MMP-9 and non-invasive markers in evaluating MASLD and atherosclerosis

Abstract

Background

Matrix metalloproteinases (MMP-9) play a vital role in extracellular matrix remodeling in metabolic-associated steatotic liver disease (MASLD) and cardiovascular diseases such as atherosclerosis. This study assesses MMP-9 levels in conjunction with non-invasive scoring systems (BAAT, BARD, FIB-4, NAFLD Fibrosis Score, and the ALT/AST ratio) aiming to enhance early diagnosis and disease monitoring.

Methods

The study included 88 participants: 25 normal individuals and 63 MASLD patients with varying grades (S1 to S3). Atherosclerosis was staged into four stages. The current study measured MMP-9 concentrations in MASLD patients and analyzed their correlation with MASLD grades, atherosclerosis stages, and non-invasive scoring systems (BAAT, BARD, FIB-4, NAFLD Fibrosis Score, and the ALT/AST ratio).

Results

Elevated MMP-9 levels were significantly associated with S2 and S3 MASLD grades (p = 0.02, 0.04; respectively) compared to controls and atherosclerosis progression (p = 0.05). Although ROC analysis showed AUC = 0.90 for discriminating advanced plaque, only two patients in our cohort had advanced plaque, which limits the generalizability of this result. Spearman correlation analysis reveals a positive correlation between MMP-9 concentration and the progression of MASLD (r = 0.4, p ≤ 0.01), atherosclerosis (r = 0.3, p ≤ 0.05), and BAAT (r = 0.254, p ≤ 0.05). BAAT score also correlated with BARD and FIB-4 scores ((r = 0.3, p ≤ 0.05, and r = 0.5 p ≤ 0.01; respectively). The highest correlation was seen between NAFLD Fibrosis score and FIB-4 (r = 0.7, p ≤ 0.01).

Conclusion

MMP-9 is a probable biomarker for assessing MASLD and may improve cardiovascular risk evaluation, especially when combined with BAAT score.

1. Introduction

Matrix metalloproteinases (MMPs), especially MMP-9, are emerging as critical drivers of disease progression, with wide range implications in liver pathology, cardiovascular health, and metabolic disorders.24, 46 The EASL, which stands for the European Association for the Study of the Liver released a consensus redefining steatotic liver disease as metabolic dysfunction-associated steatotic liver disease (MASLD), replacing NAFLD terminology, while the Asia-Pacific Association for the Study of the Liver (APASLD) continues advocating for metabolic-associated fatty liver disease (MAFLD) nomenclature.11, 15 MASLD characterized by fat accumulation exceeding 5% of hepatocytes, is increasingly prevalent due to rising obesity rates and is considered a hepatic manifestation of metabolic syndrome.9, 25, 27, 37 MASLD can advance from basic fat accumulation in the liver to metabolic-associated steatohepatitis (MASH), and eventually develop into cirrhosis and HCC.⁴⁵

Liver fibrosis is a major concern in MASLD patients. FibroScan is a non-invasive tool that closely aligns with biopsy-validated fibrosis results. However, in patients with obesity (BMI ≥30 kg/m²), it has a 2–10% failure rate. The XL probe, designed for higher BMI patients, improves accuracy, but using the wrong probe (M vs. XL) can lead to incorrect fibrosis evaluation.5, 29

Various scoring methods have been created to evaluate the extent of liver damage. Among these, the BAAT, BARD, NAFLD Fibrosis scores, and FIB-4 index are notable for their utility in clinical practice. The evaluation of the previous scores in Egypt varies due to the high prevalence of NAFLD, unique metabolic profiles, and localized clinical guidelines.⁴⁰ Approximately 32% of adults in Egypt are affected by MASLD, with 71.2% of men and 79.4% of women classified as overweight or obese, exceeding the global average with more than 7%.13, 18, 40 Implementing these non-invasive scores is crucial for effectively managing liver disease in Egypt, considering these demographic factors.6, 40, 41 Ongoing research in various health fields is focused on refining tools and assessing their effectiveness on health markers and general human health.1, 8, 10, 12, 34, 35

Importantly, MMP-9 levels are inversely related to adiponectin, suggesting a stronger association with insulin resistance related to obesity rather than obesity itself. This indicates that MMP-9 may contribute to persistent, mild inflammation referred to as metabolic inflammation or ‘metaflammation,’ characterized by metabolic disturbances and obesity.28, 32

MMPs constitute a group of enzymes that rely on zinc for their proteolytic activity, comprising 28 known members. Among them, MMP-9 that is secreted by fibroblasts and endothelial cells (ECs).3, 14, 23 MMP-9 is a zinc-dependent endopeptidase that specializes in degrading extracellular matrix (ECM) components, particularly type IV collagen and fibronectin.24, 46 MMP-9 activity is regulated by tissue inhibitors of metalloproteinases (TIMPs), emphasizing the importance of maintaining ECM balance.3, 28

In patients with MASLD, MMP-9 is significantly upregulated and serves as a key indicator of inflammatory severity.⁴⁵ MMP-9 helps degrade components of the ECM which facilitates leukocyte crossing from blood into inflamed tissues, by cleaving endothelial junctional proteins such as Platelet Endothelial Cell Adhesion Molecule (PECAM-1), which increases vascular permeability.²² This cleavage increases vascular permeability, promotes excessive leukocyte infiltration into the steatotic liver, and worsens inflammation, necrosis, and impaired regeneration.²² MMP-9 drives MASLD to MASH progression mainly through cytokine-induced secretion by neutrophils, monocytes, and macrophages, where neutrophils release pre-stored MMP-9 from granules for rapid response, while others synthesize it de novo upon stimulation by TNF-α, IL-1β, IFNγ, or TGF-β. These cytokines activate NF-κB pathway.28, 43, 45 The latter plays a prime role in enhancing MMP-9 expression that fuels inflammation, ECM remodeling, leukocyte infiltration, and anoikis resistance in steatotic hepatocytes.³

Elevated MMP-9 levels are associated not only with hepatic conditions but also with cardiovascular risk factors, including plaque deposition and plaque instability (by promoting Vascular Smooth Muscle Cells migration) in atherosclerosis, highlighting the systemic implications of MMP-9 in metabolic diseases as well as atherosclerosis.³¹ MMP-9 serum level also serves as marker of therapy effectiveness in heart failure patients and help identify those who may benefit from MMP pathway-targeted treatments.4, 31

MMP-9 is a well-established biomarker for plaque instability and inflammation in atherosclerosis, which gives a strong rationale to evaluate its levels in patients with different grades of MASLD as well as different atherosclerosis stages. Additionally, the study explores the correlation between MMP-9 levels and several non invasive scores, including the NAFLD Fibrosis Score, BAAT, BARD, FIB-4, and the ALT/AST ratio. This study comprehensively assesses MMP-9 levels in conjunction with non-invasive scoring systems aiming to enhance early diagnosis and disease monitoring.

2. Materials and methods

2.1. Patients enrollment

The study protocol was approved by the National Research Center's Ethical Review Board (NRC# 20151), and all participants provided written informed consent, adhering to the 1975 Helsinki Declaration guidelines. In this study, sixty-three patients with varying grades of MASLD categorized as S1 (mild, n = 25), S2 (moderate, n = 18), and S3 (severe, n = 20) were enrolled. Normal individuals (n = 25) were included (who were confirmed to have no metabolic-associated steatotic liver disease (MASLD) based on comprehensive diagnostic evaluations (imaging studies (ultrasound) and laboratory testings). All patients were clinically investigated in the Tropical Medicine Department at the Medical Research Centre of Excellence, from January 2024 to December 2024.

Inclusion Criteria: Adults range: 18–65 years with confirmed MASLD (via ultrasound), Concomitant atherosclerosis, Signed informed consent. Exclusion criteria: Other liver diseases (viral hepatitis), Malignancies or autoimmune conditions, inadequate or hemolyzed blood sample, Pregnancy or Lactation, Alcohol Abuse. At least one of the following metabolic conditions (BMI over 25 kg/m², hypertension, diabetes mellitus, elevated triglycerides, or reduced HDL cholesterol levels) was present in all patients¹³ In contrast, atherosclerosis was rigorously staged using carotid Doppler ultrasound—a powerful, non-invasive imaging technique that precisely measures Intima-Media Thickness (IMT) to critically assess the health of the carotid arteries. Atherosclerosis staging was conducted using carotid Doppler ultrasound, to assess Intima-Media Thickness (IMT), and the characterization of atherosclerotic plaques, with severity classification beginning from the from no plaque, minimal plaque, moderate plaque, and advanced plaque.³⁹ Blood samples were collected from all participants.

