G Model

YDLD-3906; No. of Pages 7 ARTICLE IN PRESS

Digestive and Liver Disease xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Digestive and Liver Disease

journal homepage: www.elsevier.com/locate/dld

Liver, Pancreas and Biliary Tract

Long-term follow-up of children and young adults with autoimmune

hepatitis treated with cyclosporine
Silvia Nastasio a , Marco Sciveres b , Lorenza Matarazzo c , Cristina Malaventura d ,
Francesco Cirillo b , Silvia Riva b , Giuseppe Maggiore b,d,∗
a
Division of Gastroenterology, Hepatology, & Nutrition, Boston Children’s Hospital, Boston, MA, USA
b
Pediatric Hepatology and Liver Transplantation, ISMETT UPMC Palermo, Palermo, Italy
c
University of Trieste, Trieste, Italy
d
Section of Pediatrics, Department of Medical Sciences, University of Ferrara, University Hospital Arcispedale Sant’Anna, Ferrara, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Background: Cyclosporine (CSA) is an alternative treatment for autoimmune hepatitis (AIH), however, its
Received 13 February 2018 unknown long-term safety and efficacy have limited its use.
Received in revised form 22 October 2018 Aims: Examine the long-term outcome of children and young adults with AIH treated with CSA for at
Accepted 24 October 2018
least 4 years.
Available online xxx
Methods: Twenty patients were included in this retrospective study: 15 with classical AIH and 5 with
autoimmune hepatitis/autoimmune sclerosing cholangitis overlap syndrome (ASC). CSA was adminis-
Keywords:
tered as first (12 patients) or second-line (8 patients) treatment, alone or in combination with azathioprine
Autoimmune hepatitis
Autoimmune sclerosing cholangitis
or mycophenolate mofetil and/or prednisone.
Cyclosporine Results: CSA determined initial clinical and biochemical remission in all patients. At the end of follow-up
Glomerular filtration rate (median 8.6; range 4–20.4 years), all patients are alive with their native liver; 15 in complete remission
Overlap syndrome (75%), 2 with incomplete response to treatment and 3 listed for liver transplant. Side effects were mild and
transitory after dose tapering or, in 1 case, after CSA withdrawal. Hypertrichosis and moderate gingival
hyperplasia were the most frequent. Two patients presented mild transient glomerular filtration rate
(GFR) reduction. Median GFR at the beginning and end of treatment was not statistically different for all
patients.
Conclusions: CSA was effective and safe in the long-term treatment of our cohort of patients with AIH,
tailoring the treatment remains key-points during CSA administration.
© 2018 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

1. Introduction adolescent girls. Azathioprine, on the other hand, may cause bone
marrow suppression [3–5].
Autoimmune hepatitis (AIH) is a progressive inflammatory dis- For these reasons the need for alternative treatments has been
ease of the liver, which if untreated may lead to cirrhosis and recognized by the medical community. The use of cyclosporine
terminal liver failure [1,2]. AIH usually responds to immuno- (CSA) was first reported in 1985 in an adult with type 1 AIH, and
suppressive therapy, and the “so-called” conventional treatment subsequently in 1987 in a 14-year-old boy with the same disorder
consists of prednisone (PDN) and azathioprine (AZA). Although [6,7]. Since then, several reports have been published where CSA
this treatment induces clinical and biochemical remission in most has been shown to allow recovery from liver failure and shown to
cases, it is frequently associated with incomplete response and induce and safely maintain remission in children with AIH. How-
relapses. Moreover, corticosteroids may cause moderate to severe ever, since CSA has mainly been administered to induce remission
side effects, which lead to poor treatment compliance, especially in as a bridge to a conventional treatment, its long-term safety and
efficacy is unknown [8–11].
In 2004 we reported our experience with CSA in 12 patients
∗ Corresponding author at: Section of Pediatrics, Department of Medical Sciences,
with AIH who were followed-up for at least 18 months, and
Univeristy of Ferrara, University Hospital Arcispedale Sant’Anna, Via A. Moro, 8,
achieved complete remission without significant side effects [12].
44124 Cona, Ferrara, Italy. The present study aims to report on the long-term outcome of
E-mail address: [email protected] (G. Maggiore).

https://doi.org/10.1016/j.dld.2018.10.018
1590-8658/© 2018 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018
G Model
YDLD-3906; No. of Pages 7 ARTICLE IN PRESS
2 S. Nastasio et al. / Digestive and Liver Disease xxx (2018) xxx–xxx

