Concomitant extrahepatic autoimmune diseases do not compromise the long-term outcomes of primary biliary cholangitis

Sha Chen , Meng-Qi Li , Wei-Jia Duan, Bu-Er Li, Shu-Xiang Li, Ting-Ting Lv, Lin Ma,Ji-Dong Jia

Liver Research Center, Beijing Friendship Hospital, Capital Medical University; Beijing Key Laboratory of Translational Medicine on Liver Cirrhosis; National Clinical Research Center for Digestive Diseases, 95 Yong’an Road, Beijing 10 0 050, China

Keywords:Primary biliary cholangitis Extrahepatic autoimmune diseases Sj?gren’s syndrome Rheumatoid arthritis Prognosis

A B S T R A C T

Introduction

Primary biliary cholangitis (PBC) is an autoimmune liver disease with pathogenesis not fully elucidated. Environmental factors may trigger immunological dysfunction in individuals with genetic susceptibility, eventually leading to the development of PBC [1] . Since PBC and other autoimmune diseases may share features of pathogenesis [ 2 , 3 ], PBC patients often have concomitant extrahepatic autoimmune (EHA) diseases including Sj?gren’s syndrome (SS), systemic sclerosis (SSc), rheumatoid arthritis (RA), and autoimmune thyroid disease. To date, several studies [4–6] have reported the varied prevalence of EHA diseases in PBC, with most of them either including a small sample size or without standardized diagnostic criteria. More importantly, most of these studies were from Western countries and few reports focused on Asian populations.In addition, the impact of EHA diseases on the prognosis of PBC remains controversial [6–9] .

The present study aimed to describe the prevalence of common and uncommon EHA diseases in PBC, and to explore the impact of EHA diseases including SS, RA and Hashimoto’s thyroiditis (HT) on the long-term outcomes of PBC in China.

Methods

Patients

We analyzed all PBC patients who were diagnosed at Beijing Friendship Hospital, Capital Medical University from January 20 0 0 to December 2020. We included all patients who were diagnosed with PBC according to the international guidelines [ 1 , 10 ].

Patients were excluded if they were 1) concomitant with drug-induced liver injury (DILI), autoimmune hepatitis (AIH), and chronic hepatitis B or C; 2) with missing key baseline clinical data;3) without ursodeoxycholic acid (UDCA) therapy or with poor compliance; and 4) lost to follow up or followed up for less than 1 year.

Data collection and follow-up procedures

Baseline information including age, sex, anti-mitochondrial M2 antibody (AMA-M2) status, AMA status, antinuclear antibody(ANA) status, smooth muscle antibody (SMA) status, HBsAg status, anti-HBc status, liver histology, and UDCA treatment (start time and dose) was collected. Laboratory data including blood routine, liver biochemistry, the international normalized ratio, immunoglobulin levels, and abdominal ultrasound at baseline and 12 months after UDCA therapy were recorded.

Two doctors followed up PBC patients by a structured, standardized telephone interview and systematic review of the electronic database. Smoking history, alcohol consumption, body mass index (BMI), family history, dosage and duration of UDCA treatment, concomitant EHA disease affecting all organ systems (the time of diagnosis, treatment, related clinical, laboratory or histological results were reviewed), abnormal results of imaging examinations (fatty liver, ascites, etc.), the occurrence time of hepatocellular carcinoma (HCC), decompensated cirrhosis (variceal bleeding,ascites, or hepatic encephalopathy), liver transplantation (LT), and liver-related death were recorded.

Definition and diagnostic criteria

PBC with EHA disease was defined as PBC patients with at least one EHA disorder. SS was diagnosed according to the 2002 American-European classification criteria [11] ; RA was diagnosed according to the revised 2010 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) diagnostic criteria [12] ; HT was diagnosed according to the clinical manifestations, the presence of serum anti-thyroid antibodies and thyroid ultrasound [13] ; other EHA diseases were diagnosed according to reports of other physicians (rheumatologists, dermatologist, etc.).

