Can a noninvasive test for nonalcoholic steatohepatitis be used as an alternative to liver biopsy and imaging for monitoring hepatic fibrosis in patients with psoriasis who are taking methotrexate sodium?
In this single-institution cohort study, 54 of 69 patients (78.3%) had elevated steatosis scores. Women displayed a significant correlation between worsening fibrosis score detected by NASH FibroSure and an increasing cumulative methotrexate dose; this correlation was not observed in men.
The NASH FibroSure test may be used for monitoring of hepatic fibrosis scores in patients with psoriasis who are taking methotrexate; worsening fibrosis scores can indicate the need to switch to an alternative systemic agent without performing a liver biopsy.
The long-term implications of hepatotoxic effects in patients with psoriasis remains uncharacterized, and a method is needed for the noninvasive monitoring of the development and progression of hepatic fibrosis in patients with psoriasis receiving long-term methotrexate therapy.
To evaluate if NASH FibroSure, a noninvasive test for nonalcoholic steatohepatitis (NASH) and hepatic fibrosis, can be used for patients with psoriasis to aid in determining eligibility for methotrexate sodium (MTX) therapy, monitor for the development of MTX-induced hepatotoxic effects, and monitor for worsening of hepatic fibrosis scores during MTX therapy.
Design, Setting, and Participants
A retrospective descriptive analysis was conducted among a cohort of patients with psoriasis treated with MTX who underwent NASH FibroSure testing between January 1, 2007, and December 31, 2013, at a dermatology referral center at a single institution. Data analysis was performed from January 1 to December 31, 2014.
Main Outcomes and Measures
NASH FibroSure risk scores suggesting the development and progression of hepatic fibrosis in patients with psoriasis receiving long-term MTX therapy.
Included in the institutional experience portion of the study were 129 patients with psoriasis undergoing treatment with MTX, while 107 patients (57 women and 50 men; mean [SD] age, 83.3 [13.5] years) underwent NASH FibroSure testing during MTX therapy and were eligible for correlation analysis. Of the 129 patients with psoriasis undergoing treatment with MTX, 69 (53.5%) underwent NASH FibroSure testing prior to starting MTX; 19 of those patients (27.5%) had elevated fibrosis scores, and 54 (78.3%) had elevated steatosis scores. Among the 107 patients who underwent NASH FibroSure testing during MTX therapy, the cumulative MTX dose corresponded to a statistically significant association of a higher NASH FibroSure hepatic fibrosis score in women (Spearman ρ = 0.21; P = .02) but not in men (Spearman ρ = 0.17; P = .11). All patients in the cohort except 1 were managed without a liver biopsy.
Conclusions and Relevance
The patients with psoriasis in this study had a high prevalence of elevated hepatic steatosis scores. The NASH FibroSure test can be used to monitor changes in fibrosis score in patients with psoriasis receiving MTX. In a single-institution cohort, these results suggest that NASH FibroSure may be used, especially among female patients, to help monitor for risk of worsening fibrosis during MTX therapy.
Because of its relatively low cost, infrequent dosing, and favorable side effect profile, methotrexate sodium (MTX) is a widely used first-line agent among systemic options for the long-term treatment of severe psoriasis. The most concerning potential long-term toxic effect of MTX in patients with psoriasis is hepatic fibrosis.1 Even for patients who are carefully prescreened for alcohol use and subclinical chronic viral hepatitis, the excess risk of hepatic fibrosis seems to already exist in the population.1
Several studies have demonstrated a high prevalence (20%-60%) of nonalcoholic fatty liver disease (NAFLD) in patients with plaque psoriasis.2-5 Nonalcoholic fatty liver disease encompasses a spectrum of disease ranging from mild to severe infiltration of fatty liver and ultimately hepatic scarring. Methotrexate is a potential accelerator of hepatic fibrosis progression in a population that is already at risk for hepatic fibrosis.
