The zoonotic cestode Taenia solium poses a substantial public health and economic challenge. Globally, T. solium is ranked as the foodborne parasitic infection contributing the highest number of disability-adjusted life-years (DALYs), an estimated 2.78 million DALYs in 2010 (Havelaar et al., 2015). Control and elimination of T. solium requires a One Health approach (Thomas et al., 2019; Braae et al., 2019) due to its complex multi-host life-cycle, involving infection with larval-stage cysticerci in the intermediate pig host (porcine cysticercosis, or PCC), infective stages (eggs) in the environment, and taeniasis (infection with the adult worm) in the definitive human host (human taeniasis, or HTT). When humans ingest T. solium eggs, establishment of cysticerci (human cysticercosis, or HCC) in the central nervous system results in neurocysticercosis. Neurocysticercosis is responsible for the major health burden associated with T. solium, causing approximately one third of epilepsy/seizure disorders in endemic settings (Gripper and Welburn, 2017).

Current interventions target PCC through mass treatment of pigs with oxfendazole and their vaccination with TSOL18, and HTT through mass treatment of humans with praziquantel or niclosamide. These control strategies have proven efficacious in field studies (de Coster et al., 2018; Dixon et al., 2021), but to date no large-scale interventions have been rolled out as part of national Neglected Tropical Disease (NTD) programmes, and therefore their impact alone or in combination in different epidemiological settings remains poorly understood and/or modelled (CystiTeam Group for Epidemiology and Modelling of Taenia solium Taeniasis/Cysticercosis, 2019). The milestones proposed in the new World Health Organization (WHO) NTD 2021–2030 roadmap (Havelaar et al., 2015; World Health Organization, 2020) focus on achieving intensified control in hyperendemic areas of 17 (27%) endemic countries. However, endemicity levels for T. solium have not been defined in terms of infection indicators and there exist limited data on the geographical distribution of T. solium prevalence at national scales. For other NTDs, pre-intervention levels of endemicity (according to infection prevalence, intensity, incidence or morbi-mortality) determine the magnitude, duration and likely success of control efforts (NTD Modelling Consortium Onchocerciasis Group, 2019). Therefore, successful T. solium intervention strategies will require tailoring to local epidemiological and socio-economic conditions that determine the intensity of transmission and will likely be informed by age- and/or sex-dependent contact rates and mixing patterns in endemic settings (Dixon et al., 2021; CystiTeam Group for Epidemiology and Modelling of Taenia solium Taeniasis/Cysticercosis, 2019; Welburn et al., 2015).

Global patterns in transmission rates have recently been explored for PCC by assessing the force-of-infection (FoI; the per-capita rate of infection acquisition) according to proposed endemicity levels (overall (sero)prevalence) and geography (Dixon et al., 2020). Estimates of antibody (Ab) and antigen (Ag) seroconversion (and seroreversion rates) for HCC have also been estimated from a small number of longitudinal surveys in Burkina Faso (Dermauw et al., 2018), Peru (Garcia et al., 2001), Ecuador (Coral-Almeida et al., 2014), and Zambia (Mwape et al., 2013), demonstrating substantial geographic variation.

Age-prevalence profiles are not only useful for estimating FoI and (sero)reversion rates, but also to provide insight into processes driving epidemiological dynamics, such as immunological responses or heterogeneity in exposure which may explain abrupt changes in HCC seropositivity in older individuals or distinct prevalence peaks in younger ages for HTT prevalence, as observed in the Democratic Republic of the Congo (DRC) (Kanobana et al., 2011; Madinga et al., 2017). Specific age-prevalence profiles may also provide insight into whether age-targeted interventions may be appropriate in certain settings. This study, therefore, and for the first time systematically collates and reviews all HTT and HCC age-(sero)prevalence data available from the literature and from collaborators, and uses simple and reversible catalytic models to estimate the FoI and Ab-seroreversion/infection loss rates from cross-sectional surveys. The results improve our understanding of global variation in the epidemiology of T. solium and provide estimates of transmission rates, crucial for guiding the design of interventions. In addition, a country profile of FoI estimates is presented at the sub-national level for Colombia to exemplify how detailed FoI analyses can help understand local epidemiological patterns.

This paper presents the first global FoI estimates for T. solium HTT and HCC, identifying geographical heterogeneity that supports calls to implement both locally-adapted control efforts (Johansen et al., 2017Gabriël et al., 2017), and setting-specific parameterisations of T. solium transmission models to support such efforts (Dixon et al., 2019). We have also placed our HTT FoI estimates in the context of (putative) endemic or hyperendemic settings, with 0.44 per 1,000 people per year for a (single) endemic setting, and from 9.6 to 21 per 1000 per year for two hyperendemic settings. HCC FoI estimates ranged from 0.086 to 21 per 1000 per year in (18, including Colombian departments) endemic settings (0.086–21 for Ab-based surveys; 0.17–4.4 for Ag-based surveys), and from 0.54 to 120 per 1000 per year in (19) hyperendemic settings (2.3–120 for Ab-based surveys; 0.54–110 for Ag-based surveys). Further work will be required to refine the proposed (and preliminary) characterisation of endemicity levels, perhaps linked to severity/morbidity as in other NTDs (Prost et al., 1979; Smith et al., 2013). This will be relevant to inform what constitutes a hyperendemic setting in need of intensified control interventions, as proposed in the WHO 2021–2030 goals for T. solium (Havelaar et al., 2015; World Health Organization, 2020). Our work also follows the five principles distilled by Behrend et al., 2020 for communicating policy-relevant NTD modelling to stakeholders and implementation partners (Supplementary file 8).

Across epidemiological settings, there was a clear trend of decreasing average time until humans become HCC Ab-seropositive or infected as inferred through Ag-seropositivity (the reciprocal of λ_sero or λ_inf) with increasing all-age seroprevalence. This makes intuitive sense as the FoI increases in settings with more intense transmission (reflected by a higher all-age seroprevalence). It was not possible to discern whether such trends exist for HTT given the limited availability of HTT datasets with age-stratified prevalence data. Infection loss with infection acquisition was indicated for two HTT copro-Ag-ELISA-based surveys, and five of six HCC antigen-based surveys, supporting inclusion of parameters representing recovery from HTT and HCC in T. solium transmission models (Dixon et al., 2019). Our infection loss rates for HTT indicate that the duration of adult tapeworm infection (the reciprocal of ρ_inf) ranges from 1.3 to 1.4 years (with uncertainty bounds from 1.0 to 2.8 years), aligning with literature estimates suggesting that adult tapeworms live for less than 5 years (Garcia et al., 2014), although our values are only based on two studies. In addition, reinfection of individuals with the adult tapeworm is also likely to occur, particularly in high-endemicity settings; therefore, the persistence of antibodies against the adult worm is likely to complicate the measurement of reinfection rates (where antibody seroconversion is equated to infection, although we take care to differentiate between these two processes when interpreting the λ_sero and λ_inf parameters). However, with the limited number of HTT-based surveys available to estimate antibody seroconversion and duration of antibody parameters, it is difficult to determine to what extent this is an issue. Understanding antibody kinetics in relation to spontaneous death and elimination of the adult tapeworm (either naturally or following treatment) is therefore an important area for further consideration, especially towards understanding whether seroreversion rates are related to baseline antibody levels. HCC Ab-seroreversion rates (or infection loss rates) from this analysis are also consistently larger than HCC Ab-seroconversion rates (or infection acquisition rates), in agreement with comparisons of HCC Ab-seroreversion to Ab-seroconversion (or Ag-seroreversion to Ag-seroconversion) estimates in the literature, which are generally based on cumulative seroconversion and seroreversion proportions, or rule-based modelling approaches (Dermauw et al., 2018; Garcia et al., 2001; Coral-Almeida et al., 2015; Mwape et al., 2013).

Some of the modelled estimates presented here indicate large differences between FoI and Ab-seroreversion/infection loss rates. These differences are particularly evident in datasets, such as Brazil (Gomes et al., 2002) and Risaralda Department in Colombia (Flórez Sánchez et al., 2013), where the reversible model was selected for flat age-(sero)prevalence profiles in low (sero)prevalence settings, where the seroreversion parameter provided minimal information with large uncertainty. The flat age-(sero)prevalence in these settings may also be likely explained by false-positives generated given the slightly sub-optimal performance of the serological antibody diagnostic used e.g. posterior median specificity estimates of 98% in Brazil (Gomes et al., 2002) for the antibody LLGP-EITB test (Tsang et al., 1989), highlighting issues with imperfect specificity of diagnostics in settings with very low transmission intensity.

