- Email: [email protected]
Rapidity of treatment response and outcome in type 1 autoimmune hepatitis
Journal of Hepatology 51 (2009) 161–167 www.elsevier.com/locate/jhep
Rapidity of treatment response and outcome in type 1 autoimmune hepatitisq Albert J. Czaja* Division of Gastroenterology and Hepatology, Mayo Clinic College of Medicine, 200 First Street S.W., Rochester, MN 55905, USA
Background/Aims: Corticosteroid therapy is effective in type 1 autoimmune hepatitis. This study determines if the rapidity of response affects outcome. Methods: The duration of treatment to improve clinical and laboratory indices was determined retrospectively in 146 patients and correlated with outcome. Results: Sixteen patients (11%) responded after 66 months, and 20 patients did so after 36 months (14%). The rapid responders were older than those who responded slowly (54 ± 3 years versus 41 ± 4 years, P = 0.007), and they had a lower frequency of HLA DRB1*03 (36% versus 76%, P = 0.03). Patients aged P60 years responded within 6 months more commonly than adults aged <40 years (18% versus 2%, P = 0.02), and most had responded within 24 months (94% versus 64%, P = 0.003). Elderly patients who responded quickly had HLA DRB1*04 more commonly than adults aged <40 years (75% versus 8%, P = 0.0001). Progression to cirrhosis (18% versus 54%, P = 0.03) and liver transplantation (2% versus 15%, P < 0.05) was less common in the rapid responders. The frequencies of a sustained remission (19% versus 10%, P = 0.6) and relapse after drug withdrawal (81% versus 90%, P = 0.6) were similar between patients with an early and late response. Conclusions: A rapid treatment response decreases progression to cirrhosis and liver transplantation, but it does not alter the frequency of sustained remission or relapse after drug withdrawal. Elderly patients respond more quickly to treatment than adults aged <40 years, and they are characterized by HLA DRB1*04. Ó 2009 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Corticosteroid therapy; Progression to cirrhosis; Response rate
1. Introduction Corticosteroid therapy ameliorates symptoms, resolves laboratory abnormalities, improves histological Received 16 December 2008; received in revised form 3 February 2009; accepted 23 February 2009; available online 10 April 2009 Associate Editor: V. Barnaba q The author declared that he does not have anything to disclose regarding funding or conflict of interest with respect to this manuscript. * Tel.: +1 507 284 8118; fax: +1 507 284 0538. E-mail address: [email protected] Abbreviations: AJC, Albert J. Czaja; ANA, antinuclear antibodies; AST, aspartate aminotransferase; HLA, human leukocyte antigen; LKM1, liver-kidney microsome type 1; SMA, smooth muscle antibodies; ULN, upper limit of normal range.
changes, reduces or prevents hepatic fibrosis, and enhances survival in most patients with autoimmune hepatitis [1–4]. Treatment until normal laboratory indices and normal liver tissue is reached is the ideal end point of treatment [5–8], but it is not achievable in all patients [8]. Furthermore, its relentless pursuit may be associated with drug-related side effects [5]. Treatment until improvement in the serum aspartate aminotransferase (AST) level to less than twice the upper limit of the normal range (UNL) in conjunction with normalization of serum bilirubin and c-globulin abnormalities and disappearance of interface hepatitis from the liver tissue is achievable in most patients and is well tolerated [1,9,10]. This result has been associated with the reduction and prevention of hepatic fibrosis [4],
0168-8278/$36.00 Ó 2009 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2009.02.026
162
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
improved long-term survival [1], and a sustainable remission in 50% of patients [9,10] that may be long-term in 20% [10]. It constitutes the minimum level of improvement that has been associated with a satisfactory outcome after drug withdrawal. The rapidity of a treatment response may have prognostic implications that reflect the severity of the disease, the genetic predispositions of the host, and the metabolism of the medication in individual patients [9,10]. In chronic hepatitis C, the disappearance of viremia within 12 weeks after institution of anti-viral therapy has been associated with a sustained virological response, and this realization has allowed the individualization of antiviral regimens [11]. Patients who respond rapidly do not require protracted anti-viral therapy, whereas those who respond slowly or not at all can be identified quickly and considered for alternative treatments. In autoimmune hepatitis, the association between the rapidity of the treatment response and the outcome of the disease is unknown, yet it may be an important discriminator between those who will survive without consequences and those who will progress to cirrhosis and liver failure [12]. Treatments typically average 2 years [5,8–10,12,13], but they range from 6 months [1,5] to life-long [14,15]. Progression of hepatic fibrosis in autoimmune hepatitis can be rapid, and it is fueled by inflammatory activity [16]. Corticosteroids can suppress this activity [4] and prevent stellate cell activation or promote degradation of the already established fibrotic matrix [17]. Quick suppression of this inflammatory activity may identify those individuals who will not progress. In this retrospective analysis of prospectively acquired data, the patients who respond rapidly (66 months) and those who respond slowly (>36 months) to conventional corticosteroid therapy are characterized, and the effect of rapid improvement on the frequency of relapse after drug withdrawal, progression to cirrhosis, and occurrence of death from liver failure or requirement for liver transplantation is assessed. In this fashion, the study determines if the rapidity of response to a standard treatment has prognostic value and if the rapidity of response can be used to identify problematic patients early who may need close surveillance and individualized therapy. Furthermore, the characterization of rapid responders may affect the indications and enthusiasm for their treatment.
2. Patients and methods 2.1. Study population One hundred and forty-six patients who satisfied international pretreatment criteria for the diagnosis of autoimmune hepatitis [18] and who had been evaluated and treated in accordance with a pre-established protocol [19] constituted the study population. These patients had been selected from 275 similar patients because they had
responded to their first course of corticosteroid therapy, and they had been withdrawn from medication. The 129 patients not included in the analysis included 28 patients who had failed to improve during treatment (10%), 98 patients who required continued treatment or who had developed drug intolerances that warranted adjustments in their management strategy (36%), and 3 patients who had not undergone liver tissue examinations at presentation (1%). The investigation had been approved by the Institutional Review Board of the Mayo Clinic. One hundred and eighteen patients (81%) were women, and the mean age of the population was 47 ± 1 years (range, 13–82 years; median, 49 years). One hundred and thirty patients (89%) were classified pre-treatment as having definite autoimmune hepatitis, and 16 patients (11%) were classified pre-treatment as having probable autoimmune hepatitis by the criteria of the International Autoimmune Hepatitis Group [18]. The mean pre-treatment diagnostic score was 18.3 ± 0.2 points (range, 11–22 points; median, 19 points). All patients had been tested for antinuclear antibodies (ANA) and smooth muscle antibodies (SMA), and 127 patients (87%) had been tested and negative for antibodies to liver kidney microsome type 1 (anti-LKM1) by previously published methods [4,8,10,12,16]. The serological findings justified the designation of type 1 autoimmune hepatitis [20].
2.2. Clinical and histological assessments Clinical examinations and conventional laboratory tests of liver inflammation and function had been performed at presentation and at each 6-month follow-up interval by one clinician (A.J.C.) in accordance with a previously published protocol [19]. Liver tissue examinations had been performed at accession in all patients and immediately prior to drug withdrawal in 127 patients (87%). Pre-established criteria were applied in all histological examinations, and the diagnosis of cirrhosis required demonstration of fibrosis and a complete regenerative nodule [19,21]. Our previous validation studies indicated that the consistency of pathological interpretation for grade and stage of disease was 90% and 78%, respectively [5,12,21,22].
