Issue Summary

 

BLOOD PRODUCTS ADVISORY COMMITTEE

92nd Meeting

September 11, 2008

 

Topic III: Options for Blood Donor Screening and Reentry for Malaria

 

Issue

 

FDA is seeking advice from the Blood Products Advisory Committee on options for blood donor screening for the presence of malarial antibodies as evidence for malaria exposure and on a possible mechanism to allow more rapid reentry of donors who traveled to Mexico.

 

Background

 

Transfusion-transmitted-malaria (TTM) is a rare but serious concern in transfusion medicine. TTM can be caused by any of the four Plasmodium species (P. falciparum, P. malariae, P. ovale, and P. vivax). Most cases of TTM in the United States are caused by transfusion of blood donated by infected individuals who visited or were prior residents of malaria-endemic countries [1]. According to surveillance studies conducted by the CDC, TTM occurs at a rate of about 0.25 cases per one million blood units collected, a rate that remained stable for more than two decades since 1974 [1, 2]. However, in more recent years (2000 through 2007), possibly as a result of more rigorous implementation of donor deferral policies, the rate of TTM has declined to approximately 0.027 cases per million blood units collected. There is no licensed test to screen donor blood for malaria infections in the U.S. In the absence of a screening test, blood establishments have deferred donors based on a history of possible exposure to malaria or previous malaria infection. A significant negative impact of deferral-based policies to identify malaria-risk donors is the loss of a sizeable numbers of donors who were actually never exposed to malaria; according to some estimates between (1.0 - 3.0%) of all presenting donors are deferred each year based on their history of travel or residence in malaria endemic areas [1, 3, 4]. The number of donors who self-defer for risk of malaria exposure is not known and, without conducting extensive donor surveys, total donor loss associated with malaria risk is difficult to estimate.

 

Malaria is transmitted in more than 100 countries with approximately 500 million cases and more than one million deaths annually [5, 6]. Inside the U.S., malaria currently is seen only rarely as a natural infection transmitted by mosquitoes [7, 8]. However, approximately 1,500 cases of clinical malaria are reported here annually [9], almost all due to infections acquired outside the country. Thus, it is evident that the major risk of malaria in prospective U.S. blood donors must be from possible exposure during travel or residence in countries where malaria is transmitted. Data obtained from the World Tourism Organization suggests that between 2000 – 2004 there were an average of 34 million arrivals per year of U.S. residents in countries where malaria transmission occurs. It is noteworthy that, outside sub-Saharan Africa, risk of malaria exposure may be limited to certain areas of a country. It is also important to remember that boundaries of malaria transmission are not rigid and vary from year to year dependent on seasonal and other external factors. A list of malaria-endemic countries and malaria-endemic areas is maintained by CDC at http://wwwn.cdc.gov/travel/yellowBookCh5-MalariaYellowFeverTable.aspx.  

 

            While tests for anti-malarial antibodies are not generally useful in the diagnosis of acute malarial infection, presence of malaria-specific antibodies is a reliable indicator of previous exposure with Plasmodium parasites. Accordingly, some European countries, and Australia and New Zealand permit blood establishments to test otherwise deferred at-risk donors for the presence of anti-malarial antibodies. In those countries, an otherwise deferred at-risk donor is allowed to reenter the donor pool after a shortened deferral period (4 – 6 months), if found negative for antibodies to P. falciparum and P. vivax parasites [10, 11]. It is important to keep in mind that, although P. falciparum and P. vivax are the most common cause of malaria globally, parasites of all four Plasmodium species have caused TTM in the U.S. [1]. Table 1 shows the total number of TTM cases reported in the U.S. since 1963 and the causative Plasmodium species identified in these incidents [1, 12, and 13].  Although, globally only approximately 5% of all malaria infections are caused by P. malariae, this Plasmodium species is responsible for 28.5% of TTM in the U.S. The disproportionately high prevalence of P. malariae as the causative agent of TTM can be attributed to the chronic, sub-patent infection caused by this parasite, and asymptomatic infectivity may be lifelong [14].  

 

Table 1.

