Open Forum Infectious Diseases

SUPPLEMENT ARTICLE

Diagnostics for Typhoid Fever: Current Perspectives and

Future Outlooks for Product Development and Access
Jyotshna Sapkota,1,2 Tamalee Roberts,3,4 Buddha Basnyat,5 Stephen Baker,6,7 Lee M. Hampton,8 and Sabine Dittrich1,4,9
1
FIND, the Global Alliance for Diagnostics, Geneva, Switzerland, 2Department of Microbiology, Nepal Medical College, Kathmandu, Nepal, 3Lao-Oxford-Mahosot Hospital-Wellcome Trust Research
Unit, Microbiology Laboratory, Mahosot Hospital, Vientiane, Lao People’s Democratic Republic, 4Centre for Tropical Medicine and Global Health, Nuffield Department of Medicine, University of
Oxford, Oxford, United Kingdom, 5Oxford University Clinical Research Unit, Kathmandu, Nepal, 6University of Cambridge School of Clinical Medicine, Cambridge Biomedical Campus, Cambridge,
United Kingdom, 7Department of Medicine, University of Cambridge School of Clinical Medicine, Cambridge Biomedical Campus, Cambridge, United Kingdom, 8Gavi, the Vaccine Alliance, Geneva,
Switzerland, and 9Deggendorf Institute of Technology, European Campus Rottal-Inn, Pfarrkirchen, Germany

Downloaded from https://academic.oup.com/ofid/article/10/Supplement_1/S17/7188897 by guest on 29 August 2023

Typhoid is an enteric disease caused by Salmonella Typhi. Like many febrile illnesses, typhoid presents with nonspecific symptoms.
In routine healthcare settings in low- and middle-income countries, typhoid fever is suspected and treated empirically. Though
many diagnostic tests are available for typhoid diagnosis, there are currently no diagnostic tests that meet ideal requirements for
sensitivity, specificity, speed, and cost-effectiveness. With introduction of typhoid conjugate vaccine, it is essential to explore the
current and future typhoid approach in the context of use case and access to ensure their utilization for disease control.
Keywords. point of care tests; rapid diagnostic tests; Salmonella Typhi; typhoid diagnostics.

Regional estimates of the burden of typhoid fever cannot be for typhoid and explore new technological and access develop­
measured accurately without improved disease diagnostics; ments that could improve the diagnostic landscape in the fu­
this lack of diagnostics and data impacts the ability of govern­ ture and in the context of existing target product profile
ments to plan and appropriately intervene. Given the need for drafts. We explore different areas of typhoid diagnostic chal­
disease control, funding typhoid diagnostic capacity, including lenges, including current shortcomings in an example setting,
availability and use of improved typhoid test kits, should be in­ expanding understanding of existing rapid diagnostic tests
creased, especially where the incidence of typhoid is unknown (RDTs), the future of diagnostics as part of surveillance, and
[1]. Challenges regarding typhoid diagnostics may also impact access-related considerations that aim to improve the availabil­
the implementation of new-generation typhoid vaccines ity of quality diagnostic tests in the future.
in endemic regions due to lack of surveillance tools [2].
Multidrug resistance (MDR) has been spread intercontinen­
tally due to an increase in travel connectivity, affecting those CHALLENGES IN TYPHOID DIAGNOSIS: EXAMPLE
living in endemic regions and travelers alike [3, 4]. Notably, FROM LAOS
multidrug-resistant and fluoroquinolone-resistant variants The Lao People’s Democratic Republic (Laos) is a landlocked
of Salmonella enterica subspecies enterica serovar Typhi country in Southeast Asia with a population of approximately
(S. Typhi) may be associated with more severe disease with po­ 7 million people [5]. Typhoid fever is endemic in Laos, but
tentially adverse outcomes, therefore creating clinical manage­ there are limited epidemiological data. In a hospital-based
ment challenges [3]. The spread of drug-resistant organisms as study examining blood cultures at Mahosot hospital in the cap­
well as an expected reemergence of typhoid in currently nonen­ ital, Vientiane, between 2000 and 2018, there were a total of 913
demic settings due to climate change make improved diagnos­ culture-confirmed typhoid cases from just over 60 000 blood
tics key for tracking incidence to inform public health policy in cultures (∼1.5% positivity). Most cases originated in rural areas
additional to ensuring individual patients get appropriate treat­ with the majority of patients recruited into research studies;
ment. Here, we aim to review gaps in the diagnostic landscape there were limited specimen requests outside of these studies,
particularly outside of the capital city [6].
These data suggest that the amount of typhoid in Laos is un­
Correspondence: Jyotshna Sapkota, MD, Scientist, Malaria and Fever Team, FIND, the
Global Alliance for Diagnostics, Campus Biotech, Chemin des Mines 9, 1202, Geneva,
derestimated; detection relies on blood culture, with the labora­
Switzerland. tory capacity to process blood cultures being limited outside of
Open Forum Infectious Diseases® Vientiane (capacity is increasing as a consequence of a UK
© The Author(s) 2023. Published by Oxford University Press on behalf of Infectious Diseases
Society of America. This is an Open Access article distributed under the terms of the Fleming Fund grant). Provincial hospital laboratories in the
Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which network perform manual blood cultures in-house and then
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original
work is properly cited.
send positive cultures, including suspected S. Typhi, to a
https://doi.org/10.1093/ofid/ofad120 central government or reference laboratory in Vientiane for

