Syphilis and Blood Donors: What We Know, What We Do Not Know, and What We Need to Know

Sharyn Orton

The lack of transfusion-transmitted syphilis in the Un- tied States in the past 30 years has led us to question the rationale for continued syphilis testing of blood do- nors. In addition, the significance of a confirmed positive syphilis test result in a blood donor is not clear. The following 2 questions have been raised: (1) Are we de- tecting any truly infectious donors? and (2) If not, what

is the significance of a confirmed positive test for syph- ilis in contemporary blood donors? This review will summarize what is and what is not known about syph- ilis testing in blood donors and will discuss the need for further research to answer these questions. Copyright �9 2001 by W.B. Saunders Company

S YPHILIS is an infectious disease caused by the spirochete T pallidum. This association was

first described in 1905, with the demonstration of spirochetes in Giemsa-stained fluid from syphilitic lesions. 1 Person-to-person transmission can occur during sexual intercourse, congenitally, or rarely, through other means, including accidental direct inoculation by a blood transfusion. 2

Early in the 20th century, syphilis was a major public health problem. Transfusion-transmitted syphilis was first described in 1915. By 1941, 138 cases had been described in the literature? Sero- logic tests for syphilis (STS) have been performed on blood donations for more than 60 years, which is the longest of any infectious disease test. Trans- fusion-transmitted syphilis has become nonexistent in the United States because there have been no reported cases in more than 30 years. Yet, of all the tests performed on blood donors, least is known about the current rationale for this test and the significance of the test results in the blood-banking environment. There are 2 questions that we keep asking: (1) Are we detecting any truly infectious donors? and (2) If not, what is the significance of a confirmed positive test for syphilis in the con- temporary allogeneic blood donor?

This review will summarize what is and what is not known about syphilis testing in blood donors and will suggest further work, the results of which may help answer these questions and thus the

From the Transmissible Diseases Department, Jerome Hol- land Laboratory, American Red Cross, Rockville, MD.

Address reprint requests to Sharyn Orton, MT (ASCP) SBB, MSPH, PhD, Holland Laboratory, 15601 CRABBS Branch Way, Rockville, MD 20855.

Copyright �9 2001 by W.B. Saunders Company 0887- 7963/01/1504-0003535. 00/0 doi: 10.1053/tmrv.2001.26956

potential to change our testing practices or how we manage the counseling of blood donors who have positive STS.

WHAT WE KNOW

Natural Histo~ of Disease

In nature, syphilis is contracted during sexual intercourse by passage from genital lesions of 1 partner though the mucous membranes of the other partner. Intracutaneous inoculation results in the organism multiplying at the primary site 4 and pres- ence of organisms in local lymph nodes within min- utes and systemic dissemination within hours. 1,5 Within 4 days, the spirochetes disseminate by in- vasion of the local lymph nodes causing swelling (lymphadenopathy) followed by seeding to the bloodstream in large numbers. 4 Localized lymph- adenopathy and primary lesions (chancres) de- velop in 9 to 90 days, with an average of 21 days. 6,7 The primary chancre can be found in up to 97% of infected individuals, although up to 60% may not remember having had a lesion. Chance healing will occur spontaneously in about 2 to 6 weeks, al- though lymphadenopathy persists. 4,7 Progression from primary to secondary disease occurs in ap- proximately 46 days (range 30 - 70 days). 7

If untreated, the spirochete begins migrating to the bloodstream, and the disease progresses to the secondary phase 6 to 12 weeks after exposure (2 - 8 weeks after development of the chancre), 5,7 with a corresponding vigorous antibody response. 5 Once the organism has invaded the bloodstream, it continues to infiltrate all organs and body fluidsY Peak spirochetemia with high organism load and systemic infection occur during secondary syphilis. A generalized condition with parenchymal, sys- temic, and mucocutaneous manifestations occurs. 7 Secondary phase is the most clinically apparent and is characterized by acute infection caused by

282 Transfusion Medicine Reviews, Vol 15, No 4 (October), 2001: pp 282-292

SYPHILIS AND BLOOD DONORS 283

multiplication and systemic dissemination of the organism. During secondary syphilis, spirochetes are cleared from the bloodstream, probably by a lytic process involving complement. Untreated secondary disease lasts a mean of 3.6 months, with a range of 1 to t2 months. 7 In 50% of cases, the patient will no longer be infectious or exhibit any symptomatology and will be considered cured. The other 50% will go on to a latent phase. 4 This stage progresses to the asymptomatic latent phase in which a positive serologic test for syphilis will be the only indicator of infection. 7

The latent stage of disease is characterized by complete asymptomatology with positive serology being the only evidence of infection. 7 During this time, the organisms remain dormant and are unde- tected in blood and tissues. There are no cutaneous manifestations or clinical symptoms, but the organ- ism may be actively replicating in target organs. Although most patients in late latent syphilis re- main asymptomatic, one third go on to tertiary syphilis 10 to 20 years later. 8 This latent stage may persist for life or progress several years or decades later to tertiary syphilis, which may result in gum- matous lesions, cardiovascular, or neurologic dis- orders. 9 Approximately 30% of untreated patients progress, over a period of many years, to the ter- tiary stage with progressive disease occurring in the ascending aorta and central nervous system. 7

Antibody Development and Testing

In 1906, August von Wasserman ~ developed the first test for syphilis and so began serologic testing for syphilis. This test was a nontreponemal serum reagin test, the availability of which allowed for the testing of blood donors beginning in the 1940s. m Over the years, tests have been modified and changed, improvements made in specificity,

and treponeme-specific antibody tests developed. More recently, treponemal-specific antibody tests have been automated. A summary of these tests is described in Table 1.

