health · family: it worked in the lab
curablebut unconfirmable
Buruli ulcer now has a simple oral cure but no way to confirm diagnosis in the field where patients live
Problem statement
Buruli ulcer, caused by Mycobacterium ulcerans, is a neglected tropical disease reported in 33 countries that causes progressive skin destruction and disability if untreated. A paradigm shift to simple 8-week oral antibiotic treatment (rifampicin + clarithromycin) has made the disease curable — but the gold-standard confirmatory test is PCR, which is run at reference laboratories that are usually in cities, far from the remote communities of sub-Saharan Africa where the burden is highest. Samples must be shipped to reference laboratories, and WHO says the turnaround time is too long to guide early treatment decisions. In one rural district of Ghana, where an arrangement sends samples by public minibus straight to a laboratory in Accra, health workers described waits of 48 hours to about two weeks between taking a sample and learning the result, typically about a week. Meanwhile, WHO requires that at least 70% of reported cases be laboratory-confirmed, so that patients given 8 weeks of antibiotics are true cases. In Ghana, a laboratory series covering 2008 to 2016 tested 2,203 suspected cases and found 1,020 of them (46.3%) positive by IS2404 PCR; its authors call for education on case recognition to prevent over-diagnosis as case numbers fall. The WHO's 2022 TPP calls for a rapid test usable at primary health care level, but none exists.
Why this matters
In sub-Saharan Africa, most of those affected are children under 15; of the 18,449 cases reported to WHO between 2010 and 2017 with an age recorded, 40% were in patients under 15. Without treatment the swelling ulcerates within 4 weeks and bone is occasionally affected, and an estimated 25% of those affected are left with long-term disability. A study of 476 laboratory-confirmed cases at one medical centre in Allada, Benin, found that the delay before diagnosis was statistically related to the form the disease had reached: a median of 32.5 days for lesions that had not ulcerated, 60 days for ulcers (p = 0.009) and 365 days for bone lesions. The WHO road map for 2021–2030 aims to cut the share of cases diagnosed in Category III, the most severe, from 30% to less than 10%. WHO set four programmatic targets in 2013, covering PCR confirmation above 70%, fewer Category III lesions, fewer ulcerative lesions and less limitation of movement; an analysis of the data countries report to WHO found that by 2014 only the movement target had been met, and that as of 2019 none were met on a global average. Reported cases fell to 2,217 in 2017, reported suspected cases numbered 2,713 in 2018 and 1,458 in 2020, and only 14 of the 33 countries regularly report data to WHO. The irony is stark: we now have a simple oral treatment but cannot confirm in the field who needs it.
What’s been tried and why it hasn’t worked
Microscopy (Ziehl-Neelsen staining) is the most widely available method in endemic countries but has low sensitivity. Culture of M. ulcerans requires particular conditions: the bacterium grows best at 29–33 degrees C and needs a low (2.5%) oxygen concentration. It is also slow: a systematic review of the diagnostic literature reports that the bacterium takes 6 to 8 weeks to grow, that cultures are usually positive within 9 to 12 weeks of incubation, that incubation has sometimes run to 9 months, and that culture sensitivity is about 35–60%. IS2404-targeted PCR is considered the gold standard, but the same review calls it expensive and dependent on a sophisticated laboratory, trained staff and strict quality control, and it is run at reference laboratories, which are usually in cities and far from the affected areas. The WHO Buruli Ulcer Laboratory Network for Africa (established 2019) has strengthened PCR capacity in 9 countries, but specimens still must be transported to reference sites, creating delays that undermine the benefit of having a simple oral cure.
