Xenodiagnosis to evaluate the infectiousness of humans to sandflies in an area endemic for visceral leishmaniasis in Bihar, India: a transmission-dynamics study

Summary Background Visceral leishmaniasis, also known on the Indian subcontinent as kala-azar, is a fatal form of leishmaniasis caused by the protozoan parasite Leishmania donovani and transmitted by the bites of the vector sandfly Phlebotomus argentipes. To achieve and sustain elimination of visceral leishmaniasis, the transmission potential of individuals exposed to L donovani from across the infection spectrum needs to be elucidated. The aim of this study was to evaluate the relative infectiousness to the sandfly vector of patients with visceral leishmaniasis or post-kala-azar dermal leishmaniasis, before and after treatment, and individuals with asymptomatic infection. Methods In this prospective xenodiagnosis study done in Muzaffarpur district of Bihar, India, we included patients with clinically confirmed active visceral leishmaniasis or post-kala-azar dermal leishmaniasis who presented to the Kala-Azar Medical Research Center. These participants received treatment for L donovani infection. We also included asymptomatic individuals identified through a serosurvey of 17 254 people living in 26 high-transmission clusters. Eligible participants were aged 12–64 years, were HIV negative, and had clinically or serologically confirmed L donovani infection. During xenodiagnosis, the forearms or lower legs of participants were exposed to 30–35 female P argentipes sandflies for 30 min. Blood-engorged flies were held in an environmental cabinet at 28°C and 85% humidity for 60–72 h, after which flies were dissected and evaluated for L donovani infection by microscopy and quantitative PCR (qPCR). The primary endpoint was the proportion of participants with visceral leishmaniasis or post-kala-azar dermal leishmaniasis, before and after treatment, as well as asymptomatic individuals, who were infectious to sandflies, with a participant considered infectious if promastigotes were observed in one or more individual flies by microscopy, or if one or more of the pools of flies tested positive by qPCR. Findings Between July 12, 2016, and March 19, 2019, we recruited 287 individuals, including 77 with active visceral leishmaniasis, 26 with post-kala-azar dermal leishmaniasis, and 184 with asymptomatic infection. Of the patients with active visceral leishmaniasis, 42 (55%) were deemed infectious to sandflies by microscopy and 60 (78%) by qPCR before treatment. No patient with visceral leishmaniasis was found to be infectious by microscopy at 30 days after treatment, although six (8%) were still positive by qPCR. Before treatment, 11 (42%) of 26 patients with post-kala-azar dermal leishmaniasis were deemed infectious to sandflies by microscopy and 23 (88%) by qPCR. Of 23 patients who were available for xenodiagnosis after treatment, one remained infectious to flies by qPCR on the pooled flies, but none remained positive by microscopy. None of the 184 asymptomatic participants were infectious to sandflies. Interpretation These findings confirm that patients with active visceral leishmaniasis and patients with post-kala-azar dermal leishmaniasis can transmit L donovani to the sandfly vector and suggest that early diagnosis and treatment could effectively remove these individuals as infection reservoirs. An important role for asymptomatic individuals in the maintenance of the transmission cycle is not supported by these data. Funding Bill & Melinda Gates Foundation.

: Infectiousness of PKDL patients to sand flies at pre-and post-treatment………… …8 Table S4: Cluster wise summary of total population, number of subjects participation in Serosurveys and in xenodiagnosis……………………………………………….. …9 Figure S1: Flow diagram of xenodiagnosis study………………………………………………10 Figure S2. Splenic scores and parasitemias in VL patients…………………………………….11 Figure S3: Study clusters Population pyramid…………………………………..…………. …..12 Figure S4: IGRA in asymptomatic subjects…………………………………………………......13 Figure S5: Average number of sand flies used, blood fed, and dissected following feeds on asymptomatic subjects during xenodiagnosis…………………………………………14 Each study patient was informed both orally and in writing (in English and Hindi) about the nature of the study, the anticipated risks and benefits, the discomforts to which the patient will be exposed, and their right to discontinue participation at any time of their own free will. The patients who gave written consent were enrolled in the study. In case of illiterate subjects, a thumb print in addition to signature of an independent witness was obtained. For minors under the age of 18 years, informed consent was obtained from a parent or guardian. Subjects aged <13 were excluded from the study based on recommendation of the project Scientific Advisory Committee (SAC) in order to avoid any risk associated to sand fly exposure and community perception. Subjects who were HIV-positive were also excluded from the study as per guidelines of National AIDS Control Organization (NACO).

Asymptomatic subjects and selection
Asymptomatic subjects were screened from 26 high-transmission clusters (total households =2904) with a total population of 17254 persons (48% female and 52% male), in which 167 VL cases were reported over a 3-year period (2013-2015) ( Figure S1 and Table S4) [Appendix pp [6][7][8][9]. We conducted two rounds of house to house surveys during which we collected blood samples on filter paper from all consenting individuals aged 12 years and above. Samples were tested for antileishmania antibodies by Direct Agglutination Test (DAT) and rK39 ELISA. In the first sero-survey, subjects positive by both DAT and rK39-ELISA (≥ 1:1600 and ≥ 14 percentage positivity (PP), respectively), and highly seropositive by one or both assays (≥1:25,600 and ≥ 23 PP; see Appendix 1 for how these titers were determined) and who met the other inclusion criteria described above, were invited to KAMRC within 14 days of identification for participation in the xenodiagnosis study. Subjects who were negative for both serology tests in the first sero-survey were retested in the following year. The seroconverters who tested moderately positive or higher on both of assays were invited to enroll in the study. During the second serosurvey, 5378 samples of finger prick blood were collected, and 499 rK39-ELISA seroconversion and 263 DAT seroconversions were documented. Among the rK39 ELISA seroconverters, 76.6% (382/499) subjects were moderately positive and 23.4% (117/499) subjects were strongly positive. Similarly, among the DAT seroconverters, 70.3% (185/263) subjects were moderately positive and 29.6% (78/263) subjects were strongly positive by DAT. A total of 83 asymptomatic subjects from the second serosurvey who met the other inclusion criteria participated in xenodiagnosis.
A 5 ml sample of heparinized venous blood was collected from all 184 asymptomatic subjects, and whole blood/ serum/ plasma samples were stored in aliquots for qPCR analysis. All asymptomatic subjects were monitored monthly for 24 months after enrollment in the study to observe any development of active VL.

Xenodiagnosis protocol
Sand flies from the 8 th to 25 th colony generations were used in the direct xenodiagnosis study due to insufficient number of flies in earlier generations to start the study. During xenodiagnosis, 3 -4 days old flies were starved for 12 hrs prior to use. Infectivity of the different subject groups was tested with direct feeding of 30 -35 females and 10-12 males for 30 min on each site on patients' forearm and lower leg, or the forearm only. For the PKDL patients, sand flies were also exposed on nodular and macular lesions. Flies were loaded into a 2-inch diameter polycarbonate feeding chamber with screenmesh bottom through which the flies fed, and vented on the top to prevent moisture condensation that might entrap flies inside the chamber. The loaded feeding chamber was strapped to the arm or leg by a Velcro strap. After feeding, the flies were released from the feeding chamber into a polycarbonate holding cage where they were segregated (fed and unfed). Blood-engorged female flies were placed in separate 300-ml paper holding containers and held for at least 60 -72 hrs. in an environmental cabinet at 28°C and 85% humidity. The flies were dissected individually 3-5 days post feeding and infection rates determined by microscopic examination of dissected midguts. After microscopy, midgut material on the slides was transferred to 1.5 ml centrifuge tubes containing PBS. Midgut homogenates from all of the flies fed on the same site were pooled, and parasite load estimates were made by real-time quantitative PCR (qPCR), as described below. Blood fed flies that died prior to the time of dissection were pooled separately and analyzed by qPCR. A subject was considered positive for infectivity to sand flies if an infection was observed in one or more individual flies by microscopic exam, or in one or more of the pools of flies by qPCR analysis. The percent blood fed flies infected and the qPCR values were determined as secondary end points. To get an upper confidence limit on the proportion of asymptomatic subjects that are infective, we use exact binomial methods, assuming the fly dissecting assay has sensitivity of 100%. The 100% sensitivity assumption is a convenience assumption. The fly dissecting assay is meant to determine if there are any viable Leishmania parasites in the flies, and there is no other gold standard assay to test it against. The qPCR may detect DNA from non-viable Leishmania, so it is not suitable.

Laboratory tests Direct Agglutination Test (DAT):
DAT was performed using finger prick blood collected on Whatman filter paper as described elsewhere. 1 Briefly, 100 μl blood filter paper eluate (1:400 dilutions) were serially diluted up to 1:51200 with 50 μl DAT buffer in V-bottom well microtiter plates with one positive and one negative control. Fifty μl of DAT antigen (ITM, Belgium) was then dispensed to every well and plates were incubated overnight at room temperature for agglutination. 2 The DAT results were observed against a white background. Samples with a titer of 1: 1600 or above were considered DAT seropositive. For the purpose of enrollment of asymptomatic subjects in xenodiagnosis DAT seropositives were further subdivided into moderately seropositive (DAT titer ≥ 1:1600 to ≤ 1: 25600) and highly seropositive (DAT titer ≥1:25,600), as only the later were found to be at increased risk of disease. 3 rK39 Enzyme-Linked Immunosorbent Assay ELISA against rK39 antigen was performed as described previously. [4][5][6] Briefly, high binding flat bottom 96-well Nunc ELISA plates (Thermo Fisher Scientific, USA) were coated with rK39 (25ng/well) overnight at 4 0 C. Plates were blocked with PBS containing 1% (wt/vol) bovine serum albumin (BSA) (VWR, Life Science, USA) for 2 hrs at room temperature. One hundred µl of eluted blood from 5 mm whatman filter paper was added and incubated for 30 minutes. The wells were washed four times with PBS-Tween (PBS-T) and incubated for 30 minutes with protein G-horseradish peroxidase (1:32000 dilution; Thermo Fisher Scientific, USA) in PBS containing 0.1% BSA and 1.0% Tween-20. Plates were washed four times in PBS-T and incubated with 100 μl tetramethyl benzidine (TMB) / Hydrogen peroxide (H2O2) (Bangalore Genei, India) substrate for another five minutes. The reaction was stopped by addition of 1N H2SO4 and optical density (OD) measurements were taken at 450nm using a micro-titer plate ELISA reader (Molecular Devices, USA). A positive and negative control (filter paper pooled eluates from VL patients and non-endemic healthy controls, NEHC) were run in each plate, and the positive control was used as a reference to calculate a relative value of positivity of each sample, expressed as percentage positivity (PP). Samples having ≥ 14PP were considered positive. ELISA results were again subdivided into rK39 moderately positive (≥ 14 PP to 23 PP) and rK39 strongly positive (> 23 PP).

Real-Time Polymerase Chain Reaction (qPCR) to measure parasite load in blood, skin biopsy and sand fly midguts
Quantification of parasites in whole blood and/or buffy coat, skin biopsy and sand fly midguts was done by Real-time Polymerase Chain Reaction (qPCR), as described elsewhere. [8][9][10] Briefly, DNA was extracted from 200 µl of heparinized venous blood, skin tissue biopsy and dissected midguts (in pooled) using QIAamp DNA blood and tissue mini kit (QIAGEN, GmbH) and QIAamp DNA Investigator kit (QIAGEN, GmbH), following the manufacturers recommendations. DNA concentration of each sample was measured by NanoDrop ND 2000 (Thermo Fisher Scientific, USA). TaqMan based qPCR on each DNA sample was run in triplicate on an Applied Biosystems (ABI) 7500 Real Time PCR system (Thermo Fisher Scientific, USA) to amplify the region of L. dovovani kinetoplast minicircle DNA. Nuclease free water (Thermo Fisher Scientific, USA) as well as blood DNA from NEHC and DNA from uninfected laboratory reared sand fly midguts (pooled) were used as negative controls. Absolute quantification of parasite numbers in the test samples were calculated with the specific set of standard samples (DNA from healthy human blood and uninfected sand fly midguts spiked with serial dilution of cultured Leishmania parasites) run in parallel to each set of test samples, as described previously 11 . The average qPCR value for each sand fly pool was calculated as pool qPCR value divided by the number of flies in the pool. qPCR threshold ΔCt value was ≤ 35. Above 35 Ct value was considered as qPCR negative. VL subjects were assigned to 5 groups based on blood qPCR value (parasite genomes / ml of blood): 1 = qPCR 0-10; 2 = qPCR 10-100; 3 = qPCR 100-1000; 4 = qPCR 1000-10000; 5 = qPCR 10000-100000. Six drug treated cured VL patients had peripheral blood qPCR positive with values just above threshold (1.0-2.5 parasite genomes/ml of blood). In case of PKDL, only one patient (macular PKDL) was found to have qPCR value just above threshold.

Whole blood IFN-γ release assay (IGRA)
A whole blood assay for detection of antigen -specific IFN-γ production in-vitro was conducted as described previously. 12,13 Briefly, 3 ml of heparinized whole blood was collected and cultured in the absence of antigen (diluent PBS), or stimulated with L.donovani soluble antigen (SLA) (10 µg/ml), or with phytohemagglutinin (PHA) (Sigma-Aldrich, Germany) (5 µg/ml) as a positive control. These tubes were incubated in a humified air atmosphere at 37°C with 5% CO2 for 20-24 hours. Culture supernatants were collected and stored at -20 0 C until used for cytokine measurement. Antigen-specific IFN-γ levels in the supernatant were measured using BioLegend ELISA Max Dulex Set kit (Cat# 430104; USA). IFN-γ production in response to SLA stimulation was determined by subtracting background levels measured in the non-stimulated (NIL, PBS) samples. Results were considered positive when the IFN-γ concentration in the antigen wells was >52.28 pg/mL; this cutoff was determined from the mean of IFNγ concentration of 36 Non endemic healthy controls (NEHC) + 3 standard deviations.    +  ND  ND  ND  ND  ND  ND  2907 20  Total sand flies exposed Blood fed flies dissected flies Sand flies exposed on forearm Sand flies exposed on leg