Species

Number of cases of transfusion-transmitted malaria in the United States from 1963 through 2007 categorized by Plasmodium species found to be causative agents [references 1, 12, 13]

Plasmodium falciparum

                               35 (38.5%)

 

Plasmodium vivax

 

                               25 (27.5%)

 

Plasmodium malariae

 

                               26 (28.5%)

 

Plasmodium ovale

 

                               5 (5.5%)     

 

Sub-total

                               91

Mixed species

 

                                 3

 

Species unknown

 

                                 2

   Total

                                 96         


The value of antibody tests in identifying malaria infections in prospective U.S. blood donors is evident from data showing that 58 of 59 (98%) donors implicated to cause TTM were positive in malaria serology tests [1].  Considering that, apart from P. falciparum and P. vivax, a relatively high number of cases of TTM are also caused by P. malariae and to a lesser extent by P. ovale (Table 1), antibody-based tests used to identify malaria-at-risk donors ideally should detect antibodies to all four Plasmodium species.

 

In 1999, the Blood Product Advisory Committee (BPAC) discussed the potential benefits of screening for anti-malarial antibodies as a method to rule out possible parasite exposure and thus to reduce malaria-related donor deferrals. However, no specific recommendations were offered at that meeting. On July 12, 2006, FDA convened a scientific workshop on “Testing for Malaria Infections in Blood Donors” to seek public discussion of scientific developments that might support donor testing for malaria infections as part of pre-donation testing, and/or as follow-up testing to permit a reduced deferral period for donors deferred for malaria risk. There was general agreement that donor testing could be an alternative strategy to identify and reenter some donors who would have been otherwise deferred for malaria risk. The workshop meeting minutes are available at http://www.fda.gov/cber/minutes/malaria071206t.pdf. The outcome of the workshop was summarized at the July 13, 2006 BPAC meeting in a presentation available at http://www.fda.gov/ohrms/dockets/ac/06/slides/2006-4226S1_2.pdf.

 

Current FDA Recommendations for Donor Deferral for Malaria Risk

 

The current FDA recommendation to reduce risk of TTM is based on the 1994 memorandum “Recommendations for Deferral of Donors for Malaria Risk” [15].

 

1.      Permanent residents of nonendemic countries who travel to an area considered

endemic for malaria by the Malaria Branch, CDC, U.S. Department of Health

and Human Services, should not be accepted as donors of whole blood and blood

components prior to 1 year after departure from the endemic area. After 1 year after departure, such otherwise suitable prospective donors may be accepted provided that they have been free of unexplained symptoms suggestive of malaria and regardless of whether or not they have received antimalarial chemoprophylaxis.

 

2.      Prospective donors who have had malaria should be deferred for three years after becoming asymptomatic.

 

3.      Immigrants, refugees, citizens, or residents of endemic countries should not be accepted as donors of whole blood or blood components prior to three years after departure from the area. After the three-year period, otherwise suitable prospective donors may be accepted if they have remained free of unexplained symptoms suggestive of malaria.

 

 

 

Discussion

 

1. Estimation of current magnitude of donor deferrals for malaria risk by geographic region

 

According to a study conducted by American Red Cross blood collection centers between the years 2000 through 2006, 1.1% of persons in the U.S. who presented for donation were deferred for malaria risk. A vast majority of malaria risk deferrals (91%) were for history of travel to endemic areas while only 8.6% of malaria-risk deferrals were for prior residence in an endemic area [3]. Thus, a strategy to identify the actual malaria exposure in travelers, if it were able to safely allow unexposed individuals to donate, might greatly improve donor availability without further compromising blood safety.

As noted earlier, rates of malaria transmission are dynamic and also greatly vary among different parts of the world. This scenario is well illustrated by malaria in Mexico. During the last 10 years, there has been a steady decline in malaria cases in Mexico, from 15,100 in 1998 to 2,297 in 2007 [16]. Although P. vivax has been the predominant species in Mexico, a small number of P. falciparum infections were also reported (Table 2). However, no cases of P. falciparum were reported in 2007 or to date in 2008. Infections by P. malariae and P. ovale species have not been reported in Mexico in recent years.

 

Table 2. Malaria Morbidity in Mexico by Plasmodium Species 1998 - 2007 [data from reference 16].

