Value of specific immunoglobulin a detection by two ... - Springer Link

Article

Vol. 14. No. 7

585

Eur. J. Clin. Microbiol. Infect. Dis., 1995, 14:585-590

Value of Specific Immunoglobulin A Detection by Two Immunocapture Assays in the Diagnosis of Toxoplasmosis

E Foudrinier*, C. Marx-Chemla, D. Aubert, A. Bonhornme, J.M. Pinon

The diagnosis of Toxoplasmagondii infection is currently based on immunological tests, but tests for IgG and IgM antibodies alone are often insufficient to assess the risk of active disease, especially during pregnancy and in immunodeficient subjects. The supplementary diagnostic value of testing for antitoxoplasmic IgA in cases of acute, chronic, congenital and reactivated toxoplasmosis, relative to classical immunological tests, was evaluated using two immunocapture tests, one based on tachyzoite agglutination and the other on an immunoenzymatic complex recognizing the membrane protein P30 of Toxoplasma gondii. A total of 4,541 sera from 395 uninfected subjects, 468 immunized subjects with chronic infection, 117 subjects with acute infection and 403 children, 103 of whom had congenital toxoplasmosis, was tested. Specific IgA tests were negative in the nonimmune population, but tests for this immunoglobulin subtype became positive very rapidly during primary infection, and IgA disappeared more rapidly than IgM. In the children infected in utero, specific IgAwas detected more frequently than IgM. In contrast, in a population of HIV-seropositive subjects with clinical toxoplasmosis, tests for IgA were poorly sensitive. The two tests for specific IgA produced similar results, except in the early stages of primary infection, in which immunoenzymatic testing for anti-P30 IgA was less sensitive than the agglutination method.

Toxoplasmosis, a protozoan infection, is frequently asymptomatic in humans. The diagnosis is currently based on immunological tests. Classically the study of antitoxoplasmic immunity involves titration of IgG antibodies, which reflect immunity to the parasite, and IgM antibodies which, if present, reveal acute infection. However, technical advances have shown the limitations of these tests as tests for IgM can be positive because of residual specific IgM or even in subjects free of acute infection due to the existence of natural interfering IgM (1-3). In addition, IgM can be absent in children with congenital toxoplasmosis or subjects with secondary reactivation (4-7). For many years we have screened for specific IgA antibodies using an immunocapture method which employs a suspension of Toxoplasma gondii in formol to supplement classical diagnostic methods (4). This has enabled us to improve the

management of infection in pregnant women and to diagnose congenital toxoplasmosis at an earlier stage. The emergence of a specific IgA response in an immunodeficient subject can reflect a reactivation of chronic infection (8, 9). Given the important role of the Toxoplasma gondii membrane protein P30 in early antibody synthesis (10, 11), we introduced an immunocapture technique in 1991 based on immunoenzymological technology which is selective for anti-P30 IgA. In this study we compared the results of the two techniques in the diagnosis of the different types of Toxoplasma gondii infection (acute, chronic, reactivated and congenital).

Materials and Methods Patients and Sera. A total of 4,541 sera from 1,589 patients

Laboratoire de Parasitologie-Mycologie, Equipe 4, Institut National de la Sant6 et de la Recherche M6dicale U314, Centre Hospitalier Universitare, Hopital Maison Blanche, 45 rue Cognacq Jay, 51092 Reims cedex, France.

were tested. The samples were classified into five groups as follows. Group 1 consisted of 1,104 sera from 395 uninfected subjects (IgG-negative). Group 2 consisted of 1,077 sera from 468 subjects immune to Toxoplasma gondii (IgG-positive). Group 3 comprised 308 sera from

586

E u r . J. Clin. M i c r o b i o l . Infect. Dis.

117 subjects monitored after seroconverting to Toxoplasma gondii. Group 4 consisted of 1,554 sera, from 403 children aged from 0 to 10 years. Of these 1,554 sera, 1,482 were collected from 331 children born to mothers infected during pregnancy (103 children were shown to be infected in utero and 228 were not infected, as determined on the basis of the loss of maternal IgG antibodies after 6 to 12 months). Seventy-two sera were from cord blood of neonates born to mothers free of antitoxoplasmic IgG (maternal serostatus was checked one month after delivery). Group 5 comprised 498 sera from 206 subjects infected by human immunodeficiency virus (HIV).