The current criterion standard for detecting and monitoring hepatic fibrosis in patients with psoriasis receiving long-term MTX therapy is a liver biopsy (American Academy of Dermatology consensus guidelines6). A recent systematic review by Maybury et al7 of 8 observational studies that included 429 patients with psoriasis who had a liver biopsy while undergoing MTX treatment demonstrated a pooled risk difference of 0.09 for developing significant fibrosis, as well as a pooled risk difference of 0.22 for developing any fibrosis; a clear association between progression of hepatic fibrosis and cumulative MTX dose was not detected. A liver biopsy is an invasive procedure with associated risks of complications, including sampling error (ie, missing the presence or maximum degree of scarring owing to the patchy nature of hepatic fibrosis) and diagnostic discordance among pathologists on the interpretation of fibrosis.8 Although noninvasive tests for monitoring MTX-induced hepatic fibrosis have been studied and shown to be useful in reducing the number of liver biopsies, no single test has emerged as a replacement for liver biopsy.9
Previous studies have described the use of noninvasive tests for detecting hepatic fibrosis in patients with psoriasis who are receiving long-term MTX therapy. Transient elastography is an ultrasonography-based test that measures liver stiffness. A meta-analysis indicates that transient elastography has a high degree of sensitivity (84.8%) and specificity (93.6%) for the detection of fibrosis, but it performs less well in discriminating mild from advanced fibrosis.10 However, the failure rate of hepatic ultrasonography for patients with a body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) of more than 30 is reported to be as high as 50%11; therefore, this test appears to be of questionable diagnostic value for monitoring of hepatic fibrosis in obese individuals. Procollagen III amino terminal peptide, a serologic marker of collagen turnover, has been used with some success in monitoring the progression of hepatic fibrosis in patients with psoriasis who are taking MTX, but active psoriatic arthritis may result in artificially elevated procollagen III amino terminal peptide levels.12 A family of related noninvasive tests for hepatic fibrosis is available; these tests consist of panels of serum biomarkers and patient biometrics that are put into patented mathematical models that generate scores predictive of hepatic fibrosis. A recent article by Lynch et al13 described the combined use of procollagen III amino terminal peptide levels, transient elastography, and FibroTest/FibroSure in monitoring MTX-induced hepatotoxic effects. The FibroTest/FibroSure test is a patented algorithm for generating a fibrosis score from 5 serum biomarkers.14 In the study by Lynch et al,13 a positive correlation was reported between cumulative MTX dose and FibroTest score, as well as between duration of MTX therapy and FibroTest score, suggesting that this noninvasive test has a role to play in detecting MTX-induced hepatic fibrosis.
The nonalcoholic steatohepatitis (NASH) FibroSure test (LabCorp) is similar to the FibroTest but was specifically developed for patients with NAFLD. With a computational algorithm to provide a quantitative surrogate marker of liver fibrosis, steatosis, and nonalcoholic steatohepatitis (NASH), the NASH FibroSure test uses 10 biochemical markers in combination with age, sex, height, and weight. The NASH FibroSure test is reported to have a sensitivity of 83% and a specificity of 78% in detecting risk of significant fibrosis and a sensitivity of 71% and a specificity of 72% in identifying risk of significant steatosis in patients with NAFLD.15 Given the prevalence of NAFLD in patients with psoriasis, we propose that NASH FibroSure can be used as a standalone test to monitor the progression of hepatic fibrosis in patients with psoriasis receiving long-term MTX therapy. We report our experience using this test to guide medication management in the absence of a liver biopsy.
This study was a single-institution retrospective analysis of patients with psoriasis treated with MTX who were identified through the electronic medical record system. Patients were excluded if they were younger than 18 years of age, had received MTX therapy for another indication, had received less than a 500-mg cumulative dose of MTX, or had been taking MTX for fewer than 90 days. Age, sex, BMI, diagnosis of type 2 diabetes (defined by a plasma fasting glucose level >126 mg/dL [to convert to millimoles per liter, multiply by 0.0555], a hemoglobin A1c level >6.5% [to convert to proportion of total hemoglobin, multiply by 0.01], or a random fasting glucose concentration of >200 mg/dL), and the presence of stage III chronic kidney disease (defined by 2 consecutive glomerular filtration rates <59 mL/min) were abstracted. Patient medical records were manually reviewed to ensure that NASH FibroSure had been performed as part of MTX therapy monitoring for treatment of psoriasis. The study was approved by the Marshfield Clinic institutional review board with a waiver of informed consent.