Diagnostics for Typhoid Fever • OFID 2023:10 (Suppl 1) • S17

identification or confirmation. The shipping process can take used test to diagnose typhoid despite a low performance (sensitiv­
several days mostly due to infrastructure challenges (eg, lack ity range, 57%–74%; specificity range, 43%–83% [8]) reported in
of transportation/staff, unpassable roads) with blood culture several studies. A 2017 Cochrane review summarized the evi­
bottles sitting at room temperature without any additional tem­ dence on diagnostic accuracy of available RDTs for enteric (ty­
perature regulation. This leads to isolates not being recovered phoid) fever (mostly TUBEX, Typhidot, and KIT Test-It) [9].
in the reference laboratory in Vientiane due to high tempera­ The result of the meta-analysis found TUBEX to have an average
tures and transportation conditions. A previous study recorded sensitivity of 78% and a specificity of 87%; Typhidot had an aver­
temperatures as high as 41°C in a transportation box used for age sensitivity of 84% and a specificity of 79%, and Test-It (KIT)
sample shipment [7]. The cost and availability of confirmation was found to have an average sensitivity of 69% and a specificity
tests also impacts the diagnostic result, as high costs of identi­ of 90% [8]. Numerous studies had been conducted to assess com­
fication reagents, antisera, and shipment may be prohibitive for mercial diagnostic tests for typhoid [8]; however, key opinion
laboratories in low- and middle- income countries (LMICs); for leaders highlighted that these studies are difficult to compare

Downloaded from https://academic.oup.com/ofid/article/10/Supplement_1/S17/7188897 by guest on 29 August 2023

example, API 20E identification strips cost approximately US$6 due to a lack of comparable case definitions and a lack of geo­
per test and antisera cost approximately $3 per test and can take graphical diversity. To address these data shortcomings, FIND
months to be shipped to the country. Blood culture is recog­ established a head-to-head comparison of commercial typhoid
nized to be the gold standard for typhoid diagnosis; however, tests and simultaneously generated a sample set that could be
it is a complex and expensive process as highlighted by the ex­ used in the evaluation of emerging technologies [10]. Typhoid
ample from Laos. In this example use case, financial and logistic positives and negatives were analyzed in both regions with 205
support can be provided due to research affiliation of the labo­ positives and 205 negatives from Asia and 59 positives and 59
ratories and even here things are not working in a way that pro­ negatives from Africa. Nine different RDTs were evaluated
vides reliable, high-quality data. The result of such imperfect against blood culture as the reference standard. The tests used
data is the underestimation of typhoid cases nationally but were SD Bioline Salmonella Typhi immunoglobulin G (IgG)/im­
also in the areas with access to research network. munoglobulin M (IgM), Typhidot Rapid IgG/IgM, Enterocheck
This example from Laos shows that without local capacity WB, Test-It Typhoid IgM, CTK Typhoid IgG/IgM Combo Rapid
and appropriate (cheap, long shelf life, simple to use) diagnostic Test CE, Spectrum Typhoid IgG/IgM Rapid cassette, TUBEX-TF,
tests, provincial hospitals have to continue to rely on sample Diaquick S.Typhi/Paratyphi antigen cassette, and the Widal test.
transport over large distances. This type of centralized testing The sensitivity values varied widely between the different tests,
leads to lost time and quality and as a result clinicians are in­ from 0% with the Diaquick antigen cassette to 78.8% with the
clined to skip the laboratory sample and rather go ahead with IgG component of the CTK Typhoid IgG/IgM Combo Rapid
treatments without laboratory confirmation. However, realisti­ Test CE. Overall, the study confirmed that no test currently meets
cally, blood or bone marrow cultures, which are highly specific the desired accuracy criteria [11] and diagnostic innovation is
and considered the gold standard, are not suitable for use out­ critical.
side of well-established centers and not truly close to the point
of care. This lack of accurate diagnostic testing has a negative
THE FUTURE OF TYPHOID DIAGNOSTICS
impact on patient care and reliable incidence data.
While a variety of techniques are currently in use for the diag­
nosis of typhoid, no single technique satisfies the requirement
THE LACK OF SUITABLE RAPID DIAGNOSTIC TESTS
for sensitivity and specificity while being rapid and cost-
Ultimately to most of the challenges described in the case study, effective. This was again confirmed in the most recent data gen­
well-performing, high-quality tests are needed to be performed in erated by FIND and partners [9], and the need for innovations
a decentralized manner at the point of contact and not at a central was once again made obvious. However, future innovation for
facility that requires sample shipment. While many tests exist and typhoid diagnosis should not only focus on disease diagnosis
are used at point of care (POC) by minimally trained staffs, unfor­ for immediate treatment purposes but also disease surveillance
tunately few meet the “well performing” or “high quality” bar that and the detection of carriers, to support public health interven­
is equally essential. Various RDTs and different forms of the tions. Ultimately both aspects are different sides of the same
Widal test are commonly used in health facilities around the coin and need to be advanced simultaneously to accelerate dis­
world to diagnose typhoid. These tests are cheap and simple, ease elimination as a whole.
do not require sophisticated laboratories, and deliver results in RDTs using selected antigens such as the protein HlyE and
a shorter time frame than blood culture, making them very pop­ sugars in the lipopolysaccharide are under investigation and ex­
ular. However, such tests lack sensitivity and specificity and thus hibit some potential [12, 13]. Furthermore, studies using metab­
are not of sufficient accuracy to replace blood culture as the main olomic platforms have sought to identify biomarkers specific to
diagnostic approach for typhoid fever. The Widal test is the most typhoid. Identifying a single or a combination of metabolites

S18 • OFID 2023:10 (Suppl 1) • Sapkota et al

during the course of typhoid illness could provide several prom­ prevalence is the ability of governments to determine whether
ising biomarkers [14–16]. Polymerase chain reaction (PCR)– and where to use typhoid conjugate vaccine (TCV). Two TCVs
based detection of typhoid in the blood generally shows poor have been prequalified by the World Health Organization
sensitivity. Conventional to real-time PCR and nested and mul­ (WHO) [19]. The WHO Strategic Advisory Group of Experts
tiplex PCR using different targets have been used to diagnose ty­ recommends prioritizing TCV introduction in areas with
phoid with sensitivity ranges of 40%–100% [17]. However a high typhoid fever burden and areas with high prevalence of
more recent study using machine-learning algorithms to iden­ antimicrobial resistance (AMR) [20]. Although the introduc­
tify expression signatures of host-associated genes showed tion is good news for many countries, due to the lack of quality
some promise. This study identified the transcripts of 5 key diagnostics both at the central (eg, blood culture or environ­
genes (STAT1, SLAMF8, PSME2, WARS, and ALDH1A1) that mental surveillance data) or decentralized level (eg, reliable
can differentiate enteric fever from other febrile illness; this ap­ RDT data, as part of the surveillance data set), the same coun­
proach may have some traction for a multipathogen diagnostic tries often struggle to justify the use of TCVs due to missing

Downloaded from https://academic.oup.com/ofid/article/10/Supplement_1/S17/7188897 by guest on 29 August 2023

approach [18]. The latter 2 approaches might provide better val­ data. The lack of data links both to typhoid as well as AMR
ue and may aid in identifying the cause of undifferentiated fe­ data, the latter also requiring microbiology facilities.
brile disease (including typhoid) in resource-limited LMICs Since 2017, a TCV vaccine has been approved for funding
for better patient management. At this point, however, this is support by Gavi, the Vaccine Alliance, and will be made avail­
not yet the case as none of the existing tools meet the needs of able to eligible countries [21]. Liberia, Malawi, Nepal, Pakistan,
resource-poor settings, both in terms of cost and performance and Zimbabwe have been approved for funding support and re­
[17]. A tool would have to be cheap and simple to use (akin to lated TCV introduction support. Sixteen countries theoretically
the GeneXpert) to really make it suitable for hospitals in lower- eligible for Gavi support do not have reliable typhoid surveil­
resource settings. While simple molecular tools to be used at the lance data in the public domain (ie, blood culture confirmation,
POC level were scarce pre-2020, after the scientific advance­ since at least 1995) [22]. In light of the outlined limitations of all
ments and investments made linked to the severe acute respira­ currently available typhoid RDTs, the WHO recommends that
tory syndrome coronavirus 2 (SARS-CoV-2) pandemic it might blood culture be used as the preferred diagnostic test for guid­
be more feasible to think about a workable POC device that can ing immunization program decisions [23]. Gavi is working to
be used to identify a magnitude of possible fever causes [18]. help make improved typhoid diagnostic tests available to coun­
Given the complexities of typhoid diagnosis in patients or car­ tries eligible for Gavi funding support. Such improved tests
riers using simply accessible samples, public health agencies should facilitate country efforts to scale up reliable typhoid di­
might have to resort to identifying the pathogens in the environ­ agnostic testing as part of multidisease surveillance systems,
ment as a proxy for patients. which in turn should lead to improved availability and use of
Molecular approaches look promising to detect S. Typhi in information to ensure that immunization programs are more
environmental samples. These methods are not meant as tools effective, efficient, and equitable [24].
for healthcare workers to inform patient management so will
not advance this area, yet they might open up a more promising
CONCLUSIONS
area for public health surveillance, similar to cholera.
When thinking about and envisioning the next generation of Reviewing the past, present, and future of typhoid diagnostic
diagnostic tools, it is critical that we do not confuse the different tests highlights that we really have not progressed rapidly since
use cases and that we make sure the future Target Product the introduction of the Widal test. There are many advances
Profiles account for all. As of now, the ideal approach unfortu­ still to be made to enable the timely and reliable diagnosis of
nately remains elusive as it needs to be low-cost and simple to typhoid infections. Our overview of use cases ranging from pa­
use even when deployed for environmental surveillance. tient management to environmental surveillance and vaccine
Looking toward a future with increased focus on typhoid, the allocation highlight the critical need for research and product
most likely scenario is a combined approach where more high- development work. We argue that we are in the right period
tech approaches are developed by research and public health to solve these problems, with the ongoing SARS-CoV-2 pan­
authorities and individual patient management remains to be demic showing the importance of diagnostics for disease miti­
guided by culture as well as improved RDTs. gation. Additionally, the fact that Gavi is a committed ally for
diagnostics is encouraging and may help to steer additional re­
sources toward the development of pragmatic tools.
IMPROVED DIAGNOSTICS ACCESS TO IMPROVE
VACCINE DELIVERY
Notes
Arguably, one of the biggest consequences of the current limi­ Financial support. The study mentioned on “The lack of suitable
tations in typhoid diagnostics and the resulting data gap on true rapid diagnostic tests” was conducted by FIND through DFID grant.

Diagnostics for Typhoid Fever • OFID 2023:10 (Suppl 1) • S19

 Supplement sponsorship. This article appears as part of the supplement 11. Mather RG, Hopkins H, Parry CM, Dittrich S. Redefining typhoid diagnosis: what
“Charting the Course to Meet the Challenges Ahead: Research and would an improved test need to look like? BMJ Glob Health 2019; 4:e001831.
Developments on Typhoid and Other Invasive Salmonelloses” sponsored 12. Davies DH, Jain A, Nakajima R, et al. Serodiagnosis of acute typhoid fever in
by the Coalition against Typhoid Secretariat, housed at the Sabin Nigerian pediatric cases by detection of serum IgA and IgG against hemolysin
E and lipopolysaccharide. Am J Trop Med Hyg 2016; 95:431–9.
Vaccine Institute in Washington, DC and made possible by a grant from
13. Kumar S, Nodoushani A, Khanam F, et al. Evaluation of a rapid point-of-care
the Bill & Melinda Gates Foundation. multiplex immunochromatographic assay for the diagnosis of enteric fever.
Potential conflicts of interest. All authors: No reported conflicts. mSphere 2020; 5:e00253-20.
14. Näsström E, Vu Thieu NT, Dongol S, et al. Salmonella Typhi and Salmonella
References Paratyphi A elaborate distinct systemic metabolite signatures during enteric fever.
1. Ajibola O, Mshelia MB, Gulumbe BH, Eze AA. Typhoid fever diagnosis in endem­ Elife 2014; 3:e03100.
ic countries: a clog in the wheel of progress? Medicina (Kaunas) 2018; 54:23. 15. Darton TC, Blohmke CJ, Giannoulatou E, et al. Rapidly escalating hepcidin and
2. Bhutta ZA, Gaffey MF, Crump JA, et al. Typhoid fever: way forward. Am J Trop associated serum iron starvation are features of the acute response to typhoid in­
Med Hyg 2018; 99:89–96. fection in humans. PLoS Negl Trop Dis 2015; 9:e0004029.
3. Azmatullah A, Qamar FN, Thaver D, Zaidi AKM, Bhutta ZA. Systematic review of 16. Blohmke CJ, Darton TC, Jones C, et al. Interferon-driven alterations of the host’s
the global epidemiology, clinical and laboratory profile of enteric fever. J Glob amino acid metabolism in the pathogenesis of typhoid fever. J Exp Med 2016; 213:
1061–77.

Downloaded from https://academic.oup.com/ofid/article/10/Supplement_1/S17/7188897 by guest on 29 August 2023

Health 2015; 5:020407.
4. Masuet-Aumatell C, Atouguia J. Typhoid fever infection—antibiotic resistance 17. Neupane DP, Dulal HP, Song J. Enteric fever diagnosis: current challenges and
and vaccination strategies: a narrative review. Travel Med Infect Dis 2021; 40: future directions. Pathogens 2021; 10:410.
101946. 18. Blohmke CJ, Muller J, Gibani MM, et al. Diagnostic host gene signature for dis­
5. Lao Statistics Bureau. Results of population and housing census 2015. tinguishing enteric fever from other febrile diseases. EMBO Mol Med 2019; 11:
2016. Available at: https://lao.unfpa.org/en/publications/results-population- e10431.
and-housing-census-2015-english-version. Accessed 5 August 2022. 19. World Health Organization. Prequalification of medical products (IVDs,
6. Roberts T, Rattanavong S, Phommasone K, et al. Typhoid in Laos: an 18-year per­ medicines, vaccines and immunization devices, vector control): prequalified
spective. Am J Trop Med Hyg 2020; 102:749. vaccines. Available at: https://extranet.who.int/pqweb/vaccines/prequalified-
7. Ling CL, Roberts T, Soeng S, et al. Impact of delays to incubation and storage tem­ vaccines. Accessed 5 August 2022.
perature on blood culture results: a multi-centre study. BMC Infect Dis 2021; 21: 20. World Health Organization. Typhoid vaccines: WHO position paper—March
173. 2018. Wkly Epidemiol Rec 2018; 93:153–72.
8. Amicizia D, Micale RT, Pennati BM, et al. Burden of typhoid fever and cholera: 21. Gavi the Vaccine Alliance. Typhoid vaccine support. Available at: https://www.
similarities and differences. Prevention strategies for European travelers to en­ gavi.org/types-support/vaccine-support/typhoid. 2022. Accessed 5 August 2022.
demic/epidemic areas. J Prev Med Hyg 2019; 60:271–85. 22. Kim JH, Im J, Parajulee P, et al. A systematic review of typhoid fever occurrence in
9. Wijedoru L, Mallett S, Parry CM. Rapid diagnostic tests for typhoid and paraty­ Africa. Clin Infect Dis 2019; 69(Suppl 6):S492–8.
phoid (enteric) fever. Cochrane Database Syst Rev 2017; 5:CD008892. 23. World Health Organization. Surveillance standards for vaccine-preventable dis­
10. Sapkota J, Hasan R, Onsare R, et al. Comparative analysis of commercially avail­ eases, 2nd ed. 2018. Available at: https://apps.who.int/iris/handle/10665/275754.
able typhoid point of care tests: results of a prospective and hybrid retrospective Accessed 5 August 2022.
multicentre diagnostic accuracy study in Kenya and Pakistan. J Clin Microbiol 24. Hampton LM, Johnson HL, Berkley SF. Diagnostics to make immunisation pro­
2022; 60:e0100022. grammes more efficient, equitable, and effective. Lancet Microbe 2022; 3:e242–3.

S20 • OFID 2023:10 (Suppl 1) • Sapkota et al