Rapid design and fielding of four diagnostic technologies in Sierra Leone, Thailand, Peru, and Australia: Successes and challenges faced introducing these biosensors

Shawn P. Mulvaney, Lisa A. Fitzgerald, Leila J. Hamdan, Bradley R. Ringeisen, Emily R. Petersen, Jaimee R. Compton, Nina L. McAuliff, Tomasz A. Łęski, Chris R. Taitt, David A. Stenger, Christopher A. Myers, Erin Hansen, Michelle Ricketts, Chelsea Hoegberg, Kamonthip Homdayjanakul, Rashid Ansumana, Joseph M. Lamin, Umaru Bangura, Joseph Lahai, Victoria Baio, Direk Limmathurotsakul, Gumphol Wongsuvan, Viriya Hantrakun, Supaporn Wacharapluesadee, Anek Mungaomklang, Opass Putcharoen, Pratoomtong Yatoom, Kriengsak Kruthakool, Robert D. Hontz, Christopher N. Mores, Crystyan Siles, Amy C. Morrison, Mark Mayo, Bart J. Currie, Kathryn H. Jacobsen, Kathleen Quinn, Jerold Blutman, Flavia Amariei, John R. HannanSensing and Bio-Sensing Research 2018 · DOI

TL;DR

This paper details a rapid, multi-continental deployment of four diagnostic technologies for febrile illnesses, including multiplexed lateral flow immunoassays (LFIs) and a PCR-based system. The project successfully moved technologies from concept to field testing in under a year, demonstrating a viable pathway for accelerating diagnostic development and deployment in resource-limited settings.

What Problem Does This Paper Address?

Febrile illnesses are a leading cause of hospital visits globally, but identifying the specific pathogen is crucial for effective treatment and outbreak detection. In low-income countries, traditional laboratory diagnostics are often hindered by a lack of reliable power, cold storage, and trained personnel. This creates a critical need for robust, easy-to-use diagnostic tools that can perform well in austere environments. The US Department of Defense, facing similar challenges with its globally deployed personnel, initiated the "24 Month Challenge" to rapidly develop and field diagnostic technologies that could address these limitations, moving beyond traditional lab-based methods to point-of-care solutions.

Methods

The core of this research involved a rapid, public-private partnership to develop and field-test diagnostic technologies. The process included:

  • Technology Selection and Development: Commercial partners developed custom multiplexed lateral flow immunoassay (LFI) tests and a specialized PCR assay pouch. Two LFIs were developed: the Active Dengue Melioidosis Detect (ADMD) (duplex for Dengue NS1 and B. pseudomallei CPS) and the Dual Path Platform (DPP) (4-plex for Dengue NS1, B. pseudomallei CPS, Y. pestis F1 antigen, and Plasmodium species pLDH/HRP2). A FilmArray system with a custom Severe Acute Systematic Febrile Illness (SASFI) pouch was also adapted for field use.
  • Analytical Validation: Technologies underwent rigorous testing at government laboratories (US Naval Research Laboratory, Naval Medical Research Center, USAMRIID) for sensitivity, specificity, inclusivity, and exclusivity using both inactivated and live agents under BSL2 and BSL3 conditions. Environmental robustness was also assessed under elevated temperature and humidity.
  • Field Deployment: The validated technologies were deployed to clinics and laboratories in Sierra Leone, Thailand, Peru, and Australia.
  • Data Collection and Analysis: A key component was the Deki Reader, an optoelectronic reader for LFIs. It guided users through test protocols, captured images of results, and transmitted data to a cloud database. This facilitated remote monitoring and reach-back support. Standard-of-care or gold-standard tests were run in parallel to benchmark performance.

Datasets

The paper describes the use of various biological samples for validation and field testing:

  • Human whole blood: Obtained from Valley Biomedical, spiked with target antigens or challenge samples for sensitivity and specificity testing.
  • Inactivated Yersinia pestis strain Colorado 92: Used for BSL2 testing.
  • Live agent Y. pestis samples: Used for BSL3 testing.
  • Purified Burkholderia pseudomallei CPS antigen: Provided by Dr. David AuCoin.
  • Live agent Burkholderia species: Grown in-house at USAMRIID for BSL3 inclusivity testing.
  • Recombinant Dengue NS1 antigens: Purchased from The Native Antigen Company, including all four serotypes.
  • Plasmodium falciparum strains (D6, C235, 3D7, 7G8, Malawi, Thailand): Prepared at various institutions.
  • Exclusivity panels: Included Leishmania donovani, chikungunya, yellow fever, influenza A (H3N2), hepatitis B and C viruses, and various Burkholderia species.
  • Patient samples: Over 10,000 samples were collected from febrile participants across the four continents for in-field testing.