Plasma was collected after fasting according to each biochemical marker from all individuals to overcome variability. Clinical parameters were evaluated, including a full lipid profile (glucose levels, total cholesterol, triglycerides, high/low-density lipoprotein), as well as AST and ALT, have been recorded. Anthropometric assessment involved calculating BMI as weight (kg) divided by height (m²).

2.2. Evaluation of MMP-9 levels in serum among control subjects and patients with varying grades of MASLD (S1, S2, S3)

To evaluate MMP-9 protein levels, An MMP-9 ELISA kit was utilized (Elabscience, Cat# E-EL-H6075). According to the manufacturer's protocol. Briefly, serum samples were collected, diluted as needed, and added to microplate. After further washes,TMB substrate was added to trigger color development, followed by the addition of stop solution to halt the reaction. Absorbance was measured at 450 nm, and sample concentrations were calculated in ng/ml.

2.3. Staging of carotid atherosclerosis

Carotid Doppler ultrasonography was used to assess the presence and severity of carotid artery plaque in participants, conducted by a trained radiologist following standardized imaging protocols. During plaque diagnosis, some samples were missed due to insufficient imaging quality, resulting in only a subgroup of 64 patients being suitable for reliable analysis. Specifically, 22 patient samples were excluded due to inadequate image quality or because they were missed during the neurology/vascular imaging clinic visit. Plaque burden was categorized into four groups: (No plaque, n = 15), (minimal plaque, n = 32), (moderate plaque, n = 15) and (advanced plaque, n = 2), based on stenosis degree and plaque characteristics. These findings were correlated with MASLD grades assessed through abdominal ultrasound to explore the relationship between liver MASLD and carotid atherosclerosis.

2.4. The calculated scores

BAAT score was calculated by summing points assigned for BMI ≥ 28 kg/m², age ≥ 50 years, ALT ≥2 times the normal value, and triglycerides ≥1.7 mmol/L, with each criterion contributing 1 point.⁴⁴

BARD Score was calculated by [1 × BMI (kg/m2) + 1 × presence of diabetes +2 × AST/ALT ratio (>0.8)]. Values of the score lie between 0 and 4, with higher scores indicating an increased probability of having fibrosis. A score of 0 suggests absence of Fibrosis, 1 a low likelihood of fibrosis, a score of 2–3 indicates an intermediate likelihood, and a score of 4 points to a high likelihood of the condition.³²

NAFLD fibrosis score was calculated The score is calculated as follows:

The score is calculated as negative 1.675 plus 0.037 times age (years), plus 0.094 times BMI (kg/m²), plus 1.13 times impaired fasting glucose status (1 for yes, 0 for no), plus 0.99 times the AST/ALT ratio, minus 0.013 times platelet count (×10⁹/L), and minus 0.66 times albumin (g/dL).¹⁶ The score ranges from −2.0 to 6.0, with specific cut-off values indicating different fibrosis stages. Patients were stratified by NAFLD Fibrosis Score into low probability (score < −0.6) and high probability (score > −0.6) of advanced fibrosis, and further distributed according to steatosis grade (normal, S1, S2, or S3).