The NASH FibroSure test is a semiquantitative test developed for use for patients with suspected NAFLD; it calculates the degree of hepatic fibrosis using computational algorithms to analyze serum levels of aminotransferases, α2-macroglobulins, haptoglobin, apolipoprotein A1, cholesterol, transpeptidase, glucose, and triglycerides, as well as age, sex, height, and weight. Scores range from 0.0 to 1.0, with subscores for fibrosis (reported as a METAVIR [Meta-analysis of Histological Data in Virial Hepatitis] score from stage F0 [<0.21, no fibrosis] to stage F4 [>0.74, cirrhosis]), hepatic steatosis (METAVIR score from stage S0 [<0.30, no steatosis] to stage S3 [>0.69, marked or severe steatosis]), and NASH (reported as a METAVIR score from stage N0 [0.25, no NASH] to stage N2 [0.75, clear diagnosis of NASH]). Cumulative MTX doses were manually calculated from the medications record. Race/ethnicity was included because the cohort was anticipated to be predominantly non-Hispanic white through self-identification, which reflects the demographics of Central Wisconsin and represents a potential limitation to study generalizability. Two patients of Asian ancestry and 2 of Native American ancestry were identified in the abstractions. One Native American patient was found to have latent tuberculosis on screening and stopped MTX. For the remaining 3 patients, MTX therapy failed, and they were switched to other therapies. None achieved the minimum 500-mg cumulative MTX dose required to be included in the correlation analysis.
The observational study included 107 patients who received a diagnosis of psoriasis during the study period from January 1, 2007, to December 31, 2013. Spearman (including partial) correlation coefficients were calculated to measure the strength of the linear association between cumulative MTX dose and fibrosis score. For each correlation, a P value was derived from the 2-tailed t test, and P < .05 was considered statistically significant. To consider the potential confounding effect of age, BMI, glomerular filtration rate, or diabetes on the association between MTX dose and fibrosis scores, the correlation analyses were repeated according to the status of age (<65 vs ≥65 years), BMI (<30 vs ≥30), glomerular filtration rate (normal or abnormal), and diabetes (yes or no). In addition, the Fisher exact test was performed to assess the associations between sex and each of the following variables of interest: age, BMI, and diabetes status.
Data analysis was performed from January 1 to December 31, 2014. With 107 patients planned to be included, the study has 80% statistical power to detect an observed correlation coefficient of 0.27 or higher. The estimated statistical power was derived using the following assumptions: a type I error rate of 5% and a 2-tailed hypothesis testing is of interest. Time to event was not used to assess study outcomes, and patients who were lost to follow-up during the study period were included in the correlation analysis if they otherwise met study inclusion criteria. All data analyses were carried out using SAS, version 9.3 (SAS Institute Inc), in English.
The initial cohort consisted of 129 patients who had at least 1 NASH FibroSure determination during MTX therapy (Table 1): 69 patients had a baseline NASH FibroSure test performed before starting MTX; of these 69 patients, 54 (78.3%) had elevated steatosis scores and 18 (26.1%) had elevated fibrosis scores. Table 2 displays the mean methotrexate dose at which NASH FibroSure was checked in patients who underwent serial testing. Of 107 patients who had a NASH FibroSure test performed any time during therapy, a statistically significant correlation between worsening fibrosis scores and cumulative MTX dose was observed in women (Spearman ρ = 0.21; P = .02) but not in men (Spearman ρ = 0.17; P = .11); the correlation for women remained significant after correcting for patients who had multiple NASH FibroSure tests (Table 3). For women, there was a statistically significant correlation between a BMI of 28 or more and worsening fibrosis scores (Spearman ρ = 0.2; P = .03) (Table 4) and no significant correlation between the combination of age older than 65 years and worsening of hepatic fibrosis scores with cumulative MTX dose. There were no differences between men and women with regard to prevalence of a BMI of 28 or more, diabetes, age older than 65 years, or chronic kidney disease (Table 5). These results suggest that for women, obesity influences the progression of fibrosis scores.
