Liver transplant (LTX) recipients should avoid any liver injury and therefore be immunized against hepatitis A even if the risk of exposure is low. Other nonimmune transplant recipients should also receive the vaccine if they are highly exposed to hepatitis A (e.g., travelers to endemic areas) or if they have chronic hepatitis B or C associated with an increased risk of fulminant hepatitis A (1). In LTX and renal transplant (RTX) recipients a satisfactory anti-hepatitis A virus (HAV) immune response after two vaccinations was observed, although geometric mean titres (GMT) were lower than in healthy controls (2). Follow-up data and mathematical models suggest that in healthy individuals protective HAV antibodies after complete vaccination may persist for more than 20 years (3, 4). However, no data are available on the antibody decline in vaccinated individuals on immunosuppressive therapy. Such data are needed when planning vaccination schemes (vaccine dose, timing of booster injections) in organ transplant recipients.
Thirty-eight LTX patients, 39 RTX patients, and 27 controls who had no detectable HAV antibodies received two doses of hepatitis A vaccine (Havrix 1440; SmithKline Beecham, Rixensart, Belgium) 6 months apart. Anti-HAV titres were assessed 4 weeks after the first and second vaccine dose, respectively (2). Two years after the second vaccine dose participants were again tested for anti-HAV antibodies by the same commercial ELISA (Boehringer Enzymun kit; Boehringer Mannheim, Mannheim, Germany), which was calibrated by use of World Health Organization international standard reference serum. Anti-HAV titers ≥33 mIU/ml were considered positive according to the manufacturer and previous studies (2, 5). Thus, patients who showed postimmunisation titres of ≥33 mIU/ml were defined as seroconverters. Only patients who had seroconverted after the second vaccine dose were included in the 2-yr follow-up analysis. Chi-square test and Fisher’s exact test were used to compare proportions between groups. Geometric mean anti-HAV titers were calculated and comparisons of the titres between groups were done by nonparametric tests (Mann-Whitney U test, Kruskal-Wallis test). The seroconverters followed-up for 2 years did not differ significantly from those who were lost to follow-up with respect to age, gender, immunosuppressive therapy, duration of posttransplantation period, and anti-HAV titers after the first and second vaccine dose (Table 1).
The seroconversion rates after complete immunization were 97% (37/38) in LTX patients, 72% (28/39) in RTX patients, and 100% (27/27) in controls. RTX had significantly lower GMTs than LTX and controls. Follow-up sera were available from 70 individuals (27 LTX, 23 RTX, 20 controls) who had seroconverted. The mean age was 47.6 years (SD 10.4) in LTX, 42.7 (SD 10.7) years in RTX, and 39.4 years (SD 7.6) in controls (LTX versus controls P <0.05, t test).
Two years after vaccination, protective antibody titres were detectable in 59% of LTX, 26% of RTX, and 100% of the controls (Table 1). Both LTX and RTX patients showed a significantly larger decrease in GMTs than controls. It is noteworthy that LTX had not differed significantly from controls in their GMTs after the second vaccine dose but had much lower GMTs 2 years later. All these results did not change notably if a lower anti-HAV cut-off level of 20 mIU/ml was chosen. Neither the seroconversion rates nor the GMTs 4 weeks after complete immunization and 2 years later were associated with gender, age, time interval since transplantation, or HBV or HCV serostatus. In three LTX patients, an untypical course of antibody titers was observed. They developed high titers, exceeding even the titers of most of the controls, and lost their antibodies at a much lower rate than the other transplant recipients.
The majority of LTX patients received either tacrolimus (44%) or cyclosporin A (44%) for immunosuppressive treatment, whereas all RTX patients were on combinations of at least two drugs including cyclosporin A, azathioprine, and prednisolon. None of the patients received mycophenolate mofetil. No difference in the seroconversion rate and GMTs 4 weeks after vaccination was found between LTX patients on cyclosporin A and those on tacrolimus. However, after 2 years protective anti-HAV titers were detected in 79% of LTX on tacrolimus, but only in 39% of LTX on cyclosporin A (P <0.05, χ2 test), and GMTs were 115 and 35 mIU/ml, respectively (P =0.05, Mann-Whitney test). These differences remained after adjusting for posttransplantation time interval which was somewhat but not significantly shorter in the tacrolimus group. In RTX, the antibody decline was not associated with the type of immunosuppressive treatment.
Hepatitis A vaccination should be recommended to all anti-HAV negative organ transplant recipients if their risk of HAV exposure is not negligible. In contrast to healthy controls, however, many of these patients experience a rapid decline of antibody titers after vaccination. At present the protective anti-HAV antibody level after vaccination is not exactly known and the role of cellular immunity in protection against hepatitis A in the case of low anti-HAV antibody titers in immunosuppressed patients is unclear and needs further investigation (6). However, it is possible that in some immunosuppressed vaccines protection against hepatitis A is seriously impaired or lost already after 2 years. The type of immunosuppressive treatment may have an effect on the antibody decline. In LTX patients, the larger antibody decrease in the cyclosporin A group compared with the tacrolimus group may result from differences in the immunosuppressive processes induced by the two drugs (7). In RTX patients, the different drug combinations had no impact on the anti-HAV status 2 years after immunization, although Huzly et al. (8) reported that in tetanus, diphtheria, and polio vaccination the antibody response was better in RTX on double than on triple therapy.
The relatively high antibody titers in a few LTX patients may be due to a normal immune response despite immunosuppressive treatment. However, we cannot rule out the possibility that these patients had acquired HAV infection before transplantation but lost their antibodies before vaccination. The booster effect by the vaccine would indicate that the immunological memory resulting from previous natural infection may work even if the antibody titer is well below the cut-off level.
Further studies are needed including different vaccination schemes and long-term follow-up to identify the optimum vaccination strategy for transplant recipients.
1. Keeffe EB. Is hepatitis A more severe in patients with chronic hepatitis B and other chronic liver disease? Am J Gastroenterol 1995; 90: 201.
2. Stark K, Günther M, Neuhaus R, et al. Immunogenicity and safety of hepatitis A vaccine in liver and renal transplant recipients. J Infect Dis 1999; 180: 2014.
3. Clemens R, Safary A, Hepburn A, Roche C, Stanbury WJ, André FE. Clinical experience with an inactivated hepatitis A vaccine. J Infect Dis 1995; 171 (Suppl. 1): S44.
4. Van Damme P, Thoelen S, Cramm M, De Groote K, Safary A, Meheus A. Inactivated hepatitis A vaccine: reactogenicity, immunogenicity, and long-term antibody persistence. J Med Virol 1994; 44: 446.
5. Keeffe EB, Iwarson S, McMahon BJ, et al. Safety and immunogenicity of hepatitis A vaccine in patients with chronic liver disease. Hepatology 1998; 27: 881.
6. Fujiyama S, Odoh K, Tanaka M, Kuramoto I, Tomita K. Evaluation of the timing of the booster injection after a primary vaccination against hepatitis A. J Gastroenterol Hepatol 1997; 12: 172.
7. Jiang H, Kobayashi M. Differences between cyclosporin A and tacrolimus in organ transplantation. Transplant Proc 1999; 31: 1978.
8. Huzly D, Neifer S, Reinke P, Schröder K, Schönfeld C, Bienzle U. Routine immunisations in adult renal transplant recipients. Transplantation 1997; 63: 839.