New Therapies and Biomarkers for Chagas Infection (TESEO)

Fundación Ciencia y Estudios Aplicados para el Desarrollo en Salud y Medio Ambiente (CEADES), Barcelona Institute for Global Health, Institute of Parasitology and Biomedicine Lopez Neyra, U.S. Food and Drug Administration (FDA), Drugs for Neglected Diseases, Mundo Sano Foundation, National Institute of Allergy and Infectious Diseases (NIAID)

Chagas disease (CD) is an endemic zoonotic disease with a significant global impact. Current approved treatments for CD (benznidazole (BZN) and nifurtimox (NFX)) were developed in the 1970s with regimens and dosing intervals derived from decades-old patient series and with very limited direct comparisons. Treatment recommendations vary significantly from country to country and the comparative evidence-base with the current treatment regimens is limited. The reported efficacy of both drugs in patients with T. cruzi infection is variable and depends on the disease stage, the drug dose, the age of patients, and the infecting T. cruzi strain or genotype. Due to a therapeutic failure of at least 20% after 12 months in chronic patients and the high rate of adverse events, together with the recent data that suggest that we may be overdosing patients, we propose to test new dosing regimens of these two old compounds. Hypotheses: Lowering the frequency of drug dosing of BZN and NFX, the plasma drug levels of the drugs within the therapeutic range will be maintained. The duration of treatment with BZN or NFX may be related to the effectiveness of these drugs. Blood levels of the proposed biomarkers will significantly diminish or became negative after a relatively short interval after treatment.

Chagas disease (CD) is an endemic zoonotic disease caused by the protozoan parasite, T. cruzi. It affects 8-10 million people in Latin America and is a worldwide public health issue due to migratory flows. CD has a significant economic impact. Recently, a study showed that the global costs for CD are US$7-19 billion per year, similar or even higher to those of other important diseases such as rotavirus infection or cervical cancer. Treatment of chronic CD (CCD) has been hampered, unlike other illnesses, by the paramount importance given to the autoimmune theory of the disease that prevailed for many years. As a result, several generations of health professionals were trained in the belief that CCD had no treatment. As a consequence, currently, most (>99%) chronically infected people are still not treated with specific antiparasitic drugs, and the research and development for new, more effective drugs was overlooked for many years, until very recently. Nowadays, the key role of the parasite persistence in the pathophysiology of CD is recognized, as well as the need for specific treatment. Current approved specific treatments for CD include nifurtimox (NFX) and benznidazole (BZN) and the recommended dosing regimens are 5 mg/Kg/day divided into two doses (2.5 mg/Kg b.i.d) given for 60 days for BZN, and 8 mg/Kg divided into three equal daily doses (2.7 mg/Kg t.i.d.) given for 90 days for NFX. The efficacy of both drugs in patients with T. cruzi infection is highly variable and depends on the disease stage, the drug dose, the age of patients, and the infecting T. cruzi strain or genotype. Moreover, the high rate of adverse events hampers their standard use in the field. Recent studies show that at the current doses of both drugs, more than 70% of patients suffer mild/moderate reactions and around 10-27% experience serious ones, forcing patients to stop the treatment and take appropriate medications for the adverse events. Data on the pharmacokinetics (PK) of BZN and NFX are limited and there are no recent data on PK of NFX in adults with chronic CD. Moreover, due to a lack of early BMKs of therapeutic efficacy, the true efficacy of these drugs remains unknown. Seroconversion using conventional serology (CS) is often long-term (~10-20 years) or incomplete, and a reduction in T. cruzi-specific antibody titers often takes many years, rendering the evaluation of response to treatment insensitive and lengthy, and therefore impractical in clinical settings. The need for new, safer, and more efficacious drugs against T. cruzi as well as early BMKS of therapeutic efficacy are the major challenges in the treatment of CD, particularly in chronic adults. With this project, the investigators aim to achieve specific knowledge about the safety and efficacy of new dosing regimens for BZN and NFX. The proposed new regimens for these drugs are based on recent data that suggest that with half of the dosing frequency the levels of BZN can be maintained in the therapeutic range of this drug, which could conceivably reduce the appearance of adverse events while maintaining antiparasitic efficacy. At the same time, the investigators plan to evaluate whether the drug efficacy will be maintained if the investigators reduce the length of treatment with BZN or NFX to 30 days. Furthermore, the investigators also plan to evaluate whether the efficacy of the treatment with BZN or NFX is improved by increasing its duration to 90 days and to evaluate novel potential BMKs of response to specific treatment and eventual parasitological cure in CCD patients. The information obtained in this study would also allow for better-designed clinical trials with drug combinations, in which NFX and BZN will have a central role. The results will be disseminated via publications in peer-reviewed journals, conferences, and reports to the NIH, FDA, and participating institutions. The investigators of this study are aware of and have agreed to abide by the principles for sharing research resources as described by NIH in "Principles and Guidelines for Recipients of NIH Research Grants and Contracts on Obtaining and Disseminating Biomedical Research Programs." Accordingly, resources developed in this study will be available to the scientific community as soon as the intellectual property of these resources and/or research tools have been protected or disclosed in publications. In the event that a specific research tool is requested from the TESEO investigators and is available, it will be shared with members of the scientific community. Data sharing not applicable as no datasets have been generated and/or analysed for this study yet. However, once the datasets resulting from this study are available, they will be disseminated via publications in peer-reviewed journals, national and international conferences, and reports to the NIH, FDA, and participating institutions.

