Long-term administration of hepatitis B immunoglobulin (HBIG), currently in combination with nucleoside or nucleotide analogues, remains the gold standard for hepatitis B reinfection prophylaxis after liver transplantation.1–3 With the introduction of HBIG monoprophylaxis 2 decades ago, reinfection rates were reduced significantly from about 80% to about 30%–50%, and the beneficial effect of long-term immunoprophylaxis compared to short-term prophylaxis was established.4, 5 With combined treatment of HBIG and antiviral drugs, currently lamivudine and adefovir, the risk of hepatitis B virus (HBV) reinfection has been further reduced below 10% during the first 2 years following transplantation.1–3 Combined long-term treatment is considered the safest prophylactic strategy available, at least in nonresponders to active vaccination.1–3 However, costs for combined treatment, in particular for HBIG, are tremendous.
As one strategy to reduce costs, conversion from intravenous (IV) to intramuscular (IM) HBIG administration has been proposed and evaluated. The authors of those studies concluded that IM HBIG administration may be cost-effective and even dose-sparing.6–8 Many transplant centers use HBIG preparations for IM use as standard treatment with good success9–18 because of the unavailability of IV HBIG preparations or potential cost effectiveness. However, pharmacokinetic data on IM HBIG use are rare,19 especially comparative data with IV administration.6, 7
Our study provides comparative pharmacokinetic data for IV and IM HBIG use derived from a high number of measured antibody against hepatitis B surface antigen (anti-HBs) levels in a fairly large number of patients as a basis for rationally deciding whether or not to use IM HBIG.
anti-HBs, antibody against hepatitis B surface antigen; AUC, area under the curve; BMI, body mass index; HBIG, hepatitis B immunoglobulin; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HDV, hepatitis D virus; IM, intramuscular; IV, intravenous; OLT, orthotopic liver transplantation; SC, subcutaneous; SD, standard deviation.
PATIENTS AND METHODS
Twenty-four patients receiving long-term hepatitis B reinfection prophylaxis for at least 12 months after liver transplantation for HBV-related liver failure were enrolled between December 2003 and June 2004. All patients had received high-dose HBIG in the initial phase after orthotopic liver transplantation [OLT; 10,000 IU daily until hepatitis B surface antigen (HBsAg) became negative] followed by low-dose, long-term HBIG treatment (2000 IU in irregular intervals with an anti-HBs maintenance level intended to be kept above 100 IU/L).20 Prophylaxis was combined with lamivudine or adefovir in all except 2 patients. In the 3 patients receiving adefovir, antiviral therapy had been switched from lamivudine to adefovir because of viral resistance pre-OLT. At time of enrolment, all patients showed normal liver function, negative HBsAg, and negative HBV polymerase chain reaction. Further baseline characteristics for patients are shown in Table 1. Two responders to active vaccination (discussed later) were excluded from evaluation in this study. However, 22 patients (11 in each group) did not respond to active vaccination, and anti-HBs levels of these nonresponders were evaluated. In all, 18 anti-HBs values were missing, with a maximum of 3 values in a single patient.
Table 1. Demographic Data
Abbreviations: anti-HBs, antibody against hepatitis B surface antigen; BMI, body mass index; HBIG, hepatitis B immunoglobulin; HBV, hepatitis B virus; HDV, hepatitis D virus; OLT, orthotopic liver transplantation; SD, standard deviation.