Agglutination Method. The target antigen was produced from the ascitic fluid of Swiss female mice infected by the RH strain of Toxoplasma gondii and grafted with TG 180 sarcoma cells. After three days, peritoneal tachyzoites were collected in 0.9 % NaC1 buffer, washed in phosphate buffered saline (PBS) and incubated with trypsin. The suspension was then washed in PBS and placed in 6 % formaldehyde at a density of 2 x 108 tachyzoites/ml. The suspension was stored in borate albumin buffered saline (BABS). IgG antibodies were titrated by means of high-sensitivity direct agglutination. Immunocapture detection of IgM and IgA was performed using a suspension of Toxoplasmagondii (ICT-M, ICT-A) as previously described (2), with the following modifications: microplates were sensitized with a monoclonal antig-antibody (Tibi 82; Arg~ne Biosoft, France) or an antic~-antibody (Chafi 1.4; Arg~ne Biosoft, France) at a concentration of 2 gg/ml, and 100 gl of serum diluted 1:100 in PBS (1:20 for neonatal sera) were added to three wells. After 2 h and 45 min of incubation at 37~ and three washes in PBS-0.5 % Tween 20, 1.5, 2 and 2.5 x 106 Toxoplasma gondii were added to each well in volumes of 100, 150 and 200 gl, respectively. The plates were read automatically (MR 7000; Dynatech, USA) after 18 h of incubation at room temperature. The results were analyzed using a computer programme (Logaulagg software; Orkys, France) which assigned a score between 0 (complete sedimentation) and 4 (homogeneous carpet of tachyzoites) for each well. The final index corresponded to the sum of the scores for each of the three wells for a given serum and thus ranged from 0 to 12 for

each subject. The ICT-M index cutoff values were 1 for neonates and 9 for adults, with an indeterminate interval between 6 and 8.5. The ICT-A index cutoff values were 1 for the neonates and 2 for adults.

Immunoenzymatic Method. We used the commercial kits ToxoG Ab EIA, ToxoM Ab E I A and ToxoA Ab E I A (refs K2 055, K2 056 and K2 057, Clonatec, France). All the incubations were done at room temperature. Enzymatic detection was based on the reaction of peroxidase with orthophenylene diamine. The reaction was stopped with 1N HC1, and optical density was read at 490 nm on the plate reader. After immunocapture, anti-P30 IgM or IgA (ICP30-M or ICP30-A) was detected by a double-sandwich E I A using a monoclonal anti-P30 antibody labeled with peroxidase and coupled to the antigen. The ICP30-M and ICP30-A indices were calculated from the samples' optical densities and those of the positive and negative controls. The cutoff was set at 0.70, with an indeterminate zone between 0.50 and 0.70 for both IgM and IgA.

Results

Group 1. I C P 3 0 - A results w e r e always n e g a t i v e , b u t an I C T - A i n d e x o f 4 was f o u n d in a w o m a n r e ceiving an o r a l a n t i c o a g u l a n t . I n d e t e r m i n a t e or p o s i t i v e results w e r e o b t a i n e d in 57 cases (14 % ) with I C P 3 0 - M (n = 27), I C T - M (n = 16) o r b o t h (n = 14). Specific I g G was n e v e r d e t e c t e d in r e p e a t s a m p l e s f r o m t h e s e subjects. Group 2. T w e n t y - f i v e of t h e 27 I g G - p o s i t i v e / I g M n e g a t i v e subjects w e r e I C P 3 0 - A p o s i t i v e , a n d 20 w e r e I C T - A p o s i t i v e (18 p o s i t i v e r e s u l t s in b o t h tests). T h e s e v e n I C P 3 0 - A - p o s i t i v e / I C T - A - n e g a tive results ( l o w e s t I C P 3 0 - A - i n d e x 0.83, h i g h e s t 1.46) w e r e a t t r i b u t e d to r e s i d u a l I g A in six cases a n d to s e r o l o g i c a l r e a c t i v a t i o n in o n e r e n a l graft r e c i p i e n t w h o was p r e v i o u s l y I g G p o s i t i v e a n d