Smoking habit was defined as smoking more than 100 cigarettes lifetime [ 14 , 15 ]. Alcohol abuse was defined as an average alcohol consumption > 40 g per day for men or > 20 g per day for women for longer than 5 years [16] . Past HBV infection was defined as positive anti-HBc with negative HBsAg.Cirrhosis [17] and HCC [18] were diagnosed based on the published guidelines. Aspartate aminotransferase (AST)/platelet ratio index (APRI) high risk was defined as APRI score > 0.54 [19] ;GLOBE high risk was defined as GLOBE score > 0.3 [20]. ANA positivity was defined as ANA titer ≥1:160.

The duration of follow-up was calculated as the time interval between the start date of UDCA therapy and the date of last visit or death or LT. Transplant-free survival was defined as survival without LT or liver-related death.

Statistical analyses

Categorical variables were expressed as counts and percentages. Continuous variables were described as median (interquartile range). Mann-WhitneyUtest was used for comparing characteristics of age, BMI, liver biochemistries, and immunoglobulin levels at baseline between patients with and without EHA diseases.Chi-square test was applied to compare smoking habit, alcohol consumption, family history, the status of AMA/AMA-M2, cirrhosis, histological stage, biochemical response and clinical outcomes between the two groups.

Transplant-free survival was analyzed by Kaplan-Meier curves(difference was calculated by log-rank test), and Cox proportionalhazard model was applied in order to determine the risk factors associated with long-term outcomes. A two-sidedPvalue < 0.05 was considered statistically significant. The analyses were conducted by using SPSS statistics version 26 (SPSS Inc., Chicago, IL,USA).

Table 1 Prevalence of extrahepatic autoimmune diseases in patients with primary biliary cholangitis ( n = 985).

Results

Prevalence of EHA diseases in PBC patients

A total of 1517 patients were diagnosed as PBC from January 20 0 0 to December 2020 in our institute, of which 985 fulfilled the inclusion criteria ( Fig. 1 ). Among the enrolled 985 patients, 738 were without EHA diseases (group 1), 247 with at least one EHA diseases (group 2), of which 55 had more than two EHA diseases(EHA ≥2). The prevalence of each EHA disease in PBC patients was summarized in Table 1 . SS was the most frequent disease coexisted with PBC (n= 140), followed by RA (n= 56) and HT (n= 45).

Comparison of baseline characteristics

The baseline characteristics were shown in Table 2 . Patients with EHA diseases were more common in females (P< 0.001) and in those with a family history of autoimmune disease (P= 0.017).Patients with EHA diseases had lower total bile acid (TBA) and gamma-glutamyl transpeptidase (γ-GT) than patients without EHA diseases. However, other liver biochemistry and the proportion of cirrhosis were comparable between the two groups.

Comparison of biochemical response to UDCA and long-term outcomes

UDCA response rates were comparable between the two groups according to the Barcelona, Paris I and Paris II criteria. Similarly,risk stratification defined by APRI score and GLOBE score showed no difference in the proportion of high-risk patients between the two groups ( Table 3 ).

Fig. 1. The flowchart of patients enrollment. PBC: primary biliary cholangitis; AIH: autoimmune hepatitis; DILI: drug-induced liver injury; UDCA: ursodeoxycholic acid; EHA:extrahepatic autoimmune.

Table 2 Baseline characteristics of primary biliary cholangitis patients with and without extrahepatic autoimmune diseases.

During a median follow-up duration of 5.3 (2.7-8.6) years,24 (2.4%) patients developed HCC, 24 (2.4%) patients received LT, and 76 (7.7%) patients had liver-related death.Among 845 patients without decompensated cirrhosis at baseline, 172 (20.4%) developed decompensated cirrhosis. The median follow-up time was 6.7 (3.3-9.9) years for group 1 and 5.0 (2.7-8.3) years for group 2. Overall, there were no significant differences in the incidence of decompensation, HCC, and LT/liver-related deaths between the two groups( Table 3 ).

