Pre- and post-treatment antibody levels in visceral leishmaniasis

Pre- and post-treatment antibody levels in visceral leishmaniasis

TRANSACTIONS OF THE ROYALSOCIETYOF TROPICALMEDICINEAND HYGIENE(1990)84, 673-675 Pre- and post-treatment Asrat Hailu Institute antibody of Pathobi...

339KB Sizes 6 Downloads 136 Views

TRANSACTIONS OF THE ROYALSOCIETYOF TROPICALMEDICINEAND HYGIENE(1990)84, 673-675

Pre- and post-treatment Asrat

Hailu

Institute

antibody

of Pathobiology,

673

levels in visceral leishmaniasis

Addis Ababa

Abstract

Enzyme-linked immunosorbent assay(ELISA) and the direct agglutination test (DAT) were employed to test sera obtained from a visceral leishmaniasis (VL) endemic area, the Aba-Roba focus in south-west Ethiopia. Tlurty sera of untreated VL patients, 37 sera 6-90 months after treatment, 18 sera from endemic controls, 8 sera from non-endemic controls and 23 serafrom patients with other diseases(schistosomiasis, tuberculosis, cutaneous leishmaniasis) were tested. Based on ELISA, the percentages negative were found to be 50% up to one year and 89% from 2 to 8 years after treatment. In contrast, these rates basedon DAT were 0% in one year and 33% from l-8 years after treatment. The relevance of persisting antibodies in the kinetics and profile of antibody production during treatment is discussed. The use of ELISA in the evaluation of clinical prognosis and patient follow-up is recommended. Se&m from a diffuse cutaneous leishmaniasis patient cross-reacted with the DAT and ELISA for VL. Introduction

Enzyme-linked immunosorbent assay(ELISA) and the direct agglutination test (DAT) are the two most important serological tests used in seroepidemiological -surveys and- clinical serodiagnosis of visceral leishmaniasis (VL) (HOMMEL et al., 1978; Ho et al., 1983; JAHN & DIESFIELD, 1983; BADARO et al., 1986; HARITH et al.. 1986). Although there have been attempts to improve and simplify these tests (PAPPAS et al., 1983; HARITH et al., 1988; JAHN & DIESFIELD, 1983), they have not yet been universally adopted as standard procedures. This is partly due to failure to identify a common immunogenic fraction in Leishmania (HOMMEL et al., 1978). No one test has yet proved superior to the others when sensitivity, specificity, simplicity and cost are all considered. The potential use of ELISA and DAT in early detection of asymptomatic and subclinical diseasealso needs evaluation. In this study, the antibody levels after specific antimonial treatment were examined using ELISA and DAT. Materials and Methods Subjects. Patient sera were obtained from venous

blood. Thirty serawere obtained from patients before treatment, and 8 sera 6-15 months, 9 sera 16-35 months, 10 sera 36-55 months, and-another 9 sera 56-90 months after treatment. Eight sera from persons living in a non-endemic area, i8 samplesfrom apparently healthy individuals in the endemic area, and 23 sera of individuals with schistosomiasis, cutaneous leishmaniasis or tuberculosis were included as controls. Post-treatment sera were obtained from patients considered as clinically cured, except for one case. The criteria of clinical cure were: regression of hepatosplenomegaly, rise in haemoglobin level, rever-

