Serum vitamin D level is associated with disease severity and response to ursodeoxycholic acid in primary biliary cirrhosis

Transcription

1 Alimentary Pharmacology and Therapeutics Serum vitamin D level is associated with disease severity and response to ursodeoxycholic acid in primary biliary cirrhosis G.-Y. Guo*,,1, Y.-Q. Shi*,,1, L. Wang*,,1, X. Ren*,, Z.-Y. Han*,, C.-C. Guo*,, L.-N. Cui*, J.-B. Wang*, J. Zhu*, N. Wang*,, J. Zhang*,, Y. Cai*,, Y. Han*,, X.-M. Zhou*, & D.-M. Fan*, *Division of Hepatology, Xijing Hospital of Digestive Diseases, Fourth Military Medical University, Xi`an, China. State Key Laboratory of Cancer Biology, Fourth Military Medical University, Xi`an, China. Correspondence to: Dr Y. Han and Dr X.-M. Zhou Division of Hepatology, Xijing Hospital of Digestive Diseases, Fourth Military Medical University, 127 Changle West Road, Xi`an 7132, China. hanying1@fmmu.edu.cn; zhouxmm@fmmu.edu.cn 1 These authors contributed equally to this work. Publication data Submitted 15 November 214 First decision 13 December 214 Resubmitted 6 February 215 Resubmitted 8 April 215 Resubmitted 24 April 215 Accepted 25 April 215 EV Pub Online 18 May 215 This article was accepted for publication after full peer-review. SUMMARY Background Serum vitamin D levels are associated with bone complications in patients with primary biliary cirrhosis (PBC). Increasing evidence suggests a nonskeletal role of vitamin D in various autoimmune and liver diseases. Aim To investigate the clinical relevance of vitamin D levels in PBC, especially their association with the therapeutic effects of ursodeoxycholic acid (UDCA). Methods Consecutive PBC patients were retrospectively reviewed. 25-hydroxyvitamin D [25(OH)D] levels were determined in frozen serum samples collected before initiation of UDCA treatment. Response to UDCA was evaluated by Paris-I and Barcelona criteria. Logistic regressions were performed to identify the treatment response-associated parameters. Results Among 98 patients, the mean serum 25(OH)D concentration was ng/ ml. 25(OH)D levels decreased with increasing histological stage (P =.29) and were negatively correlated with bilirubin and alkaline phosphatase levels. After 1 year of UDCA therapy,31 patients failed to achieve complete response according to Paris-I criteria. The baseline 25(OH)D level was significantly lower in nonresponders ( vs ng/ml, P =.5). Vitamin D deficiency at baseline was associated with an increased risk of incomplete response independent of advanced stages (OR = 3.93, 95% CI = , P =.47). Similar results were obtained when biochemical response was evaluated by Barcelona criteria. Furthermore, 25(OH)D levels were lower in patients who subsequently suffered death or liver transplantation ( vs ng/ml, P =.23). Conclusions 25(OH)D level is associated with biochemical and histological features in PBC. Pre-treatment vitamin D status is independently related to subsequent response to UDCA. Our results suggest that vitamin D status may have important clinical significance in PBC. Aliment Pharmacol Ther 215; 42: doi:1.1111/apt.13244

2 G.-Y. Guo et al. INTRODUCTION Primary biliary cirrhosis (PBC) is an immune-mediated chronic cholestatic liver disease characterised by ongoing inflammatory destruction of the interlobular and septal bile ducts, eventually leading to the development of biliary cirrhosis. 1 Currently, ursodeoxycholic acid (UDCA) is the only FDA-approved drug for PBC with demonstrated clinical benefits for the liver biochemistry and, potentially, for survival. 2, 3 However, up to nearly onethird of patients may not achieve a complete treatment response. 4 7 Primary biliary cirrhosis results in an increased risk of fat-soluble vitamin malabsorption and deficiency due to the loss of functioning intrahepatic bile ducts and inadequate bile salt secretion. Vitamin D deficiency in PBC patients has been well documented by early studies, in which low serum 25-dihydroxyvitamin D [25(OH)D] concentrations were detected in PBC patients Given the crucial role of vitamin D in modulation of bone metabolism, such aberrations in vitamin D endocrine system are always associated with bone disorder in PBC, a complication occurring in nearly one-third of PBC 11, 12 patients. Experimental evidence from recent years suggests potential protective abilities of the vitamin D endocrine system in cholestasis, cirrhosis and autoimmunity, all of which are crucial to the progression of PBC. A normal vitamin D VDR (vitamin D receptor) axis may be important for hepatobiliary homoeostasis and bile duct 13, 14 integrity in cholestatic mice, and vitamin D can modulate hepatic fibrogenesis during the wound-healing process. 