The high cysticercosis Ab-seroreversion rates indicated across datasets are likely indicative of substantial transient responses generated from exposure without establishment of infection. This may also explain the high all-age Ab-seroprevalence estimates observed in field studies (Garcia et al., 2001). Even with Ag-based surveys, a marker of infection rather than exposure, Ag-positivity may still indicate some degree of transient responses due to partial establishment of (immature) cysts or establishment followed by rapid degeneration of (mature) cysts (Mwape et al., 2012). One expectation was that Ab-seroreversion and infection loss rates are fundamentally biological processes so would be fairly consistent across settings. Since Ab-seroreversion or infection loss parameters were allowed to vary among settings, the consistency of the posterior estimates for these parameters indicates that rates of HCC Ab-seroreversion are stable (i.e. no relationship between all-age Ab-seroprevalence and the average duration humans remain Ab-seropositive, Figure 8b; no relationship between the Ab-seroconversion and Ab-seroreversion rates, Figure 8—figure supplement 2), confirming this a priori expectation. The average duration humans remain infected (i.e., the reciprocal of the HCC infection loss rate), however, decreases with increasing seroprevalence (and a positive relationship between infection acquisition rates and infection loss rates, Figure 8—figure supplement 2). Such a signal seen for Ag-based estimates, albeit with only 5 observations, may indicate increased transient responses in higher transmission settings due to more partial establishment of infection. There is however an absence of wider literature to support the hypothesis of exposure intensity affecting probability of successful parasite establishment, but such (transmission intensity-dependent) parasite establishment processed have been incorporated into transmission models for other NTDs (Hamley et al., 2019), highlighting an area of future research with implications for control efforts. In the case of Ab-seroreversion rates, where more data are available compared to infection loss rates, it may be suitable in future work to jointly fit the ρ_sero parameter for reversible models across settings.

This analysis also provides, for the first time, a detailed country-level analysis, focused on Colombia, exploring the geographical variation of key epidemiological metrics (FoI and antibody seroreversion) alongside understanding potential drivers of these trends, using a rich dataset generated by Flórez Sánchez et al., 2013. Our results identify substantial subnational variation in HCC FoI (measured by Ab-seroconversion) and Ab-seroreversion rates. There appears to be only a limited relationship between these epidemiological quantities, Ab-seroprevalence and other risk factors, including open defecation and pig ownership at the department level, highlighting the complex nature of exposure drivers for HCC. For example, the southeastern rural departments of Vaupés and Amazonas show the highest all-age seroprevalence and FoI estimates. But while Vaupés has one of the highest reported proportions of open defecation, pig ownership is low in both departments. Notwithstanding, substantial subnational variation indicates the need for tailored subnational approaches to control. Spatial variation has been explored formally for this dataset by including spatial structure (i.e. spatial correlation in seropositivity estimated up to approximately 140 km at the Municipality level) into a risk factor analysis, and identifying hot spots of unexplained residual clustering in northern and southern municipalities in Colombia (Galipó et al., 2021). The findings from this study further supports the requirement for designing subnational and targeted interventions (risk groups such as those with low education levels and displaced persons), alongside the need to understand epidemiological dynamics at the subnational level.

A potential limitation of this study is the assumption that diagnostic sensitivity and specificity for a single diagnostic test do not vary substantially among settings (implied by estimating a single posterior for specificity and sensitivity using data from multiple surveys that used the same diagnostic). In reality, there may be some variation in diagnostic performance between settings, particularly for HTT serological diagnostics. For example, the widely used copro-Ag ELISA for detecting adult tapeworm infection is not species specific (Ng-Nguyen et al., 2017). Literature estimates indicate high specificity of HCC diagnostics; however, a proportion of positive results by both HCC Ag- and Ab- diagnostics may be due to transient responses and cross-reactions following exposure to the eggs of other other helminths including Taenia saginata, Echinoccous granulosus and Schistosomia species (Lightowlers et al., 2016b; Noh et al., 2014; Kojic and White, 2003; Furrows et al., 2006). Limited information exists on the co-distribution and prevalence of potentially cross-reactive Taenia species (such as Taenia saginata) and other helminths to determine the relative contribution of within- and between-location variability in the performance of specific diagnostic tests.

The prevailing (and most parsimonious) assumption governing the behaviour of the catalytic models fitted in this analysis is of a constant FoI with respect to age. Collated age-(sero)prevalence profiles presented here largely reveal increasing (sero)prevalence with age, or saturation with age through Ab-seroreversion/infection loss, but this behaviour can also be observed with age-dependent infection rates (Grenfell and Anderson, 1985). However, the available data limit the testing of more complex FoI functions. A few datasets appeared indeed to deviate from the basic assumption, for example for HTT dynamics in the DRC (Madinga et al., 2017) and Bolívar in Colombia (Flórez Sánchez et al., 2013). While it was not possible to fit the current formulation of catalytic models to these datasets, the specific age-prevalence peaks in younger ages may indicate other drivers such as heterogeneity in exposure (by age) or age-related immunity mechanisms. The observations of HTT copro-Ag-ELISA age-prevalence peaks/odds of infection in children from Peru, Guatemala and Zambia (Falcon et al., 2003; Allan et al., 1996; Mwape et al., 2015) suggest that these trends might be more widespread. Age peaks in HTT prevalence could indicate the need for targeted mass drug administration (MDA) programmes, such as delivery of anthelminthics (praziquantel) to school-age children, an approach that could be integrated into existing schistosomiasis control programmes where HTT and schistosomiasis are co-endemic. Should more age-prevalence data become available, especially for HTT, fitting models with age-varying FoI may become relevant as indicated in other NTDs (Gambhir et al., 2009).

The results presented here, particularly for HCC, where many datasets were available, suggest marked geographical heterogeneity in FoI and associated all-age (sero)prevalence estimates. This indicates substantial variation in the intensity of T. solium transmission among endemic settings. There is strong evidence for both HCC Ab-seroreversion and infection loss, likely due to transient dynamics which highlights the need for careful interpretation of cross-sectional (sero)prevalence survey estimates. Catalytic models provide useful tools for interpreting such data, which are far more abundant than a few longitudinal surveys reported in the literature. We also quantify, for the first time, the presence of substantial subnational variation in exposure to HCC, highlighting the need for tailored, sub-national control strategies. More work is also required to understand whether age-prevalence peaks (as observed for HTT) are more commonplace, and whether age-targeted control strategies may be required under specific epidemiological and socioeconomic conditions.

All age-(sero)prevalence data are available in the following data repository: http://doi.org/10.14469/hpc/10047. Original data for two datasets available (under the Creative Commons Attribution License; CC BY 4.0) from the International Livestock Research Institute open-access repository (http://data.ilri.org/portal/dataset/ecozd) referenced in Holt et al. (2016) and University of Liverpool open-access repository (http://dx.doi.org/10.17638/datacat.liverpool.ac.uk/352) referenced in Fèvre et al. (2017).

1. 1. Holt HR
   2. Inthavong P
   3. Khamlome B
   4. Blaszak K
   5. Keokamphe C
   6. Somoulay V
   7. Phongmany A
   8. Burr PA
   9. Graham K
   10. Allen J
   11. Donnelly B
   12. Blacksell SD
   13. Unger F
   14. Grace D
   15. Alonso S
   16. Gilbert J

   (2016) ILRI Data Portal

   ID ecozd. Ecosystem approaches to the better management of zoonotic emerging infectious diseases in the Southeast Asia region.

   http://data.ilri.org/portal/dataset/ecozd
2. 1. Fèvre EM
   2. de Glanville WA
   3. Thomas LF
   4. Cook EAJ
   5. Kariuki S
   6. Wamae CN

   (2017) DataCat: The Research Data Catalogue

   An integrated study of human and animal infectious disease in the Lake Victoria Crescent small-holder crop-livestock production system, Kenya.

   https://doi.org/10.17638/datacat.liverpool.ac.uk/352

1. 1. Allan JC
   2. Avila G
   3. Garcia Noval J
   4. Flisser A
   5. Craig PS

   (1990) Immunodiagnosis of taeniasis by coproantigen detection

   Parasitology 101 Pt 3:473–477.

   https://doi.org/10.1017/s0031182000060686

   • PubMed
   • Google Scholar
2. 1. Allan JC
   2. Velasquez-Tohom M
   3. Garcia-Noval J
   4. Torres-Alvarez R
   5. Yurrita P
   6. Fletes C
   7. de MF
   8. Alfaro H
   9. Craig PS