Table 3 Biochemical response and long-term outcomes of primary biliary cholangitis patients with and without extrahepatic autoimmune diseases.

Table 4 Univariate Cox regression analysis for long-term outcomes of PBC patients.

Comparison of transplant-free survival and incidence of hepatic events

Consistent with the abovementioned results, the transplant-free survival and cumulative incidence of hepatic events (decompensation cirrhosis and/or HCC) were similar between the two groups( Fig. 2 A and B,P> 0.05). However, transplant-free survival was better in patients with EHA ≥2 than that in those without ( Fig. 2 C,P= 0.015). In addition, transplant-free survival of patients with RA seemed better than that of those without, although the results were not statistically significant ( Fig. 2 E,P= 0.079). Consistent with these results, the cumulative incidence of hepatic events was significantly lower in patients with EHA ≥2 (P= 0.019) and patients with RA (P= 0.001) than those without. Additionally,Kaplan-Meier plots for survival and incidence of hepatic events showed no difference between patients with and without SS or HT(Fig. S1).

Risk factors associated with long-term outcomes of PBC patients

Univariate analysis showed that patients with EHA ≥2 had significantly lower liver-related death/LT, and that EHA ≥2 and RA were protective factors of decompensation cirrhosis and/or HCC( Table 4 ). However, all results became insignificant in multivariate regression analysis after including age, sex, bilirubin, and cirrhosis(Table S1).

Discussion

To our knowledge, this is the largest cohort evaluating the impact of concomitant EHA diseases on the prognosis of PBC patients in the Asia-Pacific area. In this cohort, we reported the prevalence of common and uncommon EHA diseases in PBC patients.We found that concomitant EHA diseases were more common in female PBC patients and those with a family history of autoimmune diseases, and PBC patients concomitant with EHA diseases tended to be less severe at baseline. More importantly, concomitant EHA diseases did not compromise the long-term outcomes of PBC patients.

The top three most frequent EHA diseases in PBC patients in our cohort were SS, RA and HT, which were identical to the other two studies in Europe [ 15 , 21 ]. As the most frequent coexisting EHA disease, SS has been reported to have a prevalence of 3.5%-73%in PBC [4–7] . This great variation might be caused by different study populations, sample sizes and inconsistent diagnostic criteria. Our cohort reported a 14.2% prevalence of SS using 2002 American-European classification criteria [11] , which was similar to another cross-sectional study with a large sample size in Japan(13.5%) [22] . The reported prevalence of RA in PBC patients ranges from 1.8% to 10% [ 5 , 15 , 21-24 ], which were mainly from United States and European countries. We identified 56 cases (5.7%)with RA in our cohort, which was similar to a Japanese cohort(5.6%) [23] . HT is another common disease in PBC patients but is rarely reported in Asia, with a frequency of 9%-20.4% in Western PBC cohorts [ 5 , 15 , 21 , 25 ]. The prevalence of HT in our PBC cohort(4.6%) was lower than that of Western countries, but similar to that of two Japanese cohorts (4.4%-5.6%) [ 22 , 23 ]. Overall, the incidences of other EHA diseases including SSc were relatively low in our cohort, and their association with PBC needs to be further confirmed.

One key observation in our cohort was that concomitant EHA diseases did not compromise the long-term outcomes of PBC patients. So far, the impact of EHA diseases on the prognosis of PBC remains controversial. Some studies [ 6 , 8 ] reported higher all-cause mortality and higher incidence of spontaneous bacterial peritonitis and interstitial lung disease in PBC with SS, whereas a study found a similar UDCA response rate and comparable incidence of hepatic events between PBC alone and PBC with SS [7] . Similarly,anticentromere antibody (ACA), known as a specific antibody for SSc, was considered a risk factor for portal hypertension and associated with ductular reaction in PBC patients [26] . However, the incidences of liver-related death and LT were lower in PBC patients with SSc compared with those in PBC alone [9] .

