Circulating Candida antigens and antibodies: useful markers of candidemia

Vol. 31, No. 9 JOURNAL OF CLINICAL MICROBIOLOGY, Sep . 1993, p. 2550-2552 0095-1137/93/092550-03$02.00/0 Copy igh © 1993, Ame ican Socie y o Mic obiology Ci cula ing Candida An igens and An ibodies: Use ul Ma ke s o Candidemia JOSE GUTIERREZ,1* CARMEN MAROTO,1 GONZALO PIEDROLA,' ESTRELLA MARTIN,2 AND JOSE ANTONIO PEREZ' Depa amen o de Mic obiologia, Hospi al Uni e si a io San Cecilio, Uni e sidad de G anada, 18012 G anada,' and Se icio de Mic obiologia, Hospi al Uni e si a io de Valme, Uni e sidad de Se illa,' Se ille,2 Spain Recei ed 11 Decembe 1992/Re u ned o modi ica ion 15 Feb ua y 1993/Accep ed 14 May 1993 To in es iga e he u ili y o he 48-kDa an igen om Candida albicans in i s comme cial o m (Di ec igen; Bec on Dickinson) and h ee o he se odiagnos ic me hods (de ec ion o one an igen by Pas o ex Candida [Sano i Diagnos ics Pas eu ] and de ec ion o immunoglobulin G [IgG1 and IgM an ibodies o C. albicans blas oconidia [bioMe ieuxl) o diagnosis o in asi e Candida in ec ion, we conduc ed a p ospec i e clinical ial among 10 pa ien s wi h candidemia (g oup 1), 30 pa ien s colonized by C. albicans (g oup 2), 20 pa ien s wi h bac e emia (g oup 3), and 20 subjec s wi hou clinical o mic obiological e idence o in ec ion. The Di ec igen sys em was posi i e o a leas one se um sample each om eigh pa ien s in g oup 1. In g oups 2, 3, and 4, i was posi i e o only h ee pa ien s. The e was no eac ion o he Pas o ex sys em in any o he pa ien s in ec ed wi h o colonized by C. albicans o in he non-Candida-ca ying con ols. The IgG an ibody concen a ion oscilla ed be ween 100 and 800 (mean, 510 ± 268) IU/ml o he pa ien s in g oup 1. In his g oup, eigh pa ien s had IgG an ibody le els o >400 IU/ml. The pe cen ages o pe sons wi h IgG an ibody le els o >400 IU/ml in g oups 2, 3, and 4 we e 43.3, 0, and 0, espec i ely. Speci ic IgM an ibody was p esen in all g oup 1 pa ien s bu no in hose in g oups 2, 3, and 4. The sensi i i y and speci ici y o he Di ec igen es we e 65 and 97.1%, espec i ely. Fo he Pas o ex es , he sensi i i y was 0%Y. The sensi i i y o IgG an ibodies was 80%, wi h a speci ici y o 81.4%, while he IgM an ibodies we e 100% speci ic and sensi i e. Bo h he posi i e and nega i e p edic i e alues o speci ic IgM an ibodies appea ed o be supe io o hose o he o he h ee es s. In asi e candidiasis is di icul o diagnose and is he cause o subs an ial mo bidi y and mo ali y in immunosup- p essed pa ien s o in hose supe in ec ed wi h he ungus (12). The only cu en ly a ailable, eliable diagnos ic aid o de ec ion o in asi e candidiasis is open biopsy o deep issue. Blood cul u es a e equen ly nega i e wi h p o en deep isce al candidiasis, while pa ien s wi h cen al lines o o he wise colonized a e posi i e (11, 20). Among he sys- ems a ailable o de ec ion o ungemia by Candida spp. a e an immunodiagnos ic assay o an ibodies o mannop o ein and mannan o o he an igens hemsel es (12) and de ec- ion o se um a abini ol o mannose by gas-liquid ch oma- og aphy (3, 10, 15). An immunodominan cy oplasmic 48- kDa an igen (Candida enolase an igen) has ecen ly been iden i ied and de e mined o be p esen in many pa ien s wi h in asi e candidiasis (1, 22). To in es iga e he u ili y o his 48-kDa an igen in i s comme cial o m (Di ec igen; Bec on Dickinson) and h ee o he se odiagnos ic eagen s o in a- si e Candida in ec ion ( he de ec ion o one an igen and he wo emaining an ibodies), we conduc ed a p ospec i e clinical ial among pa ien s wi h suspec ed candidemia. Eigh y pa ien s a high isk o dissemina ed candidiasis we e s udied. Two se um samples om each we e e alua ed o he p esence o Candida albicans an igens by wo me hods and o he p esence o bo h immunoglobulin G (IgG) and IgM an ibodies o C. albicans blas oconidia. Pa ien s we e di ided in o he ollowing ou g oups: 1, 10 pa ien s wi h p o en i s - ime C. albicans sepsis as e i- denced by h ee posi i e blood cul u es (Bec on Dickinson) * Co esponding au ho . (Table 1); 2, 30 pa ien s colonized by C. albicans (coloniza- ion was de ined as he p esence o C. albicans isola ed om mucosal su aces only when he e was no e idence o deep in asi e in ec ion); 3, 20 pa ien s wi h bac e emia and no e idence o Candida in ec ion; 4, 20 subjec s who we e conside ed o ha e no clinical o mic obiologic e idence o in ec ion. When he hemocul u e was posi i e (g oups 1 and 3) and when he subjec showed no signs o e e (g oups 2 and 4), wo se um samples we e collec ed (sepa a ed by 48 h) and ozen a -70°C. The ollowing de ec ion me hods we e used: Di ec igen (liposome immunoassay o de ec ion o he C. albicans 48-kDa p o ein an igen); Pas o ex Candida (la ex pa icles conjuga ed o an an i-mannan monoclonal an ibody; Sano i Diagnos ics Pas eu ), and Candida-Spo IFA (uses a C albicans blas oconidial clone VW32 slide and luo escein-labelled an i-IgG o -IgM human globulin; bioMe ieux). Resul s o IgG we e exp essed as ecip ocal inal i e s (ini ial dilu ion, 1/100). S anda diza ion o his es was achie ed by use o a pool o se a om pa ien s wi h candidiasis (bioMe ieux). IgG concen a ions o >400 IU/ml a e conside ed indica i e o candidemia by he manu ac- u e . IgM an ibody was de e mined a an ini ial dilu ion o 1/10 and exp essed as ei he posi i e o nega i e. The iden i y o IgM was con i med wi h an i-IgG Abso ben RF (Beh ing Ins i u e). The Di ec igen sys em was posi i e o a leas one se um sample om each o he pa ien s in g oup 1, excep o pa ien s 3 and 10 (Table 2). In g oups 2, 3, and 4, i was posi i e only ou imes ( h ee pa ien s) (Table 2). The e was no eac ion o he Pas o ex sys em o any o he in ec ed o colonized pa ien s (g oups 1 and 2) o o he non-Candida- ca ying con ols (g oups 3 and 4) (Table 2). The IgG 2550 NOTES 2551 TABLE 1. In o ma ion abou pa ien s wi h candidemia (g oup 1) Pa ien ou come Unde lying diso de 1 Cu e Ch onic panc ea i is 2 Cu e B onchopneumonia 3 Cu e Duodenos omy 4 Dea h Neoplasm 5 Dea h P ema u e bi h 6 Cu e Neoplasm 7 Cu e Neoplasm 8 Cu e Neoplasm 9 Cu e Neoplasm 10 Cu e Neoplasm an ibody concen a ion oscilla ed be ween 100 and 800 (mean, 510 + 268) IU/ml o pa ien s in g oup 1. In his g oup, eigh pa ien s (80%) had IgG an ibody le els o >400 IU/ml, he posi i i y h eshold sugges ed by he manu ac- u e . The pe cen ages o IgG an ibody le els ha we e >400 IU/ml in g oups 2, 3, and 4 we e 43.3, 0, and 0, espec i ely. Speci ic IgM an ibody was p esen in all g oup 1 pa ien s bu no in hose om g oups 2, 3, and 4 (Table 2). The sensi i i y and speci ici y o he es s we e calcula ed o each pa ien . The sensi i i y and speci ici y o he Di ec igen es we e 65 and 97.1%, espec i ely. Fo he Pas o ex es , he sensi i i y was 0%. The sensi i i y o IgG an ibodies was 80%, wi h a speci ici y o 81.4%, while he IgM an ibodies we e 100% speci ic and sensi i e. While he nega i e p edic- i e alue o speci ic IgG an ibodies appea ed o be excel- len , bo h he posi i e and nega i e p edic i e alues o speci ic IgM an ibodies appea ed o be supe io o hose o he o he h ee es s (Table 3). The diagnosis o in asi e candidiasis is ex emely di icul bo h clinically and mic obiologically, and he ole o a en- dan candidemia is o en ha d o disce n (12). To shed ligh on his p oblem, some se odiagnos ic me hods based upon de ec ion o an igens o an ibodies o C. albicans ha e been p oposed (1-10, 12, 14-22). Al hough comme cial ki s a e TABLE 2. C. albicans an igen and an ibody esul s o pa ien s wi h candidemia (g oup 1), Candida coloniza ion (g oup 2), o bac e emia (g oup 3) and heal hy subjec s (g oup 4) No. o samples G oup (no. o posi i e/ o al IgG concn No. o samples pa ien s)' (IU/ml)b IgM posi i e/ o al Di ec igen Pas o ex 1 (10) 1 2/2 0/2 800 2/2 2 1/2 0/2 200 2/2 3 0/2 0/2 400 2/2 4 2/2 0/2 800 2/2 5 2/2 0/2 100 2/2 6 1/2 0/2 400 2/2 7 2/2 0/2 400 2/2 8 2/2 0/2 800 2/2 9 1/2 0/2 800 2/2 10 0/2 0/2 400 2/2 2 (30) 0/60 0/60 545 ± 140 0/60 3 (20) 1/40 0/40 100 ± 50 0/40 4 (20) 3/40 0/40 110 ± 60 0/40 a Values o indi idual pa ien s in g oup 1 and o all o he pa ien s in g oups 2 o 4 a e shown. b Mean o mean + s anda d de ia ion. TABLE 3. Sensi i i y, speci ici y, and posi i e and nega i e p edic i e alues, calcula ed pe pa ien , o Di ec igen, Pas o ex, and IgG and IgM an ibodies o C. albicans Posi i e Nega i e Tes % Sensi i i y % Speci ici y p edic i e p edic i e alue (%) alue (%) Di ec igen 65 97.1 76.5 95.1 Pas o ex 0 IgG an ibody 80 81.4 38.1 96.6 IgM an ibody 100 100 100 100 a ailable, issue in asion by C. albicans canno be eliably de ec ed by es ing o he p esence o a speci ic an igen (12). These include he la ex es o mannan de ec ion and agglu ina ion wi h liposomes o de ec he 48-kDa cy oplas- mic p o ein an igen. The sensi i i y o bo h es s is imp o ed when se ial assays using mul iple consecu i e se a a e used. Me hods in ol ing an ibodies also appea o be mo e e ec- i e when pe o med in se ies. Fo example, a nega i e inding wi h hemagglu ina ion-based an ibody es s ules ou he possibili y o C. albicans in ec ion (12). Ou indings indica e ha concen a ions o C. albicans blas oconidium- speci ic IgG an ibody le els highe han 400 IU/ml a e obse ed in he majo i y o pa ien s wi h dissemina ed candidiasis. Pe haps o equal impo ance is he high p edic- i e alue o a nega i e esul . We also demons a ed ha C. albicans blas oconidial IgM an ibodies showed e y high sensi i i y and speci ici y o de ec ion o in asi e candidi- asis. We mus conside , howe e , he ac ha hese pa ien s had no su e ed any o he p e ious candidemia; hus, he e iciency o his es is limi ed o a i s - ime in ec ion. I would be in e es ing o s udy speci ic IgM p oduc ion du ing ein ec ion and he du a ion o hese IgM an ibodies should hey appea . Se e al new compa a i e epo s ha e p o- posed he clinical u ili y o in es iga ions conce ning he cy oplasmic C. albicans 48-kDa an igen (22) and de ec ion o he C. albicans mannan an igen by la ex (13) in subjec s wi h o wi hou in asi e candidiasis. We ha e compa ed he eliabili y o ou me hods ( wo an igen de ec ion and wo an ibody de ec ion me hods) o he diagnosis o dissemi- na ed candidiasis. The cy oplasmic an igen de ec ion ki (Di ec igen) had mode a e sensi i i y and high speci ici y in ou popula ion. Walsh e al. (22) ob ained alues simila o ou s (sensi i i y, 64%; speci ici y, 96%), bu hese alues inc eased in cases o in asi e candidiasis. In ou s udy, he Pas o ex ki could no de ec he Candida mannan an igen and did no show adequa e sensi i i y o diagnosis in ei he he dissemina ed-in ec ion o colonized pa ien g oup. Ne - e heless, He en e al. (13) ecommended his es because i was mode a ely sensi i e, al hough i was di icul o ecognize i s alue conside ing he complexi y o he esul s. In addi ion, in his s udy a clea e alua ion o he ki s was u he hinde ed by he absence o clea ly de ined pa ien and con ol popula ions. Se e al ac o s may ha e con ib- u ed o he alse-nega i e de e mina ions o in asi e candi- diasis in ou s udy (g oup 1, pa ien s 3 and 10), such as low concen a ions o he Candida an igen, an ibody-media ed clea ance o he an igen, and in equen sampling. Ou esul s showed ha he iming o se um collec ion and he numbe o samples we e pa icula ly impo an in ob aining a posi i e esul . Specimens we e ob ained om bo h pa- ien s when he hemocul u es we e posi i e, bu by ha ime he an igen could ha e clea ed. I is appa en om ou esul s ha wo an igen-nega i e se um samples do no VOL. 31, 1993 J. CLIN. MICROBIOL. exclude a diagnosis o candidiasis. We he e o e sugges ha a la ge numbe o specimens be analyzed. P ocessing should in ol e a maximum o one eeze- haw cycle be o e es ing, since epea ed cycles o eezing and hawing a e known o dena u e and diminish de ec able an igen ac i i y. In conclusion, when we compa ed ou me hods, wo in ol ing an igen de ec ion and wo in ol ing an ibody de ec ion, o se odiagnosis o in asi e candidiasis, he mos use ul ma ke s in pa ien s wi h i s - ime C. albicans sepsis p o ed o be C. albicans blas oconidial IgM an ibodies. Gi en he appa en complemen a i y o de ec ion o candi- demia and p oduc ion o IgM an ibodies, o adequa ely de ec in asi e candidiasis mul iple se um samples mus be ob ained o an an ibody assay, ei he wi h each se o blood cul u es o la e . In pa ien s wi h suspec ed candidemia, in es iga ion o IgG and IgM an ibodies and he 48-kDa an igen is p oposed. The cos o his mul iple es ing may pe haps be con ained by limi ing es ing o hose pa ien s conside ed o be a high isk o in asi e candidiasis. REFERENCES 1. Anonymous. 1991. Ci cula ing candida an igen: a use ul ma ke o in asi e candidiasis. In ec . Dis. Ale 10:63. 2. A aj, G. F., R. L. Hop e , S. Chesnu , V. Fains ein, and G. P. Bodey, S . 1982. Diagnos ic alue o he enzyme-linked immu- noso ben assay o de ec ion o Candida albicans cy oplasmic an igen in se a o cance pa ien s. J. Clin. Mic obiol. 16:46-52. 3. A ms ong, D. 1989. P oblems in managemen o oppo unis ic ungal diseases. Re . In ec . Dis. 11(Suppl. 7):S1591-S1599. 4. Bailey, J. W., E. Sada, C. B ass, and J. E. Benne . 1985. Diagnosis o sys ema ic candidiasis by la ex agglu ina ion o se um an igen. J. Clin. Mic obiol. 21:749-752. 5. Benne , J. E. 1987. Rapid diagnosis o candidiasis and aspe gil- losis. Re . In ec . Dis. 9:398-402. 6. Bisbe, J., J. M. Mi 6, J. M. To es, C. Alia, J. Mallolas, and M. Ama al. 1986. Diagn6s ico se ol6gico de la candidiasis disemi- nada en he oin6manos. En e m. In ec . Mic obiol. Clin. 4:279- 280. 7. Deacon, A. G. 1986. Es ima ion o se um a abini ol o diagnos- ing in asi e candidosis. J. Clin. Pa hol. 39:842-850. 8. Delozie , J. B., m, C. W. S a on, and L. . Po s, J . 1987. Rapid diagnosis o Candida sepsis in su gical pa ien s. Am. Su g. 53:600-602. 9. Edwa ds, J. E. 1991. Especies de candida, p. 2057. In G. L. Mandell, R. Go don, and J. E. Benne (ed.), En e medades in ecciosas. P incipios y p ac ica. Paname icana, Buenos Ai es. 10. Fuji a, S.-I., and T. Hashimo o. 1992. De ec ion o se um Candida an igens by enzyme-linked immunoso ben assay and a la ex agglu ina ion es wi h an i-Candida albicans and an i- Candida k usei an ibodies. J. Clin. Mic obiol. 30:3132-3137. 11. Gu i6 ez, J., C. Le6n, R. Ma amo os, C. Nogales, and E. Ma in. 1992. Ca he e - ela ed bac e emia and ungemia. Reli- abili y o wo me hods o ca he e cul u e. Diagn. Mic obiol. In ec . Dis. 15:575-578. 12. Gu ie ez, J., and J. Liebana. 1993. Immunological me hods o he de ec ion o s uc u al componen s and me aboli es o bac e ia and ungi in blood. Clinical e ec i eness. Ann. Biol. Clin. 51:83-90. 13. He en , P., D. S ynen, F. He nando, J. F ui , and D. Poulain. 1992. Re ospec i e e alua ion o wo la ex agglu ina ion es s o de ec ion o ci cula ing an igens du ing in asi e candidosis. J. Clin. Mic obiol. 30:2158-2164. 14. Lemieux, C., G. S -Ge main, J. Vincele e, L. Kau man, and L. de Repen igny. 1990. Collabo a i e e alua ion o an igen de ec- ion by a comme cial la ex agglu ina ion es and enzyme immunoassay in he diagnosis o in asi e candidiasis. J. Clin. Mic obiol. 28:249-253. 15. Lew, M. A. 1989. Diagnosis o sys emic candida in ec ions. Annu. Re . Med. 40:87-97. 16. Mo ales, C., J. M. Noguei a, and J. Ga cia de Lomas. 1988. Candidisis sis emicas: p oblemas diagn6s icos ac uales y pe - spec i as u u as. En e m. In ec . Mic obiol. Clin. 6:107-113. 17. Philips, P., A. Dowd, P. Jewesson, G. Radigan, M. G. Tweed- dale, A. Cla ke, I. Gee e, and M. Kelly. 1990. Non alue o an igen de ec ion immunoassays o diagnosis o candidemia. J. Clin. Mic obiol. 28:2320-2326. 18. Quindos, G., A. Zan on, I. Gu ie ez, J. Pon 6n, and R. Cis- e na. 1988. U ilidad de la c oma og a ia liquida de al a p esion en el diagn6s ico de la candidiasis sis emica. En e m. In ec . Mic obiol. Clin. 6:203-207. 19. Reiss, E., R. J. Kuykendall, and L. Kau nan. 1986. An igenemia in abbi s in ec ed wi h Candida albicans se o ype B: de ec ion by enzyme immunoassay and p elimina y cha ac e iza ion o he an igen. J. Med. Ve . Mycol. 24:259-269. 20. Roboz, J., D. C. Kappa os, and J. F. Holland. 1987. Role o indi idual se um pen i ol concen a ions in he diagnosis o dissemina ed isce al candidiasis. Eu . J. Clin. Mic obiol. In- ec . Dis. 6:708-714. 21. Telen i, A., and G. D. Robe s. 1987. El diagn6s ico de candidi- asis diseminada. En e m. In ec . Mic obiol. Clin. 5:311-313. 22. Walsh, T. J., J. W. Ha ho n, J. D. Sobel, e al. 1991. De ec ion o ci cula ing candida enolase by immunoassay in pa ien s wi h cance and in asi e candidiasis. N. Engl. J. Med. 324:1026- 1031. 2552 NOTES