What would unlock progress
WHO's Diagnostic Technical Advisory Group named three approaches as diagnostic needs at its first meeting in 2019: (1) LAMP (loop-mediated isothermal amplification) and (2) recombinase polymerase amplification (RPA), as design-locked tests to replace home-brewed PCR methods; and (3) a rapid point-of-care test targeting mycolactone, the unique toxin of M. ulcerans. The TPP sets the minimum bar for any of them: sensitivity above 65%, specificity above 90%, a same-day result, no cold chain, and battery power at most. Published results for the first two are near that bar. In Ghana, a LAMP protocol that used a syringe for DNA extraction and a disposable pocket warmer for heating, so that no electricity was needed, gave 83.6% sensitivity and 100% specificity on 64 samples against reference PCR; an earlier study of a freeze-dried LAMP format put its cost at US$1–2 per reaction against US$8–10 for a PCR reaction. An RPA assay tested on samples from 67 patients with suspected Buruli ulcer and 12 patients with other lesions, also in Ghana, gave 88% sensitivity (95% CI 77–95) and 100% specificity, amplifying its target in 15 minutes at a constant 42 degrees C. The mycolactone test is further back: a prototype lateral flow assay has been used on a single patient in Japan, returns a result in about two hours, and is still being tested for efficacy rather than sold. A student team could contribute to the RPA or LAMP platform optimization for field conditions.
Entry points for student teams
A team could optimize a battery-free isothermal amplification protocol (LAMP or RPA) for M. ulcerans detection from skin swabs, focusing on the sample preparation step — which is the primary bottleneck for field deployment of molecular assays. The engineering challenge is extracting sufficient DNA from swab specimens without centrifugation, pipetting, or cold reagents. Alternatively, a team could evaluate the diagnostic accuracy of mycolactone detection via lateral flow using synthetic mycolactone analogs as positive controls. Relevant disciplines: biomedical engineering, microbiology, tropical medicine, point-of-care diagnostics.
Genes
Structural cousins: stuck the same way, in other fields
Sources
WHO, "Target product profile for a rapid test for diagnosis of Buruli ulcer at the primary health-care level," 2022, ISBN 9789240043251. WHO Buruli Ulcer Fact Sheet, 12 January 2023. ); Omansen TF, Erbowor-Becksen A, Yotsu R, van der Werf TS, Tiendrebeogo A, Grout L, Asiedu K, "Global Epidemiology of Buruli Ulcer, 2010–2017, and Analysis of 2014 WHO Programmatic Targets," Emerging Infectious Diseases, 25(12):2183-2190, 2019, accessed 2026-09-29; Yeboah-Manu D, Aboagye SY, Asare P, Asante-Poku A, Ampah K, Danso E, Owusu-Mireku E, Nakobu Z, Ampadu E, "Laboratory confirmation of Buruli ulcer cases in Ghana, 2008-2016," PLOS Neglected Tropical Diseases, 12(6):e0006560, 2018, accessed 2026-09-29; Sakyi SA, Aboagye SY, Darko Otchere I, Yeboah-Manu D, "Clinical and Laboratory Diagnosis of Buruli Ulcer Disease: A Systematic Review," Canadian Journal of Infectious Diseases and Medical Microbiology, 2016:5310718, 2016, accessed 2026-09-29; Capela C, Sopoh GE, Houezo JG, Fiodessihoué R, Dossou AD, Costa P, Fraga AG, Menino JF, Silva-Gomes R, Ouendo EM, Rodrigues F, Pedrosa J, "Clinical Epidemiology of Buruli Ulcer from Benin (2005-2013): Effect of Time-Delay to Diagnosis on Clinical Forms and Severe Phenotypes," PLOS Neglected Tropical Diseases, 9(9):e0004005, 2015, accessed 2026-09-29; Ahortor EK, Gwira TM, Mahazu S, Erber AC, Ablordey A, "Evaluation of an electricity-independent method for IS2404 Loop-mediated isothermal amplification (LAMP) diagnosis of Buruli ulcer in resource-limited settings," PLOS Neglected Tropical Diseases, 18(8):e0012338, 2024, accessed 2026-09-29; Beissner M, Phillips RO, Battke F, Bauer M, Badziklou K, Sarfo FS, Maman I, Rhomberg A, Piten E, Frimpong M, Huber KL, Symank D, Jansson M, Wiedemann FX, Banla Kere A, Herbinger KH, Löscher T, Bretzel G, "Loop-Mediated Isothermal Amplification for Laboratory Confirmation of Buruli Ulcer Disease—Towards a Point-of-Care Test," PLOS Neglected Tropical Diseases, 9(11):e0004219, 2015, accessed 2026-09-29; Frimpong M, Ahor HS, Abd El Wahed A, Agbavor B, Sarpong FN, Laing K, Wansbrough-Jones M, Phillips RO, "Rapid detection of Mycobacterium ulcerans with isothermal recombinase polymerase amplification assay," PLOS Neglected Tropical Diseases, 13(2):e0007155, 2019, accessed 2026-09-29; Sakakibara Y, Konishi M, Ueno T, Murase C, Miyamoto Y, Ato M, de Souza DK, Biamonte M, Pluschke G, Yotsu RR, "Pilot use of a mycolactone-specific lateral flow assay for Buruli ulcer: A case report from Japan," Journal of Clinical Tuberculosis and Other Mycobacterial Diseases, 36:100469, 2024, accessed 2026-09-29 go to source 1 ↗ go to source 2 ↗ go to source 3 ↗ go to source 4 ↗ go to source 5 ↗ go to source 6 ↗ go to source 7 ↗ go to source 8 ↗ go to source 9 ↗ go to source 10 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