patients who received a prolonged course of CSA. Disease remis- week during the second month, monthly for the following 3
sion, as well as the development of side effects, were analyzed months and every 3–6 months in the period after). Response
in order to assess CSA efficacy and safety in the treatment of AIH to treatment was defined as complete remission (disappearance
in children. The second end point of the study was to investigate of symptoms, normalization of aminotransferases and gamma
potential predictors of treatment outcome. glutamyl transpeptidase activities, normalization of IgG levels, neg-
ative or < 1:40 autoantibody titre), incomplete response (partial
improvement in clinical and laboratory features without normal-
2. Materials and methods
ization, with or without the need for second or third-line treatment)
or treatment failure (worsening clinical, laboratory, and histologi-
This observational study was conducted retrospectively by chart
cal features despite compliance with therapy). Relapse was defined
review, from 1993 to 2004, and prospectively, from 2004 to 2015.
as an increase in serum aminotransferases activity of greater than
The study was approved by the ethical committee of the Azienda
twice the upper limit of normal, with or without reappearance of
Ospedaliera Universitaria Pisana. Twenty patients with AIH who
symptoms, after a complete remission as defined above.
were treated with CSA for at least 4 years were included. Fifteen
Furthermore, in order to evaluate any adverse events related
patients were affected by “classical” autoimmune hepatitis and 5
to CSA treatment, serum creatinine levels and blood pressure,
patients by autoimmune hepatitis/autoimmune sclerosing cholan-
as well as the presence of gingival hyperplasia, hypertrichosis
gitis overlap syndrome (ASC).
and neurological signs, were routinely assessed in all patients.
Patients’ characteristics at diagnosis are summarized in Table 1.
To assess kidney function, glomerular filtration rate (GFR)
In all patients, any known cause of liver damage was excluded,
was estimated using the creatinine-based Schwartz equation
including viral infections by hepatotropic viruses (hepatitis A, B
in patients up to 18 years old and using the Chronic Kid-
and C virus, Epstein–Barr virus, Cytomegalovirus, Herpes sim-
ney Disease Epidemiology Collaboration creatinine equation
plex virus), ␣1-antitrypsin deficiency and Wilson’s disease, when
in patients older than 18 years. GFR was defined as normal
appropriate. Serum autoantibodies were detected using indirect
(≥90 mL/min/1.73 m2 ), mildly reduced (89–60 mL/min/1.73 m2 ),
immunofluorescence techniques. Liver biopsy was performed at
moderately reduced (59–30 mL/min/1.73 m2 ), severely decreased
diagnosis in all except 3 patients, who presented with a pro-
(29–15 mL/min/1.73 m2 ), or consistent with kidney failure
thrombin time (PT) <50%. Ishak scoring system was used for the
(<15 mL/min/1.73 m2 ).
assessment of necroinflammatory activity [13]. Biliary lesions were
graded as absent, mild, moderate or severe. Diagnosis of AIH was
made in 15 children according to the diagnostic criteria defined by
2.2. Assessment of predictive indicators of treatment outcome
the International Autoimmune Hepatitis Group [14], correspond-
ing in all patients with a definite diagnosis. The diagnosis of ASC
To identify potential indicators of treatment outcome, patients
was made in 5 patients with anti-nuclear and/or anti-smooth mus-
were divided in two groups. The group of responders consisted of
cle autoantibodies high titer reactivity, elevated immunoglobulin
patients who at the end of follow-up were in complete remission.
G (IgG) and histologic findings consistent with diffuse inflamma-
The group of non-responders comprised patients who, at the end of
tory bile duct damage as well as with interface hepatitis. Magnetic
follow-up, showed incomplete response to treatment, or showed
resonance cholangiography (MRC) and/or endoscopic retrograde
treatment failure. The prognostic value of 12 variables was analyzed
cholangiography (ERPC) were performed in 5 patients. In two of
[Table 3].
them signs of sclerosing cholangitis were present 8.8 and 6 years
after ASC diagnosis. Colonoscopy was performed in 4 of the 5 ASC
patients and showed findings consistent with ulcerative colitis in
2.3. Statistical analysis
one case.
For descriptive analysis, categorical data are presented as num-
2.1. CSA treatment bers and percentages; continuous data are presented as medians
and ranges. To assess predictive indicators of treatment outcome, a
CSA was administered either as first or second-line treatment. bivariate analysis was carried out in which each predictive variable
Indications for first-line treatment were the following: disease was explored separately with respect to treatment outcome. Differ-
severity (4), patient steroid refusal (1), and presence of relative con- ences were evaluated with the Fisher exact test for categorical data
traindication to steroid use (7) (age < 3 years, growth retardation, and with the non-parametric Mann–Whitney test for continuous
pubertal growth spurt in progress, overweight/obesity, candidia- data (a non-normal distribution of data was shown both visually
sis). CSA was used as a second-line treatment in patients who were and with the Kolmogorov–Smirnov test). To assess the long-term
non-responders (5) or intolerant to conventional treatment (3). safety of CSA, GFRs were compared using a non-parametric test for
Patients receiving second-line CSA treatment were all previously paired data (Wilcoxon test) at the beginning of CSA treatment and
treated with PDN and AZA/mycophenolate mofetil (MMF). at the end of follow-up for each patient. A double-sided p value of
Patients received CSA for a median time of 6.3 years (range <0.05 was considered statistically significant.
4–15.5), alone or in combination with AZA or MMF and/or pred-
nisone. Ursodeoxycholic acid was added in ASC patients [Table 2].
In all cases a written consent to CSA treatment was obtained. 3. Results
CSA was administered, in all cases, orally with a daily dosage
ranging between 1.5 and 8 mg/kg with the aim to obtain and main- 3.1. Response to treatment and outcome in 12 patients receiving
tain initial trough blood levels of 150–200 ng/mL. Once remission cyclosporine as first-line treatment
was achieved, CSA doses were reduced to maintain trough levels
between 100 and 150 ng/mL and, after 1 year of treatment, between CSA treatment was associated with a 100% initial clinical and
50 and 70 ng/mL. biochemical remission rate after a median period of 8.5 weeks
Once CSA treatment had been established, blood levels of (range: 4–36 weeks). No statistically significant difference was
aminotransferases, albumin, IgG, PT and cyclosporine were peri- found in the time to achieve remission in patients receiving CSA
odically monitored (weekly during the first month, every other as a first- and second-line treatment (P = 0.10).