FIB-4 Score was calculated by [(Age × AST) / (Platelet count × √ALT)], where age is in years, AST /ALT in IU/L, and platelet count is expressed in ×10⁹/L. The FIB-4 score ranges from 0.01 to 6.25. A score 1.45 suggests lack of fibrosis, values between 1.45 and 3.25 indicate significant fibrosis, from 3.25 to 6.25 correspond to advanced fibrosis, and a score above 6.25 indicates cirrhosis.²

2.5. Statistical analyses

Analyses were conducted with SPSS software (version 26). Parametric tests (e.g., one-way ANOVA) were used for continuous variables, while non-parametric tests (e.g., Kruskal-Wallis) were applied for others based on data distribution characteristics. MMP-9 protein levels across groups were compared using ANOVA for MASLD grades (followed by Tukey's HSD post-hoc) and Kruskal-Wallis test for atherosclerosis stages (Dunn's post-hoc), respectively. Associations between MASLD grades and atherosclerosis stages, as well as MASLD grades versus computed scores (e.g., BARD and NAFLD Fibrosis score) were assessed through contingency table crosstabulation followed by Chi-square tests. Fisher's exact test (or Freeman-Halton extension) was used when group frequencies were low. Spearman's rank correlation was used to evaluate relationships involving categorical variables treated as ordinal (e.g., disease stages). ROC curve analysis was used to assess the accuracy of classification and diagnostic models. Differences with a p-value ≤0.05 were deemed significant.

3. Results

3.1. Demographic data of the studied participants

In our study, the cohort included individuals aged 18–65 years, encompassing naïve patients with different grades of MASLD (ranging from S1; mild, S2; moderate, S3; severe) were analyzed for various biomarkers. Clinical parameters were compared across the normal individuals group and the three MASLD grades.

Upon comparing clinical parameters among healthy individuals and MASLD patients ranging from S1 to S3, the groups were comparable in age and key metabolic confounders except BMI and TG. BMI increased notably from 25.20 ± 4.78 in healthy individuals to 38.40 ± 10.32 in S3 MASLD patients (P = 0.0001), reflecting a significant impact of obesity on advanced MASLD disease. Similarly, TG levels elevated significantly from 96.96 ± 42.20 in healthy individuals to 184.50 ± 122.09 in S3 MASLD patients (P = 0.001). Other clinical parameters, including HOMAIR, HB, TLC, PLT, ALT, AST, Albumin, Creat, Cholest, HDL, LDL, FBS, and HBA1C, did not show significant variations across the different MASLD grades. All results are shown in Table 1.

Table 1.

Comparison between the studied cohorts (Normal individuals and MASLD groups (S1, S2, S3)) in the following parameters.

Parameters	MASLD grades	Mean ± S.D.	p-value
BMI kg/m2	Normal	25.20 ± 4.78	0.0001
	S1	31.77 ± 5.29
	S2	36.20 ± 6.22
	S3	38.40 ± 10.32
HOMAIR	Normal	0.4 ± 0.01	0.547
	S1	1.64 ± 0.78
	S2	1.90 ± 1.25
	S3	2.28 ± 1.98
HB g/dL	Normal	13.15 ± 1.39	0.632
	S1	13.06 ± 1.98
	S2	13.25 ± 2.08
	S3	13.74 ± 1.39
TLC ×109/L	Normal	5.21 ± 1.38	0.271
	S1	5.30 ± 1.58
	S2	6.09 ± 2.51
	S3	5.98 ± 2.05
PLT ×109/L	Normal	239.71 ± 53.27	0.075
	S1	222.82 ± 61.23
	S2	240.06 ± 47.68
	S3	197.65 ± 57.07
ALT U/L	Normal	19.33 ± 10.90	0.209
	S1	27.91 ± 22.32
	S2	28.50 ± 18.79
	S3	27.12 ± 14.92
AST U/L	Normal	20.97 ± 8.57	0.210
	S1	26.50 ± 16.61
	S2	29.94 ± 21.41
	S3	28.76 ± 15.01
Albumin g/dL	Normal	4.34 ± 0.26	0.394
	S1	4.26 ± 0.27
	S2	4.27 ± 0.27
	S3	4.21 ± 0.25
Creat mg/dL	Normal	0.94 ± 0.32	0.643
	S1	0.93 ± 0.32
	S2	0.86 ± 0.19
	S3	0.85 ± 0.20
Cholest mg/dL	Normal	191.18 ± 47.41	0.244
	S1	211.65 ± 46.98
	S2	219.07 ± 47.61
	S3	208.71 ± 53.16
HDL mg/dL	Normal	48.46 ± 16.42	0.786
	S1	50.09 ± 23.76
	S2	48.50 ± 17.60
	S3	43.71 ± 24.44
LDL mg/dL	Normal	120.91 ± 38.92	0.388
	S1	130.65 ± 43.12
	S2	144.00 ± 42.41
	S3	136.33 ± 51.27
Triglyceride mg/dL	Normal	96.96 ± 42.20	0.001
	S1	155.42 ± 123.48
	S2	148.56 ± 52.37
	S3	184.50 ± 122.09
Fasting Blood Sugar mg/dL	Normal	94.36 ± 55.58	0.588
	S1	101.22 ± 48.54
	S2	119.69 ± 86.89
	S3	109.50 ± 43.10
Hemoglobin A1c %	Normal	5.63 ± 1.27	0.411
	S1	5.72 ± 1.10
	S2	6.21 ± 1.51
	S3	6.01 ± 1.08

