Reuters Health Information: New score may have prognostic value in severe alcoholic hepatitis
New score may have prognostic value in severe alcoholic hepatitis
Last Updated: 2017-11-30
By Marilynn Larkin
NEW YORK (Reuters Health) - Combining gene expression signature data with the model for end-stage liver disease (MELD) score may predict survival of patients with severe alcoholic hepatitis, researchers say.
"Survival prediction of severe alcoholic hepatitis is an urgent unmet medical need given the considerable side effects of the standard of care, steroids, which is not effective in a significant fraction of the patients," Dr. Yujin Hoshida of the Icahn School of Medicine at Mount Sinai told Reuters Health.
"The currently available score predicts patient prognosis only seven days after starting steroid treatment, and its predictive performance is suboptimal," he said by email. "Our score enables prognostic prediction before steroid initiation, with substantially improved performance."
Dr. Hoshida and colleagues developed a scoring system - gs-MELD - that integrates gene expression signature (gs) data with the MELD score to better predict patient survival.
The researchers collected fixed liver biopsy samples from 71 patients (median age, 53; 62% male) diagnosed with severe alcoholic hepatitis treated in Belgium. For this derivation cohort, they then identified gene expression signatures and clinical data associated with survival without liver transplantation at 90 and 180 days after corticosteroid therapy initiation.
Findings were validated using liver biopsies from two validation cohorts: 48 patients (median age, 53; 52% male) in Belgium and Switzerland (cohort 1) and 20 patients (median age, 44; 40% male) in the U.S. (cohort 2).
As reported in Gastroenterology, online November 20, data on expression patterns of 123 genes were integrated with MELD scores to assign patients in the derivation cohort to groups with poor survival (29% survived 90 days; 26% survived 180 days) and good survival (76% survived 90 days; 65% survived 180 days).
In validation cohort 1, the gs-MELD score discriminated patients with poor survival from those with good survival at both 90 days (43% vs. 96%) and 180 days (34% vs. 84%). In this cohort, the gs-MELD score outperformed other clinical models to predict survival of patients with severe alcoholic hepatitis, according to the authors.
In validation cohort 2, the score discriminated patients with poor survival from those with good survival at 90 days (24% vs. 100%) and 180 days (12% vs. 100%).
Dr. Yoshida noted, "The gene signature is already implemented in an FDA-approved, clinically applicable assay, and the rest of the parameters in the score are readily available in routine clinical practice.
However, "the current gene signature assay requires liver biopsy tissue, which may not be routinely obtained especially at non-referral centers for alcoholic liver disease," he said. "We are now developing a blood-based surrogate marker of the gene signature for even wider clinical applicability."
Dr. Sander Florman, who is also at Mount Sinai but was not involved in the study, told Reuters Health, "The gene signature data combined with MELD data for these patients with severe alcoholic hepatitis is interesting and merits further investigation."
"These findings will hopefully lead to the development of serum assays as biopsy is not readily available everywhere and can be hazardous," he said by email.
For now, the findings "may have significant clinical applicability, especially when biopsy is available, and may allow us to stratify these patients and their care based upon who is more or less likely to survive."