University,

Ethiopia

sion of total and differential cell counts to normal, and increase in body weight and appetite. Pauents’ subjective responses were also considered. I?n.zyme-linked immunosorbent assay. Polyvinyl chloride microtitre mates (cataloeue no. 655061. Greiner, Federal Republic of Germany) were coated with the supernatant from lysed promastigotes of L. donovani prepared by 5 rapid freeze-thaw cycles using dry ice/methanol and a 37°C waterbath. The strain was originally identified as L. donovani sensu lato by restriction endonuclease analysis, Southern blotting, and using recombinant deoxyribonucleic acid (DNA) probes selected from an L. infantum cDNA library (VAN EYS et al., 1989). The optimum concentrations of antigen and serum were determined by chequerboard titration using known positive and negative sera. Serawere used at l/800 dilution in 0.05% Tween 20 in nhosnhate-buffered saline (PBS), nH 7.2. Plates were coated using 0.5 pg protein’per well in 0.05 M carbonate/bicarbonate buffer. DH 9.6. for one hour in a 37°C waterbath. The conjugate (horseradish peroxidase-labelled goat anti-human immunoglobulin G, heavy+light chains (catalogue no. 75021, Pasteur Institute) was diluted to l/1000 with serum diluent. Blocking was done with 1% bovine serum albumen in 0.01 M PBS, pH 7.2. The plates were washed 4 times with 0.05% Tween 20 in PBS except after antigen coating, when washing was done with PBS alone. All incubations were done in a 37°C waterbath for one hour. 3’,3,5’,5-tetramethylbenzidine (TMB; Sigma, T2885) was used as a substrate at 0.6 mg/ml in equal volumes of methanol and O-1 M citratephosphate buffer, pH 5. The substrate was activated by adding 5 p.l of 30% Hz02 in 10 ml of TMB immediately after use. Optical density readings were obtained at 405 mn in a Titertek@J MCC Scanner. The cut-off value (0.75) was determined as the mean+3-09 standard deviations of the optical densities of control sera. Direct agglutination test. This test was performed using antigen kindly provided by Dr Harith (batch of 7/6/89), basically as described by HARITH et al. (1988). The serum diluent contained 0.8% %-mercaptoethanol and 0.1% gelatin in 0.15 M NaCl aqueous solution. Each serum was tested in duplicate using neat serum and serum mixed with 50°i, glycerol in V-shaped, 96-well microtitre mates (Greiner). A dilution of l/3200 was taken as the cut-off t&e. Data analysis. Data were grouped into 5 categories: Co=non-endemic control sera, Ci=endemic control sera (other infections included), P,,=pre-treatment patient sera, Pi=post-treatment patient sera (0.5-3 years), and Pr=post-treatment patient sera (3-7-5 years). Student’s t test and ANOVA (SNEDECOR & COCHRAN, 1968) were both used to examine the significance of differences between the mean values of the groups. In computing the meansand variances, all titres above 1:102 400 and ontical densities above 2.0 were taken as 1:102 400 and>*0 respectively. Similarly, titres less than 1:100 were taken as 1:lOO.

674 Results Using the cut-off value for optical density readings in the ELISA of 0.75, 3 patients’ sera were classified as negative (Figure. a), a sensitivitv of 88% and specif~ity of‘ l&t%‘(Table 1). If the cut-off was lowered to an optical density of 0.6, the sensitivity could be brought to 100% with a risk of lowering specificity to 94%. In the DAT, a change of cut-off titre from 1:3200 to 1:6400 or 1: 1600 did not affect specificity or sensitivity (Figure, b). The analysis of variance on ELISA readings showed significant differences (F=92*86, PO*Ol). A similar analysis of the DAT values also showed a significant difference (F=90*34, P
ELISA Positive Negative 22

3

DAT Positive

Negative

30

0

0

8

0

8

0 0 0 0 1 23

18 13 8 1 0 51

0

18

0 0 1 31

8 1 0 35

88 100

. . . . .. . .

Table 2. Results of enzyme-linked immunosorbent assay (ELBA) and direct agglutination test (DAT) 011 sera of visceral leishmaniasis patients at different times after treatment ELISA’ Negative Positive

Months after treatment

4 (50) 1(11)

4 (50) 8 (89)

36-55

1w

9 (90)

6 (67) 5 (45)

56-90 Total

1(11) 7 (19)

8 (89) 29 (89)

8 (89) 27 (73)

“Numbers

Negative

8 (100)

0 (0)

3 (37) 6 (55) l(lU 10 (67)

in parentheses are percentages.

