Telephone: 49-511-532-3305; FAX: 49-511-532-4896
Failure of hepatitis B vaccination with conventional HBsAg vaccine in patients with continuous HBIG prophylaxis after liver transplantation
Article first published online: 22 FEB 2007
Copyright © 2007 American Association for the Study of Liver Diseases
Volume 13, Issue 3, pages 367–373, March 2007
How to Cite
Rosenau, J., Hooman, N., Hadem, J., Rifai, K., Bahr, M. J., Philipp, G., Tillmann, H. L., Klempnauer, J., Strassburg, C. P. and Manns, M. P. (2007), Failure of hepatitis B vaccination with conventional HBsAg vaccine in patients with continuous HBIG prophylaxis after liver transplantation. Liver Transpl, 13: 367–373. doi: 10.1002/lt.21003
- Issue published online: 22 FEB 2007
- Article first published online: 22 FEB 2007
- Manuscript Accepted: 19 AUG 2006
- Manuscript Received: 28 MAY 2006
- ZLB-Behring, Marburg, Germany
Hepatitis B vaccination after liver transplantation for hepatitis B–related liver disease has been investigated as an alternative strategy to reinfection prophylaxis with hepatitis B immunoglobulin (HBIG) with conflicting results. In most studies, HBIG treatment was discontinued before vaccination. An outstanding good response was achieved with vaccination under continuous HBIG administration using hepatitis B surface antigen (HBsAg)-based vaccine containing special adjuvants. Both, adjuvants and continuous HBIG administration have been discussed as crucial factors for good response. Twenty-four patients were vaccinated with conventional double dose recombinant vaccine containing 40 μg HBsAg up to 12 times at weeks 0, 2, 4 (cycle 1), 12, 14, 16 (cycle 2), 24, 26, 28 (cycle 3), and 36, 38, 40 (cycle 4). All patients received 2,000 IU HBIG every 6 weeks (4 times intravenously and 4 times intramuscularly). A significant response was defined as reconfirmed increase of anti-HBs-antigen (anti-HBs) unexplained by HBIG administration or lack of anti-HBs decrease below 100 IU/L after discontinuation of HBIG treatment after week 48. Only 2 of 24 patients (8.3%) responded significantly. Anti-HBs started to increase after the seventh vaccination (cycle 3, during intramuscular HBIG administration) in 1 patient and after 12th vaccination (cycle 4, during intravenous HBIG administration) in the other. Maximum anti-HBs levels were >1,000 IU/L in both patients and decreased significantly slower as compared to passive prophylaxis during follow-up. In conclusion, the conventional HBsAg vaccine failed to induce a significant humoral immune response in most patients despite continued HBIG treatment. Further studies should address the question, of whether the use of potent adjuvant systems results in higher response rates. Liver Transpl 13: 367–373, 2007. © 2007 AASLD.
Hepatitis B virus (HBV) reinfection of the liver graft after transplantation for HBV-related liver disease was associated with a considerably reduced graft and patient survival before the nucleoside analogue era.1, 2 Using reinfection prophylaxis with hepatitis B immunoglobulin (HBIG) reinfection rates decreased significantly.3, 4 Reinfections occurred less frequently under long-term immunoprophylaxis as compared to short-term prophylaxis.3, 4 Few reinfections were reported more than 12 months after orthotopic liver transplantation (OLT) despite ongoing HBIG treatment.4, 5 Thus, most transplant centers currently recommend continuing HBIG prophylaxis indefinitely.
After establishment of HBV therapy with the nucleoside analog lamivudine, the substitution of immunoglobulins by nucleotide analogs as reinfection prophylaxis was introduced. However, long-term results were disappointing, especially in patients with high viral replication before lamivudine initiation.6, 7 Better results were achieved by combination treatment with lamivudine (pre- and post-OLT) and HBIG (post-OLT). In several studies, reinfection rates of about 0 to 10% were shown even after longer follow-up periods.6–8
With an increasing number of patients receiving an infinite combination prophylaxis with immunoglobulins and nucleoside analog, costs are escalating. In addition, infinite treatment is uncomfortable.6, 9–11 Therefore, safe and cost-saving concepts for termination of HBIG prophylaxis are needed.
