S2439 Association of Liver Enzymes With 18FDG PET/CT Uptake in Sarcoidosis Patients

Introduction: Hepatic involvement is a common sarcoidosis manifestation with clinically significant abnormalities, including portal hypertension and cirrhosis, seen in up to 25% of patients. However, despite this high prevalence, there are unclear guidelines on interpreting liver function tests (LFTs) and 18FDG PET/CT studies in hepatic sarcoidosis. In this pilot study, we assessed the predictive value of LFTs as biomarkers of hepatic sarcoidosis based on 18FDG liver uptake. We hypothesized that LFT abnormalities would correlate with hepatic 18FDG uptake when using a standardized blood pool (aorta) rather than liver as a reference background and could serve as a screening tool for hepatic sarcoidosis. Methods: We retrospectively abstracted clinical and radiologic data on biopsy-confirmed sarcoidosis patients between January 1, 2015 and December 31, 2020. In total, 48 out of 288 patients with available 18FDG PET/CT studies were included and randomized into exploratory (n = 24) and validation (n = 24) cohorts. Patients with incomplete data or liver diseases other than sarcoidosis were excluded. LIFEx radiomic software (v7.8.0) was used to analyze 18FDG PET/CT studies, interpreting standardized uptake values (SUV) from the liver and aorta. Statistical analyses were performed using R and GraphPad. Results: In the exploratory cohort, a significant relationship was noted within groups between alkaline phosphatase (ALP), but not alanine aminotransferase or aspartate aminotransferase, with SUV ratios. Specifically, those with elevated ALP levels had increased SUV ratios, while those with normal ALP had significantly lower SUV ratios (Mann-Whitney U test P < 0.005). Receiver operating characteristic analysis determined an optimal threshold ALP value (Y-Index) at 86.5 U/L, which effectively stratified patients with elevated hepatic 18FDG uptake. Application of the Y-Index threshold value for ALP to the validation cohort demonstrated a statistically significant difference in the SUV ratios between the normal and elevated ALP cohorts. Conclusion: Although further investigation is warranted, our study demonstrates an association between ALP levels and hepatic 18FDG uptake using a standardized blood pool reference with the LIFEx radiomic software. Our findings suggest that ALP, when assessed using a modified threshold, may serve as a clinically relevant biomarker for identifying hepatic involvement in sarcoidosis. Recognition of elevated ALP on routine laboratory testing could prompt earlier imaging and/or biopsy for confirmation of hepatic sarcoidosis.