In patients who have chronic hepatitis C and undergo liver biopsy, the presence of Mallory-Denk bodies – hepatocyte cytoplasmic inclusions that occur in several chronic liver diseases – is independently associated with progression of liver fibrosis, Dr. Mina O. Rakoski and her colleagues reported in the October issue of Clinical Gastroenterology and Hepatology.
In addition, patients in whom serial liver biopsies show an increase in the number of Mallory-Denk bodies over time are more likely to have clinical decompensation and progression to cirrhosis than patients who have no Mallory-Denk bodies or who have a stable or decreasing number of them over time.
Little is known about Mallory-Denk bodies, and it is still unclear whether they "represent a benign epiphenomenon of hepatocyte injury" or are actual modifiers of disease progression. Their major constituents are keratin polypeptides 8 and 18, which "likely play an essential cytoprotective role in the liver," said Dr. Rakoski of the University of Michigan, Ann Arbor, and her associates.
In mice, gender and genetic background play critical roles in the formation of Mallory-Denk bodies. In humans, genes that encode keratin, including KRT8 and KRT18, have been associated with susceptibility to end-stage liver disease, increased fibrosis in chronic hepatitis C, and increased severity of primary biliary cirrhosis, they noted.
To explore the potential prognostic value of Mallory-Denk bodies in biopsy samples, Dr. Rakoski and her colleagues analyzed data from the HALT-C (Hepatitis C Antiviral Long-Term Treatment Against Cirrhosis) trial. This was a multicenter prospective randomized controlled trial involving 1,050 patients with chronic hepatitis C and advanced fibrosis or cirrhosis.
The HALT-C study subjects underwent liver biopsy at baseline, 18 months, and 3 years, and were followed for a median of 6 years to track their clinical and histologic outcomes. The presence or absence of Mallory-Denk bodies was recorded by expert pathologists reviewing the biopsy samples.
A total of 158 subjects (15%) had Mallory-Denk bodies present in baseline biopsy samples. Their presence was associated with laboratory markers of severe disease, including low platelets, low albumin, high AFP, and high AST/ALT ratio. It also was associated with histologic markers of severe disease, including greater periportal fibrosis, greater pericellular fibrosis, steatosis, and higher inflammation scores.
A subset of 844 HALT-C patients was studied longitudinally, and 719 of these patients showed no Mallory-Denk bodies on baseline biopsy. In all, 61 of these subjects (8.5%) did show Mallory-Denk bodies on repeat biopsy ("MDB gain").
MDB gain was significantly associated with increased fibrosis and steatosis on repeat biopsy, as well as with diabetes, female gender, and Hispanic ethnicity. The association with gender and ethnicity suggest that genetic factors play an important but as yet unknown role in MDB formation.
Of 125 patients who had Mallory-Denk bodies on baseline biopsy, 101 (81%) showed fewer inclusions on repeat biopsy ("MDB loss"). This loss was associated with a lower BMI, less baseline fibrosis, the absence of diabetes, and the absence of smoking.
It is unknown why some patients showed MDB loss over time, nor why these patients did not show improved outcomes. It is possible that the loss of MDB actually reflected liver sampling errors. It also is possible that some unknown environmental or genetic factor caused the resolution of the inclusions but did not impact overall outcomes, Dr. Rakoski and her colleagues said.
In a subset of 58 patients with MDB gain over time, half developed an adverse clinical outcome. In contrast, only 15% of subjects who did not have MDB developed an adverse clinical outcome, a significant difference.
By comparison, in a subset of 88 patients with MDB loss over time, 23% developed an adverse clinical outcome, whereas 32% of those who did not have MDB loss developed an adverse clinical outcome – a nonsignificant difference. Thus, MDB loss was not associated with either good or adverse clinical outcomes.
Histologic outcomes also were assessed in a subset of 447 patients in the longitudinal analysis. In all, 67% of those who showed MDB gain over time developed an adverse histologic outcome, compared with only 28% of patients who did not show MDB gain over time. This difference was highly significant.
This study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute of Allergy and Infectious Diseases, the National Cancer Institute, the National Center for Minority Health and Health Disparities, the National Center for Research Resources, and Hoffmann-La Roche (now Genentech). The investigators reported no conflicts of interest.