Several management patterns were observed in this study of our institutional experience using the NASH FibroSure test for management of patients with psoriasis. Elevated steatosis scores were never treated as a contraindication to starting MTX therapy within the cohort, and with only 1 exception, MTX was instituted for 18 patients who had baseline elevated fibrosis scores. Six of the 18 were lost to follow-up shortly after starting MTX therapy. Owing to the development of transaminitis within the first year of therapy, 2 patients with elevated baseline fibrosis scores were switched to different treatment regimens, specifically, etanercept and topical corticosteroids. Methotrexate therapy failed in 2 other patients, who were started on anti–tumor necrosis factor agents: 1 patient received adalimumab and the second received etanercept. Eight patients responded well to MTX and continued this therapy; 1 had progressive worsening of fibrosis in spite of reducing the weekly MTX dose and, ultimately, died of unrelated causes. Of the remaining 7 patients with baseline fibrosis, 4 continued MTX therapy with stable fibrosis scores; 1 patient was the single patient in our study to undergo a liver biopsy, which confirmed steatosis but did not detect fibrosis. The remaining 3 patients switched to biologic agents, specifically adalimumab, etanercept, and ustekinumab, owing to concern for fibrosis in spite of stable fibrosis scores. Twenty patients without baseline fibrosis had worsening fibrosis scores, most often resulting in a change in therapy. No NASH FibroSure tests were ordered for patients after MTX was discontinued, so the potential changes in hepatic fibrosis in the absence of MTX could not be assessed. The dermatologists at our institution showed a strong preference for switching to biologics without further gastrointestinal workup in patients with psoriasis who developed worsening fibrosis while receiving MTX. Of 4 patients referred to a gastroenterology specialist, only 1 received a liver biopsy and another underwent ultrasonography of the liver; neither patient showed evidence of fibrosis and continued MTX therapy. Of the remaining patients with worsening fibrosis scores, 15 were offered biologics (12 of 15 received adalimumab, and 3 of 15 received ustekinumab), 1 was switched to cyclosporine, and another to acitretin.
Several noninvasive proprietary tests based on panels of serum biomarkers are available for monitoring the risk of progression of hepatic fibrosis in patients with NAFLD, with the FibroTest being the most widely adopted test evaluating MTX-induced hepatotoxic effects.9,13,16-18 The high prevalence of NAFLD among patients with psoriasis, including in our cohort, suggests that tests validated to assess the NAFLD spectrum may be useful for patients with psoriasis receiving long-term MTX therapy, particularly considering that fatty liver is considered to be a potential precursor of hepatic fibrosis. The NASH FibroSure test is unique among these tests in that it includes height and weight in the model, thereby accounting for the contribution of BMI to the progression of steatosis to fibrosis. To our knowledge, our study is the first to evaluate the utility of this test in monitoring the progression of fibrosis scores among patients with psoriasis with elevated BMI, including obesity. Although our study design does not allow us to examine the contribution of MTX to worsening fibrosis scores, our results provide support for the application of serologic tests for the progression of NAFLD to patients with psoriasis. In addition to analyzing the use of NASH FibroSure as a quantitative test to measure hepatic fibrosis scores among patients with psoriasis, we also analyzed the associations among sex and the comorbidities of obesity and diabetes with progression of fibrosis scores during MTX therapy. Further analysis of the sex differences in the progression of hepatic fibrosis according to cumulative MTX dose will aid in the development of improved treatment and monitoring recommendations for female and male patients with psoriasis.