A currently promising strategy is the active immunization with vaccines containing hepatitis B surface antigen (HBsAg). However, study results are conflicting (Table 1). A publication by Sanchez-Fueyo et al.12 presented the new strategy with stimulating results using a standard HBsAg vaccine after discontinuation of HBIG. In contrast, a report published by Angelico et al.13 showed a very low response rate to vaccination despite frequent intramuscular and intradermal vaccine application. Next, the controversial discussion weighing risks and benefits of vaccination protocols was strikingly revitalized by the study of Bienzle et al.11 In that study, the highest response rate to date was reported, and notably this response rate was achieved without stopping the protective HBIG treatment. The continuation of HBIG treatment and the special adjuvant system in the vaccine were discussed as the most probable constitutive variables for the good response. In studies by Albeniz Arbizu et al.14 and Lo et al.,15 neither concomitant HBIG treatment nor special adjuvants were used, and response rates or response magnitude were significantly lower again.
|Sanchez-Fueyo et al.12, 16||Angelico et al.13||Bienzle et al.11||Albeniz Arbizu et al.14||Lo et al.15||Rosenau et al.17||Stärkel et al.18||Rosenau et al. (present study)|
|Number of patients||22||17||20||12||52||8||10||24|
|Age*||39 ± 8†||53 (36–63)||54 (35–69)||NA||47 (17–61)||50 (36–68)||49 (23–62)||47 (21–68)|
|HBV DNA negative before OLT‡||100%||100%||100%||NA||81%||100%||90%||79%|
|Time of vaccination (months after OLT)*||31 (18–76)||48 (25–85)||78 (24–156)||>24||14 (12–68)||60 (26–90)||55 (36–120)||40 (13–124)|
|Maximum number of vaccinations (per cycle)||3 + 3||3 + 6 + 3||5 + 3||3 + 3 + 3||3 + 3||6||5||3 + 3 + 3 + 3|
|HBsAg dose (μg), route of vaccination||40 IM/40 IM||40 IM/10 ID/40 IM||20 IM§/ 100 IM§||40 IM||40 IM||20 IM||40 IM||40 IM|
|Adjuvants to vaccine||Alum hydroxide||Alum hydroxide||Oil/water emulsion MPL + QS21||Alum hydroxide||Alum hydroxide||Alum salt MPL||Alum salt MPL||Alum hydroxide|
|HBIG prophylaxis during vaccination||No||No||Yes||No||No||No||Yes||Yes|
|Anti-HBs-increase to ≥10 IU/L % (absolute)||63.6 (14)||17.6 (3)||NA||75 (9)||7.7 (4)||12.5 (1)||50 (5)||NA|
|Anti-HBs-increase to ≥100 IU/L % (absolute)||22.7 (5)||11.8 (2)||NA||23 (3)||1.9 (1)||12.5 (1)||NA||8.3 (2)|
|Anti-HBs-increase to ≥500 IU/L % (absolute)||9.1 (2)||5.9 (1)||80 (16)||NA||0.0 (0)||12.5 (1)||40 (4)||8.3 (2)|
|Maximum anti-HBs-titers of responders*||47 (10–1,000)||253 (20–678)||25,344 (1,255–83,121)||NA||22 (12–103)||561||>1,000 (4× >1,000)||2,502 and 1,302|
PATIENTS AND METHODS
Between December 2003 and June 2004, 24 patients with normal liver function, negative HBsAg, and negative HBV after liver transplantation were enrolled in the study. The study was completed in June 2005. At the time of transplantation, all patients had been HBsAg positive. Of the 24 patients, 13 had been HBV DNA negative by polymerase chain reaction (PCR) without antiviral pretreatment. The other 11 patients had been HBV DNA positive, all with levels above 105 copies/mL. Lamivudine or lamivudine plus adefovir pretreatment reduced HBV DNA in 6 of these patients to levels below 105 copies/mL at time of transplantation. In 5 patients (2 of them with acute liver failure) antiviral treatment was initiated late; therefore, HBV DNA levels were above 105 copies/ml at the time of transplantation. The immunosuppressive regimen consisted of cyclosporine monotherapy in 13 patients, tacrolimus monotherapy in 4, mycophenolate mofetil monotherapy in 1, cyclosporine plus mycophenolate mofetil in 3, tacrolimus plus prednisolone in 2, and cyclosporine plus prednisolone plus rapamycin in 1. Immunosuppressive drugs were not changed in any of the patients throughout the study period. Six patients had hepatitis D virus coinfection. Other patient characteristics are shown in Table 1.
The vaccine consisted of recombinant purified HBsAg (Engerix B, GlaxoSmithKline, Munich, Germany) and was administered in double dose (40 μg) to the left deltoid muscle. Patients were vaccinated in 4 cycles at weeks 0, 2, 4 (cycle 1), 12, 14, 16 (cycle 2), 24, 26, 28 (cycle 3), and 36, 38, 40 (cycle 4). HBIG was administered every 6 weeks, beginning at week 0, in doses of 2,000 IU. HBIG was started intravenously (Hepatect CP, Biotest, Dreieich, Germany) in 12 patients and intramuscularly (Hepatitis B Immunoglobulin Behring, ZLB Behring, Marburg, Germany) in another 12 patients. At week 24, HBIG administration was switched from intramuscular to intravenous and vice-versa (cross-over). Intramuscular HBIG was administered in 2 portions of 1,000 IU (5 mL) into each gluteus muscle. Significant response to active vaccination was defined as a reconfirmed increase of anti-HBs-antibody (anti-HBs) unexplained by HBIG administration or lack of anti-HBs decrease below 100 IU/L after discontinuation of HBIG treatment after week 48. A total of 22 patients received combined reinfection prophylaxis with HBIG and either lamivudine (n = 19) or adefovir (n = 3). Two patients received HBIG monoprophylaxis.