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018
G Model
YDLD-3906; No. of Pages 7 ARTICLE IN PRESS
S. Nastasio et al. / Digestive and Liver Disease xxx (2018) xxx–xxx 3

Table 1
Patients’ characteristics at diagnosis.

AIH-1 AIH-2 ASC All

Number of patients 5 10 5 20
Female gender, n 3 9 3 15
Age at diagnosis [median, in years (range)] 10.5 (2.4–12.9) 3.7 (1.4–14.2) 10.1 (8.2–13.5) 9.5 (1.4–14.2)
Mode of presentation, n
• Acute hepatitis 1 3 0 4
• Insidious onset 0 0 2 2
• Incidental finding 4 7 3 14
Clinical findings, n
• Hepatomegaly 3 6 3 12
• Splenomegaly 2 2 1 5
Biochemistry
• ALT, xULN (median, range) 16 (7–25) 17 (5–73) 12 (10–20) 20 (5–73)
• GGT, xULN (median, range) 0 (0–2) 0 (0–2) 6 (2–10) 1.4 (0–10)
• PT <50%, n 2 2 2 6
• IgG, mg/dL (median, range) 3656 (1934–4430) 1860 (813–2200) 4000 (1730–4560) 2000 (813–4560)
Positive autoantibodies, n
• ANA 2 3 5
• SMA 1 1
• ANA + SMA 2 1 3
• LKM1 8 8
• LKM1 + LC1 1 1
• LKM1 + LC1 + ANA 1 1
• ANA + pANCA 1 1
Liver histology*
• Inflammatory activity Ishak Score (median, range) 12/18 (12–12) 12/18 (6–16) 6/18 (6–12) 12/18 (6–16)
• Fibrosis Ishak Score (median, range) 1/6 (0–2) 1/6 (0–5) 5/6 (4–6) 2/6 (0–6)
• Presence of biliary lesions, n of patients 2 0 5 7
Associated autoimmune disorders, n (%) 3 (60%) 5 (50%) 1 (20%) 9 (45%)
Family history of autoimmune disorders in first-degree relatives, n (%) 2 (40%) 5 (50%) 0 (0%) 7 (35%)

AIH-1: autoimmune hepatitis type 1, AIH-2: autoimmune hepatitis type 2, xULN: upper limit of normal, PT: prothrombin time, IgG: immunoglobulin G, ANA: antinuclear
antibodies, SMA: anti-smooth muscle antibody, LKM1: anti-liver kidney microsomal type 1 antibody, LC1: anti-liver cytosol type 1 antibody, pANCA: perinuclear anti-
neutrophil cytoplasmic antibodies.
*
In 3 patients (2 AIH-1, 1 ASC) with PT <50% liver biopsy was not performed at diagnosis.

Table 2
Cyclosporine treatment features of 20 patients with autoimmune hepatitis.

Drug combinations in • Cyclosporine (10)

patients receiving • Cyclosporine + azathioprine (1)
first-line cyclosporine • Cyclosporine + prednisone (1)
treatment (n = 12) • Ursodeoxycholic acid (3)a

Drug combinations in • Cyclosporine (2)

patients receiving • Cyclosporine + prednisone (1)
second-line • Cyclosporine + azathioprine + prednisone (4)
cyclosporine treatment • Cyclosporine + mycophenolate + prednisone (1)
(n = 8) • Ursodeoxycholic acid (2)a

Indications to • Relative controindication to steroid use (7)

cyclosporine first-line • Age <3 years
treatment (n) • Growth retardation
• Pubertal growth spurt in progress
• Overweight/obesity
• Candidiasis
• Patient steroid refusal (1)
• Disease severity (4)

Indications to • Conventional treatment failure (5)

cyclosporine • Steroid-related severe side effects (3)
second-line treatment • Obesity
(n) • Growth failure
• Osteoporosis

Age at the beginning of cyclosporine treatment median (range) 10.3 years (2.2–23.1 years)

Cyclosporine treatment duration, median (range) 6.3 years (4–15.5 years)

a
Ursodeoxycholic acid was added to patients with ASC.

