Multiparametric MRI liver biomarkers predict those ‘likely to achieve’ MASH resolution
Click Here to Manage Email Alerts
Key takeaways:
- Treatment responders demonstrated significant reductions in iron corrected T1 and liver fat content.
- Optimal thresholds of –82 ms for cT1 and a relative reduction of 58% for liver fat content were identified.
Changes in MRI-derived markers of iron-corrected T1 mapping and liver fat content appear effective at predicting histological response among patients treated for metabolic dysfunction-associated steatohepatitis, according to researchers.
“Efforts to identify alternative, noninvasive robust biomarkers for detecting change following treatment, and as surrogate endpoints, have been strongly encouraged by regulatory bodies such as the United States Food and Drug Administration,” Naim Alkhouri, MD, FAASLD, chief medical officer, chief of transplant hepatology and director of the fatty liver program at Arizona Liver Health, and colleagues wrote in Journal of Hepatology. “MRI proton density fat fraction (PDFF), also known as liver fat content (LFC), is a reliable and accurate measure of liver fat, and several studies have examined the association of change in LFC with histological response.”
They continued: “The longitudinal relaxation time (T1-mapping) from MRI may offer another approach to characterize the health of the liver.”
In a retrospective, pooled analysis of three interventional phase 2 trials, Alkhouri and colleagues investigated the association between changes in iron corrected T1 (cT1) and LFC and histologic improvement among 150 patients (mean age, 57 years; 68% women) with MASH. They also evaluated outcomes among treatment responders, defined as steatohepatitis resolution with no worsening of fibrosis, compared with non-responders.
Patients underwent biopsies and abdominal multiparametric MRI with the LiverMultiScan (Perspectum), which reported cT1, PDFF and T2 mapping, at baseline and at the end of each study. Evaluated thresholds included a 30% relative reduction in LFC and an 80-ms absolute reduction in cT1.
Both cT1 and LFC correlated only with steatosis at baseline (R = 0.5; P < .001), with changes in these biomarkers linked to changes in steatosis, inflammation and ballooning.
Compared with non-responders, patients who achieved MASH resolution (n = 22) experienced significant reductions in cT1 (–119 vs. –49 ms; P < .001) and LFC (–65% vs. –29%; P < .001). The area under the curve for both markers was 0.72 (cT1: 95% CI, 0.6-0.85; LFC: 95% CI, 0.59-0.85).
In addition, the Youden’s index, or optimal threshold, to distinguish responders from non-responders was –82 ms (OR = 5.4; 95% CI, 1.9-17) for cT1 and a relative reduction of 58% (OR = 6.8; 95% CI, 2.4-20.7) for LFC.
Further, patients with a cT1 reduction of at least 80 ms were five times more likely to achieve histological response (sensitivity = 0.68; specificity = 0.7), while those with a 30% reduction in liver fat were four times more likely (sensitivity = 0.77; specificity = 0.53).
“In this assessment of potential thresholds for change in multiparametric MRI using LFC and cT1, our results agree with our previous cross-sectional modelling and suggest that both imaging biomarkers will be effective at identifying those who are likely to achieve resolution of steatohepatitis,” Alkhouri and colleagues wrote. “A relative reduction in LFC of less than 30% is a sensitive threshold to rule out likelihood of histological response, but a reduction in cT1 of greater than 80 ms is a more specific rule in threshold for likelihood of steatohepatitis resolution.”
Collapse
Click Here to Manage Email Alerts