The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the appropriate institutional review committee.
Anti-HBs, HBsAg and HBV DNA were performed as described previously.10
Response to Vaccination
A total of 22 patients received the maximum number of 12 double-dose vaccinations without significant response; in other words, a reconfirmed increase of anti-HBs unexplained by HBIG administration or lack of anti-HBs decrease below 100 IU/L after discontinuation of HBIG treatment after week 48 was not observed in any of these patients. Only 2 patients responded significantly (response rate, 8.3%). In responder 1, anti-HBs started to increase after the seventh vaccination (cycle 3, during intramuscular HBIG administration) and increased to 2,503 IU/L after the ninth vaccination (Table 2, Fig. 1). The fourth vaccination cycle was not administered, because the anti-HBs level was considered sufficient. In responder 2, anti-HBs started to increase to a level of 1,303 IU/L after the 12th vaccination (cycle 4, during intravenous HBIG administration) (Table 2, Fig. 1). HBIG-treatment was discontinued in both responders.
|Responder||Gender||Age||Indication for OLT||Immuno-suppression||HBV DNA before antiviral therapy/at OLT||Anti-HDV||Antiviral Therapy||Time of 1st Vaccination After OLT, (months)||Response after Vaccination Number (route of HBIG administration)||Anti-HBs Maximum|
|1||Male||21||Cirrhosis||CsA||Negative/negative||Positive||Lamivudine||32||7 (cycle 3, IV)||2,502|
|2||Male||31||Cirrhosis||CsA + MMF||Negative/negative||Positive||None||89||12 (cycle 4, IM)||1,302|
Anti-HBs levels decreased significantly slower in the 2 responders (Fig. 1) as compared to the nonresponders (anti-HBs half-life in nonresponders: median 25; range 16–36 days) during follow-up. During follow-up, anti-HBs levels dropped to 362 IU/L and 100 IU/L (responder 1 and 2 respectively) 42 weeks and 28 weeks after the maximum level of 2,503 IU/L and 1,303 IU/L. Both patients responded to booster immunizations (3 double-dose vaccinations at weeks 0, 2, and 4) with anti-HBs levels >1,000 IU/L. HBsAg remained negative in all patients during the study period and follow-up.
Active hepatitis B vaccination after liver transplantation for HBV-induced liver disease has evolved as an alternative strategy to replace long-term prophylaxis with HBIG and/or nucleoside analog. However, study results have a broad spectrum of conflicting response rates.11–18 In the studies by Angelico13 and Lo,15 response rates and anti-HBs levels of responders were disappointing. The data of Sanchez-Fueyo12 and Albeniz Arbizu14 showed a promising overall response rate, but anti-HBs levels of most responders were low. In contrast to these studies, Bienzle11 reported not only a good response rate but also impressive anti-HBs levels among most responders. The question of whether success or failure of vaccination depends on patient-based factors on one hand or vaccine or protocol related factors on the other hand brings up a discussion that appears to be essential for optimizing the vaccination strategy. In the report by Bienzle,11 2 factors were discussed as possible constitutive variables for the strong response: the continuation of HBIG treatment, and the adjuvant contained in the vaccine (Table 1).