Data are represented as mean ± S.D. BMI: Body Mass Index; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; HB: Hemoglobin; TLC/PLT: Total Leukocyte/Platelet Count; ALT/AST: Alanine/ Aspartate Aminotransferase; Creat: Serum Creatinine; Cholest: Total Cholesterol; HDL/LDL: High/Low-Density Lipoprotein.The number of patients included in each analysis reflects only those with complete data for the respective parameters.

3.2. The prevalence of MASLD severity with carotid atherosclerosis

The association between MASLD grades and carotid Doppler findings showed a statistically significant relationship. No individuals with a normal liver had advanced plaque. The analysis showed that the majority of advanced plaque has been observed in S1 and S3 MASLD patients. Statistical analysis Fisher freeman halton exact test, p = 0.016 confirmed an association between atherosclerosis and advanced MASLD grade. Results are depicted in Fig. 1.

Fig. 1.

Distribution of carotid plaque severity categories across normal individuals (Normals, n = 25) and MASLD patients with different grades (S1; mild, n = 25), (S2; moderate, n = 18), and (S3; severe, n = 20). Statistical significance was determined by Fisher freeman halton exact test (p = 0.016), indicating a link between atherosclerosis and increasing MASLD grade.’

3.3. Comparison of MMP-9 levels among control subjects and patients with varying grades of MASLD

The MMP-9 concentration was evaluated in the serum of the studied group compared to normal individuals group. The results were shown in Fig. 2.

Fig. 2.

Diagrammatic representation of the MMP-9 protein level in MASLD patients. ELISA assay was used to quantify MMP-9 protein level in serum (ng/ml) of normal individuals (Normals, n = 25) and MASLD patients with different grades (S1; mild, n = 25), S2 (S2; moderate, n = 18), and S3 (S3; severe, n = 20).

MMP-9 levels were compared across normal individuals and MASLD patients with different grades (S1, S2, and S3). The analysis revealed an overall significant variations in MMP-9 levels among the groups (p = 0.007), with higher mean levels observed in the S2 and S3 groups (8.6 and 8.2, respectively) relative to the normal individuals. Post hoc comparisons further indicated that the lowest concentration of MMP-9 levels has been detected in normal individuals compared to both S2 (p = 0.018) and S3 (p = 0.035), highlighting an association between MASLD progression and increased MMP-9 levels. ROC analysis confirmed MMP-9's lack of discriminative power across MASLD grades (AUC < 0.6, p > 0.05 for all comparisons).

3.4. Comparison of MMP-9 levels among control subjects and patients with different stages of Atherosclerosis

To estimate the level of MMP-9 concentration in different atherosclerosis patients' groups, the mean of protein level of MMP-9 in each group was analyzed by the Kruskal wallis test and the data revealed a significant difference between the studied cohort (p = 0.05).

Then the performance of Post hoc analyses using Tukey HSD method revealed the mean difference in MMP-9 levels was −9.17765 between “No plaque” and “Moderate plaque” (p = 0.014) and − 7.76187 between “Minimal Plaque” and “Advanced plaque” (p = 0.04), underscoring the relationship between the severity of carotid atherosclerosis and high MMP-9 levels, see Fig. 3.

Fig. 3.

Diagrammatic representation of the MMP-9 protein level in atherosclerosis patients with different stages assessed by carotid Doppler imaging. ELISA assay was performed to estimate the serum level of MMP-9 protein in atherosclerosis patients with different stages (No plaque, n = 15), (minimal plaque, n = 32), (moderate plaque, n = 15) and (advanced plaque, n = 2).

