Clinical, Basic Research And The Effects Of Drug Therapy For Primary Biliary Cirrhosis

Part one Prospective, randomized, controlled study ofursodeoxycholic acid, ursodeoxycholic acid combined withprednisonlone or azatharaprine therapy for primary biliary cirrhosisObject The aim of this study was to evaluate the effects of thecombination therapy with prednisonlone/azatharaprine and ursodeoxycholic acid(UDCA) for primary biliary cirrhosis (PBC), compared to UDCA monotherapy.Materials and methods Eighty two patients with untreated PBC were dividedrandomly into three groups. Group U (28 patients) received UDCA alone for 12months, group UP (27 patients) received UDCA and prednisonlone, while group UA(27 patients) received UDCA and Azp. The clinical and laboratory data was followedafter 3, 6 and 12 months. Inclusion criteria was the guideline for management of PBCwhich AASLD presented in 2000. The dose of these dugs were defined as UDCA13-15mg·d-1·kg-1, prednisonlone started with 0.5mg·d-1·kg-1, and tapered to7.5mg·d-1 from the fifth week, Azp 1mgod-1·kg-1. Disease progression wasdescriptived as liver decompensation, liver transplantation, two times elevation ofTBIL. Biochemical remission was defined as the normalization of serum bilirubinwith gamma-glutamyl transferase (GGT) with alkaline phosphatase (ALP)＜1.5×ULN (upper limit of normalization). Result Fatigue and pruritus were improvedonly in group UP. ALT, aspartate aminotransferase (AST), ALP, GGT, total bilirubin(TBIL), direct bilirubin (DBIL), IgM all decreased below or reached the ULN values(within-subject effects, P=0.000,0.000,0.000,0.000,0.009,0.001,0.000,respectively), and there had no ststistic significance between three groups(between-subject effects, P＞0.05 for all). The number of patients with diseaseprogression were two in group U, three in group UP, and one in group UA. Thepatients with disease progression had higher Mayo risk score(P=0.018) and prolongedProthrombin time (PT) (P=0.042). No side-effects occurred in group U. In group UP, side-effect related with Glucocorticosteroids occurred in eight patients. In group UA,two patients had side-effects related with Azp and one had side-effect related toUDCA. The level of ALP,GGT,CHO in baseline were risk factor for biochemicalremission (β=-2.095, P=0.015, Exp(β)=0.123, 95%CI 0.023-0.663), high level ofDBIL,TBIL,total bile acid (TBA),PT of baseline were not beneficial forbiochemical remission. (β=-1.162, P=0.075, Exp(B)=0.313, 95%CI0.083-1.124),wether combined with prednisonlone (β=-0.747, P=0.350, Exp(B)= 0.474, 95%CI0.099-2.266) orAzp (β=-1.479, P=0.223, Exp(B)=0.228, 95%CI0.021-2.453)with UDCA were not related with biochemical remission. CONCLUSIONS Thereare no difference between three groups for improve biochemical data of liver and IgM.High level of ALP, GGT, CHO in baseline are risk factor for biochemical remission.High level of TBIL,DBIL,TBA,PT are also not beneficial for biochemical remission.The combination with prednisonlone or Azp are not related with the ratio ofbiochemical remission. The disease of patients with higher Mayo risk score andprolonged PT tend to progress.Part two Character of peripheral lymphocytic subsets and cytokinesin patients with primary biliary cirrhosis and their change aftertherapyObject The aim of this study was to describe the Character ofperipheral lymphocytic subsets and cytokines in patients with primary biliary cirrhosisand their change after therapy. Materials and methods Eighty two patients withuntreated PBC were divided randomly into three groups. Group U (28 patients)received UDCA for 12 months, group UP (27 patients) received UDCA andprednisonlone, while group UA (27 patients) received UDCA and azatharaprine (Azp).Peripheral lymphocytic subsets and cytokines were detected at 0, 3, 6 months aftertherapy using tricolour flowcytometry analysis and ELISA respectively. Twenty ninepatients with chronic hepatitis B (CHB), twenty one patients with pimary