Number of patients enrolled
Number of patients evaluated
Age (years): mean ± SD, median (range)
48 ± 14, 49 (26–68)
43 ± 12, 45 (22–57)
46 ± 13, 47 (22–68)
Weight (kg): mean ± SD, median (range)
83 ± 19, 82 (60–125)
74 ± 16, 75 (50–100)
79 ± 18, 80 (50–125)
Height (cm): mean ± SD, median (range)
177 ± 7, 175 (164–186)
174 ± 8, 174 (159–186)
175 ± 7, 175 (159–186)
BMI (kg/m2): mean ± SD, median (range)
26 ± 5, 26 (20–37)
25 ± 6, 22 (17–32)
26 ± 5, 26 (17–37)
Plasma volume (L): mean ± SD, median (range)
3.1 ± 0.4, 3.0 (2.4–3.8)
2.9 ± 0.4, 2.9 (2.2–3.4)
3.0 ± 0.4, 3.0 (2.2–3.8)
Liver failure: acute/chronic
HBV DNA before antiviral therapy (copies/mL)
< 105 N = 5; > 105 N = 6
< 105 N = 6; > 105 N = 5
< 105 N =11; > 105 N = 11
HBV DNA at time of OLT (copies/mL)<
105 N = 8; > 105 N = 3
< 105 N = 9; > 105 N = 2
< 105 N = 17; > 105 N = 5
Antiviral therapy during study period (lamivudine/adefovir/none)
Time of first study HBIG dose post-OLT (months): median (range)
Baseline anti-HBs levels at time of first study HBIG administration: mean ± SD, median (range)
214 ± 77, 227 (107–342)
190 ± 60, 182 (114–310)
202 ± 69, 194 (107–342)
The immunosuppressive regimen consisted of cyclosporine monotherapy in 12 patients, tacrolimus monotherapy in 3 patients, mycophenolate mofetil monotherapy in 1 patient, cyclosporine plus mycophenolate mofetil in 3 patients, tacrolimus plus prednisolone in 2 patients, and cyclosporine plus prednisolone plus rapamycin in 1 patient.
During the study period, 2000 IU of HBIG was administered in regular 6-week intervals ± 2 days. IM HBIG was administered in 2 portions of 1000 IU (5 mL) into each gluteus muscle. HBIG administration was started intravenously in 12 patients of group A and intramuscularly in 12 patients of group B. At week 24, HBIG administration was switched from IM to IV and vice versa (crossover study). Anti-HBs levels were measured 2, 4, and 6 weeks after each HBIG administration [for example, 4 × 3 values during the 24-week IM period and 4 × 3 values during the 24-week IV period (in all, 24 values per patient × 22 patients)].
Twenty-four patients were vaccinated with conventional double-dose recombinant vaccine containing 40 μg of HBsAg up to 12 times.
During the study period, each patient was vaccinated 12 times [at weeks 0, 2, and 4 (cycle 1), 12, 14, and 16 (cycle 2), 24, 26, and 28 (cycle 3), and 36, 38, and 40 (cycle 4)] with a double-dose HBsAg vaccine administered to the left deltoid muscle (Engerix B, GlaxoSmithKline; 2 × 20 μg).21 Response to active vaccination was defined as a reconfirmed increase of anti-HBs unexplained by HBIG administration or a lack of anti-HBs decrease below 100 IU/L after discontinuation of HBIG treatment after week 48.
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. Written informed consent was obtained from each patient.
The intravenously administered HBIG preparation was Hepatect CP, which was manufactured by Biotest (Dreieich, Germany) and contained 50 mg of plasma protein standardized to 50 IU of anti-HBs per milliliter (package insert). The intramuscularly administered HBIG preparation was Hepatitis B Immunoglobulin Behring, which was manufactured by CSL Behring (Marburg, Germany) and contained 100–170 mg of plasma protein per milliliter standardized to 200 IU of anti-HBs per milliliter (package insert). Costs per 2000 IU in the year 2006 were €1719 for Hepatect CP versus €1362 for Hepatitis B Immunoglobulin Behring. The anti-HBs content of both preparations was measured with the same assay (Axsym AUSAB microparticle enzyme immunoassay, Abbott Diagnostics, Delkenheim, Germany).
Anti-HBs Level Measurements
During the study, blood samples were collected for serum anti-HBs level measurement every 2 weeks ± 2 days, that is, 2, 4, and 6 weeks after HBIG administration. Directly after the 6-week measurement, another HBIG dose was administered. Serum anti-HBs levels were determined with the commercial Axsym AUSAB microparticle enzyme immunoassay (Abbott Diagnostics).
Calculation of Pharmacokinetic Parameters
Elimination rate constants for the calculation of elimination half-lives were determined from log-linear regression of 3 anti-HBs serum concentration values. AUC between week 2 and week 6 after HBIG administration was estimated with the trapezoidal rule.