Table 3. Summary of enzyme-linked immunosorbent assay (ELBA) and direct agglutination test (DAT) results Controls Non-endemic

ELBA

DAT”

0.25kO.06 (8)

0.38kO.74 (8)

0.33kO.15 (39) 0.31kO.79 (26)

Endemi?’ Patients

cut-

DAY Positive

615 16-35

Group

100 100

“LCL=localized cutaneous leishmaniasis patient, DCL=diffuse aneous leishmaniasis patient.

there was a significant difference between group Pr and the controls (Co and C,) when examined by the least significant difference method at the 5% level of significance (Table 3). Antibody titre after treatment declined gradually only with the ELISA (Figure). Comparisons of DAT and ELISA results with posttreatment test sera are shown in Table 2, and Table 3 summarizes the results of both tests. On average, 10% of the sera obtained after 15 months of treatment were positive by ELISA. The rate for the same group by DAT was 67%. After one year, ELISA titres fell from 50% to 11%. Of 29

91+70+0.97 (31) 7 .40f3.60 (17) 6.00+4.;0 (20)

1.56f0.49 (25) Untreated Treated (0.5-3 years) 0*60+0.29 (17) Treated (3-7.5 years) 0.58?$;: (19) Cut-off value Least significant 0.41 difference (OL=5%) 92.86 Variance ratio (F)
3.16 90.34 to.01

“ELISA values are optical densities; DAT values are transformed reciprocal titres (transformation factor=log* all results expressed as [reciprocal titrel/lOO]; mean&variance (number tested in parentheses). bIncluding those with other infections.

:::F

PI02 400 g 51200 ‘= 25600 E 12800 B 6 6400 5 3200 _ 1600 1 : 800 b ‘loo ‘D 2

. .. . . .. *

........

. . . _

-

.

,::: : l*

-

-

--

-

. . ..

-, .

L&.:: CO

-

Cl Serum

.

.-

.....

PO PI groups

P2

b

Fig. 1. Pre- and post treatment antibody titres in visceral leishmaniasis patients and controls, based on (a) enzyme-linked immunosorbent assay and (b) direct agglutination test. Serum groups: C,,, non-endemic controls; C,, endemic controls+other infections; P0, untreated visceral leishmaniasis patients; PI, 0.5-3 years after treatment; P2, s7.5 years after treatment. The dashed lines indicate the cut-off values.

675 post-treatment sera which proved negative by ELISA, 20 had positive DAT titres. Ten which were positive by DAT were all negative by ELISA. Only with 7 post-treatment sera did the results agree in both tests. Two sera, one from localized cutaneous leishmaniasis and another from diffuse cutaneous leishmaniasis (DCL), both due to L. aerhiopicu, had ELISA optical densities of 0.36 and 0.84 and DAT titres of cl:100 and >1:191 400 respectively. Only the DCL patient’s serum showed a titre as high as those of VL patients. One post-treatment patient’s serum in group PI (Figure? a) had an exceptionally high OD value, 1.42. This patient had a slightly palpable spleen during blood collection. Six other patients with an OD value above 0.75 (Figure, a) did not have any gross clinical abnormalities. Discussion

In both the ELISA and the DAT it was shown that endemic control sera were not strictly necessary. This was auite obvious. esDeciallv with the DAT when using-2-mercaptoeihhanbl in the serum diluent, which is assumed to increase specificity and sensitivity (HARITH et al., 1988). The DAT detected a high level of anti-leishmanial antibodies long after treatment, even in patients considered to be clinically cured. Unlike the DAT antibodies, the ELISA antibodies declined progressively with tnne after treatment. This observation has also been reDorted bv TAHN & DIESFIELD (1983) in Kenyan Vi patients e&mined

bv the ELISA and bv NEOGY et al. (1987) in India iith the indirect haemagglutinatiod tesi. In this study, however, a few 07~the post-treatment sera

(10%) had high OD values. One of these uatients had

an e&eption&y

high OD, and at the t&e of blood

collection was considered as incompletely cured. As thje criteria of cure are ill-defined it is difficult to explain why some of the treated patients had high OD values without obvious clinical signs of disease. Notwithstanding these exceptions, the ELISA should prove useful in assessingcure and seroconversion, and for patient follow-up. Seventy percent of the post-treatment sera negative by ELISA were positive by DAT. The DAT demonstrated high levels of antileishmanial antibodies until 75 years after treatment. One can only speculate why DAT antibodies linger so long after treatment.

