Presence and high titers of antinuclear antibodies do not correlate with clinical severity nor outcome in patients with drug-induced liver injury

Background A proportion of patients with drug-induced liver injury (DILI) present with autoantibodies. This has led to the current concept that an autoimmune-like subtype of DILI exits, especially in the setting of certain drug intake, e.g. minocycline and nitrofurantoin [1-3]. However, there is no standardized definition for autoimmune-like DILI and the clinical relevance of autoantibodies in DILI has not been studied in detail yet. We hypothesized that autoantibodies might merely be an epiphenomenon, which is also encountered with culprit drugs that have not been associated with autoimmune-like DILI. To this end we analyzed the rate of autoantibody positivity in a cohort of 144 causality proven DILI patients with a broad spectrum of causative agents.

Methods 144 patients included in the ongoing study on the effects of potentially hepatotoxic drugs (NCT02353455) with DILI were used for the current analysis. Blood samples for monocyte-derived hepatocyte-like (MH) cell testing were acquired from all patients involved. MH cell generation and testing were performed as described previously [4, 5]. The DILI diagnosis was based on the MH cell test results and supported by Roussel Uclaf Causality Assessment Method (RUCAM) and expert adjudication. Testing for antinuclear antibodies (ANA) was performed by the standards of our hospital’s Institute for Laboratory Medicine using immunofluorescence. Titers ≥ 1:100 were considered positive and ≥ 1:400 clinically significant.

Results 67 % of the patients exhibited ANA titers ≥ 1:100; 28 % of the patients presented with ANA titers ≥ 1:400. There was no significant correlation between the ANA titers and the class of the causative drug. Regarding laboratory parameters, there was no difference in the maximum serum levels of ALT, AST, ALP, TB and INR between patients with high and non-detectable or low ANA titers, while ANA titers ≥ 1:400 correlated with higher levels of AST and TB at onset. Conversely, the proportion of Hy’s law positivity was significantly higher in patients with ANA ≥ 1:400, however the severity of liver injury and the rate of acute liver failure was similar between both groups. Moreover, significantly more patients with high ANA titers were treated with corticosteroids (34 % vs. 16 %, p = 0.01). Yet, the outcome of the patients was not different in both groups, with similar proportions developing chronic liver injury, undergoing transplantation or having a fatal outcome.

Conclusion A closer look in a causality proven DILI cohort proved that there is no evidence that presence of ANA is specific for DILI caused by a certain type of medication. Further, our findings indicate that high ANA titers are not associated with the severity of the DILI episode nor with patient outcome.

Tab. 1 Clinical characteristics of DILI patients with moderate to high ANA titers (≥1:400) and undetectable or low ANA titers (<1:400).

	ANA ≥ 1:400 (41; 28 %)	ANA < 1:400 (103; 72 %)	p
Age [ 1 ]	53 (18-79)	50 (1-83)	0.45
BMI [ 1 ]	24.8 (18.1-42.5)	23.4 (15.6-34.5)	0.16
Female sex [ 2 ]	23 (56 %)	59 (57 %)	0.90
AMA positive [ 2 ]	3 (7 %)	14 (14 %)	0.28
Other Autoantibodies [ 2 ]	8 (20 %) 3 ASMA 3 centromer B 1 dsDNA 1 NUMA	7 (7 %) 7 ASMA	0.02 [*]
Histology [ 2 ]	20 (49 %)	39 (38 %)	0.23
Type of injury [ 2 ] Hepatocellular Mixed Cholestatic	32 (78 %) 3 (7 %) 6 (15 %)	69 (67 %) 14 (14 %) 20 (19 %)	0.39
Hy’s Law criteria [ 2 ]	25 (61 %)	44 (43 %)	0.05 [*]
Acute Liver failure [ 2 ]	11 (27 %)	24 (23 %)	0.62
Severity [ 2 ] Mild (Grade 1) Moderate (Grade 2) Severe (Grade 3) Fatal or Transplantation (Grade 4)	8 (20 %) 22 (54 %) 8 (20 %) 3 (7 %)	26 (25 %) 54 (52 %) 9 (9 %) 13 (13 %)	0.23
Corticosteroid treatment [ 2 ]	14 (34 %)	16 (16 %)	0.01 [*]
Outcome [ 2 ] Remission Chronicity Death or transplantation	35 (85 %) 3 (7 %) 3 (7 %)	84 (82 %) 6 (6 %) 13 (13 %)	0.73

^{1 1} median (range)

^{2 2} n (%)

^* *statistically significant (p ≤ 0.05)

Publication History

Article published online:
10 June 2021