Hepatotoxicity rates do not differ in patients with rheumatoid arthritis and psoriasis treated with methotrexate

Abstract

Objective. MTX hepatotoxicity is considered to occur more frequently in patients with psoriasis than in patients with RA. However, toxicity guidelines are based on reports from studies with small sample sizes and limited follow-up periods. The current study's objective was to examine the long-term risk of MTX hepatotoxicity based on a database review of patients with RA or psoriasis, and to examine whether the two populations differed.

Methods. We conducted a retrospective cohort review among members of a large health maintenance organization (HMO) in Israel who were diagnosed with either RA (n = 119) or psoriasis (n = 690) and who had purchased at least one dose of MTX. Liver function analyses were performed serially in these patients during the follow-up. All abnormal assays were recorded in the computerized database of the HMO.

Results. Both groups had hepatic enzyme elevation; the pre-disposing factors predictive of liver damage were female gender and a higher cumulative dose of MTX (hazard ratios, 1.46 and 1.07, respectively, P < 0.001). Age, concurrent diseases and type of disease had no influence on susceptibility to liver damage. No statistically significant difference was detected in any abnormal liver function test among patients with either RA or psoriasis.

Conclusion. Our study did not corroborate previous findings of significant differences between psoriasis patients and RA patients concerning susceptibility to hepatotoxicity from MTX therapy. The only significant factor predicting a higher risk of hepatic damage was female gender.

Methotrexate, Psoriasis, Rheumatoid arthritis, Psoriatic arthritis, Hepatotoxicity

Introduction

MTX is a folic acid antagonist derived from aminopterin. It inhibits the enzyme dihydrofolate reductase, which is seminal in the pathway supplying methyl donor groups for DNA, RNA and protein synthesis [1]. MTX has been used for the treatment of diseases characterized by inflammation or cellular proliferation. It is currently considered to be the most widely used DMARD [2]. The first reported use of MTX was over 50 years ago when aminopterin was studied as an alternative to cortisone for patients with psoriasis and RA [3, 4].

MTX toxicity remains an important issue and is one of the main reasons for discontinuation of treatment. MTX has a well-defined toxicity profile, and patients are monitored for gastrointestinal, hepatic and pulmonary toxicity, bone marrow suppression and stomatitis [5].

Hepatotoxicity is a major concern with respect to the use of low-dose MTX. This observation had been first noted in leukaemia patients receiving high daily doses of the drug [6]. Previous studies on hepatotoxicity after long-term MTX therapy in patients with PsA found that the risk of developing cirrhosis may be as high as 25% [7]. However, similar studies among RA patients reported substantially lower rates of liver cirrhosis of <2% [8, 9] as well as a low risk of mild liver fibrosis [10] and abnormal liver tests [11].

In 1994, the ACR published guidelines for monitoring the development of hepatotoxicity related to MTX [12]. These recommendations included measuring aspartate aminotransferase (AST), alanine aminotransferase (ALT) and albumin every 4–8 weeks. In addition, a complete blood count and serum creatinine concentration are evaluated at baseline and repeated at intervals by most rheumatologists. Those guidelines differed from dermatology guidelines used at that time by not requiring a liver biopsy before MTX treatment. Since then, the guidelines have been revised and current follow-up recommendations consist of performing a liver biopsy after 1–1.5 g of MTX has been administered to low-risk patients, and close to initiation of therapy (within 2–4 months) in patients with risk factors [13].

Earlier observational studies assessing MTX hepatotoxicity, including the studies on which the guidelines were based, have been generally limited to short follow-up periods and to relatively small study populations and thus were underpowered to detect the true risk of MTX hepatotoxicity [14].

Methods

This study was based on data from Maccabi Healthcare Services (MHS), Israel's second largest health maintenance organization with 1.7 million members nationwide (∼24% of the total population). According to the 1994 National Health Act in Israel, MHS may not exclude an applicant on any grounds, including age or state of health. Thus, all sectors of the population are represented in MHS. All MHS members’ diagnoses are downloaded daily to a central computer. The database is automatically updated with all hospitalizations and outpatient visits. Medications prescribed by MHS physicians are recorded as well. The study was approved by the ethics committee of the MHS.