   (1996) Epidemiology of intestinal taeniasis in four, rural

   Guatemalan Communities Annals of Tropical Medicine and Parasitology 90:157–165.

   https://doi.org/10.1080/00034983.1996.11813039

   • Google Scholar
3. 1. Behrend MR
   2. Basáñez MG
   3. Hamley JID
   4. Porco TC
   5. Stolk WA
   6. Walker M
   7. de Vlas SJ
   8. NTD Modelling Consortium

   (2020) Modelling for policy: the five principles of the neglected tropical diseases modelling consortium

   PLOS Neglected Tropical Diseases 14:e0008033.

   https://doi.org/10.1371/journal.pntd.0008033

   • PubMed
   • Google Scholar
4. 1. Braae UC
   2. Gabriël S
   3. Trevisan C
   4. Thomas LF
   5. Magnussen P
   6. Abela-Ridder B
   7. Ngowi H
   8. Johansen MV

   (2019) Stepwise approach for the control and eventual elimination of taenia solium as a public health problem

   BMC Infectious Diseases 19:182.

   https://doi.org/10.1186/s12879-019-3812-y

   • PubMed
   • Google Scholar
5. 1. Brandt JR
   2. Geerts S
   3. De Deken R
   4. Kumar V
   5. Ceulemans F
   6. Brijs L
   7. Falla N

   (1992) A monoclonal antibody-based ELISA for the detection of circulating excretory-secretory antigens in taenia saginata cysticercosis

   International Journal for Parasitology 22:471–477.

   https://doi.org/10.1016/0020-7519(92)90148-e

   • PubMed
   • Google Scholar
6. 1. Conlan JV
   2. Vongxay K
   3. Khamlome B
   4. Dorny P
   5. Sripa B
   6. Elliot A
   7. Blacksell SD
   8. Fenwick S
   9. Thompson RCA

   (2012) A cross-sectional study of taenia solium in a multiple taeniid-endemic region reveals competition may be protective

   The American Journal of Tropical Medicine and Hygiene 87:281–291.

   https://doi.org/10.4269/ajtmh.2012.11-0106

   • PubMed
   • Google Scholar
7. 1. Coral-Almeida M
   2. Rodríguez-Hidalgo R
   3. Celi-Erazo M
   4. García HH
   5. Rodríguez S
   6. Devleesschauwer B
   7. Benítez-Ortiz W
   8. Dorny P
   9. Praet N

   (2014) Incidence of human taenia solium larval infections in an ecuadorian endemic area: implications for disease burden assessment and control

   PLOS Neglected Tropical Diseases 8:e2887.

   https://doi.org/10.1371/journal.pntd.0002887

   • PubMed
   • Google Scholar
8. 1. Coral-Almeida M
   2. Gabriël S
   3. Abatih EN
   4. Praet N
   5. Benitez W
   6. Dorny P

   (2015) Taenia solium human cysticercosis: A systematic review of sero-epidemiological data from endemic zones around the world

   PLOS Neglected Tropical Diseases 9:e0003919.

   https://doi.org/10.1371/journal.pntd.0003919

   • PubMed
   • Google Scholar
9. 1. CystiTeam Group for Epidemiology and Modelling of Taenia solium Taeniasis/Cysticercosis

   (2019) The world health organization 2030 goals for taenia solium: insights and perspectives from transmission dynamics modelling

   Gates Open Research 3:1546.

   https://doi.org/10.12688/gatesopenres.13068.2

   • Google Scholar
10. 1. de Coster T
    2. Van Damme I
    3. Baauw J
    4. Gabriël S

    (2018) Recent advancements in the control of taenia solium: a systematic review

    Food and Waterborne Parasitology 13:e00030.

    https://doi.org/10.1016/j.fawpar.2018.e00030

    • PubMed
    • Google Scholar
11. 1. Dermauw V
    2. Carabin H
    3. Ganaba R
    4. Cissé A
    5. Tarnagda Z
    6. Gabriël S
    7. Dorny P
    8. Millogo A

    (2018) Factors associated with the 18-month cumulative incidence of seroconversion of active infection with taenia solium cysticercosis: a cohort study among residents of 60 villages in burkina faso

    The American Journal of Tropical Medicine and Hygiene 99:1018–1027.

    https://doi.org/10.4269/ajtmh.18-0294

    • PubMed
    • Google Scholar
12. Website

    1. DiagAutom

    (2016) Human Cysticercosis IgG (T. Solium) ELISA Test kit

    Accessed March 9, 2020.

    http://www.rapidtest.com/index.php?i=Parasitology-ELISA-kits&id=170&cat=17
13. 1. Diggle PJ

    (2011) Estimating prevalence using an imperfect test

    Epidemiology Research International 2011:1–5.

    https://doi.org/10.1155/2011/608719

    • Google Scholar
14. 1. Dixon MA
    2. Braae UC
    3. Winskill P
    4. Walker M
    5. Devleesschauwer B
    6. Gabriël S
    7. Basáñez MG

    (2019) Strategies for tackling taenia solium taeniosis/cysticercosis: A systematic review and comparison of transmission models, including an assessment of the wider taeniidae family transmission models

    PLOS Neglected Tropical Diseases 13:e0007301.

    https://doi.org/10.1371/journal.pntd.0007301

    • PubMed
    • Google Scholar
15. 1. Dixon MA
    2. Winskill P
    3. Harrison WE
    4. Whittaker C
    5. Schmidt V
    6. Sarti E
    7. Bawm S
    8. Dione MM
    9. Thomas LF
    10. Walker M
    11. Basáñez MG

    (2020) Force-of-infection of taenia solium porcine cysticercosis: a modelling analysis to assess global incidence and prevalence trends

    Scientific Reports 10:17637.

    https://doi.org/10.1038/s41598-020-74007-x

    • PubMed
    • Google Scholar
16. 1. Dixon MA
    2. Winskill P
    3. Harrison WE
    4. Basáñez MG

    (2021) Taenia solium taeniasis/cysticercosis: from parasite biology and immunology to diagnosis and control

    Advances in Parasitology 112:133–217.

    https://doi.org/10.1016/bs.apar.2021.03.003

    • PubMed
    • Google Scholar
17. Software

    1. Dixon MA

    (2022) Human_tsol_FoI_modelling, version swh:1:dir:fbf523fa5d981757bd3bfa28fd300cd441a7f5a3

    GitHub.

    https://github.com/mrc-ide/human_tsol_FoI_modelling
18. 1. Dorny P
    2. Vercammen F
    3. Brandt J
    4. Vansteenkiste W
    5. Berkvens D
    6. Geerts S

    (2000) Sero-epidemiological study of taenia saginata cysticercosis in belgian cattle

    Veterinary Parasitology 88:43–49.

    https://doi.org/10.1016/s0304-4017(99)00196-x

    • PubMed
    • Google Scholar
19. 1. Edia-Asuke AU
    2. Inabo HI
    3. Mukaratirwa S
    4. Umoh VJ
    5. Whong CMZ
    6. Asuke S
    7. Ella EE

    (2015) Seroprevalence of human cysticercosis and its associated risk factors among humans in areas of kaduna metropolis, nigeria

    Journal of Infection in Developing Countries 9:799–805.

    https://doi.org/10.3855/jidc.5415

    • PubMed
    • Google Scholar
20. 1. Falcon N
    2. Bernal T
    3. Lescano AG
    4. Gavidia C
    5. Gonzalez AE
    6. García HH
    7. Tovar M
    8. Tsang VCW
    9. Verastegui M
    10. Moulton LH
    11. Rodriguez S
    12. Gilman RH

    (2003) HYPERENDEMIC human and porcine taenia solium infection in perú

    The American Journal of Tropical Medicine and Hygiene 68:268–275.

    https://doi.org/10.4269/ajtmh.2003.68.268

    • Google Scholar
21. 1. Fèvre EM
    2. de Glanville WA
    3. Thomas LF
    4. Cook EAJ
    5. Kariuki S
    6. Wamae CN

    (2017) An integrated study of human and animal infectious disease in the lake victoria crescent small-holder crop-livestock production system, kenya

    BMC Infectious Diseases 17:457.

    https://doi.org/10.1186/s12879-017-2559-6

    • PubMed
    • Google Scholar
22. 1. Fleury A
    2. Hernández M
    3. Avila M
    4. Cárdenas G
    5. Bobes RJ
    6. Huerta M
    7. Fragoso G
    8. Uribe-Campero L
    9. Harrison LJS
    10. Parkhouse RME
    11. Sciutto E

    (2007) Detection of HP10 antigen in serum for diagnosis and follow-up of subarachnoidal and intraventricular human neurocysticercosis

    Journal of Neurology, Neurosurgery, and Psychiatry 78:970–974.

    https://doi.org/10.1136/jnnp.2006.107243

    • PubMed
    • Google Scholar
23. 1. Flórez Sánchez AC
    2. Pastrán SM
    3. Vargas NS
    4. Beltrán M
    5. Enriquez Y
    6. Peña AP
    7. Villarreal A
    8. Salamanca L
    9. Rincón E
    10. Garzón IP
    11. Muñoz L
    12. Guasmayan L
    13. Valencia C
    14. Parra S
    15. Hernandez N

    (2013)

    Cisticercosis en colombia estudio de seroprevalencia 2008-2010

    Acta Neurológica Colombiana 29:0120–8748.

    • Google Scholar
24. 1. Furrows SJ
    2. McCroddan J
    3. Bligh WJ
    4. Chiodini P

    (2006) Lack of specificity of a single positive 50-kda band in the electroimmunotransfer blot (EITB) assay for cysticercosis

    Clinical Microbiology and Infection 12:459–462.

    https://doi.org/10.1111/j.1469-0691.2006.01381.x

    • PubMed
    • Google Scholar
25. 1. Gabriël S
    2. Dorny P
    3. Mwape KE
    4. Trevisan C
    5. Braae UC
    6. Magnussen P
    7. Thys S
    8. Bulaya C
    9. Phiri IK
    10. Sikasunge CS
    11. Makungu C
    12. Afonso S
    13. Nicolau Q
    14. Johansen MV

    (2017) Control of taenia solium taeniasis/cysticercosis: the best way forward for sub-saharan africa?

    Acta Tropica 165:252–260.

    https://doi.org/10.1016/j.actatropica.2016.04.010

    • PubMed
    • Google Scholar
26. 1. Galipó E
    2. Dixon MA
    3. Fronterrè C
    4. Cucunubá ZM
    5. Basáñez MG
    6. Stevens K
    7. Flórez Sánchez AC
    8. Walker M

    (2021) Spatial distribution and risk factors for human cysticercosis in colombia

    Parasites & Vectors 14:590.

    https://doi.org/10.1186/s13071-021-05092-8

    • PubMed
    • Google Scholar
27. 1. Gambhir M
    2. Basáñez MG
    3. Burton MJ
    4. Solomon AW
    5. Bailey RL
    6. Holland MJ
    7. Blake IM
    8. Donnelly CA
    9. Jabr I
    10. Mabey DC
    11. Grassly NC

    (2009) The development of an age-structured model for trachoma transmission dynamics, pathogenesis and control

    PLOS Neglected Tropical Diseases 3:e462.

    https://doi.org/10.1371/journal.pntd.0000462

    • PubMed
    • Google Scholar
28. 1. Garcia HH
    2. Gonzalez AE
    3. Gilman RH
    4. Palacios LG
    5. Jimenez I
    6. Rodriguez S
    7. Verastegui M
    8. Wilkins P
    9. Tsang VC
    10. Cysticercosis Working Group in Peru

    (2001) Short report: transient antibody response in taenia solium infection in field conditions-a major contributor to high seroprevalence

    The American Journal of Tropical Medicine and Hygiene 65:31–32.

    https://doi.org/10.4269/ajtmh.2001.65.31

    • PubMed
    • Google Scholar
29. 1. García HH
    2. Gonzalez AE
    3. Evans CAW
    4. Gilman RH

    (2003) Cysticercosis working group in perú

    Taenia Solium Cysticercosis Lancet 362:547–556.

    https://doi.org/10.1016/S0140-6736(03)14117-7

    • Google Scholar
30. 1. Garcia HH
    2. Rodriguez S
    3. Friedland JS
    4. Cysticercosis Working Group in Peru

    (2014) Immunology of taenia solium taeniasis and human cysticercosis

    Parasite Immunology 36:388–396.

    https://doi.org/10.1111/pim.12126

    • PubMed
    • Google Scholar
31. 1. Gomes I
    2. Veiga M
    3. Embiruçu EK
    4. Rabelo R
    5. Mota B
    6. Meza-Lucas A
    7. Tapia-Romero R
    8. Carrillo-Becerril BL
    9. Alcántara-Anguiano I
    10. Correa D
    11. Melo A

    (2002) Taeniasis and cysticercosis prevalence in a small village from northeastern brazil

    Arquivos de Neuro-Psiquiatria 60:219–223.

    https://doi.org/10.1590/S0004-282X2002000200006

    • Google Scholar
32. 1. Gonzalez G
    2. Rosenheck RA

    (2002) Outcomes and service use among homeless persons with serious mental illness and substance abuse

    Psychiatric Services 53:437–446.

    https://doi.org/10.1176/appi.ps.53.4.437

    • PubMed
    • Google Scholar
33. 1. Grenfell BT
    2. Anderson RM

    (1985) The estimation of age-related rates of infection from case notifications and serological data

    The Journal of Hygiene 95:419–436.

    https://doi.org/10.1017/s0022172400062859

    • PubMed
    • Google Scholar
34. 1. Gripper LB
    2. Welburn SC

    (2017) The causal relationship between neurocysticercosis infection and the development of epilepsy - A systematic review

    Infectious Diseases of Poverty 6:31.

    https://doi.org/10.1186/s40249-017-0245-y

    • PubMed
    • Google Scholar
35. 1. Hamley JID
    2. Milton P
    3. Walker M
    4. Basáñez MG

    (2019) Modelling exposure heterogeneity and density dependence in onchocerciasis using a novel individual-based transmission model, epioncho-ibm: implications for elimination and data needs

    PLOS Neglected Tropical Diseases 13:e0007557.

    https://doi.org/10.1371/journal.pntd.0007557

    • PubMed
    • Google Scholar
36. 1. Hancock K
    2. Pattabhi S
    3. Whitfield FW
    4. Yushak ML
    5. Lane WS
    6. Garcia HH
    7. Gonzalez AE
    8. Gilman RH
    9. Tsang VCW

    (2006) Characterization and cloning of T24, a taenia solium antigen diagnostic for cysticercosis

    Molecular and Biochemical Parasitology 147:109–117.

    https://doi.org/10.1016/j.molbiopara.2006.02.004

    • PubMed
    • Google Scholar
37. 1. Harrison LJ
    2. Joshua GW
    3. Wright SH
    4. Parkhouse RM

    (1989) Specific detection of circulating surface/secreted glycoproteins of viable cysticerci in taenia saginata cysticercosis

    Parasite Immunology 11:351–370.

    https://doi.org/10.1111/j.1365-3024.1989.tb00673.x

    • PubMed
    • Google Scholar
38. 1. Havelaar AH
    2. Kirk MD
    3. Torgerson PR
    4. Gibb HJ
    5. Hald T
    6. Lake RJ
    7. Praet N
    8. Bellinger DC
    9. de Silva NR
    10. Gargouri N
    11. Speybroeck N
    12. Cawthorne A
    13. Mathers C
    14. Stein C
    15. Angulo FJ
    16. Devleesschauwer B
    17. World Health Organization Foodborne Disease Burden Epidemiology Reference Group

    (2015) World health organization global estimates and regional comparisons of the burden of foodborne disease in 2010

    PLOS Medicine 12:e1001923.

    https://doi.org/10.1371/journal.pmed.1001923

    • PubMed
    • Google Scholar
39. 1. Holt HR
    2. Inthavong P
    3. Khamlome B
    4. Blaszak K
    5. Keokamphe C
    6. Somoulay V
    7. Phongmany A
    8. Durr PA
    9. Graham K
    10. Allen J
    11. Donnelly B
    12. Blacksell SD
    13. Unger F
    14. Grace D
    15. Alonso S
    16. Gilbert J

    (2016) Endemicity of zoonotic diseases in pigs and humans in lowland and upland lao pdr: identification of socio-cultural risk factors

    PLOS Neglected Tropical Diseases 10:e0003913.

    https://doi.org/10.1371/journal.pntd.0003913

    • PubMed
    • Google Scholar
40. 1. Jayaraman T
    2. Prabhakaran V
    3. Babu P
    4. Raghava MV
    5. Rajshekhar V
    6. Dorny P
    7. Muliyil J
    8. Oommen A

    (2011) Relative seroprevalence of cysticercus antigens and antibodies and antibodies to taenia ova in a population sample in south india suggests immunity against neurocysticercosis

    Transactions of the Royal Society of Tropical Medicine and Hygiene 105:153–159.

    https://doi.org/10.1016/j.trstmh.2010.10.007

    • PubMed
    • Google Scholar
41. 1. Johansen MV
    2. Trevisan C
    3. Gabriël S
    4. Magnussen P
    5. Braae UC

    (2017) Are we ready for taenia solium cysticercosis elimination in sub-saharan africa?

    Parasitology 144:59–64.

    https://doi.org/10.1017/S0031182016000500

    • PubMed
    • Google Scholar
42. 1. Kanobana K
    2. Praet N
    3. Kabwe C
    4. Dorny P
    5. Lukanu P
    6. Madinga J
    7. Mitashi P
    8. Verwijs M
    9. Lutumba P
    10. Polman K

    (2011) High prevalence of taenia solium cysticerosis in a village community of bas-congo, democratic republic of congo

    International Journal for Parasitology 41:1015–1018.

    https://doi.org/10.1016/j.ijpara.2011.06.004

    • PubMed
    • Google Scholar
43. 1. Kojic EM
    2. White AC

    (2003) A positive enzyme-linked immunoelectrotransfer blot assay result for A patient without evidence of cysticercosis

    Clinical Infectious Diseases 36:e7–e9.

    https://doi.org/10.1086/344445

    • PubMed
    • Google Scholar
44. Book

    1. Lambert B

    (2018)

    Chapter 8: A Student’s Guide to Bayesian Statistics

    London: SAGE Publications.

    • Google Scholar
45. 1. Lescano AG
    2. Garcia HH
    3. Gilman RH
    4. Gavidia CM
    5. Tsang VCW
    6. Rodriguez S
    7. Moulton LH
    8. Villaran MV
    9. Montano SM
    10. Gonzalez AE

    (2009) Taenia solium cysticercosis hotspots surrounding tapeworm carriers: clustering on human seroprevalence but not on seizures

    PLOS Neglected Tropical Diseases 3:e371.

    https://doi.org/10.1371/journal.pntd.0000371

    • Google Scholar
46. 1. Lightowlers MW
    2. Donadeu M
    3. Elaiyaraja M
    4. Maithal K
    5. Kumar KA
    6. Gauci CG
    7. Firestone SM
    8. Sarasola P
    9. Rowan TG

    (2016a) Anamnestic responses in pigs to the taenia solium TSOL18 vaccine and implications for control strategies

    Parasitology 143:416–420.

    https://doi.org/10.1017/S0031182016000202

    • PubMed
    • Google Scholar
47. 1. Lightowlers MW
    2. Garcia HH
    3. Gauci CG
    4. Donadeu M
    5. Abela-Ridder B

    (2016b) Monitoring the outcomes of interventions against taenia solium: options and suggestions

    Parasite Immunology 38:158–169.

    https://doi.org/10.1111/pim.12291

    • PubMed
    • Google Scholar
48. 1. López M
    2. Murillo C
    3. Sarria N
    4. Nicholls R
    5. Corredor A

    (1988)

    Estandarización y evaluación de la prueba de ELISA para la detección de anticuerpos en LCR y suero en neurocisticercosis

    Acta Neurológica Colombiana 4:164–169.

    • Google Scholar
49. 1. Madinga J
    2. Kanobana K
    3. Lukanu P
    4. Abatih E
    5. Baloji S
    6. Linsuke S
    7. Praet N
    8. Kapinga S
    9. Polman K
    10. Lutumba P
    11. Speybroeck N
    12. Dorny P
    13. Harrison W
    14. Gabriel S

    (2017) Geospatial and age-related patterns of taenia solium taeniasis in the rural health zone of kimpese, democratic republic of congo

    Acta Tropica 165:100–109.

    https://doi.org/10.1016/j.actatropica.2016.03.013

    • PubMed
    • Google Scholar
50. 1. Moher D
    2. Shamseer L
    3. Clarke M
    4. Ghersi D
    5. Liberati A
    6. Petticrew M
    7. Shekelle P
    8. Stewart LA
    9. Group P

    (2015) Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement

    Systematic Reviews 4:1.

    https://doi.org/10.1186/2046-4053-4-1

    • PubMed
    • Google Scholar
51. 1. Moro PL
    2. Lopera L
    3. Bonifacio N
    4. Gilman RH
    5. Silva B
    6. Verastegui M
    7. Gonzales A
    8. Garcia HH
    9. Cabrera L
    10. Cysticercosis Working Group in Peru

    (2003) Taenia solium infection in a rural community in the peruvian andes

    Annals of Tropical Medicine and Parasitology 97:373–379.

    https://doi.org/10.1179/000349803235002371

    • PubMed
    • Google Scholar
52. 1. Muench H

    (1934) Derivation of rates from summation data by the catalytic curve

    Journal of the American Statistical Association 29:25–38.

    https://doi.org/10.1080/01621459.1934.10502684

    • Google Scholar
53. 1. Mwape KE
    2. Phiri IK
    3. Praet N
    4. Muma JB
    5. Zulu G
    6. Van den Bossche P
    7. de Deken R
    8. Speybroeck N
    9. Dorny P
    10. Gabriël S

    (2012) Taenia solium infections in a rural area of eastern zambia-a community based study

    PLOS Neglected Tropical Diseases 6:e1594.

    https://doi.org/10.1371/journal.pntd.0001594

    • PubMed
    • Google Scholar
54. 1. Mwape KE
    2. Phiri IK
    3. Praet N
    4. Speybroeck N
    5. Muma JB
    6. Dorny P
    7. Gabriël S
    8. Flisser A

    (2013) The incidence of human cysticercosis in a rural community of eastern zambia

    PLOS Neglected Tropical Diseases 7:e2142.

    https://doi.org/10.1371/journal.pntd.0002142

    • PubMed
    • Google Scholar
55. 1. Mwape KE
    2. Phiri IK
    3. Praet N
    4. Dorny P
    5. Muma JB
    6. Zulu G
    7. Speybroeck N
    8. Gabriël S

    (2015) Study and ranking of determinants of taenia solium infections by classification tree models

    The American Journal of Tropical Medicine and Hygiene 92:56–63.

    https://doi.org/10.4269/ajtmh.13-0593

    • PubMed
    • Google Scholar
56. 1. Ng-Nguyen D
    2. Stevenson MA
    3. Dorny P
    4. Gabriël S
    5. Vo TV
    6. Nguyen VAT
    7. Phan TV
    8. Hii SF
    9. Traub RJ

    (2017) Comparison of a new multiplex real-time PCR with the kato katz thick smear and copro-antigen ELISA for the detection and differentiation of taenia spp: in human stools

    PLOS Neglected Tropical Diseases 11:e0005743.

    https://doi.org/10.1371/journal.pntd.0005743

    • PubMed
    • Google Scholar
57. 1. Nguekam JP
    2. Zoli AP
    3. Zogo PO
    4. Kamga ACT
    5. Speybroeck N
    6. Dorny P
    7. Brandt J
    8. Losson B
    9. Geerts S

    (2003) A seroepidemiological study of human cysticercosis in west cameroon

    Tropical Medicine & International Health 8:144–149.

    https://doi.org/10.1046/j.1365-3156.2003.01000.x

    • PubMed
    • Google Scholar
58. 1. Noh J
    2. Rodriguez S
    3. Lee YM
    4. Handali S
    5. Gonzalez AE
    6. Gilman RH
    7. Tsang VCW
    8. Garcia HH
    9. Wilkins PP

    (2014) Recombinant protein- and synthetic peptide-based immunoblot test for diagnosis of neurocysticercosis

    Journal of Clinical Microbiology 52:1429–1434.

    https://doi.org/10.1128/JCM.03260-13

    • PubMed
    • Google Scholar
59. 1. NTD Modelling Consortium Onchocerciasis Group

    (2019) The world health organization 2030 goals for onchocerciasis: insights and perspectives from mathematical modelling: NTD modelling consortium onchocerciasis group

    Gates Open Research 3:1545.

    https://doi.org/10.12688/gatesopenres.13067.1

    • PubMed
    • Google Scholar
60. 1. Praet N
    2. Rodriguez-Hidalgo R
    3. Speybroeck N
    4. Ahounou S
    5. Benitez-Ortiz W
    6. Berkvens D
    7. Hul AV
    8. Barrionuevo-Samaniego M
    9. Saegerman C
    10. Dorny P

    (2010) Infection with versus exposure to taenia solium: what do serological test results tell us?