15 The immunoregulatory properties of vitamin D in immune-cell activation and proliferation, T-helpercell differentiation and humoral immunity are well demonstrated. 16 These nonskeletal properties indicate that vitamin D may have important clinical significance in PBC beyond the simple link with bone complications. In recent years, serum Vitamin D status has been associated with disease activity, severity and prognosis in various autoimmune and liver diseases, including systemic lupus erythematosus (SLE), 17 multiple sclerosis (MS), 18, 19 2, 21 inflammatory bowel disease (IBD), chronic hepatitis C (CHC), 22 chronic hepatitis B (CHB), 23 non-alcoholic fatty liver disease (NAFLD) 24 and alcoholic liver disease (ALD). 25 Adeficit in vitamin D appears to be a risk factor for surgery, hospitalisations and cancer in IBD 26, 27 ; poor response to anti-viral therapy in CHC 22, 28 ; and mortality in ALD 25 and hepatocellular carcinoma. 29 Consequently, we performed this study to examine whether vitamin D serum levels are associated with histological severity, liver function degree, prognosis, and especially, the response to UDCA in PBC. PATIENTS AND METHODS Study population This study was conducted in 98 consecutive PBC patients diagnosed at Xijing Hospital of Fourth Military Medical University (Xi`an, China) between January 27 and December 29. Diagnosis criteria were: (i) intrahepatic cholestasis, (ii) presence of AMA and (iii) typical bile ducts lesions on liver histology. All patients fulfilled at least two of these three criteria and were treatmentnaive for UDCA at presentation. Serum samples 2, 3 were collected at the time of diagnosis. Patients concomitant with HBV, HCV, human immunodeficiency virus infection, excessive alcohol consumption, Wilson s disease, hepatic or extra-hepatic malignancies, inflammatory bowel disorders, or coeliac disease were excluded, as well as those who showed features such as high transaminase and IgG levels that were suggestive of AIH/PBC overlap syndrome. We also specifically excluded patients who used vitamin D supplements, corticosteroids, cholestyramine, anticonvulsants, anti-tuberculosis or cimetidine within 3 months before diagnosis, due to the effects of these drugs on serum vitamin D levels. 31 Informed consent was obtained in writing from each patient, and the study protocol was approved by the ethical committee of the Fourth Military Medical University. Clinical and laboratory features at presentation A total of 84 patients underwent a liver biopsy at baseline in the present study. The histological stage was defined based on Ludwig s classification. 32 Clinical examinations and conventional biochemical tests were performed in all patients before the initiation of therapy. The presence of serum AMA was determined at diagnosis by indirect immunofluorescence of Hep-2 cells and rat tissue slides. The serum immunoglobulin levels were assessed by immunonephelometry, and the Mayo risk scores were calculated according to the original model. Records of demographic and clinical characteristics at baseline were extracted from case records, including age, sex, AMA, body mass index (BMI), ALT (alanine aminotransferase), AST (aspartate aminotransferase), ALP (alkaline phosphatase), GGT (gamma-glutaryl-transferase), albumin, total bilirubin, international normalised 222 Aliment Pharmacol Ther 215; 42:

3 Clinical relevance of vitamin D level in PBC ratio (INR) and Mayo risk score. The records of bone mineral density based on dual-energy X-ray absorptiometry were collected. The lowest T-score was used for analyses if two or more records from different bones existed. Follow-up and response definition Upon PBC diagnosis, all patients were treated with UDCA (Dr. Falk Pharma GmbH, Freiburg, Germany) at stable doses of mg/kg/day and regularly followed up for more than 12 months. The biochemical response to UDCA was assessed by the Paris-I and Barcelona criteria. 4, 5 Serum samples and serum vitamin D assessment Blood samples were routinely collected from each subject at the time of diagnosis. They were drawn in the fasting condition and centrifuged within 3 min of collection. After centrifugation, the serum samples were frozen at 8 C or in liquid nitrogen before analysis. 25-hydroxyvitamin D [25(OH)D] is the main circulating form of vitamin D and is considered as the most reliable indicator for assessing vitamin D status. The serum 25(OH)D levels in the present study were measured by batched electrochemiluminescence immunoassay (Elecsys 25(OH)D Total Assay; Roche diagnosis, Basel, Schweiz). The within-batch coefficient of variation for the test at 5.6% and the measuring range was from 4 to 1 ng/ml. Seasonality was defined as winter spring from December to April and summer autumn from May to October according to previous report 33 and local seasonal climate. Vitamin D deficiency was defined as a 25(OH)D level below 2 ng/ml based on recommendations of the IOM and the Endocrine Society. Statistical analyses Statistical analyses were performed with SPSS 19. (IBM, Armonk, NY, USA) and GraphPad Prism5 (GraphPad Software Inc., La Jolla, CA, USA). All analyses were twosided, and P <.5 were considered statistically significant. Continuous variables with normal distribution are presented as means s.d., and comparisons were made between groups using Student s t-tests. Where data were distributed nonnormally, they are presented as median and 25th 75th interquartile range, and comparisons were made by the Mann Whitney U-test. Differences in categorical variables were determined using the v 2 and Fisher s exact tests. Pearson correlation analysis was done to assess the correlation of 25(OH)D level with other continuous biochemical tests values, while Spearman s rank correlation coefficient for histological stage. The values of ALP, ALT, AST, GGT and bilirubin analysed in Pearson correlation were log e transformed to ensure normal distributions. Univariate and multivariate logistic regression analyses were performed to identify the baseline variables associated with a subsequent response to UDCA. The candidate variables included the parameters linked to disease severity and those having been reported to be associated with UDCA response in previous studies. 25(OH)D levels in regression analyses were all adjusted for seasonality. Variables that achieved significance (P <.5) in the univariate analysis were entered into a backward-step multivariate logistic regression with an entry probability of.1 and a removal probability of.15. The rate of transplant-free survival was estimated by the Kaplan Meier method. RESULTS General characteristics of the study population The demographic and clinical features of the study cohort at baseline and after 1 year of UDCA treatment are presented in Table 1. A total of 98 patients with PBC were included in the study, namely, 85 women and 13 men, with a mean age at diagnosis of 53 years. AMA was detected in 88 (9%) patients at baseline. The Mayo risk score at baseline was A liver biopsy was obtained in 84 patients, 16 of whom were stage I, 37 were stage II, 21 were stage III and 1 were stage IV. The biochemical parameters of each disease stage are presented in Table S1. Serum 25(OH)D level and its relationship with other baseline parameters The mean serum 25(OH)D concentration of the patients was ng/ml. Fifty-six (57%) patients had 25 (OH)D levels below 2 ng/ml and were classified as having a vitamin D deficiency. Additionally, 43% of the serum samples were obtained during the winter spring season, with lower 25(OH)D levels compared to those obtained in the summer autumn season ( vs ng/ml, P <.1). No difference in 25(OH)D levels was found between genders ( ng/ml in men vs ng/ml in women, P.5). Seven patients underwent bone density scan at diagnosis or during the first year of follow-up, and bone complications were found in five of them (three with osteopenia and two with osteoporosis). No differences in 25(OH)D levels were observed in these patients ( , P.5 compared with others without bone complications), and Aliment Pharmacol Ther 215; 42:

4 G.-Y. Guo et al. Table 1 General characteristics of the study population (N = 98) At baseline After 1 year Age (years) BMI Female gender (%) 85 (87%) AMA-positive (%) 88 (9%) ALT (IU/L) 58 (35 91) 33 (24 56)** AST (IU/L) 61 (37 73) 31 (21 47)** ALP (IU/L) 245 (163 41) 163 ( )** GGT (IU/L) 29 ( ) 117 (62 251)** Total bilirubin (mg/dl) 1.1 (.9 1.7).8 (.7 1.1)** Albumin (g/l) 4.4 ( ) 41.6 ( )* Platelet count (91 9 /L) * IgM (g/l) ** Mayo risk score Histology stage (%, n = 84) I 16 (19%) II 37 (44%) III 21 (25%) IV 1 (12%) Late (III, IV) 34 (37%) UDCA response Responders Paris-I criteria (%) 67 (68.4%) Responders Barcelona criteria (%) 61 (62.2%) Vitamin D Serum 25(OH)D (ng/ml) Vitamin D deficiency (%) 56 (57%) Winter spring season (%) 42 (43%) AMA, anti-mitochondrial antibody; BMI, body mass index; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; GGT, gamma-glutaryl-transferase; IgM, immunoglobulin M; UDCA, ursodeoxycholic acid. Continuous data are presented as means s.d. or as medians with interquartile range and