At the end of a median follow-up of 8.6 years (range 4–20.4 successfully shifted to azathioprine monotherapy after receiving
years), all patients are alive with their native liver. One patient is cyclosporine treatment for a median time of 8.5 years (range 4–15.5
listed for liver transplant. years); (2) one patient has been receiving cyclosporine and aza-
thioprine for 4.6 years; (3) one patient received cyclosporine for
3.1.1. Complete remission at the end of follow-up (n = 10) 1.5 years before azathioprine was added to her treatment main-
Ten of 12 patients are in complete remission at the end of follow- taining a complete remission for 6.5 years. She then experienced
up (AIH-1, n = 2; AIH-2, n = 6; ASC n = 2): (1) seven patients were a relapse and shortly after was diagnosed with Systemic Lupus

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018
G Model
YDLD-3906; No. of Pages 7 ARTICLE IN PRESS
4 S. Nastasio et al. / Digestive and Liver Disease xxx (2018) xxx–xxx

Table 3
Analysis of prognostic indicators of treatment outcome in 20 children with autoimmune hepatitis.

Parameters Responder Non responder Total p

n = 15 n=5 N = 20

Sex, n (%) 1.00

M 4 (80.0%) 1 (20.0%) 5 (100.0%)
F 11 (73.3%) 4 (26.7%) 15 (100.0%)
Diagnosis, n (%) 0.56
AIH 12 (80.0%) 3 (20.0%) 15 (100.0%)
OVLS 3 (60.0%) 2 (40.0% 5 (100.0%)
Age at diagnosis, years 8.2 (1.4–14.2) 12.4 (8.7–13.2) – 0.12
Signs of cirrhosis at diagnosis, n (%) 0.03
Yes 1 (25.0%) 3 (75.0%) 4 (100.0%)
No 14 (87.5%) 2 (12.5%) 16 (100.0%)
Jaundice at diagnosis, n (%) 1.00
Yes 5 (71.4%) 2 (28.6%) 7 (100.0%)
No 10 (79.6%) 3 (20.4%) 13 (100.0%)
ALT (xULN) 20.0 (7.0–73.0) 10.0 (5.0–42.0) – 0.07
GGT (xULN) 1.0 (1.0–6.5) 2.5 (1.0–10.0) – 0.07
PT (%) 75.0 (32.0–114.0) 54.0 (34.0–88.0) – 0.25
Presence of autoimmune comorbidities, n (%) 0.13
Yes 5 (55.6%) 4 (44.4%) 9 (100.0%)
No 10 (90.9%) 1 (9.1%) 11 (100.0%)
Age at beginning CSA treatment, years 10.0 (2.2–23.1) 12.9 (9.9–18.8) – 0.08
Previous treatment before CSA, n (%) 0.35
Yes 5 (62.5%) 3 (37.5%) 8 (100.0%)
No 10 (83.3%) 2 (16.7%) 12 (100.0%)
Relapses during CSA treatment, n (%) 0.04
Yes 6 (54.5%) 5 (45.5%) 11 (100.0%)
No 9 (100%) 0 9 (100.0%)

Row percentages.
Continuous data are expressed as median and range.

Erythematosus and continued treatment with prednisone alone; At the end of a median follow-up of 8.6 years (range 4–20.4
(4) one patient has been in stable remission for 6 months without years) all patients are alive with their native liver. Two are listed
any immunosuppressive maintenance drug, after 7.4 years of treat- for liver transplant.
ment with cyclosporine and prednisone followed by azathioprine
monotherapy. 3.2.1. Complete remission at the end of follow-up (n = 5)
Four of the 10 patients had one or more episodes of relapse. Five of 8 patients are in complete remission at the end of a
follow-up (AIH-1, n = 2; AIH-2, n = 2; ASC, n = 1): (1) two patients are
3.1.2. Incomplete response to treatment at the end of follow-up on azathioprine monotherapy; (2) two patients are on cyclosporine
(n = 1) and azathioprine; (3) one patient is on triple immunosuppressive
One patient with AIH-1, liver cirrhosis, biochemical signs of liver therapy with cyclosporine, prednisone and MMF.
failure (PT 43%) and severe hypergammaglobulinemia was initially Two of the 5 patients had one or more episodes of relapse dur-
started on cyclosporine achieving first remission after 36 weeks. ing treatment. One patient relapsed 6 months after withdrawal of
Prednisone and azathioprine were added to her treatment after 2.6 immunosuppressive treatment; complete remission was achieved
and 7.9 years respectively due to failure to maintain a complete 4 weeks after cyclosporine administration was restarted.
response.
3.2.2. Incomplete response to treatment at the end of follow-up
(n = 1)
3.1.3. Treatment failure (n = 1) One patient with AIH-2 and celiac disease was initially started
In one patient with ASC, cyclosporine allowed for remission on prednisone and azathioprine; despite initial first remission,
lasting 2 years after which the patient relapsed. Subsequently he the patient relapsed and a repeat biopsy 2.3 years after diagno-
presented recurrent episodes of ascending bacterial cholangitis sis showed persistent inflammatory activity without worsening of
and 3.9 years after ASC diagnosis, ulcerative colitis was diag- fibrosis. Cyclosporine was added allowing achievement of remis-
nosed. MRC performed 2 years later demonstrated extrahepatic sion and tapering of the steroid. After 7.6 years from diagnosis,
bile duct changes characteristic of sclerosing cholangitis. Despite however, and despite switching azathioprine to MMF, the patient
triple immunosuppressive therapy with cyclosporine, prednisone shows incomplete response to triple immunosuppressive treat-
and azathioprine, liver disease progressed and enrollment in the ment.
liver transplant waiting list was necessary 6.3 years after diagnosis.

3.2.3. Treatment failure (n = 2)

3.2. Response to treatment and outcome in 8 patients receiving In two patients (AIH-2, n = 1; ASC, n = 1) with liver cirrhosis at
cyclosporine as second-line treatment diagnosis, cyclosporine was added to prednisone and azathioprine
after 1.8 and 6.3 years, respectively, due to incomplete response
CSA treatment was associated with initial clinical and biochem- and allowed for transient remission. In the patient with AIH-2,
ical remission in a median period of 6 weeks (range: 4–8 weeks) cyclosporine was replaced with tacrolimus after 5.7 years of ther-
in all patients but two who were shifted to CSA despite being in apy due to a reduction of the GFR. In the patient diagnosed with
remission with conventional treatment, because of severe obesity ASC, a MRC showed multifocal structuring and dilatation of intra-
and severe growth impairment. and extrahepatic bile ducts 8.8 after diagnosis. Liver disease pro-

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018
G Model
YDLD-3906; No. of Pages 7 ARTICLE IN PRESS
S. Nastasio et al. / Digestive and Liver Disease xxx (2018) xxx–xxx 5

Fig. 2. Renal function of 20 patients with autoimmune hepatitis before and after
cyclosporine treatment (median follow-up 8.6 years, range 4–20.4 years).

To assess the long-term safety of cyclosporine, GFR at the begin-

ning of cyclosporine treatment and at the end of follow-up of each
patient were compared. No statistically significant difference was
found (p = 0.41) [Fig. 2].
Neither arterial hypertension nor any other known CSA-related
side effects were registered throughout the entire follow-up period.

Fig. 1. Glomerular filtration rate and cyclosporine blood level relationship in two
patients who presented mild transient GFR reduction. 4. Discussion
(A) Patient 1, mild GFR reduction after 5.7 years of cyclosporine administration.
(B) Patient 2, mild GFR reduction after 1.7 years of cyclosporine administration.
This study is the first to report on the long-term efficacy and
safety of cyclosporine treatment in children and young adults with
gressed in both patients, and they were listed for liver transplant AIH.
11.2 and 14.4 years after diagnosis, respectively. Conventional treatment, has been preferred over the last 30
years [1–3,15]. Its well-known benefits, however, need to be care-
3.3. Prognostic indicators of treatment outcome fully weighed against the risks and flaws of such treatment. It is
reported that 69% to 94% of pediatric patients relapse after drug
Univariate analysis showed that 2 of the investigated variables withdrawal [16–18,28,29] and that 3% to 55% of children need to
were significantly associated with treatment outcome. Specifically, undergo liver transplant, despite treatment, with a recurrence rate
an unfavorable outcome (non-responder group) was associated with of de novo AIH in transplanted patients ranging from 38% to 83%
signs of cirrhosis at diagnosis (p = 0.03) and presence of relapses [2,16–18,28,29,31]. Furthermore, high dose steroids carry the risk
during CSA treatment (p = 0.04) [Table 3]. The limited sample size of severe side effects. This risk increases with repetitive treatment
and complete separation problems did not allow for development of multiple relapses. Such side effects are often poorly tolerated,
of a multivariate model. particularly by adolescent patients, and potentially cause scarce
treatment compliance [1,5,18,19]. Azathioprine monotherapy has
3.4. Cyclosporine safety been demonstrated to maintain remission in most patients with
AIH, although only in retrospective adult series, and its long-term
Hypertrichosis was observed in 9 of 20 patients (45%), it resolved use is potentially burdened by the risk of developing malignan-
in all patients either spontaneously or after cyclosporine tapering. cies. All of these are compelling reasons to refine current treatment
Eleven patients (55%) presented gingival hyperplasia which dis- strategies and pursue new management options [20–23].
appeared in all once cyclosporine dosage was reduced. Cyclosporine was initially used to prevent allograft rejection
A significant increase in serum creatinine levels up to 1.2 mg/dL after solid organ transplantation. Subsequently, it was also used for
and 0.96 mg/dL was observed in 2 patients (10%), respectively. the treatment of numerous autoimmune conditions such as uveitis,
In the first patient mild GFR reduction (76.5 mL/min/1.73 m2 ) rheumatoid arthritis, and psoriasis. The major CSA side effects are
occurred after 5.7 years of cyclosporine administration. nephrotoxicity, arterial hypertension and gastrointestinal and neu-
Cyclosporine, which at the time was administered at the dose of rological toxicity [24–27].
1.9 mg/kg/day, was therefore switched to tacrolimus with a subse- Recently Czaja reported that since 1985, CSA use has been doc-
quent persistent normalization of serum creatinine (0.89 mg/dL) umented in 10 reports including 133 adult patients with AIH. CSA
and GFR (103.2 mL/min/1.73 m2 ) after 2 months. In the second was used as a salvage therapy with overall positive response in
patient, mild GFR reduction (88.8 mL/min/1.73 m2 ) was observed about 93% of patients and a negative response in 7% [20].
after 1.7 years of cyclosporine administration. Cyclosporine dosage As for pediatric patients, there are only four major publications
was accordingly reduced from 3.7 to 2.8 mg/kg/day; serum cre- that include large series of patients with autoimmune liver dis-
atinine levels and GFR gradually returned in the normal range ease. The reported remission rate achieved with CSA administration
(0.85 mg/dL and 101.6 mL/min/1.73 m2 , respectively) within 8 ranges between 83% and 100% with the most commonly reported
months [Fig. 1]. side effects being hypertricosis and gingival hyperplasia. Transient

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018
G Model
YDLD-3906; No. of Pages 7 ARTICLE IN PRESS
6 S. Nastasio et al. / Digestive and Liver Disease xxx (2018) xxx–xxx

Table 4
Clinical features, outcome and side effects of cyclosporine treatment in selected studies featuring large series of patients with autoimmune liver disease.

Study Number of Diagnosis Median age at the Median duration % of remission Side effects Comments on
patients beginning at CSA of CSA treatment (% of side effects
treatment (years) (years) patients)

Alvarez et al. [14] 32 AIH-1 10.1 0.5 83% Mild hypertricosis Tolerance to CSA was very
AIH-2 (2–16.6) Moderate gingival good. No patient presented
hypertrophy with hypertension or renal
complications.

EASL [15] 15 AIH-2 Not reported Not reported 100% Mild Hypertricosis (80%) Renal function returned to
Mild gingival hypertrophy normal after CSA dose
(33%) reduction or discontinuation.
GFR reduction > 20% (20%) Transient mild hypertension
Hypertension (6%) occurred in a child with
Tremor (6%) associated autoimmune
Headache (13%) glomerulonephritis receiving
also prednisone. All other side
effects occurred within the first
6 months of treatment and
then disappeared.
Maggiore et al. 12 AIH-1 8 (3–12) 2.9 (0.7–7.4) 100% Hypertricosis (17%) Hypertricosis and gingival
[18] AIH-2 Gingival hypertrophy (25%) hypertrophy improved on
AC Creatinine elevation changing to conventional
GCH (8%) treatment; renal function
returned to normal after CSA
dose reduction.

Saadah et al. [16] 84 AIH-1 9.9 (2.3–16) 0.5 94% Mild hypertricosis (55%) Tolerance to CSA was
AIH-2 Mild gingival hypertrophy satisfactory. All side effects
(39%) were transient. Creatinine level
Mild creatinine elevation increase and hypertension
(10%) were detected in 1 occasion in
Mild hypertension (4%) 5 and 3 patients respectively
(during the first month of
treatment).
AIH-1: autoimmune hepatitis type-1, AIH-2: autoimmune hepatitis type-2, AC: autoimmune cholangitis, GCH: giant cell hepatitis, CSA: cyclosporine.

GFR reduction and creatinine elevation is reported in 8–20% of the response rate to second-line agents in pediatric refractory AIH.
patients in 3 of the studies [8–10,12] [Table 4]. The highest response rate at 6 months was estimated to be with
In all of these studies, however, CSA has mostly been adminis- cyclosporine (86%) followed by MMF (38%) [30].
tered for short periods. It is likely that the lack of data concerning In our study, only 5 patients received CSA as a second-line treat-
its long-term efficacy as well as its potential risks limited its use to ment, however despite the small number of patients, the follow-up
short-term initial therapy in alternative to steroid-azathioprine or was prolonged and our results support the finding that CSA has a
to salvage treatment. high-response rate as a second-line treatment, even well beyond
This study was therefore undertaken to evaluate the long-term the 6 month period.
efficacy and safety of CSA in patients with AIH. Our patients were In our cohort of patients, there were no deaths. Three patients
treated with CSA for a median period of 6.3 years (range: 4–15.5 are on transplant waiting list 6.3, 11.2 and 14.4 years after diag-
years), the longest reported in literature to our knowledge. nosis, respectively. These three presented with more severe and
Our results confirm the efficacy of CSA in treatment-naïve aggressive features than the others. Two of them had ASC, which
patients with AIH. Specifically, the efficacy of CSA in the induction was associated with cirrhosis at diagnosis in one. Both of these pre-
and maintenance of clinical and biochemical remission in AIH was sented with sclerosing cholangitis 8.8 and 6 years after diagnosis,
shown to be equal to that of conventional treatment reported in respectively. The third patient presented initially with cirrhosis, PT
other studies [1,2,29]. <50% and portal hypertension, and was diagnosed with AIH-2 as
Initial clinical and biochemical remission was indeed achieved well as with multiple extra-hepatic autoimmune disorders.
in all of our patients; independent of the degree of hepatic impair- The association with inflammatory bowel disease is a well-
ment, CSA was equally effective in inducing remission both as first- known feature of autoimmune liver disorders and is more
and as second-line treatment. commonly reported in patients with ASC (45%) than with AIH-1
CSA also determined long-term sustained clinical and biochem- (20%). In our study population, only one of 5 patients with ASC was
ical remission in more than 80% of the treatment naïve patients, diagnosed with ulcerative colitis and he was one of the patients
alone or in combination with azathioprine or prednisone. Shift to listed for liver transplant.
AZA monotherapy was achieved in 60% of patients and even after The analysis of prognostic indicators showed that cirrhosis at
CSA withdrawal no relapse occurred, suggesting that CSA lead to a diagnosis and relapses during treatment were associated with an
complete and persistent resolution of the liver inflammatory pro- unfavorable outcome. The limited sample size and complete sepa-
cess. In 1 patient (8%), treatment was stopped and she has currently ration problems, however, did not allow for the development of a
been out of therapy for the past 6 months. multivariate model, withholding the possibility to draw any defini-
In patients who received CSA as second-line treatment, CSA tive conclusion.
allowed complete prolonged remission in 62.5% of cases, alone or The frequency of relapses during CSA treatment in our study
in combination with azathioprine and/or prednisone and/or MMF. appeared to be equal to that which was reported in other studies
Recently, a meta-analysis by Zizzo et al. compared the short-term with conventional treatment [1,2,29].