Nontreponemal tests. Nontreponemal tests in individuals with early primary syphilis may be negative in 30% to 50% of cases. Humoral anti- bodies are detected by these tests 1 week after development of the primary chancre, a 3 to 5 weeks after acquiring infection. 11 Nontreponemal tests are made with cardiolipin antigen designed to de- tect the IgM (also known as reagin) and IgG anti- body produced in response to syphilis infection. These antibodies are directed at a lipid hapten antigen called cardiolipin, which can be found in many tissues as well as within the spirochete. The source of the cardiolipin that induces the nonspe- cific antibody development is unclear. Serologic screening tests have been developed to detect the antibody-cardiolipin complex by using comple- ment fixation, precipitation, or flocculation. 12 An- tibody titers decline after the secondary phase and may become undetectable in both treated (3 months - 1 year) and untreated individuals. H,13

The rapid plasma reagin (RPR) is the most com- monly used nontreponemal test for donor screen- ing today. RPR uses a combination of cardiolipin antigen and charcoal and measures antilipin anti- bodies formed in the immune response. The result- ing flocculation is a result of a suspended antigen antibody complex that is formed. This test result will revert to nonreactive over time after treatment but varies depending on the interval of time be- tween contact and treatment and severity of illness. Persistence of these tests usually indicates persis- tent infection, reinfection, or a false-positive test result. 7 A second test that is specific for trepo- nemes should be used for confirmation. ~2

Table 1. Antibody Detection by Test in Untreated Syphilis

Nonspecific Specific

Test RPR FTA-ABS Methodology Charcoal flocculation Immunofluorescence Becomes positive - 2 weeks post-exposure - 2 weeks post-exposure Remains positive

Treated 3 months - 1 year Probably for life Untreated Decades; decreasing with latency Probably for life

Sensitivity Primary phase 86% (77-100) 84% (70-100) Secondary phase 100% 100% Latent 98% (95-100) 100%

Specificity 98% (93-99) 97% (94-100)

PK TM TP

Hemagglutination 2 weeks post-exposure

Probably for life Probably for life

76% (69-90) 100% 97% (97-100) 99% (98-100)

284 SHARYN ORTON

Treponemal tests. Because of the nonspecific nature of the tests, further research led to the de- velopment of treponemal specific tests. The latter are designed to detect antibodies directed against treponemal cellular antigens. The fluorescent treponemal antibody absorption test (FTA-ABS) is the most commonly used for confirmation of a positive STS. 12 This test has been available since the mid 1960s and uses the pathogenic Nichols strain of Tpallidum as the antigen. The test detects 2 different antibodies; the first, called a group antibody, reacts with antigens shared with other treponemes. This antibody is present in low titres in most normal nonsyphilitic sera and may be a biological balance between natural and immune antibody, or it may be produced after exposure to other commensal treponemes in the body. 14 Re- moval of the group antibody by absorption allows for the detection of a second T pallidum-specific antibody. Both antibodies are found in patients with T pallidum infection, and the group antibody actually predominates in all phases except second- ary syphilis. The addition of the fluorescein tag results in fluorescent T pallidum, and sample flu- orescence is measured against a standard control. As a manual test, 1 downfall of this test is the subjective evaluation required by the observer. Be- cause the FTA-ABS test is manually performed, there is a certain amount of misclassification be- cause of the subjective nature of result interpreta- tion. (This is also true for RPR.) In the blood donor setting, the practice is often to interpret the results erring on the side of caution, thus increasing the number of reactive (RPR) and positive (FTA-ABS) test reports and thereby increasing disease misclas- sification. Borderline or equivocal reading have less than a 5% chance of being associated with syphilis. 8

In 1965, hemagglutination assay technology was applied for the development of treponeme-specific testing. The resulting Tpallidum hemagglutination assay (TPHA) was further modified to a microhe- magglutination (MHA-TP) format. This test could be used qualitatively, quantitatively, and was simple to use. Recently, the MHA-TP technology has been automated (PKSMTP, Olympus, Irving Texas), eliminating the subjectivity associated with the manual test interpretation. TPHA tests have detectable reactivity approximately 4 weeks after exposure. Antibody titers increase dramatically during secondary syphilis. During latent syphilis,

or after treatment, the titers decline and may be- come undetectable, 9 although even after successful treatment, most infected individuals test positive for life. Because nonpathogenic treponemes cannot be immunologically distinguished from T palli- dum, the TPHA tests will also detect antibodies to these organisms. 7

Agreement between the TPHA and FTA-ABS tests vary from 87% to 93%, depending on the study population. It has been suggested that the MHA-TP test may not be appropriate for routine use as screening procedures because 1% of the general population will have false-positive results. 8 However, current use of the PKXMTP for syphilis testing is standard in the blood-banking environ- ment because of the ease of automation.

Sensitivity of all STS varies with an increase seen as time from exposure increases. 15 There is conflict in literature reviews regarding the earliest positive results for syphilis testing when compar- ing nonspecific cardiolipin tests with the trepo- neme specific tests. However, studies specifically evaluating this issue suggest that treponeme-spe- cific antibody detection occurs at approximately the same time as the nonspecific 6,8 and possibly slightly sooner (Table 1). 9'15

Other Research Tests

Rabbit infectivity testing. The "gold standard" for syphilis infectivity is rabbit infectivity testing (RIT) because it is a direct measure of infectivity. It appears that RIT requires approximately 10 vi- able organisms per inoculum to transmit infec- tion. 16 Intertesticular or interdermal inoculation causes formation of a localized tissue lesion that remains infective for life, may be transferred to another animal, and causes positive STS. Incuba- tion may require up to 90 days and multiple animal passages. Because it is impractical and expensive, it is used only in research settings and as a sensi- tivity reference for PCR testing.*