Our cohort demonstrated a high prevalence of baseline fatty liver changes when the NASH FibroSure test was used as a screening tool. The NASH FibroSure scores suggested that fibrosis was also detected in some patients before MTX therapy, although the drug was generally initiated in these patients. The study also suggests that patients with psoriasis receiving long-term MTX therapy can be managed without a liver biopsy or gastroenterology consultation. Although cirrhosis and hepatic failure were not detected in any of the patients in our cohort, the safe upper level of an elevated fibrosis score among patients taking MTX and the rate of progression to cirrhosis among patients with psoriasis not treated with MTX are not known.
Multiple European studies have demonstrated that combinations of serum biomarkers and imaging can be used to reduce the number of liver biopsies performed.9,13,16-18 Given the significant risks and limitations of a liver biopsy, we suggest that our data support managing most patients with psoriasis receiving long-term MTX without the use of a liver biopsy. Female patients with psoriasis showed a significant correlation between fibrosis score progression and cumulative MTX dose (Table 5). A sex effect has also been reported by Amital et al,19 who observed that women with either rheumatoid arthritis or psoriasis treated with MTX had a higher risk of an abnormal liver function test result than men. In our cohort, obesity was significantly correlated with elevated fibrosis scores and cumulative MTX dose. The lack of differences between cumulative MTX dose and prevalence of obesity, diabetes, or elevated steatosis scores in the male and female subgroups within the cohort strengthens the case for a potential sex effect on MTX-induced progression of hepatic fibrosis. No differences between the cohorts were observed that would explain the association of worsening fibrosis scores and cumulative MTX dose among women but not men. The retrospective nature of our study does not allow us to assess medication adherence. However, given the implications of the progression of hepatic fibrosis, we still recommend discontinuing MTX for male patients who demonstrate worsening fibrosis scores.
There are several limitations to this study. The NASH FibroSure test is a mathematical model that generates a risk score for hepatic fibrosis, not an actual measure of fibrosis. With the exception of 2 patients, liver biopsies and imaging were not used to validate the fibrosis scores. In addition, the cohort was uncontrolled, and the change in fibrosis scores or the progression of NAFLD in patients with psoriasis remains unknown. There was heterogeneity in the departmental use of NASH FibroSure as well, with some patients already receiving long-term MTX therapy when their first NASH FibroSure test was performed. Finally, the entire correlation cohort self-identified as non-Hispanic white, limiting the ability to generalize the data to other races.
Our results do not suggest that a liver biopsy can be entirely replaced by the NASH FibroSure test; they do, however, support the idea that the number of liver biopsies can be significantly reduced by the use of noninvasive tests such as the NASH FibroSure. To address the study limitations, a prospective, randomized, multi-institutional analysis of NASH FibroSure and liver biopsies for patients with psoriasis receiving MTX vs other treatments, including a larger cohort of men and women with different racial and ethnic backgrounds, would be necessary.
Corresponding Author: Clayton Green, MD, PhD, Department of Dermatology, Marshfield Clinic, 1000 N Oak Ave, Marshfield, WI 54449 (email@example.com).
Accepted for Publication: May 5, 2017.
Published Online: August 23, 2017. doi:10.1001/jamadermatol.2017.2083
Author Contributions: Dr Green had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Bauer, Green.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Bauer, Chyou, Green.
Critical revision of the manuscript for important intellectual content: Bauer, Stratman, Green.
Statistical analysis: Bauer, Chyou, Green.
Obtained funding: Bauer, Green.
Administrative, technical, or material support: Green.
Study supervision: Stratman, Green.
Conflict of Interest Disclosures: Dr Green reported receiving research funding from T2 Biosystems. No other disclosures were reported.
Funding/Support: Funds for this study were provided by the Marshfield Clinic Resident Research Program.
Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: Deb Hansen, BSN, The Marshfield Clinic, provided administrative assistance in preparing the initial grant submission. She was not compensated for her contribution.