2.3. Treatment regimens All patients had been treated with either prednisone alone (45 patients) or a lower dose of prednisone in conjunction with a fixed dose of azathioprine (101 patients) in accordance with schedules published previously [19,23]. Each regimen had been shown to be comparable to each other [24]. Prednisone alone (60 mg daily) or a lower dose of prednisone (30 mg daily) in conjunction with azathioprine (50 mg daily) was started as an induction regimen and continued for one week [19,23]. The dose of prednisone was then reduced to 40 mg daily if given alone or 20 mg daily if given with azathioprine for one week. During the third and fourth weeks of treatment, patients received prednisone, 30 mg daily if given alone or 15 mg daily if given with azathioprine. After the fourth week of treatment, patients were maintained on prednisone, 20 mg daily if given alone or 10 mg daily if given with azathioprine, until a response had been achieved.
2.4. Definition of response Response was defined as the absence of symptoms, normalization of serum bilirubin and c-globulin levels, and improvement of serum AST levels to normal or less than twice ULN during treatment. Improvement to this degree has been associated with enhanced survival [1] and the prevention or reduction of hepatic fibrosis [4,16]. Liver tissue examinations in the 127 patients who had responded in this fashion indicated that the clinical and laboratory improvements typically reflected histological improvement [25]. Treatment was terminated according to protocol over a 6-week period with reductions in the dose of prednisone by 2.5 mg or 5 mg every 2 weeks depending on the maintenance dose. Azathioprine was continued at 50 mg daily for 3 weeks during the withdrawal period and then 25 mg daily for 3 weeks before its discontinuation. Serum AST, bilirubin and c-globulin levels were assessed at 3 week intervals from mailed
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
163
specimens for the first 3 months and then at 6 month intervals thereafter in accordance with a previously reported protocol [12,19,23].
2.5. Definitions of outcomes after treatment withdrawal Outcomes after treatment withdrawal were defined as relapse, sustained remission, progression to cirrhosis, death from liver failure, and liver transplantation [19]. Relapse connoted an increase in the serum AST level to more than three-fold ULN with or without the reappearance of symptoms after discontinuation of the medication [12,19]. Previous studies had indicated that an increase of the serum AST level to more than three-fold ULN after discontinuation of treatment uniformly indicated the re-appearance of interface hepatitis [25]. Sustained remission connoted the absence of relapse and need for retreatment [9,10,19]. Progression to cirrhosis connoted the presence of cirrhosis at a subsequent liver tissue examination in patients without this finding at presentation. Death from liver failure connoted demise associated with hepatic encephalopathy, renal failure or gastrointestinal bleeding, and liver transplantation connoted performance of the transplant procedure because of manifestations of severe intractable liver failure.
2.6. HLA determinations One hundred and twenty-seven patients (87%) had been evaluated for HLA DRB1*03 and HLA DRB1*04 by polymerase chain reaction with sequence-specific oligonucleotide probes (49 patients), restriction fragment length polymorphism (75 patients), or microlymphocytotoxicity (3 patients), as reported previously [26].
2.7. Statistical analyses The Fisher exact probability test was used to compare categorical variables, and the Mann–Whitney test was used to compare differences in the means of quantitative variables with a skewed distribution. The cumulative percentage of response in various patient groups was calculated using the Kaplan–Meier method, and differences between the groups were compared using the Log–Rank (Mantel–Cox) test. Correlation between variables was determined by Spearman rank correlation analysis. The variables for comparison had been formulated a priori and then assessed systematically, and an unadjusted P-value <0.05 was used to determine statistical significance. The mean ± standard error of the mean, median and ranges for all the quantitative variables are presented in the tables and text.
3. Results 3.1. Duration of treatment to response The duration of treatment to a response was 22 ± 2 months (range, 1–180 months; median, 15 months). Sixteen patients (11%) had responded after 66 months of treatment, including two patients who had responded after one and two months of therapy, respectively (Fig. 1). The majority of patients (63%) who responded had done so within 18 months of treatment, and 77% had done so within 24 months. The frequency of response for each 6 months of treatment through the first 2 years averaged 19%. This average decreased to 3% for each 6 months of treatment during the next 4 years. 3.2. Features associated with rapid response The 16 patients who responded after 66 months of treatment were older than the 130 patients who
Fig. 1. Cumulative percentage of response and duration of treatment. The number of responses during each 6 month interval and the cumulative number of responses that constitute the percentages depicted at 6 month intervals on the curve are shown at the bottom of the figure for the 146 patients. The cumulative percentage of response was determined by Kaplan–Meier analysis.
responded later (54 ± 3 years versus 46 ± 1 years, P = 0.04), but they were otherwise indistinguishable by gender, laboratory indices, histological features, HLA phenotype, and treatment regimens (Table 1). The two patients who responded within two months of treatment were 59 years and 69 years old, respectively, and one had cirrhosis at presentation. When the 16 patients who responded after 66 months were compared to the 20 patients who responded after 36 months, the distinction by age (54 ± 3 years versus 41 ± 4 years, P = 0.01) was maintained, and the rapid responders were further distinguished by a lower frequency of HLA DRB1*03 (36% versus 76%, P = 0.03). Other clinical, laboratory, and histological features between these groups were similar (Table 1). Comparisons between the 59 patients who entered remission within 12 months of therapy and the 20 patients who entered remission after 3 years of treatment continued to demonstrate that the rapid responders were older than the late responders (50 ± 2 years versus 41 ± 4 years, P = 0.03) and less commonly positive for HLA DRB1*03 (47% versus 76%, P = 0.04). 3.3. Rapidity of response and age at presentation Patients aged P60 years responded more rapidly than patients aged <40 years (Fig. 2). They responded within 6 months of therapy more commonly than patients aged <40 years (18% versus 2%, P = 0.02), and 94% had responded within 24 months. In contrast, only 64% of patients aged <40 years had responded within 24 months and only 81% had responded within 36 months (Fig. 2). The cumulative percentage of response was significantly greater for patients aged P60 years than for patients aged <40 years through the first 36 months of treatment (94% versus 81%, P = 0.004) (Fig. 2). The cumulative percentages of response for patients in the other age ranges were similar
164
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
Table 1 Clinical features at presentation and treatment durations to response. Clinical features
Treatment duration to response 66 mo (N = 16)
>6 mo (N = 130)
>36 mo (N = 20)
Age (years) Female:male AST (nl, 631 U/L) Bilirubin (nl, <1.1 mg/dL) c-Globulin (nl, 0.7–1.7 g/dL) Immunoglobulin G (nl, 600–1500 mg/dL) Interface hepatitis Bridging necrosis Multi-acinar necrosis Cirrhosis at entry DRB1*03 DRB1*04 Prednisone alone Combination therapy
54 ± 3a,c (55; 24–71) 13:3 698 ± 121 (614; 205–1800) 5 ± 1.2 (2.4; 0.6–16) 2.6 ± 0.3 (2.7; 0.8–4.8) 2181 ± 288 (1966; 1600–7870) 8 (50) 0 (0) 3 (19) 5 (31) 5/14 (36)b 8/14 (57) 7 (44) 9 (56)
46 ± 1c (48; 13–82) 105:25 541 ± 37 (447; 105–2000) 3.6 ± 0.4 (1.9; 0.3–24) 3 ± 0.1 (3; 0.7–6.2) 2799 ± 128 (2610; 422–7200) 70 (54) 13 (10) 17 (13) 30 (23) 60/113 (53) 54/113 (48) 38 (29) 92 (71)
41 ± 4a (41; 13–62) 14:6 520 ± 72 (464; 119–846) 3.7 ± 1 (2.2; 04–18.6) 3.1 ± 0.3 (2.8; 0.9–6.2) 2840 ± 386 (2330; 838–6400) 12 (60) 1 (5) 3 (15) 4 (20) 13/17 (76)b 6/17 (35) 5 (25) 15 (75)
Significantly different from each other at level of aP = 0.01, bP = 0.03, and cP = 0.04. AST = serum aspartate aminotransferase level. Numbers in parentheses are median and ranges for the quantitative variables and percentages for the categorical variables.
