The association of HLA-DR13 with lower graft survival rates in hepatitis B and primary sclerosing cholangitis caucasian patients receiving a liver transplant

Authors


Abstract

We investigated an association of human leukocyte antigen (HLA)-DR13 to graft survival in liver transplantation among Caucasian recipients. 28,708 deceased liver transplants performed between January 1990 and December 2002 in the United States as reported to the United Network for Organ Sharing registry were utilized to compare survival rates. We utilized Caucasian adult patients (>20 years) by Kaplan-Meier curves, log-rank tests, and Cox proportional hazard analyses. HLA-DR13-negative hepatitis B virus (HBV) and primary sclerosing cholangitis (PSC) recipients yielded significantly lower graft survival rates than those of DR13-negative patients (P = 0.002, P = 0.015, respectively). This negative association was still significant after adjusting potential confounding factors. The Cox test demonstrated that HLA-DR13-positive groups have a significantly higher hazard ratio in PSC (1.40; P = 0.029; 95% confidence interval, 1.04-1.90) and HBV patients (1.78; P = 0.032; 95% confidence interval, 1.05-3.02). In conclusion, our data suggest that HLA-DR13 is a strong, positive predictor of increased risk for graft loss in HBV and PSC liver transplant recipients. Further study is needed to test the hypothesis that DR13-related immune responses may play a role in mediating graft loss in HBV and PSC liver transplantations. Liver Transpl 12:600–604, 2006. © 2006 AASLD.

Liver transplantation has become a widely accepted form of treatment for end-stage liver diseases. Interestingly, unlike kidney or heart transplantation, human leukocyte antigen (HLA) matching has not been used for the allocation of hepatic allografts, because the liver is considered to be less susceptible than the kidney or the heart to hyperacute or other humoral rejection. Nevertheless, the exact role of various HLA antigens in the long-term results of liver transplantation is not yet clearly defined and is still a subject of recalcitrant controversy.

Recently, HLA-DR13-positive hepatitis B virus (HBV) patients have been reported to be protected against persistent HBV infection, possibly through highly efficient CD4+ cell responses.1 HLA-DR13 haplotype has also been found to be associated with an increased risk of primary sclerosing cholangitis (PSC).2–4 These results led us to hypothesize that prevalence of DR-13 has the effects on long terms liver transplant outcome. The aim of this study is to evaluate the effect of HLA-DR13 on graft survival in different liver transplant recipients.

Abbreviations

HLA, human leukocyte antigen; HBV, hepatitis B virus; PSC, primary sclerosing cholangitis.

PATIENTS AND METHODS

Study Population

Liver transplants performed between January 1990 and December, 2002 in the United States as reported to the United Network for Organ Sharing Registry were utilized to examine the relationship between graft survival rates and HLA-DR by primary diseases. Multiorgan transplant grafts and regrafts were excluded. The HLA distribution differs by different races. In the DR-13-positive patient group, the African-American patients accounted for 12% and the Caucasian patients 68%, while in the DR-13-negative patient group, they accounted for 5% and 59%, respectively. To minimize the race effect, we focused only on Caucasian recipients, who were the major liver transplant recipients. We limited our study to Caucasian adult patients (>20 years) who received an organ from a deceased donor. Data included 28,708 liver deceased cases. In Table 1, we showed the overall transplantation-related characters by HBV, PSC, and other primary diseases.

Table 1. Characteristic by Recipients With HLA-DR13 or Those Without It at PSC, HBV, and All Other Diseases
Primary diseasesRecipient groupCase numberAntigen frequency5-year Graft survival (%)10-year Graft survival (%)Donor age (yr) ave. ± S.D.Recip age (yr) ave. ± S.D.ICU patient distribution (%)HCV antibody positive ratio (%)Distribution of male recipient (%)
PSCDR 13(−)177378.80%74.259.934.9 ± 16.747.9 ± 11.39.82.169.5
 DR 13(+)47621.20%70.253.934.9 ± 16.548.5 ± 10.37.13.269.1
HBVDR 13(−)85292.40%71.553.736.9 ± 16.250.2 ± 10.019.310.686.0
 DR 13(+)707.60%58.035.238.8 ± 16.049.8 ± 9.417.18.785.7
all other diseasesDR 13(−)2239886.07%67.651.036.7 ± 17.250.8 ± 9.817.044.461.7
 DR 13(+)313913.93%66.050.836.7 ± 16.850.9 ± 9.815.747.162.5
         p = 0.006 

Statistical Analysis

Kaplan-Meier survival curve estimates and log-rank tests were used to describe and compare survival rates. All P values were 2-sided, and P < 0.05 was considered significant. All analyses were performed with the Stata program (Ver.7.0, College Station, TX). The chi-square test was used to compare distribution differences. Student's t test was used to compare group mean differences. Patients' deaths were counted as graft loss.