Moreover, the ROC analysis of MMP-9 protein in serum demonstrated its modest predictive ability (AUC = 0.903; p-value 0.019), in distinguishing advanced plaque than minimal and moderate plaque, but its usefulness for grading is limited without further validation. At a cutoff of 9.15, sensitivity is maximized to (1) with specificity at 0.833, achieving a high diagnostic performance. This finding highlights the prognostic utility of MMP-9 biomarker for clinical assessment, see Fig. 4.

Fig. 4.

ROC Curve analysis comparing advanced-stage atherosclerosis patients to those in earlier stages. AUC is 0.903, cutoff of 9.15, sensitivity is maximized to (1) with specificity at 0.833; p = 0.019.

3.5. Relationship between Scores across patients with varying MASLD grades; S1, S2, S3

In this study, the association between various computed scores and MASLD grading (normal, S1, S2, S3) was assessed. The results indicated that the BARD score and NAFLD Fibrosis Score showed significant associations with MASLD grades. Association analysis between MASLD grades and BARD score (varying from absence of fibrosis, low, intermediate, high likelihood of fibrosis) revealed a significant association, with majority of cohort S3 fell into high likelihood of Fibrosis. The distribution of patients in normal MASLD grade (absence n = 3, low n = 0, intermediate n = 9, high n = 6). While S1 (absence n = 2, low n = 3, intermediate n = 5, high n = 10). Also, S2 (absence n = 0, low n = 2, intermediate n = 2, high n = 9). Finally, S3 (absence n = 0, low n = 5, intermediate n = 2, high n = 8).

The majority of S2&S3 patients fall into the high likelihood of NASH group (p = 0.04). According to the NAFLD Fibrosis Score, with cut-off values of < −0.6 indicating low probability and > −0.6 indicating high probability of advanced fibrosis, the majority of the analyzed cohort (n = 50; 83.3%) fell into the low fibrosis risk group—distributed as normal (n = 17), S1 (n = 16), S2 (n = 10), and S3 (n = 7)—while a smaller group (n = 15; 16.7%) was classified in the high fibrosis risk group, distributed as normal (n = 1), S1 (n = 4), S2 (n = 3), and S3 (n = 8). The Chi-square test was significant (P = 0.014), indicating that the computed scores align well with our cohort's characteristics. In contrast, the BAAT and FIB-4 scores were not significantly associated, indicating potential limitations in their applicability within the studied cohort, as shown in Table 2.

Table 2.

Analysis of Predictive Scores against MASLD Grades.

Predictor	Formula	Grading category	N	P-value
BAAT44	(BMI ≥ 28) + (Age ≥ 50) + (ALT ≥2 N) + (TG ≥ 1.7 mmol/L)⁎	Low	14	0.3
		Moderate	49
		High	3
BARD32	(BMI ≥ 28) + (Age ≥ 50) + (Diabetes = 1 if yes, 0 if no) + (AST ≥ 2 N)	Absence of Fibrosis	5	0.014
		Low likelihood of Fibrosis	10
		Intermediate likelihood of Fibrosis	18
		High likelihood of Fibrosis	33
FIB-42	(Age × AST) / (Platelet count × √ALT)	No significant Fibrosis	51	0.25
		Intermediate	13
		Advanced Fibrosis	2
NAFLD Fibrosis Score16	1.18 × Metabolic Syndrome +0.45 × (Diabetes = 2 if yes, 0 if no) + 0.15 × FSI + 0.04 × AST – 0.94 × (AST/ALT) - 2.89	Low	50	0.014
		High	15

N refers to the number of patients in each group.

^⁎Each parameter is assigned 1 point if its corresponding condition is met: BMI of 28 kg/m² or higher, age 50 years or older, ALT levels at least twice the normal limit (≥ 2 N), and triglycerides equal to or exceeding 1.7 mmol/L.