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018
G Model
YDLD-3906; No. of Pages 7 ARTICLE IN PRESS
S. Nastasio et al. / Digestive and Liver Disease xxx (2018) xxx–xxx 7

Cyclosporine treatment was very well tolerated, as observed in [4] Murray-Lyon IM, Stern RB, Williams R. Controlled trial of prednisone and aza-
previous studies [8–12]; however both CSA treatment duration and thioprine in active chronic hepatitis. Lancet 1973;1:735–7.
[5] Maggiore G, Bernard O, Hadchouel M, Alagille D. Life-saving immunosup-
follow-up were significantly longer in this study. pressive treatment in severe autoimmune chronic active hepatitis. J Pediatr
CSA side effects were mild and transitory in all patients Gastroenterol Nutr 1985;4:655–8.
either spontaneously, after dose tapering or, in 1 case, after CSA [6] Mistilis SP, Vickers CR, Darroch MH, McCarthy SW. Cyclosporin, a new treat-
ment for autoimmune chronic active hepatitis. Med J Aust 1985;143:463–5.
withdrawal. Transient hypertrichosis and moderate gingival hyper- [7] Hyams JS, Ballow M, Leichtner AM. Cyclosporine treatment of autoimmune
plasia were the most frequent side effects. Mild GFR reduction with chronic active hepatitis. Gastroenterology 1987;93:890–3.
return to normal renal function occurred in two patients. Median [8] Alvarez F, Ciocca M, Cañero-Velasco C, Ramonet M, de Davila MT, Cuarterolo
M, et al. Short-term cyclosporine induces a remission of autoimmune hepatitis
GFR was not statistically different at the beginning and at the end
in children. Hepatology 1999;30:222–7.
of treatment for all patients, and no patients presented with hyper- [9] Debray D, Maggiore G, Girardet JP, Mallet E, Bernard O. Efficacy of cyclosporin
tension. A in children with type 2 autoimmune hepatitis. J Pediatr 1999;135:111–4.
[10] Cuarterolo M, Ciocca M, Velasco CC, Ramonet M, González T, López S, et al.
Severe and permanently disabling side effects such as the ones
Follow-up of children with autoimmune hepatitis treated with cyclosporine. J
seen with prolonged steroid treatment are therefore not present Pediatr Gastroenterol Nutr 2006;43:635–9.
even after prolonged CSA administration. [11] Cuarterolo ML, Ciocca ME, López SI, de Dávila MT, Alvarez F. Immunosuppres-
Satisfactory CSA tolerability in our patients seems to be related sive therapy allows recovery from liver failure in children with autoimmune
hepatitis. Clin Gastroenterol Hepatol 2011;9:145–9.
to the administration of a low CSA dose regimen and to its strict [12] Sciveres M, Caprai S, Palla G, Ughi C, Maggiore G. Effectiveness and safety of
blood level monitoring. ciclosporin as therapy for autoimmune diseases of the liver in children and
This study presents several limitations, the main ones being the adolescents. Aliment Pharmacol Ther 2004;19:209–17.
[13] Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, et al. Histological
relatively small number of patients, the heterogeneity of diagno- grading and staging of chronic hepatitis. J Hepatol 1995;22:696–9.
sis (AIH-1, AIH-2, ASC) and CSA treatment (first- or second-line) [14] Alvarez F, Berg PA, Bianchi FB, Bianchi L, Burroughs AK, Cancado EL, et al. Inter-
and the study design (partially retrospective and partially prospec- national Autoimmune Hepatitis Group Report: review of criteria for diagnosis
of autoimmune hepatitis. J Hepatol 1999;31:929–38.
tive). Moreover the presence of co-medication constitutes another [15] European Association for the Study of the Liver. EASL clinical practice guide-
potential confounder. lines: autoimmune hepatitis. J Hepatol 2015;63:971–1004.
Despite these limitations, the major strength of this study is [16] Saadah OI, Smith AL, Hardikar W. Long-term outcome of autoimmune hepatitis
in children. J Gastroenterol Hepatol 2001;16:1297–302.
that it is the first one to describe the long-term outcome of CSA
[17] Radhakrishnan KR, Alkhouri N, Worley S, Arrigain S, Hupertz V, Kay M, et al.
treatment in children and young adults with autoimmune hepati- Autoimmune hepatitis in children-impact of cirrhosis at presentation on nat-
tis, reporting data on CSA efficacy and safety acquired over more ural history and long-term outcome. Dig Liver Dis 2010;42:724–8.
[18] Maggiore G, Veber F, Bernard O, Hadchouel M, Homberg JC, Alvarez F, et al.
than 20 years of experience.
Autoimmune hepatitis associated with anti-actin antibodies in children and
adolescents. J Pediatr Gastroenterol Nutr 1993;17:376–81.
5. Conclusions [19] Maggiore G, Bernard O, Homberg JC, Hadchouel M, Alvarez F, Hadchouel P,
et al. Liver disease associated with anti-liver-kidney microsome antibody in
children. J Pediatr 1986;108:399–404.
In conclusion, CSA is an effective and safe long-term treatment [20] Czaja AJ. Advances in the current treatment of autoimmune hepatitis. Dig Dis
for children and young adults with AIH constituting a viable alter- Sci 2012;57:1996–2010.
native to conventional treatment. CSA administration should not [21] Van Thiel DH, Wright H, Carroll P, Abu-Elmagd K, Rodriguez-Rilo H, McMichael
J, et al. Tacrolimus: a potential new treatment for autoimmune chronic
be limited to a bridge to steroid-azathioprine treatment or to a active hepatitis: results of an open-label preliminary trial. Am J Gastroenterol
salvage therapy, but could also be considered as a first-line and 1995;90:771–6.
maintenance treatment. [22] Burak KW, Urbanski SJ, Swain MG. Successful treatment of refractory type 1
AIH with methotrexate. J Hepatol 1998;29:990–3.
Monitoring the occurrence of potential adverse effects, specifi- [23] Richardson PD, James PD, Ryder SD. Mycophenolate mofetil for maintenance
cally the evaluation of renal function, and tailoring the treatment of remission in autoimmune hepatitis in patients resistant to or intolerant of
remain key points during CSA administration. azathioprine. J Hepatol 2000;33:371–5.
[24] Minuk GY. Cyclosporin A in non-transplant-related liver diseases. Am J Gas-
troenterol 1989;84:1345–50.
Conflict of interest [25] Treem W, Hyams JS. Cyclosporine therapy for gastrointestinal disease. J Pediatr
None declared. Gastroenterol Nutr 1994;18:270–8.
[26] Mattos AM, Olyaei AJ, Bennett WM. Nephrotoxicity of immunosuppressive
drugs: long-term consequences and challenges for the future. Am J Kidney Dis
Acknowledgements 2000;35:333–46.
[27] Choudhry S, Bagga A, Hari P, Sharma S, Kalaivani M, Dinda A. Efficacy and safety
The authors would like to thank Dr. Luca Ronfani for assistance of tacrolimus versus cyclosporine in children with steroid-resistant nephrotic
syndrome: a randomized controlled trial. Am J Kidney Dis 2009;53:760–9.
with statistical analysis. [28] Jiménez-Rivera C, Ling SC, Ahmed N, Yap J, Aglipay M, Barrowman N,
et al. Incidence and characteristics of autoimmune hepatitis. Pediatrics
References 2015;136:e1237–48.
[29] Lee WS, Lum SH, Lim CB, Chong SY, Khoh KM, Ng RT, et al. Characteris-
tics and outcome of autoimmune liver disease in Asian children. Hepatol Int
[1] Maggiore G, Nastasio S, Sciveres M. Juvenile autoimmune hepatitis: spectrum 2015;9:292–302.
of the disease. World J Hepatol 2014;6:464–76. [30] Zizzo AN, Valentino PL, Shah PS, Kamath BM. Second-line agents in pediatric
[2] Gregorio GV, Portmann B, Reid F, Donaldson PT, Doherty DG, McCartney M, patients with autoimmune hepatitis: a systematic review and meta-analysis. J
et al. Autoimmune hepatitis in childhood: a 20-year experience. Hepatology Pediatr Gastroenterol Nutr 2017;65:6–15.
1997;25:541–7. [31] Mieli-Vergani G, Vergani D, Baumann U, Czubkowski P, Debray D, Dezsofi
[3] Maggiore G, Bernard O, Hadchouel M, Hadchouel P, Odievre M, Alagille D. A, et al. Diagnosis and management of pediatric autoimmune liver disease:
Treatment of autoimmune chronic active hepatitis in childhood. J Pediatr ESPGHAN hepatology committee position statement. J Pediatr Gastroenterol
1984;104:839–44. Nutr 2018;66:345–60.

Please cite this article in press as: Nastasio S, et al. Long-term follow-up of children and young adults with autoimmune hepatitis treated
with cyclosporine. Dig Liver Dis (2018), https://doi.org/10.1016/j.dld.2018.10.018