Experience with lamivudine therapy for hepatitis B virus infection before and after liver transplantation, and review of the literature
Abstract. Ben-Ari Z, Mor E, Tur-Kaspa R (Beilinson Campus, Petah Tikva, and Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel). Experience with lamivudine therapy for hepatitis B virus infection before and after liver transplantation, and review of the literature. J Intern Med 2003; 253: 544–552.
Objectives. To analyse the results of lamivudine therapy on suppression of hepatitis B virus (HBV) replication before transplantation and on preventing graft reinfection postoperatively.
Design. Long-term clinical study.
Setting. Liver Institute and Department of Transplantation of a tertiary-care university-affiliated centre.
Subjects. (1) 14 candidates for liver transplantation with decompensated liver disease caused by active replication of HBV; (2) six patients with recurrent HBV infection after transplantation.
Intervention. Lamivudine 100 mg daily; administered in group 1 before surgery and continued after in nine patients who underwent transplantation; administered in group two postoperatively only. antihepatitis B surface antigen immunoglobulin (HBIg) was administered postoperatively in both groups.
Main outcome measures. Immunoassay evaluation of serum hepatitis B surface antigen, serum hepatitis Be antigen and serum HBV DNA (hybridization and PCR); sequencing through the tyrosine-methionine-aspartate-aspartate locus of the HBV polymerase gene in patients with lamivudine breakthrough; inflammation and fibrosis scoring on liver biopsy before and at least 2 years after lamivudine therapy in group 2.
Results. Pretransplantation therapy (group 1) significantly suppressed HBV replication and enabled nine patients (64.2%) to undergo transplantation. Only one patient (7.1%) had lamivudine breakthrough, and one (7.1%) had recurrent HBV. Lamivudine administration begun after transplantation (mean 48.0 months, range 30–60 months) because of graft reinfection (group 2) was associated, over the long-term, with the emergence of high mutation rates (83.3%), histological disease progression (66.6%), and hepatic failure (33.3%).
Conclusions. In patients with chronic HBV infection and active viral replication, lamivudine therapy is effective when started before transplantation. However, its long-term administration after transplantation for recurrent HBV leads to high resistance rates. Combination therapy with lamivudine and HBIg immunoglobulin can substantially reduce the recurrence rate. Further studies on combination antiviral therapy are needed in this patient population.
Chronic hepatitis B virus (HBV) infection is a common cause of decompensated end-stage liver disease. Hepatitis B virus-infected patients form a large indication for orthotopic liver transplantation (OLT). Until recently, it was considered to be a relative contraindication for liver transplantation because of poor outcome posttransplantation. In 1991, Todo et al.  reported that recurrent HBV infection was the most common cause of death at more than 60 days after liver transplantation in patients who underwent the procedure for end-stage hepatitis B. Recurrence almost invariably leads to chronic infection, with an often aggressive course and rapid progression to cirrhosis . The introduction of posttransplant immunoprophylaxis with polyclonal antihepatitis B surface antigen immunoglobulin (HBIg) has significantly reduced recurrence rate [2, 3], and today, chronic HBV infection accounts for a large proportion of OLTs. However, HBIg prophylaxis is very expensive, has to be given indefinitely in large amounts, and cannot be used to treat recurrent hepatitis B, if it occurs . Furthermore, even with prophylaxis, 50% of patients with evidence of active viral replication before transplantation will have graft failure within 3 years [2, 5], probably because of preexisting circulating virions or enhanced replication of HBV in extrahepatic sites stimulated by immunosuppressive therapy [6, 7].
Lamivudine (3TC) is the negative enantiomer of 3′thiacytidine, a 2′3′-dideoxynucleoside. It serves as a potent inhibitor of HBV replication in patients with chronic HBV infection  by causing chain termination of RNA-dependent HBV polymerase . Several studies have reported that lamivudine administered before and after liver transplantation to prevent HBV-induced graft reinfection caused complete and sustained suppression of viral replication [10–12]. However, its prolonged use has been associated with the emergence of lamivudine-resistant HBV. Rates of emergence range from 17 to 46% in patients treated for 1 year [13, 14] to 67–75% after 3–4 years of continuous therapy . Drug-resistant mutants have also been observed in 27–62.5% of patients after liver transplantation for HBV infection [12, 16]. The emergence of viral variants results from one or more mutations in the tyrosine-methionine-aspartate-aspartate (YMDD) locus of the HBV polymerase gene. It is the nucleotide-binding domain for virally encoded DNA polymerase [17, 18].
