Liver injury after pulsed methylprednisolone therapy in multiple sclerosis patients

Abstract Objectives High‐dose pulsed methylprednisolone‐related liver injury cases have been reported in the literature, but a prospective study in patients with multiple sclerosis (MS) has never been performed. The aim of this study was to evaluate the prevalence and severity of liver injury in patients with MS after pulsed methylprednisolone therapy. Methods We performed a prospective observational single‐center study on patients with MS treated with i.v. methylprednisolone 1,000 mg/day for 5 days. We tested the liver functionality before and 2 weeks after the treatment. In case of severe liver injury, defined according to “Hy's law,” a comprehensive hepatologic workup was performed. Results During a 12‐month observation period, we collected data on 251 cycles of i.v. steroid treatment of 175 patients with MS. After excluding eight cycles presenting a basal alteration of the biochemical liver tests, we observed a prevalence of 8.6% of liver injury in MS patients treated with pulsed methylprednisolone for clinical and neuroradiological relapses. In 2.5% of the patients, the liver injury was severe according to Hy's law; after a comprehensive hepatologic workup, three of them received a diagnosis of drug‐induced liver injury and the other three of autoimmune hepatitis. Conclusions Liver injury after pulsed methylprednisolone therapy in patients with MS is not infrequent, and a close monitoring of aminotransferase level before treatment and 2 weeks later seems advisable.

Most of MS exacerbations are usually followed by a period of repair leading to a clinical remission, and in some cases, to a complete recovery (Berkovich, 2013). However, the residual deficits after a MS relapse may persist contributing to the stepwise progression of disability; therefore, treating MS relapses is important as it may help shortening the disability associated with their course as well as impacting subsequent disease progression (Lublin, Baier, & Cutter, 2003). Corticosteroids are the main pharmacological treatment of MS relapses with a demonstrated effect, eventually supplemented by a plasmapheresis cycle (3-5 treatments) in nonresponders patients (Galea, Ward-Abel, & Heesen, 2015). A high-dose short-term methylprednisolone therapy is the currently recommended first-line treatment, acting by reducing the inflammation in the brain and the spinal cord, thus increasing the probability of ameliorating the clinical episode speeding up the patient recovery (Citterio et al., 2000;Sellebjerg et al., 2005).

| ME THODS
We performed a prospective observational study on adult patients with MS (defined according to the 2010 revisions of McDonald criteria) (Polman et al., 2010) treated with methylprednisolone pulses for clinical and neuroradiological relapses (including disease onset) from April 2014 to April 2015 at the multiple sclerosis center of the "Fondazione Policlinico Universitario Agostino Gemelli," Catholic University of Sacred Heart, in Rome, Italy. All the patients with MS who experienced a relapse and with no contraindications to methylprednisolone treatment have been enrolled. All patients received i.v. methylprednisolone at the dosage of 1,000 mg/day for 5 days. Biochemical liver function tests, namely aspartate aminotransferase (AST, normal value 7-45 IU/L), alanine aminotransferase (ALT, 7-45 IU/L), total bilirubin (normal value 0.3-1.2 mg/dl), alkaline phosphatase (ALP, normal value 40-129 IU/L), were performed in our laboratory before treatment and 2 weeks after the end of therapy. In case of abnormality, liver function tests were repeated every 2 weeks until normalization.

| Definitions
Liver injury was defined as elevation of ALT (with or without abnormalities of the other liver function tests) above the upper normal limit of our local laboratory (45 UI/L). Causality of ALT elevation was assessed through the Naranjo scale (Naranjo et al., 1981) and the World Health Organization (WHO) -Uppsala Monitoring Centre (UMC) causality assessment.
In case of a severe liver injury, defined according to "Hy's law" (hepatocellular injury of any grade associate with bilirubin increase) (Navarro & Senior, 2006), a comprehensive hepatological workup was performed. A careful medical history was collected, including previous exposure to methylprednisolone, concomitant diseases, eating habits, alcohol consumption, eventual consumption of mushrooms, and a detailed history on the use of prescribed and nonpre- antibodies), and metabolic screening (serum ferritin, copper, ceruloplasmin, and alfa-1-antitrypsin level). The patients underwent a liver ultrasound examination in order to exclude biliary abnormality, arterial thrombosis or Budd-Chiari syndrome. When clinically appropriate, a liver biopsy was performed and, in selected cases, repeated.
Liver injury was classified as hepatocellular, cholestatic, or mixed according to R-ratio score (R = (ALT value/ALT upper normal limit)/(ALP value/ALP upper normal limit); briefly, R ratios of >5 define a hepatocellular pattern of liver injury, <2 a cholestatic pattern of liver injury and between 2 and 5 a mixed pattern of liver injury (Chalasani et al., 2014).
The probability of a drug-induced liver injury (DILI) was assessed by the Roussel UCLAF Causality Assessment Method (RUCAM) that depends on the value of R-ratio and assigns a score to factors such as time to onset, course, risk factors, concomitant drugs, nondrug causes of liver injury, previous information on the hepatotoxicity of the drug and response to rechallenge; briefly, RUCAM indicate that a drug is a possible (Berkovich, 2013;Galea et al., 2015;Lublin et al., 2003), probable (Citterio et al., 2000;Schäcke et al., 2002;Sellebjerg et al., 2005), or highly probable (>8) cause of liver injury .

| Ethics statement
The local ethic committee approved the study, and all patients signed an informed consent.