Cox proportional hazard analyses were performed to estimate the total effect of HLA-DR13 on liver graft survival. According to a previous study,5 the adjusted potential risk factors were as follows: donor age, recipient age, patient status, anti–hepatitis C antibody, recipient sex, recipient serum creatinine, and transplantation year. In Cox proportional hazard analyses, grafts that survived less than 1 year were excluded.

RESULTS

Transplantation-Related Characteristics

Table 1 shows some transplant-related characteristics. Basically, there was no significant difference between HLA-DR13-positive patients and HLA-DR13-negative patients for those characters in each disease category listed. A higher ratio of hepatitis C virus antibody in HLA-DR13-positive patients at all other primary diseases was significant, compared to HLA-DR13-negative recipients. Frequency of HLA-DR13 antigen in PSC patients is 21.2%, while that of HBV and all other diseases are 7.6% and 13.9% respectively.

Graft Survival in HLA-DR13-Positive HBV and PSC Patients vs. DR13-Negative HBV and PSC Patients

HLA-DR13-positive HBV patients had significantly lower graft survival rates than those without DR13 (P = 0.002). The 10-year graft survival rates of HLA-DR13-positive recipients and HLA-DR13-negative patients were 35.2% and 53.7%, respectively. The 5-year graft survival rates were 58.0% and 71.5%, respectively (Fig. 1A, Table 1).

Figure 1.

Graft survival rates by HLA-DR13-positive and negative recipients. Analyses were limited to adult Caucasian population who primarily received an organ from a deceased donor. (A) In HBV patients, HLA-DR13-positive patients (solid line; n = 70) had a significantly lower graft survival rate (P = 0.002) than DR13-negative patients (dotted line; n = 852). (B) In PSC patients, HLA-DR13-positive patients (solid line; n = 476) had a significantly lower graft survival rate (P = 0.015) than DR13-negative patients (dotted line; n = 1,773).(C) In other diseases, there was no significant difference (P = 0.212) between HLA-DR13-positive patients (solid line; n = 3,139) and HLA-DR13-negative patients (dotted line; n = 22,398).

Similarly, PSC patients with HLA-DR13 had significantly lower graft survival rates than those without it (P = 0.015). The 10-year graft survival rates of HLA-DR13-positive recipients and negative patients were 53.9% and 59.9%, respectively. The 5-year corresponding rates were 70.2% and 74.2% (Fig. 1B, Table 1).

However, the HLA-DR13 effect was not observed in other primary diseases' graft survival rates. The 10-year graft survival rates of HLA-DR13-positive recipients and HLA-DR13-negative patients were 50.8% and 51.0%, respectively (not significant) (Fig. 1C, Table 1).

All DR13-positive HBV patients lost their grafts within 6 years when they received a DR13-positive liver transplant. The 6-year graft survival for DR13-positive patients who received DR13-negative grafts was 62.7%. The difference in graft survival rates between those 2 groups was statistically significant (Fig. 2A, P < 0.05). There was no difference between DR13-positive and DR13-negative donor groups when recipients were DR13-negative (Fig. 2A). In PSC patients, a similar phenomenon was observed in DR13-positive patients. However, statistically, there was no difference between DR13-positive and DR13-negative donors (Fig. 2A and 2B).

Figure 2.