Western blot. Western immunoblot technol- ogy is used in the blood donor environment to perform confirmatory testing of retroviruses. Ex- tension of this technology has led to the develop- ment of a test using whole-cell, Tpallidum antigen. Detection of at least 3 or 4 major antigens having molecular masses of 15.5, 17, 44.5, and 47 Kd indicate a positive test. With clinically confirmed samples, the sensitivity of the assay is 93.8%, and the specificity is 100%. Agreement between the

SYPHILIS AND BLOOD DONORS 285

Western blot (WB) assay and the FTA-ABS is greater than 95%, with the WB having equal or greater sensitivity. The WB assay showed no false positive or equivocal reactivities for nonsyphilitic specimens from individuals with previously false- positive test results, elevatated gamma globulin, or antinuclear antibody, but 1 false-positive test re- sults in a normal blood donor. Unfortunately, the number of samples tested in each group was small (19, 10, and 15, respectively). Restriction of the result interpretation to exclude bands of molecular mass between 30 and 40 Kd minimizes cross- reactivity with flagellar and nonspecific proteins of other treponemes. ~7 Testing by WB allows for the detection of both IgM and IgG antibodies, with detection of IgM antibody suggesting recent or active infection.

Polymerase chain reaction. Polymerase chain reaction (PCR) has been used extensively for vi- ruses and currently is in use for the detection of hepatitis C virus and human immunodeficiency virus (HIV) in blood donors. The principle of PCR technology is the use of successive rounds of in vitro replication of nucleic acid sequences using primers complementary to specific targets. Mil- lions of copies of the target gene are made, making it possible to detect minute quantities of a trans- missible agent in the blood. Although this tech- nology is being used with transfusion-transmitted viruses, little work has been done related to trans- fusion-transmitted bacteria (ie, Tpallidum). 1.

DNA. The first description of PCR for testing of T pallidum DNA was in 1990. This technique amplified the TmpA and 4D genes, had a sensitiv- ity to approximately 65 organisms per 0.5 mL, and a specificity of approximately 96%. 16 In 1991, the description of the use of the highly conserved gene for the T paIlidum 47 Kd immunogeneic mem- brane protein was first reported. 19 This target is commonly used in PCR technology because it is rarely found to cross-react with other commensal spirochetes. 8 Early studies using normal sera and cerebrospinal fluid from 9 adults with untreated syphilis reported a sensitivity of 100% in samples with at least 10 organisms, but inhibitory effects caused a lower sensitivity at lower organism con- centrations. Specificity was 100% against other bacterial flora and human DNA. 19 ~

More recently, the 47 Kd target has been used in a multiplex test format that includes testing for T pallidum DNA PCR with known infected samples

compared with dark-field microscopy with a re- ported sensitivity of 91% and specificity of 100%. Analysis of unrelated organisms and other com- mensal treponemes indicated no cross-reactivity. 8 When compared with individual confirmatory PCR assays, concordance was 100%. When comparing PCR with STS, the specificity discrepancy (PCR negative, STS positive) between the 2 tests was attributable to a possible sampling error or a lower specificity of the dark-field microscopy because of commensal organisms and/or serologic false posi- tives. The sensitivity discrepancies (PCR positive, STS negative) were reported to potentially occur in patients with early syphilis when STS specificity may be as tow as 75%. 2o

One limitation of PCR testing for T pallidum is the inability to differentiate between the DNA of viable or dead organisms. 8 Persistence of the T pallidum DNA in successfully treated individuals with neurosyphitis may be because of the very stable nature of the T pallidum DNA biopolymer. Therefore, the length of time that there is persis- tence of dead organism is unknown. 21

RNA. New tests based on the amplification of Tpallidum RNA have been developed. Although 2 copies of ribosomal DNA have been identified in T pallidum, the number of copies of ribosomal RNA per cell is not known, although it is assumed that like other bacterial species there are many copies. No cross-reactivity with other spirochetes has been shown, including testing with B burgdorferi and the nonpathogenic spirochetes. Because it is un- known if DNA detection is derived from viable or dead organisms, it is possible that RNA detection suggests the presence of living versus dead organ- isms because RNA is rapidly degraded after organ- ism death occurs, making it a more reliable indi- cator of potential infectivity. 22 Reverse transcriptase PCR (RT-PCR) is a powerful tool for detection of rare RNA and is more sensitive than DNA PCR because multiple copies of RNA are found per organism, allowing for the detection of less than 1 genome equivalent. 23

Current RNA RT-PCR methodology uses the 16S ribosomal RNA of T pallidum as a template with the production of complementary DNA and has used amniotic fluid, sera, and cerebrospinal fluid samples. Samples spiked with T pallidum organisms have shown sensitivity to detect as low as 10 .2 to 10 -3 T pallidum equivalents of organ- isms (that is, 102 or 103 RNA copies per organism

286 SHARYN ORTON

compared with l DNA). This means that a sample can be diluted over 100-fold more than for DNA- based PCR while still detecting organisms (pro- vided there is at least 1 organism per sample). Sensitivity testing of 20 clinical isolates that had been infectious using RIT was 100%. No cross- amplification was detected by using other species as templates (100% specificity). 22

Recent Historic Perspectives of STS for Blood Donation

In 1985, although the requirement for syphilis testing was dropped from the American Associa- tion of Blood Banks Standards, the Food and Drug Administration (FDA) failed to support this change. The FDA's position at that time was that STS be retained as a potential surrogate for high- risk behavior for HIV infection, rather than as evidence of syphilis, and syphilis screening con- tinued to be a mandatory requirement? ~ In 1995, the National Institutes of Health convened a Con- sensus Development Conference, which included discussion of the extent to which tests for syphilis contributed to transfusion safety and whether their use as in current practice be continued or modified. The resulting recommendation was to continue syphilis screening of all blood donors until such time as more information was available regarding the effect of component storage conditions on T pallidum survival and molecular techniques could assess the absence of T pallidum in serologically positive donated blood, w In 1999, the FDA re- quested the submission of available data to again re-evaluate the need for continued syphilis testing of blood donors. After presentation of data at the September 2000 Blood Product Advisory Commit- tee Meeting, the committee voted that the current scientific data were insufficient to warrant discon- tinuation of donor testing for antibodies to syphilis.