Sj(?)gren'ssyndrome (PSS) and twenty healthy people were also included as controls. ResultCompared with PSS, the percent of B cells were decreased (P=0.010), the percent of CD4+T cells (P=0.003) and memory cells (CD4+CD45RO+) (P=0.051) wereincreased, and the level of IFN-γ, TNF-α, IL-2 were all increased (P=0.004, P=0.003P=0.000 respectively) in PBC. Compared with healthy control (HC), the percent ofCD8+T cells (P=0.277) decreased companied with the increase of percent of CD4+Tcell (P=0.044) and the ratio of CD4/CD8 (P=0.042) in PBC. And the patients withPBC also had higher percent of memory cells (P=0.049) than HC. The level of IFN-γ(P=0.001), TNF-α(P=0.000), IL-2 (P=0.000), IL-4 (P=0.010), IL-6 (P=0.029)were increased in PBC comparing with HC. The lower level of IL-10 was observed inPBC than in HC but didn't reach the statistic significance. The number of CD8+Tcells in peripheral were correlated with lower Mayi risk score (MRS) (P=0.001), TBIL(P=0.010), DBIL (P=0.013), and higher albumin (P=0.008). The level of TNF-αinperipheral were correlated with higher ALT (p=0.005) , AST (P=0.002) , TBIL(P=0.001) , DBIL (P=0.002) , MRS (P=0.020) . The percent of B cells decreasedin patients with PBC after therapy and reached the statistic significance in group U atsixth month. The percent of CD8+T cells increased in the three groups(P=0.002 ingroup UP at the sixth month). The percent of CD4+T cell (P=0.047, P=0.025 in groupU at the third and sixth months) and the ratio of CD4/CD8 (P=0.020 in group UP atthe sixth month) were decreased in group U and UP. Though didn't reach the statisticsignificance, the percent of memory cells decreased in the three groups. The percentof CD8+CD28+T cells in CD8+T cells was dereased in group U after UDCA therapy.The level of IFN-γ, IL-2, IL-4, IL-6 all decreased after therapy at the third month ingroup U, and rebounded thereafter. In group of UP and UA, all these cytokines weresurpressed below the level of baseline. TNF-αdecreased in all the three group aftertherapy (P=0.042 and P=0.043 in group UP and UA at the sixth month).CONCLUSIONS The reduction of CD8+T cells and the correlations with diseaseactivity revealve it's important role in the pathogenesis of PBC. PBC is a disease withpredominant cellular immunity as described with the pattern of peripheral cytokines.TNF-αis an important cytokine in the pathogenesis of patients with PBC, andcorrelated with the severity of the disease. The effects on the peripheral lymphocyticsubsets and cytokines of these three regime for the therapy of PBC are not identical.But the increase of CD8+T and the suppression of cytokines after therapy wereresponsive for the clinical amelioration of the disease. Part three Character of self antibody in patients with primary biliarycirrhosis and their change after therapyObject The aim of this study was to describe the character ofAnti-mitochondrial antibodies (AMA), Anti-mitochondrial M2 antibodies (AMA-M2),anti GP210 antibody (anti-GP210 Ab) and anti SP100 antibody (anti-SP100 Ab) inpatients with primary biliary cirrhosis (PBC) and their correlation with therapy.Materials and methods Eighty two patients with untreated PBC were dividedrandomly into three groups. Group U (28 patients) received UDCA, group UP (27patients) received UDCA and prednisonlone, while group UA (27 patients) receivedUDCA and azatharaprine (Azp). AMA (IIF), AMA-M2 (ELISA), anti-GP210antibody and anti-SP 100 antibody (IBT) were detected in these patients before therapy.AMA and AMA-M2 were repeated after therapy at the third and sixth months. ResultThere had no correlations between the titer of AMA, concentration of AMA-M2 andsymptoms, biochemical data of liver, histopathologie character in liver. AMA wascorrelated with peripheral IL-6 and IL-1β. The three protocol of therapy had noimpact on the level of AMA or AMA-M2 (within-subjects effects, P=0.270,between-subjects effects, P=0.648 for AMA; within-subjects effects, P=0.613,between-subjects