| Statistical analyses
Independent-samples Student's t test and chi-squared test were used for group comparisons, as appropriate. Mean ± SD was given for continuous measurements. Frequencies and percentiles were given for categorical data. Binary logistic regression was used to evaluate the risk factors associated with liver injury. Differences were reported as statistically significant if the p-value was less than .05. Statistical analysis was performed using SPSS version 19.0 (SPSS Inc., Chicago, IL).

| Study population
In this study, we included one hundred and seventy five patients treated with pulsed methylprednisolone therapy for a clinical or neuroradiological relapse during 12 months of observation period. The baseline clinical characteristics of the patients are shown in Table 1.
The majority of the patients were women (65,1%) with a mean age of 40.8 ± 12.2 years; most of them (73.7%) presented a relapsing-remitting form of MS with a mean duration of disease of 9.7 ± 8.4 years and a mean EDSS score of 2.8 ± 2.1; 61.7% of patients were receiving a disease-modifying drug (DMD).

| Drug causality assessment in individual cases
Causality assessment for every case of ALT elevation 2 weeks after methylprednisolone therapy (without basal ALT alteration) was reported as probable according to both Naranjo scale and WHO-UMC causality assessment (Table 3).

| Clinical characteristics of patients with severe liver injury
Patient 1 was a 31-year-old woman, treated with glatiramer acetate, who presented an increase in the AST (49 IU/L) and ALT (110 IU/L) values 2 weeks after the pulsed methylprednisolone therapy.
Biochemical liver tests progressively increased, but she remained asymptomatic. Glatiramer acetate was stopped, and ademetio- hepatitis with centrolobular necrosis, consistent with a diagnosis of probable autoimmune hepatitis (IAHG score 13). The patient was treated with ursodeoxycholic acid and started a DMD with glatiramer acetate. Six months later, liver function tests were normal.

| Univariate analysis and binary logistic regression analysis for factors associated with DILI outcomes
Univariate analysis (Table 1)  To further investigate the risk factors associated with liver injury, a multivariate binary logistic regression analysis was performed, including age, sex, smoking habit, disease duration, clinical form of MS, and disease-modifying treatment. No positive or negative association was found between these variables and risk of liver injury.

| D ISCUSS I ON
Our study showed a prevalence of 8.6% of liver injury in patients affected by MS and treated with pulsed methylprednisolone for clinical and neuroradiological relapses. In 2.5% of patients, the liver injury was severe (according to Hy's law) and attributable either to a drug-induced liver injury or to the onset of an autoimmune hepatitis.
To the best of our knowledge, this is the first prospective observation study on methylprednisolone hepatotoxicity in a population pulse therapy for Graves' ophthalmopathy, but they included many patients with chronic HBV or HCV hepatitis (Eguchi et al., 2015).
We analyzed separately the population of patients with and without liver injury in order to find relevant risk factors for liver injury. Both populations were in fact quite similar for most of the clinical features, but we were not able to identify any risk or protective factor also because of the limited number of events observed.
In case of severe methylprednisolone-related liver injury, a careful hepatic workup is warranted in order to accurately attribute the responsibility of the toxic effect to the drug rather than to the an ex novo onset of autoimmune hepatitis (Carrier et al., 2013;Maàmouri et al., 2009;Reuβ et al., 2007;Salvi et al., 2004;Takahashi et al., 2008). In our cohort, three patients received a final diagnosis of probable or definite autoimmune hepatitis (patients 2, 3, and 5) and their baseline clinical characteristics were almost indistinguishable from those of patients who received a final diagnosis of DILI.. The diagnostic challenge is made even more difficult by the fact that there is a continuous spectrum crossing the diagnosis of hepatoxicity with immunoallergic mechanism with autoimmune hepatitis due to immune rebound phenomenon (De Boer et al., 2017). The distinction between these two pathological entities has an important therapeutic relevance for the neurologist. In case of methylprednisolone-induced liver injury, the neurologist has to take this into account to set the goal of reducing the clinical disease activity as much as possible. In fact, as the typically relapsing course of MS carries the need to repeat pulsed methylprednisolone treatment several times, in patient with high disease activity an early treatment with a second-line DMD such as natalizumab or fingolimod should be considered to more effectively reduce disease activity. The use of alemtuzumab in these patients may be limited by the need to coadminister methylprednisolone during the infusion cycle.
In case of severe relapse, a rescue therapy with plasma exchange should be considered in patients with methylprednisolone-induced liver injury.
In case of autoimmune hepatitis, instead, a concomitant treatment with azathioprine represents an important therapeutic limitation to the use of all second-line DMD. In addition, interferon beta is contraindicated in autoimmune hepatitis because of the possible worsening of liver disease activity. Treatment with glatiramer acetate is an option. There are no available safety data on the association of azathioprine with dimethyl-fumarate, neither on the use of teriflunomide in autoimmune hepatitis instead of azathioprine.
Our data highlight the importance of a close follow-up of the liver function tests following pulsed methylprednisolone treatment.
According to our experience, we suggest to test transaminases 2 weeks after pulsed methylprednisolone treatment in all patients, and particularly, before the beginning of a new DMD therapy, in order to avoid a misdiagnosis of DMD-induced liver injury. In case of abnormality, the start of the DMD therapy should be postponed and biochemical liver tests have to be repeated until normalization, while it is advisable to refer the patient to a hepatologist in case of bilirubin elevation.
In conclusion, our study confirms that liver injury after pulsed methylprednisolone therapy in patients with MS is not infrequent.
A complete baseline assessment of the liver function and a close monitoring after each cycle of high-doses methylprednisolone is recommended. In our experience, 2 weeks later seems to be the right interval for such monitoring.

ACK N OWLED G M ENTS
The authors have nothing to disclose.

CO N FLI C T O F I NTE R E S T
None to be declared.