categorical data are shown as number with percentage. Comparisons between biochemical variables before and after 1 year of UDCA treatment were performed using the paired Student s t-test or Wilcoxon signed-rank test. *P <.1; **P <.1. Paris-I criteria: ALP <3-fold the upper limit of normal, AST <2-fold the upper limit of normal and bilirubin 1 mg/dl after 12 months of UDCA treatment. Barcelona criteria: ALP decrease >4% of pre-treatment levels or normalisation after 12 months of UDCA treatment. Seasonality was defined as winter spring from December to April and summer autumn from May to October according to previous report and the local seasonal climate. no correlation between the 25(OH)D levels and T-scores was identified. No patient received vitamin D supplementation before or at the time of diagnosis. In six patients, supplementation was started during the first year of follow-up at doses ranging from 8 to 1 IU daily. Ludwig s classification was used to evaluate the histological severity, and the mean 25(OH) D levels in patients with stage I, II, III and IV were , , and respectively (P for trend =.29) (Figure 1). Moreover, a higher rate of vitamin D deficiency was observed among patients in the later stages (stage III IV) than those in the moderate stages (74% vs. 49%, P =.38) (Figure 1). The 25(OH) D level was inversely correlated with typical cholestatic parameters of PBC, including the bilirubin and ALP levels (r =.38 and.37, respectively, both P <.1). The results of correlation analyses for other biochemical indexes of liver function are presented in Table S2. Similar inverse correlations of 25(OH)D with AST levels, Mayo risk scores, and APRI values were identified, as well as a positive correlation with the albumin level. The clinical and laboratory characteristics were compared between patients with and without a deficiency in 25 (OH)D (Table 2). The age, gender and BMI were similar between the two groups. Patients with vitamin D deficiency had a higher prevalence of advanced stages (47% vs. 23%, P =.38) and significantly higher serum values 224 Aliment Pharmacol Ther 215; 42:

5 Clinical relevance of vitamin D level in PBC (a) 25(OH)D (ng/ml) (c) Bilirubin (Ln-mg/dL) (b) 5 P = ± 7.5 Stage I 18.9 ± 7.8 Stage II 15.9 ± 6.3 Stage III 14.3 ± 6.4 Stage IV Proportion of vitamin D deficiency (%) % P =.38 74% (d) 2 1 r =.38 (P <.1) r =.37 (P <.1) (OH)D (ng/ml) 25(OH)D (ng/ml) ALP (Ln-IU/L) Stage I-II Stage III-IV Figure 1 Lower 25(OH)D levels are associated with more severe histological and cholestatic features in patients with PBC. (a) Distribution of 25(OH)D levels according to histological stages in patients with PBC. 25(OH)D levels decreased with increasing histological stage (ANOVA for trend with P =.292). (b) Patients in late histological stages (stage III IV) had higher rate of vitamin D deficiency (74% vs. 49%, P =.38). (c and d) Linear coefficient between 25(OH)D levels and serum cholestasis parameters in patients with PBC. Decreased 25(OH)D levels was significantly correlated with increased bilirubin and alkaline phosphatase values (r =.38 for bilirubin and r =.37 for bilirubin, both P <.1). The serum values of bilirubin and alkaline phosphatase were log e logarithmic transformded for normal distribution. of ALP and bilirubin, as well as higher Mayo risk scores and INR values. Association between pre-treatment vitamin D status and subsequent response to UDCA After 1 year of UDCA treatment, the abnormal values of AST, ALT, ALP, GGT, bilirubin and albumin were significantly improved (P <.1 for all variables) compared with baseline (Table 1). A total of 67 patients (responders, 68.4%) had a favourable response to UDCA according to the Paris-I criteria, while application of the Barcelona criteria identified 61 (62.2%) patients as responders. The biochemical parameters of responders and nonresponders are presented in Table S3. Nonresponders had higher AST, ALP and bilirubin values on presentation. The 25(OH)D levels were significantly lower in the nonresponders than responders ( ng/ml vs ng/ml, P =.5 according to the Paris-I criteria, and ng/ml vs ng/ml, P =.2 according to the Barcelona criteria) (Figure 2). Similarly, a higher number of vitamin D-deficient patients failed to achieve a complete response after 1 year of UDCA therapy according to the Paris-I criteria (41% vs. 19%, P =.28). The Barcelona criteria yielded similar results (48% vs. 24%, P =.2) (Figure 2). We performed univariate and multivariate analyses to examine the factors associated with response to UDCA (Table S4). In the final multivariate model, vitamin D