Syphilis Screening as a Surrogate Marker for HIV

Although donor deferral strategies and routine testing for HIV have excluded nearly all HIV- positive donations, the usefulness of STS as a surrogate marker for HIV infection in seronegative blood donors is still being debated. In 1995, the National Institutes of Health also addressed the rationale of continuing syphilis screening as a sur- rogate marker for other transfusion-transmitted in- fections and concluded that the low efficacy was

not sufficient by itself to warrant the continued use of syphilis screening in blood donors. 1~

The most extensive evaluation of syphilis screening as a surrogate marker for HIV was pub- lished in 1997. The source population included 4,468,570 donations collected between 1992 and 1994 at 18 American Red Cross (ARC) sites. Of an estimated t3 HIV infectious window-period dona- tions potentially donated during this time, 0.2 would have been removed because of a reactive STS alone at a cost of over $16 million. 24 With the incorporation of HIV p24 antigen testing in 1995 and the addition of nucleic acid testing for HIV in 1999, the residual risk of HIV infectious window- period donations has been reduced even further, making the value of STS as a surrogate marker negligible, if of any value at all.

Routine Blood Donor Testing for Syphilis

Over the years, blood centers have shifted from the use of the nontreponemal tests (ie, RPR) to treponemal tests (ie, PKTMTP hemag- glutination) for donor screening. This has been primarily because of the ease of automation and the minimalization of technologist-dependent in- terpretation of test results. However, this has led also to detecting lifelong antibodies (versus an- tibodies suggesting current, recent, or persistent infection).

The current routine blood donor testing algo- rithm for syphilis includes the use of either a nontreponemal test (RPR) or an automated treponemat test (PKTMTP) as the screening test. If this test is reactive, it is confirmed by testing with an additional treponemal test, the FTA- ABS or an enzyme immunoassay. Sensitivity of the PKTMTP test varies by phase of disease but has clinical sensitivity (that is, tested in individ- uals with clinically diagnosed disease) of 93.9% and specificity of 95.6%. The FTA-ABS has a relative sensitivity (that is, compared with clin- ical samples tested by PKTMTP) of 99% and specificity of 93%. Blood centers that use the PK~MTP as a screening test may further test the donor using the RPR as a measure of recent exposure for the purposes of donor counseling.

It is important to point out that all of these tests have been developed for diagnostic use and not for use as screening tests. The use of diagnostic tests for screening low-risk populations results in a rel- ative increase in false-positive test results, which

SYPHILIS AND BLOOD DONORS 287

leads to a low-positive predictive value (PPV or percent of individuals who test positive and are currently infected) of a positive test. By using Baye's theorem, 25 the PPV of the combined tests (PKTMTP and FTA-ABS) can be calculated using the following formula:

PPV = disease prevalence

• sensitivity/([prevalence • sensitivity] + [(1

- prevalence) • (1 - specificity)]).

Today, by using the incidence of clinical syphilis in the United States (2.6/100,000) as an estimate of prevalence, the PPV for disease detection in the general population can be calculated as being < 0.1%. Because blood donors are a low-risk popu- lation (and current methodology also detects anti- body from old, treated disease), the PPV for this group may be even lower.

To show the number of donors in the United States per year affected by positive STS results, data from the ARC may be used and extrapolated. Between July 1998 and October 1999, approxi- mately 7.45 million donations were tested for syphilis. Of these, 16,907 (--13,525 per year) had a reactive syphilis screening test result, and 40% had a confirmed positive FTA-ABS (--5,400 per year). Of the confirmed positives, 27% were RPR reactive. Because the ARC draws approximately 50% of US blood donors, doubling of these figures would yield an estimate of nationwide figures.

Interpretation of Positive Syphilis Test Results in Blood Donors

It is unclear if our multiple testing algorithms are truly identifying any infectious donations. A cases series report of platelet concentrates and se- rum samples from blood donors with PKTMTP+, FTA-ABS + STS, tested for T pallidum DNA and RNA was presented at the September 2000 Blood Products Advisory Committee meeting in Be- thesda, MD. These samples were tested using the DNA test that was Tpallidum pol A-gene specific, a DNA test that was a mnltiplex PCR using a T pallidum 47 Kd gene target and/or an RT-PCR test for RNA using Tpallidum 16S ribosomal RNA as a template for production of a complementary DNA target, as previously described: At least 100 samples (50 RPR+ and 50 R P R - ) were tested by each method. All samples were negative for evi- dence of T pallidum DNA or RNA. The lack of

demonstrable T pallidum DNA or RNA suggests that the probability that the blood of donors who have confirmed positive syphilis test results is in- fectious for syphilis is 0% to 3%. 26 These results are in contrast to recent data reported from the analysis of 2 syphilis outbreaks in the United States, which found that Tpallidum DNA could be detected in the blood of untreated, infected indi- viduals during all phases of syphilis infection. 2v,28 By extension, it is unlikely that blood donors with confirmed positive syphilis tests are truly infected with syphilis.

To what can we attribute confirmed and uncon- firmed positive syphilis test results in blood do- nors? There is ample literature addressing condi- tions associated with false-positive syphilis test results. 6,7,8,t2,29,3~ To determine if these condi- tions applied to blood donors, a history of condi- tions reported to be associated with false-positive syphilis tests (Table 2), and/or a previous history of syphilis infection was assessed by a case-control study of donors with reactive (both confirmed and nonconfirmed) and nonreactive PKTMTP test re- sults by using an anonymous mail survey. Among responding blood donors, approximately 50% of donors with FTA-ABS-positive test results re- ported a syphilis history. There was no difference in reported false-positive conditions for either FTA-ABS-positive or FTA-ABS-negative cases, compared with controls. These results suggest that (1) a large proportion of confirmed positive test results are a result of old, treated infection and (2) although historically conditions in the literature associated with false-positive test results have been used to explain both unconfirmed and confirmed

Table 2. Published Causes of a False Positive STS

Nonspecific Specific Tests Tests

Pregnancy X X Some viral/bacterial infections X X Liver disease~hepatitis X X Vaccinations/immunizations X Hypergammaglobulinernia X Leptospirosis X Narcotic addiction X X Lupus erythematosus X X Rheumatoid arthritis X X Other autoimmune/collagen diseases X X Aging X Diabetes X Lyme disease X X

288 SHARYN ORTON

false-positive STS test results in healthy blood donors, this may not be appropriate. 32

Biological Properties Relevant to Potential Transfusion Transmission

The properties necessary to classify an infec- tious agent as potentially transfusion transmissible include presence in blood; infectious asymptom- atic, carrier or latent stages; and stability or growth of the agent in stored blood or blood components, 33 such as red blood cells and platelet concentrates.

Several additional factors influence the proba- bility of a blood donor being truly infected with syphilis and thus the probability of the presence of treponeme in blood. First, the frequency of syphilis infection in the general population has changed over time. In 1947, before the discovery and use of penicillin, the incidence of primary and secondary syphilis was 66.4 cases per 100,000 persons. ~ The last syphilis epidemic was in 1990 when the inci- dence was 20.3 per 100,000 persons. More re- cently, the incidence of syphilis has decreased as a widespread disease (2.6/100,000 in 1998). 34 This low incidence in the general population decreases the probability of a blood donor being infectious for syphilis. Second, properties of the different phases of syphilis infection may be related to the likelihood that an individual would present as a blood donor and/or that an individual's blood is infectious. Early primary syphilis is characterized by rapid dissemination that may precede antibody development and symptomotology and has re- sulted in documented transfusion acquired infec- tion from blood donors in the seronegative window period. 8 Although this potential has always been present, we still have not seen a documented case of transfusion-tranmitted syphilis in the United States in over 30 years. Conversely, secondary phase is the most clinically apparent and is char- acterized by acute infection caused by multiplica- tion and systemic dissemination of the organism, with peak spirochetemia. 8 It is unlikely that an infected individual in this phase would present as a blood donor. Finally, untreated syphilis can progress to a latent phase. Although there is data regarding T pallidum DNA detection, the infec- tious potential for blood in this phase is not well described.

Inoculation studies have shown that the infec- tious potential of whole blood stored at 4~ is directly proportional to the size of the inoculum.

Inoculum concentrations of 5.0 • 10e4 trepo- nemes/mL, 1.25 • 10e6 treponemes/mL, and 2.5 • 10e7 treponemes/mL were able to induce syphilis in rabbits up to 36 hours, 72 hours, and 108 hours, respectivelyY ,36 Because current trans- fusion demands for red blood cells often lead to the transfusion of blood that is a few days old, it cannot be assumed that storage at 4~ would pre- clude transfusion transmission of the agent.~~

WHAT WE DO NOT KNOW

Several hypotheses have been put forward to explain the lack of reports of transfusion-transmit- ted syphilis. These include (1) a low incidence rate of the disease, (2) movement to all volunteer blood donors, (3) deferral of individuals with high-risk behaviour using donor history questionnaires, (4) serologic testing of blood donors, (5) the impact of refrigeration on the survival of the spirochete, (6) use of antibiotics in hospitalized patients, and (7) transmission with an as yet undetected infection30

Preparation and storage of individual blood components may have an impact on the potential for transfusion transmission. However, several key factors that could affect this potential are unknown. First, the concentration of treponeme in the blood of a naturally infected individual at different phases of infection has not been published. Prelim- inary data on 5 blood samples from individual's studied in a syphilis outbreak in Maricopa County, AZ, suggest that the concentration of treponeme varies from approximately 50 to 100,000 organ- isms per mL, depending on the phase of infec- tion. 27 Second, the early 4~ storage studies make no comparison to any other storage temperature. Therefore, no reliable information is available on the true effect 4~ storage has on T pallidum survival. Conversely, platelet concentrates are stored in bags designed to facilitate oxygen flow. This property of the platelet bag leads to an inter- nal oxygen content (--15%) higher than that which can be withstood by the Tpallidum spirochete (1% - 4 % ) , 37 although the actual impact on survival time is unknown.

Evidence that the Tpallidum spirochete attaches to the surface membrane of phagocytes 38 may have implications for the increased use of leukoreduc- tion filters. Transfusion medicine is moving toward a policy of universal leukoreduction of all cellular blood components (ie, red blood cells and plate- lets) for the removal of white blood cells. This

SYPHILIS AND BLOOD DONORS 289

universal leukoreduction may impact the potential for transfusion-transmitted syphilis.

It is unlikely that a transfusion-transmitted case of syphilis would not be identified, based on past descriptions of transfusion-transmitted syphilis, which presented as acute, fulminant symptomatic secondary syphilis. 39 In addition, very specific an- tibiotic use would have to occur to cure a patient with transfusion-transmitted syphilis. The acute symptomatology usually associated with the spiro- chetemic phase of the disease may also make it unlikely that an infected individual would present as a blood donor. However, none of the hypotheses stated earlier have been fully tested.

