Blood Products Advisory Committee

94th Meeting, April 1, 2009

Rockville, MD

 

Issue Summary

                                                

Topic I.A.:  Blood Donor Screening for Hepatitis B Virus Infection by Nucleic Acid Testing 

            

Issue:  FDA seeks advice and input from the Committee on issues related to blood donor screening for hepatitis B virus (HBV) DNA by nucleic acid testing (NAT) to prevent transfusion transmission of HBV.  

 

Introduction:

 

At the present time, to reduce the risk of transmission of hepatits B by blood, all blood for transfusion in the U.S. is tested for hepatitis B surface antigen (HBsAg) and antibody to hepatitis B core antigen (anti-HBc).  The residual risk of post-transfusion hepatitis B infection is estimated to be about 1:205,000 to 1:269,000 per transfusion. 1, 2   There are currently three FDA-licensed HBV NAT assays for screening blood available in the U.S.:

 

1.      April 2005: COBAS AmpliScreen HBV Test (Roche Molecular Systems, Inc.) licensed for donor screening using pools of up to 24 samples. 

                                 

2.   August 2008: Procleix Ultrio Assay (Gen-Probe, Inc.) licensed for donor    screening using pools of up to 16 samples.  The Tigris automated instrument, capable of handling smaller pools of samples and single samples, i.e., individual donation (ID) NAT, was licensed (May 2007) for use with Ultrio. 

                     

3.   December 2008: COBAS TaqScreen MPX Test for use on cobas s 201 system   

      (Roche Molecular Systems, Inc.) licensed for donor screening using pools of up to 6 samples.   

 

(See Appendix I for additional information on the analytical sensitivity of these assays in the ID NAT and minipool formats.)

 

Although these assays have approved indications for donor screening, and although some U.S. blood establishments are using them, HBV NAT has not been recommended by the FDA for routine donor screening based on a previous assessment of a limited added value to public health.  We wish to reconsider the scientific value of routine donor screening by HBV NAT based on a) recent approval of “multiplex” NAT for HIV, HCV and HBV; b) newer data on the “yield” of DNA positive, but HBsAg negative/seronegative donations in routine screening; and c) the implication of HBV vaccine “breakthrough” infections in donors. 

 

 

Background:

 

Depending on the relative sensitivities of HBsAg and HBV NAT assays used, HBV DNA can be detected 2 to 5 weeks after infection, and up to 40 days (mean 6 to 15 days) before HBsAg.3 HBV DNA levels rise slowly and are present at relatively low levels during the seronegative/HBsAg-negative period of early infection.  HBV DNA can be detected along with HBsAg and anti-HBc in chronic hepatitis B infections, and rarely in recovered infections that are negative for HBsAg and positive for anti-HBs and anti-HBc. 

 

As blood for transfusion in the U.S. is also tested for anti-HBc, which develops a few days after HBsAg, and as anti-HBc usually remains detectable life-long, irrespective of whether the individual recovers from acute hepatitis B or whether chronic HBV infection develops, HBV NAT’s potential utility in further reducing risk of hepatitis B transmission by blood transfusion is restricted solely to the early HBsAg-negative phase of infection.  The question of HBV NAT’s possible added value in blood safety revolves around the number of HBV DNA positive, HBsAg negative units a particular NAT assay detects compared to a particular HBsAg assay.  This number—referred to as “yield” —depends on the relative sensitivities of the NAT and HBsAg assays used.  Yield data can be arrived at in two different ways:       

 

1)      The actual numbers of HBV DNA positive, HBsAg and anti-HBc negative units that HBV NATs have detected compared to HBsAg assays in studies at blood establishments, including data from clinical trials prior to licensure and any additional data gathered during donor screening post-licensure. 

 

2)      Estimated yields (“modeling”) that use formulae for calculations of yield. 2, 4  These estimations include use of, a) incidence rates of hepatitis B in blood donors and, b) relative window period reductions accomplished by HBV NAT compared to HBsAg tests when testing plasma from serial bleeds from seroconversion panels from individuals who became infected with hepatitis B.  

 

As mentioned above, despite approval of donor screening assays, HBV NAT has not been recommended by the FDA for routine use. 

 

Following licensure of Roche’s COBAS AmpliScreen HBV NAT in April 2005, FDA stated that it “regards use of this test for donor screening as voluntary because the estimated individual and public health benefits of adding this test to available screening tests are thought to be very limited.”  (See Frequently Asked Questions Regarding Implementation of the Roche Molecular Systems COBAS AmpliScreen HBV Test, FDA website, April 21, 2005 at http://www.fda.gov/cber/products/hbvroc042105qa.htm.)  This decision took into consideration a recommendation from the DHHS Secretary’s Advisory Committee on Blood Safety and Availability on August 27, 2004, which considered societal issues such as cost-effectiveness, feasibility, and overall public health benefit in addition to scientific data on detection of HBV in donors. 

 

The reasons for not recommending universal donor HBV NAT screening at that time were as follows:

 

1)  It was estimated that using the 24-sample pool NAT, in addition to the then current HBsAg and anti-HBc testing, approximately an additional 50 infectious donations/year (potentially contaminating an estimated 85 blood components) would be interdicted.  This number compared with approximately 20,000 positive Whole Blood donations detected annually by screening donors for HBsAg and anti-HBc at that time.  Only a small proportion of such recipients will develop acute or chronic HBV disease with morbid sequelae (estimated 0.16 Quality Life Years saved per infection prevented.)

 

2)  Single sample individual donor (ID) NAT or small-pool NAT was considered infeasible because instrumentation and software was not available for high throughput and it was not practical logistically. 

                                                

3)  Increasing vaccination of donors and recipients was thought to play a role in protecting them against hepatitis B, thereby reducing transmission of HBV by transfusion.

                                                                                                                           

Discussion:

 

We are bringing the topic of routine donor screening by HBV NAT to the Committee for discussion at this time for the following reasons:

 

1)  There are now three FDA-licensed HBV NAT assays with indications for blood screening.  Two of the licensed HBV NAT assays, Gen-Probe’s multiplex Procleix Ultrio and Roche’s multiplex MPX, both licensed in the past year, make it possible to simultaneously test for HIV, HCV, and HBV in a single sample.

 

2)  With the recent advance in technology and increased automation enabling the performance of NAT with small pools (and perhaps individual samples), more sensitive HBV NAT testing of blood donations is now possible, thus potentially increasing the number of window period HBV DNA positive/HBsAg negative units detected, and hence increasing the safety of blood and thereby improving public health.

 

[As shown in Appendix I, the analytical sensitivity of the Roche MPX test in the 6 sample minipool format (22.8 IU/mL) is similar to the sensitivity of the Roche COBAS AmpliScreen HBV NAT in the ID NAT format (15.99 IU/mL).  The analytical sensitivity of the Gen-Probe Procleix Ultrio assay in a minipool of 8 samples is estimated to be 83.2 IU/mL similar to the analytical sensitivity of the Roche COBAS AmpliScreen HBV NAT on minipools of 24 samples (105.84 IU/mL)]. 

 

3)  There is now more information available on the role of vaccination of donors and recipients against hepatitis B infection, which seems to indicate that protection against hepatitis B infection for the long term is not absolute, i.e. “breakthrough” infections can occur.  However, the infectivity of units obtained from hepatitis B vaccinated donors with “breakthrough” hepatits B infections is unknown at the present time.  Studies using animal models to assess infectious potential of units from hepatitis B vaccinated donors with “breakthrough” hepatits B infections are being considered.     

 

Hepatitis B Vaccination and “Breakthrough” Infections

In the past the two decades there has been widespread introduction of vaccination against HBV in the U.S., with overall higher rates in blood donors than in the general population, and very high rates in young donors.  (According to a preliminary analysis 74% of 18-22 year old donors are vaccinated, compared to 21% of 58-62 year old donors, using the National Health Interview Survey, 2006.5,6,7 Additional information will be provided at the Advisory Committee meeting.)  Vaccination with hepatitis B vaccine (recombinant antigenic HBsAg) elicits neutralizing/protective anti-HBs (not anti-HBc).  However, it appears that in some cases a vaccinated donor, who is exposed to HBV, can still become infected despite development of anti-HBs, and can undergo a “breakthrough” infection characterized by HBV NAT positivity with low viral load and, evidently, no symptoms, but sometimes with subsequent development of detectable HBsAg and anti-HBc.  It is not known at the present time whether the virus is intact in the circulation, is infectious, and whether it can cause chronic infection in a new host.   

 

The question of infectivity of circulating virus in vaccine “breakthrough” infections is relevant to estimates of the benefit of screening blood donors by HBV NAT.  This is because it has been observed that about half of the “yields” in the HBV NAT clinical trials occurred in previously hepatitis B-vaccinated blood donors with “breakthrough” hepatitis B infections.  As described above, they are HBsAg negative, anti-HBc negative, HBV NAT positive and anti-HBs positive.  Some of these donors subsequently develop detectable HBsAg and/or anti-HBc indicative of active infection.  

 

This finding raises two issues:

 

First, the finding may affect estimates of NAT yield.   Window period yield estimations of HBV NAT compared to HBsAg testing, using older donor incidence rates for hepatitis B and window period reduction data, from non-vaccinated donors with well-known disease course, are not applicable to these breakthrough infections.  The dynamics of breakthrough infections in vaccinees seem different from those infections in non-vaccinated individuals. 

 

Second, much of the available literature indicates that HBV NAT positive/anti-HBs positive/HBsAg negative blood, irrespective of anti-HBc test results, does not transmit hepatitis B.8-14 However, it is FDA’s opinion that there can be no assumption of non-infectivity of units from donors with breakthrough infections when transfused into recipients, and no assumption of lack of morbidity and mortality in recipients, especially when many recipients have compromised immune systems.  This is also true for children and neonates, who have a higher tendency for developing hepatitis B chronicity than adults: according to CDC about 1% of adults without other preexisting conditions are estimated to get chronic hepatitis B if infected, but 2 to 10% of children > 5 years of age get chronic hepatitis B, and 30 to 90% of children < 5 years of age get chronic hepatitis if infected. 3

 

        

Utility (Yield) of Donor Screening by HBV NAT

At the Committee meeting data regarding the utility of HBV NAT compared to HBsAg will be presented.  This will be presented in the form of “yield,” as detailed above, i.e., the number of HBV DNA positive, HBsAg negative/anti-HBc negative units a particular NAT detects compared to a particular HBsAg assay.  As mentioned above, different yields depend on the relative sensitivities of the different NAT formats and HBsAg assays being compared.  Estimates of residual risks of post-transfusion hepatitis B will also be given for the different yields.  (A summary of interim data on yield of HBV NAT available to the FDA at this time is provided in Appendix II.) 

 

Yield data will be presented in the two different ways:

 

1)  Actual yields obtained in recent U.S studies, including data from clinical trials prior to licensure and any data available at the time of the meeting from blood donor testing post-licensure.  There may be an undercount of yield data, because many deferred donors do not return for follow-up NAT and serology testing that is needed to ensure that an initial HBV NAT reactive result is not a false positive, and that the donor is actually infected.     

            

2)  Estimated yields (“modeling”), which incorporate into the formulae for calculations of yield recently established donor incidence rates of hepatitis B and relative window period reductions of NAT compared to HBsAg testing, and which take into account vaccination

effects.  Data regarding donor vaccination rates, broken down by age, will also be discussed.  The methodologies used in estimating yields and donor vaccination rates will

be presented.  It is hoped that this will result in more accurate estimations of yield, and that it will be possible to calculate the proportions of yield from non-vaccinated and vaccinated donors.

 

Proposals for studies using animal models to assess the infectious potential of units from hepatitis B vaccinated donors with “breakthrough” hepatits B infections will also be discussed.     

                                                                                                                                 

 

Questions for the Committee:

 

1.      Does the Committee agree with FDA that units from donors with apparent vaccine breakthrough HBV infections (HBV NAT positive and anti-HBs positive) should be presumed infectious pending further studies?

 

 

2.      Please comment on the value and design of candidate studies using animal models to assess the infectious potential of units from hepatitis B vaccinated donors with “breakthrough” hepatitis B infections.

 

3.   Considering the estimated yield of HBV infected window period donations, and the

      answer to question 1, please comment on the benefit of routine screening of blood    

      donors by HBV NAT if testing were performed using available licensed tests on

      minipools, assuming a sensitivity of at least 100 IU/mL for individual samples.

 

References

 

1.       Dodd, R, et al.  Current prevalence and incidence of infectious disease markers and 

       estimated window-period risk in the American Red Cross blood donor population.   

       Transfusion 2002; 42: 975-979. 

 

2.       Busch, MP. Transfusion-transmitted viral infections: building bridges to transfusion   

       medicine to reduce risks and understand epidemiology and pathogenesis.”  

      Transfusion 2006; 46: 1624-1640.

 

3.    Alter, MJ.  Centers for Disease Control and Prevention.  Epidemiology of HBV   

       Infection and Prevention Programs.  Presentation to the Advisory Committee on  

       Blood Safety and Availability, August 27, 2004.

 

4.    Kleinman, SH, Busch MP.  Assessing the impact of HBV NAT on window period  

       reduction and residual risk.  J. of Clinical Virology 2006; 36 Suppl. 1: S23-S29

 

5.   National Center for Health Statistics and Centers for Disease Control and Prevention.     

     National Health Interview Survey, 2006 (machine readable data file and  

     documentation).  2007. Hyattsville, Maryland.

 

6.   Royston, P. and W. Sauerbrei. "Multivariable modeling with cubic regression splines: 

      A principled approach." The Stata Journal 7.1 (2007): 45-70.7. 

