Objectives. T follicular helper (Tfh) cells are the principal T helper cell subset that provides help to B cells for potent antibody responses against various pathogens. In this study, we took advantage of the live-attenuated yellow fever virus (YFV) vaccine strain, YF-17D, as a model system for studying human antiviral immune responses in vivo following exposure to an acute primary virus challenge under safe and highly controlled conditions, to comprehensively analyse the dynamics of circulating Tfh (cTfh) cells. Methods. We tracked and analysed the response of cTfh and other T and B cell subsets in peripheral blood of healthy volunteers by flow cytometry over the course of 4 weeks after YF-17D vaccination. Results. Using surface staining of cell activation markers to track YFV-specific T cells, we found increasing cTfh cell frequencies starting at day 3 and peaking around 2 weeks after YF-17D vaccination. This kinetic was confirmed in a subgroup of donors using MHC multimer staining for four known MHC class II epitopes of YF-17D. The subset composition of cTfh cells changed dynamically during the course of the immune response and was dominated by the cTfh1-polarised subpopulation. Importantly, frequencies of cTfh1 cells correlated with the strength of the neutralising antibody response, whereas frequencies of cTfh17 cells were inversely correlated. Conclusion. In summary, we describe detailed cTfh kinetics during YF-17D vaccination. Our results suggest that cTfh expansion and polarisation can serve as a ª prognostic marker for vaccine success. These insights may be leveraged in the future to improve current vaccine design and strategies.

There is an urgent need for better diagnostic and analytical methods for vaccine research and infection control in virology. This has been highlighted by recently emerging viral epidemics and pandemics (Zika, SARS-CoV-2), and recurring viral outbreaks like the yellow fever outbreaks in Angola and the Democratic Republic of Congo (2016) and in Brazil (2016–2018). Current assays to determine neutralising activity against viral infections in sera are costly in time and equipment and suffer from high variability. Therefore, both basic infection research and diagnostic population screenings would benefit from improved methods to determine virus-neutralising activity in patient samples. Here we describe a robust, objective, and scalable Fluorescence Reduction Neutralisation Test (FluoRNT) for yellow fever virus, relying on flow cytometric detection of cells infected with a fluorescent Venus reporter containing variant of the yellow fever vaccine strain 17D (YF-17D-Venus). It accurately measures neutralising antibody titres in human serum samples within as little as 24 h. Samples from 32 vaccinees immunised with YF-17D were tested for neutralising activity by both a conventional focus reduction neutralisation test (FRNT) and FluoRNT. Both types of tests proved to be equally reliable for the detection of neutralising activity, however, FluoRNT is significantly more precise and reproducible with a greater dynamic range than conventional FRNT. The FluoRNT assay protocol is substantially faster, easier to control, and cheaper in per-assay costs. FluoRNT additionally reduces handling time minimising exposure of personnel to patient samples. FluoRNT thus brings a range of desirable features that can accelerate and standardise the measurement of neutralising anti-yellow fever virus antibodies. It could be used in applications ranging from vaccine testing to large cohort studies in systems virology and vaccinology. We also anticipate the potential to translate the methodology and analysis of FluoRNT to other flaviviruses such as West Nile, Dengue and Zika or to RNA viruses more generally.