    The American Journal of Tropical Medicine and Hygiene 83:413–415.

    https://doi.org/10.4269/ajtmh.2010.10-0121

    • PubMed
    • Google Scholar
61. 1. Praet N
    2. Verweij JJ
    3. Mwape KE
    4. Phiri IK
    5. Muma JB
    6. Zulu G
    7. van Lieshout L
    8. Rodriguez-Hidalgo R
    9. Benitez-Ortiz W
    10. Dorny P
    11. Gabriël S

    (2013) Bayesian modelling to estimate the test characteristics of coprology, coproantigen ELISA and a novel real-time PCR for the diagnosis of taeniasis

    Tropical Medicine & International Health 18:608–614.

    https://doi.org/10.1111/tmi.12089

    • PubMed
    • Google Scholar
62. 1. Prost A
    2. Hervouet JP
    3. Thylefors B

    (1979)

    Les niveaux d’endémicité dans l’onchocercose

    Bulletin of the World Health Organization 57:655–662.

    • Google Scholar
63. Software

    1. R Development Core Team

    (2021) R: A language and environment for statistical computing, version 4.0.5

    R Foundation for Statistical Computing, Vienna, Austria.

    https://www.R-project.org
64. 1. Sahlu I
    2. Carabin H
    3. Ganaba R
    4. Preux PM
    5. Cissé AK
    6. Tarnagda Z
    7. Gabriël S
    8. Dermauw V
    9. Dorny P
    10. Bauer C
    11. Millogo A

    (2019) Estimating the association between being seropositive for cysticercosis and the prevalence of epilepsy and severe chronic headaches in 60 villages of rural burkina faso

    PLOS Neglected Tropical Diseases 13:e0007101.

    https://doi.org/10.1371/journal.pntd.0007101

    • PubMed
    • Google Scholar
65. 1. Sato MO
    2. Sato M
    3. Yanagida T
    4. Waikagul J
    5. Pongvongsa T
    6. Sako Y
    7. Sanguankiat S
    8. Yoonuan T
    9. Kounnavang S
    10. Kawai S
    11. Ito A
    12. Okamoto M
    13. Moji K

    (2018) Taenia solium, taenia saginata, taenia asiatica, their hybrids and other helminthic infections occurring in a neglected tropical diseases’ highly endemic area in lao pdr

    PLOS Neglected Tropical Diseases 12:e0006260.

    https://doi.org/10.1371/journal.pntd.0006260

    • PubMed
    • Google Scholar
66. 1. Smith JL
    2. Flueckiger RM
    3. Hooper PJ
    4. Polack S
    5. Cromwell EA
    6. Palmer SL
    7. Emerson PM
    8. Mabey DCW
    9. Solomon AW
    10. Haddad D
    11. Brooker SJ

    (2013) The geographical distribution and burden of trachoma in africa

    PLOS Neglected Tropical Diseases 7:e2359.

    https://doi.org/10.1371/journal.pntd.0002359

    • PubMed
    • Google Scholar
67. 1. Spiegelhalter DJ
    2. Best NG
    3. Carlin BP
    4. van der Linde A

    (2002) Bayesian measures of model complexity and fit

    Journal of the Royal Statistical Society 64:583–639.

    https://doi.org/10.1111/1467-9868.00353

    • Google Scholar
68. 1. Theis JH
    2. Goldsmith RS
    3. Flisser A
    4. Koss J
    5. Chionino C
    6. Plancarte A

    (1994)

    Detection by immunoblot assay of antibodies to taenia solium cysticerci in sera from residents of rural communities and from epileptic patients in bali

    Indonesia The Southeast Asian Journal of Tropical Medicine and Public Health 25:7777908.

    • Google Scholar
69. 1. Thomas LF
    2. Cook EAJ
    3. Fèvre EM
    4. Rushton J

    (2019) Control of taenia solium; A case for public and private sector investment

    Frontiers in Veterinary Science 6:176.

    https://doi.org/10.3389/fvets.2019.00176

    • PubMed
    • Google Scholar
70. 1. Tsang VC
    2. Brand JA
    3. Boyer AE

    (1989) An enzyme-linked immunoelectrotransfer blot assay and glycoprotein antigens for diagnosing human cysticercosis (taenia solium)

    The Journal of Infectious Diseases 159:50–59.

    https://doi.org/10.1093/infdis/159.1.50

    • PubMed
    • Google Scholar
71. 1. Wardrop NA
    2. Thomas LF
    3. Atkinson PM
    4. de Glanville WA
    5. Cook EAJ
    6. Wamae CN
    7. Gabriël S
    8. Dorny P
    9. Harrison LJS
    10. Fèvre EM

    (2015) The influence of socio-economic, behavioural and environmental factors on taenia spp: transmission in western kenya: evidence from a cross-sectional survey in humans and pigs

    PLOS Neglected Tropical Diseases 9:e0004223.

    https://doi.org/10.1371/journal.pntd.0004223

    • PubMed
    • Google Scholar
72. 1. Weka RP
    2. Ikeh EI
    3. Kamani J

    (2013) Seroprevalence of antibodies (igg) to taenia solium among pig rearers and associated risk factors in jos metropolis, nigeria

    Journal of Infection in Developing Countries 7:67–72.

    https://doi.org/10.3855/jidc.2309

    • PubMed
    • Google Scholar
73. 1. Welburn SC
    2. Beange I
    3. Ducrotoy MJ
    4. Okello AL

    (2015) The neglected zoonoses--the case for integrated control and advocacy

    Clinical Microbiology and Infection 21:433–443.

    https://doi.org/10.1016/j.cmi.2015.04.011

    • PubMed
    • Google Scholar
74. 1. Wilkins PP
    2. Allan JC
    3. Verastegui M
    4. Acosta M
    5. Eason AG
    6. Garcia HH
    7. Gonzalez AE
    8. Gilman RH
    9. Tsang VC

    (1999) Development of a serologic assay to detect taenia solium taeniasis

    The American Journal of Tropical Medicine and Hygiene 60:199–204.

    https://doi.org/10.4269/ajtmh.1999.60.199

    • PubMed
    • Google Scholar
75. Book

    1. World Health Organization

    (2020)

    Ending the Neglect to Attain the Sustainable Development Goals: A Road Map for Neglected Tropical Diseases 2021–2030

    Geneva, Switzerland: WHO.

    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Global Force-of-Infection Trends for Human Taenia solium Taeniasis/Cysticercosis" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Marcelo U Ferreira as Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Miles Davenport as the Senior Editor.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1) Please add a more detailed discussion about the limitations of the dataset analyzed (reviewer #1). Address the potential differences between antigen and antibody detection, the differences between laboratory methods used in each primary study (e.g., species-specificity, sensitivity) and openly discuss how these differences may affect the current analysis.

2) Please specifically address why the reversible catalytic model has been used to fit some data, while the simple (non-reversible) version has been used to fit other datasets. What is the rationale for using such models with such different assumptions (reviewer #2)?

3) Please correct the typos listed by reviewers #2 and #3.

Reviewer #1 (Recommendations for the authors):

Some suggestions are listed below. Given that no new data are to be generated by the authors, I primarily focus on ways of acknowledging the limitations of available data and discussing how can they impact this study.

1. I would like to see a much more comprehensive discussion on "seroreversion" in human taeniasis and cysticercosis. The basic question is: how quickly antibodies become undetectable after treatment or following spontaneous worm death and elimination? Do seroreversion rates depend on average levels of antibodies at baseline, before treatment (i.e., are individuals heavily exposed to infection within the past few years more likely to remain seropositive long after treatment even in the absence of reinfection)?

2. If antibodies are not quickly cleared, then seroconversion rates are not synonymous with "infection rates". In a population with heterogeneous exposure to infection, individuals more frequent exposed to infection may remain seropositive for longer (even if treated) because of memory responses and their antibody status will not change upon reinfection (i.e., they will remain positive). There is a potential bias towards underestimating infection rates from seroconversion data -- especially among individuals who are most likely to be (re)infected..

3. If the antibody-detection assay used in the Colombian study (described in a local publication, not available to this reviewer) is not strictly species- and stage-specific, and also short-lived (i.e., if only individuals currently exposed to T. solium cysticerci are positive), the age-related prevalence of "cysticercosis" may have been grossly overestimated in this country.

4. If prevalence data from selected study sites in a dozen countries cannot be taken as representative of the countries/regions of origin, the study is far from being able to estimate "global force-of-infection trends". Please provide further information regarding the primary studies, in addition to those in Table S1, to help the reader decide whether or not some extrapolation is possible.

Reviewer #2 (Recommendations for the authors):

Although the work is well written and very organized, I have some concerns about the model and assumptions used in the analysis.

Equation 3 is different than the one reported by Diggle, 2011. One essential set of parentheses is missing. Please, clarify it.