The statistical evaluation was performed with SPSS 14. Anti-HBs levels, half-lives, and AUC values after IV HBIG administration versus IM HBIG administration were compared pairwise with the t test for independent samples (Table 2). Anti-HBs levels at consecutive time points were compared pairwise by analysis of variance for repeated measures (Table 2). Intraindividual and interindividual mean variances of anti-HBs levels after IV and IM HBIG administration was compared by the nonparametric Mann-Whitney test (shown later in Fig. 4). Box plots were produced by means of SPSS 14.0 (shown later in Figs. 3 and 4).
Table 2. Anti-HBs Levels, Half-Lives, and AUC Values After IV and IM HBIG Administration
NOTE: The mean ± standard deviation, median, and range of single patients' mean anti-HBs concentrations, half-lives, and AUC values are given for separate groups and all patients together. P values far above 0.05 point out the absence of significant differences between administration modes, whereas P values below 0.0005 demonstrate the highly significant decline of anti-HBs levels at consecutive time points (2, 4, and 6 weeks after HBIG administration).
Abbreviations: anti-HBs, antibody against hepatitis B surface antigen; AUC, area under the curve; HBIG, hepatitis B immunoglobulin; IM, intramuscular; IV, intravenous.
Anti-HBs level at week 2
518 ± 175, 490
484 ± 174, 513
442 ± 154, 456
429 ± 161, 406
Anti-HBs level at week 4
343 ± 123, 360
336 ± 164, 294
294 ± 130, 281
284 ± 130, 226
Anti-HBs level at week 6
251 ± 115, 222
238 ± 118, 207
190 ± 91, 180
198 ± 106, 174
26.7 ± 7.0, 25.0
26.3 ± 6.2, 24.7
22.7 ± 4.9, 22.4
24.6 ± 5.9, 23.0
10.2 ± 3.7, 10.6
9.7 ± 4.3, 8.4
8.5 ± 3.5, 8.6
8.4 ± 3.7, 7.2
Anti-HBs level at week 2
480 ± 166, 471 (225–867)
457 ± 166, 461 (179–833)
<0.0005 (week 2 versus 4)
Anti-HBs level at week 4
319 ± 126, 298 (129–599)
310 ± 147, 269 (91–618)
<0.0005 (week 4 versus 6)
Anti-HBs level at week 6
221 ± 106, 197 (67–484)
218 ± 112, 187 (60–485)
24.7 ± 6.2, 23.8 (16.1–39.6)
25.5 ± 6.0, 23.6 (17.2–36.1)
9.4 ± 3.6, 8.8 (3.8–16.8)
9.0 ± 3.9, 8.1 (2.9–17.9)
Pairwise comparisons of anti-HBs levels at 3 different time points, anti-HBs half-lives, and AUC values showed no difference for IV administration mode versus IM administration mode (Table 2). Kinetics of mean anti-HBs levels of single patients 2, 4, and 6 weeks after HBIG administration are given in Fig. 1. Figure 2 illustrates a direct comparison of anti-HBs trough levels measured 6 weeks after IM HBIG administration versus IV HBIG administration (mean of 4 values for each data point). Figure 3 illustrates the similarity of the anti-HBs concentrations, half-lives, and AUC values after IV and IM HBIG administration.
In contrast, anti-HBs levels at consecutive time points (2 versus 4 weeks and 4 versus 6 weeks after HBIG administration) declined continuously and significantly (Table 2) without any difference between IV and IM administration modes.
Regression analysis of single patients' anti-HBs trough levels over time (from first to eighth HBIG administration) showed no significant trend of increasing or decreasing values in any of the patients.
To compare intraindividual and interindividual variation of anti-HBs levels 2, 4, and 6 weeks after HBIG administration, variances of single patients' values at different time points (4 IM and 4 IV values for each patient) were compared with variances of all patients' values at defined time points (22 values per time point; Fig. 4). After IV HBIG administration, the mean intraindividual variance of anti-HBs levels (15,660, 6430, and 5430 IU/L) was significantly lower than the interindividual variation (39,200, 20,590, and 15,260 IU/L) at all 3 time points (P = 0.007, P = 0.002, and P = 0.021, respectively), just as after IM HBIG administration (12,860, 4750, and 2590 IU/L for intraindividual variation and 35,530, 24,620, and 14,340 IU/L for interindividual variation with P = 0.004, P < 0.001, and P = 0.001, respectively), as illustrated in Fig. 4.
