Primary prophylaxis with lamivudine of hepatitis B virus reactivation in chronic HbsAg carriers with lymphoid malignancies treated with chemotherapy


Dr Giuseppe Rossi, Sezione Ematologia, Spedali Civili, p.le Spedali Civili, 1, 25100 Brescia, Italy. E-mail: rossi@


Hepatitis B virus (HBV) reactivation of various degrees of severity, including fulminant hepatitis, may develop in 20–50% of hepatitis B virus surface antigen (HbsAg)-positive patients undergoing immunosuppressive or cytostatic treatment. Lamivudine is a nucleoside analogue that can directly suppress HBV replication. We have performed a pilot study to test the efficacy and tolerability of lamivudine as a primary prophylaxis of HBV reactivation in 20 consecutive patients treated for haematological malignancies, mainly of lymphoid origin. Lamivudine, 100 mg/d, was given orally from the start until 1 month after the end of chemotherapy, which included corticosteroids and/or purine analogues in 85% of cases. It was well tolerated and did not cause any unexpected reduction of cytostatic drugs dosages. The chemotherapy programme was completed in all patients without modifications. A transient threefold increase in serum amylase was observed in one case. HBV-DNA levels decreased in six out of six patients (P = 0·039) and ALT levels in five out of six patients (P = 0·057) whose serum levels were abnormal at the onset of therapy. Two patients developed transient hepatitis. HBV reactivation was documented in only one of these patients who had stopped lamivudine 1 month before. No signs of HBV reactivation were detected both during and after treatment in 18 patients with a median follow-up of 6 months (range 3–12). Thus, primary prophylaxis with lamivudine may be a well tolerated and effective method to reduce the frequency of chemotherapy-induced HBV reactivation in chronic HBsAg carriers.

Cancer chemotherapy in chronic carriers of hepatitis B virus surface antigen (HbsAg) is known to promote viral replication and, when immunosuppressive treatment is stopped, the return of immune competence can be followed by liver damage of varying degrees of severity, including fulminant hepatitis. Hepatitis B flare-up, during or shortly after chemotherapy, has been reported for more than 20 years (Hoofnagle et al, 1982). It is a frequent problem, occurring in 21–53% of chronic HbsAg carriers (Nakamura et al, 1996; Kumagai et al, 1997; Yeo et al, 2000). In our experience, it occurred in three out of seven (43%) HbsAg-positive patients treated for lymphoma during a 20 month observation period and in none of 74 concurrent patients with negative or unknown HbsAg status. Acute hepatitis caused by hepatitis B virus (HBV) reactivation may be severe, with jaundice occurring in 10–22% of cases and mortality rates from acute liver failure ranging from 4% to 41% of affected patients (Lok et al, 1991; Kumagai et al, 1997; Liang et al, 1999; Markovic et al, 1999). Moreover it causes delays and reductions in the delivery of cytostatic drugs in most patients, and may hinder the continuation of the treatment programme.

Lamivudine (3′-thiacytidine) is a reverse-transcriptase inhibitor used for antiviral therapy in human immunodeficiency virus (HIV) infection, and it is effective against HBV in chronic hepatitis B infection. It is active and well tolerated and has little if any haematological toxicity. The most frequently reported side-effect has been elevation in amylase levels. Lamivudine has been used successfully at dosages of 100–300 mg/d for the treatment of acute HBV reactivation in single patients undergoing chemotherapy for solid tumours (Maguire et al, 1999), lymphoma (Clark et al, 1998; Ahmed & Keeffe, 1999) or after allogeneic bone marrow transplantation (Picardi et al, 1998). In two small series, it was able to control hepatitis in about 80% of cases (Yeo et al, 1999), particularly in hepatitis E virus surface antigen (HbeAg)-negative patients (Wong et al, 1998). Recently, it has been used in five patients who had experienced HBV reactivation during chemotherapy in order to allow the completion of cytostatic treatment and/or the delivery of high-dose therapy and autologous stem cell transplantation (Al-Taie et al, 1999; Silvestri et al, 2000). We have performed a pilot study to test the tolerability and efficacy of low-dose lamivudine as primary prophylaxis of HBV reactivation in HbsAg-positive patients undergoing chemotherapy for haematological malignancies.

Patients and methods

Twenty consecutive adult patients starting chemotherapy for a haematological malignancy between August 1999 and August 2000 were enrolled. They had to be chronic HbsAg carriers with any serum HBV-DNA level. Patients with acute hepatitis at the time of initiating chemotherapy as well as patients with chronic liver failure or with any condition contraindicating the use of chemotherapy were excluded. Pregnancy or lactation, pancreatitis, the presence of 6-mercaptopurine or azathioprine among scheduled cytostatic drugs and the absence of written informed consent were further exclusion criteria.