Cases of psoriasis and RA diagnosed between 1 January 1998 and 1 July 2007 were identified by computerized record linkage of the entire cohort to the MHS central computerized database. We identified all prevalent patients diagnosed with psoriasis, PsA or RA according to the International Classification of Diseases, Ninth Revision, Clinical Modification. All dispensed prescriptions of MTX from 1998 through July 2007 were examined. The following data were obtained for every participant: gender, age at first MTX purchase, number and dosage of dispensed prescriptions of MTX and folic acid. The normal values of the liver functions tests were considered as following: serum γ-glutamyltransferase (GGT) (10–49 U/l); alkaline phosphatase (ALKP) (34–104 U/l); AST (10–37 U/l); and albumin (3.5–5.2 g/dl).

Statistical analysis

Baseline characteristics for psoriasis and RA patients were compared with the chi-square test for categorical variables and Mann–Whitney or Student's t-test for continuous variables where appropriate. We used Cox regression with days from first dispensed prescription of MTX as the time scale to estimate hazard ratios (HRs) and 95% CIs of first abnormal liver test among psoriasis patients compared with RA patients. Follow-up began with earliest diagnosis and ended with first date of abnormal test or last day of normal lab test. We then adjusted for gender, age at baseline and cumulative dose of MTX as a time-dependent variable. To assess the effect of the frequency of the testings, the model was recalculated for patients with at least 10 tests during the follow-up period.

Results

During the study period, 704 (3.5%) patients with psoriasis (out of the 19 921 psoriasis patients in MHS) and 133 (4.7%) patients with RA (out of the 2825 RA patients in MHS) had purchased at least one dose of MTX following the diagnosis of their disease. Fifteen patients who had both RA and psoriasis and 20 patients with missing data on at least one liver test (AST, GGT, albumin or ALKP) during the study period were excluded from the study, leaving a total of 809 patients (690 with psoriasis and 119 with RA) for further analysis. The mean follow-up period was 883 days for the psoriasis group, and 843 days for the RA group. Of the study groups, 48.3% of the psoriasis patients and 34.5% of the RA patients were men. The mean age was 52.6 years for the psoriasis group and 59.9 years for the RA group. The percentage of comorbid conditions was similar with ∼20% incidence of diabetes, 12% cardiovascular diseases and 8% asthma in each group.

Cumulative MTX dosing for each patient was calculated. In the psoriasis group, the median amount was 1000 mg of MTX dispensed at an average amount of 182 mg. In the RA group, the cumulative dose of the medicine was 3625 mg dispensed at an average of 195 mg per prescription (Table 1).

Table 1.

Demographical characteristics of RA and psoriasis patients included in the study

	Psoriasis, n = 690	RA, n = 119	P-value
Gender, men, n (%)	333 (48.3)	41 (34.5)	0.002
Age, n (%), years
<45	242 (35.1)	18 (15.1)
45–54	135 (19.6)	17 (14.3)
54–64	153 (22.2)	39 (32.8)
65–74	107 (15.5)	25 (21)
>74	53 (7.7)	20 (16.8)
Mean ± s.d.	52.6 ± 15.2	59.9 ± 14.2	<0.001
Cumulative dose, median, mg	1000	3625
MTX dose, n (%), mg
<375	189 (27.4)	3 (2.5)	<0.001
375–875	133 (19.3)	8 (6.7)
875–2000	137 (19.9)	18 (15.1)
2000–4750	122 (17.7)	43 (36.1)
>4750	109 (15.8)	47 (39.5)
Weekly dose, median, mg	19.0	20.0	0.795
Days of MTX treatment, median	878	1471	<0.001
Mean follow-up, median, days	883.8	843	0.65
Comorbid conditions, n (%)
Diabetes mellitus	130 (19)	25 (21)	0.45
Cardiovascular disease	84 (12)	14 (12)	1.00
Asthma	52 (8)	8 (7)	0.93
Folic acid use
Any form of vitamin	576 (84)	13 (10)	<0.001

A total of 45% of the tested patients had at least one abnormal result. The most commonly performed tests were ALKP and AST. GGT and albumin were tested in nearly 95% of the RA group, and 78% of the psoriasis patients.

There were statistically significant differences in the rate of at least one abnormal result of ALKP and albumin, with a proportionally higher rate of elevated ALKP in the RA group (33.6 vs 23.7%; P = 0.012) and a decreased albumin level in the same group (11.7 vs 7.8%; P = 0.17). No significant difference was found between the rate of abnormal GGT and AST levels (Table 2).

Table 2.