- The treatment-diagnosis inversion (cure exists, confirmation doesn't) makes Buruli ulcer structurally unusual among neglected tropical diseases. Most NTDs face the opposite problem.
- WHO says confirmation is essential to make sure that patients treated with antibiotics for 8 weeks are true cases — differential includes cutaneous leishmaniasis, deep fungal infections, and diabetic ulcers. The Ghana series, in which fewer than half of the suspected cases tested were confirmed by PCR, is the evidence that clinical diagnosis on its own is not enough.
- Mycolactone-based detection is attractive because mycolactone is unique to M. ulcerans and, being secreted, could be detected throughout the lesion.
Reconciliation 2026-09-29: Both cited sources were fetched and read in full on 2026-09-29. The WHO target product profile (14 pages) was read as the PDF linked from the cited page, at https://iris.who.int/server/api/core/bitstreams/2c3ea930-4817-4177-803b-9882ae247074/content. The WHO fact sheet was read at the cited address. Neither source gives sensitivity figures for existing tests or any performance or cost figures for experimental tests, so most of the numbers in the earlier text had no support in them. (1) The Source line dated the fact sheet 2024. The page is dated 12 January 2023, and the Source line now says so. (2) The text said the oral treatment made the disease curable "without surgery". The fact sheet says surgery is still used to speed up healing, so those two words were removed. (3) The text called PCR "the only reliable confirmatory test". The fact sheet lists four laboratory methods that can confirm a case, and the profile calls PCR the gold standard. The text now says gold standard. (4) The text said more than 90% of cases occur in rural African communities. The profile says only that the highest burden is in sub-Saharan Africa. The text now says that. (5) The text gave PCR turnaround as "days to weeks". The fact sheet says the turnaround time is too long to guide early treatment decisions and gives no length. The text now says that. (6) The text cited a Ghana study in which more than 50% of clinically diagnosed cases were misdiagnosed. Neither source mentions such a study, and the fact sheet says experienced health workers can in most cases make a reliable clinical diagnosis. The sentence and the matching Source Note now give the profile's own reason for confirmation: WHO requires at least 70% of reported cases to be laboratory-confirmed, so that patients treated for 8 weeks are true cases. The list of look-alike conditions in that note now follows the fact sheet, which does not list trauma. (7) The text said about 50% of those affected are children under 15. The profile says most of those affected in sub-Saharan Africa are children under 15. (8) The text said diagnostic delay is statistically associated with ulcers and bone infection. The sources give no such statistic. They say that without treatment the lesion ulcerates within 4 weeks, that bone is occasionally affected, and that an estimated 25% of those affected have long-term disability. (9) The text said WHO targets set in 2013 were largely unmet by 2019. Neither source says this. The sentence now gives the road map target that the profile does state: to cut the share of cases diagnosed in Category III from 30% to less than 10% by 2030. (10) The text gave a reported case count of about 2,200 a year and said it understates the true burden. The profile gives 1961 suspected cases in 2016, 2713 in 2018 and 1458 in 2020, and says 14 of the 33 countries regularly report data. It does not estimate the true burden. (11) The text called the treatment "effective, affordable". Neither source gives its cost or describes how well it works, and the fact sheet calls it a "simple oral treatment". The text now says simple. The same sentence said we cannot confirm who needs the treatment. It now says we cannot confirm this in the field, because the sources say PCR at reference laboratories does confirm cases. (12) The text gave microscopy a sensitivity of 30-40%, PCR a sensitivity of 54-84%, culture a time of weeks to months, and a growth temperature of 30-32 degrees C. The sources give no sensitivity figures and no culture time. The profile says microscopy is the most widely available method and has low sensitivity. The fact sheet gives the growth temperature as 29–33 degrees C with a low (2.5%) oxygen concentration. The list of what PCR requires was replaced by what the profile says, that reference laboratories are usually in cities and far from affected areas. (13) The text gave performance and cost figures for LAMP (84-87% sensitivity, 100% specificity, $1-2 per reaction), for RPA (results in about 40 minutes, 88% sensitivity), and said a mycolactone lateral flow test would be the first true rapid diagnostic and is at prototype stage. None of this is in either source. The paragraph now reports what the profile says: the WHO advisory group named these three kinds of test as needs in 2019, and the profile's minimum requirements are sensitivity above 65%, specificity above 90%, a same-day result, no cold chain and battery operation. (14) The Source Note said mycolactone is difficult to produce and that antibody development has been slow. Neither source says so. The note now gives the profile's reason for the target, that mycolactone is secreted and could be detected throughout the lesion. Confirmed against the sources and left as written: the profile's title, year and ISBN; 33 countries; the 8-week treatment with rifampicin and clarithromycin; the laboratory network set up in 2019 for 9 countries; the need to transport specimens to reference laboratories; mycolactone as a toxin unique to the bacterium. Not checked, because the sources do not address them and they are the write-up's own judgment or advice: the word "cure" in the titles, where WHO says "simple oral treatment"; the statement in the first Source Note that most neglected tropical diseases face the opposite problem; and the statement in the entry points that sample preparation is the main bottleneck for field use of molecular tests. These are left as written. Genes untouched.
Reconciliation 2026-09-29 (re-sourcing): The editor asked whether the claims taken out or softened in the previous pass could be put back with primary sources behind them. Ten could, and each was fetched and read in full on 2026-09-29 before anything went back into the text. (1) The Problem Statement had said a Ghana study found more than half of clinically diagnosed cases were misdiagnosed. The nearest real study says something narrower: a laboratory series in Ghana covering 2008 to 2016 tested 2,203 suspected cases and found 1,020 of them, 46.3%, positive by IS2404 PCR, and its authors call for education on case recognition to prevent over-diagnosis as case numbers fall (Yeboah-Manu et al., PLOS Neglected Tropical Diseases, 2018, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6003692/fullTextXML on 2026-09-29). The text now gives that series and its figure instead of the misdiagnosis claim, and the matching Source Note says what it shows. (2) The text had said turnaround was days to weeks without saying where. A 2024 study of an electricity-free LAMP method includes interviews with health workers in a rural district of Ghana, who described waits of 48 hours to about two weeks, typically about a week, with samples sent by public minibus to a laboratory in Accra (Ahortor et al., PLOS Neglected Tropical Diseases, 2024, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC11346967/fullTextXML on 2026-09-29). The text now says that, at that scope, beside WHO's statement that the turnaround is too long. (3) The text had said about 50% of those affected are children under 15. Of the 18,449 cases reported to WHO between 2010 and 2017 with an age recorded, 40% were in patients under 15 (Omansen et al., Emerging Infectious Diseases, 2019, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6874257/fullTextXML on 2026-09-29). The text gives 40% and says what it is a proportion of. (4) The text had said diagnostic delay is statistically associated with ulcers and bone infection. A study of 476 laboratory-confirmed cases at one medical centre in Allada, Benin, found median delays of 32.5 days for lesions that had not ulcerated, 60 days for ulcers and 365 days for bone lesions, with p = 0.009 between the first two (Capela et al., PLOS Neglected Tropical Diseases, 2015, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4565642/fullTextXML on 2026-09-29). The text now reports that one cohort and its numbers. The same paper found delay was not associated with the severer phenotypes, so the text claims nothing about those. (5) The text had said the 2013 WHO targets were largely unmet by 2019. The same epidemiology paper says WHO set four targets in 2013, covering PCR confirmation, Category III lesions, ulcerative lesions and limitation of movement, that in 2014 only the movement target was met, and that in 2019 none were met on a global average. The text now says that, and adds the paper's figure of 2,217 reported cases in 2017. (6) The text had said culture takes weeks to months. A systematic review of Buruli ulcer diagnosis says the bacterium takes 6 to 8 weeks to grow, that cultures are usually positive within 9 to 12 weeks of incubation, that incubation has sometimes run to 9 months, and that culture sensitivity is about 35 to 60%; it also calls PCR expensive and dependent on a sophisticated laboratory, expertise and strict quality control (Sakyi et al., Canadian Journal of Infectious Diseases and Medical Microbiology, 2016, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4931084/fullTextXML on 2026-09-29). Those figures are now in the text, attributed to the review. (7) The text had given LAMP 84 to 87% sensitivity. The 2024 Ghana study gives 83.6% sensitivity and 100% specificity on 64 samples, using a syringe for DNA extraction and a disposable pocket warmer for heating so that no electricity is needed. The text now gives those figures and that setup. (8) The text had put LAMP at about US$1 to 2 a reaction. A 2015 study of a freeze-dried LAMP format gives US$1 to 2 per LAMP reaction against US$8 to 10 per PCR reaction (Beissner et al., PLOS Neglected Tropical Diseases, 2015, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4643924/fullTextXML on 2026-09-29). (9) The text had said RPA gives results in about 40 minutes in a mobile suitcase laboratory. The study of the assay reports 88% sensitivity (95% CI 77 to 95) and 100% specificity on samples from 67 patients with suspected Buruli ulcer and 12 with other lesions in Ghana, with the target amplified in 15 minutes at a constant 42 degrees C, and does not describe a suitcase format (Frimpong et al., PLOS Neglected Tropical Diseases, 2019, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC6373974/fullTextXML on 2026-09-29). The text now says 15 minutes and drops the suitcase. (10) The text had said a mycolactone lateral flow assay remains at prototype stage. A case report from Japan describes a prototype used on one patient, which shows mycolactone within two hours and is still being tested for efficacy rather than sold (Sakakibara et al., Journal of Clinical Tuberculosis and Other Mycobacterial Diseases, 2024, read at https://www.ebi.ac.uk/europepmc/webservices/rest/PMC11338991/fullTextXML on 2026-09-29). All eight new sources are on the Source line after the two WHO documents, which are why the problem is in the collection. Six claims were searched for and stay out. The text used to say the oral treatment made the disease curable without surgery; no source says surgery is no longer used, and the WHO fact sheet says it is used to speed healing. It used to say more than 90% of cases occur in rural African communities; the epidemiology paper reports cases by country and region, not by rural or urban setting, and gives no such figure. It used to give microscopy a sensitivity of 30 to 40% and PCR a sensitivity of 54 to 84%; the systematic review gives neither, reporting instead positivity rates between 40% and 78% for microscopy in Ghana and Benin and calling PCR the most sensitive method without a number, so the text keeps the plain statement that microscopy has low sensitivity. It used to say mycolactone is difficult to produce and that antibody development has been slow; nothing in the sources read says either. The claim that the reported case count understates the true burden, and the word affordable for the treatment, also stay out, because no source read gives a burden estimate or a price. Everything else the previous pass corrected is left as it was. Genes untouched.