The characteristics of patients treated are summarized in Table I. Median age was 60·5 years. Six patients had elevated alanine aminotransferase (ALT) levels (range 48–230 U/l). Serum hepatitis E virus surface antigen (HbeAg) was present in four patients (20%), reflecting HBV epidemiology in Italy. Three out of four of the HbeAg-positive and three out of fifteen of the HbeAg-negative patients had detectable serum HBV-DNA levels (range 406–14830 mmol/l) at study entry. Two patients had hepatitis C virus co-infection. Non-Hodgkin's lymphoma (NHL) was the most frequent haematological malignancy (out of 12 cases, seven were low grade and five were high grade); five patients had chronic lymphocytic leukaemia (CLL). The chemotherapy regimens used included corticosteroids in 13 cases and purine analogues in five out of the seven cases not treated with corticosteroids. Lamivudine was given orally at the dose of 100 mg, once daily. Treatment started on the first day of chemotherapy and was discontinued 1 month after the end of the last chemotherapy cycle. Patients gave their written informed consent. Statistical analysis of the differences between pre- and post-treatment levels of ALT and HBV-DNA was performed using a paired t-test.

Table I.   Clinical characteristics of patients treated with lamivudine.
   Pre-treatment levels    





  • *

    NHL, non-Hodgkin's lymphoma; AML, acute myeloid leukaemia; CLL, chronic lymphocytic leukaemia; HD, Hodgkin's disease, MM, multiple myeloma, NN, normal values.

  •   †CHL, chlorambucil; PD, prednisone; FL, fludarabine; 2-CDA, 2-clorodeoxyadenosine; HD-CY, high-dose cyclophosphamide; miniALLO, reduced-intensity allogeneic stem cell transplantation; CHOP, cyclophosphamide, hydroxorubicin, vincristine, prednisone; ACVBP; intensified CHOP-like regimen; FLAG-IDA fludarabine, cytarabine. filgrastin, idarubicin.

  • Neg, negative; Pos, positive.

152FNN4409NegNegNHL high gradeACVBP
261F54/48NegNegNegNHL low gradeCHOP/HD-CY
344FNNNegNegNegNHL low gradeCHOP
455MNNNegNegPosAML fourth relapseFLAG-IDA
676FNNNegNegNegNHL low gradeCHL,PD
761FNNNegNegNegNHL high gradeACVBP
829MNNNegPosNegHD relapseFL, miniALLO
1070MNNNegNegNegNHL high gradeCHOP
1251FNNNDNegNegNHL high gradeCHOP
1345FNNNegNegNegNHL Low gradeFL,CY
1660MNN14 830PosNegNHL low gradeCY,PD
1750FNNNegNegNegNHL low grade2-CDA
1866M270/230NegNegNegNHL high gradeCHOP
1956M112/46NegNegNegNHL low gradePD,CHOP


Results of treatment are summarized in Table II. Lamivudine was given at the planned dose and for the planned duration of therapy in all cases, except in one patient who started treatment after four chemotherapy courses and before high-dose cyclophosphamide for stem cell mobilization. Lamivudine was well tolerated. The only side-effect was a transient threefold increase of serum amylase levels in one patient, which did not require treatment modification. Haematological toxicity was similar to that expected for the same chemotherapy regimens administered without lamivudine. The planned haematological treatment was carried out without modifications in 17 cases. In three patients, reduction of the dose of cytostatic agents or delay in chemotherapy administration were made. The response rate to chemotherapy was 100% in five patients receiving first-line treatment for high-grade NHL and 55% in 12 patients treated for low-grade NHL/CLL, nine of whom had resistant or relapsed disease.

Table II.   Toxicity and results of lamivudine treatment.




  • *

    Started lamivudine during the fourth cycle of chemotherapy.

  •   CR, complete response, PR, partial response; Rel; relapse; Progr: progression. Chemotherapy cycles are defined in the legend to Table I.

1*NoNo1487/975NoNoACVBPYesCR8 +
2NoNo2300/3154Yes10CHOP/HD-CYNoCR12 +
15Pancr (2)NoNNNoNoHD-CYYesNR7 +
17NoNoNNNoNo2-CDANoPR5 +

At the end of treatment, ALT levels normalized in five out of six patients, and HBV-DNA levels significantly decreased in six out of six patients with abnormal values before treatment respectively (Fig 1). Acute hepatitis developed in two patients. In the first, who was the only patient to start lamivudine after the initiation of chemotherapy, acute hepatitis developed after the fourth cycle. HBV reactivation was suspected but not proven as HBV-DNA levels did not increase and HbcAb IgM was negative. Virological studies were negative. A tentative diagnosis of toxic hepatitis was made. The clinical course was mild, and both chemotherapy and lamivudine treatment could be resumed and completed without further problems. In the second patient, hepatitis developed 2 months after high-dose cyclophosphamide and 1 month after discontinuation of lamivudine. HbcAb IgM was positive. HBV-DNA levels increased slightly over baseline levels. Hepatitis was clinically mild and resolved within 1 month. After a median follow-up of 6 months (range 3–12 months), no reactivation of HBV, as documented by an increase in ALT or HBV-DNA serum levels, has been observed among the other 18 patients.

