T. cruzi Incidence Study in Blood Donors and its Implications for One-time Testing of Blood Donors Robert Duncan, Ph.D. DETTD, CBER, FDA Blood Products Advisory Committee August 2, 2011

2 Background-1 Trypanosoma cruzi: causative agent of Chagas’ disease Small protozoan parasite Chronic, asymptomatic infection Difficult or impossible to treat, severe symptoms late in about 30% of cases Endemic to portions of Mexico, Central America, and S. America, 10 million people infected worldwide with 300,000 in the U.S. Transmission: feces of an infected triatomine insect, congenital, organ transplant, transfusion, oral (breast milk, cane juice), via conjunctiva, laboratory accident Blood transfusion transmission is a recognized problem in endemic areas. An infected unit is estimated to have a 12-20% probability of causing infection in the recipient (WHO TRS 905, 2002)

3 Background-2 9 cases of transfusion transmission documented in US/Canada 5 reported cases of solid organ transplant transmission. Seroprevalence in the US donor population based on testing with licensed assays ranges from % with the higher rates in areas with large numbers of immigrants from Central and South America. Increasing rates of immigration raises concern about the potential for increased transmission.

4 Background-3 FDA licensed the Ortho T. cruzi ELISA Test System, Dec. 2006, the ABBOTT PRISM Chagas, Apr as donor screening tests for blood donors. From the end of January, 2007 to mid 2009 a number of blood centers, representing a large proportion of U.S. blood collections tested every donation using a licensed assay. Repeatedly reactive (RR) specimens are mostly retested with the unlicensed Radio Immune Precipitation Assay (RIPA) because no licensed supplemental test is available. 4

5 Background-4 All RR donors are indefinitely deferred (4,936 have been detected through 7/26/2011) RIPA positive donors are counseled and prior donations are traced for quarantine and in order to contact and test recipients. (1,443 were confirmed through 7/26/2011) An FDA draft guidance recommending universal testing of blood and tissue donors was released on March 26, Since the start of testing no transmissions have been reported due to a seronegative transfusion. 5

6 Background-5 April 2009 BPAC voted in favor of one-time testing for antibodies to T. cruzi in blood donors contingent on the outcome of studies to establish the rate of incidence of new T. cruzi infections. ARC and BSI volunteered to perform studies. June 6, 2009, ARC and BSI submitted a proposal for an Incidence study defined on the next slide Aug. 2009, ARC and Apr. 2010, BSI initiated one-time testing of all donors. Dec. 2010, the FDA published Final Guidance recommending one-time testing for T. cruzi that would be reevaluated based on the outcome of incidence studies.

7 ARC and BSI Proposed T. cruzi Incidence Study Aim: To attempt to determine the rate of newly acquired T. cruzi infections among blood donors who had previously tested negative for T. cruzi in the areas of the US where there is risk from travel to endemic areas or indigenous risk of infection.

8 ARC and BSI Proposed T. cruzi Incidence Study (cont.) a)Continue universal testing in 5 regions selected based on high prevalence b)Continue testing for 3 additional years to a total of 5 years from the initiation of universal testing in Feb c)Continue testing until 5 million person-years had been accumulated (the sum of the intervals from the first negative test to the last donation for all repeat donors in the study)

9 ARC and BSI Proposed T. cruzi Incidence Study (cont.) d)Continue testing until a mean observation period of 1.9 years was reached. Chosen as the extrapolation to 5 years of the 0.9 year mean observation period achieved in the first 2 years of testing. e)The ARC and BSI proposed that the study would conclude that the risk of T. cruzi incident infections was acceptable if the outcome was less than 1 case per million person-years and the 95% upper confidence limit of the estimated rate was less than or equal to 2.4 cases per million person-years.

10 ARC and BSI Proposed T. cruzi Incidence Study (cont.) f)The criteria for an incident case are: “For the purposes of this protocol, an incident case/sero- converter is defined as an ELISA-repeatedly reactive and RIPA-positive donor who had a prior ELISA nonreactive donation (S/CO 2.) while remaining RIPA positive. The donor may have positivity in PCR or HC in follow-up testing and/or may have identified risk factors.” FDA emphasizes that this definition was intended to favor true positives that are incident cases and to exclude chronically infected individuals who test reactive and individuals who inconsistently test reactive.

11 Total Repeat Reactive RIPA Positive RIPA Negative RIPA Ind RIPA Pending/NT 4,9361,4433, Los Angeles Central Coast Community Blood Centers of Florida 7/27/11 Three Travel Risk Sites

12 * * * * ARC Greater Ozarks (AR) ARC Southwest (OK/TX) * * * Seven published human vector-associated cases Triatome insects Reservoir mammals documented * Five BSI Regions (TX, AR, MS1, MS2, LA) Seven Indigenous Risk Sites In a study of T. cruzi infection among U.S. blood donors, 14 donors had two positive blood center tests and two positive CDC tests, all donors had resided in a state with documented vector or infected mammalian reservoirs and many of the donors had lived in rural areas. Cantey et al. Transfusion 2010.