Graft survival rates by the combinations of HLA-DR13 in donors and recipients. (A) In HBV, DR13-positive recipients who received DR13-positive transplants (thick solid line; n = 13) yielded the lowest graft survival rate. In contrast, DR13-negative recipients who received DR13-negative transplants yielded the highest graft survival rate (dotted line; n = 697). Donors' DR13 was a determinant when recipients were DR13-positive. Dashed line represents DR13-negative recipients and DR13-positive donors (n = 155). Thin solid line represents DR13-positive recipients and DR13-negative donors (n = 57). P values are as follows: DR13-negative recipient and DR13-negative donor vs. DR13-positive recipient and DR13-positive donor, P = 0.001; DR13-negative recipient and DR13-negative donor vs. DR13-positive recipient and DR13-negative donor, P = 0.0510; DR13-positive recipient and DR13-negative donor vs. DR13-positive recipient and DR13-positive donor, P = 0.0143; and DR13-negative recipient and DR13-positive donor vs. DR13-positive recipient and DR13-positive donor, P < 0.001.(B) In PSC, the trend was similar to HBV. The combination of DR13-positive donors and recipients (n = 384) yielded the lowest graft survival rate and DR13-negative donors and recipients (n = 1,480) yielded the highest graft survival rate. However, the difference was not significant. Dashed line represents DR13-negative recipients and DR13-positive donors (n = 293). Thin solid line represents DR13-positive recipients and DR13-negative donors (n = 392). P values are as follows: DR13-negative recipient and DR13-negative donor vs. DR13-positive recipient and DR13-positive donor, P = 0.135 (not significant); DR13-negative recipient and DR13-negative donor vs. DR13-positive recipient and DR13-negative donor, P = 0.035; and DR13-positive recipient and DR13-negative donor vs. DR13-positive recipient and DR13-negative donor, P = 0.732 (not significant).

Obvious DR13 Effect

Transplantation-related factors that may have an independent influence on liver graft survival, such as donor age, recipient age, recipient sex, medical condition, hepatitis C virus infection status, recipient serum creatinine, and transplantation year period were considered by the Cox hazard test (Table 2). Even after adjusting these factors, the HLA-DR13-positive group had a significantly higher hazard ratio in PSC (hazard ratio = 1.40, P < 0.05) and HBV patients (hazard ratio = 1.78, P < 0.05). Conversely, other primary disease groups showed no significant differences (hazard ratio = 1.04, P = 0.37).

Table 2. Results of Cox Hazard Tests by Recipients With HLA-DR13 or Those Without It at PSC, HBV, and All Other Diseases
Primary diseasesRecipient groupCase numberHazard ratio95% confidence intervalP value
  1. Cox proportional hazard analyses were performed to adjust for confounding factors which were classified as follows: 1) donor age variable:0-20, 21-40, 41-60, 61-2) recipient age: 21-40, 41-60, 61-. 3) Medical condition: not hospitalized, hospitalized, in ICU. 4) HCV antibody: positive, negative. 5) recipient sex. 6) recipient pretransplant serum creatinine: ≤1.0, 1.1-1.5, 1.6<. 7)transplantation year: 1990-1992, 1993-1996, 1997-2002. Hazard ratio was computed with HLA-DR13 negative recipients as the reference.

PSCDR 13(−)17731.00  
 DR 13(+)4761.401.04–1.90P= 0.029
HBVDR 13(−)8521.00  
 DR 13(+)701.781.05–3.02P= 0.032
all other diseasesDR 13(−)223981.00  
 DR 13(+)31391.040.95–1.13P = 0.373

DISCUSSION

In this study, we showed that recipients' HLA-DR13 may be one of the strong, independent predictors of lower graft survival rates in HBV and PSC liver transplant Caucasian patients. DR13-positive patients may be forced to avoid liver transplants with DR13.

HLA-DR13 has been reported to be associated with resistance to several pathogenic virus infections, such as hepatitis C, human immunodeficiency virus, malaria, and papilloma virus.6–8 In fact, HLA-DR13-positive patients in all other diseases have a significantly higher ratio of having hepatitis C virus antibody in this analysis (Table 1). A strong association between the HLA-DR13 and a self-limited course of HBV infection was first reported in the African population and later confirmed in Caucasian and Asian populations.9–11 Our study showed that the frequencies of HLA-DR13 antigen in HBV (7.6%) and all other original liver diseases (13.93%), except PSC (21.2%), were significantly lower than that of general Northern American Caucasian adults (21.7%). These data confirmed that DR13-positive individuals less likely develop severe liver diseases that may finally need to be treated by liver transplantation.

