Assessment of the immunological surveillance value of humoral and lymphocyte assays in severe human cystic echinococcosis

TRANSACTIONS

OF THE ROYAL

SOCIETY

OF TROPICAL

MEDICINE

AND HYGIENE

Assessment of the immunological surveillance assays in severe human cystic echinococcosis

(2000) 94,97-102

value of humoral

and lymphocyte

I Departamento de Parasitologia, P. S. Craig3, I. Almeida,’ and D. Da Rosa’ R. Bonifacinol, S. D. Carter’, Institute de Higiene, Facultad de Medicina, Montevideo, Uruguay; ‘Department of Veterinary Pathology, Veterinay Immunology, University of Liverpool, UIZ; 3Department of Biological Sciences, University of Salford, UK

Abstract Thirty cystic echinococcosis (CE) patients in Uruguay with severe bone or secondary disseminated echinococcosis were immunologically assessed using cellular (lymphocyte transformation assay, LTA) and humoral (specific antibody and subclass responses, circulating antigen and circulating immune-complexes) immunological assays during the course of chemotherapy (albendazole and/or praziquantel). CE patients were divided into 4 groups, according to clinical treatment and outcome: (I) surgery and chemotherapy, (II) chemotherapy with outcome unchanged, (III) chemotherapy with outcome improved, and (IV) chemotherapy considered cured. Increased circulating antigen was of prognostic value in some severe CE cases where levels remained high and/or increased. The lymphoproliferative response in vitro to Echinococcus granulosus antigen was statistically greater in all patient groups compared to normal individuals but at lower levels in improved or cured CE patients. Levels of non-specific LTA response were significantly lower than controls for all groups during albendazole treatment (P < 0.001) but returned to normal levels in cured patients, a result consistent with parasite-induced suppression of cellular responses. This study suggests that, at least in severe osseous and secondary CE, immunosurveillance by specific antibodies, especially total specific immunoglobulin, was overall of more practical use than antigen-specific in-vitro lymphocyte transformation assays. Keywords:

cystic echinococcosis,

Echinococcus granulosus, immunosurveillance,

Introduction Cystic echinococcosis (or cystic hydatid disease) (CE), due to infection with the larval stage of the dog tapeworm Echinococcus ganulosus, is a major chronic zoonosis which is endemic over a large area of the world including South America KRAIG et al., 1996; COHEN et al., 1998). In addition to imaging tech-niques, immunological assays have proven particularly useful for diagnosis, especially by detection of specific Echinococcus antibodies (COLTORTI et al., 1988), but in addition detection of circulating antigen (CAg) and of immunecomplexes has been considered (CRAIG, 1986; BARBIERI et al., 1994). There are also reports of specific cellmediated responses in human CE (SIRACUSANO et al., 1988; IOPPOLO et aZ., 1996). There has, however, been little information concerning the monitoring of immune responses of CE patients -throughout the course of disease or the effects of treatment (CRAIG, 19971. The treatment of CE is still &imarily by surgical excision of cystic material, which is successful in most uncomplicated hepatic and pulmonary cases (WHO, 1996). There are some patients for whom surgery is not an option, primarily because of the location and/or disseminated nature of the hydatid cysts. In these cases, anthelmintic chemotherapy is considered beneficial, the main drug of choice being albendazole (ABZ) (HORTON, 1989; TODOROV et al., 1992). Assessment of the effect of chemotherapy has to date mainly been restricted to the use of imaging techniques, such as computerized tomography (CT) scanning, to determine not only the location and number of cysts but also their size and status. However, imaging methods do not always indicate clear changes and hence treatment assessment is difficult. In addition to imaging methods, other means to monitorthe effects of different treatments (especially chemotherapy) would be advantageous, to provide a better, perhaps earlier insight of changes to cysts and thus clinical improvement. Immunological assessment of CE has been largely restricted to single-sample analysis and there have been limited studies following individual patients (e.g., AWAR et al., 1991;

MATOSSIAN

et al., 1992);

none

chemotherapy,

albendazole, Uruguay

cellular immunological changes. vestigates the use of both cellular logical assays to monitor a group with severe and largely inoperable of extended chemotherapy. Materials

The present study inand humoral immunoof patients in Uruguay CE, during the course

and Methods

Patients Thirty

severe

CE

patients,

with

bone

or secondary

disseminated echinococcosis, from rural areas of Uruguay were referred by surgeons to the Clinical Parasitology Unit of the Institute of Hygiene (Montevideo). Referral was mainly because of surgical failure or the inoperable nature of their CE. Chemotherapy with ABZ, or in combination with praziquantel (PZQ), was the primary form of treatment for these patients. However, when the cyst diameter was > 5-7 cm, combined surgery and chemotherapy was generally attempted. Of the 30 CE patients, 18 were male (mean age 39 years * 15) and 12 female (mean age 42 years * 14). The previous duration of the disease ranged from 1 to 13 (mean 7.7) years. The location of the hydatid cysts were bone including spine (9 patients), lung (5), liver (6), both lung and liver (6), peritoneum (1) and brain (3) (Table 1). For all 30 CE patients, immunoassays (serum antibody, antigen or immune-complex detection) were done before and after chemotherapy. Cell-mediated responses, i.e., lymphocyte transformation assays (LTA) were assessed when out-patients came for clinical examination. Where surgery was combined with chemotherapy, humoral and LTA tests were performed before surgery and after each course of chemotherapy. Chemotherapy

ABZ was given at 15 mg/kg bodyweight daily, over 28-day courses, for 3 to 12 courses [mean number of courses: 5.26 according to WHO (1996) guidelines]. PZQ was given at 40 mg/kg once weekly for patients who had side-effects or exhibited a bad response to ABZ alone (3-4 months, in 6 patients).

has consid-

ered changes relating to long-term follow-up of patients. Furthermore, whilst assays for serum antibodies or CAgs have been reported in human CE (GOTTSTEIN et al., 1991; CRAIG, 1986), few post-treatment studies have been undertaken in CE patients which have included

Surge y In bone CE cases surgery involved aspiration, drainage, removal of daughter cysts, curetage of the pathological bone and where appropriate the addition of cement, and in CE of the femur a prosthesis was added owing to

R. BONIFACINO ETAL.

98

Table 1. Summary of clinical status of the disease

data for 30 severe cystic echinococcosis

No. of patients

Age, mean (SD) in years

M:F

I. Surgery + chemotherapy II. Unchanged

7

39 (18)

6: 1

9

38 (12)

7:2

III. Improved

7

45 (15)

4:3

IV. Cured

7

41 (16)

1:6

Group”

Site of the cyst 3 Lun & Li, 2 Spi, 1 Lun, 1 Li 1 Lun &‘Li, 2 Spi, 1 Bra, 3 Li, 1 Bon, 1 Lun 1 Lun & Li, 3 Lun, 1 Bon, 1 Bra, 1 Per 1 Bra, 2 Li, 2 Bon, 1 Lun & Li, 1 Spi

patients

in Uruguay

with different

4 (2-8)

3 (3-3)

Follow-up mean in months k-w) 14 (S-24)

3 (1-14)

5 (3-12)

35 (12-81)

1 (l-3)

3 (3-5)

21 (11-39)

3 (l-10)

5 (3-12)

68 (50-74)

Past surgervb - -

No. of courses chemotherapy”

M, male; F, female; Lun, lung; Li, liver; Spi, spine; Bra, brain; Bon, bone; Per, peritoneal. “See text for explanation. bMean number of operations (number range). ‘Albendazole. Praziquantel was given to 6 patients (3-4 courses).

the possibility of fracture. In the spine location, laminectomy and anterior decompression of the spinal cord were used as the common surgical treatment. In all the other locations radical surgical resection (where possible) of the bigger cysts was performed. CT scanning Each patient was scanned at the beginning of the study and then once every 12 months subsequently. The CT images were evaluated by image analysis to determine the size and status ofindividual cysts, i.e., increase in density, loss of daughter cysts, or laminations and appearance of calcification (HORTON, 1989). In patients with a bone location of the disease, X-ray examination was also used. Follow-up and treatment efficacy Clinical examination and clinical laboratory tests, i.e., liver function tests, haematology recount (Coulter 660), and urinalysis were done during each course of ABZ therapy to assess possible side-effects. In cases of toxic ABZ effects, i.e., glutamic-oxaloacetic transaminase (GOT) > 37 mUI/mL, glutamic-pyruric transaminase (GPT) >40 mUI/mL or leucopenia (white cells > 4000/mm3) the drug was stopped until these parameters returned to normal values. ABZ was resumed at a reduced dose, and if the described side-effects returned on the lower dose of ABZ then PZQ was used instead. Follow-up periods for the 30 CE patients ranged from 8 to 81 months (mean 36.3 months). CE patients were grouped into 4 groups, according to clinical treatment and outcome (Table 1). Group I: Surgery and chemotherapy. This group contained the most severe cases (n = 7) which included spine (2 patients) and secondary CE infection (5 patients) and where at least 1 cyst(s) had diameter > 5 cm. The 2 uatients with snine CE location were also treated with surgery owing tithe compression of the spinal cord by the metacestode tissues. Viability of protoscoleces was assessed in all hydatid cyst samples from patients receiving surgery by microscopy examination of flame-cell activity and by eosin exclusion (CHINNERY & MORRIS, 1986).

Group II: Chemotherapy-outcome unchanged. This group included 9 patients whose clinical symptoms and the CE cysts/lesion had no change in the size/appearance on CT over a chemotherapy period of 31 year. Most were treated for long periods owing to lack of response and 3 patients were given combined treatment with ABZ and PZQ. The follow-up period for this group was 12-8 1 (mean 35) months. Group .III: Chemotherapy-outcome improved. This group included 7 CE patients with reduction of > 25% of the original size of the cyst(s), who also exhibited an

increase in cyst density and apparent calcification by imaging (CT scan). The follow-up period was 11-39 (mean 21) months. Group IV: Chemotherapy-outcome cure. The criterion for cure was the complete disappearance of hydatid cyst(s) by CT scan image, with a period of follow-up of at least 4 years. This group (n = 7) included 3 cases who had prolonged treatment combining ABZ and PZQ to improve the effectiveness of the chemotherapy. The follow-up period for this group was 50-74 (mean 68) months. Blood samples Serum samples were collected before and during each cycle of chemotherapy. For humoral assays, serum was prepared from clotted blood and stored in aliquots at -70°C until used. For cellular immunological assays, blood was taken into sterile bottles (Vacutainer) containing heparin. Thirty healthy blood donors supplied control samples. Humoral immunoassays Specifi antibodies. Anti-Echinococcus antibodies were assayed in sera using a peroxidase enzyme-linked immunosorbent assay (ELISA) as described by COLTORTI (1986) and modified by BONIFACINO et al. (1991). In brief, microtitre plates were coated with sheep hydatid cyst fluid (SHCF). Non-reactive sites were blocked with 1% bovine serum albumin (BSA). Patient serum samples were absorbed (1 h at 37°C) with normal sheep serum, prior to addition to the ELISA plate. Peroxidase-conjugated goat anti-human polyvalent immunoglobulin (Ig) (Sigma) was used to detect specific antibodies and developed with 5-aminosalicylic acid (Sigma) as the chromogen. Optical density (OD) readings were made at 450 nm using a microplate reader (Multiskan MS, Labsystem). Anti-Echinococcus IgG and IgG-subclass antibodies were assayed in a similar way but using the assay described by WEN & CRAIG (1994). Microplates were coated with crude hydatid cyst fluid antigen and, after serum incubation, specific IgG subclasses were revealed using mouse monoclonal antibodies to human IgGl or IgG4 (Oxoid-Unipath, UK). Circulating antipen and immune-comblexes. Soecific CAg was assayed as described by BARBIERI *et al. (1994). In brief, ELISA plates were coated with 10 pg/mL rabbit anti-hydatid cyst lipoprotein (HBLF) (BARBIERI et al., 1993). Plates were blocked with 1% BSA. Sera were added. To avoid interference due to rheumatoid factor 10% of normal rabbit serum was added to the buffer used to dilute serum samples. Peroxidase-conjugated rabbit anti-HBLF was added in

IMMUNOLOGICAL

SURVEILLANCE

subjects. Linear regression analysis was used to identify any parameter correlation.

1% BSA and 2% fetal bovine serum. The substrate was added and OD values were read. HBLF antigen added to normal human serum was used as a standard in the range of 3-8-250 ng/mL and antigen concentration in patient samples was obtained by linear regression. ELISA titres were thus expressed as ng/mL of the reference serum. A titre of less than the mean value ulus 3 SD (90 ng/mL) of the 30 sera from non-hydatid individuals was cc&idered as negative for CAg. Specific circulating immune-complexes (CICs) were assayed in whole human sera using a capture antibody method as described previously (CRAIG, 1986; BONIFACINO et al., 1993). In brief, CICs were precipitated with polyethylene glycol6000 and the precipitateswere mixed with 0.2-M nlvcineiHC1. nH 3. overnight and annlied directly to r&roplates with 20 & 0.054 bicarbdnatecarbonate buffer, pH 9.6, for 1 h. Plates were incubated for 1 h at room temperature with peroxidase-conjugated sheep anti-SHCF diluted 1:200 (100 pL/well). After washing as above, wells were developed using 5-aminosalicylic acid, and OD600 was read as described above. Sera were considered antibody, CAg or CIC-positive when exhibiting OD values equal to or greater than the mean value of control sera plus 3 SD. Lymphocyte

Results Clinical

Chemotherapy side-effects were observed in 2 (6.7%) ofthe 30 patients but in only 1 did the levels ofboth GOT and GPT liver enzvmes increase and then the ABZ therapy was replaced by PZQ. The mean time of surgery after start of chemotherapy was 0.9 months. Specific antibodies

All CE patients (except 1) were seropositive for antiSHCF antibodies at some period during the study, whereas all the normal donors were seronegative. Statistical analysis (Kruskal-Wallis H-test) showed that all the serum samples from the 4 clinical groups showed greater antibody Ievels compared to normal donors (Table 2). The CE patients from Group I (surgery and chemotherapy) had the highest levels of antibodies, followed by the Group II unchanged CE patients. Patients who improved (Group III) or were considered cured (Group IV) had the lowest levels of total specific antibodies. All of the patients who were eventually cured (Group IV), except 1, exhibited reduced serum specific antibodies (total Ig) during the assessment period. One cured CE patient had antibodies that remained high and this patient was unusual in having some remnants of cyst membranes in 1 hydatid cyst. Both improved (Group III) and unchanged (Group II) CE patients failed to show a consistent and statistically significant change in total serum specific antibodies during the study period.

transformation

The LTA was standardized using 30 healthy blood donors. Peripheral blood was taken by venepuncture into preservative-free heparin, and peripheral blood mononuclear cells (PBMC) were isolated using FicollHipaque (Pharmacia). Cells were resuspended at 2 X lo6 cells/ml in RPM1 medium (with 10% bovine calf serum) in round-bottomed microplate wells (Nunc) in triplicate with a final volume of 100 uL/well. Phytohaemagglutinin (PHA; Sigma) or SHCF antigen were added at 8 ug/mL (optimum derived from titration). Lymphocytes without PHA or SHCF antigen were used as-controls (unstimulated lymphocytes). After 48 h of incubation, 3H-thymidine (Amersham, UK) was added to 1 @i/well. After a further 16 h of incubation cells were harvested on to tibre-glass filter papers. Each dried tibre-glass filter disc was immersed in 2 mL of scintillator liquid (Sigma) and subjected to scintillation counting in a Beckman LS 7500 liquid scintillation spectrometer recording 3H-thymidine incorporation as countsimin (cpm). The stimulation index (SI) for each samnle was calculated bv dividing mean cpm for stimulated wells by mean cpm ior unstimulated wells. Statistical

99

OF SEVERE ECHINOCOCCOSIS

and IgG4 antibodies Not all CE patients had detectable raised specific IgGl antibodies to SHCF antigen. However, all of the groups showed significantly raised mean IgGl antibody levels compared to normal donors. The cured patients (Group IV) exhibited the lowest IgGl antibody levels, while the improved group (Group III) exhibited slightly lower IgGl levels (non-significant) than CE patients from Group I (combined surgery and chemotherapy). The unchanged group (Group II) exhibited the highest IgGl levels. Analysis of group trends during treatment showed that specific IgGl antibody levels were consistently reduced, even when patients showed no evidence of improvement based on CT scan. IgG4 specific antibody levels were significantly higher in patients from Group I (surgery and chemotherapy) and Group II (unchanged following chemotherapy). Not all CE patients, however, had raised IgG4 antibodies. In seropositive patients from the cured group (Group IV), IgG4 levels did usually reduce to zero, but over a long time-scale, and in 1 case did not fall. For the group showing improvement (Group III) patients showed a pattern of reduced IgG4 antibody responses over a

IgGl

analysis

All the patient clinical and laboratory data were analysed by EpiInfo version 5 software (Centers for Disease Control, Atlanta, USA). A non-parametric statistical test (the Kruskal-Wallis H-test) was used to test differences between patient groups and control Table 2. Specific antibodies (Ab), circulating with severe cystic echinococcosis

Group” I. Surgery + chemotherapy II. Unchanged III. Improved IV. Cure Donors

No. of samples studied

antigen

(CAg) and immune-complexes

7 9 7 7

37 54 30 55

Ab mean (SD) 2.9 (1.2) 2.4 (1.4) 2.1 (1.2) 1.53 (1.2)

30

30

0

No. of patients

IgGI mean (SD) 1.5 (1.1) 1.6 (1.2) 1.2 (1.1) 0.8 (1.2)

IgG4 mean (SD) 2.5 (2.2) 0.8 (1.1) 1.1 (1.6)

0

0

1.9 (1.7)

(CIC) in 30 patients

0.4 (1.0)

CIC mean (SD) 0.4 (0.8) 0.8 (1.3) 0.2 (0.4) 0.1 (0.3)

0

0

CAg mean WY 0.3 (0.4) 0.7 (1.2) 0.9 (1 .O)

All the results were from enzyme-linked immunosorbent assays. The OD values were converted for each individual in relation to values for the control group (donors): 0, <(x + 3 SD); 1, 2 (X + 3 SD), <(x + 4 SD); 2, 2 (x + 4 SD), <(x + 6 SD); 3, 2 (x + 6 SD), <(x + 8 SD); 4, 2 (X + 8 SD). “See text for explanation.

100

R. BONIFACINO

time span (30 months). The unchanged group II) showed no consistent pattern in IgG4 levels over a-long period of analysis. Serum IgG4 levels in the clinically improved or cured CE patients combined compared to patients from Group I and II combined were, however, statistically significantly reduced. shorter (Grouv

Specific circulating

Total white cell and lymphocyte

antigen and immune-complexes

Table 3. Proliferative cystic echinococcosis

response in vitro of peripheral and from a control group

cellular responses Lymphocytes from CE patients exhibited lower LTA responses to PHA than normal controls (Table 3). The PBMC samples from the cured patient group (Group IV) were taken after they were considered cured and showed normal responses. In the 7 patients in which surgery was combined with chemotherapy (Group I) the first LTA sample result (before surgery) was significantly lower (P < 0.002) than the subsequent response from the same group after treatment (mean 28 * 22 versus 130 rt 87, respectively) (Figure). These results are consistent with an E. granulosus-induced suppression of cellular immune responses. The levels of LTA response to PHA were significantly higher for all groups during ABZ treatment (PCO.01). Normal individuals showed little or no response to E. granulosus SHCF, as reflected in their very low stimulation indices (mean 1.1). In contrast, all of the CE patient groups showed statistically increased LTA responses to E. granulosus SHCF antigen (Table 3). In particular there were strong responses in the patients from poorresponding groups, i.e., Groups I (surgery and chemotherapy) and II (chemotherapy-unchanged) (mean SI 4.8 and 4.6; P < 0.001 and P < 0.01, respectively). In-vitro

blood mononuclear Stimulation

I. Surgery

No. of patients +

chemotherapy II. Unchanged III. Improved IV. Cure Donors

counts

The mean lymphocyte count for all samples from CE patients was 2 155 * 889/mm3 which was significantly lower than the mean value from healthy donors (Table 3). The lowest mean lymphocyte count was observed in CE patients who had both surgery and chemotherapy treatment (Group I). The total eosinophil count from all the patients was 451 + 204/mm3 (normal 305 * 45/mm3).

CAg was detected in only 20 (66.7%) CE patients; however, levels of CAg were higher in patients with the worse clinical outcome. Chronological analysis indicated (not shown) that those patients with raised CAg levels in the cured group (Group IV) subsequently exhibited the biggest reductions during the initial period of treatment. One cured patient showed persistently high CAg levels which were eventually reduced to zero after 45 months and this was consistent with the longer period of treatment needed to cure this patient. Interestingly, 2 CE patients with high CAg levels had bone CE (spine and tibia). In the group showing improvement (Group III) there were only a few patients with raised CAg levels and then only for short periods of time; most were CAg negative over the study period. This pattern was repeated in the unchanged patient group (Group II) with the exception of 1 CE patient who showed high CAg levels in most samples and this patient was the most severely affected of the group, showing no response to either ABZ or PZQ. Five patients were positive for specific CICs. Only 2 patients, however, in the whole study showed any increase in CIC levels and this was very low. No conclusions can be made from the CIC data.

Group”

ETAL.

No. of samples studied

PHA

7

34

9 7 7 30

cells from 30 patients

with severe

index White cells no./mm3

Lymphocytes no./mm’

115 (99)

Ag 4.8 (4.1)

6685 (1441)

1887 (772)

24 14 12

93 (78) 97 (67) 180 (210)

4.6 (3.8) 2.6 (2.1) 1.8 (1.1)

7276 (2026) 7246 (1327) 7240 (1350)

2124 (900) 2998 (933) 2360 (693)

30

153 (85)

1.1 (0.3)

6969 (1540)

3159 (985)

Unless otherwise stated, values are means (standard deviation in parentheses). PHA, phytohaemagglutinin; Ag, Echinococcus granulosus sheep cyst fluid at 8 pg/mL. “See text for explanation.

300 /

I

(9

(ii)

Ii

J

Figure. (i) Stimulation indices (SI) obtained with fresh blood mononuclear cells (PBMC), stimulated with phytohaemagglutinin in cystic echinococcosis (CE) patients. A, patients from Group 1, before surgery; B, patients from Group 1, after surgery; C, control blood donors. (ii) SI from PBMC stimulated with E. granulosus antigen in CE patients. A, patients from Group 1, before surgery; B, patients from Group 1, after surgery; C, control blood donors. Group 1 comprised 7 patients treatedwith albendazole and/or praziquantel as well as surgery.

IMMUNOLOGICALSURVEILLANCE OFSEVERE ECHINOCOCCOSIS The cured CE patients (Group IV) almost all responded like normal donors except for 1 patient who also had consistently elevated total Ig antibodies to E. nranuZosus SHCF. Virtuallv all of the patients who were treated by surgery and chemotherapy
I

Correlations of assays In each patient group the total anti-E. granulosus antibody levels (Ig) correlated strongly with levels of specific IgGl and IgG4 antibodies which also correlated strongly with each other. Specific IgGl and IgG4 antibody levels were also correlated with levels of circulating E. granulosus antigen in CE patients classed as improved or cured. In general, a concordance was observed between nonsuecific SI resnonses of PBMCs to PHA with those to specific hydatid antigen. This was highly statistically significant in Group I (P < 0.0 l), which was the group from which most samples were analysed by LTA (n = 34). Discussion Human CE is still predominantly treated by surgery; however, chemotherapy alone or in combination with surgery is now recognized as an important alternative (WHO, 1996). Medical treatment of CE is primarily by benzimidazole drugs, of which ABZ currently shows greatest clinical efficacy (HORTON, 1989; AA&MN &

ECKERT, 1995). Imaaina techniaues. such asultrasound scanning or CT, are ;ery uskful~ in assessment of chemotherapeutic efficacy on hydatid cysts (DE ROSA & TEGGI, 1990; TODOROV et al., 1992). Nevertheless,

sensitive immunological surveillance could provide additional follow-up data particularly when imaging techniques are unable to indicate clearly changes in hydatid cyst size, shape or integrity, particularly for severe-cases, for examnle osseous CE or disseminated CE (BONIFACINO et al., 1997). Significant group differences in immunological parameters were apparent when the clinically unimproved CE cases treated by chemotherapy (Group II) were compared to the eventually cured chemotherapy group (Group IV). These differences were manifested by lower mean total specific Ig, IgGl, and IgG4 antibody levels as well as lower CAg and CIC levels in the latter group, together with significantly lower LTA responses in vitro to hydatid antigen. There was also a trend for reduction in all these humoral and in-vitro cell parameters when the good-responder patient groups (Groups III and IV) were compared to the non-responder/aggravated groups (I and II). The patterns of serum immunoassays in some individual CE patients also provided useful data to interpret the changes occurring as a result of treatment. Total antiE. granulosus cyst fluid antibodies (Ig) were reduced to low levels (mostly to background) in the patients eventually classed as cured and thus may be a useful long-term serological marker for good response to treatment. Specific IgGl antibodies generally reduced to low or background levels in the good responders but also to some extent in all groups, including those clinically

101

unimproved. In general, mean levels of specific IgG4 antibodies declined in the clinicallv imoroved or cured groups in contrast to the 2 poor-responder groups of CE patients. RIGAN et al. (1995) observed that elevated IgG4 antibodies in ABZ-treated hepatic CE patients were correlated with poor response to chemotherapy. Furthermore, levels of specific IgG4 antibodies were found to be significantly higher in a group of untreated clinically advanced CE-patients compared to asymptomatic CE cases(SHAMBESHet al.. 1997). Tonether those findings suggest that specific IgG4 is a potemially useful marker for poor prognosis even in osseous CE. CAg was not detectable in enough CE patients to justify its recommendation for general prognostic use, although its measurement may be of value in some individual severe cases, as previously observed in hepatic CE (CRAIG, 1997). It is of interest that the 2 most severe CE natients (both with bone hvdatid) freauentlv had high-levels of serum CAg over thetreatment or follow-up period. LTAs whether non-specific (PHA-LTA) or parasite specific (SHCF-LTA) did not show strong correlations. However, in general, stronger cellular responses occurred in the poor-responding CE groups (I and II). LTAs are notoriously variable, even within replicates and obtaining significant data is difficult. Also, the low SI obtained with specific antigens can present problems of insensitivity, although values obtained in this study are good in this respect. Bone CE is almost always a primary CE infection derived from oncospheres and not from protoscoleces (i.e., secondary echinococcosis); this form of the infection exhibits a different growth pattern in the bone location where the host does not usually respond as in other parenchyma. The pericystic fibrotic capsule or ectocyst in CE is characterized by mononuclear cells, eosinophils, giant endothelial cells and fibroblasts; however, it does not usually occur in bone CE. The growth pattern of CE in bone in some ways is probably more similar to alveolar echinococcosis due to infection with E. multiloculavis. The resistant nature of osseous tissue restricts the parasite’s growth and leads to exogenous daughter cyst spread of E. granulosus along the medullary and trabecular canal with little resistance. In the spine, articular cartilage and intervertebral discs offer little resistance to hydatid cyst growth (SZYPRYT et al., 1987). This closer host-parasite interface in bone CE could explain the higher CAg levels previously observed in this group of CE patients (BONIFACINOet al., 1997). It is also possible that high levels of CAg might contribute to immunosuppression through polyclonal stimulation. Immunosuppression appeared to occur in the current study for the most severe CE patient who exhibited poor lymphocyte responses in vitro to PHA and E. granulosus cyst fluid but with concomitant medium-to-high specific IgG4 and total Ig antibody levels. Despite this patient, however, a general negative correlation between humoral and cellular responses was not found. Previously RAKHA et al. (199 1) observed a significant negative correlation between humoral and cell-mediated responses in experimental secondary CE in rodents, and such a correlation has also been observed with primary hepatic CE in sheep (JUDSONet al., 1985). All the CE cases investigated in the present study were classed as severe, in contrast to those observed in previous follow-up studies. One of the main conclusions that can be drawn from this CE post-treatment immunosurveillance study is that there is a potential problem in single-sample analysis of CE patients. Not only is there a potent effect on immune parameters during treatment, as also observed by IOPPOLO et al. (1996), but there also appears to be variation unrelated to treatment or progress of CE disease. Evaluation of CE treatment by surge& and/or chemotherapy continues to be difficult. Imaaina methods do not alwavs indicate clear changes incys’t viability and the immunological assays used in the current study were not consistently

102

R. BONIFACINO

able to define the real status of the parasite earlier than CT scan for this groun of uatients with severe CE. Nevertheless, speck a&ibo& and LTA tests broadly showed a pattern of reduced response post-treatment for those CE patients classed as improved or cured. Such laboratory

tests therefore

do provide

additional

or sup-

porting clinical data for patients with long-term-treated severe CE. Finally, the use of PZQ after poor response to ABZ was beneficial in 4 of the 6 CE patients treated in this way and therefore the role of PZQ in chemotherapy of severe CE requires further evaluation. Acknowledgements We are grateful to Professor Albert0 Nieto and his staff for their co-operation. We are also grateful to Alejandra Puglia, Ana Richieri, Ana Combo1 and Beatriz Cammbula for their technical assistance and to Dr John Horton (Smith, Kline, Beecham, UK) for supply of albendazole. We also gratefully acknowIedge financial support from a European Commission International Co-operation Research Project No. CIl *-CT93-0024. References Amman, R. & Eckert, J. (1995). Clinical diagnosis and treatment of echinococcosis in humans. In: Echinococcus and Hydatid Disease, Thompson, R. C. A. & Lymbery, A. J. (editors). Wallingford: CAB International, pp. 41 l-463. Awar, G. N., Matossian, R. M., Radwan, H. & Meshefedjian, G. A. (199 1). Monitored medico-surgical approach to the treatment of cystic hydatidosis. Bulletin of the World Health Organization, 69, 477-482. Barbieri, M., Steria, S., Battistoni, J. & Nieto, A. (1993). High performance latex reagents for hydatid serology using an Echinococcus Pranulosus liooorotein mutinen fraction vurified from cyst fl&l in one steb. international~ournalfor Farasitology, 23,565-572. Barbieri, M., Severi, M. A., Pirez, M. I., Battistoni, J. & Nieto, A. (1994). Use of soecific antibodv and circulating antigen serum levels in the-hydatid immunodiagnosis of asymctomatic population. International Journal for Parasitology, 24, 937-942. Bonifacino, R., Malgor, R., Barbeito, R., Balleste., R., Rodriguez, M. J., Botto, C. & Klug, F. (1991). Seroprevalence of Echinococcus granulosus infection in a Uruguayan rural human population. Transactions of the Royal Society of Tropical Medicine and Hygiene, 85,769-772. Bonifacino, R., Craig, I’. S., Carter, S., Malgor, R. & Dixon, J. (1993). Partial characterization of antigens in circulating immune complexes in cystic hydatid patients treated with albendazole. Transactions of the Royal Society of Tropical Medicine and Hygiene, 87, 97- 102. Bonifacino, R., Dogliani, E. & Craig, I’. S. (1997). Albendazole treatment and serologic follow-up in cystic echinococcosis of the bone. International Orthopaedics, 21, 127-132. Chinnery, J. B. & Morris, D. L. (1986). Effect of albendazole sulphoxide on viability of hydatid protoscoleces in vitro. Transactions of the Royal Society of Tropical Medicine and Hygiene, 80,815-817. Cohen, H., Paolillo, E., Bonifacino, R., Botta, B., Parada, L., Cabrera, I’., Snowden, K., Gasser, R., Tessier, R., Dibarboure, L., Wen, H., Allan, J. C., Soto de Al Faro, H., Rogan, M. T. & Craig,P. S. (1998). Humancysticechinococcosisin a Uruguayan community: a sonographic, serologic and epidemioloaical studv. American 7oumal of Tropical Medicine and Hygiene;59, 620--627. Coltorti, E. A. (1986). Standardization and evaluation of an enzyme immunoassay as a screening test for the seroepidemiology of human hydatidosis. American Journal of Tropical Medicine and Hygtene, 35, lOOO- 1005. Coltorti, E. A., Fernandez, E., Guarnera, E., Lago, J. & Iriarte, J. (1988). Field evaluation of an enzvme immunoassav for dete&on of asymptomatic patients in a hydatid control proaramme. American Journal of Tropical Medicine and Hy&iene, 38, 603-607. -

ET/X.

Craig, I’. S. (1986). Detection of specific circulating antigens, immunecomplexes and antibodies in human hydatidosis from Turkana (Kenya) and Great Britain, by enzyme-immunoassay. Parasite Immunology, 8, 171-188. Craig I’. S. (1997). Immunodiagnosis of Echinococcus granulosus. In: Compendium on Cystic Echincoccosis in Africa and in Middle Eastern Countries with Special Reference to Morocco, Anderson, F. L. (editor). Provo: Brigham Young University Print Services, UT84602, pp. 85-l 18. Craig, P. S., Rogan, M. T. & Allan, J. C. (1996). Detection, screening and community epidemioiogy of taeniid cestode zoonoses: cystic echinococcosis, alveolar echinococcosis and neurocysticercosis. Advances in Parasitology, 38, 169-250. De Rosa, R. & Teggi, A. (1990). Treatment of Echinococcus granulosus hydatid disease with albendazole. Annals of Tropical Medicine and Parasitology, 84, 467-472. Gottstein, B., Mesarina, B., Tanner, I., Ammann, R., Wilson, J. F., Eckert, J. & Lanier, A. (199 1). Specific cellular and humoral responses in patients with different long-term courses of alveolar echinococcosis (infection with Echinococcus multilocularts) . American yournal of Tropical Medicine and Hygiene, 45,734-742. Horton, R. J. (1989). Chemotherapy of Echinococcus infection in man with albendazole. Transactions of the Royal Society of Tropical Medicine and Hygiene, 83, 97- 102. Ioppolo, S., Notargiacomo, S., Frofumo, E., Franchi, C., Ortona, E., Rigano, R. & Siracusano, A. (1996). Immunological responses to antigen B from Echinococcus granulosus cyst fluid in human hydatid patients. Parasite Immunology, 18, 571-578. Judson, D. G., Dixon, J. B., Clarkson, M. J. & Pritchard, J. (1985). Ovine hydatidosis: some immunological characteristics of the seronegative host. Parasitology, 91,349-357. Matossian, R., Awar, G. N., Radwan, H., Craig, I’. S. & Meshefedjian, G.A. (1992). Immune status during albendazole therapy for hydatidosis. Annals of Tropical Medicine and Parasitology, 86, 67-75. Rakha, N. K., Dixon, J. B., Skerrit, G. C., Carter, S. D., Jenkins, P. & Clarke-Marshall, I. (1991). Lymphoreticular responses to metacestodes: Taenia multiceps (Cestoda) can modify interaction between accessory cells and responder cells during lymphocyte activation. Parasitology, 102, 133- 140. Rigano, R., Profumo, E., Ioppolo, S., Notargiacomo, S., Ortona. E.. Tenei. A. & Siracusano. A. (1995). Immunological markers mdicating the effectiveness‘of pharmacological treatment in human hydatid disease. Clinical and Experimental Immunolo&y, 102,281-285. Shambesh, M. K., Craia, I’. S.. Iven, H., Roaan, M. T. & Paolillo,-E. (1997). IgGl andIgG4 se-rumantibodyresponses in asymptomatic and clinically expressed cystic echinococcosis in patients. Acta Tropica, 64, 53-63. Siracusano, A., Teggi, A., Quintieri, F., Notargiacomo, S., de Rosa, F. & Vicari, G. (1988). Cellular immune responses of hvdatid vatients to Echinococcus eranulosus antieens. Clinical and Expdrimental Immunology, 72,400-405. n Szvptyt, E. I’., Morris, D. L. & Mulholland, R. C. (1987). Combined chemotherapy and surgery for hydatid bone disease. journal ofBone and Point Suraerv. 69. 141-144. Todorov, T., Mechkov, G., Vu?ova, K.,-G&g&, I’., Lazarova, I., Tonchev, Z. & Nedelkov, G. (1992). Factors influencing the response to chemotherapy in human cystic echinococcosis. Bulletin ofthe World Health Organization. 70. 347-358. Wen, H. & Craig, I’. S. (1994). 1gGs;bclass responses in human cystic and alveolar echinococcosis. American Journal of Tropical Medicine and Hygiene, 51, 741-748. WHO (1996). Informal Working Group on Echinococcosis. Guidelines for treatment of cystic and alveolar echinococcosis in humans. Bulletin of the World Health Organization, 14,23 l242.

Received 12 March 1999; revised 25 August 1999; accepted for publication 7 September 1999