The methods section would be improved by presenting the motivations for the models' choice and a brief description of them, including assumptions and limitations. Although the reversible catalytic model (rho>0) is a generalization of the simple form (rho=0), there are two mutually exclusive assumptions: either only conversion or conversion and reversion. Because of it, unless well justified, it seems unlike that a particular pattern of an infectious disease could be described considering two fundamentally different biological processes. As observed in Figures 2, 3 and 4, the data sometimes support the simple model, sometimes support the reversible, and sometimes it is not possible to decide. It means to me that it is not possible to decide between both designs which encapsulates the mechanism driving the disease age-prevalence presented by the data. Please, clarify it.

Another concern is regarding to the saturation of the reversible model. The current formulation assumes that all individuals are exposed/susceptible to the disease and, therefore, its saturation is controlled by the reversion rate parameter (rho). The greater rho, the lower the saturation. Hypothetically, one can think of a situation in which just a fraction of the population is exposed/susceptible to an underlined disease. In this scenario, the greater the fraction of individuals not exposed/susceptible, the greater the interference bias of the reversion rate. I wonder whether the authors believe that this is a reasonable possibility and, in the affirmative case, please acknowledge the limitation of the methodology.

Essential revisions:

1) Please add a more detailed discussion about the limitations of the dataset analyzed (reviewer #1). Address the potential differences between antigen and antibody detection, the differences between laboratory methods used in each primary study (e.g., species-specificity, sensitivity) and openly discuss how these differences may affect the current analysis.

We thank Reviewer #1 and the Reviewing Editor for providing this feedback. A more detailed response to these points can be found in Author responses 1.3 and 1.8, where we focus on providing clarification on the diagnostic used in Flórez Sánchez et al., (2013). We highlight the difference between inferences made from antibody-based surveys (relating to exposure) and antigen-based surveys (active infection dynamics) in Authors’ responses 1.2 and 1.7.

We also provide a detailed analysis of the limitations regarding current diagnostics and how these may impact our analysis on lines 411 – 415 regarding imperfect specificity, especially in low (sero)prevalence settings, and on lines 465 – 473 in the Discussion, focussing on implications regarding variation in sensitivity and specificity estimates across settings, as well as the presence of transient responses/ cross-reactions. In particular, our responses on diagnostic test limitations beyond those already mentioned can be found in Authors’ responses 1.1 in relation to taeniasis diagnostics.

2) Please specifically address why the reversible catalytic model has been used to fit some data, while the simple (non-reversible) version has been used to fit other datasets. What is the rationale for using such models with such different assumptions (reviewer #2)?

We thank Reviewer #2 and the Reviewing Editor for providing this feedback. We fitted both models (simple and reversible) to each dataset, to determine whether there is more evidence for biological processes including or excluding seroreversion / infection loss across settings, as this is a key knowledge gap in the field. The best-fit model was then selected and presented. A more detailed response to these points can be found under Authors’ response 2.2.

3) Please correct the typos listed by reviewers #2 and #3.

We thank Reviewers #2 and #3 and the Reviewing Editor for indicating these typographical errors. We will address the specific typos under our responses in the Reviewers #2 and #3 (Public Review and Recommendations) sections.

Reviewer #1 (Recommendations for the authors):

Some suggestions are listed below. Given that no new data are to be generated by the authors, I primarily focus on ways of acknowledging the limitations of available data and discussing how can they impact this study.

1. I would like to see a much more comprehensive discussion on "seroreversion" in human taeniasis and cysticercosis. The basic question is: how quickly antibodies become undetectable after treatment or following spontaneous worm death and elimination? Do seroreversion rates depend on average levels of antibodies at baseline, before treatment (i.e., are individuals heavily exposed to infection within the past few years more likely to remain seropositive long after treatment even in the absence of reinfection)?

We agree with the reviewer that this is an important area for further consideration. There is very limited literature available on the persistence of antibodies raised to the adult tapeworm after spontaneous worm death and elimination, and, as evidenced by the minimal number of available studies with suitable ageprevalence data, we were not able to estimate and assess human taeniasis seroreversion rates in different epidemiological settings. Thus, it is difficult to discuss these dynamics further in our paper at present. We therefore introduce further text to present these considerations as important further research questions that should form part of the T. solium research agenda:

“Understanding antibody kinetics in relation to spontaneous death and elimination of the adult tapeworm (either naturally or following treatment) is therefore an important area for further consideration, especially towards understanding whether seroreversion rates are related to baseline antibody levels.” (lines 396 – 399)

We have also included some further text, as indicated in Authors’ response 1.1, regarding challenges with measuring reinfection rates should antibodies to the adult tapeworm persist long after worm death.

We currently discuss, in detail, considerations around human cysticercosis (antibody and antigen) seroreversion rates, exploring possible underlying mechanisms to explain the presence/absence and variation in these rates through transient responses and partial establishment of cysts (in the paragraph on lines 416– 440).

2. If antibodies are not quickly cleared, then seroconversion rates are not synonymous with "infection rates". In a population with heterogeneous exposure to infection, individuals more frequent exposed to infection may remain seropositive for longer (even if treated) because of memory responses and their antibody status will not change upon reinfection (i.e., they will remain positive). There is a potential bias towards underestimating infection rates from seroconversion data -- especially among individuals who are most likely to be (re)infected.

The reviewer raises an important point. We have taken care not to conflate seroconversion rates with infection acquisition rates, relating the former to exposure and the latter to active infection dynamics. We have further clarified in the text that antibody seroconversion refers to exposure and not to active infection in lines 127 – 129, as follows:

“The FoI parameters of acquisition (λ) and reversion/loss (ρ) (for antibody datasets, a marker of exposure: λ_sero and ρ_sero; for antigen datasets, a marker of active infection: λ_inf and ρ_inf) were estimated for each dataset.”

We have also amended the title of Table 1 (see Authors’ response 1.2.), Table 2 and Table 3 to indicate that “Antibody seroconversion and seroreversion rates represent markers of exposure” and “Antigen-based infection acquisition and infection loss rates represent markers of active infection”.

3. If the antibody-detection assay used in the Colombian study (described in a local publication, not available to this reviewer) is not strictly species- and stage-specific, and also short-lived (i.e., if only individuals currently exposed to T. solium cysticerci are positive), the age-related prevalence of "cysticercosis" may have been grossly overestimated in this country.

We thank the reviewer for this question and have detailed in Authors’ responses 1.3 that the López et al., diagnostic used in the Flórez Sánchez et al., 2013 study is indicated as both highly stage- and species-specific (97.6% specific). Therefore, we do not believe that the age-prevalence is overestimated in this study. We agree with the reviewer that this information should be made available in the manuscript, therefore we include more details in the footnote section of Supplementary File 1:

“** Flórez Sánchez et al., (2013) indicate that this diagnostic is suitable to determine exposure through detection of anti-cysticercus immunoglobulin G (IgG) antibodies and was evaluated in Colombian patients to assess cross-reactions with different infectious agents (Taenia saginata, Hymenolepis nana, Echinococcus sp., Fasciola hepatica, Entamoeba histolytica, Ascaris lumbricoides, Mansonella ozzardi, Treponema pallidum, Cryptococcus neoformans and HIV)”.

For the benefit of the reviewer, we have also managed to locate a reference from 1996 describing the assay, which is downloadable from:

https://revistabiomedica.org/index.php/biomedica/article/view/903

Corredor et al., (1996) [Standardization and evaluation of ELISA in dried blood eluates collected on filter paper for the diagnosis of cysticercosis]. Biomedica 16, 131–133 (in Spanish).

4. If prevalence data from selected study sites in a dozen countries cannot be taken as representative of the countries/regions of origin, the study is far from being able to estimate "global force-of-infection trends". Please provide further information regarding the primary studies, in addition to those in Table S1, to help the reader decide whether or not some extrapolation is possible.

We agree and thank the reviewer for this feedback. We have addressed the issue of representativeness of each study (where information is available) under Authors’ response 1.4. In several studies, the authors indicated that the study sites, selected from different areas, were representative of specific socioeconomic factors across a region (e.g., Conlan et al., 2012; Jayaraman et al., 2011). We acknowledge that it was difficult to determine how representative some of the other studies were at country level. Having said that, we strongly contend that our analyses, taken in their entirety, do indicate substantial variation in FoI/seroreversion or infection loss estimates across a range of different epidemiological settings representing the major global endemic areas (e.g., South America, sub-Saharan Africa and Asia). The distribution of study sites is visually presented in Figure S2, which is now Figure 1—figure supplement 1: Geographical distribution of studies with human taeniasis (HTT) and human cysticercosis (HCC) age-(sero) prevalence data included in the final analysis (n = 16) by diagnostic method. For this reason, we believe this study reflects the variation in global trends, and therefore propose modifying the manuscript title as follows:

“Global variation in Force-of-Infection trends for Human Taenia solium Taeniasis/Cysticercosis”.

Reviewer #2 (Recommendations for the authors):

Although the work is well written and very organized, I have some concerns about the model and assumptions used in the analysis.

Equation 3 is different than the one reported by Diggle, 2011. One essential set of parentheses is missing. Please, clarify it.

We thank the reviewer for spotting this. We have amended equation 3 (line 555-556) as follows:${\displaystyle p^{\prime }(a)=(1-sp)+(se+sp-1)\times p(a)}$

The methods section would be improved by presenting the motivations for the models' choice and a brief description of them, including assumptions and limitations. Although the reversible catalytic model (rho>0) is a generalization of the simple form (rho=0), there are two mutually exclusive assumptions: either only conversion or conversion and reversion. Because of it, unless well justified, it seems unlike that a particular pattern of an infectious disease could be described considering two fundamentally different biological processes. As observed in Figures2, 3 and 4, the data sometimes support the simple model, sometimes support the reversible, and sometimes it is not possible to decide. It means to me that it is not possible to decide between both designs which encapsulates the mechanism driving the disease age-prevalence presented by the data. Please, clarify it.

The comments raised by the reviewer are well received. There is indeed paucity of information and data available in the wider literature to understand (a) whether and where antibody seroreversion occurs; (b) the potential degree of variation in seroreversion rates between settings; (c) how long, on average, taeniasis and cysticercosis persist in humans in different epidemiological settings (estimated in our work by taking the reciprocal of infection loss rates), and (d) whether a lack of infection loss better explains trends in age-(sero)prevalence profiles. These are key knowledge gaps that are not described/explored in detail elsewhere. Our modelling approach, albeit more phenomenological rather than truly mechanistic, provides a start to identify potentially underlying drivers of epidemiological patterns. However, we agree that we should describe better our rationale / motivation to justify the approach taken to investigate two models which include these two different biological processes as follows (lines 537 – 546):

“These two model configurations represent fundamentally different processes (i.e., presence or absence of Ab-seroreversion or infection loss). Therefore, it was the intention of this analysis to explore which of these processes best captures available age-(sero)prevalence data across different epidemiological settings, given the current knowledge gaps in the literature. In particular, there is minimal knowledge relating to how long taeniasis and cysticercosis infections and seropositivity persist in human hosts, on average (estimated here by the reciprocal of infection loss and Abseroreversion rates). This enabled us to consider the possibility that transmission intensity drives the presence and extent of infection loss and Ab-seroreversion rates (due to transient antigen responses and/or partial parasite establishment, as explored in the Discussion).”

We then postulate possible explanations for a stronger degree of infection loss rates that relate to a higher proportion of transient antigen responses/partial establishment of infection in higher transmission settings in the Discussion (lines 432 – 440), and the need for further investigation into this area.

The reviewer is also correct in highlighting that across Figures 2-4 (note now Figures 3-5), there is not always a definitive answer to which biological mechanism prevails in terms of explaining age-prevalence dynamics with the datasets found for this study. This is generally a reflection of fitting these models to diverse, but often limited data sources (for example where there is minimal information particularly in the early and latter part of the curves (the very young or older groups in the population) to adequately capture the age-prevalence saturation dynamics). Flat age-(sero)prevalence profiles in low prevalence settings, not necessarily representing limited data but more a reflection of specific epidemiological conditions, also proved challenging for model fitting with these types of parsimonious models.

However, some trends emerged, particularly across human cysticercosis age(sero)prevalence profiles, in which the majority (71%) of the studies derived from the general systematic literature search indicated that the model with antibody seroreversion or infection loss (reversible model) was preferable across epidemiological settings. (This is 10 out of 14 studies, excluding the Colombian data, but it was also true for the models fitted to all Colombian departments.) This was a key observation also when models were jointly fitted across multiple datasets (e.g., LLGPEITB antibody assay and B158/B60 Ag-ELISA), therefore providing a strong signal that this mechanism (with antibody seroreversion/infection loss) may be driving the dynamics across the majority of settings.

Another concern is regarding to the saturation of the reversible model. The current formulation assumes that all individuals are exposed/susceptible to the disease and, therefore, its saturation is controlled by the reversion rate parameter (rho). The greater rho, the lower the saturation. Hypothetically, one can think of a situation in which just a fraction of the population is exposed/susceptible to an underlined disease. In this scenario, the greater the fraction of individuals not exposed/susceptible, the greater the interference bias of the reversion rate. I wonder whether the authors believe that this is a reasonable possibility and, in the affirmative case, please acknowledge the limitation of the methodology.

We thank the reviewer for raising this consideration around the behaviour of the reversible model. The catalytic model configurations as presented in this analysis assume an average FoI to which all age groups are subjected. Thus, in terms of potential exposure heterogeneity among individuals or by age, we acknowledge that the current model formulations would not be able to reflect these dynamics. Equally, other heterogeneities in exposure may exist, such as by sex (if for example household roles or occupational practices drive exposure, as the case with other NTDs). Currently there are limited or no data for T. solium to indicate how these heterogeneities may underpin dynamics. We discuss in some detail the limitations with the current models in capturing these more complex FoI dynamics, such as age-dependent exposure patterns, in the paragraph (lines 474 – 492) of the Discussion.

Author details

1. Matthew A Dixon

   1. Department of Infectious Disease Epidemiology and London Centre for Neglected Tropical Disease Research (LCNTDR), Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom
   2. MRC Centre for Global Infectious Disease Analysis, Department of Infectious Disease Epidemiology Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom
   3. SCI Foundation, Edinburgh House, London, United Kingdom

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   For correspondence

   m.dixon15@imperial.ac.uk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1710-6237

2. Peter Winskill

   MRC Centre for Global Infectious Disease Analysis, Department of Infectious Disease Epidemiology Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom

   Contribution

   Formal analysis, Supervision, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

3. Wendy E Harrison

   SCI Foundation, Edinburgh House, London, United Kingdom

   Contribution

   Supervision, Writing – review and editing

   Competing interests

   No competing interests declared

4. Charles Whittaker

   1. Department of Infectious Disease Epidemiology and London Centre for Neglected Tropical Disease Research (LCNTDR), Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom
   2. MRC Centre for Global Infectious Disease Analysis, Department of Infectious Disease Epidemiology Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom

   Contribution

   Software, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

5. Veronika Schmidt

   1. Department of Neurology, Center for Global Health, Technical University Munich (TUM), Munich, Germany
   2. Centre for Global Health, Institute of Health and Society, University of Oslo, Oslo, Norway

   Contribution

   Formal analysis, Writing – review and editing

   Competing interests

   No competing interests declared

6. Astrid Carolina Flórez Sánchez

   Grupo de Parasitología, Instituto Nacional de Salud, Bogotá, Colombia

   Contribution

   Data curation, Writing – review and editing

   Competing interests

   No competing interests declared

7. Zulma M Cucunuba

   1. Department of Infectious Disease Epidemiology and London Centre for Neglected Tropical Disease Research (LCNTDR), Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom
   2. MRC Centre for Global Infectious Disease Analysis, Department of Infectious Disease Epidemiology Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom

   Present address

   Departamento de Epidemiología Clínica y Bioestadística, Facultad de Medicina, Pontificia Universidad Javeriana, Bogotá, Colombia

   Contribution

   Data curation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8165-3198

8. Agnes U Edia-Asuke

   Ahmadu Bello University, Zaria, Nigeria

   Contribution

   Data curation, Writing – review and editing

   Competing interests

   No competing interests declared

9. Martin Walker

   Department of Pathobiology and Population Sciences and London Centre for Neglected Tropical Disease Research (LCNTDR), Royal Veterinary College, Hatfield, United Kingdom

   Contribution

   Formal analysis, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

10. María-Gloria Basáñez

    1. Department of Infectious Disease Epidemiology and London Centre for Neglected Tropical Disease Research (LCNTDR), Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom
    2. MRC Centre for Global Infectious Disease Analysis, Department of Infectious Disease Epidemiology Faculty of Medicine, School of Public Health, Imperial College London, London, United Kingdom

    Contribution

    Formal analysis, Supervision, Methodology, Writing – original draft, Writing – review and editing

    Competing interests

    No competing interests declared

• 715

  Page views

• 139

  Downloads

• 3

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.