The aim of the present study was to analyse the results in our centre with lamivudine, administered before liver transplantation to suppress HBV replication and after liver transplantation in combination with HBIg to prevent graft reinfection or to treat recurrent or de novo HBV infection. Related findings in the literature are reviewed.
Between January 1992 and December 2000, 134 patients underwent OLT at the Rabin Medical Center, Israel. Thirty were referred because of HBV-related end-stage liver disease.
The study group consisted of the 20 patients (16 males, four females) who were treated with oral lamivudine 100 mg daily on a compassionate-use basis before or after liver transplantation to prevent graft reinfection or to treat recurrent HBV. Mean age at transplantation was 46.3 ± 8.6 years. Before initiation of treatment, all patients had evidence of active viral replication: positive serum HBV DNA (by hybridization) and positive serum hepatitis B surface antigen (HBsAg); 11 patients (55%) were positive for serum hepatitis Be antigen (HBeAg) and nine (45%) were positive for serum HBe antibody. Three patients were coinfected with hepatitis D virus (HDV), and one with hepatitis C virus (HCV). Patients received standard immunosuppression treatment with cyclosporin or tacrolimus, azathioprine or CellCept, and corticosteroids, as well as HBIg immunoprophylaxis (Hepatect R; Biotest Pharma GmbH, Dreieich, Germany). An initial dose of 10 000 U HBIg intraoperatively was followed by 10 000 U during the first postoperative week, with continuous treatment to maintain anti-HB titres above 300 mIU L−1.
Patients were divided into two groups (Table 1).
Table 1. Patient characteristics at baseline (before transplantation)
|On waiting list||3||–|
|Died whilst on waiting list||2||–|
|Mean age at OLT (year)||43.9a||53.8|
|Serum HBsAg positivity||14||6|
|Serum HBeAg positivity||8 (57%)||3 (50%)|
|Serum HBV DNA positivityb||14||6|
Treatment was initiated before transplantation in 14 patients (11 males, three females) with decompensated replicative HBV cirrhosis who were otherwise candidates for liver transplantation. Mean age was 47.8 ± 8.8 years. All were serum HBV DNA-positive (by hybridization) and HBsAg-positive; eight (57%) were serum HBeAg-positive. Nine patients (mean age 43.9 years) underwent OLT when serum HBV DNA became negative in at least two consecutive serum samples, 2 months apart; all received HBIg after transplantation combined with lamivudine therapy. Of the remainder, two patients died of spontaneous bacterial peritonitis (n = 1) and massive oesophageal bleeding (n = 1) after 2 and 4 months of lamivudine therapy, respectively, and three are still on our waiting list after 4–10 months of lamivudine therapy, although serum HBV DNA became negative in two consecutive serum samples. Median duration of lamivudine therapy before and after liver transplantation was 8 months (range 2–21 months) and 21 months (range 1–35 months), respectively. Total duration of therapy was 17.3 months (range 2–42 months).
Six (four males, two females) of the 29 patients (20%) who underwent OLT caused by HBV infection (all were serum HBV DNA-negative prior to transplantation) had recurrent HBV infection after transplantation. Median age of this group was 53.8 ± 10.1 years. The median time from transplantation to recurrent infection was 10 months (range 6–60 months). All had received HBIg monotherapy after transplantation. The diagnosis of recurrent HBV infection was based on increased levels of serum liver enzymes, the reappearance of serum HBsAg, positive HBV DNA (by hybridization) and/or HBeAg (50%), and positive staining of liver tissue for hepatitis B core antigen (HBcAg) and HBsAg. Median duration of lamivudine treatment was 48.1 months (range 30–60 months).
All 20 patients were closely monitored during lamivudine treatment for serum liver enzymes, HBsAg, HBeAg and HBV DNA. Patients with negative findings on hybridization DNA analysis were tested for HBV DNA with polymerase chain reaction (PCR) technique. Liver biopsies were performed in patients from group 2 before administration of lamivudine and for at least 2 years after.