T a b l e 1" M e a n values of immunocapture tests for antitoxoplasmic IgA during the ten months following infection. ICT-A = immunocapture of IgAdetermined by tachyzoite detection, cutoff = 2; I C P 3 0 - A = immunocapture of IgA determined by anti-P30 monoclonal antibody detection, cutoff = 0.70. ICT-A

ICP30-A

Time No. sera 15days 1 month 2 months 3 months 4 months 5months 6 months 7months 8months 9 months 10months

22 34 39 34 34 23 25 14 11 5 4

Mean + SD 5.59 8.59 8.87 7.93 5.49 5.98 5.26 4.21 3.86 3.30 0.25

+ 3.87 + 3.45 + 3.61 + 3.79 + 3.20 + 3.04 + 3.51 + 3.70 + 4.20 + 2.39 _+ 0.50

No. sera 23 42 50 38 38 31 29 20 13 9 7

Mean + SD 0.62 1.77 2.71 2.30 1.75 1.36 1.33 1.16 1.26 1.06 0.49

+ 0.61 + 1.39 +_ 1.33 +_ 1.51 + 1.38 + 1.19 + 1.29 + 0.67 + 1.09 + 0.77 + 0.35

Vol. 14, 1995

587

Table 2: Comparison of IgAsignals in sera of children. ICT-A or ICT-M = i mmunocapture of IgA or IgM,

determined by tachyzoite detection; ICP30-A or ICP30-M = immunocapture of IgA or IgM, determined by anti-P30 monoclonal antibody detection. Cord-blood samples" (seronegative mothers)

Test result

No.of repeat IgA responses in uninfected infants a 1st month

2-18 months

No. of repeat IgA responses in infected infants a'b 1st month

2 months-10 years

ICT-A-/ICP30-AICT-A+/ICP30-A ICT-A-/ICP30-A § ICT-A+/ICP30-A +

72 (0) 0 0 0

219 (0) 4 (2) 4 (1) 1 (1)

792 (0) 0 1 (0) 0

22 (5) 10 + 2c (5) 8 (4) 53 (48)

294 21 (14) 14 (13) 37 (25)

Total

72

228

793

95

366

aNumerals in brackets indicate number of IgM-positive sera (ICT-M and/or ICP30-M). bEight were not evaluated in the first month. ~ of the 12 infants were found to be IgA positive at birth (their mother being IgA positive) and IgA negative 10 days later.

Table3: Sensitivity and specificity of immunocapture

tests for uninfected subjects (group 1), recently seroconverted subjects (group 3), infants (group 4) and HIVpositive subjects (group 5). Techniquea

Sensitivity (%)

Group 1 ICT-A ICP30-A ICT-M ICP30-M Group 3 ICT-A ICP30-A ICT-M ICP30-M

Specificity (%)

99.7 100.0 92.4 89.6

96.9 95.7 100.0 100.0

Group4 ICT-A ICP30-A ICT-M ICP30-M

66.3 b 64.2 b 55.8 b 54.7 b

98.3 b 98.3 b 99.0 b 99.3 b

Goup 5 ICT-A ICP30-A [CT-M ICP30-M

23.5 20.5 13.2 7.4

94.9 95.6 98.6 98.6

a patient with an autoimmune disease, a patient undergoing treatment for a lymphoma, two patients consulting for eye problems (one had toxoplasmic chorioretinitis documented by a study of the aqueous humor) and 14 subjects thought to have chronic toxoplasmosis (although 3 were referred by laboratories that detected specific IgM). In this group of 468 subjects we also found indeterminate (n -- 33) or positive (n = 82) results for IgM in 115 samples which contained no specific IgA.

Group 3. The mean IgA index during the ten months following Toxoplasma gondii infection is shown in Table 1. No IgA signal above the cutoff was found in three patients (ICT-A) and five patients (ICP30-A). All the patients in this group had specific IgM at the time of seroconversion.

ICT-A or ICT-M = immunocapture of IgA or IgM, determined by tachyzoite detection; ICP30-A or ICP30-M = immunocapture of IgA or IgM, determined by anti-P30 monoclonal antibody detection. bSensitivity and specificity were calculated from the samples collected in the first month of life.

Group 4. Table 2 shows the results of the two tests in the different categories of children. The performance of each method in the early diagnosis of congenital toxoplasmosis was as follows: 65.2 % (62 of 95) of children showed IgM positivity (ICP30-M, n -- 59; and/or ICT-M, n = 57), 74.7 % (71 of 95) were IgA positive (ICP30-A, n = 61, and/or ICT-~, n = 63) (significant difference relative to IgM)~ and 80 % (76 of 95) w~re IgA-positive and/or IgM-positive.

both IgM and IgA negative. The two ICT-A-positive/ICP30-A-negative results corresponded to one leukemic child receiving chemotherapy at the time of sampling (ICT-A index: 10; this patient later became ICT-A negative) and to one pregnant woman (ICT-A index: 6.5). The 18 ICT-Apositive/ICP30-A-positive results corresponded to

Group 5. Among the 206 HIV-ser0positive subjects, 68 had clinical toxoplasmosis (cerebral, pulmonary or disseminated forms). Specific IgA was detected in t7 (25 %) of these symptomatic subjects (ICP30-A, n = 14; ICT-A, n - 16). However, we also detected specific IgA (ICP30-A, n = 6; ICT-A, n - 7) in seven patients (5 %) with no clinical manifestations of Toxoplasma gondii in-

a

588

fection. In this group, specific IgM was detected in ten symptomatic patients (ICT-M, n = 9; ICP30-M, n -- 5) and in four asymptomatic patients (ICT-M, n = 2; ICP30-M, n = 2).

Sensitivity and Specificity. Table 3 compares the sensitivity and specificity of the two techniques in the different patient groups.

Discussion

The specificity of the two techniques was calculated on the basis of data for group 1 (uninfected subjects). The results confirmed the absence of natural IgA, whereas natural IgM antibodies frequently interfered in both ICT-M and ICP30M. This specificity made IgA a powerful tool for identifying seroconverters among IgG-negative/IgM-positive subj ects. In group 2 (chronic toxoplasmosis, IgG-positive/IgM-negative), a high IgA index raised the possibility of persistent or recurrent IgA antibodies. Most of these patients also had a high specific IgG titer. In practice, this led us to check routinely for specific IgA in subjects with high IgG antibody titers, irrespective of IgM status. The presence of these IgA antibodies suggests that toxoplasmic seroconversion occurred during the preceding months. The possibility of seroconversion without detectable IgM cannot be ruled out although this situation is rare with new IgM immunocapture methods. In group 3 (immune subjects), we established the chronology of seroconversion whenever possible, but in a few cases the initial samples were not available. Results were designated as 'false negative' if specific IgA was absent during the four months following the emergence of specific IgM, a period chosen on the basis of our experience with ICT-A, in which values peak four months after seroconversion and usually become negative in less than eight months (4). The overall spread of values fits with this interpretation although values were widely dispersed, as previously reported (12-14). In this series the rate of IgA-negative responses at four months did not exceed 3 % in ICT-A and 5 % in ICP30-A. This absence of detection might correspond to an IgA response which is eliminated by early treatment (15, 16). Indeed, most of these subjects were pregnant women, and the detection of specific IgM usually leads the treating physician to give early preventive therapy which might influence the

Eur. J. Clin. Microbiol. Infect. Dis.

kinetics of antibody production. We found that the ICT-A signal occurred early (usually concomitant with IgM), whereas the ICP30-A signal often occurred later. The ICT-A test thus appears to be particularly suited to the early diagnosis of seroconversion (IgG-negative/IgM-positive). In addition, while the highest IgA values were found during the 2 or 3 months following seroconversion, there were several cases of persistent IgA (more than a year after infection ). These patients also had high titers of IgG and IgM antibodies and developed clinical toxoplasmosis at the time of seroconversion (fever and/or adenopathies and/or fatigue, or severe pulmonary toxoplasmosis) requiring treatment. Finally, IgA was sometimes present despite the early disappearance (< 8 months) of IgM (7 cases of 117), confirming previous reports (7) and resembling the findings in group 2 (immune subjects). We routinely tested the subjects with chronic IgA antibodies (with or without IgM) for the presence of circulating Toxoplasma gondii. Eleven months after seroconversion, we detected free trophozoites in the venous blood of a female patient with persistent adenopathies who still had high titers of specific IgG, IgM and IgA. The contribution of IgA detection to the early diagnosis of congenital toxoplasmosis is illustrated by the results for group 4 (children born to infected mothers). Like other authors (11, 12, 14, 17), we detected IgA more often than IgM in the children with congenital toxoplasmosis, ICT-A being more sensitive than ICP30-A in this study. However, the specificity of the IgA detected at birth must be confirmed, as indeterminate and positive IgA-index values were found in nine children during the first few days of life, while subsequent sera became negative within less than ten days, with the exception of one patient whose IgA index remained in the indeterminate range until the age of 2 months. We later demonstrated that this was a case of false positivity, as these children lost their maternal IgG without synthesizing new specific antibodies, in the absence of specific treatment. These results are similar to published data (14, 17) and may have been due to the high sensitivity of the test for IgA antibodies of maternal origin in neonatal serum, as all the mothers of these children had moderate or high titers of specific IgA at delivery. In four cases the same phenomenon was observed with IgM. In the same way, two infected children born to IgA-positive mothers were IgA positive at birth but not a few days later. This means that IgA (or IgM) positivity in a neonate born to a mother with antitoxo-

Vol. 14, 1995

plasmic IgA (or IgM) antibodies must be interpreted with care and checked at least ten days later. Contrary to previous reports (8, 18), we found a low proportion of IgA responses among the symptomatic patients with HIV infection. This may have been due to widespread use of chemoprophylaxis against Pneumocystiscarinii,which is also effective against Toxoplasmagondii (19) but does not prevent clinical manifestations when the CD4+ cell count falls (antibody production being less frequent at this stage). The meaning of the IgA signals in the asymptomatic subjects is unclear although they may be related to parasitemia, as circulating Toxoplasma gondii has been isolated during cases of serological reactivation (20-22). In conclusion, our results confirm the diagnostic value of testing for antitoxoplasmic IgA antibodies. These antibodies are absent from uninfected subjects and are detected rapidly after primary infection; they generally emerge and disappear rapidly, making them an isotype of choice for interpreting the presence of IgM antibodies, which can be natural, residual or specific. In addition, abnormal persistence of specific IgA should be interpreted as reflecting progressive infection, especially in women who are, or wish to become, pregnant. In the case of immunodeficient subjects, tests for specific IgA, although they yield more interesting results than those for IgM, lack sensitivity. The determination of IgA complements IgM determination for the diagnosis of congenital toxoplasmosis, and the determination of the two isotypes is particularly useful for early diagnosis as long as positive results at birth are confirmed after ten days of life. The two tests compared here produced similar results, except during the early stage of acute infection in which ICT-A was more sensitive.

Acknowledgement This work was supported by the Programme Hospitalier de Recherche Clinique, Direction des H6pitaux, Minist~re des Affaires Sociales, de la Santd et de la Ville, Paris, France. We thank David Young for checking the English.

References 1. Potasman I, Araujo F, Remington J: Toxoplasmaantigens recognized by naturally occurring human antibodies. Journal of Clinical Microbiology 1986, 24: 1050-1056. 2. Thulliez P, Dutriat P, Saulnier M, Vernes A, Desmonts G: Evaluation de trois r~actifs de d~tection par immunocapture des IgM sp6cifiques de la toxoplasmose. Revue Frans des Laboratoires 1988, 169: 25-31.

589

3. Verhofstede C, Van Renterghem L, Plum J: Comparison of six commercial linked immunosorbent assays for detecting IgM antibodies against Toxoplasmagondii. Journal of Clinical Pathology 1989, 42: 1285-1290. 4. Pinon JM, Thoannes H, Pouletty P, Poirriez J, Damiens J, Pelletier P: Detection of IgA specific for toxoplasmosis in serum and cerebrospinal fluid using a non-enzymatic IgA-capture assay. Diagnostic Immunology 1986, 4: 223227. 5. Pinon JM, Thoannes H, Gruson N: An enzyme-linked immuno-filtration assay to compare infant and maternal antibody profiles in toxoplasmosis. Journal of Immunological Methods 1985, 77: 15-23. 6. Derouin F, Thuilliez Ph, Garin YJF: Int~r6t et limites de la s6rologie de toxoplasmose chez les sujets VIH positifs. Pathologie Biologie 1991, 39, 255-259. 7. Suzuki Y, Remington J: Insufficient sensitivity of the ToxoTest-MT kit for serodiagnosis of toxoplasmic encephalitis in AIDS patients. European Journal of Clinical Microbiology & Infectious Diseases 1990, 9: 238-239. 8. Pinon JM, Foudrinier F, Mougeot G, Marx C, Aubert D, Toupance O, Niel G, Danis M, Camerlynck P, R6my G, Frottier J, Jolly D, Bessi6res MH, Richard-Lenoble D, Bonhomme A: Evaluation of risk and diagnostic value of quantitative assays for anti-Toxoplasma gondii Immunoglobulin A (IgA), IgE and IgM and analytical study of specific IgG in immunodeficient patients. Journal of Clinical Microbiology 1995, 33: 878-884. 9. Zufferey J, Sugar A, Rudaz P, Bille J, Glauser MP, Chave JP: Prevalence of latent toxoplasmosis and serological diagnosis of active infection in HIV-positive patients. European Journal of Clinical Microbiology & Infectious Diseases 1993, 12: 591-595. 10. Santoro F, Afchain D, Pierce JR, Cesbron JY, Ovlaque G, Capron A: Serodiagnosis of Toxoplasma infection using a purified parasite protein (P30). Clinical and Experimental Immunology 1985, 62: 262-269. 11. Decoster A, Caron A, Darcy F, Capron A: IgA antibodies against P30 as markers of congenital and acute toxoplasmosis. Lancet 1988, ii: 1104-1107. 12. Stepick-Biek P, Thulliez P, Araujo FG, Remington JS: IgA antibodies for diagnosis of acute congenital and acquired toxoplasmosis. Journal of Infectious Diseases 1990, 162: 270-273. 13. Francis JM, Joynson D: Duration of specific immunoglobulin A antibody following acute toxoplasmosis as determined by enzyme immunoassay and immunosorbent agglutination assay. European Journal of Clinical Microbiology & Infectious Diseases 1993, 12: 556-560. 14. Patel B, Young Y, Duffy K, Tanner R, Johnson J, Holliman R: Immunoglobulin A detection and the investigation of clinical toxoplasmosis. Journal of Medical Microbiology 1993, 38: 286-292. 15. Le Fichoux Y, Marry P, Chan H: Les IgA s~riques sp6cifiques dans le diagnostic de la toxoplasmose. Annales de P6diatrie 1987, 34: 375-379. 16. Piens MA, Garin JP: New perspectives in the chemoprophylaxis of toxoplasmosis. Chemotherapy 1989, 1: 46-51. 17. Decoster A, Slizewicz B, Simon J, Bazin C, Darcy F, Vittu G, Boulanger C, Champeau Y, Demory JL, Duhamel M and Capron A: Platelia-Toxo IgA, a new kit for early diagnosis of congenital toxoplasmosis by detection of anti-P30 immunoglobulin A antibodies. Journal of Clinical Microbiology 1991, 10: 2291-2295.

590

18. Darcy F, Foudrinier F, Mougeot G, Decoster A, Caron A, Marx-Chemla C, Capron A, Pinon JM: Diagnostic value of specific IgA antibodies in AIDS patients with Toxoplasma infection: a bicentric evaluation. Immunology Letters 1991, 30: 345-348. 19. Clotet B, Romeu J, Sirera G: Cerebral toxoplasmosis and prophylaxis for Pneumocystiscarinfipneumonia. Annals of Internal Medicine 1992, 2: 169. 20. Hoffin JM, Remington JS: Tissue culture isolation of Toxoplasmafrom blood of a patient with AIDS. Archives of Internal Medicine 1985, 145: 925-926.

Eur. J. Clin. Microbiol. Infect. Dis.

21. Derouin F, Vittecoq D, Beauvais B, Bussel A: Toxoplasma parasitaemia associated with serological reactivation of chronic toxoplasmosis in a patient with the acquired immunodeficiency syndrome. Journal of Infection 1987, 14: 189-190. 22. Tirard V, Niel G, Rosenheim M, Katlama C, Ciceatlama C, Ciceron L, Ogunkolade W, Danis M, Gentilini M: Diagnosis of toxoplasmosis in patients with AIDS by isolation of the parasite from the blood. New England Journal of Medicine 1991, 324: 634.