Noninvasive Diagnosis of Visceral Leishmaniasis: Development and Evaluation of Two Urine-Based Immunoassays for Detection of Leishmania donovani Infection in India

Abstract

Background: Visceral Leishmaniasis (VL), a severe parasitic disease, could be fatal if diagnosis and treatment is delayed. Post kala-azar dermal leishmaniasis (PKDL), a skin related outcome, is a potential reservoir for the spread of VL. Diagnostic tests available for VL such as tissue aspiration are invasive and painful although they are capable of evaluating the treatment response. Serological tests although less invasive than tissue aspiration are incompetent to assess cure. Parasitological examination of slit-skin smear along with the clinical symptoms is routinely used for diagnosis of PKDL. Therefore, a noninvasive test with acceptable sensitivity and competency, additionally, to decide cure would be an asset in disease management and control.

Methodology/principal findings: We describe here, the development of antibody-capture ELISA and field adaptable dipstick test as noninvasive diagnostic tools for VL and PKDL and as a test of cure in VL treatment. Sensitivity and specificity of urine-ELISA were 97.94% (95/97) and 100% (75/75) respectively, for VL. Importantly, dipstick test demonstrated 100% sensitivity (97/97) and specificity (75/75) in VL diagnosis. Degree of agreement of the two methods with tissue aspiration was 98.83% (κ = 0.97) and 100% (κ = 1), for ELISA and dipstick test, respectively. Both the tests had 100% positivity for PKDL (14/14) cases. ELISA and dipstick test illustrated treatment efficacy in about 90% (16/18) VL cases when eventually turned negative after six months of treatment.

Conclusions/significance: ELISA and dipstick test found immensely effective for diagnosis of VL and PKDL through urine samples thus, may substitute the existing invasive diagnostics. Utility of these tests as indirect methods of monitoring parasite clearance can define infected versus cured. Urine-based dipstick test is simple, sensitive and above all noninvasive method that may help not only in active VL case detection but also to ascertain treatment response. It can therefore, be deployed widely for interventions in disease management of VL particularly in poor resource outskirts.

Fig 1. Comparison of urine antibody absorbances between VL patients and healthy controls using four leishmanial antigens.

Reactivity of urine (undiluted) from visceral leishmaniasis (VL) patients (n = 5) and healthy controls (HC) (n = 6) against 2.5 μg/well of leishmanial membrane antigen, LAg; recombinant glycoprotein 63, rGP63; recombinant cysteine protease A, rCPA and soluble leishmanial antigen, SLA. The standard error of mean for each group is shown as a solid line.

Fig 2. ELISA to determine relative antibody isotype titres in urine of VL patients and negative controls.

Graph showing comparative optical density values of LAg-specific antibody isotypes; IgG, IgA, IgM, and IgE present in the urine samples of VL patients (n = 5), healthy controls (HC) (n = 3) and other diseases (OD) (n = 4) at 492nm.

Fig 3. ELISA results with urine samples obtained from VL and PKDL patients, healthy individuals and other clinical groups.

Comparative evaluation of ELISA reactivity of anti-Leishmania antibodies in urine against LAg at 0.1875 cut-off. The study groups were composed of a panel of VL patients (VL; n = 97), PKDL patients (n = 14), endemic healthy controls (EHC; n = 25), non-endemic healthy controls (NEHC; n = 25) and other diseases (OD; n = 25) including malaria (n = 2), viral fever (n = 2), tuberculosis (n = 4), typhoid (n = 4), liver abscess (n = 3), systemic lupus erythematosus (SLE) (n = 3), sepsis (n = 2), pancreatitis (n = 1), pneumonia (n = 1), filariasis (n = 1), abdominal pain and loose motion (n = 1), and mouth ulcer (n = 1). Each sample was tested in duplicate and the mean was taken. Each dot represents mean of a single sample.

Fig 4. Receiver operator characteristics (ROC) curve for urine ELISA.

ROC curve obtained from the ELISA values for the detection of anti-Leishmania IgG antibodies in urine samples. Cut-off value (0.1875), sensitivity (97.94%), specificity (100%) and AUC (0.9984) were determined by this curve using GraphPad Prism software (version 5.0).

Fig 5. Comparison of urine antibody titres of VL patients at different time points of treatment.

Absorbance value of LAg-specific urine antibody titres in VL at day 0, before the treatment started; at day 30, after drug therapy and at days >180, more than 6 months of treatment started.

Fig 6. Dot blot assay to identify L. donovani infection.

Results showing dot blot assay with VL; 1–10, malaria; 11 and 12, viral fever; 13 and 14, tuberculosis; 15 and 16, typhoid; 17, endemic healthy control; 18 and 19 and non-endemic healthy control; 20 and 21.

Fig 7. Representative image of dipstick immunochromatographic test.

Dipstick assay shows positive results with double band at the test and control lines and negative results with bands only at the control line.