It is

possible that (i) agglutination antibodies have a longer half-life than antibodies detected by ELISA;

(ii) a

subclinical infection could have persisted, eliciting agglutinating antibodies, treatment having not been sufficient to result in parasitological cure; or (iii) treated patients had been re-exposed to Leishmania in the endemic area. It would be interesting to know how the DAT performs with sera from subclinical or preclinical subjects. Its potential use in the detection of subclinical diseaseor asymptomatic casesneeds to be explored. Also the potential use of the DAT in

immunological

studies needs attention.

Acknowledgements

I am very grateful to Dr A. E. Harith, who kindly provided the DAT antigen. My sincere thanks also go to S. Kuiper, Dr G. J. J. M. Van Eys, Dr A. H. Kolk, and G. S. Linghart, all from the Swellengrebel Laboratory of Tropical Hygiene at the Royal Tropical Institute in Amsterdam. I am also grateful to Dr P. Wright, Dr J. Leeuwenburg and I. Yo for their helo and assistance. This study received financial support partly from the International Development Research Centre (IDRC) and also from the UNDP/World BankWHO Special Programme for Research and Training in Tropical Diseases.

References

Badaro, R., Jones,T. C., Lorenco, R., Cerf, B. J., Sampaio, D., Carvalo, E. M., Rocha, H., Teixeira, R. & Johnson, W. D., Jr (1986). A prospective study of visceral leishmaniasis in an endemic area of Brazil. Journal of Infectious Diseases. 154. 639-649.

Harith, A. E., Kolk, 6. H.; Kager, P. A., Leeuwenburg, J., Mugai, R., Kiugu, S. & Laarman, J. J. (1986). A simple and economical direct agglutination test for serodiagn&is and sero-epidemiological studies of visceral leishmaniasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 80, 583-586.

Harith, A. E., Kolk, A. H., Leeuwenburg, J., Muigai, R., Huigen, E., Jelsma, T. & Kager? P. A. (1988). Improvement of a direct agglutinauon test for field studies of visceral leishmaniasis. Journal of Clinical Microbiology, 26, 1321-1325. Ho, M., Leeuwenburg, J., Mbugna? G., Wamachi, A. & Voller, A. (1983). An enzyme-lmked immunosorbent assay (ELISA) for field diagnosis of visceral leishmaniasis. American Journal of Tropical Medicine and Hygiene, 32, 943-946.

Hommel, M., Peters, W., Ranque, J., Quilici, M. & Lanotte, G. (1978). The micro-ELISA technique in the serodiagnosisof visceral leishmaniasis.Annals of Tropical Medicine and Parasitology, 72, 213-218. Jahn, A. & Diesfield, H. J. (1983). Evaluation of a visually read ELISA for serodiagnosis and seroepidemiological studies of kala-azar in the Baring0 District, Kenya. Transactions of the Royal Society of Tropical Medicine and Hygiene, 77, 451-454.

Neogy, A. B., Nandy, A., Ghosh Dastidar! B. & Chowdhury, A. B. (1987). Antibody kinetics 111kala-azar in response to treatment. Annals of Tropical Medicine and Parasitology,

81, 727-729.

Pappas, M. G., Hajkowski, R. & Hockmeyer, W. T. (1983). Dot enzyme-linked immunosorbent assay (DOT-ELISA): a micro technique for the rapid diagnosis of visceral leishmaniasis.Journal of Immunological Methods, 64. 205-214.

Snedecor, G. W. & Cochran, W. G. (1968). Statistical Methods, 6th edition. New Delhi: Oxford & IBH Publishing Company. Van Eys, G. J. J. M., Schoone,G. J., Ligthart, G. S., Alvar, J., Evans, D. A. & Terpstra, W. J. (1989). Identification of Old World Leishmania bv DNA recombinant nrobes. Molecular and Biochemical
21 December 1989; accepted 1990

for publication

6