to each other and to those of the patients aged <40 years or aged P60 years (Fig. 2). Increasing age was associated with shorter treatment times to response by Spearman rank correlation analysis (Rho 0.24, P = 0.003). 3.4. Rapidity of response and HLA status The 29 patients aged P60 years who underwent HLA typing were distinguished from the 39 patients aged <40 years who underwent HLA typing by having a lower frequency of HLA DRB1*03 (31% versus 69%, P = 0.003) and higher frequency of HLA DRB1*04 (76% versus 20%, P < 0.0001). Three of the 4 patients aged P60 years who underwent HLA typing and who responded to treatment within 6 months had HLA DRB1*04 as
did 9 of the 12 elderly patients who had been HLA typed and who responded within 12 months. In contrast, only 1 of the 13 patients aged <40 years who had been HLA typed and who responded within 12 months had HLA DRB1*04 (75% versus 8%, P = 0.0001). HLA DRB1*03 was present in 10 of the 13 patients aged <40 years who underwent HLA typing and who responded after more than 24 months of treatment (77%). This group could not be compared to patients aged P60 years since only two of the 33 elderly patients required treatment of this duration. Other features at presentation, including female gender (88% versus 79%, P = 0.4), frequency of concurrent immune diseases, and histological findings did not distinguish the elderly patients from those aged <40 years. 3.5. Reversion to normal liver tests and tissue
Fig. 2. Cumulative percentage of response and duration of treatment in different age groups. The cumulative percentage of response by Kaplan– Meier analysis was greater during the same time intervals for the patients aged P60 years (thick solid line) than for adults aged <40 years (thin solid line), and this difference was significantly greater by the Log–Rank (Mantel–Cox) test (P = 0.004). The cumulative percentage of response in patients aged >40 years and <60 years (dashed and dotted lines) was intermediate between those for the elderly and the young and not significantly different from each other or the other age groups.
Twenty-one of the 127 patients who had undergone liver tissue examination immediately prior to treatment withdrawal (16%) had reverted to normal liver tests and tissue. Treatment durations were shorter in these 21 patients (median, 11 months; range, 5–96 months) than in the 106 patients (median, 18 months; range, 2– 180 months) who had improved to near-normal serum AST level and liver tissue prior to drug withdrawal (16 ± 4 months versus 23 ± 2 months, P = 0.01). There were no other differences between these groups, including age at presentation (45 ± 3 years versus 46 ± 2 years, P = 0.8), female gender (86% versus 77%, P = 0.6), and frequencies of HLA DRB1*03 (55% versus 52%, P = 0.8) and DRB1*04 (39% versus 46%, P = 0.6). Patients aged <40 years reverted to normal liver tests and tissue as commonly as patients aged P60 years (17% versus 11%, P = 0.7).
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
3.6. Outcomes associated with rapidity of response Patients who responded within 12 months of therapy had lower frequencies of progression to cirrhosis (18% versus 54%, P = 0.03) and requirement for liver transplantation (2% versus 15%, P < 0.05) than patients whose response was delayed for more than 36 months. These advantages were also apparent in patients who responded within 18 months and 24 months of treatment compared to those whose response was delayed for more than 36 months (Table 2). Progression to cirrhosis (48% versus 24%, P = 0.05) and requirement for liver transplantation (15% versus 3%, P = 0.05) increased in frequency when the treatment response was delayed for more than 24 months. The frequencies of sustained remission (19% versus 10%, P = 0.6) and relapse (81% versus 90%, P = 0.6) after drug withdrawal were similar during all response intervals (Table 2).
4. Discussion This study demonstrates that a rapid treatment response improves the outcome of type 1 autoimmune hepatitis. Patients who were able to suppress the clinical and laboratory manifestations of inflammatory activity within 24 months after institution of corticosteroid therapy progressed to cirrhosis less frequently and required liver transplantation less often than patients whose response was delayed (Table 2). Furthermore, patients aged P60 years responded more rapidly to treatment than patients aged <40 years, and they were characterized by a high frequency of HLA DRB1*04. In contrast, patients aged <40 years responded more slowly to conventional corticosteroid treatment than the elderly patients and tended to have HLA DRB1*03. Previous studies have indicated that age and HLA status are not independent features of autoimmune hepatitis and that elderly patients commonly have HLA DRB1*04 and an excellent long-term outcome after treatment [27]. The findings in this study suggest that the basis for their good prognosis may relate to the rapidity of their response to treatment.
165
Previous studies have demonstrated that hepatic inflammation stimulates transition of perivascular hepatic stellate cells into myofibroblasts which can proliferate, migrate, and synthesize fibrillar collagens and other matrix proteins [17]. The matrix proteins then accumulate because tissue inhibitors retard the counteractive degradative actions of matrix metalloproteinases [17]. Corticosteroids can facilitate anti-fibrotic actions by suppressing liver inflammation and promoting disappearance of the metalloproteinase inhibitors [4,16]. The findings in this study suggest that the rapidity by which the inflammation is suppressed is important in preventing or retarding hepatic fibrosis. A treatment response within 24 months affords the best opportunity to prevent disease progression, and it is an appropriate goal of initial therapy in all adult age groups (Table 2). Eighty-five percent of elderly patients responded within 18 months of treatment and 94% did so within 24 months (Fig. 2). In contrast, only 64% of patients aged <40 years responded within 24 months. These propensities suggest that different timelines for the expected response based on age and possibly HLA status may be useful in individualizing the treatment strategy. Slow responders can be identified after 24 months of treatment and further evaluated for alternative diagnoses or therapies. The basis for the rapid response of the elderly to corticosteroid therapy is unknown, but it may reflect agerelated changes in immune reactivity. Aging alters immune responsiveness by decreasing the expression of HLA class II molecules [28] and reducing the stimulation and proliferation of antigen-stimulated T cells [29]. These effects might make the pathogenic process less severe and more responsive to corticosteroid therapy. Similarly, the mechanisms by which the HLA phenotype may influence the rapidity of the treatment response are uncertain. HLA DRB1*03 may favor the presentation of a triggering antigen that is less frequently encountered in the elderly and more likely to generate a vigorous immune response in the young [27]. This increased immune response may in turn result in longer treatment requirements to suppress inflammatory activity.
Table 2 Outcomes associated with rapidity of treatment response. Outcome
Treatment duration to response 66 mo (N = 16)
612 mo (N = 59)
618 mo (N = 92)
624 mo (N = 112)
>36 mo (N = 20)
Sustained remission Relapse Progression to cirrhosis Death from liver failure Liver transplantation Duration follow-up (mo)
3 (19) 13 (81) 2/10 (20) 0 (0) 0 (0) 117 ± 27 (88; 2–328)
15 (25) 44 (75) 6/33 (18)c 2 (3) 1 (2)g 91 ± 11a (61; 2–365)
21 (23) 71 (77) 10/51 (20)d 4 (4) 3 (3) 102 ± 9b (79; 2–380)
27 (24) 85 (76) 15/63 (24)e 5 (5) 3 (3)f 110 ± 9 (92; 2–380)
2 (10) 18 (90) 7/13 (54)c–e 1 (5) 3 (15)f,g 164 ± 26a,b (145; 2–409)
Significantly different from each other at level of aP = 0.01, bP = 0.03, c,dP = 0.03, e,fP = 0.04 and gP = 0.048. Numbers in parentheses are the median and range for the quantitative variables and percentages for the categorical variables.