Key Insights and Ideas

  • Accelerated Development is Possible: The "24 Month Challenge" successfully demonstrated that diagnostic technologies can be moved from ideation to field deployment much faster than traditional timelines, highlighting the effectiveness of a structured, public-private partnership.
  • The Deki Reader is a Game Changer: This optoelectronic reader significantly enhances LFI usability by providing guided protocols, objective result interpretation, and crucial data transmission capabilities for remote monitoring and quality control. It mitigates user bias, especially at the limit of detection.
  • Multiplexing is Key for Febrile Illnesses: The ability to test for multiple pathogens simultaneously (e.g., Dengue, Malaria, Melioidosis, Plague) is essential for accurate diagnosis in regions where these diseases co-exist.
  • Field Challenges are Multifaceted: Beyond technical performance, successful deployment hinges on factors like user instructions (pictorial and video are effective), sample collection methods (finger prick challenges), and reagent storage/transport in diverse environmental conditions.
  • PCR Offers High Sensitivity but Has Throughput Limitations: While the FilmArray SASFI pouch provided broad diagnostic coverage and high sensitivity, its sample-to-answer time and lower throughput compared to LFIs present trade-offs for high-volume screening.

Key Results

  • Over 10,000 assays were performed across four continents using the developed technologies.
  • High concordance (often >85%) was observed between the new diagnostic technologies and standard-of-care or gold-standard tests, validating their performance in real-world settings.
  • Environmental robustness testing showed that both DPP and ADMD LFIs remained operational under stressed conditions (37°C/95% humidity), though some loss in sensitivity was noted near the limit of detection.
  • The Deki Reader demonstrated high reproducibility (>97%) in intra- and inter-instrument reliability tests.
  • No false positives were observed for the ADMD and DPP assays on a BSL2 exclusivity panel.
  • The SASFI PCR pouch successfully detected multiple targets across four diseases, with limits of detection significantly lower than LFIs.

Limitations and Future Work

  • Limited Timeframe for Optimization: The rapid development schedule (~6 months for custom assay development) limited the extent of assay optimization, potentially impacting sensitivity and specificity for certain targets.
  • Malaria Concordance: The concordance for malaria detection was lower than other diseases, attributed to the technical demands of smear microscopy, varying user expertise, and the nature of HRP2 markers lingering in vivo. This highlights an area for future improvement in malaria diagnostics.
  • DPP Assay Limitations: While the DPP was designed for multiplexing, some targets showed lower sensitivity or missed detections (e.g., Y. pestis strains Kim 5 and Java 9, and lower dengue detection compared to ADMD).
  • Cost of PCR: The high cost of FilmArray assays, while offering comprehensive diagnostics, may limit their use as a primary tool for routine diagnostics in resource-limited settings, suggesting their role as confirmatory tests.
  • Future Work: The authors suggest improving LFI sensitivity and specificity, further optimizing malaria diagnostics, and leveraging the established infrastructure and lessons learned for future rapid diagnostic development and regulatory approval pathways. The tiered selection method and public-private partnership model are also areas for continued exploration.

Summary updated

BibTeX
@inproceedings{mulvaney2018rapid,
  author = {Shawn P. Mulvaney and Lisa A. Fitzgerald and Leila J. Hamdan and Bradley R. Ringeisen and Emily R. Petersen and Jaimee R. Compton and Nina L. McAuliff and Tomasz A. Łęski and Chris R. Taitt and David A. Stenger and Christopher A. Myers and Erin Hansen and Michelle Ricketts and Chelsea Hoegberg and Kamonthip Homdayjanakul and Rashid Ansumana and Joseph M. Lamin and Umaru Bangura and Joseph Lahai and Victoria Baio and Direk Limmathurotsakul and Gumphol Wongsuvan and Viriya Hantrakun and Supaporn Wacharapluesadee and Anek Mungaomklang and Opass Putcharoen and Pratoomtong Yatoom and Kriengsak Kruthakool and Robert D. Hontz and Christopher N. Mores and Crystyan Siles and Amy C. Morrison and Mark Mayo and Bart J. Currie and Kathryn H. Jacobsen and Kathleen Quinn and Jerold Blutman and Flavia Amariei and John R. Hannan},
  title = {{Rapid design and fielding of four diagnostic technologies in Sierra Leone, Thailand, Peru, and Australia: Successes and challenges faced introducing these biosensors}},
  booktitle = {Sensing and Bio-Sensing Research},
  year = {2018},
  doi = {10.1016/j.sbsr.2018.06.003},
}
Figure