 

 

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

P.falciparum and P. vivax  (mixed)

21

16

124

70

16

43

47

22

15

0

P. vivax

15,100

6,386

7,272

4,867

4,596

3,607

3,268

2,804

2,414

2,297

    

Approximately 12 million U.S. residents travel to Mexico each year. Although, the vast majority of these travelers visit areas in Mexico where there is no risk of malaria, many travel to the areas where malaria transmission continues to occur. A large proportion of visitors returning  from Mexico are deferred from blood donation in the U.S. due to malaria risk, but very few of them are actually infected. Deferral of these potential donors represents a significant donor loss; if unexposed individuals could be identified, their reentry would eliminate an unnecessary loss to the U.S blood donor pool.

 

            A recent survey, conducted as a part of the Retrovirus Epidemiology Donor Study (REDS) involving six U.S. blood centers, determined the number of donor deferrals for malaria risk and the geographical locations of their visits [4 and Bryan Spencer, American Red Cross, personal communication; see Table 3]. The study included a total of 1,278,782 allogeneic donations (from 716,641 actual donors). By proportional extrapolation to the total number of national collections, the authors estimated 61,516 donor deferrals each year for travel to endemic areas in Mexico out of an estimated total of 150,468 travel-related deferrals for malaria risk. Thus, travel to Mexico alone constituted a sizeable part (about 41%) of all malaria-related donor deferrals.

  

Table 3:  Distribution of prospective blood donors deferred for history of travel to or residence in malaria-endemic regions and projected number deferred for travel to each region (columns sum to 100%).  Data are shown for six participating REDS-II Blood Centers in 2006 as number of donors deferred, because of exposure in one of eight geographical regions and as percent of total deferrals (%). Data were provided by Bryan Spencer, June 27, 2008.

 

 

 

REDS-II Blood Centers*

 

 

 

 

 

 

BCP

 

BCW

 

HOX

 

ITxM

 

NEARC

 

SARC

Total (%)

Projected annual number of donors deferred nationally

 

Africa

16

 

6

 

13

 

14

 

11

 

18

 

78

(3.7)

5502

 

Asia

100

27

 

37

 

44

 

46

 

69

 

323  

(15.3)

24533

 

Caribbean

9

 

37

 

59

 

88

 

63

 

24

 

280 (13.3)

18866

Central America

75

 

41

 

87

 

74

 

97

 

88

 

462 (21.9)

33507

 

Middle East

2

 

6

 

4

 

7

 

6

 

12

 

37

(1.8)

2533

North  America** (Mexico)

120

229

148

118

131

124

870 (41.3%)

61516

Oceania***

0

0

0

0

0

0

0

0

South America

12

 

7

 

8

 

14

 

6

 

11

 

58

(2.7)

4011

Total

334

353

356

359

360

346

2,108

150, 468

 

*The following blood centers had participated in the study: Blood Centers of the Pacific, San Francisco, CA (BCP); Blood Center of Wisconsin, Milwaukee, WI (BCW); Hoxworth Blood Center, Cincinnati, OH (HOX); Institute for Transfusion Medicine, Pittsburgh, PA (ITxM); New England Region, Dedham, MA (NEARC); American Red Cross, Southern Region, Atlanta, GA (SARC).

**In North America, endemic malaria transmission occurs only in Mexico.

***Endemic countries in Oceania include Vanuatu, Solomon Islands, and Papua New Guinea.

 

2. Evaluation of possible options for testing blood donors for malarial infection

 

A computer-based risk assessment model was developed to estimate the potential risks and benefits of the current deferral policy and possible alternative policy options for persons at-risk for malaria who present to donate in the U.S. The FDA model was developed using data on the annual number of donor deferrals from the REDS-II study [4] and Table 3. The U.S. population blood donation rates for travelers and immigrants were estimated using the National Health Interview Survey for the years 2000 - 2006. Annual estimates of malaria cases for U.S. travelers and immigrants were calculated using data from the CDC for the years from 2001 - 2005 [9, 12, 13].

