Graphical abstract During epidemics, such as the frequent and devastating Ebola virus outbreaks that have historically plagued regions of Africa, serological surveillance efforts are critical for viral containment and the development of effective antiviral therapeutics. Antibody serology can also be used retrospectively for population-level surveillance to provide a more complete estimate of total infections. Ebola surveillance efforts rely on enzyme-linked immunosorbent assays (ELISAs), which restrict testing to laboratories and are not adaptable for use in resource-limited settings. In this manuscript, we describe a paper-based immunoassay capable of detecting anti-Ebola IgG using Ebola virus envelope glycoprotein ectodomain (GP) as the affinity reagent. We evaluated seven monoclonal antibodies (mAbs) against GP—KZ52, 13C6, 4G7, 2G4, c6D8, 13F6, and 4F3—to elucidate the impact of binding affinity and binding epitope on assay performance and, ultimately, result interpretation. We used biolayer interferometry to characterize the binding of each antibody to GP before assessing their performance in our paper-based device. Binding affinity (K D ) and on rate (k on ) were major factors influencing the sensitivity of the paper-based immunoassay. mAbs with the best K D (3–25 nM) exhibited the lowest limits of detection (ca. μg mL −1 ), while mAbs with K D > 25 nM were undetectable in our device. Additionally, and most surprisingly, we determined that observed signals in paper devices were directly proportional to k on . These results highlight the importance of ensuring that the quality of recognition reagents is sufficient to support desired assay performance and suggest that the strength of an individual’s immune response can impact the interpretation of assay results. Supplementary Information The online version contains supplementary material available at 10.1007/s00216-021-03317-4.

The correct interpretation of the result from a point-of-care device is crucial for an accurate and rapid diagnosis and to guide subsequent treatment. Lateral flow tests (LFTs) use a well- established format that was designed to simplify the user experience. However, the LFT device architecture is inherently limited to detecting analytes that can be captured by molecular recognition. Microfluidic paper-based analytical devices (μPADs), like LFTs, have the potential to be used in diagnostic applications at the point of care. However, μPADs have not gained significant traction outside of academic laboratories, in part, because they have often demonstrated a lack of homogeneous shape or color in signal outputs, which consequently can lead to inaccurate interpretation of results by users. Here, we demonstrate a new class of μPADs that generate colorimetric signals at the interfaces of converging liquid fronts (i.e., lines) to control where colorimetric signals are formed without relying on capture techniques. We demonstrate our approach by developing assays for three classes of analytes—an ion, an enzyme, and a small molecule—using iron (III), acetylcholinesterase, and lactate, respectively. Additionally, we show these devices have the potential to support multiplexed assays by generating multiple lines in a common readout zone. These results highlight the ability of this new paper-based device architecture to aid the interpretation of assays that create soluble products by using flow to constrain those colorimetric products in a familiar, line-format output.

The correct interpretation of the result from a point-of-care device is crucial for an accurate and rapid diagnosis to guide subsequent treatment. Lateral flow tests (LFTs) use a well-established format...