3.6. Correlation analysis of scores across patients with varying MASLD grades; S1, S2, S3

Correlations between the MMP-9 concentration and scores were assessed. Spearman correlation analysis demonstrated a positive correlation among MMP-9 concentration with the progression of MASLD, which was treated as an ordinal variable (r = 0.4, p ≤ 0.01), as well as atherosclerosis progression (r = 0.3, p ≤ 0.05), and BAAT score (r = 0.254, p ≤ 0.05).

The NAFLD Fibrosis score exhibited a strong positive correlation with MASLD grades (r = 0.396, p ≤ 0.01) and FIB-4 (r = 0.726, p ≤ 0.01), and moderate correlations with atherosclerosis progression (r = 0.327, p ≤ 0.01) and BARD score (r = 0.460, p ≤ 0.01). Another significant relationship between carotid Doppler and MASLD grades (r = 0.430, p ≤ 0.01) as well as FIB-4 (r = 0.491, p ≤ 0.01) have been detected. BAAT showed significant positive correlations with FIB-4 (r = 0.520, p ≤ 0.01) and atherosclerosis stage (r = 0.385, p ≤ 0.01). BARD was significantly correlated with NAFLD Fibrosis score (r = 0.460, p ≤ 0.01) and FIB-4 (r = 0.245, p ≤ 0.05). A weak negative correlation was observed between BARD score and the ALT/AST ratio, which was not statistically significant (see Table 3). All categorical clinical variables (MASLD grade, atherosclerosis progression, BAAT, BARD, and FIB-4 scores) were treated as ordinal variables reflecting increasing severity or progression for the purpose of correlation analysis. A positive correlation was observed between MMP-9 concentration and BARD, BAAT, and atherosclerosis progression, suggesting that elevated MMP-9 levels are associated with greater liver disease severity and more advanced vascular remodeling in the studied cohort.

Table 3.

Correlation analysis between MMP-9 concentration and the computed scores.

Numbers represent Spearman's rank correlation coefficient (r). Statistical significance was denoted by p values, with p ≤ 0.05 indicated by a single asterisk (^⁎) and p ≤ 0.01 indicated by two asterisks (^⁎⁎).

4. Discussion

MASLD is considered one of the most significant challenges faced by non-Western populations, particularly among obese individuals. This challenge underscores the need for clinicians to validate non-invasive scores for effective screening, early diagnosis, and follow-up.9, 36 Additionally, it is essential to ensure that these tools are applicable for MASLD stratification across diverse populations and various countries.20, 33 In the current cohort, elevated clinical parameters included BMI, and TG levels, suggesting a direct link between MASLD progression, increased body fat, and TG levels. Similar to our finding, Nassir et al., reported the elevation of BMI and lipid profile.³⁰

MMP-9 is a significant biomarker in various medical conditions, particularly in inflammatory and cardiovascular diseases.21, 38 Our findings showed a significant association between increased MMP-9 concentrations and advanced MASLD grades (S2, S3). A western study identified MMP-9 as a predictive marker for the progression from MASLD to MASH and ending with cirrhosis and HCC.7, 43, 45, 46 Conversely, others reported that circulating MMP-9 levels were lower in advanced MASLD patients.16, 17

In steatotic hepatocytes, MMP 9 is rapidly released from neutrophil granules, while other cells produce it de novo in response to TNF-α, IL-1β, IFN-γ, or TGF-β by activating the NF-κB pathway.28, 43, 45 MMP-9 production, maintains inflammation, ECM remodeling, leukocyte infiltration, and anoikis resistance, that is the main way that MMP-9 contributes to the transition from MASLD to MASH. The systemic implications of MMP-9 in metabolic diseases and atherosclerosis are underscored by the observation that elevated MMP-9 degrades the vascular basement membrane and internal elastic lamina, thereby promoting leukocyte extravasation, endothelial barrier dysfunction and plaque instability.4, 31

Our results indicate significant associations and positive linear correlations between MASLD grades and the progression of atherosclerosis. Notably, MMP-9 highest concentrations were found in advanced atherosclerosis patients, supporting research that links elevated MMP-9 levels to plaque instability and an increased risk of cardiovascular events.²⁴ Additionally, ROC curve analysis suggests some clinical utility, but its usefulness for grading is limited without further validation. Increased MMP-9 contributes to brain extravasation and edema by compromising the blood-brain barrier and thinning the fibrous cap on plaques.14, 24, 47 The treatment strategy of aortic atherosclerotic plaques, MMP-9 is silenced to lower the levels of C-reactive protein, suggesting that MMP-9 deficiency could contribute to plaque stabilization through preventing inflammation.²⁴

Several computed scores have been validated to assess fibrosis and NASH risk across different population.20, 24, 27, 42 It is essential to compare the calculated scores with established diagnostic test results. The association between various computed scores and MASLD grading revealed a significant link between MASLD and both BARD and NAFLD Fibrosis score. The majority of low degree of MASLD were categorized into low risk for NASH and liver fibrosis. The lack of significant association between the MASLD grading and BAAT score in our cohort may stem from genetic variations among our cohort, which can lead to differences in disease manifestation not accurately captured by the score.

Up to our knowledge, this is the first study to investigate the role of MMP9 in MASLD patients with atherosclerosis. Hence, spearman analysis has been performed to explore the correlations between MMP9 and several computed scores in the studied cohort. The obtained results showed significant correlations between MMP-9, MASLD grades, atherosclerosis stages, and the BAAT score. The NAFLD Fibrosis score is significantly associated with MASLD grades, BARD, BAAT, and FIB-4 but not with MMP-9. In the British population, MMP-9 showed a significant negative correlation with FIB-4.¹⁶ Aligned with current results, the BAAT score effectively assessed NAFLD risk among populations such as Hispanics and Chinese. In Asian populations, particularly China, BAAT score predicts NAFLD risk in hypertensive patients and demonstrated strong utility for cardiovascular risk.42, 44 While the BAAT score is recognized for its ease of implementation, some critics argue that it may not adequately reflect the prevalence of NAFLD in specific populations, such as the Korean demographic.²⁰ Additionally, in populations with limited healthcare access, the BAAT score may be less reliable due to under diagnosis or late-stage presentations of MASLD.

The absence of the correlation significant between MASLD grades and both BARD and FIB4 may be attributed to the high prevalence of obesity in EGYPT.13, 19 The correlation analysis showed that the MASLD grades didn't correlate with BARD. This may be explained by the fact that the BARD score equation includes diabetes while many NASH patients are not diabetic.²⁶

All these findings underscored the importance of choosing appropriate scores for specific populations when aiming to understand the complexities of MASLD progression. A Caucasian study demonstrated that BARD index had a strong negative predictive value of advanced fibrosis in NAFLD patients.³³

Combining MMP-9 inhibitors with anti-inflammatory treatments may provide synergistic effects in reducing liver inflammation and fibrosis. By using viral vectors to deliver MMP-9 mutants or inhibitors, researchers aim to selectively target liver fibrosis and improve liver function.¹⁴ The limitations of our study were that it employed a single-center design, MASLD diagnosis was not established via biopsy, the sample size was relatively small, and we did not measure MASLD using the Controlled Attenuation Parameter.

5. Conclusion

MMP-9 is a probable biomarker for assessing MASLD and may improve cardiovascular risk evaluation. BARD and NAFLD Fibrosis scores effectively identified patients at risk for advanced MASLD in our populations. MMP-9 may be particularly useful when combined with the BAAT score, but its lack of correlation with other scores suggested it didn't align closely with their criteria. Targeting MMP-9 may also offer new therapeutic options for MASLD and atherosclerosis.

Ethics clearance

The study adhered to ethical standards and all participants provided informed consent prior to involvement, Approval was granted by the Institutional Ethical Review Board (Approval No. [20151]) following the principles outlined in the Helsinki Statement.

CRediT authorship contribution statement

Ghada M. Salum: Writing – original draft, Methodology, Data curation. Mai Abd el Meguid: Methodology, Data curation. Basma E. Fotouh: Methodology, Data curation. Mohamed Mokhles: Investigation. Reham Ibrahim siddik: Investigation, Data curation. Sherif Hassan Elwan: Investigation. Ahmed Saleh: Resources. Reham M. Dawood: Writing – review & editing, Supervision, Conceptualization.

Funding

The authors confirm that this study did not receive any financial support.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.