CH. Diagnostic accuracy of noninvasive markers of liver fibrosis in patients with psoriasis taking methotrexate: a systematic review and meta-analysis. Br J Dermatol
. 2014;170(6):1237-1247.PubMedGoogle ScholarCrossref
G. Non-alcoholic fatty liver disease in patients with chronic plaque psoriasis. J Hepatol
. 2009;51(4):758-764.PubMedGoogle ScholarCrossref
et al. Prevalence, characteristics and severity of non-alcoholic fatty liver disease in patients with chronic plaque psoriasis. J Hepatol
. 2009;51(4):778-786.PubMedGoogle ScholarCrossref
S. The increased prevalence of non-alcoholic fatty liver disease in psoriatic patients: a study from South India. Australas J Dermatol
. 2012;53(3):190-197.PubMedGoogle ScholarCrossref
et al. The prevalence of NAFLD and NASH among patients with psoriasis in a tertiary care dermatology and rheumatology clinic. Aliment Pharmacol Ther
. 2015;41(3):293-300.PubMedGoogle ScholarCrossref
CH. Methotrexate and liver fibrosis in people with psoriasis: a systematic review of observational studies. Br J Dermatol
. 2014;171(1):17-29.PubMedGoogle ScholarCrossref
CO. Monitoring methotrexate-induced hepatic fibrosis in patients with psoriasis: are serial liver biopsies justified? Aliment Pharmacol Ther
. 2004;19(4):391-399.PubMedGoogle ScholarCrossref
et al. Methotrexate in psoriasis: a systematic review of treatment modalities, incidence, risk factors and monitoring of liver toxicity. J Eur Acad Dermatol Venereol
. 2011;25(suppl 2):12-18.PubMedGoogle ScholarCrossref
et al. Diagnostic accuracy and reliability of ultrasonography for the detection of fatty liver: a meta-analysis. Hepatology
. 2011;54(3):1082-1090.PubMedGoogle ScholarCrossref
CT. Liver fibrosis screening for patients with psoriasis taking methotrexate: a cross-sectional study comparing transient elastography and liver biopsy. Br J Dermatol
. 2012;166(5):1125-1127.PubMedGoogle ScholarCrossref
L. Serum aminoterminal propeptide of type III procollagen in psoriasis and psoriatic arthritis: relation to liver fibrosis and arthritis. J Am Acad Dermatol
. 1991;25(1, pt 1):50-53.PubMedGoogle ScholarCrossref
et al. The use of transient elastography and FibroTest for monitoring hepatotoxicity in patients receiving methotrexate for psoriasis. JAMA Dermatol
. 2014;150(8):856-862.PubMedGoogle ScholarCrossref
et al; LIDO Study Group; CYTOL Study Group. Diagnostic value of biochemical markers (FibroTest-FibroSURE) for the prediction of liver fibrosis in patients with non-alcoholic fatty liver disease. BMC Gastroenterol
. 2006;6:6.PubMedGoogle ScholarCrossref
J, de Jong
et al. Biochemical and biophysical assessment of MTX-induced liver fibrosis in psoriasis patients: FibroTest predicts the presence and FibroScan predicts the absence of significant liver fibrosis. Liver Int
. 2007;27(5):639-645.PubMedGoogle ScholarCrossref
et al. Assessment of liver fibrosis with transient elastography and FibroTest in patients treated with methotrexate for chronic inflammatory diseases: a case-control study. J Hepatol
. 2010;53(6):1035-1040.PubMedGoogle ScholarCrossref
P. Assessment of methotrexate hepatotoxicity in psoriasis patients: a prospective evaluation of four serum fibrosis markers. J Eur Acad Dermatol Venereol
. 2013;27(8):1007-1014.PubMedGoogle ScholarCrossref
V. Hepatotoxicity rates do not differ in patients with rheumatoid arthritis and psoriasis treated with methotrexate. Rheumatology (Oxford)
. 2009;48(9):1107-1110.PubMedGoogle ScholarCrossref