In Table 3, the percentage of patients in whom anti-HBs did not drop below predefined levels (for example, 100 IU/L) after administration of 2000 IU of HBIG in 6-week intervals is displayed.
Table 3. Anti-HBs Levels 2, 4, and 6 Weeks After IV and IM HBIG Administration
NOTE: The data are presented as percentage of patients with a mean anti-HBs concentration above the given level (mean of 4 values)/percentage of patients with repeated anti-HBs measurements above the given level (4 values each).
Abbreviations: anti-HBs, antibody against hepatitis B surface antigen; HBIG, hepatitis B immunoglobulin; IM, intramuscular; IV, intravenous.
IV HBIG Administration
IM HBIG Administration
During the study period, HBsAg remained negative in all patients, and no laboratory or clinical signs of HBV reinfection were seen in any patient.
Our study demonstrates for the first time that crucial pharmacokinetic parameters, including anti-HBs trough levels at time of HBIG readministration, do not differ significantly after IV and IM HBIG administration. All available previous trials on HBIG prophylaxis after liver transplantation postulated a reduction of HBIG costs of 50% or more by a switch from IV HBIG administration to IM HBIG administration.6–8 However, these trials had substantial shortcomings: Adequate comparative pharmacokinetic data measured after IV and IM HBIG administration were not presented, the number of patients and measured data points was to small, or the study protocol was too confusing to draw meaningful conclusions. Our data reveal that anti-HBs levels tend to differ between IV and IM administration in single patients, underlining the importance of a sufficient number of patients and data points. However, our analysis of a large number of anti-HBs levels, resulting half-lives, and AUC values measured in a high number of patients with a crossover design does not show a pharmacokinetic advantage of IM HBIG administration compared to IV HBIG administration and unmasks trends in single patients as coincidental.6, 7 On the other hand, it is important to stress that our analysis does not show a pharmacokinetic disadvantage of IM HBIG administration that may have been postulated because of potentially lower bioavailability. Therefore, we confirm the cost-effectiveness of IM HBIG, which is easy to calculate because it is restricted to the HBIG price: IM HBIG is cost-effective if the price per unit of anti-HBs is lower compared to IV HBIG (20% in our study) or vice versa.
On the background of the additional repeated active vaccination with conventional HBsAg vaccine in our study, which failed to induce a significant anti-HBs response in almost all patients as reported previously,21 it is important to note that this evaluation does not even show a trend of increasing anti-HBs levels toward the end of the study. The crossover design of the study was performed for 2 reasons. First, we wanted to prevent the difficulty of interpreting rising anti-HBs levels toward the end of the study. However, such a trend did not occur in either of the 2 groups. Second, the crossover design prevented an uneven distribution of patients with overall higher versus overall lower anti-HBs levels in the IV versus the IM group, which is indeed visible as a trend in groups A and B.
In principle, these data are consistent with published pharmacokinetic data measured after IV infusion or IM injection of antibodies. Previously published kinetics clearly show an exponential antibody decrease well described by a terminal elimination half-life. Indeed, the measured half-lives of 25 ± 6 and 26 ± 6 days in the present study perfectly fit the reported terminal elimination half-lives of 22 to 25 days.22–24
Kinetics measured after IM injection of antibodies (in contrast to IV administration) are influenced not only by distribution and elimination but also by absorption processes. Systemic absorption of antibodies following IM administration most likely occurs via the lymphatic system. As lymph fluid drains slowly into the vascular system, absorption of antibodies from the site of administration continues for a prolonged period.22 In animals and in man, the time to maximal anti-HBs plasma concentrations for different monoclonal antibodies has been reported to be 3, 5.5, and 7 to 8 days.22 Little pharmacokinetic data about IM administration of HBIG are available. Partovi et al.19 determined the time of maximum anti-HBs levels to be between 4 and 20 days (mean: 10.5 ± 6.7) after HBIG administration.19 However, because of the variability of predose anti-HBs levels, variable HBIG doses, and a small number of patients (6), the power of this study is limited. In principle, prolonged absorption of drugs, as observed after IM or subcutaneous (SC) injection, may result in increased drug levels at later time points, if absorption is complete, because AUC values after IV, IM, or SC drug administration are equal if bioavailability is 100%. However, after IM or SC administration, bioavailability may be reduced because of local degradation of antibodies by proteolytic enzymes. Indeed, most investigations of antibody absorption following IM injection have reported incomplete absorption, with bioavailabilities ranging from 50% to 100%.22
On this background, our pharmacokinetic data interestingly do not show significantly lower AUC values after IM HBIG administration versus IV HBIG administration. Although AUC values during the first 2 weeks after HBIG administration, which were not measured in our study, presumably were lower after IM HBIG administration because of incomplete absorption, this did not influence anti-HBs levels after 2, 4, and 6 weeks significantly. On the other hand, no effect of protracted absorption was visible after IM HBIG administration, as there was no trend to higher anti-HBs levels at any time point.