Chagas disease, Trypanosoma cruzi, Chemotherapy, Benznidazole, Nifurtimox, Biomarkers, Bolivia, Antiparasitic Agents, Parasitic Diseases

Three of 6 parallel groups will receive one of three different oral benznidazole (BZN) dosing regimens and the other three will receive one of the three different nifurtimox (NFX) regimens.

The primary efficacy endpoint or measure is a binary 'cured' (success), 'not-cured' (failure) variable based on a total of eight qPCR timepoints from end-of-treatment (EOT) up to 36 months of follow-up. Each of the timepoints includes a total of three sequential qPCR examinations on blood samples collected during one visit. Therefore, for a patient to be considered "cured", a total of 24 negative qPCR results should be documented. Blood samples to be collected at EOT (30, 60, or 90 days), and at 4, 6, 12, 18, 24, 30 and 36 months of follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

Evaluate changes over time in blood parasitic load as measured by qPCR. The unit for the qPCR is parasite equivalents/milliliter of blood (Par. Eq./mL). The limit of detection (LOD) and limit of quantification (LOQ) for the qPCR assay is 0.69 and 1.5 Par. Eq./mL blood, respectively. Kaplan-Meier survival curves will be drawn to depict the time-to-reappearance of blood parasitic load for patients who have cleared blood parasitic load at EOT (as such this is a time-to-relapse for 'success' patients) across treatment groups. A discrete-time survival model will be used to analyze this event time data. Change in parasite load over time assessed at days 14-17, 59-62, and EOT (30, 60, or 90 days), according to the treatment arm, and months 4, 6, 12, 18, 24, 30, and 36 of follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months), as measured by qPCR.

Days 14-17, 59-62; EOT-30, EOT-60, and EOT-90; and months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

Evaluate changes over time in serological titers and incidence of negative seroconversion through conventional serology. Chagas disease serology commercial kit using a parasite antigenic mixture (lysate) will be used to perform this analysis. Serum samples from all timepoints in the study for each patient will be analyzed in the same ELISA plate to avoid bias due to interassay variations. Generalized linear mixed-effects models will be used to evaluate the conventional serological response over time. The dependent variable in both models will be a binary conventional serology parameter. The treatment arm will be included in the model as a fixed effect. The significance of the treatment covariate will be tested at the 0.05 two-sided significance level with Likelihood Ratio Test.

Day -28; EOT-30, EOT-60, and EOT-90; and months 4, 6, 12, 18, 24, 30, and 36 (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months) during follow-up.

Evaluate changes over time in serological titers and incidence of negative seroconversion through conventional serology. Chagas disease serology commercial kit using recombinant parasite antigens will be used to perform this analysis. Serum samples from all timepoints in the study for each patient will be analyzed in the same ELISA plate to avoid bias due to interassay variations. Generalized linear mixed-effects models will be used to evaluate the conventional serological response over time. The dependent variable in both models will be a binary conventional serology parameter. The treatment arm will be included in the model as a fixed effect. The significance of the treatment covariate will be tested at the 0.05 two-sided significance level with Likelihood Ratio Test.

Day -28; EOT-30, EOT-60, and EOT-90; and months 4, 6, 12, 18, 24, 30, and 36 (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months) during follow-up.

Evaluate changes over time in the non-conventional serology biomarker (BMK) "Lytic Anti-α-Gal Antibodies" and incidence of negative seroconversion for it, as measured by chemiluminescent (CL)-ELISA using T. cruzi tGPI-mucins as antigens. CL-ELISA unit is the Titer (T). T is the ratio of the relative luminescence units (RLU) of the tested serum to the cutoff value (CV). The CV is calculated by defining the upper prediction limit (as standard deviation, SD), multiplied by a factor according to the number of negative controls (NC) and a confidence interval of 99.5%. For each test plate in which six NC are included, the CV is defined as the NC mean plus 4.355 times the SD. A serum sample is positive when T = or >1.000; negative when T = or < 0.900; and inconclusive when T > 0.900, < 1.000. Generalized linear mixed-effects models (GLMM) will define how this BMK affects binary outcomes of parasite clearance. The BMK variable will be included in the model, with repeated measures.

Days 0, 14-17, 59-62; EOT-30, EOT-60, and EOT-90; and at months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

This BMK is measured by Enzyme-Linked Immunosorbent Assay (ELISA), whose unit is Optical Density (OD) at 492 nm. Cutoff: A negative result will be considered when OD values = or <0.200, and positive when OD values >0.300, at 1/200 serum dilution; indeterminate results will be OD values comprised between 0.200 and 0.300. Therapeutic efficacy will imply a continuous and substantial drop in the reactivity after treatment. A substantial decrease has to be at least 40% compared with the reactivity at T0 (before treatment). A slight increase (= or <20%) in the reactivity against the BMK shortly after treatment versus OD at T0 will be allowed. Generalized linear mixed-effects models (GLMM) will define how this BMK affects binary outcomes of parasite clearance. The BMK variable will be included in the model, with repeated measures.

Days 0, 14-17, 59-62; EOT-30, EOT-60, and EOT-90; and at months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

This BMK is measured by Enzyme-Linked Immunosorbent Assay (ELISA), whose unit is Optical Density (OD) at 492 nm. Cutoff: A negative result will be considered when OD values = or <0.300, and positive when OD values >0.450, at 1/200 serum dilution; indeterminate results will be OD values comprised between 0.300 and 0.450. Therapeutic efficacy will imply a continuous and substantial drop in the reactivity after treatment. A substantial decrease has to be at least 30% compared with the reactivity at T0 (before treatment). A slight increase (= or < 20%) in the reactivity against the BMK shortly after treatment versus OD at T0 will be allowed.

Days 0, 14-17, 59-62; EOT-30, EOT-60, and EOT-90; and at months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

This BMK is measured by Enzyme-Linked Immunosorbent Assay (ELISA), whose unit is Optical Density (OD) at 492 nm. Cutoff: A negative result will be considered when OD values = or <0.250 at 1/400 dilution, and positive when OD values >0.350, at 1/400 serum dilution; indeterminate results will be OD values comprised between 0.250 and 0.350. Therapeutic efficacy will imply a continuous and substantial drop in the reactivity after treatment. A substantial decrease has to be at least 40% compared with the reactivity at T0 (before treatment). A slight increase (= or <20%) in the reactivity against the BMK shortly after treatment versus OD at T0 will be allowed. Generalized linear mixed-effects models (GLMM) will define how this BMK affects binary outcomes of parasite clearance. The BMK variable will be included in the model, with repeated measures.

Days 0, 14-17, 59-62; EOT-30, EOT-60, and EOT-90; and at months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

This BMK is measured by Enzyme-Linked Immunosorbent Assay (ELISA), whose unit is Optical Density (OD) at 492 nm. Cutoff: A negative result will be considered when OD values = or <0.250, and positive when OD values >0.350, at 1/400 serum dilution; indeterminate results will be OD values comprised between 0.250 and 0.350. Therapeutic efficacy will imply a continuous and substantial drop in the reactivity after treatment. A substantial decrease has to be at least 40% compared with the reactivity at T0 (before treatment). A slight increase (= or <20%) in the reactivity against the BMK shortly after treatment versus OD at T0 will be allowed. Generalized linear mixed-effects models (GLMM) will define how this BMK affects binary outcomes of parasite clearance. The BMK variable will be included in the model, with repeated measures.

Days 0, 14-17, 59-62; EOT-30, EOT-600, and EOT-90; and at months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

Changes Over Time in the Non-Conventional Serology Biomarker "Trypomastigote Excreted/Secreted Antigens (TESA)"

Evaluate changes over time in the non-conventional serology biomarker (BMK) "Trypomastigote Excreted/Secreted Antigens (TESA)" and incidence of negative seroconversion for it, as measured by the aptamer assay. The unit of the aptamer assay is relative fluorescence unit (RFU). A Signal to Cutoff (S/CO) ratio is calculated for each clinical specimen by dividing the RFU value of the test sample by the cutoff (CO). The CO is the highest RFU value obtained from the endemic control specimens included in each run. The S/CO of a positive control included in the same run must meet pre-established acceptance criteria to be valid. A specimen with an S/CO > 1.0 is interpreted as positive for the presence of the BMK (=infected). A specimen with an S/CO < 1 is interpreted as negative (=non-infected). Generalized linear mixed-effects models (GLMM) will define how this BMK affects binary outcomes of parasite clearance. The BMK variable will be included in the model, with repeated measures.

Days 0, 14-17, 59-62; EOT-30, EOT-60, and EOT-90; and at months 4, 6, 12, 18, 24, 30, and 36 during follow-up (with a window period of +/- 7 days from 4 to 12 months and +/- 14 days from 18 to 36 months).

Inclusion Criteria: Adults, 18-50 years. Weight: 88-198 pounds (40-90 Kg). Individuals diagnosed as being infected with T. cruzi by conventional serology (two positive tests with different antigens) with at least one positive qualitative RT-PCR assay out of three during the screening. Patient classified as being in the indeterminate form (without clinical manifestations) or early cardiac form (Kushnir 1) of chronic Chagas disease. Signed informed consent form (ICF). Exclusion Criteria: Clinical signs of dilated cardiomyopathy (dyspnea, legs' edema, syncope, pulmonary crackles). Patients with an EKG showing the following characteristics: sinus tachycardia or atrial fibrillation, ventricular arrhythmias, left atrial enlargement, left bundle-branch block (LBBB) accompanied by right axis deviation (RAD), and/or patients with Calculation of Fridericia's corrected QT interval (QTcF) > 450ms, a formula for calculating the QT interval on an electrocardiogram (ECG). History of Chagas disease treatment with BZN or NFX or any triazole drug(s) in the last five years. Clinical signs and/or symptoms of digestive form of Chagas disease, which is characterized by the presence of two or more of the following criteria *: Excessive exertion in at least 25% of bowel movements Hard stools in at least 25% of stools (type 1-2 of Bristol) Feeling of incomplete evacuation in at least 25% of bowel movements Feeling of obstruction or anorectal block in at least 25% of bowel movements Manual maneuvers to facilitate defecation in at least 25% of bowel movements Less than 3 complete spontaneous stools per week Criteria must be met for at least the last three months and symptoms must have been started for at least six months before diagnosis. Hypersensitivity to the active substances (BZN or NFX) or to the excipient. Previous diagnosis of porphyria. Any other acute or chronic health conditions that in the opinion of the PI, may interfere with the efficacy and/or safety evaluation of the study drug. Formal contraindication to BZN or NFX. Any concomitant or anticipated use of drugs that are contraindicated with the use of BZN or NFX. Individuals currently known to abuse alcohol and/or drugs. Furthermore, if throughout the course of the study the team becomes aware that a participant is using drugs/alcohol that participant will be excluded from the treatment but will continue with the follow-up visits. The study manual outlines how abuse and dependence will be measured for this study. Pregnancy. Females of childbearing potential will be required to complete a pregnancy test prior to enrollment and throughout the course of treatment. Women in reproductive age must have a negative serum pregnancy test at screening, must not be breastfeeding, and consistently use and/or have partner consistently use a highly effective contraceptive method during the entire treatment phase of the trial. Transaminases (alanine aminotransferase-ALT and aspartate aminotransferase- AST). AST must be within the normal range, within an acceptable margin of 25% above the upper limit of normality for both, according to the insert of the biochemical kit being used in this study. Creatinine must be within an acceptable range, within an acceptable margin of 10% above the upper limit of normality, according to the insert of the biochemical kit being used. The normal ranges of transaminases (ALT and AST) and creatinine are defined by the inserts of the commercial biochemical kits selected to be used in the present study. All treatment centers (Chagas Platforms in Cochabamba, Sucre, and Tarija) are going to use the same biochemical kits. The participating clinical laboratories at the Platforms (in Cochabamba, Sucre, and Tarija) will use the Common Terminology Criteria for Adverse Events (CTCAE, v.5.0; ttps://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/ctcae_v5_quick_reference_5x7.pdf). Total bilirubin must be within the normal range, within an acceptable margin of 15% above the upper limit of normality for both sexes, according to the insert of the biochemical kit being used in this study. For other standard exclusion criteria, a detailed explanation for each criterion is provided in the Manual of Operations and Procedures (MOP).

Bern C, Montgomery SP, Herwaldt BL, Rassi A Jr, Marin-Neto JA, Dantas RO, Maguire JH, Acquatella H, Morillo C, Kirchhoff LV, Gilman RH, Reyes PA, Salvatella R, Moore AC. Evaluation and treatment of chagas disease in the United States: a systematic review. JAMA. 2007 Nov 14;298(18):2171-81. doi: 10.1001/jama.298.18.2171.

Bustamante JM, Tarleton RL. Potential new clinical therapies for Chagas disease. Expert Rev Clin Pharmacol. 2014 May;7(3):317-25. doi: 10.1586/17512433.2014.909282. Epub 2014 Apr 9.

Carod-Artal FJ, Gascon J. Chagas disease and stroke. Lancet Neurol. 2010 May;9(5):533-42. doi: 10.1016/S1474-4422(10)70042-9.

Gascon J, Vilasanjuan R, Lucas A. The need for global collaboration to tackle hidden public health crisis of Chagas disease. Expert Rev Anti Infect Ther. 2014 Apr;12(4):393-5. doi: 10.1586/14787210.2014.896194. Epub 2014 Mar 3.

Jackson Y, Alirol E, Getaz L, Wolff H, Combescure C, Chappuis F. Tolerance and safety of nifurtimox in patients with chronic chagas disease. Clin Infect Dis. 2010 Nov 15;51(10):e69-75. doi: 10.1086/656917. Epub 2010 Oct 8.

Kierszenbaum F. Chagas' disease and the autoimmunity hypothesis. Clin Microbiol Rev. 1999 Apr;12(2):210-23. doi: 10.1128/CMR.12.2.210.

Lee BY, Bacon KM, Bottazzi ME, Hotez PJ. Global economic burden of Chagas disease: a computational simulation model. Lancet Infect Dis. 2013 Apr;13(4):342-8. doi: 10.1016/S1473-3099(13)70002-1. Epub 2013 Feb 8.

Pinazo MJ, Thomas MC, Bua J, Perrone A, Schijman AG, Viotti RJ, Ramsey JM, Ribeiro I, Sosa-Estani S, Lopez MC, Gascon J. Biological markers for evaluating therapeutic efficacy in Chagas disease, a systematic review. Expert Rev Anti Infect Ther. 2014 Apr;12(4):479-96. doi: 10.1586/14787210.2014.899150.

Perez-Molina JA, Sojo-Dorado J, Norman F, Monge-Maillo B, Diaz-Menendez M, Albajar-Vinas P, Lopez-Velez R. Nifurtimox therapy for Chagas disease does not cause hypersensitivity reactions in patients with such previous adverse reactions during benznidazole treatment. Acta Trop. 2013 Aug;127(2):101-4. doi: 10.1016/j.actatropica.2013.04.003. Epub 2013 Apr 11.

Rassi A Jr, Rassi A, Marin-Neto JA. Chagas disease. Lancet. 2010 Apr 17;375(9723):1388-402. doi: 10.1016/S0140-6736(10)60061-X.

Regueiro A, Garcia-Alvarez A, Sitges M, Ortiz-Perez JT, De Caralt MT, Pinazo MJ, Posada E, Heras M, Gascon J, Sanz G. Myocardial involvement in Chagas disease: insights from cardiac magnetic resonance. Int J Cardiol. 2013 Apr 30;165(1):107-12. doi: 10.1016/j.ijcard.2011.07.089. Epub 2011 Sep 9.

Soy D, Aldasoro E, Guerrero L, Posada E, Serret N, Mejia T, Urbina JA, Gascon J. Population pharmacokinetics of benznidazole in adult patients with Chagas disease. Antimicrob Agents Chemother. 2015;59(6):3342-9. doi: 10.1128/AAC.05018-14. Epub 2015 Mar 30.

Tarleton RL. Parasite persistence in the aetiology of Chagas disease. Int J Parasitol. 2001 May 1;31(5-6):550-4. doi: 10.1016/s0020-7519(01)00158-8.

Tarleton RL, Reithinger R, Urbina JA, Kitron U, Gurtler RE. The challenges of Chagas Disease-- grim outlook or glimmer of hope. PLoS Med. 2007 Dec;4(12):e332. doi: 10.1371/journal.pmed.0040332.

Urbina JA, Docampo R. Specific chemotherapy of Chagas disease: controversies and advances. Trends Parasitol. 2003 Nov;19(11):495-501. doi: 10.1016/j.pt.2003.09.001. No abstract available.

Viotti R, Vigliano C, Lococo B, Alvarez MG, Petti M, Bertocchi G, Armenti A. Side effects of benznidazole as treatment in chronic Chagas disease: fears and realities. Expert Rev Anti Infect Ther. 2009 Mar;7(2):157-63. doi: 10.1586/14787210.7.2.157.

Viotti R, Alarcon de Noya B, Araujo-Jorge T, Grijalva MJ, Guhl F, Lopez MC, Ramsey JM, Ribeiro I, Schijman AG, Sosa-Estani S, Torrico F, Gascon J; Latin American Network for Chagas Disease, NHEPACHA. Towards a paradigm shift in the treatment of chronic Chagas disease. Antimicrob Agents Chemother. 2014;58(2):635-9. doi: 10.1128/AAC.01662-13. Epub 2013 Nov 18.

Zhang L, Tarleton RL. Parasite persistence correlates with disease severity and localization in chronic Chagas' disease. J Infect Dis. 1999 Aug;180(2):480-6. doi: 10.1086/314889.

Schelldorfer, J., Meier, L., and Buhlmann, P. (2014). GLMMLasso: An algorithm for high-dimensional generalized linear mixed models using l1-penalization. J Comput Graph Stat 23, 460-477.

Alonso-Vega C, Urbina JA, Sanz S, Pinazo MJ, Pinto JJ, Gonzalez VR, Rojas G, Ortiz L, Garcia W, Lozano D, Soy D, Maldonado RA, Nagarkatti R, Debrabant A, Schijman A, Thomas MC, Lopez MC, Michael K, Ribeiro I, Gascon J, Torrico F, Almeida IC. New chemotherapy regimens and biomarkers for Chagas disease: the rationale and design of the TESEO study, an open-label, randomised, prospective, phase-2 clinical trial in the Plurinational State of Bolivia. BMJ Open. 2021 Dec 31;11(12):e052897. doi: 10.1136/bmjopen-2021-052897.