HBsAg, HBeAg and the corresponding antibodies (anti-HBs, anti-HBe) were tested by commercially available enzyme-linked immunoassays (Abbott Laboratories, Chicago, IL, USA). For serum qualitative HBV DNA, 50 μL serum was alkaline-treated, transferred to Biodyne nylon membrane (Pall Corp., Glen Cove, NY, USA) and neutralized. The membrane was hybridized to 32P-labelled HBV DNA (lower limit of detection <106 copies per mL) . Since 2002, the HBV DNA PCR technique was performed using the commercial Amplicor HBV monitor test (Roche Diagnostics, Branchburg, NJ, USA) (lower limit of detection, <400 copies per mL), for quantifying HBV viral DNA.
Whenever breakthrough was documented (reappearance of serum HBV DNA following an undetectable levels by hybridization and increase in serum liver enzyme levels in patients with serum positive HBsAg and HBeAg or HBeAb), HBV DNA was amplified from serum and sequenced through a conserved polymerase domain, the YMDD locus, before and after development of lamivudine resistance (Glaxo-Wellcome Research and Development Laboratories, UK), as described previously .
The paired liver biopsies were scored blindly with the Hepatic Activity Index (HAI) for degree of necroinflammation and fibrosis. Improvement was defined as a 2-point or more reduction in score.
The virological findings and outcome are shown in Table 2, and the histological findings in the patients with lamivudine resistance are shown in Table 3. Table 4 summarizes the genotypic resistance and outcome.
Table 2. Virological response and outcome during lamivudine therapy
|Duration of therapy (mo):||20||48|
|Serum HBsAg clearance||9 (64.2%)||0%|
|Serum HBeAg clearance||9 (64.2%)||0%|
|Serum HBV DNA clearance||14 (100%)||6 (100%)|
|Time of HBV DNA clearance (months)|
|Median Range||1.5 (1–2)||1.5 (1–2)|
|Recurrent HBV infection||1 (7.1%)||–|
|HBV DNA breakthrough||1 (7.1%)||5 (83.3%)|
|Time to breakthrough (months)||22||14|
|HBV DNA copies per mLa||265 × 103||1.8 × 106|
|Median range||(200–330 × 103)||(348 × 103−8.4 × 106)|
|Stable||6b (66%)||4 (66.6%)|
|Hepatic decompensation||–||1 (16.7%)|
|Died||3b (33%)||1 (16.7%)|
Table 3. Hepatic Activity Index before and during lamivudine therapy (Knodell's score) in patients who developed lamivudine resistance
|Duration of lamivudine therapy (months)||20||52|
|Median range|| ||(40–60)|
| Baseline||2||3 (2–5)|
| During therapy||4||5 (4–9)|
| Baseline||0.5||1 (0–1)|
| During therapy||2||3 (2–4)|
Table 4. Genotypic resistance to lamivudine and outcome
|Time of YMDD analysis (months)a||8||10|
|Median range|| ||(9–20)|
|YMDD mutation (amino acids)||M552V and L558M||Two patients (M552I)|
|Three patients (M552 and L558M)|
|Outcome (months after breakthrough)||Stable (5)||Three patients stable 43 (40–51)|
|One patient hepatic failure (25)|
|One patient hepatic failure: died (14)|
Viral replication disappeared during the first 2 months of treatment with lamivudine in all patients. Two patients showed a significant improvement on liver function tests (Child–Pugh score decreased to <7), and they were placed on inactive status (UNOS status 7) for liver transplantation for 8–9 months. However, progressive liver disease recurred in both, and both underwent OLT. In all, nine patients in this group underwent OLT. All were given lamivudine 100 mg and HBIg after transplantation. Seven patients remained HBV DNA-negative over a median follow-up period of 8 months (range 0.5–24 months). Lamivudine breakthrough (diagnosed by sequencing through the YMDD locus) was noted in one patient (7.1%) and recurrent HBV infection in another (7.1%) after completion of 12–32 months of lamivudine therapy. At the last follow-up, both patients retained stable graft function. A repeated liver biopsy performed in the patient who developed lamivudine breakthrough 24 months after onset of lamivudine therapy demonstrated a worse score for the inflammatory and the fibrosis component of the HAI (Table 3). Mutations in the YMDD locus (L528M and M552V) were confirmed in the patient with lamivudine breakthrough (Table 4). No mutation was found in the YMDD locus in the patient with recurrent HBV infection. Three of the nine patients who underwent OLT died: one of cerebrovascular accident, one of sepsis, and one of hepatic artery rupture.