166
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
The rapid suppression of the clinical and laboratory indices of liver inflammation was unable to prevent relapse or ensure a sustained remission after drug withdrawal (Table 2). Rapid suppression of disease activity is important to retard fibrosis, but treatment until full resolution of liver tests and tissue is the preferred strategy to impact on relapse after drug withdrawal [5–8]. The nature of the underlying liver disease is the likely basis for the different response rates since the treatment schedules were similar in all patients and those who achieved normal tests and tissue actually required shorter durations of treatment than others. Complete clinical and laboratory improvement has been achieved elsewhere in 91% of patients after a mean treatment duration of 3 ± 3 months [30], and higher doses of medication than those used in this study may have accounted for these quick responses. A high dose treatment strategy may achieve the goals of preventing fibrosis and normalizing liver tests and tissue, and treatment trials evaluating the efficacy and safety of high dose schedules are warranted to optimize and standardize management. The observations in this study apply to a select subgroup of individuals that constitute only 53% of the treated population. The 28 patients who failed treatment were no longer candidates for a favorable response by conventional treatment, but the 98 patients who required continued therapy or adjusted regimens could theoretically respond to treatment at a later date. Seventy of these patients had received 636 months of treatment (mean, 11 ± 1 months; median, 7 months; range, 1–36 months) and 28 patients had received >36 months of treatment (mean, 100 ± 13 months; median, 77 months; range, 40–324 months). These patients were similar in age (47 ± 2 versus 47 ± 1 years), gender, HLA status, and histological features, including cirrhosis at presentation (24% versus 24%), to those who responded, and it is unlikely that their outcomes would alter the findings in this study if they ultimately achieved a response with continued treatment. Multiple analyses were performed using variables that had been defined a priori, and the possibility that the associations with age, HLA status, progression to cirrhosis or liver transplantation were by chance alone cannot be excluded. Nevertheless, the findings suggest that milestones for improvement during fixed treatment schedules can be developed based on clinical phenotypes. Future studies are warranted to improve the rapidity and completeness of the response through better patient selection, more effective dosing, and the identification and application of superior therapies. In summary, the rapidity of the response to conventional corticosteroid therapy is important in preventing the development of cirrhosis and requirement for liver transplantation in patients with type 1 autoimmune hepatitis. Suppression of the clinical and laboratory mani-
festations of active inflammation within 24 months is effective in achieving these results, but it is insufficient to prevent relapse after drug withdrawal. Patients aged P60 years respond more quickly to treatment than adults aged <40 years, and the timeline of expected responses varies by age group. References [1] Soloway RD, Summerskill WHJ, Baggenstoss AH, Geall MG, Gitnick GL, Elveback LR, et al. Clinical, biochemical, and histological remission of severe chronic active liver disease: a controlled study of treatments and early prognosis. Gastroenterology 1972;63:820–833. [2] Floreani A, Niro G, Rosa Rizzotto E, Antoniazzi S, Ferrara F, Carderi I, et al. Type 1 autoimmune hepatitis: clinical course and outcome in an Italian multicentre study. Aliment Pharmacol Ther 2006;24:1051–1057. [3] Seo S, Toutounjian R, Conrad A, Blatt L, Tong MJ. Favorable outcomes of autoimmune hepatitis in a community clinic setting. J Gastroenterol Hepatol 2008;23:1410–1414. [4] Czaja AJ, Carpenter HA. Decreased fibrosis during corticosteroid therapy of autoimmune hepatitis. J Hepatol 2004;40:644–650. [5] Czaja AJ, Davis GL, Ludwig J, Taswell HF. Complete resolution of inflammatory activity following corticosteroid treatment of HBsAg-negative chronic active hepatitis. Hepatology 1984;4:622–627. [6] Verma S, Gunuwan B, Mendler M, Govindrajan S, Redeker A. Factors predicting relapse and poor outcome in type 1 autoimmune hepatitis: role of cirrhosis development, patterns of transaminases during remission, and plasma cell activity in the liver biopsy. Am J Gastroenterol 2004;99:1510–1516. [7] Miyake Y, Iwasaki Y, Terada R, Takagi S, Okamaoto R, Ikeda H, et al. Persistent normalization of serum alanine aminotransferase levels improves the prognosis of type 1 autoimmune hepatitis. J Hepatol 2005;43:951–957. [8] Montano-Loza A, Carpenter HA, Czaja AJ. Improving the end point of corticosteroid therapy in type 1 autoimmune hepatitis to reduce the frequency of relapse. Am J Gastroenterol 2007;102:1005–1012. [9] Czaja AJ, Beaver SJ, Shiels MT. Sustained remission following corticosteroid therapy of severe HBsAg-negative chronic active hepatitis. Gastroenterology 1987;92:215–219. [10] Czaja AJ, Menon KVN, Carpenter HA. Sustained remission after corticosteroid therapy for type 1 autoimmune hepatitis: a retrospective analysis. Hepatology 2002;35:890–897. [11] Ferenci P, Fried MW, Shiffman ML, Smith CI, Marinos G, Goncales Jr FL, et al. Predicting sustained virological responses in chronic hepatitis C patients treated with peginterferon alfa-2a (40 kD)/ribavirin. J Hepatol 2005;43:425–433. [12] Montano-Loza A, Carpenter HA, Czaja AJ. Consequences of treatment withdrawal in type 1 autoimmune hepatitis. Liver Int 2007;27:507–515. [13] Kanzler S, Gerken G, Lohr H, Galle PR, Meyer zum Buschenfelde KH, Lohse AW. Duration of immunosuppressive therapy in autoimmune hepatitis. J Hepatol 2001;34:354–355. [14] Johnson PJ, McFarlane IG, Williams R. Azathioprine for longterm maintenance of remission in autoimmune hepatitis. N Engl J Med 1995;333:958–963. [15] Czaja AJ. Low dose corticosteroid therapy after multiple relapses of severe HBsAg-negative chronic active hepatitis. Hepatology 1990;11:1044–1049. [16] Czaja AJ, Carpenter HA. Progressive fibrosis during corticosteroid therapy of autoimmune hepatitis. Hepatology 2004;39:1631–1638.
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167 [17] Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 2000;275:2247–2250. [18] Alvarez F, Berg PA, Bianchi FB, Bianchi L, Burroughs AK, Cancado EL, et al. International Autoimmune Hepatitis Group report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:929–938. [19] Czaja AJ, Freese DK. Diagnosis and treatment of autoimmune hepatitis. Hepatology 2002;36:479–497. [20] Czaja AJ, Manns MP. The validity and importance of subtypes of autoimmune hepatitis: a point of view. Am J Gastroenterol 1995;90:1206–1211. [21] Czaja AJ, Carpenter HA. Sensitivity, specificity and predictability of biopsy interpretations in chronic hepatitis. Gastroenterology 1993;105:1824–1832. [22] Soloway RD, Baggenstoss AH, Schoenfield LJ, Summerskill WHJ. Observer error and sampling variability tested in evaluation of hepatitis and cirrhosis by liver biopsy. Am J Dig Dis 1975;20:1087–1090. [23] Montano Loza A, Czaja AJ. Current therapy for autoimmune hepatitis. Nat Clin Pract Gastroenterol Hepatol 2007;4: 202–214.
167
[24] Summerskill WHJ, Korman MG, Ammon HV, Baggenstoss AH. Prednisone for chronic active liver disease: dose titration, standard dose, and combination with azathioprine compared. Gut 1975;16:876–883. [25] Czaja AJ, Wolf AM, Baggenstoss AH. Laboratory assessment of severe chronic active liver disease (CALD): correlation of serum transaminase and gamma globulin levels with histologic features. Gastroenterology 1981;80:687–692. [26] Montano-Loza A, Carpenter HA, Czaja AJ. Clinical significance of HLA DRB103–DRB104 in type 1 autoimmune hepatitis. Liver Int 2006;26:1201–1208. [27] Czaja AJ, Carpenter HA. Distinctive clinical phenotype and treatment outcome of type 1 autoimmune hepatitis in the elderly. Hepatology 2006;43:532–538. [28] Villanueva JL, Solana R, Alonso MC, Pena J. Changes in the expression of HLA class II antigens on peripheral blood monocytes from aged humans. Dis Markers 1990;8:85–91. [29] Murasko DM, Goonewardene IM. T cell function in aging: mechanisms of decline. Ann Rev Gerontol Geriatr 1990;10:71–96. [30] Kanzler S, Lohr H, Gerken G, Galle PR, Lohse AW. Long-term management and prognosis of autoimmune hepatitis (AIH): a single center experience. Z Gastroenterol 2001;39:339–341.