Testing Limitations

A serious limitation of many published articles is the lack of a gold standard for diagnosis of syphilis in blood donors. The gold standard for syphilis testing, RIT, is not used routinely in the evaluation of licensed test kits for diagnosis and/or screening of blood donors. For the purposes of overall sensitivity determination, clinical diagnosis has been used as the gold standard. As new tests are developed, sensitivity and specificity testing are calculated and compared relative with the stan- dard test in use at that time, rather than by testing large groups of samples from clinically diagnosed individuals. This type of comparison may be useful in comparing positive tests (ie, reactivity or titers) in individuals with independent assessment of dis- ease but can be misleading when used with both negative and positive tests in populations without a gold standard disease classification. The resulting relative sensitivity and specificity values reported by the manufacturer may result in a high PPV in diagnosing populations likely to be infected but has questionable applicability when used for screening in low-risk populations, such as blood donors, and may result in disease misclassification.

It should also be noted that although reactive STS are confirmed with an additional test, the use of a multiple test algorithm with tests that lack independence of test results (have commonality of false positives or detection of old antibody) means that the additional test results may not provide additional diagnostic information. 15 An assump- tion of increased specificity of a con~firmatory test calcu]ated using the results found when screen- reactive individuals are tested is misleading. This perceived increase in specificity is a result of a

greater probability of disease after a second posi- tive test result and reflects the increased prevalence in the population tested. As an example, for the FTA-ABS test, studies of unselected sera run in parallel do not support the increased specificity reported in the literature. 15 In individuals who have no evidence of syphilis, even multiple positive tests are not a substitute for clinical judgment. 39

Information for Blood Donors

The relevance of a confirmed positive syphilis test result in an apparently healthy blood donor is an issue today. Although there is ample literature addressing false-positive test results associated with specific conditions when using the current methodology, these conditions may not be relevant to the contemporary interpretation of positive STS in blood donors.

For pKTMTp unconfirmed donors, post-donation counseling information regarding the implication of a false-positive test result is necessarily incom- plete as it is unknown what factors cause a false- positive test result in this population. 32 For donors with FTA-ABS-confirmed results, a recent expo- sure to syphilis appears to be unlikely, 26 and ex- posure to syphilis some time in the past appears to explain approximately 50% of FTA-ABS-con- firmed positive test results. However, false-posi- tive test results also occur without association to the conditions reported to cause false positives, for reasons which are currently unknown. 32 It is known, however, that the majority (90%) of blood donors with confirmed positive syphilis test results continue to test confirmed positive on subsequent donations. 41

WHAT WE NEED

1. Follow-up of transfusion recipients for evi- dence of transfusion-transmitted syphilis would answer the question of whether or not this event ever occurs today. However, the broad scope and associated cost of this type of large-scale study of transfusion recipients would be prohibitive.

2. Follow-up of blood donors with confirmed positive syphilis tests for evidence of infec- tion would answer the question of whether or not any infected blood donors present for donation. Again, the broad scope and associ- ated cost of this type of study would also be prohibitive. However, collaboration with the

290 SHARYN ORTON

Publ ic Heal th Service in in terv iewing indi-

viduals who represent subsequent to having a conf i rmed posit ive syphilis test at donat ion might be informative.

3. Specific studies on the compar tmenta l iza t ion of spirochetes in stored blood components and the effect of storage condi t ions and con- tainers are needed to clarify what impact refr igerat ion (for red blood cells), oxygen tens ion (for platelets), or leukofil tration has on spirochete viabili ty, and by extension, t ransmiss ion potential. In addition, quantifi- cat ion of t reponeme concentrat ions in the blood of natural ly infected individuals is needed.

4. Manufacturers should provide sensit ivity and specificity informat ion result ing from the testing of samples from individuals with and without cl inical ly diagnosed disease (as the gold standard) rather than relative sensit ivi ty and specificity.

5. Because predict ive accuracy varies with the cl inical characteristics of the test populat ion, a descript ion of the predictive value in low- risk popula t ions based on the prevalence of

.

.

disease in that populat ion should be de-

scribed in the manufac ture r ' s package in-

serts. This is especial ly impor tant when a

diagnostic test is used for screening a low-

prevalence populat ion, as is required in the

blood bank ing envi ronment .

With the characterization of the complete T

pallidum genome, ant igens and antibodies

produced in response to these antigens, 5 as

well as the m a n y test methodologies avail-

able (ie, PCR, Western blot), manufacturers

should at tempt to analyze and correct speci-

ficity problems. Alternat ively, if donor spiro-

chetemia is of pr imary concern, deve lopment

of an inexpensive , sensitive, specific, auto-

mated test methodology for presence of spi-

rochete (ie, PCR) is warranted.

Review of the current data on syphil is testing

and the implicat ions of both unconf i rmed and

confirmed posit ive STS in b lood donors

should be done to assure that predonat ion

informat ion and/or post -donat ion counsel ing

of blood donors with posit ive STS is as ac-

curate and complete as possible.