It has been indicated that HLA-DR13 may have a beneficial effect, protecting against chronic HBV, possibly through the induction of more vigorous hepatitis B core antigen specific CD4+ T-cell response.1 HLA-DR13-restricted CD4+ T-cell epitopes of hepatitis B core antigen were later discovered.12 These data provided strong evidence to prove how DR13-positive individuals overcome chronic HBV infection. Here, for the first time, we showed that once HBV reinfection occurs in a transplanted liver, the infected hepatocytes could still become targets of immunoresponse. That might be one of the explanations of lower graft survival rates of HLA-DR13-positive HBV patients, when they received a DR13-positive liver transplant compared with a DR13-negative donor (Fig. 2). A DR13-positive transplant may be infected by HBV, and virus peptides are processed and presented by DR13, therefore triggering a patient's immune response.

PSC is considered an immune-mediated liver disease of multifactorial and multigenetic etiology. Any single genetic allele is not detected as an attributable locus. Four key HLA haplotypes associated with PSC have developed.4 HLA-B8, DR3, and DQ2 have been well established. Another key haplotype is DRB3*0101 - DRB1*1301-DQA1*0103-DQB1*0603. The HLA-DR13 epitope is also associated with some diseases of probable autoimmune origin, such as multiple sclerosis.13

Even though the etiology of PSC is still obscure, an association of DR13 to low graft survival in PSC patients may also imply an immune-related cause of graft failure. The trigger factor for the initiation of the immune response might be the access of intestinal bacteria (possibly acting as molecular mimics for autoantigens) or other toxic metabolics into the portal circulation through a diseased and permeable bowel wall.4, 14, 15 An abnormal immune response to these antigens in an immunogenetically susceptible host might be sufficient to precipitate the cascade of immunoreaction. On the supposition that recipients with HLA-DR13 mount more effective CD4+ T-cell reaction to bacterial antigens similar to virus antigens (e.g., HBV), HLA-DR13 PSC transplant patients might cause stronger or more rapid reaction to transplanted organs, especially when they receive a DR13-match transplant. Another reason for lower graft survival rates from HLA-DR13-positive PSC patients may be the HLA-DR13-associated strong immunoresponse.

Oertrl et al. suggested the increased risk of acute rejection of hepatic allografts from HLA-DR13-positive donors from 35 liver transplants.16 Our data showed that when DR13-positive livers were transplanted into DR-13-positive patients, graft survivals were lower compared with DR13-negative donor transplants. This difference reached statistical significance in HBV patients. These data suggested immune response against DR13 may play an important role in mediating graft loss, through either direct or indirect recognition.

To examine the effect of DR13 considering the changes in management in patients over time, we conducted a stratified analyses by the years of transplantation among HBV and PSC patients and compared the survivals between DR-13-positive and DR-13-negative patients. The patients were classified into 2 groups based on the year they received transplantation: 1st period (1990-1995) and 2nd period (1996-2002). Although the difference in graft survivals between DR-13-positive recipients and DR-13-negative recipients was not statistically significant in the 1st period due to a small number of DR13-positive recipients by stratification, the effect of DR-13 was maintained in both periods (data not shown). Furthermore, our Cox hazard regression model included the year of transplantation as one of the confounding factorsthat take into account the improvement in transplantation and the management of patients. These results also confirmed the effect of DR-13 independent of the year of transplantation.

Since not all failed graft cases were reported to the United Network for Organ Sharing database, it is difficult to investigate the exact explanation for the cause of graft failure in this study. However, 280 cases out of 560 graft-failed transplants with DR13-negative recipients and 104 out of 173 cases with HLA-DR13-positive recipients were available in PSC. The distribution of rejection-related graft failure (acute and chronic rejection) in all reported graft failure cases was 21.2% in the HLA-DR13-positive recipients (30 cases), and 10.7% in the DR-13-negative recipients (22 cases). Higher incidences of acute and chronic rejection in DR13-positive recipients further supported the destructive role of DR13-related immunological factors in hepatic graft failure.

In conclusion, HLA-DR13 seems to be a strong, positive predictor of increased risk for graft loss in HBV and PSC liver transplantation. Patients with HLA-DR13 may need intensified follow-up and to avoid receiving a transplant from a DR13-positive donor.

Ancillary