Although a considerable number of studies have shown that it is possible to maintain adequate postoperative anti-HBs levels in stable patients after liver transplantation by IM HBIG administration,6–17 the pros and cons of IM prophylaxis have to be weighed carefully against IV prophylaxis on the background of lacking pharmacokinetic benefits. On the one hand, IM administration may be comfortable because the injection is performed quickly. On the other hand, this route of administration can be painful because of the large volumes injected, and local side effects of injections affect some patients' quality of life and bear a minimal risk of infection and bleeding. On the basis of these pros and cons, IM prophylaxis may be considered if IV HBIG is unavailable or if a significantly lower price per HBIG unit favors the use of an IM preparation.
Principles of anti-HBs pharmacokinetics after IM HBIG administration can be expected to be applicable to SC administration, which is currently under investigation. Our study underlines that alternative routes of HBIG administration have to be compared directly with the gold standard of IV HBIG administration to assess economic efficiency adequately.
Interindividual and intraindividual variations of anti-HBs levels measured after administration of defined HBIG doses in regular intervals have been observed in previous studies.20, 25 Besides intrarun and interrun measurement inaccuracies, the interindividual variation in particular presumably mainly results from interindividual differences in distribution and elimination processes after IV administration. Differences in anti-HBs levels may be additionally caused by variable absorption processes after IM injection. Thus, interindividual and intraindividual variations did not differ after IV and IM HBIG administration and therefore do not favor one of the two administration modes.
As a result of anti-HBs level variability, dosing schemes with fixed HBIG doses and dosing intervals appear to be inadequate for a cost-effective prophylaxis with patient-independent defined anti-HBs trough levels. Although anti-HBs levels were kept above 100 IU/L in most patients at most time points by 6-week administration of 2000 IU of HBIG in the present study (Table 3), this could have been achieved by lower HBIG doses or longer HBIG intervals in the majority of patients. Individualized fixed dosing intervals would presumably be more cost-effective, as intraindividual anti-HBs level variability is significantly lower than interindividual variability. However, an individualized anti-HBs level–dependent HBIG administration can be expected to be most cost-effective. If anti-HBs levels are measured at regular intervals, for example, every 2 weeks, individual anti-HBs levels and dynamics of anti-HBs decrease can be assessed, and measurement intervals can be adjusted. If a defined trough level, such as 100 IU/L, is sought, this strategy presumably not only is cost-effective because it prevents overdosing but also ensures a high level of safety because it simultaneously prevents underdosing.20 Finally, it is important to emphasize that pharmacokinetic data in the present study were measured in stable HBsAg-negative patients with at least 12 months follow-up after OLT. HBIG doses required to decrease HBsAg levels and raise anti-HBs levels in the initial posttransplant period strongly depend on quantitative HBsAg levels at the time of liver transplantation, as shown previously.26
In conclusion, pharmacokinetic data measured in the present study favor neither IM nor IV HBIG administration. Therefore, the advantage of a quick and easy injection of a usually cheaper IM HBIG preparation has to be weighed against the disadvantage of potential discomfort for patients and risk of complications after IM HBIG administration. HBIG dosing intervals should be individualized independently of the administration mode.