REFERENCES

1. Singh AE, Romanowski B: Syphilis: Review with empha- sis on clinical, epidemiologic, and some biologic features. Clin Microbiol Rev 12:187-209, 1999

2. Felman YM, Nikitas JA: Sexually transmitted diseases. Primary syphilis. Cutis 29:122-136, 1982

3. de Schryver A, Meheus A: Syphilis and blood transfusion: A global perspective. Transfusion 30:844-847, 1990

4. Goens JL, Janniger CK, De Wolf K: Dermatologic and systemic manifestations of syphilis. Am Fam Physician 50: 1013-1020, 1999

5. Norris S: Polypeptides of Treponema pallidum: Progress toward understanding their structural, functional, and immuno- logic roles. Microbio Rev 57, 750-779, 1993

6. Johnson P, Farnie M: Testing. ~for syphilis. Use of the Laboratory in Dermatology. Dermato Clin 12:9-17, 1994

7. Tramont E: Spirochetes. Treponema pallidum syphilis. Mandell, Douglas and Bennett's Principals and Practice of Infectious Disease (ed 4). New York, Churchill Livingstone, 1995, pp 2117-2132.

8. Larsen S, Steiner B, Rudolph A: Laboratory diagnosis and interpretation of tests for syphilis. Clin Microbio Rev 8:1-21, 1995

9. Young H: Syphilis: New diagnostic directions. Interna- tional J STD & AIDS 3:391-413, 1992

10. Infectious Disease Testing for Blood Transfusions. NIH Consensus Statement, Vol 13, 1-27, 1995

11. Moore M, Knox J: Sensitivity and specificity in syphilis serology: Clinical implications. South Med J 58:963-968, 1965

12. Cruikshank R: Treponema: Borrelia. Medical Microbi- ology (ed 12). New York: Longman, Inc., 1973, pp 386-392

13. Young H: Syphilis serology. Sexually transmitted dis- eases. Dermatol Clin 16:691-698, 1998

14. Hunter E: The fluorescent treponemal antibody-absorp- tion (FTA-ABS) test for syphilis. CRC Crit Rev Clin Lab Sciences 5:315-330, 1975

15. Hart G: Syphilis tests in diagnostic and therapeutic de- cision making. Ann Intern Med 104:368-376, 1986

16. Hay PE, Clarke JR, Strugnel RA, et al: Use of the polymerase chain reaction to detect DNA sequences specific to pathogenic treponemes in cerebrospinal fluid. FEMS Microbiol Letters 68:233-238, 1990

17. Byrne RE, Laska S, Bell M, et al: Evaluation of a Treponema pallidum western immunoblot assay as a confirma- tory test for syphilis. J Clin Microbiol 30:115-122, 1992

18. Vyas G, Yang G, Murphy E: Transfusion-related Trans- missible diseases: Detection by polymerase chain reaction-am- plified genes of the microbial agents. Trans Med Rev 8:253- 266, 1994

19. Burstain J, Grimpel E, Lukehar S, et al: Sensitive detec- tion of Treponema pallidum by using the polymerase chain reaction. J Clin Microbiol 29:62-69, 1991

20. Orle K, Gates C, Martin D, et al: Simultaneous PCR detection of Haemophilus ducreyi, Treponema pallidum, and herpes simplex virus types 1 and 2 from genital ulcers. J Clin Microbiol 34:49-54, 1996

21. Noordhoek G, Wolters E, de Jonge M, et al: Detection by

SYPHILIS AND BLOOD DONORS 291

polymerase chain reaction of Treponema pallidum DNA in the cerebrospinal fluid from neurosyphilis patients before and after antibiotic treatment. J Clin Microbiol 29:1976-1984, 1991

22. Centurion-Lara A, Castro C, Shaffer J, et al: Detection of Treponema pallidum by a sensitive reverse transcription PCR. J Clin Microbiol 35:1348-1352, 1997

23. Ni H, Blajchman M: Understanding the polymerase chain reaction. Trans Med Rev 8:242-252, 1994

24. Herrera G, Lackritz E, Jansssen R, et al: Serological test for syphilis as a surrogate marker for human immunodeficiency virus infection among Unites States blood donors. Transfusion 37:836-840, 1997

25. Sox H: Probability theory in the use of diagnostic tests. Ann Intern Med 104:60-66, 1986

26. Orton SL, Liu H, Dodd RY, et al: Prevalence of circu- lating T pallidum DNA and RNA in blood donors with eon- firmed positive syphilis tests. Transfusion (in press)

27. Liu H, McCaustland K, Holloway B: Evaluating the concentration of Treponema pallidum in blood and body fluids using a selni-quantitative PCR method (abstr). 13 th International Society for Sexually Transmitted Disease Research, June 24-27, 2001, Berlin, Germany

28. Valle P, Pimpinelli F, Maini A, et al: Diagnostiee rele- vance of polymerase chain reaction technology for Tpallidum in subjects with syphilis in different phases of infection. Mi- crobiologica 22:99-104, 1999

29. Nandwani R: Modem diagnosis and management of acquired syphilis. Brit J Hospital Med 55:399-403, 1996

30. Celli C, Gharavi A: Origin and pathogenesis of antiphos- pholipid antibodies. Brazilian J Med Biolog Res 31:723-732, 1998

31. Vila P, Hamdandez M, Fernandex M, et al: Prevalence, follow-up and clinical significance of the anticardiolipin anti-

bodies in normal subjects. Thrombosis Haemostasis 72:209- 313, 1994

32. Orton SL, Dodd RY, Williams AE: Absence of risk factors for false positive test results in blood donors with PKXMTP reactive test results for syphilis. Transfusion 41:744- 750, 2001

33. Barbara J: Challenges in transfusion microbiology. Trans Med Rev 7:96-103, 1993

34. CDC Primary and secondary syphilis--United States, 1998. Morbidity and Mortality Weekly 48:873-878, 1998

35. Van der Sluis J, Onvlee P, Kothe F, et al: Transfusion syphilis, survival of Treponema pallidum in donor blood. Vox Sang 47:197-204, 1984

36. Van der Slnis J, Tare K, Vuzeski V, et al: Transfusion syphilis, survival of Treponemapallidum in stored donor blood. Vox Sang 49:1390-399, 1985