Rapidity of treatment response and outcome in type 1 autoimmune hepatitisq Albert J. Czaja* Division of Gastroenterology and Hepatology, Mayo Clinic College of Medicine, 200 First Street S.W., Rochester, MN 55905, USA
Background/Aims: Corticosteroid therapy is effective in type 1 autoimmune hepatitis. This study determines if the rapidity of response affects outcome. Methods: The duration of treatment to improve clinical and laboratory indices was determined retrospectively in 146 patients and correlated with outcome. Results: Sixteen patients (11%) responded after 66 months, and 20 patients did so after 36 months (14%). The rapid responders were older than those who responded slowly (54 ± 3 years versus 41 ± 4 years, P = 0.007), and they had a lower frequency of HLA DRB1*03 (36% versus 76%, P = 0.03). Patients aged P60 years responded within 6 months more commonly than adults aged <40 years (18% versus 2%, P = 0.02), and most had responded within 24 months (94% versus 64%, P = 0.003). Elderly patients who responded quickly had HLA DRB1*04 more commonly than adults aged <40 years (75% versus 8%, P = 0.0001). Progression to cirrhosis (18% versus 54%, P = 0.03) and liver transplantation (2% versus 15%, P < 0.05) was less common in the rapid responders. The frequencies of a sustained remission (19% versus 10%, P = 0.6) and relapse after drug withdrawal (81% versus 90%, P = 0.6) were similar between patients with an early and late response. Conclusions: A rapid treatment response decreases progression to cirrhosis and liver transplantation, but it does not alter the frequency of sustained remission or relapse after drug withdrawal. Elderly patients respond more quickly to treatment than adults aged <40 years, and they are characterized by HLA DRB1*04. Ó 2009 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Corticosteroid therapy; Progression to cirrhosis; Response rate
1. Introduction Corticosteroid therapy ameliorates symptoms, resolves laboratory abnormalities, improves histological Received 16 December 2008; received in revised form 3 February 2009; accepted 23 February 2009; available online 10 April 2009 Associate Editor: V. Barnaba q The author declared that he does not have anything to disclose regarding funding or conflict of interest with respect to this manuscript. * Tel.: +1 507 284 8118; fax: +1 507 284 0538. E-mail address: [email protected] Abbreviations: AJC, Albert J. Czaja; ANA, antinuclear antibodies; AST, aspartate aminotransferase; HLA, human leukocyte antigen; LKM1, liver-kidney microsome type 1; SMA, smooth muscle antibodies; ULN, upper limit of normal range.
changes, reduces or prevents hepatic fibrosis, and enhances survival in most patients with autoimmune hepatitis [1–4]. Treatment until normal laboratory indices and normal liver tissue is reached is the ideal end point of treatment [5–8], but it is not achievable in all patients [8]. Furthermore, its relentless pursuit may be associated with drug-related side effects [5]. Treatment until improvement in the serum aspartate aminotransferase (AST) level to less than twice the upper limit of the normal range (UNL) in conjunction with normalization of serum bilirubin and c-globulin abnormalities and disappearance of interface hepatitis from the liver tissue is achievable in most patients and is well tolerated [1,9,10]. This result has been associated with the reduction and prevention of hepatic fibrosis [4],
0168-8278/$36.00 Ó 2009 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2009.02.026
162
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
improved long-term survival [1], and a sustainable remission in 50% of patients [9,10] that may be long-term in 20% [10]. It constitutes the minimum level of improvement that has been associated with a satisfactory outcome after drug withdrawal. The rapidity of a treatment response may have prognostic implications that reflect the severity of the disease, the genetic predispositions of the host, and the metabolism of the medication in individual patients [9,10]. In chronic hepatitis C, the disappearance of viremia within 12 weeks after institution of anti-viral therapy has been associated with a sustained virological response, and this realization has allowed the individualization of antiviral regimens [11]. Patients who respond rapidly do not require protracted anti-viral therapy, whereas those who respond slowly or not at all can be identified quickly and considered for alternative treatments. In autoimmune hepatitis, the association between the rapidity of the treatment response and the outcome of the disease is unknown, yet it may be an important discriminator between those who will survive without consequences and those who will progress to cirrhosis and liver failure [12]. Treatments typically average 2 years [5,8–10,12,13], but they range from 6 months [1,5] to life-long [14,15]. Progression of hepatic fibrosis in autoimmune hepatitis can be rapid, and it is fueled by inflammatory activity [16]. Corticosteroids can suppress this activity [4] and prevent stellate cell activation or promote degradation of the already established fibrotic matrix [17]. Quick suppression of this inflammatory activity may identify those individuals who will not progress. In this retrospective analysis of prospectively acquired data, the patients who respond rapidly (66 months) and those who respond slowly (>36 months) to conventional corticosteroid therapy are characterized, and the effect of rapid improvement on the frequency of relapse after drug withdrawal, progression to cirrhosis, and occurrence of death from liver failure or requirement for liver transplantation is assessed. In this fashion, the study determines if the rapidity of response to a standard treatment has prognostic value and if the rapidity of response can be used to identify problematic patients early who may need close surveillance and individualized therapy. Furthermore, the characterization of rapid responders may affect the indications and enthusiasm for their treatment.
2. Patients and methods 2.1. Study population One hundred and forty-six patients who satisfied international pretreatment criteria for the diagnosis of autoimmune hepatitis [18] and who had been evaluated and treated in accordance with a pre-established protocol [19] constituted the study population. These patients had been selected from 275 similar patients because they had
responded to their first course of corticosteroid therapy, and they had been withdrawn from medication. The 129 patients not included in the analysis included 28 patients who had failed to improve during treatment (10%), 98 patients who required continued treatment or who had developed drug intolerances that warranted adjustments in their management strategy (36%), and 3 patients who had not undergone liver tissue examinations at presentation (1%). The investigation had been approved by the Institutional Review Board of the Mayo Clinic. One hundred and eighteen patients (81%) were women, and the mean age of the population was 47 ± 1 years (range, 13–82 years; median, 49 years). One hundred and thirty patients (89%) were classified pre-treatment as having definite autoimmune hepatitis, and 16 patients (11%) were classified pre-treatment as having probable autoimmune hepatitis by the criteria of the International Autoimmune Hepatitis Group [18]. The mean pre-treatment diagnostic score was 18.3 ± 0.2 points (range, 11–22 points; median, 19 points). All patients had been tested for antinuclear antibodies (ANA) and smooth muscle antibodies (SMA), and 127 patients (87%) had been tested and negative for antibodies to liver kidney microsome type 1 (anti-LKM1) by previously published methods [4,8,10,12,16]. The serological findings justified the designation of type 1 autoimmune hepatitis [20].
2.2. Clinical and histological assessments Clinical examinations and conventional laboratory tests of liver inflammation and function had been performed at presentation and at each 6-month follow-up interval by one clinician (A.J.C.) in accordance with a previously published protocol [19]. Liver tissue examinations had been performed at accession in all patients and immediately prior to drug withdrawal in 127 patients (87%). Pre-established criteria were applied in all histological examinations, and the diagnosis of cirrhosis required demonstration of fibrosis and a complete regenerative nodule [19,21]. Our previous validation studies indicated that the consistency of pathological interpretation for grade and stage of disease was 90% and 78%, respectively [5,12,21,22].
2.3. Treatment regimens All patients had been treated with either prednisone alone (45 patients) or a lower dose of prednisone in conjunction with a fixed dose of azathioprine (101 patients) in accordance with schedules published previously [19,23]. Each regimen had been shown to be comparable to each other [24]. Prednisone alone (60 mg daily) or a lower dose of prednisone (30 mg daily) in conjunction with azathioprine (50 mg daily) was started as an induction regimen and continued for one week [19,23]. The dose of prednisone was then reduced to 40 mg daily if given alone or 20 mg daily if given with azathioprine for one week. During the third and fourth weeks of treatment, patients received prednisone, 30 mg daily if given alone or 15 mg daily if given with azathioprine. After the fourth week of treatment, patients were maintained on prednisone, 20 mg daily if given alone or 10 mg daily if given with azathioprine, until a response had been achieved.
