Agile 3+ and 8.7% [1130 or 12948] for Agile 4) in the intermediate-risk group than the other fibrosis scores, including assessment of all 10,678 patients who had all of the  studied fibrosis markers done

        --- the AUROC for LREs of both Agile scores were significantly higher than histologic fibrosis staging and other comparator fibrosis tests at 3 and 5 years, except for LSM at 3 years: here is the relevant AUROC score comparison (Agile scores topped the list, though there was significant statistical overlap with many of the other assessments):

 

 

 

-- all of the fibrosis tests were better at predicting hepatic decompensation than HCC

 

-- specific assessment of Agile 3+ and incidence of LREs:

    – Agile 3+ score of <0.451: incidence of LREs was 0.7 (0.5-1.0) per 1000 person-years

    – Agile 3+ score of 0.451-0.678: incidence of LREs was 3.3 (2.1-5.2) per 1000 person-years

    – Agile 3+ score of >0.678: incidence of LREs was 24.9 (21.7-28.6) per 1000 person-years

        this trend was highly statistically significant, p<0.001

 

-- specific assessment of Agile 4 and incidence of LREs:

    – Agile 4 score of <0.251: incidence of LREs was 1.2 (1.0-1.5) per 1000 person-years

    – Agile 4 score of 0.251-0.842: incidence of LREs was 23.5 (19.6-28.2) per 1000 person-years

    – Agile 4 score of >0.842: incidence of LREs was 105.5 (84.4-191.9) per 1000 person-years

        this trend was highly statistically significant, p<0.001

 

-- both Agile scores outperformed the other tests more so in those >60yo

 

Serial model (assessing changes in scores over time in those having repeated VCTE exams)

-- 10,920 patients (66% of total) had repeated VCTE exams at median interval of 15 months

    -- during that time, 12% developed diabetes and hypertension

-- 7208 of 8810 patients (82%) had stable Agile 3+ scores and 8163 of 8810 (92%) had stable Agile 4 scores

    – there were more patients having an increase in Agile scores when there was a longer time interval between scans

    – though, overall, Agile scores were more stable than other scores assessed

-- incidence of LREs:

    -- people with persistently low Agile 3+ scores: 0.6 per 1000 person-years

    -- those with persistently high Agile 3+ scores: 30.1 per 1000 person-years

    -- people with high Agile 3+ scores at baseline that decreased by more than 20%: they had a substantial reduction in the risk of LRE’s

    – and,  if the Agile 3+ scores increased, they had an increased risk of LREs

    -- Agile 4 had a similar trend, though it missed more LRE’s in the low-risk group

 

Commentary:

-- MASLD is remarkably common, affecting approximately 30% of the worldwide adult population, and is one of the leading causes of cirrhosis as well as hepatocellular carcinoma in both middle-income and high-income countries

-- the estimated annual direct medical costs in the US was $103 billion and in Europe €35 billion in a 2018 study

-- though MASLD is predominantly a disease of overweight people, about 10-20% of those with MASLD are lean, and lean individuals seem to have an even worse prognosis: https://gmodestmedblogs.blogspot.com/2023/10/lean-nafld-higher-mortality-than-non.html

-- there is typically a dose-response relationship between the severity of liver fibrosis and the future risk of LRE’s

-- prior studies have shown that VCTE is a good predictor of HCC, portal hypertension, and varices

 

-- This study found that both single and serial Agile scores in routine clinical practice were highly accurate in predicting LRE’s in patients with MASLD

-- the Agile scores did eek out the other measurements, they were generally stable over time, and changes in the scores were associated with either more or fewer LREs if the scores increased or decreased

-- the Agile 3+ score seemed to provide more prognostic information about developing LREs, and the Agile 4 score was more useful for the diagnosis of MASLD-related cirrhosis

    -- that all being said, the following seem to be just about as good as Agile scores: fibrosis stage, FIB-4 (Fibrosis-4 index), LSM (liver stiffness measurement, ie FibroScan alone is just about as good as the Agile scores), and the NFS (NAFLD fibrosis score)

-- all of the scores using different methodologies were much better at predicting liver decompensation vs HCC


-- 80% of Agile scores remained stable 
in this study on repeated testing over the median time of 15 months between the tests (MASLD progresses from no or minimal fibrosis to cirrhosis typically over a period of 20 years)

    -- but 20% had reduced vs increased scores on retesting. Why were there changes over such a short period of time??

        -- were  lifestyle changes so significant that they led to major changes in hepatic fibrosis?

        -- were there false positives on one of the tests done?

            -- FibroScans can have false positives in those patients with extreme body build, acute hepatitis, heart failure, biliary obstruction, amyloidosis, and recent food intake

            -- was the issue that FibroScan results are operator-dependent, and the liver may well have specific areas of more fibrosis than other areas and there were sampling inconsistencies?

            -- given the possibility of false positives, the differing results on repeated FibroScan testing does suggest that any significant change should lead to another FibroScan, prior to doing a more invasive liver biopsy

        --a recent study found that FIB-4 score was helpful but not completely reliable (vs liver biopsy), VCTE was helpful but a  multivariate analysis including hypertension, waist circumference, triglycerides, AST, serum albumin and serum creatinine had the highest predicted value over any single marker: https://gmodestmedblogs.blogspot.com/2023/10/nafld-utility-of-lower-cutpoint-of-alt.html  

                – perhaps the issue is having a multivariable analysis and not relying on a single test (eg VCTE), and the Agile scores do combine several variables???

 

-- in terms of Agile scores:

    -- there does not seem to be a calculator on-line, though i suspect many of us could perform this calculation in our heads...:

        

    -- there is a smartphone app, but that seems to require clinicians to have their own FibroScan machine, or at least a registration number for it

 

Limitations:

-- in order to calculate an Agile score, it seems that we would need to provide the patient’s age, sex, diabetes status, platelet count, and AST and ALT levels when the VCTE was being done.

-- we should probably be repeating VCTE to assess stability of findings. unclear how often. the 20 year progression to cirrhosis would suggest repeated VCTE assessments over time?? But how often? this would need to be studied, assessing clinical outcomes.  

-- outcomes studied should include cardiovascular ones (not just liver ones), since there is a significantly increased risk of cardiovascular disease: MI, ischemic stroke and all-cause mortality (https://www.bmj.com/content/384/bmj-2023-076388)

-- we need more information about the actual effects of therapies (the various nonpharmacologic ones (eg weight loss), as well as the evident improvements by some meds (including GLP-1 receptor agonists, perhaps vitamin E and pioglitazone, as well as resmetirom, the new one on the block (a thyroid hormone receptor-beta agonist that seemed to improve fibrosis by at least one stage)

 

so, 

-- MASLD is an increasingly common disease (likely related to increasing obesity globally), which has potentially serious long-term sequelae (liver- and cardiovascular-related).

-- it is still not clear what the optimal diagnostic strategy is, but the use of FibroScans and Agile scores seem to be high on the list. More research is needed to determine who should be screened and to confirm this or other approaches to screening

    -- we also need more information on the appropriate frequency of these assessments. And this all should be based on actual clinical outcomes

-- and, we do need longterm clinical followup on the effects of both nonpharmacologic interventions as well as meds on the putative surrogate markers that make the cut (eg VCTE, Agile score, FIB-4)

    -- without an accurate surrogate marker, we need to rely on invasive and potentially harmful liver biopsies to diagnose MASLD and assess changes with any interventions.

    -- and we do know that it is not easy to find great surrogate markers: https://gmodestmedblogs.blogspot.com/2024/06/using-surrogate-markers-for-disease-are.html

 

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would the article, please email me.

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list

Comments

Popular posts from this blog

air pollution and heart disease

UPDATE: ASCVD risk factor critique

visceral fat loss is sustained long-term