Intended for healthcare professionals

Practice Lesson of the Week

Fatal reactivation of hepatitis B after chemotherapy for lymphoma

BMJ 2008; 337 doi: (Published 01 July 2008) Cite this as: BMJ 2008;337:a423
  1. Richard Dillon, senior house officer1,
  2. Gideon M Hirschfield, clinical lecturer2,
  3. Michael E D Allison, consultant2,
  4. Kanchan P Rege, consultant1
  1. 1Department of Haematology, Hinchingbrooke Hospital, Huntingdon
  2. 2Department of Hepatology, Addenbrooke’s Hospital, Cambridge CB2 2QQ
  1. Correspondence to: G M Hirschfield gideon.hirschfield{at}
  • Accepted 18 July 2007

Routine screening for hepatitis B in patients receiving chemotherapy or immunotherapy will save lives

One third of the world’s population has evidence of previous infection with the hepatitis B virus (hepatitis B core antibodies), and 350 million people have chronic infection (hepatitis B surface antigen).1 Global migration will change the prevalence of disease in the UK; currently 200 000 people are chronically infected, and around 1500 acute and 8000 chronic new infections are diagnosed annually ( Although intravenous drug users and homosexual men are at notable risk, most cases are in people coming from high prevalence areas, where vertical transmission is common.2

Patients receiving chemotherapy or immunomodulatory drugs who have been exposed to hepatitis B virus are at risk of viral reactivation.3 4 5 In this context, and particularly when steroids are included in the treatment protocol, it is thought that immune mechanisms keeping viral replication under control are suppressed, allowing unchecked viraemia. This occurs in a large proportion of patients who have been infected with hepatitis B virus, and it can be fatal. However, screening for hepatitis B virus before starting immunosuppressive treatments or chemotherapy is not done throughout the United Kingdom.

We present a case of fatal hepatitis B virus reactivation in a young woman treated for lymphoma. With predictions of a rising prevalence of hepatitis B virus in the UK, clinicians prescribing chemotherapy or immunosuppressive treatments (including biological agents such as rituximab) should adopt strategies for screening for hepatitis B and give prophylaxis where required to prevent similar occurrences.

Case report

A 21 year old woman originally from West Africa presented to hospital with a two month history of pleuritic chest pain and weight loss, having previously been fit and well. Chest x ray confirmed a large anterior mediastinal mass and associated left sided pleural effusion. After computed tomography, mediastinoscopic biopsy was performed. Histology confirmed a high grade mediastinal large B cell lymphoma. Bone marrow was normal, as were results of routine blood tests, including liver function tests. An HIV test was negative.

Treatment with R-CHOP chemotherapy (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone) was started. After the fourth cycle of treatment, imaging showed that the mediastinal mass was smaller.

Two weeks later she presented with nausea and vomiting; although clinical examination was normal, liver function tests showed serum transaminase values more than 10 times the upper limit of normal. Further history confirmed no other risk factors for liver disease, but she mentioned transient childhood jaundice. In the next three days transaminase values continued to rise, and jaundice and coagulopathy developed. Doppler ultrasound examination of the liver was normal but hepatitis B testing confirmed ongoing infection (hepatitis B surface antigen was positive; e antigen was negative and e antibody was positive; hepatitis B core antibody was IgM negative and IgG positive; and hepatitis B DNA was 4.522×108 IU per ml). Alternative diagnoses were excluded by an extended hepatitis screen, along with a comprehensive biochemical liver screen. Testing of a stored serum sample taken before chemotherapy showed that before treatment she was positive for hepatitis B surface antigen and negative for e antigen, suggesting chronic infection with a precore stop-codon variant.

Lamivudine antiviral therapy was started, and her care was transferred to the regional liver unit. Despite this, fulminant hepatic failure developed, with associated encephalopathy, renal failure, sepsis, uncontrolled haemorrhage, and acute lung injury. She met the criteria for super urgent liver transplantation, but the aggressive, partially treated lymphoma, which was confirmed by repeat imaging, precluded transplantation because it was felt she would not derive overall long term survival benefit from the procedure. Full supportive care was given on the intensive care unit, but a week later she died.


Reactivation of hepatitis B virus with immunosuppressive treatments is well described. It is most widely reported in connection with chemotherapy used to treat lymphoma. In a prospective study from China in which 27 patients positive for hepatitis B surface antigen underwent chemotherapy for lymphoma, 18 experienced viral reactivation, six became jaundiced, one developed non-fatal liver failure, and one died of liver failure.3 Another study of 305 patients from Slovenia found that 78% of patients carrying hepatitis B surface antigen had viral reactivation, and death from liver failure occurred in 37% of these.4 One study of the risk factors predicting reactivation found that using steroids, being hepatitis B virus DNA positive before chemotherapy, and having lymphoma or breast cancer all predicted viral reactivation.6