All six patients in group 2 were serum HBV DNA-positive (by hybridization) and three were also serum HBeAg-positive. Following 6 weeks of treatment, serum HBV DNA was no longer detected in any of the patients. However, after 52 weeks of treatment, it was not detectable in only one patient (16.6%); it remains undetectable in this patient at 24 months after therapy. Serum HBsAg and HBeAg remained positive in all patients; serum ALT level normalized in three patients.
In five patients (83.3%), serum HBV DNA breakthrough occurred after a median period of 13 months (range 9–20 months). All these cases were associated with increased serum ALT level and detectable serum HBV DNA (by hybridization). Sequencing through the YMDD locus of the HBV polymerase gene showed resistance because of a mutation: M552V and L558M in three patients and M552I in two (Table 4). In four of the five patients, liver histology findings at 24–30 months after lamivudine breakthrough showed progressive fibrosis with a significant increase (≥2 points) in the HAI score (Table 3). Two of them had evidence of clinical deterioration after the YMDD mutant virus appeared, and progressive hepatic failure. One of them died of massive uncontrolled oesophageal bleeding at 14 months after breakthrough.
Five of the six patients in this group are still alive. Three remain clinically stable at 40–51 months after lamivudine breakthrough.
Lamivudine was well tolerated. None of the patients discontinued the drug, and there were no side-effects.
The safety and efficacy of lamivudine has been evaluated in transplant patients with end-stage liver disease caused by chronic HBV infection [20–22]. In the present study, lamivudine administration before transplantation (group 1) extended the transplantation-free time in two patients, served as a bridge to OLT, and led to clearance of HBV from serum in all patients, thereby enabling nine of them to undergo OLT. Lamivudine therapy was continued after transplantation in these patients, in combination with HBIg. In one patient (7.1%), lamivudine resistance developed 6 months later, after a total of 12 months of therapy. Mutations in the YMDD locus (L528M and M552V) were confirmed in our patient with lamivudine resistance, but he retained stable graft function at 4 months after breakthrough. Another patient (7.1%) developed recurrent HBV infection 10 months after transplantation. No mutation was found in the YMDD locus, and he also retained stable graft function 6 months later. Lamivudine was well tolerated prior to and following liver transplantation.
Six (20%) of our patients who received HBIg perioperatively had recurrent HBV at 6–60 months after transplantation (group 2). Long-term lamivudine therapy (34–60 months, median 48 months) resulted in HBV clearance from serum in only one (16.6%), and none of the patients lost serum HBsAg or HBeAg; 83.4% of these patients developed lamivudine resistance. Patients who developed lamivudine resistance did so within 9–20 months (mean 13 months); in all cases, mutations were confirmed by sequencing through the YMDD locus of the HBV polymerase gene. The histological outcome after the emergence of lamivudine-resistant HBV showed disease progression in 62.5%. Amongst our patients with virological breakthrough, the clinical disease outcome was mild in 66%; 33% had hepatic failure. The HBV recurrence rate was lower in patients treated with lamivudine in combination with HBIg after transplantation as immunoprophylaxis (group 1) than in patients given HBIg monotherapy (group 2) (7.1% vs. 20%).
In a previous study , our group found that the administration of lamivudine in patients with active viral replication before transplantation suppressed HBV replication and induced clinical and biochemical improvement, extending the transplantation-free time. Accordingly, others [22, 23] noted an improvement in Child–Pugh–Turcot scores in 69% of patients with HBV infection treated with lamivudine; 38% were placed on inactive status for liver transplantation .
Two groups of mutations confer HBV resistance to lamivudine, as indicated by a decreased replication capacity in tissue cultures and in vivo: substitution of amino acids in the B and C domain (L528M +M552V) and substitution in the YMDD motif of the C domain (M552I) of the DNA polymerase [17, 18, 24–27]. The L528M + M552V mutants seem to be the more frequently encountered.
Perioperative and long-term postoperative administration of HBIg contributes significantly to the prevention of HBV recurrence [2, 3]. It is especially effective in patients with fulminant hepatitis B and hepatitis D cirrhosis , and less so in patients with HBV cirrhosis, with a 67% actuarial risk of recurrence at 3 years in patients who were HBV DNA-negative before transplantation . Patients who have active viral replication (HBV DNA-positive) are at significantly increased risk for reinfection (83%) . All previous studies of lamivudine therapy for recurrent HBV infection after liver transplantation were used on open-label, nonrandomized, uncontrolled design [12, 16, 28–35](Table 5). In most of the studies, the patient sample was small and heterogeneous, and in some, HBIg immunoprophylaxis was not administered. Lamivudine was administered orally at a dose of 100–150 mg dL−1 (depending on the study design). The duration of therapy was 11–48 months. Serum ALT normalized in 11.1–71% of patients, and HBV DNA disappeared from serum in 16.7–87%, although the majority achieved early serum HBV DNA negativity. Loss of serum HBsAg was noted in only 0–45%. Andreone et al. , who treated 11 liver transplant patients with recurrent HBV infection, reported clearance of HBsAg from serum in five of them (45%). However, in that study, treatment was started within 8 weeks of the reappearance of HBsAg in serum, so the results represent early rather than delayed exposure to therapy. Loss of serum HBeAg was also achieved in minority, 0–31%. Only Andreone et al.  reported that seroconversion occurred in two patients who were serum HBeAg-positive prior to lamivudine therapy.
Table 5. Studies of lamivudine therapy for recurrent HBV infection after liver transplantation a
|Nery et al. (28)|| 8||27.5||NA||87 (PCR)||NA||NA||NA||28.5||Mild clinical |
|Andreone et al. (29)||11||17||45.4||45 (PCR)||–||45||12.5||27||Mild clinical |
|Dodson et al. (30)||26||18||NA||33 |
| 0|| 0||NA||67||Hepatic failure |
|Perrillo et al. (12)||52||12||71||60 |
|31|| 6||56||27||Mild clinical |
|McCaughan et al. (31)||10b||11||NA||30 |
|–||NA||NA||60||Hepatic failure |
|Malkan et al. (32)a||15||25||NA||86.6 |
|NA||27||NA||13||Hepatic failure |
|Seehofer et al.c (33)||34||12–36||NA||40 |
| 0||26||NA||41.1||Death |
|Ben-Ari et al. (16)|| 6||48||50||16.7 |
| 0|| 0||33.3||83.4||Hepatic failure |
|Fontana et al. (34)||33||21||27||36 |
| 9|| 0||NA||39.3||Hepatic failure |
|Chu et al. (35)|| 9||28||11.1||66 |
|11.1|| 0||NA||33||Hepatic failure |
In only one previous study were repeated liver biopsies performed (Table 5); Perrillo et al.  noted an improvement in the histological HAI, specifically in the inflammatory component, in 56% of 52 patients given lamivudine prophylaxis against hepatitis B infection after liver transplantation. However, they performed the repeated liver biopsy after 24 weeks of therapy.
In the large series of Atkins et al. , lamivudine resistance seldom occurred during the first 36 weeks of therapy. The emergence of lamivudine resistance after liver transplantation (13–83.4%) has been previously described (Table 5). Resistance was either proven or assumed to be associated with the emergence of lamivudine-resistant HBV. Perrillo et al.  reported a mutation rate of 27%, but they administered lamivudine for only 12 months. Dodson et al.  noted a 67% rate of resistance at 18 months. In another study, the resistance rate increased with the duration of lamivudine therapy at an approximate rate of 20% per annum . The higher frequency in the transplant setting may be attributable to the enhancing effect of glucocorticoid therapy on viral replication, which is associated with an increased risk of viral mutation [38, 39]. We recently published a study of the histological outcome of four patients with lamivudine resistance , also included in the present study. In three of them, progressive hepatic fibrosis was noted at 24 months. Mutimer et al.  also reported that the lamivudine resistant phenotype can cause severe graft damage, and in the Perrillo study , lamivudine resistance was associated with worse histological findings in 50%. Liaw et al.  also noted an exacerbation of liver disease in 41% of patients after the emergence of YMDD mutants during continued use of lamivudine.
In the other published studies (Table 5) there was either no clinical deterioration, or some patients (7.6–33%) developed hepatic failure. However, our study, though small in size, had the longest duration of lamivudine treatment. The reason for the differences between the studies is unclear, and may be related to study design, the nonhomogeneous patient selection, and the combined use of antiviral therapy in some studies.
Patients with YMDD variants still need to be maintained on prolonged lamivudine monotherapy despite its relatively high rate of virological breakthrough and association with graft failure, because resurgence of the wild type virus may occur on lamivudine withdrawal, leading to clinical deterioration and disease progression . Therefore, controlled studies to evaluate the safety and efficacy of combination antiviral therapy are warranted for this patient group. Recently, the new nucleoside analogue, adefovir dipivoxil, a potent inhibitor of HBV polymerase, has been reported to be effective in patients with lamivudine-resistant strains of HBV [44, 45], inducing profound suppression of HBV replication and leading to clinical improvement.
It is well known that donor organs from patients who are serum HBsAg-negative but hepatitis B core antibody-positive have the potential to transmit HBV to the recipient . However, the scarcity of organ donors has made it necessary to consider transplantation of organs with a potential to transmit HBV. Results vary, depending on the status of the recipient, with respect to antibodies directed at the virus. Anti-HBc-positive and anti-HBs-positive recipients are generally resistant to HBV infection. However, the probability of de novo HBV infection in recipients of anti-HBc-negative and anti-HBs-negative, is high (about 70%) . Several authors have suggested that de novo hepatitis B after transplantation is not the same disease as recurrent hepatitis B, which is usually associated with a mild clinical course . However, others have reported that it may also be a significant cause of graft dysfunction .
Despite the favourable long-term outcome with HBIg after liver transplantation, HBV recurrence rates are still about 20–36%. Several studies have recently reported the development of HBV recurrence with strains containing surface antigen-encoding mutations of the polymerase protein that confer resistance to HBIg [50–52]. Prophylactic lamivudine monotherapy reportedly prevented significant graft reinfection [53–55], but viral breakthrough developed in 36–40% of patients. Failure of lamivudine prophylaxis has been associated with high baseline (pretreatment) serum HBV titres . Mechanistic evidence suggests that the combination of lamivudine/HBIg may have synergistic efficacy in preventing recurrent HBV infection in the graft [56, 57]. This effect has been attributed to lamivudine's potent inhibition of viral replication, which makes it less likely that the viral binding capacity of HBIg would be overwhelmed and HBIg-resistant mutations selected. In addition, HBIg may provide humoral immunity limiting viral spread and indirectly inhibiting viral replication, thereby reducing the chances of the emergence of YMDD variants . Several long-term studies have recently investigated this combination, using different schedules, dosages and modes of administration (intravenous, intramuscular) of HBIg and lamivudine [56, 57, 59–62]. They found that in patients with hepatitis B who were HBV DNA-negative at the time of transplantation, results were excellent when lamivudine was administered before transplantation and a combination of lamivudine and HBIg after transplantation indefinitely [62, 63]. HBV recurred in 4% of patients at a median follow-up of 31 months . To minimize the disadvantages of HBIg therapy (mainly high cost), Terrault et al.  suggested replacing long-term HBIg monotherapy with a combination of lamivudine and short-term HBIg in patients who were serum HBV DNA-negative before transplantation, and Yoshida et al.  suggested combining lamivudine with low-dose intramuscular HBIg.
In conclusion, our study supports earlier findings indicating that in patients with chronic HBV infection and active viral replication, lamivudine administration before transplantation is associated with a significant suppression of HBV replication and even extends the transplantation-free time. However, lamivudine therapy should be started when OLT is imminent, preferably no more than 1–2 months before. This gives ample time to decrease viral replication and avoids the appearance of the YMDD mutation. Long-term lamivudine administration after transplantation for recurrent HBV leads to high mutation rates and might be associated with histological evidence of progression and hepatic failure. Combination therapy with lamivudine and HBIg can reduce the recurrence rate substantially.
Confirmation of the results in a larger sample is warranted. Major therapeutic advances in the prevention of HBV reinfection after liver transplantation were made with the application of lamivudine and HBIg prophylaxis and the availability of safe and effective antiviral therapies for patients with recurrent HBV and lamivudine resistance after liver transplantation.
Conflict of interest statement
No conflict of interest was declared.
Received 15 August 2002; revision received 19 December 2002; accepted 28 January 2003.
Ziv Ben-Ari, MD, Liver Institute, Rabin Medical Center, Beilinson Campus, PO Box102, Petah Tikva 49100, Israel (Tel.: 972-3-9377253; fax: 972-3-9377277; e-mail: firstname.lastname@example.org).