An international study on the serological differential diagnosis of human cystic and alveolar echinococcocis

Bulletin of the World Health Organization, 64 (1): 101-105 (1986) © World Health Organization 1986 An international study on the serological differential diagnosis of human cystic and alveolar echinococcosis BRUNO GOTTSTEIN,1 PETER M. SCHANTZ,2 TODOR TODOROV,3 ADRIEN G. SAIMOT,4 & PATRICK JACQUIER5 An enzyme-linked immunosorbent assay (ELISA) for the serological differentiation of cystic (Echinococcus granulosus) and alveolar (E. multilocularis) echinococcosis in man has been evaluated. A discrimination rate of 95. 1% was found for 82 sera from patients ofgeographically disparate endemic areas. This rate was essentially the same as thatfound for 57 Swiss patients, indicating that inter- and intraspecific strain differences do not influence the test results. The assay method is suitable for immunodiagnostic purposes as well as for seroepidemiological studies. Echinococcosis in humans is a serious problem of worldwide importance. The disease is mainly caused by the larval stage (metacestode) of two species of tapeworm, Echinococcus granulosus (cystic echino- coccosis) and E. multilocularis (alveolar echinococ- cosis). In some countries of the northern hemisphere both species occur sympatrically. For E. granulosus the occurrence of intraspecific variants or strains is well established (1-3).' Substantial differences have been found between the protein patterns of different strains (4, 5), which could also result in different immunogenic epitopes of the synthesized antigens. Little is known about the possible influence of strain differences on the immunodiagnostic potential of the antigens employed in serology. To our knowledge no reports have appeared on the characterization of dif- ferent strains of E. multilocularis. On the basis of the results obtained with E. granulosus, similar strain differences can be postulated that may influence the antigenic pattern of the parasite proteins and thus the response pattern of patients' B lymphocyte clones and the associated humoral immune response. Changes in antigen determinants may therefore occur that could be of major importance in the purification ' Senior Research Assistant, Institute of Parasitology, University of Zurich, Winterthurerstrasse 266a, CH-8057 Zurich, Switzerland. Requests for reprints should be sent to this author. 2 Assistant Chief, Helminthic Diseases Branch, Center for Infec- tious Diseases, Centers for Disease Control, Atlanta, GA, USA. 3 Professor, Clinic for Tropical and Parasitic Diseases, Sofia, Bulgaria. 4 Professor, H6pital Claude Bernard, Paris, France. s Assistant, Laboratory of Parasitic Diseases, University of Nancy, Nancy, France. a See also FAO/UNEP/ WHO Guidelines for surveillance, pre- vention, and control of echinococcosis/hydatidosis. Unpublished document WHO/VPH/81.28, 1981. of antigenic components for immunodiagnostic purposes. In some cases of human echinococcosis differential diagnosis can be achieved by clinical methods, especially by computed tomography (6), as well as by X-ray examinations and sonography. However, in other cases diagnosis is not possible, and in some instances computed tomography is not feasible (7). A recently reported serological test has enabled correct serological diagnosis of 95% of a set of 57 cases of alveolar and cystic echinococcosis diagnosed in Switzerland (8). In this test two different antigen fractions were isolated from E. multilocularis meta- cestode. The first fraction (Eml) consisted of the antigenic components shared by both parasite species (E. granulosus and E. multilocularis), but a second fraction (Em2) was shown to be highly specific for E. multilocularis infections (9, 10). Simultaneous application of both the Eml and Em2 fractions to ELISA-plates permitted serological differentiation to be made between infections caused by the two species. The possible influence of antigenic differ- ences in Echinococcus strains on the host humoral immune response to antigens was discussed pre- viously with regard to influencing the discrimination rate of the ELISA test (8). The present study was designed to evaluate ELISA as a method for the serological differential diagnosis of patients from different geographical areas. We also investigated whether the discrimination rate was independent of the geographical origin of the patients. Furthermore, the organ in which the para- site was located, the stage of the disease when the serum was sampled, and the age of the patients were also taken into account in the evaluation. 