Variables related to liver testing in RA and psoriasis patients included in the study

	Patients with test, %		Average follow-upa		Average number of tests per patient		Patients with at least one positive testb, %			Positive tests, %
Test	Psor.	RA	Psor.	RA	Psor.	RA	Psor.	RA	P-value	Psor.	RA	P-value
ALKP	96	100	1488	1677	15.60	20.91	23.7	33.6	0.012	7.27	7.70	0.78
GGT	70	97	1228	1025	8.25	10.34	6.8	7.8	NS	15.21	15.70	0.83
Albumin	86	93	1428	1561	9.95	14.16	7.8	11.7	0.17	1.83	2.45	0.37
AST	96	100	1440	1605	15.46	20.58	39.3	40.3	NS	9.88	7.17	0.09

^aDays from first MTX purchase to last test.^bNormal values: ALKP (34–104 U/l); GGT (10–49 U/l); AST (10–37 U/l); and albumin (3.5–5.2 g/dl). Psor.: patients with psoriasis.

Since total purchased dose of folic acid was not associated with any abnormal test, either on univariate analysis (p = 0.78) or on multivariable analysis (P = 0.47), it was not included in the study model.

RA patients did not differ from patients with either psoriasis or PsA in the rates of elevated liver function tests. This finding was also demonstrated by analysing each liver function test separately (Table 3). Figure 1 depicts the cumulative risk of any abnormal liver test among study participants, indicating a higher risk among RA patients in the first 4 years of follow-up although these differences did not reach statistical significance (P = 0.07).

Fig. 1.

Risk over time of a positive liver test according to type of disease (log rank test, P = 0.07).

Table 3.

HR and 95% CI for abnormal liver laboratory test

	AST		GGT		ALKP		Albumin
	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value
Women vs men	1.46 (1.16, 1.84)	<0.001	1.31 (0.71, 2.42)	0.39	1.33 (1, 1.78)	0.05	1.25 (0.74, 2.11)	0.40
Age, years	1 (0.99, 1.01)	0.86	1.01 (0.99, 1.03)	0.31	1.01 (1, 1.02)	0.07	1.04 (1.02, 1.06)	<0.001
Cumulative MTX dose	1.07 (1.02, 1.12)	<0.001	0.86 (0.70, 1.04)	0.12	1.01 (0.95, 1.08)	0.69	0.97 (0.70, 1.34)	0.85
RA vs psoriasis	0.85 (0.62, 1.17)	0.33	1.27 (0.59, 2.74)	0.55	1.17 (0.81, 1.69)	0.39	1.34 (0.69, 2.60)	0.39

HR estimates are mutually adjusted to all variables in Table 3.

Interesting results emerged from the analysis of the test population according to subgroups (Tables 3 and 4). Women were significantly more vulnerable than men for hepatic enzyme elevation (HR 1.43; P < 0.01). The enzymes markedly elevated were AST and ALKP. An elevation in liver enzymes was found to be related to the cumulative dose of MTX (P < 0.01; HR = 1.07) (Table 4).

In multivariable model, there was no statistically significant (P = 0.54) difference between RA and psoriasis patients in the risk of abnormal liver test after adjusting to age, gender and MTX purchased dose. Similar results were obtained when analyses were limited to RA and PsA patients (Table 5).

Table 4.

HR and 95% CI for abnormal liver laboratory test (any test), including psoriasis (n = 647) and RA patients (n = 108)

	GGT/ALKP/AST
	HR (95% CI)	P-value
Women vs men	1.43 (1.18, 1.75)	<0.001
Age, years	1 (1, 1.01)	0.55
Cumulative MTX dosea	1.07 (1.01, 1.12)	0.01
RA vs psoriasis	1.09 (0.83, 1.42)	0.54

^aCumulative annual dose (per mg) defined as a time-dependent variable. HR estimates are mutually adjusted to all variables in Table 4.

Table 5.

HR and 95% CI for abnormal liver laboratory test (any test), including only PsA (n = 306) and RA patients (n = 108)

	GGT/ALKP/AST
	HR (95% CI)	P-value
Women vs men	1.21 (0.93, 1.57)	0.15
Age, years	1 (1, 1.01)	0.33
Cumulative MTX dose	1.07 (1.02, 1.17)	0.02
RA vs PsA	0.86 (0.58, 1.29)	0.47

Cumulative annual dose (per mg) defined as a time-dependent variable. HR estimates are mutually adjusted to all variables in Table 5.