deficiency was independently associated with an increased risk of an incomplete response to UDCA [odds ratio (OR): 3.93; 95% confidence interval (CI): ; P =.47 according to the Paris-I criteria]. Each 1 ng/ml increase in serum 25(OH)D was associated with a 1% reduction in the risk (OR:.9; 95% CI:.81.99; P =.26). In the final model, severe histological stages, abnormal bilirubin values and elevated ALP levels were also associated with an incomplete response. Similar results were obtained when the biochemical response was evaluated with the Barcelona criteria. Association of vitamin D with long-term mortality The patients were followed up for a mean period of months. Seven patients died due to liver Aliment Pharmacol Ther 215; 42:

6 G.-Y. Guo et al. Table 2 Baseline characteristics of patients according to vitamin D status Vitamin D deficiency (25(OH)D < 2 ng/ml) n = 56 Vitamin D adequate (25(OH)D 2 ng/ml) n = 42 P-value Age (years) N.S. BMI N.S. Female gender (%) 49 (88%) 36 (86%) N.S. Serum 25(OH)D (ng/ml) <.1 ALT (IU/L) 57 (34 91) 6 (38 93) N.S. AST (IU/L) 63 (35 78) 59 (37 73) N.S. ALP (IU/L) 313 (217 43) 187 ( ).5 GGT (IU/L) 353 (174 55) 231 (95 378).72 Total bilirubin (mg/dl) 1.7 (.9 2.2) 1. (.7 1.2).1 Albumin (g/l) 39.8 ( ) 41.4 ( ).8 Platelet count (x1 9 /L) APRI INR IgM (g/l) N.S. Mayo risk score Late stage (III, IV) 47% 23%.38 BMI, body mass index; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; GGT, gammaglutaryl-transferase; APRI, AST/platelet ratio index; INR, internationalised ratio; IgM, immunoglobulin M; N.S., not significant. Continuous data are presented as means s.d. or as medians with interquartile range and categorical data are shown as percentages. Student s t-test, Mann Whitney U-test and Fisher s exact test were used. related incidences after months of treatment, one as a consequence of hepatocellular carcinoma and six due to terminal liver failure. One patient underwent liver transplantation during the study period. The serum 25 (OH)D levels in patients who suffered death or transplant were significantly lower compared to others ( vs ng/ml, P =.23; Figure 3). The Kaplan Meier curve, followed by the log-rank test, showed a trend towards decreased transplantation-free survival among patients with vitamin D deficiency (P =.63; Figure S1). DISCUSSION This study investigated the nonskeletal clinical significance of vitamin D status in PBC patients. We show here that 25(OH)D levels are associated with cholestatic severity, liver dysfunction and histological stage. A vitamin D deficiency at baseline was independently related to a high risk of subsequent incomplete response to UDCA. In recent years, low serum levels of vitamin D have been shown to be associated with the severity of liver damage in NAFLD, ALD, CHB and CHC , 34, 35 In line with these observations, we found that 25(OH)D levels were significantly correlated with major baseline hepatic dysfunction indexes in PBC patients, including the histological stage, Mayo risk score and serum values of AST, ALP, albumin and bilirubin. Our findings confirm and extend the limited data from previous vitamin D studies in PBC, which were mainly concerned with fat-soluble vitamin malabsorption and bone disorders. 36, 37 Recently, a multi-centre study by Levin et al. 38 which included 79 PBC patients (75% were UDCA treated) showed that vitamin D level was associated with disease severity. They found that 25(OH)D levels were lower in patients with advanced liver disease. 38 Taken together, our data in treatment-na ıve Chinese patients (and other studies) suggest that vitamin D status seems to be an additional marker of PBC severity. As the parameters associated with vitamin D such as Mayo risk score, serum bilirubin and ALP value are well-established predictors of prognosis in PBC, 39, 4 it is interesting to see if vitamin D may correlate with nonbaseline characteristics. The main interesting finding of this study is that pre-treatment 25(OH)D serum levels were associated with the response to UDCA. A vitamin D deficiency at baseline may be associated with an increased risk of an incomplete response to UDCA, and tends to be associated with long-term survival. These findings are consistent with the important role of vitamin D in cholestasis and immune response. Vitamin D was shown to regulate bile acid metabolism and 226 Aliment Pharmacol Ther 215; 42:

7 Clinical relevance of vitamin D level in PBC (a) 5 P =.5 P =.2 Figure 2 Association of pretreatment vitamin D status and the subsequent response to UDCA evaluated by Paris-I criteria and Barcelona criteria respectively. (a) Distribution of baseline serum 25(OH)D levels stratified by response to UDCA. The patients failed to achieve complete response (nonresponders) had lower baseline 25(OH)D levels than responders no matter which response criteria was used. (b) Patients with vitamin D deficiency (25(OH) D < 2 ng/ml) at baseline had incomplete response to UDCA more frequently (41% vs. 19%, P =.28 via Paris-I criteria; 48% vs. 24%, P =.2 via Barcelona criteria). (b) 25(OH)D (ng/ml) Proportion of non-responder (%) ± % 25(OH)D 2 ng/ml 14.8 ± 6.4 Responders Non-responders Paris-I criteria 19.7 ± 7.7 P =.28 P =.2 41% 25(OH)D<2 ng/ml Paris-I criteria 24% 25(OH)D 2 ng/ml 14.9 ± 6.3 Responders Non-responders Barcelona criteria Barcelona criteria 48% 25(OH)D<2 ng/ml transport and repress the cholestasis-induced inflammatory response. 13 Aberrations of the vitamin D VDR axis restrict the liver adaptive response and increase bile duct rupture during cholestasis. 14 Moreover, vitamin D was found to regulate the inappropriate immune responses by inducing tolerogenic dendritic cells, suppressing the development of proinflammatory effector T cells, promoting the induction and function of regulatory T cells, and modulating the functions of B cells. 16, 41 Therefore, it is conceivable that a reduced 25(OH)D level in PBC patients may impair the protective functions of vitamin D system, and thus limit the effects of anti-cholestasis treatment. Both vitamin D and UDCA can activate VDR to induce the expression of cathelicidin, which is an antimicrobial peptide with strong immunoreactivity for innate immune defences, in biliary epithelial cells. The combination of UDCA and vitamin D could increases cathelicidin expression to a greater extent than vitamin D or UDCA alone. 42 This may be another possible mechanism for the association of vitamin D deficiency and poor UDCA response. Moreover, our findings are in agreement with data from patients with hepatitis C 28 and liver cancer 29 that show associations between vitamin D and treatment response or prognosis. And as concerned in the previous reports, 43, 44 the relationship between vitamin D and liver disease is a chicken-and-egg kind of problem. The vitamin D level can be affected by abnormal bile formation or hepatic synthesis function in liver diseases, and in turn, the deficit in vitamin D may have negative effects on disease progression and therapy outcome. For this reason, our data may not be adequate to determine the cause of the association or indicate a definite benefit of supplementation, as only few patients underwent vitamin D supplementation in the present study. While it should be noted that vitamin D supplementation is currently recommended by AASLD and EASL to prevent bone complications in PBC 2, 3. But no trial data definitely support this preventive approach, 3 and long-term vitamin D supplementation may result in high costs, especially for many unwealthy Chinese patients. It would be interesting to investigate the relationship between vitamin D and UDCA responsiveness in another independent vitamin D- supplemented cohort. We should note that this study has several limitations. The primary limitation is the retrospective and monocentric nature of this study. Further validation in a different cohort is needed. The second limitation is the limited number of patients who underwent bone density Aliment Pharmacol Ther 215; 42:

8 G.-Y. Guo et al. 25(OH)D (ng/ml) 5 P = Alive Died or with transplant Figure 3 Serum 25(OH)D levels and long-term mortality. Patients who suffered death or transplant during the follow-up had lower baseline 25(OH)D value ( ) compared to those who did not develop such outcomes ( ng/ml, P =.23). detection which might explain the failure of identifying an association between vitamin D and bone disorders. Another limitation is that we did not examine the bioactive 1,25(OH)D concentration, although serum 25(OH)D level is the most reliable indicator for assessing vitamin D status. In conclusion, this study showed a correlation of vitamin D status with the severity of disease and response to UDCA in the patients with PBC. Vitamin D deficiency at baseline may increase the risk of a subsequent incomplete response to UDCA. Our results suggest a potential relationship between vitamin D and the therapeutic effects of UDCA in PBC. Further studies are needed to confirm our findings. AUTHORSHIP Guarantor of the article: Ying Han. Author contributions: Ying Han & Xinmin Zhou: Conception, design, analysis, interpretation of results and manuscript edit. Guanya Guo & Yongquan Shi: Design, data