2.4. Definition of response Response was defined as the absence of symptoms, normalization of serum bilirubin and c-globulin levels, and improvement of serum AST levels to normal or less than twice ULN during treatment. Improvement to this degree has been associated with enhanced survival [1] and the prevention or reduction of hepatic fibrosis [4,16]. Liver tissue examinations in the 127 patients who had responded in this fashion indicated that the clinical and laboratory improvements typically reflected histological improvement [25]. Treatment was terminated according to protocol over a 6-week period with reductions in the dose of prednisone by 2.5 mg or 5 mg every 2 weeks depending on the maintenance dose. Azathioprine was continued at 50 mg daily for 3 weeks during the withdrawal period and then 25 mg daily for 3 weeks before its discontinuation. Serum AST, bilirubin and c-globulin levels were assessed at 3 week intervals from mailed
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
163
specimens for the first 3 months and then at 6 month intervals thereafter in accordance with a previously reported protocol [12,19,23].
2.5. Definitions of outcomes after treatment withdrawal Outcomes after treatment withdrawal were defined as relapse, sustained remission, progression to cirrhosis, death from liver failure, and liver transplantation [19]. Relapse connoted an increase in the serum AST level to more than three-fold ULN with or without the reappearance of symptoms after discontinuation of the medication [12,19]. Previous studies had indicated that an increase of the serum AST level to more than three-fold ULN after discontinuation of treatment uniformly indicated the re-appearance of interface hepatitis [25]. Sustained remission connoted the absence of relapse and need for retreatment [9,10,19]. Progression to cirrhosis connoted the presence of cirrhosis at a subsequent liver tissue examination in patients without this finding at presentation. Death from liver failure connoted demise associated with hepatic encephalopathy, renal failure or gastrointestinal bleeding, and liver transplantation connoted performance of the transplant procedure because of manifestations of severe intractable liver failure.
2.6. HLA determinations One hundred and twenty-seven patients (87%) had been evaluated for HLA DRB1*03 and HLA DRB1*04 by polymerase chain reaction with sequence-specific oligonucleotide probes (49 patients), restriction fragment length polymorphism (75 patients), or microlymphocytotoxicity (3 patients), as reported previously [26].
2.7. Statistical analyses The Fisher exact probability test was used to compare categorical variables, and the Mann–Whitney test was used to compare differences in the means of quantitative variables with a skewed distribution. The cumulative percentage of response in various patient groups was calculated using the Kaplan–Meier method, and differences between the groups were compared using the Log–Rank (Mantel–Cox) test. Correlation between variables was determined by Spearman rank correlation analysis. The variables for comparison had been formulated a priori and then assessed systematically, and an unadjusted P-value <0.05 was used to determine statistical significance. The mean ± standard error of the mean, median and ranges for all the quantitative variables are presented in the tables and text.
3. Results 3.1. Duration of treatment to response The duration of treatment to a response was 22 ± 2 months (range, 1–180 months; median, 15 months). Sixteen patients (11%) had responded after 66 months of treatment, including two patients who had responded after one and two months of therapy, respectively (Fig. 1). The majority of patients (63%) who responded had done so within 18 months of treatment, and 77% had done so within 24 months. The frequency of response for each 6 months of treatment through the first 2 years averaged 19%. This average decreased to 3% for each 6 months of treatment during the next 4 years. 3.2. Features associated with rapid response The 16 patients who responded after 66 months of treatment were older than the 130 patients who
Fig. 1. Cumulative percentage of response and duration of treatment. The number of responses during each 6 month interval and the cumulative number of responses that constitute the percentages depicted at 6 month intervals on the curve are shown at the bottom of the figure for the 146 patients. The cumulative percentage of response was determined by Kaplan–Meier analysis.
responded later (54 ± 3 years versus 46 ± 1 years, P = 0.04), but they were otherwise indistinguishable by gender, laboratory indices, histological features, HLA phenotype, and treatment regimens (Table 1). The two patients who responded within two months of treatment were 59 years and 69 years old, respectively, and one had cirrhosis at presentation. When the 16 patients who responded after 66 months were compared to the 20 patients who responded after 36 months, the distinction by age (54 ± 3 years versus 41 ± 4 years, P = 0.01) was maintained, and the rapid responders were further distinguished by a lower frequency of HLA DRB1*03 (36% versus 76%, P = 0.03). Other clinical, laboratory, and histological features between these groups were similar (Table 1). Comparisons between the 59 patients who entered remission within 12 months of therapy and the 20 patients who entered remission after 3 years of treatment continued to demonstrate that the rapid responders were older than the late responders (50 ± 2 years versus 41 ± 4 years, P = 0.03) and less commonly positive for HLA DRB1*03 (47% versus 76%, P = 0.04). 3.3. Rapidity of response and age at presentation Patients aged P60 years responded more rapidly than patients aged <40 years (Fig. 2). They responded within 6 months of therapy more commonly than patients aged <40 years (18% versus 2%, P = 0.02), and 94% had responded within 24 months. In contrast, only 64% of patients aged <40 years had responded within 24 months and only 81% had responded within 36 months (Fig. 2). The cumulative percentage of response was significantly greater for patients aged P60 years than for patients aged <40 years through the first 36 months of treatment (94% versus 81%, P = 0.004) (Fig. 2). The cumulative percentages of response for patients in the other age ranges were similar
164
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
Table 1 Clinical features at presentation and treatment durations to response. Clinical features
Treatment duration to response 66 mo (N = 16)
>6 mo (N = 130)
>36 mo (N = 20)
Age (years) Female:male AST (nl, 631 U/L) Bilirubin (nl, <1.1 mg/dL) c-Globulin (nl, 0.7–1.7 g/dL) Immunoglobulin G (nl, 600–1500 mg/dL) Interface hepatitis Bridging necrosis Multi-acinar necrosis Cirrhosis at entry DRB1*03 DRB1*04 Prednisone alone Combination therapy
54 ± 3a,c (55; 24–71) 13:3 698 ± 121 (614; 205–1800) 5 ± 1.2 (2.4; 0.6–16) 2.6 ± 0.3 (2.7; 0.8–4.8) 2181 ± 288 (1966; 1600–7870) 8 (50) 0 (0) 3 (19) 5 (31) 5/14 (36)b 8/14 (57) 7 (44) 9 (56)
46 ± 1c (48; 13–82) 105:25 541 ± 37 (447; 105–2000) 3.6 ± 0.4 (1.9; 0.3–24) 3 ± 0.1 (3; 0.7–6.2) 2799 ± 128 (2610; 422–7200) 70 (54) 13 (10) 17 (13) 30 (23) 60/113 (53) 54/113 (48) 38 (29) 92 (71)
41 ± 4a (41; 13–62) 14:6 520 ± 72 (464; 119–846) 3.7 ± 1 (2.2; 04–18.6) 3.1 ± 0.3 (2.8; 0.9–6.2) 2840 ± 386 (2330; 838–6400) 12 (60) 1 (5) 3 (15) 4 (20) 13/17 (76)b 6/17 (35) 5 (25) 15 (75)
Significantly different from each other at level of aP = 0.01, bP = 0.03, and cP = 0.04. AST = serum aspartate aminotransferase level. Numbers in parentheses are median and ranges for the quantitative variables and percentages for the categorical variables.
