Positron emission tomography combined with serum biomarkers detects fibrotic MASH – Scientific Reports

Our study demonstrates the effectiveness of using dynamic FDG-PET/CT imaging combined with serum-based test FIB-4 as a non-invasive tool to detect fibrotic MASH, complementing studies that use ultrasound or MRI15,20,21. Newsome et al. evaluated the FibroScan-AST (FAST) score in identifying MASH patients that combined liver stiffness measurement (LSM) and controlled attenuation parameter (CAP) from the FibroScan® with the lab test aspartate aminotransferase (AST): alanine aminotransferase (AST:ALT) ratio20. The FAST test showed a sensitivity of 48%. Jung et al. found that MRI-based MR elastography (MRE) alone demonstrated clinically significant diagnostic accuracy for the detection of fibrotic MASH15. However, when MRE was combined with FIB-4 together, Jung et al. found it to have a higher AUROC with a positive predictive value (PPV) of 97.1% but a sensitivity of 56%15. While Truong et al. combined AST with MRI sores for steatosis (MR-PDFF) and fibrosis (MR elastography) to give the MAST score that also was effective in delineating fibrotic MASH21. MAST also fared well as a prognostic tool for predicting major adverse liver outcomes (MALO)21. A critical finding from these studies is that combining serum markers with the imaging biomarkers for MASH improved accuracy and enabled a more holistic approach to diagnosis and risk stratification. Our triple-variate model enabled combination of all the available imaging and biochemical markers relevant to the PET imaging space. Our model complements the other non-invasive methods and provides a sensitive tool for determining fibrotic MASH with a sensitivity of 83%. We will be collecting longer-term data to elucidate the ability of this tool for predicting major adverse liver outcomes (MALO) and extrahepatic outcomes (namely, cardiovascular, renal etc.).

The addition of the FIB-4 index creates a novel triple-variate model that detects a wider range of MASH severity in patients’ diagnoses with adequate sensitivity. This can serve as a noninvasive substitute for liver biopsies and circumvent the use of invasive methods. Furthermore, because the main cause of morbidity in MASH patients remains cardiovascular disease, FDG-PET, which has been used for clinical cardiac imaging, can enable taking a more comprehensive approach in predicting comorbidities affecting the heart or kidney22,23,24. The new PET/CT model will not only detect MASH, but it will also detect increased risk of myocardial impairment via identification of focal myocardial 18F-FDG reuptake patterns and assess renal function using multiple positrons emitting radiolabeled racers22,23. Lastly, a critical functionality of FDG-PET is in cancer imaging. Pre-determination of liver disease stage in such patients undergoing PET imaging will enable personalizing therapies based on risk of liver toxicity25. Increasing the reliability of the PET/CT model will provide physicians with a clinical advantage to address both MASH symptoms and extrahepatic disease and risk factors. We envision the utilization of this technology in a variety of settings. In an initial phase it can have significant applicability where FDG-PET remains prevalent, such as in oncologic settings. It can provide diagnostic capability to determine fibrotic MASH in patients to help choose treatment regimen based on liver risk stratification. For general clinical purposes, we do not foresee our method replacing easily available point-of-care imaging tools such as transient elastography. Detailed imaging modalities such as MRE/MR-PDFF will continue to play a niche role for clinical trials or difficult to diagnose patients. With the significant progress in the PET imaging space especially with newer tracers (e.g. FAPI26) and whole body or total-body imaging, we envision our method will have an impact in determining liver disease in the context of systemic diseases such as diabetes and obesity. As the armamentarium of MASH therapies increase especially with treatments that target multiple systems (e.g. Tirzepatide, Semaglutide), it will become increasingly essential to have a tool that can concomitantly evaluate systemic disease. Annual monitoring of response to treatment, not just from the liver standpoint (steatosis, inflammation, and fibrosis) but also overall cardiac and renal disease24 will be essential.

The current methodology is intended to establish a technology that can be applied to a widely prevalent disease. Currently, a 60 min dynamic scan would largely be restricted to clinical trial settings in academic medical centers. Advances in technology including improved tracer data capture, and analysis along with machine learning applications that is enabling substantial decrease in dynamic scan protocol to obtain comparable data. Thus, we envision a near future where dynamic PET protocol can be achieved in ~ 15 min that can capture data with high sensitivity relevant for both oncology and metabolic applications. This will make it attractive and relevant to Onco-PET practices and patient care. Although radiation exposure is a risk, effective dose remains low and comparable to abdominal CT<a data-track="click" data-track-action="reference anchor" data-track-label="link" data-test="citation-ref" aria-label="Reference 27" title="Radiological Society of North America, I. R. Radiation Dose. (2024).” href=”#ref-CR27″ id=”ref-link-section-d212782785e1267″>27. The total effective radiation dose from this PET/CT scan is lower than that of a clinical PET/CT and is accepted to be below the levels thought to result in a significant risk of harmful effects<a data-track="click" data-track-action="reference anchor" data-track-label="link" data-test="citation-ref" aria-label="Reference 27" title="Radiological Society of North America, I. R. Radiation Dose. (2024).” href=”#ref-CR27″ id=”ref-link-section-d212782785e1271″>27.

Some of the limitations of this study include its small sample size and localization to a single center. Adapting this tool in larger and diverse cohorts will enable establishing this method for determination of clinical changes in MASLD patients. The triple variate model predicted fibrotic MASH with a sensitivity of 83% and specificity of 64%. Liver biopsies are the gold standard, and our triple variate model is not intended as a substitute but rather an improved non-invasive method relative to existing non-invasive methods. With liver biopsies, we can encounter sampling errors, intra-and inter-observer variability28; high cost; and adverse effects such as pain, risk of infection, bleeding, perforation, and though rare, potentially death29, can deter patients from getting screened for MASH.

In conclusion, the triple variate model provides effective determination of fibrotic MASH, enabling utilization of PET/CT for larger scale studies in MASLD patients.