 

            The current donor deferral policy defers prospective donors who visited malaria-endemic areas during the past year and prospective donors who were prior residents of malaria-endemic countries or had malaria during the past three years [15]. Data in Table 3 suggest that 150,468 persons were deferred for malaria risk in 2006. Our model assumed 85%-99% efficiency of the donor questionnaire in deferring prospective donors who are at-risk for malaria. Results from the model (Table 4) indicate that under the current donor deferral policy the risk of collecting blood units from infected donors is very small, approximately 1.4 units per year.

 

The assessment estimated both the risk of TTM and the effects on donor availability for the current deferral-based policy and for the following three model scenarios involving early donor reentry based on antibody testing: (1) antibody testing of all presenting blood donors (universal testing); (2) antibody testing only of all malaria-at-risk donors (defined by an admitted history of travel to or residence in malaria-endemic areas anywhere in the world); or (3) antibody testing only those travelers returning from malaria-endemic areas in Mexico. Lacking reliable data, we did not consider the number of donors lost due to self-deferral for the risk of malaria exposure in this risk assessment model. To address the issues of varying lengths of pre-patency for different Plasmodium species (time from exposure to first appearance of circulating parasites), the effect of partial immunity on patency and time between parasite exposure and appearance of detectable antibody levels, the model assumed a four-month donor deferral period after return from an endemic area before antibody testing is performed in each of the scenarios involving antibody testing. Identification of malaria-at-risk donors is based on a donor questionnaire that relies on the history of travel or residence in malaria endemic areas [15]. As noted above, in countries where antibody testing is currently used to identify malaria-risk donors, testing is conducted only after deferral of donors for 4 – 6 months following their return from endemic areas [10, 11].

 

Table 4 shows the FDA’s risk-benefit assessment comparing the current donor deferral-based policy with three other possible scenarios in which donor populations with different levels of malaria risk would be tested for malaria antibodies. The hypothetical ELISA-based test for malaria antibodies used in this risk-benefit assessment is assumed to detect antibodies to P. falciparum and P. vivax and to be 99.0% sensitive and 99.8% specific. Whether malarial recombinant protein based ELISA tests can achieve such levels of sensitivity and specificity remains to be determined. Tests of lower sensitivity might increase the risk of TTM, while assays of lower specificity would reduce donor availability, especially in the universal-testing scenario where all presenting donors would be tested (8,406,000 donors, Table 4).  As an illustration, in the universal testing scenario while a test of 99.8% specificity will cause the loss of approximately 16,810 donors due to false-positive reactions; at 99% specificity, donor loss due to false-positive reactions would be 84,000 donors. Test kits to detect antibodies to P. falciparum and P. vivax malarias in blood donors are commercially available outside the U.S. [10, 11]; to our knowledge, no commercial ELISA test is available that is designed to detect antibodies to P. malariae and P. ovale.  In each antibody testing scenario, additional donors might be gained through a reduction in self-deferrals, but the model does not permit that number to be estimated.

 

Results of the FDA assessment suggest that both universal and malaria-at-risk testing scenarios with a test having the described performance characteristics are expected to increase the number of suitable donors without having any significant effect on blood safety relative to the current policy (Table 4). The estimated annual risks were between 1.4 to 1.7 blood units from infected donors entering the blood supply when all four model scenarios were considered (Table 4). Universal testing and testing of malaria-at-risk donors should add approximately 87,000 and 103,000 donors respectively annually to the current donor pool. However, as noted earlier, a major limitation for both testing scenarios is that currently available ELISA tests are designed to detect antibodies to only P. falciparum and P. vivax malarias, while antibodies to P. malariae and P. ovale are not detected. All four species of Plasmodium have caused TTM in the U.S. [1]. Therefore, in the present situation, several factors could deter us from adopting either of the first two model testing scenarios. First, available commercial tests do not detect antibodies to P. malariae and P. ovale, thus leaving open the prospect of TTM caused by these Plasmodium species. Second, testing the entire U.S. donor population may be both inefficient and costly, especially when a similar level of improvement in both donor availability and blood safety from malaria risk can be achieved by testing only a subset of the donor population (8,406,000 donors would be tested in the universal testing model versus only 103,000 donor would be tested in the at-malaria-risk model, Table 4).  