 

7.   StataCorp. Stata Statistical Software: Release 10.  2007. College Station, TX, 

      StataCorp LP.

    

8.    Hollinger, FB. Hepatitis B infection and transfusion medicine: science and the

       occult. Transfusion 2008; 48:1001-1026.

 

9.     Dreier, J, et al. Low-level viremia of hepatitis B virus in an anti-HBc- and anti-HBs-

       positive blood donor.  Transfusion Medicine 2004;14;97-103.

 

10.   Prince AM, et al. Infectivity of blood from PCR-positive, HBsAg-negative, anti-

       HBs-positive cases of resolved hepatitis B infection. Transfusion 2001;41:329-32

 

11.   Gerlich, WH. Breakthrough of hepatitis Bus escape mutants after vaccination and   

         virus reactivation. J. Clin. Virol. 2006;36:S18-S22.      

 

 12.   Mosley et al.  Donor screening for antibody to hepatitis B core antigen and hepatitis  

         B virus infection in transfusion recipients.  Transfusion 1975;35:5-12.

 

13.    Hui, CK et al. Occult hepatitis B virus infection in hematopoietic stem cell donors 

         in a hepatitis B virus endemic area.  J. of Hepatology 2005;42:813-819.

 

14.    Satake, et al.  Infectivity of blood components with low hepatitis B virus DNA  

         levels identified in a lookback program.  Transfusion 2007;47: 1197-205.

                                                         

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

     

 

 

 

 

Appendix I

       

 

The three currently licensed HBV NAT assays for screening blood, with lower limits of detection in International Units (IU)/mL, for individual donation (ID) and minipool HBV NAT: 

 

1.   April 2005: COBAS AmpliScreen HBV Test (Roche Molecular Systems, Inc.) licensed for donor screening using pools of up to 24 samples. 

 

ID-NAT (standard prep. procedure used for ID testing)                                  15.99 IU/mL

Pools of 24 samples (multiprep. procedure used for pooled sample testing) 105.84 IU/mL   

     

2.  August 2008:  Procleix Ultrio Assay (Gen-Probe, Inc.) licensed for donor screening using pools of up to 16 samples.  The Tigris automated instrument, capable of handling smaller pools and ID NAT, was licensed (May 2007) for use with Ultrio. 

 

         ID-NAT on Tigris                                                            10.40 IU/mL

         Pools of 8 samples on Tigris                                             83.20 IU/mL

   Pools of 16 samples on Tigris                                         166.40 IU/mL

 

3.  December 2008:  COBAS TaqScreen MPX Test (Roche Molecular Systems, Inc.) licensed for donor screening using pools of up to 6 samples. 

 

          ID-NAT                                                                       3.80  IU/mL 

    Pools of 6 samples                                                      22.80 IU/mL

                     

                                     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

                                                    Appendix II

 

The following table contains data available to the FDA at this time on yield of HBV NAT obtained from donor screening.  A “yield” case is defined as HBV NAT positive/HBsAg negative/anti-HBc negative.  (Point estimates of the yield rate are provided to permit comparisons, but it should be understood that there is a large uncertainty in each ratio due to a small numerator.)

      

                                                                                    Yield               Yield Rate

Assay                                      Format               (No.+/No. Tested)     per donation*

 

AmpliScreen[1]                   24-sample pools                2/704,902         1:352,451 donations

MPX Test[2]                          6-sample pools                 1/62,652          1:62,652   donations

Procleix Ultrio Assay[3]      16-sample pools                 5/1,992,160     1:398,432 donations

Procleix Ultrio Assay[4]         8-sample pools                2/977,264        1:488,632 donations

Procleix Ultrio Assay3                  ID                           1/572,569       1:572,569 donations

 

* Point estimate

 

 Updated information on yield will be presented at the Blood Products Advisory Meeting.

 



[1] Kleinman SH, et al. Hepatitis B virus (HBV) DNA screening of blood donations in minipools with the COBAS AmpliScreen HBV test. Transfusion 2005;45:1247-57.

[2] Linauts S, Saldanha J, Strong DM. PRISM hepatitis B surface antigen detection of hepatits B virus minipool nucleic acid testing yield samples Transfusion 2008;48:1376-1382.

[3] Yield data obtained from donor screening obtained from the Procleix Ultrio Assay was provided by Susan Stramer, Ph.D., American Red Cross, in a personal communication to the FDA.  

[4] Yield data obtained from donor screening obtained from the Procleix Ultrio Assay was provided by Patricia Bakke, MT(ASCP)SBB, Gen-Probe, Inc., in a personal communication to the FDA.