effects P=0.764 for AMA-M2) . Whether the patient had diseaseprogression or biochemical remission, the difference of AMA or AMA-M2 didn'treach the statistic significance. The patients with anti-GP210 antibody had moresevere icterus(P=0.005), more frequent incident of esophageal varices (P=0.005),higher Mayo risk score (P=0.005). DBIL (P=0.035 ), TBA (P=0.002) in patients withthe antibody were higher and ALB was lower than those without. The patients withanti-GP210 antibody had later pathologic stage (P=0.013) and severe fibrosis(P=0.002). The biochemical remission was lower in patient with anti-GP210 antibodythan those without (16.7% vs 52.0%, respectively, P=0.012) . Being matched withBIL and MRS, the ratio of biochemical remission in two groups didn't reach thestatistic significance (16.7 vs 33.3%, P=0.266). There had no difference betweenpatients with anti-SP100 and those withought in symptoms, biochemical data of liver,histopathologic character in liver. CONCLUSIONS AMA, AMA-M2, anti-SP100 antibody were valuble in the diagnosis of PBC, but they are not related with theseverity of the disease. Patients with anti-GP210 antibody had more severe diseaseand this antibody correlate with lower ratio of biochemical remission. But it's notanti-GP210 antibody itself resist the role of drugs for PBC but because there are morerelatively severe cases in patients with positive anti-GP210 antibody.Part four Clinicopathological features of primary biliary cirrhosisObject The aim of this study was to describe the clinicopathologicalfeatures and character of lymphocytic subsets in portal tract of primary biliarycirrhosis (PBC). Materials and methods The liver biopsy speciments were obtainedfrom twenty four patients with PBC who participate in a prospective, randomized,controlled trial (Part one) through percutaneously needle puncture. These sampleswere fixed in formaldehyde and embedded in paraffin for routine histologicalexamination. Pathologic stage according to Ludwig (stage 1-4), fibsosis (grade 0-3),portal and periportal inflammation (grad 0-3), lymphocytic periportal piecemealnecrosis (grade 0-2), ductular proliferation (grade 0-2), intralobular hepatocytenecrosis (grade 0-2) and the degree of ductopenia (0-2) were record. Serial sections,4um-thick, were prepared for immunohistochemistry (CD20, CD45RO, CD8). Theaverage numbers of every lymphocytic subsets per portal area were computed. ResultThe pathologic stages, degree of fibrosis were positively correlated with ictetus, totalbilirubin (TBIL), total bile acid (TBA),Cholesterol (CHO),IgG, and negtivelycorrelated with albumin (ALB),percent of eosinophilic cell ( EOS ). Fibrosis was alsopositively correlated with Mayo risk score (MRS). Lymphocytic periportal piecemealnecrosis was positively correlated with the degree of icterus and pruritus, MRS,alkaline phosphatase (ALP), total bilirubin (TBIL), direct bilirunbin (DBIL), total bileacid (TBA), and also TNF-α(correlations 0.617, P=0.006). the degree ofductopeniawas positively correlated with the extent of icterus, MRS, aspartate aminotransferase(AST), TBIL, DBIL, TBA, IgG, and negtively correlated with ALB. No correlationswere observed between the numbers of CD20+T cells in portal area and biochemicaldata of liver. But the numbers of CD45RO+, CD8+ T cell were positively with ALT, AST,gamma-glutamyl transferase (GGT),TBA, and CD8+T cells also positivelycorrelated with ALP (correlations 0.775, P=0.041 ) and IFN-Γ(correlations 0.901,P=0.006). CONCLUSIONS The pathologic stage and the degree of fibsosis,lymphocytic periportal piecemeal necrosis, ductopenia are correlated with the diseaseseverity and reflect the process of the disease. TNF-αis a usful cytokine which canreflect the degree of lymphocytic periportal piecemeal necrosis, and they all reflectthe progression of the disease. Cellular immunity is important in the process of PBC.Acompanied with elevated IFN-γ, CD8+T cells destroy the interlobular bile duct.