When analysis was restricted to patients with at least 10 laboratory tests during the follow-up period, female gender and cumulative MTX dose were associated with an increased risk of abnormal liver test results (Table 6) but not the type of the disease.

Table 6.

HRs of variables limited to patients with liver tests in at least 10 different days during follow-up (n = 400)

	HR (95% CI)	P-value
Women vs men	1.37 (1.08, 1.74)	0.01
Age, years	1 (0.99, 1.01)	0.78
Cumulative MTX dose	1.04 (0.98, 1.11)	0.15
RA vs psoriasis	1.03 (0.78, 1.37)	0.83

HR estimates are mutually adjusted to all variables in Table 6.

Discussion

The objective of this study was to test the widely held assumption that PsA patents receiving MTX therapy are at greater risk for liver damage than RA patients receiving similar treatment. The population in our study was larger than that used as a basis for the current toxicity guidelines. The mean follow-up period was also longer than the time mentioned in other reviews and papers.

MTX is widely used in dermatology and rheumatology and has proven efficacy for the treatment of psoriasis and PsA [15]. It is the most commonly used DMARD, both alone and in combination therapies to treat RA [16]. Although the dosage is much lower than that used in oncology, toxicity has been considered a major problem and demands patient monitoring [17]. The toxicity profile of MTX given weekly at a low dose (7.5–25 mg) is markedly different from that of the drug given at a high dose (100–1000 mg/m² of body surface area per cycle), as in cancer chemotherapy. Even with low doses, side effects commonly occur.

The current study focuses on the adverse-effect profile of MTX on the liver when used in the treatment of RA and psoriasis. Different studies report different incidence of hepatic damage during MTX therapy. A previous study of 33 RA patients showed a high rate of liver function test abnormalities in 57% of the patients [18]. In the current study, there was no evidence of a difference in the risk of hepatotoxicity between RA and psoriasis patients after adjusting for the cumulative dose of MTX.

Psoriasis patients are considered to be more prone to liver damage than RA patients. Alcoholic beverage consumption, advanced age and long-term therapy compound the risk of hepatotoxicity in patients with psoriasis. The risk of serious liver disease among patients with RA who receive low-dose MTX has been reported to be less than 1 per 1000 cases after 5 years of treatment [19]. Therefore, many dermatologists recommend that liver biopsies should be performed based on the cumulative drug dose [20]. The ACR has developed guidelines for monitoring patients with RA, who are receiving MTX. These include regular monitoring (every 4–8 weeks) of the results of liver function tests, with adjustment of drug doses but without routine liver biopsies in patients with normal test results [21].

The most reliable information about organ damage is obtained by performing a liver biopsy, although this procedure is not without risks and complications. The combined sensitivity of AST, ALT and bilirubin for detecting an abnormal liver biopsy has been rated at 0.86, whereas the negative predictive value of these test results was estimated at 0.93 [22].

Hepatic injury was found in both RA and psoriasis patients. Females were affected more often than males, regardless of primary disease or age. The risk for hepatic damage rises moderately in correlation to the cumulative dose of MTX; but in contradiction to previous assumptions, the cumulative amount of MTX was more than 3.5 times greater in the RA group probably reflecting the concerns dermatologists developed prescribing MTX.

There was no significant difference between the extent of liver damage marked by liver enzyme elevation between RA and psoriasis patients. These findings contradict previous results obtained in studies comparing MTX hepatotoxicity in these two populations [23].

Our study had several limitations; the rates of other hepatotoxic factors such as use of alcoholic beverages was not known, similarly the occurrence of other hepatic comorbidities such as non-alcoholic fatty liver disease or the prevalence of infectious causes were not available. Additionally, the liver function tests were not performed uniformly, but rather with varying frequency as might be expected for ‘real life populations’. Finally, we had no available data concerning liver biopsies or non-invasive markers of fibrosis.

Conclusion

MTX is a relatively safe drug for long-term use in the doses used to treat chronic inflammatory diseases. Hepatotoxicity may develop in a large percentage of patients who are treated with this drug and is marked by an elevation of any one of the liver enzyme test results.

In our study, the predictive factors for a higher risk of liver failure were female gender and cumulative MTX dosages. We found no significant difference between the toxicity profiles of MTX in psoriasis and RA patients.

Disclosure statement: The authors have declared no conflicts of interest.

Author notes

*Howard Amital and Yoav Arnson equally contributed to this work.