 

The fourth risk-benefit assessment is for the scenario in which only those prospective U.S. donors who traveled to malaria-endemic areas in Mexico are considered for accelerated reentry and tested four months after their return. This assessment is based on the estimate of 61,516 such donors deferred in 2006, as shown in Table 3. We assume that a similar number is deferred annually because of travel to malaria endemic areas in Mexico. Adjusting for the four-month donor deferral before antibody testing would be performed, the model predicts an annual net gain of 37,000 donors of the 61,516 donors who would have been otherwise deferred. Given the low-rates of malaria transmission in endemic areas of Mexico, it is reasonable to believe that most currently deferred donors were never exposed to malaria and would not have malarial antibodies. Such donors would be allowed to reenter after the shortened deferral period.  The scenario predicts essentially no change in risk of TTM and only 75 false-positive test results (Table 4).

 

As noted above, only P. falciparum and P. vivax malarias, and in most recent years only P. vivax malaria, is now transmitted in Mexico [16]. Available ELISA tests that detect antibodies to P. falciparum and P. vivax only, while not appropriate for testing all at-risk blood donors, might be acceptable for those returning from endemic areas of Mexico; this testing strategy might allow the early reinstatement of approximately 37,000 donors who would currently be deferred for malaria risk for 12 months after their return to the U.S. (Table 4). However, the strategy would clearly not detect all malaria infections acquired in those other parts of world where all four Plasmodium species might be present.

 

Table 4. Model results for mean risk and benefit of possible testing scenarios for malarial infection compared to current deferral policy. The model assumed a 4-month deferral before testing after departure from malaria endemic areas.   The projected annual number of donor deferrals is based on data provided in reference 3 and Table 1 as inputs.

 

 

Current Policy

(donor questioning)

 

Universal Testing*

(all presenting donors)

 

Testing* of only malaria risk donors (including travelers to and prior residents of endemic areas)

 

Testing* of only donors who traveled to malaria endemic areas in Mexico

 

Donors tested**

n/a

8,406,000

 

103,000

 

37,000

 

Donor gain over current policy**

n/a

87,000

 

103,000

 

37,000

 

False positive

n/a

16,810

 

207

 

75

 

Total potential blood units from infected donors***

1.4

 

1.5

 

1.7

 

1.4

 

*Model results are derived by use of an assumed ELISA that detects antibodies to two malaria species: P. falciparum and P. vivax.  Calculations are based on a 4-month deferral policy, and assume 99% sensitivity and 99.8% specificity of the antibody test.

**Donor gain shown takes into account for the loss due to false positive units.  Results are rounded to the nearest thousand.

***The predicted number of infected donors and units was calculated using published malaria surveillance data from CDC MMWR reports for the years 2001 – 2005.

 


 

Summary

 

An FDA risk-benefit assessment model was constructed to estimate effects on donor availability and risk of TTM for three model scenarios compared with the current deferral-based policy (Table 4). The three scenarios included antibody testing of the following donor populations: (1) all presenting blood donors (universal testing); (2) all malaria-at-risk donors only; or (3) travelers who visited malaria-endemic areas in Mexico only. The model assumed a four-month deferral after donors left malaria-endemic areas and before antibody testing.

 

The risk-benefit assessment model predicted the following effects on blood safety and availability:

 

a. Using a highly sensitive and specific test for antibodies to P. falciparum and P. vivax four months after returning from endemic areas, the universal-testing strategy and the testing of all malaria-at-risk donors would yield an annual gain of 87,000 and 103,000 donors respectively. None of the testing scenarios increased or decreased the risk of TTM relative to the current policy. The universal-testing model would require the screening of approximately 8,406,000 blood donors per year while  the malaria-at-risk donor testing model test only 103,000 donors. In both these testing model scenarios there remains a residual risk of TTM caused by P. malariae and P. ovale malarias, because the currently available ELISA tests detect antibodies only to P. falciparum and P. vivax.

 

b. Antibody testing of only those donors who traveled to malaria-endemic areas in Mexico and reentry of those donors with negative test results four months after their return would gain 37,000 donors annually without any increased risk of TTM from P. malariae and P. ovale since these two Plasmodium species are currently not present in Mexico.