Figure 1.

 Decrease of HBV-DNA and ALT levels during chemotherapy and lamivudine in six patients with abnormal values at the onset of treatment.


The prevalence of chronic carriers of HBsAg in cancer patients varies among different geographical areas and may be as high as 12% in endemic regions (Yeo et al, 2000). With the more widespread use of chemotherapy and stem cell transplantation, the possibility of HBV reactivation is becoming a problem of increasing importance that may adversely affect the final outcome of treatment. It is not possible to predict the occurrence and the clinical severity of HBV reactivation in single patients. Patients with haematological malignancies may be at increased risk of HBV reactivation as corticosteroids are more often components of their chemotherapy regimen. Indeed, the presence of corticosteroids among the protocol drugs is considered the most important predisposing factor for treatment-induced HBV reactivation (Ohtsu et al, 1991; Nakamura et al, 1996). As observed by Cheng (1996) the frequency of reactivation after chemotherapy cycles that did or did not include corticosteroids was 47% and 8% respectively. It has also been reported that the frequency as well as the severity of HBV flare-up was higher in HBeAg-negative patients (Nakamura et al, 1996; Kumagai et al, 1997; Markovic et al, 1999), who represent the majority of chronic HBsAg carriers in our country, further supporting the opportunity of adopting prophylactic measures to avoid HBV reactivation.

Several active antiviral agents against HBV are available. Recombinant α-interferon has been the most widely used, but its haemopoietic toxicity precludes its use as a prophylactic agent in association with aggressive chemotherapy. Nucleoside analogues, such as famciclovir and lamivudine, are also active against HBV by interfering with viral DNA replication. Famciclovir was effective in reducing HBV reactivation after allogeneic bone marrow transplantation (Lau et al, 1998). Lamivudine has been used more extensively and has proven effective both as a treatment and as a secondary prevention of chemotherapy-related HBV reactivation (Wong et al, 1998; Yeo et al, 1999; Silvestri et al, 2000). However waiting for HBV reactivation before starting antiviral treatment does not avoid the significant risk of fulminant hepatitis, particularly in HbeAg-negative patients (Kosaka et al, 1991). Furthermore, the development of chronic hepatitis may preclude the subsequent completion of the treatment programme for the haematological neoplasm. In our experience, of the six patients with acute hepatitis caused by HBV reactivation, three died and chemotherapy was stopped in two of the three surviving patients.

The toxicity profile of lamivudine is particularly favourable because it does not overlap with that of cytostatic agents, making this agent particularly suitable for a simultaneous use with chemotherapy. In the present study, lamivudine was given as a primary prophylaxis for HBV reactivation during the entire period of chemotherapy. Results confirm that lamivudine may be used safely without compromising the dose intensity of cytostatic drugs and the results of anti-neoplastic treatment. At the daily dose of 100 mg, lamivudine did not cause significant side-effects and confirmed its excellent tolerability. We did not observe the development of pure red cell aplasia that has been reported recently as a possible adverse affect of lamivudine (Majluf-Cruz et al, 2000).

The efficacy of lamivudine in the prevention of hepatitis B flare-up was remarkable. Indexes of HBV replication and of hepatic cytolysis decreased or normalized during lamivudine prophylaxis in spite of concomitant chemotherapy, which included corticosteroids and/or highly immunosuppressive nucleoside analogues in 90% of cases. Hepatitis occurred in two patients. It was mild and was not associated with signs of liver failure or of developing chronic hepatitis. In only one of the two cases, hepatitis was clearly attributable to HBV reactivation because HBcAb IgM was detected in the serum together with rising, albeit low, HBV-DNA levels. Of note, because, in this patient, hepatitis developed 2 months after chemotherapy and 1 month after lamivudine discontinuation, HBV reactivation could not actually be documented in any of the study patients while they were on lamivudine. Relapses of hepatitis B have been observed after lamivudine treatment cessation, and several reports have described the development of lamivudine resistance as a result of mutations in the YMDD motif of the hepatitis B virus polymerase (Tipples et al, 1996; Niesters et al, 1998), which occurred especially during long-term treatment (Honkoop et al, 1997). Because of that concern, our treatment protocol limited the duration of lamivudine prophylaxis to 1 month after completion of the chemotherapy programme. However the optimal duration of lamivudine prophylaxis remains to be determined by further studies.

In conclusion, the present study shows that lamivudine can be given safely in association with chemotherapy without causing significant side-effects, a reduction of the dosage or the efficacy of cytostatic drugs. The frequency and the severity of chemotherapy-related HBV reactivation may be significantly decreased by lamivudine prophylaxis. Further studies are indicated in order to confirm these preliminary results, and to determine the optimal duration of treatment as well as its possible impact on the development of HBV mutants.