37. Fitzgerald TJ: Treponema, in Balow A, Hainsler WJ, Hermannn KL, et al (eds): Manual of Clinical Microbiology (ed 5). Washington, DC, ASM, 1991, pp 567-578

38. Brause B, Roberts R: Attachment of virulent Treponema pallidum to human mononuelear phagocytes. Brit J Venereal Dis 54, 218-224, 1978

39. Morgan HJ: Factors conditioning the transmission of syphilis by blood transfusion. Am J Med Sci 189:808-813, 1935

40. Deacon W, Lucas J, Price E: Fluorescent treponemal antibody-absorption (FTA-ABS) test for syphilis. JAMA 198, 156-160, 1966

41. Aberle-Grasse J, Orton S, Notari IV E, et al: Predictive value of past and current screening tests for syphilis in blood donors: changing from a rapid plasma reagin test to an auto- mated specific treponemal test for screening. Transfusion 39: 206-211, 1999

Genetically Modified Dendritic Cells in Cancer Therapy: Implications for Transfusion Medicine

Ronan Foley, Richard Tozer, and Yonghong Wan

Dendritic cells (DCs) are a heterogeneous population of antigen-presenting cells (APCs) identified in various tis- sues, including the skin (Langerhans cells), lymph nodes {interdigitating and follicular DCs), spleen, and thymus. Properties of DCs include the ability to (1) capture, pro- cess, and present foreign antigens; (2) migrate to lym- phoid-rich tissue; and (3) stimulate innate and adaptive antigen-specific immune responses. Until recently, the ability to study DCs has been limited by their absence in most culture systems. It is now known that specific cytokines can be used to expand DCs to numbers suffi- cient for their in vitro evaluation and for their use in human immunotherapy trials. Human DCs can be de- rived from hematopoietic progenitors (CD34+-derived DCs) or from adherent peripheral blood monocytes (monocyte-derived DCs). Cultured DCs can be recog- nized by a typical veiled morphologic appearance and expression of surface markers that include major histo- compatibil ity complex class II, CD86/BT.2, CD80/BT.1, CD83, and CDla. DCs are susceptible to a variety of gene transfer protocols, which can be used to enhance bio- logical function in vivo. Transduction of DCs wi th genes for defined tumor antigens results in sustained protein expression and presentation of multiple tumor peptides

to host T cells. Alternatively, DCs may be transduced with genes for chemokines or immunostimulatory cyto- kines. Although the combination of ex vivo DC expan- sion and gene transfer is relatively new, preliminary studies suggest that injection of genetically modified autologous DCs may be capable of generating anti-tu- mor immune responses in patients wi th cancer. Preclin- ical animal studies showing potent antigen-specific tu- mor immunity after DC-based vaccination support this hypothesis and provide rationale to further evaluate this approach in patients. Preliminary human studies are now required to evaluate optimal DC dose, schedule of vaccination, route of delivery, and maturational state of cultured cells. Initiation of these phase I/U cell therapy- based studies wi l l occur in collaboration wi th hospital- based transfusion facilities. Issues relating to cell harvesting, storage, culture methodology, and adminis- tration require the collaborative efforts of basic scien- tists, immunologists, clinical investigators, and transfu- sion medicine staff to ensure strict quality control of injected cellular products. This review is intended to provide a brief overview of clinical DC-based gene transfer. Copyright �9 2001 by W.B. Saunders Company

DMINISTRATION of high-dose cytotoxic therapy has shown efficacy in a selected

proportion of patients with cancer. High-dose, che- motherapy/radiation-based treatments are now safely used in a variety of settingsJ -3 Despite this advance, there remains a significant number of patients who fail to achieve a durable complete remission or who relapse after an initial favorable response. As a result, there is a need to evaluate alternative therapeutic strategies that may be used in combination with conventional cytotoxic-based therapies. Immunotherapy with cancer vaccines has attracted significant recent attention. Interest in

From the Department of Laboratory Medicine, Hamilton Health Sciences Corporation; Department of Pathology and Molecular Medicine, McMaster University; and Department of Medicine, Hamilton Health Sciences Corporation and McMas- ter University, Hamilton, Canada.

Address reprint requests to Renan Foley, AID, Hamilton Regional Laboratory Medicine Program, Henderson General Hospital Site, 711 Concession Street, Hamilton, Ontario, Can- ada L8V 1C3.

Copyright �9 2001 by W.B. Saunders Company 0887- 7963/01/1504-0004535. 00/0 doi: l O. 1053/tmrv.2001.26960

immunotherapy is based on recent advances in our understanding of tumor immunology. These ad- vances address concepts relating to (1) tumor-as- sociated antigens (TAAs), (2) antigen presenting cells (APCs), and (3) effector cell subsets.

Dendritic cells (DCs) are specialized APCs ca- pable of inducing a primary immune response. 4-7 Autologous DCs can be expanded and genetically modified ex vivo. 8-12 The technology to perform gene transfer to expanded DCs has led to the development of vaccination protocols that aim to enhance tumor antigen-specific immunity by intro- ducing genes for tumor peptides or immunostimu- latory cytokines into autologous DCs. 13-16 Geneti- cally modified autologous DCs are injected in an effort to stimulate an immune response often against a specific protein(s) on target cancer cells. Generation of an anti-tumor immune response may then be able to prevent cancer recurrence (protec- tive immunity) or induce regression of existing tumor(s) (therapeutic immunity). Although prelim- inary human studies suggest that DC-based vacci- nations are both safe and feasible, questions remain regarding (1) the most effective dose and route of delivery, (2) the optimal state of DC maturation,

292 Transfusion Medicine Reviews, Vol 15, No 4 (October), 2001: pp 292-304