We therefore conducted a preliminary trial17 using an adjuvant system similar to that in the Bienzle study,11 but we discontinued HBIG treatment in contrast to the Bienzle study. However, the response rate was significantly lower as compared to the Bienzle trial, even if the number of patients enrolled was small: Only 1 of 8 patients responded. Considering this low response rate despite using the adjuvant 3-, we speculated that the successful induction of a strong response in the Bienzle study may be attributed to the concomitant HBIG administration rather than to the adjuvant in the vaccine. Therefore, we carried out the present study to investigate the hypothesis that formation of antigen-antibody complexes (formed by HBsAg as the immunogenic component of the vaccine and anti-HBs as the protecting component of HBIG) more likely enhances immunogenicity than neutralizing and eliminating the vaccine, as suggested by the results of studies in mice.19, 20 But again, the response rate in our trial is significantly lower compared to that of the Bienzle trial. Although a weak response to vaccination in some patients, as seen in other studies,12–17 cannot be excluded in our trial because of the continued HBIG administration, a strong response with anti-HBs levels above 500 IU/L was seen in only 2 patients (8.3%, as compared to 80% in the Bienzle trial).11
Since the results of the Bienzle trial11 differ significantly from all other studies, it is worth taking a closer look at other possible predictive factors for response. Our present study, as well as the data of Sanchez-Fueyo12 and Bienzle,11 clearly shows the benefit of a high number of vaccinations—in other words, an accelerated course with repeated vaccination cycles. However, the number and sequence of vaccinations does not explain the different response rates among reported protocols. Immunosuppression is supposed to reduce response rates significantly. Reports on HBV vaccination in non-HBV adult liver transplant recipients have shown very low rates of seroconversion.21–23 A reduced humoral response to vaccinations was reported particularly for patients receiving mycophenolate mofetil.24, 25 However, the thorough analysis of Sanchez-Fueyo12 did not reveal a significant difference between responders and nonresponders in regard to type and dose of immunosuppression, though the time between steroid withdrawal and vaccine administration was significantly lower in nonresponders. Furthermore, in our previous study,17 the patient with the strongest immunosuppressive regimen containing mycophenolate mofetil was the only responder, and the overall response rates reported by Angelico13 and Lo15 and in our present study were very low, although most patients received steroid- and mycophenolate mofetil–free immunosuppressive regimens. In contrast, patients with double or triple immunosuppression did not show a lower seroconversion rate than patients with monotherapy in the Bienzle study. Regarding the pretransplant course of HBV infection, a better response to vaccination could be expected in patients with fulminant liver failure, since some of these patients show an HBsAg-specific immunologic activity at time of transplantation. However, the number of patients with acute liver failure was low in all studies except in the cohort of Sanchez-Fueyo,12 with 36% fulminant courses, and no significant difference between responders and nonresponders was found in that cohort. The interval between OLT and vaccination showed no correlation to response in any of the studies, although it tended to be longer in the Bienzle11 study. A concomitant nucleoside analog treatment had been shown to restore cytotoxic T-lymphocyte reactivity,26, 27 but none of the 5 patients receiving lamivudine responded to vaccination in the Sanchez-Fueyo trial, and in the studies with lowest response rates all patients received lamivudine, suggesting that lamivudine may rather deteriorate than improve the response to HBV vaccination. With respect to age and gender, no significant difference between responders and nonresponders was found. Due to the small number of patients with hepatitis C or D virus coinfection, no conclusions can be drawn regarding the effect of viral coinfections to vaccination responses in any of the studies. The benefit of a double vaccine dose, as suggested for patients with a week immune response,21–23 remains unclear. In the Bienzle study, the immune response tended to be better in the group with the lower HBsAg dose. A potential benefit of an intradermal administration of the vaccine28 was not seen in the study of Angelico.13 The number of patients with positive pretransplant HBV DNA in the present trial and the study of Lo15 was relatively high, and Lo speculated that response to vaccination may be worse in this subgroup. However, pretransplant HBV DNA was negative not only in the study with strong response11 but also in most studies with weak response.12–17 Therefore, it remains unclear whether pretransplant replication has an impact on response.
In conclusion, our study supports that only a very limited number of patients respond to active vaccination with standard HBsAg-containing vaccines. Neither immunocomplex formation under continued HBIG administration nor the factors discussed herein appear to be decisive for success or failure of vaccination. The Bienzle study11 has raised the hope that a substantial proportion of hepatitis B patients may profit from post-OLT active vaccination. It remains unclear whether the right adjuvant system to the vaccine, or the combination of potent adjuvants and continued HBIG treatment is the key to a high response rate. The most recent study published by Stärkel et al.18 does not answer this question, since vaccination under continued HBIG treatment with the same –containing vaccine as used in our previous study17 resulted in an intermediate response rate of 40–50% with high anti-HBs levels. Salmonella minnesota, a chemically-modified derivative of lipopolysaccharide of minnesota with greatly reduced toxicity, has been used extensively in clinical trials as a component in prophylactic and therapeutic vaccines targeting infectious diseases, cancer, and allergies. In the vaccine used by Bienzle, was formulated in an oil-in-water emulsion in combination with another adjuvant, Quillaja saponaria (QS21), a pure fraction of Quil A saponin derived from the plant saponaria, which is known to induce the production of interferon gamma, interleukin 2, and antibody formation of the immunoglobulin G2a isotype.29 The adjuvant composition in the Bienzle study may have been decisive for the superior response. Therefore, further studies should address the question of whether vaccination with potent adjuvant systems under ongoing protection of HBIG confirm the high response rates reported by Bienzle. As the adjuvant system in that study appears to be the most promising to date, patients—including nonresponders in previous studies—could be vaccinated in a multicenter trial.
- 10Lamivudine and low-dose hepatitis B immune globulin for prophylaxis of hepatitis B reinfection after liver transplantation possible role of mutations in the YMDD motif prior to transplantation as a risk factor for reinfection. J Hepatol 2001; 34: 895–902., , , , , , .