Fig. 2. Kaplan-Meier plots for transplant-free survival and cumulative incidence of hepatic events in primary biliary cholangitis patients with and without extrahepatic autoimmune diseases. EHA: extrahepatic autoimmune; RA: rheumatoid arthritis.

In our cohort, we found that the transplant-free survival and incidence of hepatic events had no differences between PBC patients alone and PBC with EHA, there were also no significant differences between PBC with and without SS or HT (Fig. S1), which were consistent with the previous studies [ 5 , 7 , 21 , 25 ]. We also found that PBC patients with RA or EHA ≥2 had better prognosis compared with those with PBC alone. In addition, UDCA response rates were higher in patients with RA according to Paris I and Paris II criteria(Table S2). Two hypotheses may explain this result.

Firstly, PBC patients with RA or EHA ≥2 were less severe at baseline. We found that PBC patients with RA had higher platelet count (PLT), lower baseline transaminase,γ-GT, total bilirubin,TBA, and proportion of cirrhotic patients than PBC alone (Table S3),indicating that patients with RA were more often diagnosed in the early stage of PBC. PBC patients with EHA ≥2 had similar clinical manifestations to those with RA (Table S4). Efe and coworkers also observed a trend that PBC with EHA had less advanced fibrosis and milder liver damage (lower AST and bilirubin, etc.) at diagnosis [5] .Indeed, patients with RA are more prone to specific symptoms like joint pain, so they are more likely to seek medical attention early.More patients (n= 26) diagnosed with RA first in our cohort, with a median of 13 (5-26) years before the diagnosis of PBC, which were consistent with another study [27] .

Secondly, there is a possibility that concomitant EHA diseases might relieve the damage to the liver but aggravate systemic diseases. One supporting evidence is that PBC with SSc had lower liver-related mortality and incidence of liver failure, but higher all-cause mortality than patients with PBC alone [9] . In line with this, studies have reported a similar incidence of hepatic events and higher all-cause mortality in PBC patients with SS than in PBC alone [ 7 , 8 ]. Nevertheless, this hypothesis needs to be further confirmed.

This study has several limitations. Firstly, the diagnosis of EHA diseases except for SS, RA and HT was mainly made by other physicians instead of the investigators of the present study. Secondly,nearly 19% of patients were lost to follow up and some of the laboratory data were missing, but we hope the large sample size and long follow-up duration can make up for this weakness. Thirdly,we are unable to determine the causal relationship between EHA diseases and PBC. Lastly, we did not discuss the impact of medications for concomitant EHA disease on the natural history of PBC.

In conclusion, we found that EHA diseases were common in PBC patients but the concomitant EHA diseases did not compromise the long-term outcomes of PBC.

Acknowledgments

We thank Dr. Dong Zhang, Hong Ma, Xiao-Juan Ou and Hong You from Beijing Friendship Hospital, Capital Medical University for helpful advice on this study and Dr. Yuan-Yuan Kong from Beijing Friendship Hospital, Capital Medical University for assistance with statistical analysis.

CRediT authorship contribution statement

Sha Chen: Conceptualization, Data curation, Formal analysis,Writing original draft. Meng-Qi Li: Data curation, Formal analysis, Investigation. Wei-Jia Duan: Conceptualization, Data curation.Bu-Er Li: Data curation, Investigation. Shu-Xiang Li: Data curation,Funding acquisition, Investigation. Ting-Ting Lv: Data curation, Investigation. Lin Ma: Data curation, Investigation. Ji-Dong Jia: Conceptualization, Funding acquisition, Supervision, Writing review &editing.

Funding

This study was supported by grants from the National Natural Science Foundation of China (820 0 0533 and 81770598).

Ethical approval

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with theHelsinkiDeclarationof1975, as revised in 2008. The Ethics Committee of Beijing Friendship Hospital, Capital Medical University reviewed and approved the study.

Competing interest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.hbpd.2022.05.009 .