4642 101- 102 B. GOTTSTEIN ET AL. MATERIALS AND METHODS Eighty-two sera from patients with alveolar or cystic echinococcosis were obtained from different centres involved in the serodiagnosis and treatment of echinococcosis: Centers for Disease Control, Atlanta, GA, USA (28 sera); Hopital Claude Ber- nard, Paris, France (15 sera); Clinic for Tropical and Parasitic Diseases, Sofia, Bulgaria (17 sera); Uni- versity of Nancy, Vandceuvre-les-Nancy, France (22 sera). The diagnosis was confirmed by clinical, surgical, and histopathological methods in 57 of the cases. Twenty-five inoperable cases were diagnosed by clinical examinations (computed tomography, sonography, and scintigraphy) and by serology. Sera collected at the Institute of Parasitology, University of Zurich, Switzerland,b using a method described previously (8), were used as positive and negative reference sera on each assay microplate: pools of sera originating from patients with E. multi- locularis or E. granulosus infections were used separately as positive controls, and a pool of sera from 50 healthy Swiss blood donors was used as a negative control. The technique used for the serological differential diagnosis has been described previously (8). Two antigen fractions obtained from E. multilocularis metacestodes and isolated by affinity chromato- graphy and immunosorption were used simul- taneously to test each patient's sera. The first antigen fraction (Eml) consisted of the antigen components shared by E. granulosus and E. multilocularis, and the second fraction (Em2) contained a single im- munogenic parasite polypeptide specific for E. multi- locularis (9, 10) plus several host components. Briefly, ELISA-microplatesc were sensitized alternately with antigen solution EmI and Em2 in each row for 16 h at 4 °C; the solutions contained 10 sg protein/ml buffer (0.1 mol/l NaHCO3/Na2CO3, pH 9.6,0.2 g/l NaN3). The sera were diluted 1: 200 (optimized by checker-board titration) in a 0.1 mol/l buffer (0.01 mol/l Na2HPO4/KH2PO4, pH 7.2,0.15 mol/l NaCl) containing Tween 20 (0.5 g/l), NaN3 (0.2 g/l), and BSA (5 g/l). The same buffer solution was used to dilute the conjugate, which consisted of rabbit anti-human IgG (specific for Fc-fragment)d coupled covalently to alkaline phosphatase with the hetero- bifunctional reagent N-succinimidyl-3-(2-pyrodyl- dithio)propionate.e The substrate was incubated at a concentration of 1 mg 4-nitrophenylphosphate/ml buffer (0.05 mol/l NaHCO3/Na2CO3, pH 9.3, con- b Reference Centre for a WHO trial on chemotherapy of human echinococcosis in Europe. c Nunc Immoplates I-96F. d Dako, Copenhagen, Denmark. ' Pharmacia Fine Chemicals, Uppsala, Sweden. Table 1. Discrimination rates for patient groups infected with E. multilocularis or E. granulosus Correct serological No. of differential Group patients diagnosis (%) E. granulosus 50 94.0 E. multilocularis 32 96.9 Both groups combined 82 95.1 E. granulosus° 26 92.3 E. multilocularis° 31 96.7 Both groups combined' 57 94.7 a Data obtained from ref. (8) (i.e., patients with echino- coccosis diagnosed in Switzerland). taining 1 mmol/I MgC12) for 10 minutes, and the enzyme activity was stopped with 50 1l of a solution of 3 mol/I NaOH. Measurements were made using a Dynatech AM 115 reader.f For distinguishing between E. granulosus and E. multilocularis infections, a quotient (x) (desig- nated the "differentiation factor") was calculated using the absorbance A at X = 404 nm obtained from the patients' sera on both antigen fractions (cf. ref. 8): AEmi - Aneg. control X = AEm2 - Aneg. control The limit of discrimination at the 95% confidence level for both species has been determined previously as x = 2.64 (8), and this value was used in the present study. Values of x greater than 2.64 were taken to be diagnostic of E. granulosus, while values less than this were taken to be diagnostic of E. multilocularis. The results obtained for analysis of sera of patients with echinococcosis were grouped into categories, which were compared statistically using a standard, pooled Student's t-test performed on a desk calculator. RESULTS Table 1 shows the discrimination rates for the two species groups treated separately and then combined. No significant difference was observed between each group. For comparison, data obtained from echino- coccosis patients diagnosed in Switzerland (8) are also shown. The results of the serological differential diagnosis, the differentiation factor x, the absorbance A at f Dynatech Produkte AG, Kloten, Switzerland. HUMAN CYSTIC AND ALVEOLAR ECHINOCOCCOSIS Table 2. Results of serological differential diagnosis by enzyme-linked immunosorbent assay (ELISA) of sera from 82 patients with cystic or alveolar echinococcosis (A = absorbance at X = 404 nm) No. of Differentiation Age of patients patients factor (x) (years) AEml AEm2 E. granulosus Correct diagnosis 47 7.72 ± 2.42 39.7 ± 15.8 0.67 ±0.23 0.13 ±0.05 Incorrect diagnosis 3 1.93 ± 0.50 18.7 ± 11.5 0.45 ± 0.08 0.22 ± 0.02 E. multilocularis Correct diagnosis 31 1.22±0.56 52.3 ± 19.4 0.62 ±0.27 0.58 ±0.40 Incorrect diagnosis 1 3.2 54 0.60 0.19 404 nm, and the age of the patients are shown in Table 2. Of 82 patients investigated, 78 were assigned correctly, corresponding to a success rate of 95.107o for the diagnosis. Statistical evaluation of the differ- entiation factors x indicated significant differences (P <0.01) between the group of sera from patients with E. granulosus and that from patients with E. multilocularis that had been correctly diagnosed serologically. In accordance with this finding, the values ofA obtained with the species-specific antigen Em2 were significantly different between these two groups (P <0.05), whereas such differences did not exist when the nonspecific antigen Eml was used. Table 3. Distribution of organs in which the parasite was localized in the patients Serological diagnosis/organ No. of patients E. granulosus (correct)a Liver 1 8 Lungs 4 Liver and lungs 6 Liver and other organs 9 Other organs 5 E. granulosus (incorrect) Liver 1 Lungs 1 Liver and other organs 1 E. multilocularis (correct) Liver 30 Liver and lungs 1 E. multilocularis (incorrect) Liver 1 a Six patients in this category are not included owing to insuf- ficient information. The mean age of patients infected with E. granu- losus who were correctly diagnosed serologically was significantly lower than that of patients with E. multi- locularis. Of note is the finding that patients with E. granulosus who were incorrectly diagnosed serologi- cally were significantly younger than other patients (P < 0.05). Table 4. Geographical origin of patients with alveolar or cystic echinococcosis (N= 82) No. of patients No. of patients Origin (E. granulosus) (E. multilocularis) Patients with correct serological diagnosis Albania 1 Algeria 4 Bulgaria 1 7 Canada 1 Chile 1 - France 7 21 Iran 2 - Iraq 1 Italy 3 Lebanon 1 Morocco 1 - Portugal 1 Syria 1 Turkey 1 - USA 3 10 Patients with incorrect serological diagnosis Algeria 1 - France - 1 Iraq 1 - USA 1 - 103 B. GOTTSTEIN ET AL. Table 5. Clinical history of patients surgery and chemotherapy with regard to Operative state No. without No. undergoing of patients chemotherapy chemotherapy E. granulosus Pre-operative 20 6 Post-operative 13 a 1 1 E. multilocularis Pre-operative - 11 Post-operative - 21 b 0 Three of these patients were incorrectly diagnosed serologi- cally. b One patient incorrectly diagnosed serologically. Table 3 shows the distribution of patients infected with each species in terms of the organ in which the parasite was located. Comparison of patients diag- nosed correctly with those diagnosed incorrectly shows no apparent association between any organ and incorrect serological diagnosis. A similar result was found when the geographical origins of patients were compared (Table 4): the four patients incorrectly diagnosed serologically orig- inated from different areas of the world. Data on serum specimens indicating patients' clinical history are shown in Table 5. Three patients infected with E. granulosus who were incorrectly diagnosed serologically belonged to the group of post-operative patients who had not received chemo- therapy. No significant conclusion about the results in Table 5 can be drawn; however, the tendency of those with E. granulosus who were incorrectly diag- nosed serologically to belong to the group of post- operative patients who had not received chemo- therapy should be noted. For these three cases the time between surgery and serum sampling was 1 month in one instance and 9 months in another, but no equivalent data were available for the third case. No statements can be made about patients with E. multilocularis since only one case occurred with incorrect serological diagnosis. Since the set of patients in this study was selected using specific criteria, the resulting data cannot be used to investigate the sensitivity of the antigens employed. Information on the sensitivity and speci- ficity of the antigen fractions used is available else- where (8-10). DISCUSSION We have previously described a serological pro- cedure that can accurately differentiate E. granulosus from E. multilocularis infections in man (8). Ninety- five per cent of 57 Swiss cases of echinococcosis could be attributed to the correct species of parasite in the study. The specificity of this test has important prac- tical applications for areas where E. granulosus and E. multilocularis occur sympatrically. The current study demonstrates the potential of the method for use in other areas where the parasites are endemic. A discrimination rate of 95% was also found in 82 sera from echinococcosis patients originating from different geographical areas. These 82 sera were pre- viously tested by classical serological tests, such as immunoelectrophoresis, indirect haemagglutination, latex agglutination, or indirect immunofluorescence, and were positive in at least one of them. Serological differential diagnosis was equivocal in only four cases, three of which had a serological differentiation factor (x) very close to the discrimination borderline at the 95% confidence level. These three cases there- fore correspond to the 5% excluded by the confidence level employed. There were no significant differences between the above four patients and the others with regard to the localization of the parasite in host organs and to their geographical origin. Further- more, the absorbance at X = 404 nm of these sera and the antigens Em I and Em2 did not differ significantly from the average absorbance at 404 nm of groups with correct differential diagnosis. Two trends were observed in the group of patients with incorrect sero- logical diagnosis: the serum samples of two of the three cases with E. granulosus that were diagnosed serologically as E. multilocularis infections were sampled only a few months after surgical intervention from patients not receiving chemotherapy. This might lower the significance of the incorrect sero- logical diagnosis since most of the cases that would be examined are likely to be at the preoperative stage. However, in this study the short time between surgery and sampling the sera precludes post-surgical alter- ation of humoral immune response. Secondly, the average age of E. granulosus cases that were incorrectly diagnosed serologically is miuch lower than that of other groups in the study. Two of these cases were 12 years of age: one from the USA with a single cyst in the lungs, and another from Iraq with multiple cysts in the liver, kidneys, and heart, which had only been partially removed by surgery. Similar observations have been reported previously(8, 11), indicating that in young patients a somewhat divergent serological response may occur, and this should be considered in interpreting the immunodiag- nostic results obtained by the method described here. The other two cases with incorrect serological diag- nosis were a 32-year-old patient from Algeria with a single hydatid cyst in the liver and a 54-year-old E. multilocularis patient from France with multiple 104 HUMAN CYSTIC AND ALVEOLAR ECHINOCOCCOSIS 105 dissemination of the metacestode in the liver and bones. Our findings suggest that the use of the Eml and Em2 fractions from E. multilocularis metacestode in ELISA may be used in areas where E. granulosus and E. multilocularis occur sympatrically. The differen- tiation rate is apparently not impaired by local strains or variations. Moreover, the specificity and sensi- tivity of these antigen fractions, described previously (8-10), should qualify them as antigenic reagents for seroepidemiological studies in this field. ACKNOWLEDGEMENTS This work was supported by a grant from the Swiss National Science Foundation (Project No. 3.965-0.82). The authors wish to thank Ms E. Hemmeler for technical assistance and the Swiss Study Group on Echinococcosis for the supply of sera from patients with echinococcosis diagnosed in Switzerland. RESUME ETUDE INTERNATIONALE SUR LE SERODIAGNOSTIC DIFFtRENTIEL DE L'ECHINOCOCCOSE HYDATIQUE ET ALVEOLAIRE HUMAINE L'echinococcose hydatique et 1'echinococcose alveolaire humaines sont provoquees par la larve d'un tenia- Echinococcus granulosus dans le premier cas et E. multi- locularis dans le deuxieme. Cette etude montre qu'un dosage immunoenzymatique (ELISA) mis au point recem- ment se prete a la differenciation serologique de ces deux infections. On a observe un taux de discrimination de 95% chez des malades suisses et des malades originaires d'autres regions d'end6mie. On a examine le serum de 82 patients. Le serodiagnostic s'est reve1l exact pour 78 d'entre eux, soit 95%. Deux des quatre sujets pour lesquels on avait pose un diagnostic errone etaient ages de 12 ans et presentaient une echino- coccose hydatique. Les resultats obtenus demontrent que les fractions antigeniques sont satisfaisantes comme agents antigeniques pour l'immunodiagnostic et pour les Etudes sero-epidemiologiques des 6chinococcoses. REFERENCES 1. THOMPSON, R. C. A. ET AL. Observations on Echino- coccus granulosus of cattle origin in Switzerland. Inter- national journalfor parasitology, 14: 283-291 (1984). 2. KUMARATILAKE, L. M. & THOMPSON, R. C. A. A review of the taxonomy and speciation of the genus Echino- coccus Rudolhi 1801. Zeitschrift fuir Parasitenkunde, 61: 121-146 (1982). 3. SMYTH, J. D. & DAVIES, Z. Occurrence of physiological strains of Echinococcus granulosus demonstrated by in vitro culture of protoscoleces from sheep and horse hydatid cysts. International journal for parasitology, 4: 443-445 (1974). 4. MCMANUS, D. P. A biochemical study of adult and cystic stages of Echinococcus granulosus of human and animal origin from Kenya. Journal of helminthology, 55: 21-27 (1981). 5. KUMARATILAKE, L. M. & THOMPSON, R. C. A. A stan- dardized technique for the comparison of tapeworm soluble proteins by thin-layer isoelectric focusing in polyacrylamide gels, with particular reference to Echinococcus granulosus. Science tools, 26: 21-24 (1979). 6. GRABBE, E. ET AL. Human echinococcosis: diagnostic value of computed tomography. Tropenmedizin und Parasitologie, 32: 35-38 (1981). 7. PANDOLFO, G. ET AL. CT findings in hepatic involve- ment by Echinococcus granulosus. Journal of com- puter assisted tomography, 8: 839-845 (1984). 8. GOTTSTEIN, B. ET AL. Serological differentiation be- tween Echinococcus granulosus and E. multilocularis infections in man. Zeitschrift fur Parasitenkunde, 69: 347-356 (1983). 9. GOTTSTEIN, B. Isolation of an antigen fraction from Echinococcus multilocularis with high species speci- ficity in ELISA and its identification with the Western blotting technique. In: Avrameas, S. et al., ed., Immunoenzymatic techniques. Amsterdam, Elsevier, 1983, pp. 299-302. 10. GOTTSTEIN, B. Purificati ,n and characterization of a specific antigen from Echinococcus multilocularis. Parasite immunology, 7: 201-212 (1985). 11. TODOROV, T. ET AL. [Response of the immune system in children to echinococcosis]. Medizinische Monats- schrift, 30: 59-63 (1976) (in German).