13 Results of the Incidence Study reported by ARC and BSI a)All ARC and BSI regions over the initial period of universal testing combined with 3 ARC sites and one BSI center that continued universal testing under the Incidence Study tested a total of 4,222,285 repeat donors. b)Donor observation period data were not collected from The Community Blood Centers of Florida and therefore not included in the analysis. c)The study accumulated over 6 million person-years of observation. d)ARC/BSI incidence study showed 0 confirmed incident cases yielding a mean estimate of 0 ( , 95%CI) incident cases per million person-years. e)The mean observation period of a donor for travel risk was years and for indigenous risk was years.

14 Site Number of Donors Mean Observation Period (years) Incident cases per million p-y (95%Confidence Interval) Incidence Study (high risk areas) Travel Risk ARC Los Angeles (CA, 006), 54 months244, ( ) UBS Central Coast, (CA, 034), 48 months49, ( ) Subtotal 294, ( ) Indigenous Risk ARC Southwest (OK/TX; 049) 48 months59, ( ) ARC Greater Ozarks (AR; 055) 48 months74, ( ) Subtotal ARC133, ( ) BSI Regions (TX, MS1&2, LA, AR; 11, 16, 20, 26 & 31) 38 months 175, ( ) Subtotal ARC + BSI 309, ( ) Overall Risk (~65% of donations nationally) ARC, all Regions3,523, ( ) BSI, all Regions698, ( ) Total 4,222, ( ) Results of Incidence Study for each Site

15 ARC and BSI Interpretation of Results ARC and BSI concluded that the scope of the study was large enough to meet the study goals and there were no incident cases observed; thus the study was ended on January 31, 2011

16 Observation Periods of the Study Total Indigenous Travel Schematic of Observation Periods under universal testing All ARC Regions Feb 2007 Months (licensed testing) Jan 2011 All BSI Regions IND 294,089 Donors 309,576 4,222,285

17 Frequency Histogram of the Number of Donors by Observation Interval-All Regions 4,222,285 donors total 1,141,859 donors (27%) were observed for more than 2 years

18 Frequency Histogram of the Number of “Travel Risk” Donors by Observation Interval ARC Los Angeles Collection Center, IND+Licensed testing plus UBS Central Coast 294,089 donors total 105,020 donors (38%) were observed for more than 2 years

19 Frequency Histogram of the Number of “Indigenous Risk” Donors ARC Texas/Oklahoma/Arkansas Collection Centers, BSI TX, AR, MS1&2 and Louisiana. 309,576 donors total 103,679 donors (33.5%) were observed for more than 2 years

20 FDA Analysis to Estimate Maximum Potential Risk Out of a total of 17,568,120 donations from Feb 2007 to Aug 2009, the ARC and The Community Blood Centers of Florida identified a total of 18 donors with: –prior donations testing non-reactive for antibodies to T. cruzi. –repeatedly reactive index donation. –RIPA positive results at least once at index or follow up. –Follow-up testing showing inconsistent EIA and RIPA results. Out of 1,414,402 donations during 17 months in high-prevalence areas (the Incidence Study), 4 other “potential seroconverters” were identified who had a similar testing profile. None of these 22 donors is likely to be an incident case by the study criteria (to be described in detail in the ARC presentation). Some or all of these 22 may have a chronic T. cruzi infection that may not be detected by one-time testing.

21 FDA Analysis of T. cruzi Testing Results to Evaluate Potential Residual Risk to Recipients TestingTime FrameTotal Donations “Potential Seroconverters ” Maximum Potential Risk per Donation Range of Positive Donations per Year* from Donors with a Prior Negative Test All regions universal testing Feb 2007-Aug 2009 (31 months) 17, 568, /976,007 0 – 18 Incidence Study (high prevalenc e areas) Aug 2009-Jan 2011 (17 months) 1,414, /353, *The range of positive donations per year is based on million allogeneic donations in the U.S. per year, and 742,000 allogeneic donations per year in the high prevalence areas.

22 Summary ARC/BSI incidence study showed 0 confirmed incident cases among over 4,222,285 donors in more than 4 years who represented 6 million person-years of observation yielding a mean estimate of 0 ( ) incident cases per million person-years. The mean observation period of a donor for travel risk was years and for indigenous risk was years.

23 Summary (cont.) The maximum potential risk of a positive donation in the blood supply based on a worst case interpretation of all potential seroconverters being considered infected donors would be 1/976,007 for universal testing in all ARC and BSI regions and 1/353,601 for universal testing in the high prevalence areas of the U.S. The number of positive donations that would be estimated to occur in the blood supply per year would range from 0 to 18 in all regions and from 0 to 2 in the high prevalence areas.

24 Summary (cont.) The results of BSRI’s study of donor travel clearly documents that there is travel to T. cruzi endemic areas among repeat donors. The CDC study of T. cruzi infection in blood donors shows evidence for indigenous T. cruzi infection and seropositivity.

25 Questions for the Committee 1.Are the scientific data on the risk of incident infections among blood donors sufficient to conclude that a one- time negative test for antibodies to T. cruzi can qualify the donor for all future donations without further testing? 2.If not, are the scientific data on the risk of incident infections among blood donors sufficient to conclude that a one-time negative test for antibodies to T. cruzi can qualify the donor for a limited period of time? If so, what does the Committee recommend as a testing interval?