MASLD aspirin may help decrease liver fat

 

a recent article found that aspirin reduced the severity of metabolic dysfunction-associated steatotic liver disease (MASLD): NAFLD aspirin dec liver fat JAMA2024 in dropbox, or doi:10.1001/jama.2024.1215. The commentary section below details the new nomenclatures for steatotic liver diseases, per the AASLD (the American Association for the Study of Liver Diseases)

 

Details:

-- 80 participants 18 to 70 years old who had established MASLD were enrolled in a six-month phase-2 randomized double-blind placebo-controlled clinical trial

    -- patients were excluded if they had other causes of hepatic steatosis, including significant alcohol intake (at least three drinks per day for men or two drinks a day for women), diagnosis of viral hepatitis, hemochromatosis, alpha-1 anti-trypsin deficiency, Wilson’s disease, autoimmune hepatitis, or HIV

-- patients were randomized to aspirin 81 mg versus matching placebo, along with standard nutritional counseling for MASLD; medication adherence rates were determined by the number of returned pills at month six

-- mean age 48, 55% women, 80% white (n=66)/15% Latino (n=12)

-- BMI 34, visceral adipose tissue volume 190 cm³

-- type II diabetes 40%, hypertension 35%

-- medications used: metformin 16%, statins 14%, vitamin E 9%, GLP-1 7.5%

-- ALT 52 IU/L AST 18 IU/L, hemoglobin A1c 5.7%, LDL 112 mg/dL, HDL 44 mg/dL

-- mean hepatic fat content 35% (indicating moderate stenosis), hepatic fat by MRI protein density fat fraction 15%, liver stiffness by VCTE (e.g. fibroscan) 7 kPA with 40% having fibrosis of at least stage 2

-- 84% had a liver biopsy, with fibrosis stage 0 found in 30%, stage I in 34%, stage II in 32%, and only one individual in each group had stage III

 

-- primary endpoint: absolute change in hepatic fat content, measured by proton magnetic resonance spectroscopy (MRS) at the 6 month follow-up

-- secondary outcomes: mean percentage change in hepatic fat content by MRS, the proportion of patients achieving at least 30% reduction in hepatic fat, and the mean absolute and relative reductions of hepatic fat content, measured by magnetic resonance imaging proton density fat fraction (MRI-PDFF). The minimally clinically important differences for study outcomes were not prespecified.

--Post-hoc assessment as exploratory outcomes included attainment of an ALT reduction of at least 17  IU/L, a 30% point or greater reduction in hepatic fat fraction, and a 50%  or greater reduction in hepatic fat fraction

 

Results:

-- 71 of the 80 patients (89%) completed the six-month follow-up

 

-- mean absolute change in hepatic fat content by MRS (primary endpoint):

    -- aspirin group: -6.6%

    -- placebo: 3.6%

        -- difference: -10.2% (-27.7% to -2.6%), p=0.009

 

-- relative fat content (secondary endpoint):

    -- aspirin group: -8.8 percentage points

    -- placebo: 30.0 percentage points

        -- difference: - 38.8 percentage points (- 66.7 to -10.8 %), p=0.007

 

-- Proportion of patients with 30% or greater relative reduction in hepatic fat (secondary endpoint):

    -- aspirin group: 42.5%

    -- placebo: 12.5%

        -- difference: -30.0% (11.6% to 48.4%), p=0.006

 

-- Reduced absolute hepatic fat content by MRI-PDFF (secondary endpoint):

    -- aspirin group: -2.7%

    -- placebo: 0.9%

        -- difference: -3.7% (-6.1% to -1.2%) p=0.004

 

-- Reduced relative hepatic fat content by MRI-PDFF (secondary endpoint):

    -- aspirin group: -11.7 percentage points

    -- placebo: 15.7 percentage points

        -- difference: -27.3 percentage points (-45.2 to -9.4) p=0.003

 

-- ALT:

    -- aspirin group: 53.1 IU/L decreasing to 37.2 IU/L, -16.1 IU/L

    -- placebo: 51.2 IU/L decreasing to 50.4 IU/L

        -- difference -15.6 IU/L (-22.7 to -8.4), p<0.001

 

-- AST:

    -- aspirin group: 40.5 IU/L decreasing to 26.0 IU/L, difference of -14.6 IU/L

    -- placebo: 39.8 IU/L decreasing to 39.6 IU/L

        -- difference -14.3 IU/L (-19.0 to -9.7), p<0.001

 

-- VCTE (e.g. fibroscan):

        -- aspirin group: 7.1 kPA, decreasing to 6.0, difference of -1.1kPA

        -- placebo: 6.9 IU/L increasing to 8.7

        -- difference –2.8 kPA, p<0.001 (-19.0 to -9.7), p<0.001

 

--Neither body weight nor weight loss were different between the groups

 

-- Per protocol analysis (included the 71 of the 80 patients who completed the trial): results were similar to the findings noted above

-- sensitivity analysis: the above numbers were only marginally different by controlling for age, sex, race and ethnicity, type II diabetes, bodyweight, and visceral adipose tissue volume

 

-- subgroup analyses:

    -- participants with at least stage II fibrosis at baseline (15 in the aspirin group and 17 in the placebo group): aspirin still significantly reduced absolute fat fraction compared with placebo (-11.7% versus 1.9%) mean difference of -13.7%

    -- aspirin significantly reduced relative hepatic fat compared with placebo by -23.3 percentage points versus 33.0 percentage, with difference of -56.3 percentage points

 

-- post hoc analyses

    -- ALT decreasing by at least 17 IU/L: 32.4% versus 8.8%, difference of 23.6% (5.8% to 41.4%), p=0.02

    -- ALT reduction of at least 17 IU/L plus hepatic fat reduction of at least 30 percentage points: 20.6% in the aspirin group versus 2.9% in the placebo group, difference 17.7% (3.3% to 31.9%), p=0.04

    -- hepatic fat reduction of at least 50 percentage points or greater: 24.3% in the aspirin group versus 5.9% in the placebo group, difference of 18.4% (2.5% to 34.3%), p=0.05`        

 

Adverse events:

    upper respiratory infections (10% in each group, not likely drug related), arthralgias (5.0% per aspirin versus 7.5% for placebo); one participant on aspirin (2.5%) had drug-related heartburn

 

 

Commentary:

-- first, a review of the new nomenclature (a tad more complex than before...):

     -- MASLD (metabolic dysfunction-associated steatotic liver disease): a renaming of NAFLD (nonalcoholic fatty liver disease). These individuals have fatty liver with >5% hepatic steatosis and with at least one risk factor for cardiometabolic dysfunction such as dyslipidemia, hypertension, type II diabetes, or obesity based on BMI, and minimal or no alcohol consumption (<20g/d for females and <30g/d for males)

    -- MASH (MASLD with metabolic dysfunction-associated steatohepatitis): a renaming of NASH (nonalcoholic steatohepatitis)

    -- MASH cirrhosis: patients who have documented cirrhosis with prior histologic evidence of MASH or MASLD

    -- MetALD (metabolic dysfunction-and alcohol-associated liver disease): patients with hepatic steatosis and at least one metabolic risk factor, along with a history of moderate but not heavy alcohol use; this is a new category which allows for moderate but higher amounts of alcohol (30-60g/d for men and 20-50g/d for women)

    -- ALD (alcohol-related liver disease) is for patients who have higher intake of alcohol with associated liver disease

-- one big issue with the above classification, to me, is the role of alcohol and how well-defined it is in terms of progression of liver disease (as well as how accurately the patients quantify their alcohol consumption…). A summary article assessing patients with already defined NAFLD found that moderate alcohol consumption, as defined above, sometimes prevents the progression of fibrosis in the liver, but other studies have found worsening of fibrosis: https://pmc.ncbi.nlm.nih.gov/articles/PMC7641550/ .

    -- there is a pretty significant clinical concern now about the dangers of any alcohol consumption, and that the real goal is zero (https://gmodestmedblogs.blogspot.com/2023/11/alcohol-use-disorder-meds.html ). So, though the question of alcohol consumption in those with MASLD may be somewhat moot, the data on cancer risk and cardiovascular disease strongly suggest that zero alcohol is best.

        -- this issue reflects a potential concern with guidelines written by specialty societies: they (not surprisingly) focus on the specific area of their concern (liver disease, per the AASLD).  Reading their guidelines, the clear focus is on adverse hepatic outcomes: cardiovascular outcomes, which may well be the major ones leading to morbidity/mortality overall, are not even mentioned. But we in primary care, in particular, are interested in taking care of the whole patient (and holistically so), which really means that we should not settle for mild-to-moderate alcohol consumption as acceptable for disease prevention

 

--MASLD is the most common cause of chronic liver disease in the West, affecting 30% of adults

    -- up to one third of patients develop progressive steatohepatitis and fibrosis, which may be associated with cirrhosis, hepatocellular carcinoma, and death

-- preclinical studies of steatohepatitis found that platelets were the first cells to infiltrate the liver, promoting inflammation and releasing pro-inflammatory cytokines. Aspirin prevents fibrosis and hepatocellular carcinoma by inhibiting pro-inflammatory cyclooxygenase-2 and platelet-derived growth factor signaling

-- observational studies have supported the positive effect of aspirin in lowering rates of hepatic disease progression to advance fibrosis, hepatocellular carcinoma, and liver-related mortality

-- by the way, aspirin has fallen off the prior use for primary prevention of cardiovascular disease vascular disease, given the association with increased harms (esp major hemorrhage) with no clear benefit: https://gmodestmedblogs.blogspot.com/2018/09/aspirin-in-healthy-adults-harm-without.html ; and the USPSTF downgraded aspirin's use in primary prevention: https://gmodestmedblogs.blogspot.com/2022/05/uspstf-aspirin-for-primary-cvd.html

    -- which does bring up the issue of whether there might be different but effective other medications as opposed to aspirin:

        -- clopidogrel and other antiplatelet agents may work (??), and they seem to have decreased risk of GI bleeds than aspirin: https://pmc.ncbi.nlm.nih.gov/articles/PMC10199733/

        -- what about other medications that decrease systemic inflammation (eg colchicine)? we do know that chronic inflammation itself is associated with fibrosis and with cancers

            -- however, these medications do not affect the cyclooxygenase system, which these researchers feel may be the key to aspirin effectiveness in MASLD.  so other studies would need to be done to see if these meds worked (as well as larger confirmatory studies that aspirin really works)

 

 -- this study found that a 6-month course of aspirin 81mg was associated with a 10.3% reduction in liver fat, including a 42.5% increased likelihood of attaining a 30% or greater relative reduction in hepatic fat (this 30% mark has been considered a clinically meaningful treatment in some other studies, based on histologic improvement in steatohepatitis and fibrosis). Also several markers of hepatic inflammation were reduced with aspirin (ALT, AST) along with decreased liver stiffness by VCTE. the placebo group had progression of several of the parameters assessed above

--though a small study, there were essentially no adverse effects of the aspirin (1 case of heartburn)

    -- as a side note, it is reasonably recommended that people who will be on aspirin (or NSAIDs) should have H Pylori testing/treatment when positive in order to decrease the risk of major GI bleeding: https://gmodestmedblogs.blogspot.com/2015/05/h-pylori-and-nsaids-increased-gi.html

        -- a 2015 study found that the there was a significantly higher risk of peptic ulcer bleeding in those of either NSAIDs or low-dose aspirin who had  H Pylori infection (see hpylori gi bleed asa nsaid amjgastro 2015 in dropbox, or doi: 10.1038/ajg.2015.98)

        -- in the 2012 Maastrich IV/Florence Consensus Report on H Pylori, they comment that "For aspirin, even given at low dose, Hpylori eradication can prevent gastropathy and should be undertaken in patients with a history of peptic ulcers. In such patients, the residual risk of peptic ulcer bleeding due to continued aspirin use after Hpylori has been successfully treated is very low". https://core.ac.uk/reader/43323548?utm_source=linkout .  

            -- of note, this study did not differentiate patients who were symptomatic, and many cases of H pylori infection are in asymptomatic patients who still benefit from H Pylori eradication in terms of decreased gastric cancer risk

        -- this was also found in a newer study in the UK, finding that H Pylori eradication in people over 60yo had less peptic ulcer bleeding: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(22)01843-8/fulltext

-- another issue is who to screen for MASLD?? not so clear. screening is reasonable in patients with abnormal LFTs without another explanation, but many with NAFLD/MASLD have normal LFTs. of note, there is argument in the GI literature that the normal range of ALT and AST should be lowered; eg see https://gmodestmedblogs.blogspot.com/2023/10/nafld-utility-of-lower-cutpoint-of-alt.html)? it is pretty clear that those with diabetes and evidence of insulin resistance (including metabolic syndrome) should have screening, primarily with abdominal ultrasound and serum fibrosis tests for further risk stratification (eg FIB-4): https://pmc.ncbi.nlm.nih.gov/articles/PMC8475001/

    -- for a further discussion of this, see  https://gmodestmedblogs.blogspot.com/2023/10/nafld-utility-of-lower-cutpoint-of-alt.html 

    -- the mainstay of current MASLD treatment is weight loss, but it is important to keep in mind that MASLD can happen in lean individuals, and they may well have higher mortality than non-lean ones: https://gmodestmedblogs.blogspot.com/2023/10/lean-nafld-higher-mortality-than-non.html

 

Limitations:

-- this was a relatively small study in a single institution in Boston and with only 6-month followup. For a study like this one to be integrated into our clinical practice, it would need to be replicated in larger studies in different areas and with more diverse patient populations

-- the study did not include repeated liver biopsies which are pretty clearly associated with important clinical hepatic outcomes, so the markers of improvement that this study assessed were other surrogate markers, and we know that these types of markers may not fully reflect the actual clinical concerns (which for MASLD really are for progression to cirrhosis/hepatocellular carcinoma/death: see https://gmodestmedblogs.blogspot.com/2024/06/using-surrogate-markers-for-disease-are.html

-- there were statistical manipulations (multiple imputations) for the 9 participants with missing outcome data in the main analysis. but the per-protocol analysis, which only included the 71 participants who completed the study, had similar results. but per-protocol analysis is less statistically rigorous than intention-to-treat analysis

-- the fact that minimally clinically important differences for study outcomes were not prespecified really undercuts the actual clinical understanding of their results (ie, a very significant "p value" connoting statistical signficance may have no real clinical significance. however, there was some justification that a 30% or greater relative reduction in hepatic fat is clinically meaningful, though not prespecified in this study

 

so,

--another challenge for us in primary care: how to keep up with the ever-changing nomenclature of different specialty societies?? and i thought the infectious disease crew were particularly bad in continually renaming their various microbes.....

-- this study is important, especially since it reinforces the relevance of MASLD in clinical practice as an entity that is quite common and is associated with several bad outcomes when it progresses

-- it also brings to light that aspirin could be beneficial and appeared to be so consistently in the many diverse outcomes measured in this study. Though weight loss is an important avenue to decreasing the progression to MASH, it is important to remember that many lean individuals can get MAFLD/MASH and seem to have an even higher likelihood of a terrible prognosis

-- it also would be interesting to compare aspirin use in MASLD to its combination with other meds (eg vitamin E in patients without diabetes, GLP-1 receptor agonists)

-- and, i think it is important to realize that the association between metabolic syndrome/insulin resistance doesn't affect just the liver, but has profound implications for the whole body, and it should be identified and treated aggressively with nonpharmacologic interventions (diet/exercise/wt loss if indicated) and meds (eg statins to aggressively decrease hyperlipidemia) as needed

 

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

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