collection, analysis, interpretation of results and manuscript writing. Zheyi Han, Lu Wang & Changcun Guo: analysis, interpretation of results and manuscript edit. Jiangyi Zhu: bone density scan records and longterm data collection. Lina Cui &Jingbo Wang: Conception and interpretation of data. XiaoliRen, Jingwen Zhang, Yun Cai& Na Wang: Acquisition of data and analysis. All authors approved the final version of the manuscript. ACKNOWLEDGEMENTS Declaration of personal interests: None. Declaration of funding interests: The study was funded by National Natural Science Foundation of China (No , and ), National Science and Technology Major Projects for New Drugs Innovation and Development (No. 214ZX and 214ZX9142-3) and Foundation of Shaanxi Province (No. 213HM-1 and 212KTCL3-4). SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Figure S1. Transplant-free survival in patients with PBC according to vitamin D status. The survival of patients with vitamin D deficiency at baseline, as presented by green curve, tended to be lower than patients with 25(OH)D 2 ng/ml (P =.63 by log-rank test for Kaplan Meier curve). Table S1. Baseline characteristics of patients according to histological stages. Table S2. Correlation coefficients of Vitamin D and stage, risk score and serum parameters in patients with PBC. Table S3. Baseline parameters of responders and nonresponders. Table S4. Baseline risk factors associated with incomplete response to UDCA therapy, analysed by univariate and multivariate logistic regression. REFERENCES 1. Hirschfield GM, Gershwin ME. The immunobiology and pathophysiology of primary biliary cirrhosis. Ann Rev Pathol 213; 8: Lindor KD, Gershwin ME, Poupon R, et al. Primary biliary cirrhosis. Hepatology 29; 5: Boberg KM, Wisloff T, Kjollesdal KS, Stovring H, Kristiansen IS. Cost and health consequences of treatment of primary biliary cirrhosis with ursodeoxycholic acid. Aliment Pharmacol Ther 213; 38: Corpechot C, Abenavoli L, Rabahi N, et al. Biochemical response to ursodeoxycholic acid and long-term 228 Aliment Pharmacol Ther 215; 42:

9 Clinical relevance of vitamin D level in PBC prognosis in primary biliary cirrhosis. Hepatology 28; 48: Pares A, Caballeria L, Rodes J. Excellent long-term survival in patients with primary biliary cirrhosis and biochemical response to ursodeoxycholic Acid. Gastroenterology 26; 13: Kumagi T, Guindi M, Fischer SE, et al. Baseline ductopenia and treatment response predict long-term histological progression in primary biliary cirrhosis. Am J Gastroenterol 21; 15: Kuiper EM, Hansen BE, de Vries RA, et al. Improved prognosis of patients with primary biliary cirrhosis that have a biochemical response to ursodeoxycholic acid. Gastroenterology 29; 136: Wagonfeld JB, Nemchausky BA, Bolt M, Horst JV, Boyer JL, Rosenberg IH. Comparison of vitamin D and 25- hydroxy-vitamin-d in the therapy of primary biliary cirrhosis. Lancet 1976; 2: Herlong HF, Recker RR, Maddrey WC. Bone disease in primary biliary cirrhosis: histologic features and response to 25-hydroxyvitamin D. Gastroenterology 1982; 83(1 Pt 1): Hodgson SF, Dickson ER, Wahner HW, Johnson KA, Mann KG, Riggs BL. Bone loss and reduced osteoblast function in primary biliary cirrhosis. Ann Intern Med 1985; 13(6 Pt 1): Matloff DS, Kaplan MM, Neer RM, Goldberg MJ, Bitman W, Wolfe HJ. Osteoporosis in primary biliary cirrhosis: effects of 25-hydroxyvitamin D3 treatment. Gastroenterology 1982; 83(1 Pt 1): Eastell R, Dickson ER, Hodgson SF, et al. Rates of vertebral bone loss before and after liver transplantation in women with primary biliary cirrhosis. Hepatology 1991; 14: Ogura M, Nishida S, Ishizawa M, et al. Vitamin D3 modulates the expression of bile acid regulatory genes and represses inflammation in bile ductligated mice. J Pharmacol Exp Ther 29; 328: Firrincieli D, Zuniga S, Rey C, et al. Vitamin D nuclear receptor deficiency promotes cholestatic liver injury by disruption of biliary epithelial cell junctions in mice. Hepatology 213; 58: Ding N, Yu RT, Subramaniam N, et al. A vitamin D receptor/smad genomic circuit gates hepatic fibrotic response. Cell 213; 153: Peelen E, Knippenberg S, Muris AH, et al. Effects of vitamin D on the peripheral adaptive immune system: a review. Autoimmun Rev 211; 1: Ritterhouse LL, Crowe SR, Niewold TB, et al. Vitamin D deficiency is associated with an increased autoimmune response in healthy individuals and in patients with systemic lupus erythematosus. Ann Rheum Dis 211; 7: Weinstock-Guttman B, Zivadinov R, Qu J, et al. Vitamin D metabolites are associated with clinical and MRI outcomes in multiple sclerosis patients. J Neurol Neurosurg Psychiatry 211; 82: Runia TF, Hop WC, de Rijke YB, Buljevac D, Hintzen RQ. Lower serum vitamin D levels are associated with a higher relapse risk in multiple sclerosis. Neurology 212; 79: Garg M, Rosella O, Lubel JS, Gibson PR. Association of circulating vitamin D concentrations with intestinal but not systemic inflammation in inflammatory bowel disease. Inflamm Bowel Dis 213; 19: Mouli VP, Ananthakrishnan AN. Review article: vitamin D and inflammatory bowel diseases. Aliment Pharmacol Ther 214; 39: Petta S, Camma C, Scazzone C, et al. Low vitamin D serum level is related to severe fibrosis and low responsiveness to interferon-based therapy in genotype 1 chronic hepatitis C. Hepatology 21; 51: Farnik H, Bojunga J, Berger A, et al. Low vitamin D serum concentration is associated with high levels of hepatitis B virus replication in chronically infected patients. Hepatology 213; 58: Kwok RM, Torres DM, Harrison SA. Vitamin D and nonalcoholic fatty liver disease (NAFLD): is it more than just an association? Hepatology 213; 58: Trepo E, Ouziel R, Pradat P, et al. Marked 25-hydroxyvitamin D deficiency is associated with poor prognosis in patients with alcoholic liver disease. J Hepatol 213; 59: Ananthakrishnan AN, Cagan A, Gainer VS, et al. Normalization of plasma 25- hydroxy vitamin D is associated with reduced risk of surgery in Crohn s disease. Inflamm Bowel Dis 213; 19: Ananthakrishnan AN, Cheng SC, Cai T, et al. Association between reduced plasma 25-hydroxy vitamin D and increased risk of cancer in patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol 214; 12: Garcia-Alvarez M, Pineda-Tenor D, Jimenez-Sousa MA, Fernandez- Rodriguez A, Guzman-Fulgencio M, Resino S. Relationship of vitamin D status with advanced liver fibrosis and response to hepatitis C virus therapy: a meta-analysis. Hepatology 214; 6: Finkelmeier F, Kronenberger B, Koberle V, et al. Severe 25- hydroxyvitamin D deficiency identifies a poor prognosis in patients with hepatocellular carcinoma - a prospective cohort study. Aliment Pharmacol Ther 214; 39: European Association for the Study of the Liver. EASL Clinical Practice Guidelines: management of cholestatic liver diseases. J Hepatol 29; 51: Dietary Reference Intakes for Calcium and Vitamin D. Washington, DC: The National Academies Press, Ludwig J, Dickson ER, McDonald GS. Staging of chronic nonsuppurative destructive cholangitis (syndrome of primary biliary cirrhosis). Virchows Arch A Pathol Anat Histol 1978; 379: Webb AR, Kline L, Holick MF. Influence of season and latitude on the cutaneous synthesis of vitamin D3: exposure to winter sunlight in Boston and Edmonton will not promote vitamin D3 synthesis in human skin. J Clin Endocrinol Metab 1988; 67: Cholongitas E, Theocharidou E, Goulis J, Tsochatzis E, Akriviadis E, Burroughs K. Review article: the extra-skeletal effects of vitamin D in chronic hepatitis C infection. Aliment Pharmacol Ther 212; 35: Eliades M, Spyrou E, Agrawal N, et al. Meta-analysis: vitamin D and non-alcoholic fatty liver disease. Aliment Pharmacol Ther 213; 38: Kaplan MM, Elta GH, Furie B, Sadowski JA, Russell RM. Fat-soluble vitamin nutriture in primary biliary cirrhosis. Gastroenterology 1988; 95: Phillips JR, Angulo P, Petterson T, Lindor KD. Fat-soluble vitamin levels in patients with primary biliary cirrhosis. Am J Gastroenterol 21; 96: Agmon-Levin N, Kopilov R, Selmi C, et al. Vitamin D in primary biliary cirrhosis, a plausible marker of Aliment Pharmacol Ther 215; 42:

10 G.-Y. Guo et al. advanced disease. Immunol Res 215; 61: Lammers WJ, van Buuren HR, Hirschfield GM, et al. Levels of alkaline phosphatase and bilirubin are surrogate end points of outcomes of patients with primary biliary cirrhosis: an international follow-up study. Gastroenterology 214; 147: e5; quiz e Chan CW, Tsochatzis EA, Carpenter JR, Rigamonti C, Gunsar F, Burroughs AK. Predicting the advent of ascites and other complications in primary biliary cirrhosis: a staged model approach. Aliment Pharmacol Ther 21; 31: Mora JR, Iwata M, von Andrian UH. Vitamin effects on the immune system: vitamins A and D take centre stage. Nat Rev Immunol 28; 8: D Aldebert E, Biyeyeme Bi Mve MJ, Mergey M, et al. Bile salts control the antimicrobial peptide cathelicidin through nuclear receptors in the human biliary epithelium. Gastroenterology 29; 136: Kitson MT, Roberts SK. D-livering the message: the importance of vitamin D status in chronic liver disease. J Hepatol 212; 57: Dan YY, Lim SG. Commentary: vitamin D deficiency and liver cancer cause, effect or myth? Aliment Pharmacol Ther 214; 39: Aliment Pharmacol Ther 215; 42: