Statistical modeling based on structured surveys of Australian native possum excreta harboring Mycobacterium ulcerans predicts Buruli ulcer occurrence in humans

Abstract

Background: Buruli ulcer (BU) is a neglected tropical disease caused by infection of subcutaneous tissue with Mycobacterium ulcerans. BU is commonly reported across rural regions of Central and West Africa but has been increasing dramatically in temperate southeast Australia around the major metropolitan city of Melbourne, with most disease transmission occurring in the summer months. Previous research has shown that Australian native possums are reservoirs of M. ulcerans and that they shed the bacteria in their fecal material (excreta). Field surveys show that locales where possums harbor M. ulcerans overlap with human cases of BU, raising the possibility of using possum excreta surveys to predict the risk of disease occurrence in humans.

Methods: We thus established a highly structured 12 month possum excreta surveillance program across an area of 350 km² in the Mornington Peninsula area 70 km south of Melbourne, Australia. The primary objective of our study was to assess using statistical modeling if M. ulcerans surveillance of possum excreta provided useful information for predicting future human BU case locations.

Results: Over two sampling campaigns in summer and winter, we collected 2,282 possum excreta specimens of which 11% were PCR positive for M. ulcerans-specific DNA. Using the spatial scanning statistical tool SaTScan, we observed non-random, co-correlated clustering of both M. ulcerans positive possum excreta and human BU cases. We next trained a statistical model with the Mornington Peninsula excreta survey data to predict the future likelihood of human BU cases occurring in the region. By observing where human BU cases subsequently occurred, we show that the excreta model performance was superior to a null model trained using the previous year's human BU case incidence data (AUC 0.66 vs 0.55). We then used data unseen by the excreta-informed model from a new survey of 661 possum excreta specimens in Geelong, a geographically separate BU endemic area to the southwest of Melbourne, to prospectively predict the location of human BU cases in that region. As for the Mornington Peninsula, the excreta-based BU prediction model outperformed the null model (AUC 0.75 vs 0.50) and pinpointed specific locations in Geelong where interventions could be deployed to interrupt disease spread.

Conclusions: This study highlights the One Health nature of BU by confirming a quantitative relationship between possum excreta shedding of M. ulcerans and humans developing BU. The excreta survey-informed modeling we have described will be a powerful tool for the efficient targeting of public health responses to stop BU.

Funding: This research was supported by the National Health and Medical Research Council of Australia and the Victorian Government Department of Health (GNT1152807 and GNT1196396).

Keywords: Buruli ulcer; Mycobacterium ulcerans; One Health; disease transmission; epidemiology; global health; infectious disease; infectious diseases; microbiology; zoonosis.

Figure 1.. Overview of the Mornington Peninsula M. ulcerans DNA concentrations in IS2404 positive excreta stratified by sampling season and possum species.

Box plots depict the median, the 25^th (lower hinge), and the 75^th percentiles (upper hinge) of real-time PCR Ct (IS2404) values for each category. The whiskers extend from the Inter-Quartile Range (IQR) hinges to the largest value no further than 1.5 * IQR from the hinges. The null hypothesis (no difference between Ct (IS2404) means) was rejected for p<0.05 and assessed using an unpaired t-test while assuming equal variance (checked with an F-test). ‘Summer’ n=111, ‘Winter’ n=199, ‘Brushtail’ n=8, ‘Ringtail’ n=302’.

Figure 1—figure supplement 1.. Geography of the Mornington Peninsula and Geelong.

Panel A, B, C: zoomed inset maps at successively smaller scales are used to establish the location of the western tip of the Mornington Peninsula (Panel D) and Geelong (Panel E) in Australia. Panel D, E: land use categories of 2011 census mesh blocks as detailed in the legend. The GPS coordinates of all sample points visited during the study are rendered as black squares. The mesh block outlines of the suburb of Belmont are rendered in blue (Panel E).

Figure 1—figure supplement 2.. A set of screenshots demonstrating electronic data collection on the Open Data Kit (ODK) Collect Android app running on a survey phone.

Figure 1—figure supplement 3.. Schematic 96-well plate stack of all experimental controls used during 30 separate extraction and real-time PCR runs.

DNA extraction negative controls were included randomly at a frequency of 5–10% per extraction batch of samples and processed blinded. Additionally, every real-time PCR run included a no-template control (NTC) and a positive (M. ulcerans gDNA) real-time PCR control.

Figure 2.. Geographical surveillance of M. ulcerans in the Mornington Peninsula during the southern hemisphere’s summer.

The distribution of points where possum excreta was sampled along a 200 m grid pattern is presented alongside with IS2404 molecular screening results. The pink to purple color gradient visualizes inferred mycobacterial loads in analyzed excreta as estimated from IS2404 cycle thresholds. The dashed red circles represent significant (p<0.005) non-random clustering of IS2404 positive possum excreta identified with spatial scan statistics. All Buruli ulcer (BU) patients notified to the Department of Health (DH) with an inferred exposure time that overlapped with the excreta survey organized during summer are tabulated here by mesh block. A gradient is used to illustrate BU case counts per mesh block. The dashed blue circles represent geographical areas with higher rates of BU than would be expected if the risk of contracting BU was evenly distributed across the Peninsula.

Figure 2—figure supplement 1.. Selection of two populations of Buruli ulcer (BU) patients with exposure intervals that aligned with the excreta possum surveys organized during the southern hemispheres’ summer and winter.

The two populations comprise BU patients notified to the Victorian Department of Health (DH) who resided in or visited the Mornington Peninsula and had not reported recent (<12 months) contact with any other known BU endemic areas in the state. The reported onset of disease was used to infer the exposure interval during which patients were likely infected based on the mean incubation period of BU in Victoria of 143 days (IQR 101–171) (Loftus et al., 2018).

Figure 3.. Geographical surveillance of M. ulcerans in the Mornington Peninsula during the southern hemisphere’s winter.

The locations where the standardized roadside collection of possum excreta was organized are illustrated alongside with IS2404 molecular screening results. The winter survey was performed along a 200 m grid pattern although three limited regions were additionally sampled at higher resolution along a 50 m grid (detailed in panel B). All Buruli ulcer (BU) patients notified to the Department of Health (DH) with an inferred exposure time that overlapped with the excreta survey organized during winter are tabulated here by mesh block. The dashed circles represent geographical areas with higher rates of IS2404-positive possum excreta (red) or BU disease in humans (blue).

Figure 3—figure supplement 1.. Comparative seasonal distribution of M. ulcerans in the Mornington Peninsula during the southern hemisphere’s summer and winter.

The distribution of points where IS2404 positive possum excreta was sampled along the 200 m sampling grid pattern is presented.

Figure 4.. Statistical modeling approaches to prospectively predict the likelihood that a mesh block will contain a Buruli ulcer (BU) case.

Mornington Peninsula and Geelong have been abbreviated as (MP) and (G), respectively. All paired boxplots show the predicted probabilities for mesh blocks with a case (red) and mesh blocks without a case (blue) and the area-under-the-curve (AUC) value below each graph. (A) Paired boxplot of the Mornington Peninsula excreta-informed model. (B) Paired boxplot of the Mornington Peninsula excreta-informed model when predicting on the Geelong data. (C) Paired boxplot of the Mornington Peninsula previous year’s incidence (2018) null model. (D) Paired boxplot of the Geelong previous year’s incidence (2019) null model. All paired boxplots show the median and interquartile range (E) Ranked performance of all predictive models. Ranking cutoff intervals included the top 5%, 10%, 20%, and 50% of mesh blocks, ordered by their declining predicted class probabilities.

Figure 4—figure supplement 1.. Spatial block cross-validation approach.

The mesh block dataset was split into three spatially contiguous regions, shown in blue, green, and red. The statistical model was fitted on each 2/3 of the data from two regions to predict the probability of observing one or more Buruli ulcer (BU) cases in the remaining 1/3 of the data from the third region. This approach was used to validate the predictive ability of the statistical model using the excreta survey data from the Mornington Peninsula region.

Figure 5.. Geographical surveillance of M. ulcerans in the Geelong region.

The distribution of points where possum excreta was sampled along a 200 m grid pattern is presented alongside with IS2404 molecular screening results. The pink to purple color gradient visualizes inferred mycobacterial loads in analyzed excreta as estimated from IS2404 real-time PCR results. All BU patients were notified to the Department of Health (DH) with an inferred exposure time that overlapped with the Geelong excreta survey organized between January 16, 2020 and April 28, 2020 tabulated here by mesh block. M. ulcerans DNA was detected by IS2404 real-time PCR in 21 of the 661 (3.2%) excreta specimens (Table 4). As observed with the Mornington Peninsula survey, Ringtail possum excreta was more frequently found than Brushtail possum excreta but the proportion IS2404-positive was not significantly different (p=1, Fisher’s exact test) (Table 4).

Figure 5—figure supplement 1.. Population of Buruli ulcer (BU) patients with exposure intervals that aligned with the Geelong excreta possum survey organized between January 16 and April 28, 2020.

The population comprises BU patients notified to the Victorian Department of Health (DH) who resided in or visited the Geelong area and had not reported recent (<12 months) contact with any other known BU endemic areas in the state. The reported onset of disease was used to infer the exposure interval during which patients were likely infected based on the mean incubation period of BU in Victoria of 143 days (IQR 101–171) (Loftus et al., 2018).

Figure 6.. Geographical distribution of the top 20% of probability-ranked mesh blocks for all excreta-informed and null models.

Black indicates mesh blocks in the top 20% of probability-ranked results while grey indicates mesh blocks in the bottom 80%. Black circles with white borders indicate location of confirmed BU cases that occurred within the transmission window following the excreta sampling. (A) Mornington Peninsula null-model top 20% mesh block predictions. (B) Mornington Peninsula excreta-informed model top 20% mesh block predictions. (C) Geelong null-model top 20% mesh block predictions. (D) Geelong excreta-informed model top 20% mesh block predictions. Insets show zoom-in of mesh blocks where Buruli ulcer (BU) cases occurred.