 

Questions for the Committee

 

1. Given the historic risk of TTM from Plasmodium malariae and Plasmodium ovale, is testing for antibodies only to Plasmodium falciparum and Plasmodium vivax after a four-month deferral, practical and appropriate for use to screen and reenter donors deferred for any risk of malaria? 

 

2. Can selective testing for antibodies to Plasmodium falciparum and Plasmodium vivax four months after possible exposure be used as a criterion to screen and reenter donors deferred for travel to Mexico?   

 

References

1.                  Mungai M, Tegtmeier G, Chamberland M, Parise M. Transfusion-transmitted malaria in the United States from 1963 through 1999. New England Journal of Medicine 2001, 344:1973-1978.

2.                  Guerrero I, Weniger B, Schultz M. Transfusion malaria in the United States, 1972 – 1981. Annals of Internal Medicine 1983, 99: 221-226.

3.                  Leiby D, Nguyen M, Notari E. Impact of donor deferrals for malaria on blood availability in the United States. Transfusion 2008: In press.

4.                  Spencer B, Custer B, Kakaiya R, Hillyer K, Wilkinson S, Gottschall J Steele W. Low risk for malaria transmission from presenting donors excluded for travel to Mexico (abstract). Transfusion 2006:46 Suppl:27A.

5.                  Hay S, Snow R. The malaria atlas project: Developing global maps of malaria risk. PLOS Medicine 2006, 12: 2204-2208.

6.                  Greenwood B, Fidock D, Kyle D, Kappe S, Alonso P, Collins, Duffy P. Malaria: Progress, perils, and prospects for eradication. The Journal of Clinical Investigation 2008, 118: 1266-1276.

7.                  Zucker J. Changing patterns of autochthonous malaria transmission in the United States: A review of recent outbreaks. Emerging Infectious Diseases 1996, 2:37-43.

8.                  Local transmission of Plasmodium vivax malaria-Palm Beach county, Florida, 2003. Morbidity and Mortality Weekly Report 2003; 52(38):908-911.

9.                  Mali S, Steele S, Slutsker L, Arguin P. Malaria survelliance – United States, 2006. MMWR Surveillance Summaries 2008, 57(SS05): 24-39.

10.              Kitchen A, Chiodini P. Malaria and Blood Transfusion. Vox Sanguinis 2006, 90: 77-84.

11.              Elghouzzi M, Senegas A, Steinmetz T, Guntz P, Barlet V, Assal A, Gallian P, Volle P, Chuteau C, Beolet M, Berrebi S, Filisetti D, Doderer C, Abdelrahaman T, Candolfi E. Muticentric evaluation of the DiaMed enzyme-linked immunosorbent assay malaria antibody test for screening of blood donors for malaria. Vox Sanguinis 2008, 94: 33-40.

12.              Shah S, Filler S, Causer L, Rowe A, Bloland P, Barber A, Roberts J, Desai M, Parise M, Steketee R. Malaria surveillance – United States, 2002. MMWR Surveillance Summaries 2004, 53(SS01): 21-34.

13.              Eliades M, Shah S, Nguyen-Dinh P, Newman R, Barber A, Roberts J, Mali S, Parise M, Steketee R. Malaria surveillance – United States, 2003. MMWR Surveillance Summaries 2005, 54(SS02):25-39.

14.              Vinetz J, Li J, McCutchan T, Kaslow D. Plasmodium malariae infections in an asymptomatic 74-year-old Greek woman with splenomegaly. The New England Journal of Medicine 1998, 338:367-371. 

15.              “Recommendations for deferral of donors for malaria risk.” FDA Memorandum, July 26, 1994.

16.              The Epidemiology Division of the Secretariat of Health of Mexico. Sources: http://www.dgepi.salud.gob.mx/boletin/2007/sem52/pdf/cua7.pdfhttp://www.dgepi.salud.gob.mx/boletin/2006/sem52/pdf/cua7.pdf; http://www.dgepi.salud.gob.mx/boletin/2005/sem52/pdf/cua7.pdf.