to each other and to those of the patients aged <40 years or aged P60 years (Fig. 2). Increasing age was associated with shorter treatment times to response by Spearman rank correlation analysis (Rho 0.24, P = 0.003). 3.4. Rapidity of response and HLA status The 29 patients aged P60 years who underwent HLA typing were distinguished from the 39 patients aged <40 years who underwent HLA typing by having a lower frequency of HLA DRB1*03 (31% versus 69%, P = 0.003) and higher frequency of HLA DRB1*04 (76% versus 20%, P < 0.0001). Three of the 4 patients aged P60 years who underwent HLA typing and who responded to treatment within 6 months had HLA DRB1*04 as
did 9 of the 12 elderly patients who had been HLA typed and who responded within 12 months. In contrast, only 1 of the 13 patients aged <40 years who had been HLA typed and who responded within 12 months had HLA DRB1*04 (75% versus 8%, P = 0.0001). HLA DRB1*03 was present in 10 of the 13 patients aged <40 years who underwent HLA typing and who responded after more than 24 months of treatment (77%). This group could not be compared to patients aged P60 years since only two of the 33 elderly patients required treatment of this duration. Other features at presentation, including female gender (88% versus 79%, P = 0.4), frequency of concurrent immune diseases, and histological findings did not distinguish the elderly patients from those aged <40 years. 3.5. Reversion to normal liver tests and tissue
Fig. 2. Cumulative percentage of response and duration of treatment in different age groups. The cumulative percentage of response by Kaplan– Meier analysis was greater during the same time intervals for the patients aged P60 years (thick solid line) than for adults aged <40 years (thin solid line), and this difference was significantly greater by the Log–Rank (Mantel–Cox) test (P = 0.004). The cumulative percentage of response in patients aged >40 years and <60 years (dashed and dotted lines) was intermediate between those for the elderly and the young and not significantly different from each other or the other age groups.
Twenty-one of the 127 patients who had undergone liver tissue examination immediately prior to treatment withdrawal (16%) had reverted to normal liver tests and tissue. Treatment durations were shorter in these 21 patients (median, 11 months; range, 5–96 months) than in the 106 patients (median, 18 months; range, 2– 180 months) who had improved to near-normal serum AST level and liver tissue prior to drug withdrawal (16 ± 4 months versus 23 ± 2 months, P = 0.01). There were no other differences between these groups, including age at presentation (45 ± 3 years versus 46 ± 2 years, P = 0.8), female gender (86% versus 77%, P = 0.6), and frequencies of HLA DRB1*03 (55% versus 52%, P = 0.8) and DRB1*04 (39% versus 46%, P = 0.6). Patients aged <40 years reverted to normal liver tests and tissue as commonly as patients aged P60 years (17% versus 11%, P = 0.7).
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
3.6. Outcomes associated with rapidity of response Patients who responded within 12 months of therapy had lower frequencies of progression to cirrhosis (18% versus 54%, P = 0.03) and requirement for liver transplantation (2% versus 15%, P < 0.05) than patients whose response was delayed for more than 36 months. These advantages were also apparent in patients who responded within 18 months and 24 months of treatment compared to those whose response was delayed for more than 36 months (Table 2). Progression to cirrhosis (48% versus 24%, P = 0.05) and requirement for liver transplantation (15% versus 3%, P = 0.05) increased in frequency when the treatment response was delayed for more than 24 months. The frequencies of sustained remission (19% versus 10%, P = 0.6) and relapse (81% versus 90%, P = 0.6) after drug withdrawal were similar during all response intervals (Table 2).
4. Discussion This study demonstrates that a rapid treatment response improves the outcome of type 1 autoimmune hepatitis. Patients who were able to suppress the clinical and laboratory manifestations of inflammatory activity within 24 months after institution of corticosteroid therapy progressed to cirrhosis less frequently and required liver transplantation less often than patients whose response was delayed (Table 2). Furthermore, patients aged P60 years responded more rapidly to treatment than patients aged <40 years, and they were characterized by a high frequency of HLA DRB1*04. In contrast, patients aged <40 years responded more slowly to conventional corticosteroid treatment than the elderly patients and tended to have HLA DRB1*03. Previous studies have indicated that age and HLA status are not independent features of autoimmune hepatitis and that elderly patients commonly have HLA DRB1*04 and an excellent long-term outcome after treatment [27]. The findings in this study suggest that the basis for their good prognosis may relate to the rapidity of their response to treatment.
165
Previous studies have demonstrated that hepatic inflammation stimulates transition of perivascular hepatic stellate cells into myofibroblasts which can proliferate, migrate, and synthesize fibrillar collagens and other matrix proteins [17]. The matrix proteins then accumulate because tissue inhibitors retard the counteractive degradative actions of matrix metalloproteinases [17]. Corticosteroids can facilitate anti-fibrotic actions by suppressing liver inflammation and promoting disappearance of the metalloproteinase inhibitors [4,16]. The findings in this study suggest that the rapidity by which the inflammation is suppressed is important in preventing or retarding hepatic fibrosis. A treatment response within 24 months affords the best opportunity to prevent disease progression, and it is an appropriate goal of initial therapy in all adult age groups (Table 2). Eighty-five percent of elderly patients responded within 18 months of treatment and 94% did so within 24 months (Fig. 2). In contrast, only 64% of patients aged <40 years responded within 24 months. These propensities suggest that different timelines for the expected response based on age and possibly HLA status may be useful in individualizing the treatment strategy. Slow responders can be identified after 24 months of treatment and further evaluated for alternative diagnoses or therapies. The basis for the rapid response of the elderly to corticosteroid therapy is unknown, but it may reflect agerelated changes in immune reactivity. Aging alters immune responsiveness by decreasing the expression of HLA class II molecules [28] and reducing the stimulation and proliferation of antigen-stimulated T cells [29]. These effects might make the pathogenic process less severe and more responsive to corticosteroid therapy. Similarly, the mechanisms by which the HLA phenotype may influence the rapidity of the treatment response are uncertain. HLA DRB1*03 may favor the presentation of a triggering antigen that is less frequently encountered in the elderly and more likely to generate a vigorous immune response in the young [27]. This increased immune response may in turn result in longer treatment requirements to suppress inflammatory activity.
Table 2 Outcomes associated with rapidity of treatment response. Outcome
Treatment duration to response 66 mo (N = 16)
612 mo (N = 59)
618 mo (N = 92)
624 mo (N = 112)
>36 mo (N = 20)
Sustained remission Relapse Progression to cirrhosis Death from liver failure Liver transplantation Duration follow-up (mo)
3 (19) 13 (81) 2/10 (20) 0 (0) 0 (0) 117 ± 27 (88; 2–328)
15 (25) 44 (75) 6/33 (18)c 2 (3) 1 (2)g 91 ± 11a (61; 2–365)
21 (23) 71 (77) 10/51 (20)d 4 (4) 3 (3) 102 ± 9b (79; 2–380)
27 (24) 85 (76) 15/63 (24)e 5 (5) 3 (3)f 110 ± 9 (92; 2–380)
2 (10) 18 (90) 7/13 (54)c–e 1 (5) 3 (15)f,g 164 ± 26a,b (145; 2–409)
Significantly different from each other at level of aP = 0.01, bP = 0.03, c,dP = 0.03, e,fP = 0.04 and gP = 0.048. Numbers in parentheses are the median and range for the quantitative variables and percentages for the categorical variables.
166
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167
The rapid suppression of the clinical and laboratory indices of liver inflammation was unable to prevent relapse or ensure a sustained remission after drug withdrawal (Table 2). Rapid suppression of disease activity is important to retard fibrosis, but treatment until full resolution of liver tests and tissue is the preferred strategy to impact on relapse after drug withdrawal [5–8]. The nature of the underlying liver disease is the likely basis for the different response rates since the treatment schedules were similar in all patients and those who achieved normal tests and tissue actually required shorter durations of treatment than others. Complete clinical and laboratory improvement has been achieved elsewhere in 91% of patients after a mean treatment duration of 3 ± 3 months [30], and higher doses of medication than those used in this study may have accounted for these quick responses. A high dose treatment strategy may achieve the goals of preventing fibrosis and normalizing liver tests and tissue, and treatment trials evaluating the efficacy and safety of high dose schedules are warranted to optimize and standardize management. The observations in this study apply to a select subgroup of individuals that constitute only 53% of the treated population. The 28 patients who failed treatment were no longer candidates for a favorable response by conventional treatment, but the 98 patients who required continued therapy or adjusted regimens could theoretically respond to treatment at a later date. Seventy of these patients had received 636 months of treatment (mean, 11 ± 1 months; median, 7 months; range, 1–36 months) and 28 patients had received >36 months of treatment (mean, 100 ± 13 months; median, 77 months; range, 40–324 months). These patients were similar in age (47 ± 2 versus 47 ± 1 years), gender, HLA status, and histological features, including cirrhosis at presentation (24% versus 24%), to those who responded, and it is unlikely that their outcomes would alter the findings in this study if they ultimately achieved a response with continued treatment. Multiple analyses were performed using variables that had been defined a priori, and the possibility that the associations with age, HLA status, progression to cirrhosis or liver transplantation were by chance alone cannot be excluded. Nevertheless, the findings suggest that milestones for improvement during fixed treatment schedules can be developed based on clinical phenotypes. Future studies are warranted to improve the rapidity and completeness of the response through better patient selection, more effective dosing, and the identification and application of superior therapies. In summary, the rapidity of the response to conventional corticosteroid therapy is important in preventing the development of cirrhosis and requirement for liver transplantation in patients with type 1 autoimmune hepatitis. Suppression of the clinical and laboratory mani-
festations of active inflammation within 24 months is effective in achieving these results, but it is insufficient to prevent relapse after drug withdrawal. Patients aged P60 years respond more quickly to treatment than adults aged <40 years, and the timeline of expected responses varies by age group. References [1] Soloway RD, Summerskill WHJ, Baggenstoss AH, Geall MG, Gitnick GL, Elveback LR, et al. Clinical, biochemical, and histological remission of severe chronic active liver disease: a controlled study of treatments and early prognosis. Gastroenterology 1972;63:820–833. [2] Floreani A, Niro G, Rosa Rizzotto E, Antoniazzi S, Ferrara F, Carderi I, et al. Type 1 autoimmune hepatitis: clinical course and outcome in an Italian multicentre study. Aliment Pharmacol Ther 2006;24:1051–1057. [3] Seo S, Toutounjian R, Conrad A, Blatt L, Tong MJ. Favorable outcomes of autoimmune hepatitis in a community clinic setting. J Gastroenterol Hepatol 2008;23:1410–1414. [4] Czaja AJ, Carpenter HA. Decreased fibrosis during corticosteroid therapy of autoimmune hepatitis. J Hepatol 2004;40:644–650. [5] Czaja AJ, Davis GL, Ludwig J, Taswell HF. Complete resolution of inflammatory activity following corticosteroid treatment of HBsAg-negative chronic active hepatitis. Hepatology 1984;4:622–627. [6] Verma S, Gunuwan B, Mendler M, Govindrajan S, Redeker A. Factors predicting relapse and poor outcome in type 1 autoimmune hepatitis: role of cirrhosis development, patterns of transaminases during remission, and plasma cell activity in the liver biopsy. Am J Gastroenterol 2004;99:1510–1516. [7] Miyake Y, Iwasaki Y, Terada R, Takagi S, Okamaoto R, Ikeda H, et al. Persistent normalization of serum alanine aminotransferase levels improves the prognosis of type 1 autoimmune hepatitis. J Hepatol 2005;43:951–957. [8] Montano-Loza A, Carpenter HA, Czaja AJ. Improving the end point of corticosteroid therapy in type 1 autoimmune hepatitis to reduce the frequency of relapse. Am J Gastroenterol 2007;102:1005–1012. [9] Czaja AJ, Beaver SJ, Shiels MT. Sustained remission following corticosteroid therapy of severe HBsAg-negative chronic active hepatitis. Gastroenterology 1987;92:215–219. [10] Czaja AJ, Menon KVN, Carpenter HA. Sustained remission after corticosteroid therapy for type 1 autoimmune hepatitis: a retrospective analysis. Hepatology 2002;35:890–897. [11] Ferenci P, Fried MW, Shiffman ML, Smith CI, Marinos G, Goncales Jr FL, et al. Predicting sustained virological responses in chronic hepatitis C patients treated with peginterferon alfa-2a (40 kD)/ribavirin. J Hepatol 2005;43:425–433. [12] Montano-Loza A, Carpenter HA, Czaja AJ. Consequences of treatment withdrawal in type 1 autoimmune hepatitis. Liver Int 2007;27:507–515. [13] Kanzler S, Gerken G, Lohr H, Galle PR, Meyer zum Buschenfelde KH, Lohse AW. Duration of immunosuppressive therapy in autoimmune hepatitis. J Hepatol 2001;34:354–355. [14] Johnson PJ, McFarlane IG, Williams R. Azathioprine for longterm maintenance of remission in autoimmune hepatitis. N Engl J Med 1995;333:958–963. [15] Czaja AJ. Low dose corticosteroid therapy after multiple relapses of severe HBsAg-negative chronic active hepatitis. Hepatology 1990;11:1044–1049. [16] Czaja AJ, Carpenter HA. Progressive fibrosis during corticosteroid therapy of autoimmune hepatitis. Hepatology 2004;39:1631–1638.
A.J. Czaja / Journal of Hepatology 51 (2009) 161–167 [17] Friedman SL. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 2000;275:2247–2250. [18] Alvarez F, Berg PA, Bianchi FB, Bianchi L, Burroughs AK, Cancado EL, et al. International Autoimmune Hepatitis Group report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:929–938. [19] Czaja AJ, Freese DK. Diagnosis and treatment of autoimmune hepatitis. Hepatology 2002;36:479–497. [20] Czaja AJ, Manns MP. The validity and importance of subtypes of autoimmune hepatitis: a point of view. Am J Gastroenterol 1995;90:1206–1211. [21] Czaja AJ, Carpenter HA. Sensitivity, specificity and predictability of biopsy interpretations in chronic hepatitis. Gastroenterology 1993;105:1824–1832. [22] Soloway RD, Baggenstoss AH, Schoenfield LJ, Summerskill WHJ. Observer error and sampling variability tested in evaluation of hepatitis and cirrhosis by liver biopsy. Am J Dig Dis 1975;20:1087–1090. [23] Montano Loza A, Czaja AJ. Current therapy for autoimmune hepatitis. Nat Clin Pract Gastroenterol Hepatol 2007;4: 202–214.
167
[24] Summerskill WHJ, Korman MG, Ammon HV, Baggenstoss AH. Prednisone for chronic active liver disease: dose titration, standard dose, and combination with azathioprine compared. Gut 1975;16:876–883. [25] Czaja AJ, Wolf AM, Baggenstoss AH. Laboratory assessment of severe chronic active liver disease (CALD): correlation of serum transaminase and gamma globulin levels with histologic features. Gastroenterology 1981;80:687–692. [26] Montano-Loza A, Carpenter HA, Czaja AJ. Clinical significance of HLA DRB103–DRB104 in type 1 autoimmune hepatitis. Liver Int 2006;26:1201–1208. [27] Czaja AJ, Carpenter HA. Distinctive clinical phenotype and treatment outcome of type 1 autoimmune hepatitis in the elderly. Hepatology 2006;43:532–538. [28] Villanueva JL, Solana R, Alonso MC, Pena J. Changes in the expression of HLA class II antigens on peripheral blood monocytes from aged humans. Dis Markers 1990;8:85–91. [29] Murasko DM, Goonewardene IM. T cell function in aging: mechanisms of decline. Ann Rev Gerontol Geriatr 1990;10:71–96. [30] Kanzler S, Lohr H, Gerken G, Galle PR, Lohse AW. Long-term management and prognosis of autoimmune hepatitis (AIH): a single center experience. Z Gastroenterol 2001;39:339–341.