health · family: it worked in the lab
years hiding beforethe skin shows
Leprosy could be eliminated in a generation through contact prophylaxis but no test can identify who is subclinically infected
Problem statement
Leprosy infects over 170,000 new people per year, including 9,400 children, with an estimated 4 million additional undiagnosed cases. Single-dose rifampicin post-exposure prophylaxis (SDR-PEP) reduces leprosy risk in contacts by 50-60%, and the LPEP programme has shown that tracing and treating 170,000 contacts could near-eliminate leprosy within a generation if scaled globally. But PEP is currently given blindly to all contacts without knowing infection status, because no field-deployable test exists to detect subclinical M. leprae infection. The WHO's 2023 TPP calls for a point-of-care test with >=81% sensitivity and >=99.5% specificity usable in household and outdoor settings — a test that would transform contact-tracing from mass prophylaxis into targeted intervention.
Why this matters
Leprosy causes irreversible nerve damage, limb deformity, and blindness if not treated early. In 2024, 256 children were diagnosed with Grade 2 disability — indicating years of missed detection during which irreversible damage accumulated. The incubation period runs from months to decades, with an estimated average of about four years for the tuberculoid form and ten years for the lepromatous form (Froes et al. 2022), and infected contacts can remain without symptoms for up to 20 years (Kukkaro et al. 2024); that interval provides a window for intervention, but only if infection can be identified. Without a diagnostic, enhanced PEP regimens (PEP++) being tested for higher-risk contacts cannot be rationally allocated, and the distinction between infected contacts who need treatment and uninfected contacts who don't remains invisible.
What’s been tried and why it hasn’t worked
M. leprae has never been successfully cultured in vitro — unlike virtually every other bacterial pathogen — making standard assay development approaches impossible. Anti-PGL-I antibody tests detect multibacillary patients reliably but miss paucibacillary patients, who mount cell-mediated rather than humoral immune responses. The NDO-LID and LID-1 fusion antigen lateral flow tests show similar multibacillary bias. Multi-biomarker approaches combining 5 markers (CCL4, CRP, IL-10, IP-10, anti-PGL-I IgM) can detect both forms but have not been simplified to field-deployable formats. Transcriptomic signatures (RISK4LEP 4-gene panel, AUC 0.86) can predict leprosy development 4-61 months before onset but require laboratory RNA extraction and analysis. The fundamental challenge is that subclinical M. leprae infection is immunologically ambiguous — distinguishing it from cross-reactive responses in endemic areas requires >=99.5% specificity, which no current biomarker achieves alone.
What would unlock progress
The WHO TPP sets stringent parameters: finger-stick blood or nasal swab, <10-100 microliters, zero-infrastructure conditions (households, outdoor settings), capital costs <$2,000. Meeting this likely requires a multiplex approach combining pathogen-specific markers (anti-PGL-I) with host immune response markers (cytokines, transcriptomic signatures) on a single lateral flow or microfluidic platform. A student contribution could address the biomarker combination problem: systematically testing which minimal set of commercially available markers best discriminates subclinical infection from cross-reactive immunity in an endemic contact population.
Entry points for student teams
A modeling team needs no lab at all: take the WHO TPP's >=81% sensitivity and >=99.5% specificity as variables rather than givens and work out, from the published LPEP contact-tracing and SDR-PEP effectiveness numbers, at what performance a test actually changes who gets prophylaxis — how many contacts a 95%-specific test wrongly clears, what the blanket-PEP strategy costs per case averted versus a test-and-treat one, and whether the TPP's thresholds are the binding ones or whether specificity alone decides. A bench team could build the multiplex lateral flow prototype combining anti-PGL-I IgM with 2–3 host cytokine markers (IP-10, CCL4, CRP) on one finger-stick strip, using purified PGL-I antigen and monoclonal antibodies from BEI Resources' leprosy reagent collection, since the engineering problem — quantitative discrimination on a cheap strip with ambient-stable reagents — is testable on characterized reagents before any patient sample exists. A third team could develop the simplified isothermal assay targeting M. leprae RLEP elements and characterize its limit of detection and inclusivity by spiking BEI's M. leprae genomic DNA (strains NHDP, Br4923, Thai-53) into a nasal-swab matrix; whether nasal carriage tracks infection risk is a multi-year household-contact cohort in an endemic country, so the semester deliverable is the assay plus that cohort study's design. The two bench doors need a BSL-2 molecular lab and a BEI Resources account — free but registration-gated, held by a faculty PI, and the M. leprae materials sit under NIAID's Leprosy Research Support contract. Relevant disciplines: biomedical engineering, immunology, infectious disease, diagnostics design, decision analysis.
Genes
Structural cousins: stuck the same way, in other fields
Sources
WHO, "Target product profile for a diagnostic test to detect Mycobacterium leprae infection among asymptomatic household and familial contacts of leprosy patients," 2023, ISBN 9789240074231. Kukkaro et al., "Target product profiles: leprosy diagnostics," Bull World Health Organ 102:288-295, 2024. Froes, L.A.R., Sotto, M.N., Trindade, M.A.B., "Leprosy: clinical and immunopathological characteristics," An Bras Dermatol, 97(3):338-47, 2022, accessed 2026-09-29 go to source 1 ↗ go to source 2 ↗ go to source 3 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
- The inability to culture M. leprae in vitro is a fundamental biological constraint that distinguishes leprosy diagnostic development from virtually all other bacterial diagnostics.
- The LPEP programme data (170,000 contacts traced, 150,000 treated) provides a compelling natural experiment for evaluating diagnostic-guided vs. mass prophylaxis strategies.
- Structurally related to health-tb-poc-diagnostic-sputum-barrier (diagnostic for a mycobacterial disease that fails in the highest-burden settings) but the leprosy problem is more fundamental — no gold-standard test exists even in centralized laboratories.
- The `failure:tech-limitation-now-resolved` tag reflects that the biomarker identification barrier has been resolved: the RISK4LEP 4-gene transcriptomic panel (AUC 0.86 for predicting leprosy onset 4–61 months ahead) and multi-biomarker combinations (CCL4, CRP, IL-10, IP-10, anti-PGL-I) were validated in the 2020s. The remaining barrier is simplifying these into a field-deployable point-of-care format for low-resource endemic settings.
Reconciliation 2026-08-21: The C37 entry-point triage flagged the second suggestion for asking students to test whether nasal carriage correlates with infection risk in household contacts — multi-year field epidemiology in an endemic country. Confirmed, and a second defect the triage row did not list: the unflagged first suggestion had the same shape of problem, because both original doors were wet-lab doors and the brief offered nothing a team without a molecular lab could start, so it failed the ≥2-doors rule outright rather than in one sentence. Repaired three ways: a genuinely facility-free decision-analysis door now leads, treating the WHO TPP's 81%/99% thresholds as the thing to test rather than assume against published LPEP and SDR-PEP numbers already cited in this brief; both bench doors are retargeted onto characterized reagents so they run before any patient specimen exists; the household-contact cohort becomes the validation study to design; and one access line names the BSL-2 lab and the registration-gated reagent account the bench doors require. Resource verified by fetch: BEI Resources' Mycobacterium leprae holdings under NIAID Leprosy Research Support — 62 reagents including genomic DNA for strains NHDP, Br4923 and Thai-53, gamma-irradiated whole cells, purified PGL-I and derivatives, and anti-mycobacterial monoclonals — https://www.beiresources.org/Organism/66/Mycobacterium-leprae.aspx ; the site states only registered users can order, so it is named here as gated, not public.
Reconciliation 2026-09-29: Both cited sources were fetched and read in full on 2026-09-29. The WHO target product profile was read as the 18-page PDF linked from the cited WHO page (https://iris.who.int/server/api/core/bitstreams/edfa39e1-7153-49d4-a348-c39ccc9bf49b/content). The Kukkaro paper was read as the full text held by Europe PMC under PMC10976862. Three changes were made. (1) The text gave the minimum specificity of the test as >=99% in three places: the Problem Statement, What's Been Tried, and the first entry point. The WHO profile, row 3.3, gives 99.5% or higher as the minimum and 99.9% or higher as the ideal, and the Kukkaro paper gives the same figures. The 99% figure in the Kukkaro paper belongs to a different profile, the one for a test that confirms leprosy in people who already have symptoms. All three places now say >=99.5%. The note above this one says "81%/99%"; it is left as written. (2) The text said 289 children were diagnosed with Grade 2 disability in 2024. Neither cited source has figures for 2024. WHO's Global leprosy (Hansen disease) update, 2024 (Weekly Epidemiological Record No 37, 12 September 2025, https://iris.who.int/server/api/core/bitstreams/b4ad088c-c66a-46b9-ac80-30576097b57b/content), fetched the same day, says "In 2024, a total of 256 new child cases were diagnosed with G2D." The text now says 256. (3) The Source line gave the title of the second source as "Target product profiles for leprosy diagnostics." The record gives "Target product profiles: leprosy diagnostics." The title is now copied from the record. Confirmed against the two cited sources and left as written: the WHO title, year and ISBN; the journal, volume, pages and year of the Kukkaro paper; the 50-60% reduction in risk from single-dose rifampicin; the 81% sensitivity; finger-stick blood and nasal swab as sample types; sample volume under 100 microliters at minimum and under 10 at best; use in households and outdoors with no infrastructure; capital cost under US$ 2,000; and the purpose of the test, which is to decide which contacts need a stronger preventive regimen. Several figures in this write-up are in neither cited source. They were checked against other documents fetched the same day and left as written. The same WHO update reports 172,717 new cases and 9,397 new child cases for 2024. Smith et al., "The Missing Millions: A Threat to the Elimination of Leprosy" (PLoS Negl Trop Dis, 2015, PMC4408099) estimates that the gap between expected and detected cases would pass 4 million by 2020. The abstract of the LPEP paper (Richardus et al., Lancet Glob Health, 2021, PMID 33129378) reports 174,782 contacts traced and screened and 151,928 given rifampicin. The RISK4LEP paper (Tió-Coma et al., EBioMedicine, 2021, PMC8182229) reports a 4-gene signature with an AUC of 86.4% in people who developed leprosy 4 to 61 months later. The abstract of van Hooij et al. (EBioMedicine, 2019, PMC6796558) names the five markers CCL4, CRP, IL-10, IP-10 and anti-PGL-I IgM. Scollard et al. (Clin Microbiol Rev, 2006, PMC1471987) says M. leprae has never been grown on artificial media. The WHO leprosy topic page (https://www.who.int/health-topics/leprosy) says symptoms can appear within one year or take 20 years or more. NOT checked, and left as written: that the LPEP programme showed leprosy could be nearly eliminated within a generation, which also stands behind the main title. The LPEP abstract describes a feasibility study that counted contacts traced, screened and treated, and the full paper is behind a subscription. Also not checked: the average incubation period of 2-5 years and the five years in the short title; that the NDO-LID and LID-1 tests favour multibacillary cases; that the multi-marker tests have not been made field-deployable; that no single biomarker reaches the required specificity; and the name "PEP++" for the stronger preventive regimens. None of the documents read gives these. On the multi-marker tests, the two van Hooij papers describe lateral flow strips tried with finger-stick blood in the field, so that sentence may need another look by someone who has the full papers. Later the same day the editor asked for the five-year figure in the short title to be checked. Neither cited source gives an average incubation period. Kukkaro et al. say infected contacts "can remain asymptomatic for up to 20 years," citing among others Froes, Sotto and Trindade 2022, which was read in full from the PubMed Central copy on 2026-09-29 and says: "The incubation period for leprosy is long, ranging from months to decades, with an estimated average time of 10 years for the lepromatous form and four years for the tuberculoid form." The short title now says "Years Hiding Before the Skin Shows", the sentence in Why This Matters follows the two sources, and Froes et al. was added to the Source line. Genes untouched.