Loss of hepatitis A virus antibodies after bone marrow transplantation

Bone Marrow Transplantation (2006) 38, 37–40 & 2006 Nature Publishing Group All rights reserved 0268-3369/06 $30.00

www.nature.com/bmt

ORIGINAL ARTICLE

Loss of hepatitis A virus antibodies after bone marrow transplantation ER Godoi1, VAUF de Souza1, S Cakmak2, AF Machado1, LS Vilas Boas1 and CM Machado1 1

Virology Laboratory, Saõ Paulo Institute of Tropical Medicine, University of Saõ Paulo, Saõ Paulo, Brazil and 2Health Canada, HECS Branch, Ottawa, Ontario, Canada

Reimmunization guidelines have recommended the inactivated HAV vaccine for hematopoietic stem cell transplant (HSCT) recipients living in or traveling to areas where hepatitis A is endemic. As a shift from high to medium hepatitis A endemicity has been observed in several countries in Latin America, we conducted a retrospective study to evaluate the prevalence of hepatitis A pre-bone marrow transplant (BMT) and the loss of specific antibodies in consecutive stored serum samples from 77 BMT recipients followed up from 82 to 1530 days. The prevalence of HAV antibodies was 92.2% before BMT. As vaccine was not available in Brazil when the samples were taken, it was assumed that this prevalence reflects natural infection. Survival analysis showed that the probability of becoming seronegative was 4.5% (72.6%), 7.9% (73.4%), 10.1% (74.0%), 23.4% (79.6%) at 1, 2, 3 and 4 years after transplant, respectively. The loss of HAV antibodies was significantly associated with longer follow-up (P ¼ 0.0015), younger age (P ¼ 0.049) and acute graft-versus-host disease (P ¼ 0.035). As most reimmunization protocols start around day þ 365, in developing countries with similar HAV endemicity, BMT recipients should have serological screening before HAV vaccination and the inactivated vaccine should be advised to those seronegative. Bone Marrow Transplantation (2006) 38, 37–40. doi:10.1038/sj.bmt.1705391; published online 22 May 2006 Keywords: hepatitis A; seroprevalence; antibody loss; vaccination; epidemiology

Introduction Hepatitis A virus (HAV) generally causes mild and selflimited liver disease in immunocompetent hosts. However, few patients can develop fulminant hepatitis with fatality rates around 60%.1 As no case of HAV disease has been

Correspondence: Dr CM Machado, Virology Laboratory, Saõ Paulo Institute of Tropical Medicine, University of Saõ Paulo, Av Dr Eneas de C Aguiar, 470, Saõ Paulo, SP 05403-000, Brazil. E-mail: [email protected] Received 22 November 2005; revised 10 April 2006; accepted 11 April 2006; published online 22 May 2006

reported in hematopoietic stem cell transplant (HSCT) recipients, the morbidity or the risk of a fulminant course of hepatitis A in these patients is unknown. Seroepidemiological studies have demonstrated that the prevalence of HAV infection is decreasing in Brazil and the peak age of infection moving from early childhood to young adults. During this transitional phase of decreasing endemicity, HAV can circulate among a growing number of susceptible individuals and cause an outbreak.2,3 Post transplant reimmunization guidelines have indicated the inactivated HAV vaccine for susceptible HSCT recipients living in or traveling to areas where HAV infection is endemic.4 However, few data are available concerning the loss of specific HAV immunity over time5 and ultimately, the need of including inactivated hepatitis A vaccine in reimmunization protocols of developing countries. In the present study, we evaluate the probability of loss of HAV antibodies in bone marrow transplant (BMT) recipients living in the highly populous city of Saõ Paulo, Brazil, and the risk factors associated with the loss of HAV antibodies over time.

Methods Population The prevalence of pre-transplant HAV infection and the probability of loss of specific antibodies were evaluated in stored serum samples from 77 consecutive BMT recipients. The patients included in this study were transplanted from 1994 to 1997 and followed up at the BMT program of the University of Saõ Paulo Medical School. Inactivated HAV vaccine was not available in Brazil during this period; consequently, it was assumed that the detection of specific IgG antibodies reflects naturally acquired HAV infection. Sample selection Serum samples were routinely taken biweekly as part of the viral surveillance policy adopted in our Institution and stored at 201C. For this study, the serum samples were selected with a 3-month interval, approximately, from conditioning up to the end of serological follow-up or patient’s death. Three hundred and twenty samples were included, median five samples (2–11) per patient.

Loss of HAV antibodies after BMT ER Godoi et al

38

HAV serology Hepatitis A virus antibodies were detected using a qualitative enzyme immunoassay (DiaSorin, Saluggia, Italy) according to the manufacturer’s recommendations. This assay has been shown to have good sensitivity, with a lower limit of detection of 20 mIU/ml in comparison with other tests that have lower limit of detection of 100 mIU/ml.6 Results were expressed as positive or negative. Detection of IgM HAV antibodies was performed in pre-transplant seronegative patients who showed transient IgG seroconversion during follow-up to differentiate acute HAV infection from passively acquired antibodies through blood transfusion.

Statistical analysis The cumulative probability of loss of specific HAV antibodies was evaluated by cumulative incidence analysis by accounting for competing risks. The event is considered the loss of HAV antibodies and remaining events (death and relapse) as censoring.7 Additional analyses were performed using (simple and multiple) logistic regressions. The variables evaluated included age, gender, type of BMT, underlying disease, conditioning regimen, acute and chronic graft-versus-host disease (GVHD) and time to the event or follow-up.

Table 1

Patient characteristics (N ¼ 71)

Variable

Number (%)

Median age (range)

26 (3–58) years old

Type of BMT Allogeneic related Allogeneic unrelated Autologous

64 (90.1) 01 (1.4) 06 (8.4)

Conditioninga BUCY BUMEL Other

20 (30.3) 38 (57.6) 08 (12.1)

Underlying diseases CML SAA AML NHL MM Other

27 23 09 04 03 05

Acute GVHD ograde II Xgrade II

29 (45.3) 35 (54.7)

Chronic GVHDb No Limited Extensive

23 (36.5) 05 (7.9) 35 (55.5)

(38) (32.4) (12.7) (5.6) (4.2) (7)

a

Information not available for five patients. Allogeneic patients surviving day +100. Abbreviations: AML ¼ acute myeloid leukemia; BMT ¼ bone marrow transplant; BUMEL ¼ busulfan+melphalan; BUCY ¼ busulfan+cyclosphosphamide; CML ¼ chronic myeloid leukemia; MM ¼ multiple myeloma; NHL ¼ non-Hodgkin’s lymphoma; GVHD ¼ graft-versus-host disease; SAA ¼ severe aplastic anemia.

Results Seventy-one patients (92.2%) had HAV antibodies before BMT. Median patients’ age was 26 (3–58) years old and median follow-up was 644 days (ranging from 82 to 1530 days). Patients’ characteristics are shown in Table 1. Eight of the 71 seropositive patients (11.2%) lost specific antibodies at a median of 467 (189–1124) days after transplantation. Survival analysis showed that the probability of becoming seronegative was 4.5% (72.6%), 7.9% (73.4%), 10.1% (74.0%), 23.4% (79.6%) at 1, 2, 3 and 4 years after transplant, respectively (Figure 1). Table 2 shows the cumulative incidence rate along with 95% confidence intervals. Using the survival analysis technique and adjusting for competing risks, the loss of HAV antibodies was significantly associated with longer follow-up (P ¼ 0.0015), younger age (P ¼ 0.049) and acute GVHD (P ¼ 0.035) (Table 2). Four of the six seronegative patients showed a transient ‘seroconversion’ during follow-up. The patients were asymptomatic, with normal levels of liver transaminases and IgM antibodies persistently negative during those episodes, suggesting passively transferred antibodies through blood transfusion rather than natural HAV infection. Among the eight previously seropositive patients who lost HAV antibodies during follow-up, two patients also showed a transient ‘seroconversion’ coinciding with the administration of intravenous immunoglobulin and returned to the seronegative status in samples taken after 3–5 months. Bone Marrow Transplantation

CI

b

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

0.5

1

1.5 2.5 2 3 Years after BMT

3.5

4

4.5

Figure 1 Cumulative incidence (CI) rate of hepatitis A virus (HAV) antibody loss in bone marrow transplant (BMT) recipients.

Discussion The seroprevalence of HAV antibodies was high (92.2%) in patients undergoing BMT from 1994 to 1997. As HAV vaccine was not available in Brazil during this period and intravenous immunoglobulin was never used before transplantation, seroprevalence probably indicates naturally acquired HAV infection. Hepatitis A virus infection was highly prevalent in Brazil up to the eighties. However, in some cities median


39 Table 2 Cumulative incidence rate of loss of hepatitis A antibodies after BMT according to prognostic factors Factors

N

Cumulative incidence7s.e. (%) Year 1

Year 3

*P-value

Age p20 years 420 years

26 45

9.076.2 2.372.3

27.5711.8 2.372.3

0.049

Gender Female Male

30 41

7.174.9 2.772.7

16.778.0 5.373.7

0.4587

BMT Allogeneic Autologous

65 6

4.972.8 0

8.573.7 25.0726.4

0.7083

Underlying disease Acute leukemia Other

11 60

0 5.273.0

35.4722.6 7.173.5

0.2688

Conditioning BUMEL BUCY Other

38 20 8

2.772.7 5.675.6 12.5712.6

6.374.5 11.277.8 25.0716.7

0.2942

Time since BMT p1 year 41 year

24 47

22.7713.8 0

No data 7.474.2

0.0015

Acute GVHD No Yes

16 48

13.479.2 2.172.1

13.479.2 7.174.0

0.0355

Chronic GVHD No Yes

24 40

10.377.2 2.572.5

15.578.5 5.373.7

0.4885

*Gray test. Abbreviations: BMT ¼ bone marrow transplant; BUMEL ¼ busulfan+ melphalan; BUCY ¼ busulfan+cyclosphosphamide; GVHD ¼ graft-versushost disease.

prevalence rates have dropped from 65.6% in children younger than 7 years old to 32.1% in adolescents and young adults from 1978 to 1995.2 Recent epidemiological studies have shown a shift from high to medium endemicity in Brazil as well as in other countries of Latin America increasing the risk for outbreaks.3 As BMT recipients can lose their immunity against several diseases such as polio,8 tetanus,9 diphtheria,10 measles,11 etc, the loss of HAV antibodies must be well known in these countries. Patients who lose HAV antibodies may probably be at risk of acquiring hepatitis A. Actually, the absolute lower limit of antibody required to prevent HAV infection has not been defined. In vitro studies using cell culture-derived virus indicate that antibody levels lower than 20 mIU/ml can be neutralizing. Experimental studies in chimpanzees showed that passively transferred antibodies in concentration lower than 10 mIU/ml did not protect against infection but did prevent clinical hepatitis and virus shedding.6 Also, the persistence of B memory cells could promote a rapid increase of HAV antibodies and protect against disease in the case of natural exposure to HAV, as observed in other viral infections such as measles.11,12

Using a sensitive assay for HAV antibody detection (lower limit detection of 20 mIU/ml), we observed that the probability of loss of specific antibodies was low 1 year after transplant (4.5% (72.6%)), but increased over time up to 23.4% (79.6%) 4 years after BMT, in patients with previous HAV infection. A similar study in Argentina, including a smaller number of patients, have shown lower prevalence rates (68%) but increased loss of HAV antibodies (14%) up to one year after BMT. It is important to highlight that the mentioned study was cross-sectional and eight of the 36 patients were tested before the sixth month of transplantation, when loss of HAV antibodies is unlikely. The authors commented that higher rates of antibody loss would probably be observed if all patients had been tested after the first year of BMT.5 On the other hand, one could argue that the authors could be assigning as ‘negative’, seropositive patients but with antibodies titers lower than the limit of detection of the test, depending on the sensitivity of the assay. In this case, the loss of HAV antibodies would be overestimated in the mentioned study. The lower loss of HAV antibodies observed in the present study could also be explained by the possibility of natural booster during follow-up, as the virus is still circulating with great activity in our country. Among the variables tested, we could demonstrate that younger age; acute GVHD and longer time since BMT were predictors of HAV antibody loss. Dignani et al.5 also did not find any correlation between loss of HAV immunity and age, gender, underlying disease, time of serology after BMT, type of BMT or number of CD34 þ cells infused. Another difference between our study and the study of Dignani et al.5 is concerned to the source of hematopoietic stem cells (HSC); bone marrow versus peripheral blood, respectively. However, the role of the source of HSC on the loss antibodies to HAV or to other viral infections remains to be demonstrated. It is possible that in areas where the prevalence rates and the intensity of virus circulation are decreasing, the loss of HAV antibodies may be accelerated due to the lack of natural boosters, creating the ideal condition for the emergence of outbreaks. These areas would benefit of hepatitis A vaccine inclusion in post transplant reimmunization protocols, even though no data are available on the efficacy or immunogenicity of HAV vaccine in BMT recipients. Based on the data of the present study, we recommend that BMT recipients should perform serological screening around day þ 365, in developing countries with similar HAV endemicity. The HAV inactivated vaccine should be advised to those seronegative. The data from the present study are concerned to patients with naturally acquired HAV infection and should not be extrapolated to BMT recipients who had received HAV vaccine before transplantation. As vaccineinduced immunity is generally shorter than that acquired by natural infection, similar studies are necessary to evaluate the loss of HAV antibodies in vaccinated BMT recipients. Bone Marrow Transplantation


40

Acknowledgements We thank Drs FL Dulley, MCA Macedo, RL Silva and RS Saboya, from the BMT Program of University of Saõ Paulo Medical School, for providing excellent patient care and without whose support this study would not have been possible.

6

7

References 1 Debray D, Cullufi P, Devictor D, Fabre M, Bernard O. Liver failure in children with hepatitis A. Hepatology 1997; 26: 1018– 1022. 2 Vitral CL, Yoshida CF, Lemos ER, Teixeira CS, Gaspar AM. Age-specific prevalence of antibodies to hepatitis A in children and adolescents from Rio de Janeiro, Brazil, 1978 and 1995. Relationship of prevalence to environmental factors. Mem Inst Oswaldo Cruz 1998; 93: 1–5. 3 Tanaka J. Hepatitis A shifting epidemiology in Latin America. Vaccine 2000; 18 (Suppl 1): S57–S60. 4 Ljungman P, Engelhard D, de la Camara R, Einsele H, Locasciulli A, Martino R et al. Infectious Diseases Working Party of the European Group for Blood and Marrow Transplantation. Vaccination of stem cell transplant recipients: recommendations of the Infectious Diseases Working Party of the EBMT. Bone Marrow Transplant 2005; 35: 737–746. 5 Dignani MC, Miceli MH, Rosa CM, Gatica J, Martı´ nezRoloń J, Pizzolato M. Loss of hepatitis A virus (HAV)

Bone Marrow Transplantation

8

9

10

11

12

antibodies after peripheral stem cell transplantation (PSCT). Bone Marrow Transplant 2003; 31: 809–812. Glode MP, Evans GS, Fleming DW, Guerra FA, Le CT, Offit PA. Prevention of hepatitis A through active or passive immunization: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 1999; 48: 1–37. Gray RJ. A class of k-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988; 16: 1141–1154. Ljungman P, Duraj V, Magnius L. Response to immunization against polio after allogeneic marrow transplantation. Bone Marrow Transplant 1991; 7: 89–93. Hammarstrom V, Pauksen K, Bjorkstrand B, Simonsson B, Oberg G, Ljungman P. Tetanus immunity in autologous bone marrow and blood stem cell transplant recipients. Bone Marrow Transplant 1998; 22: 67–71. Parkkali T, Ruutu T, Stenvik M, Kuronen T, Kayhty H, Hovi T et al. Loss of protective immunity to polio, diphtheria and haemophilus influenzae type b after allogeneic bone marrow transplantation. APMIS 1996; 104: 383–388. Machado CM, Gonçalves FB, Pannuti CS, Dulley FL, Souza VAUF. Measles in bone marrow transplant recipients during an outbreak in Saõ Paulo, Brazil. Blood 2002; 99: 83–87. Pauksen K, Linde A, Ljungman P, Bolme P, Lonnerholm G, Oberg G et al. Specific T and B cell immunity to measles after allogeneic and autologous bone marrow transplantation. Bone Marrow Transplant 1995; 16: 807–813.