Although it is widely recognised that reactivation can occur with conventional immunosuppression, fewer clinicians are aware of the risk of hepatitis B reactivation with newer drugs, in particular rituximab, which has a higher risk of hepatitis than does conventional chemotherapy for lymphoma. Although lysis of hepatocytes infected with hepatitis B virus is mediated mainly by cytotoxic (CD8 positive) T cell immunity, B cells have important roles in primingthe cytotoxic T cell responses in hepatitis B infection.5 Reactivation of hepatitis B virus is also widely reported in patients undergoing immunosuppressive treatments for non-malignant disorders such as rheumatoid arthritis and inflammatory bowel disease.7 The American Association for the Study of Liver Diseases has recommended that people at high risk of hepatitis B virus infection should be tested for hepatitis B surface antigen before they are given chemotherapy or immunotherapy.2 However, reactivation can occur in people who are negative for hepatitis B surface antigen but positive for hepatitis B core antibody.8

Before antiviral agents were widely available, the only strategy for patients who tested positive for hepatitis B virus was to adjust their treatment, an approach that prejudices the chance of successfully treating the underlying disease. In our case, forgoing rituximab would have reduced the chance of developing viral reactivation, but the chance of five year survival would have been reduced by about 15%.9 A study that omitted prednisolone from standard treatment for non-Hodgkin’s lymphoma in patients positive for hepatitis B surface antigen found that overall survival at four years was lower—but jaundice occurred in only 4% of those not receiving steroids, compared with 28% of those who did.10

The antiviral drug lamivudine has been shown to be highly efficacious in preventing hepatitis B virus reactivation in these circumstances.11 Although there are no randomised control trials, there are case-control studies. For example, Li et al compared a group of 40 patients positive for hepatitis B surface antigen who received lamivudine before starting and for eight weeks after finishing chemotherapy and a group, comparable in most characteristics, of 116 historical controls.12 The lamivudine group had significantly less hepatitis (17.5% v 51.7%); hepatitis was of a lower grade in the lamivudine group (2.5% v 36.2% with grade 3 or 4 hepatitis), and death from hepatitis was less likely (0% v 5.2%). Data on mortality support the use of pre-emptive lamivudine, rather than waiting to see if hepatitis does occur: by the time of impending liver failure, lamivudine is less likely to be clinically efficacious.13

International guidelines recommend specific screening for hepatitis B core antibodies (or hepatitis B surface antigen) before chemotherapy or immunotherapy in people defined as high risk on the basis of geographic origin or behaviour patterns.2 The group at increased risk includes anyone with multiple sexual partners or a history of a sexually transmitted disease. Since the assays are readily available and relatively inexpensive, and the intervention largely prevents disastrous consequences, the safest strategy is for all patients undergoing immunosuppressive chemotherapy to be tested for hepatitis B core antibodies.14 15 Patients found to be positive for core antibodies should be tested for surface antigen and for hepatitis B virus DNA status. People positive for hepatitis B surface antigen are at risk of fulminant hepatitis and should receive prophylaxis with lamivudine or an equivalent antiviral agent throughout treatment and for at least six months afterwards.16 Patients positive for hepatitis B core antibodies but negative for hepatitis B surface antigen are at lower risk of reactivation, and a reasonable approach may be to await hepatitis B virus DNA status in such patients. If they are positive for hepatitis B virus DNA, prophylactic lamivudine is prudent; those who are negative can be observed, with liver function monitored regularly and hepatitis B virus DNA levels at least monthly, as the titre of hepatitis B virus DNA starts to rise several weeks before a rise in liver enzymes becomes evident.17 If there is a significant rise in either, antiviral therapy can be considered.

Managing the risk of hepatitis B virus reactivation in patients receiving chemotherapy or immunotherapy

  • In treatment of haematological and solid organ malignancy, hepatitis B recurrence can be fatal; use of immunosuppressive agents such as rituximab has also been associated with recurrence

    • Viral reactivation can occur during or after treatment, and can largely be avoided with appropriate antiviral therapy

  • All patients from high risk areas or with a history of high risk practices should be screened for hepatitis B before treatment

    • Since any patient with a history of multiple sexual partners or sexually transmitted disease is defined as high risk, where possible it is safest to screen ALL patients

  • Screening consists initially of determining hepatitis B core antibody status; patients positive for antibodies should be tested for hepatitis B surface antigen and levels of hepatitis B virus DNA

    • Patients positive for hepatitis B surface antigen or DNA should receive lamivudine (or an equivalent agent) for the duration of their treatment and for at least six months afterwards

    • Patients positive for hepatitis B core antibodies but negative for DNA should be monitored throughout and for at least six months after treatment by determining alanine transaminase and hepatitis B virus DNA levels

    • Patients negative for hepatitis B core antibodies, or positive with a low surface antibody titre, should be offered vaccination against hepatitis B

  • Hepatology consultation is advised for patients receiving prophylaxis or if interpretation of results is in doubt


Cite this as: BMJ 2008;337:a423


  • GMH, KPR, and MEDA had the idea of writing the paper, GMH and RD wrote the paper; KPR and MEDA reviewed the manuscript. All were involved in looking after the patient. GMH is guarantor.

  • Funding: None.

  • Competing interests: None declared.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract