Clinical Diagnostics For Immune Response Assessment Through Multiplexed Biosensor Immunoassays In At-Risk Populations

Abstract

Immunoassays are a vital tool for clinical quantification of relevant biomarkers, which often lead to diagnostic utility. Plate-based enzyme-linked immunosorbent assays (ELISAs) are a gold-standard technique for quantitatively measuring a variety of biomarkers. However, sample consumption, time-to-result, laborious handling, biological matrix effects, and analyte plexity limitations hinder novel applications in disease diagnostics. While many diagnostic tests focus on a single analyte, multiplexed analyses can contribute a signature profile that can differentiate disease states. Commercialization of multiplexable technologies has led to their increase in clinical relevance. Therefore, analytical advancements, in the form of technology commercialization and multiplexing, are vital for productive clinical translation. In my doctoral dissertation, I present the validation, development, and implementation of custom microring sensor-based multiplexed immunoassays for a variety of clinical challenges. Chapter 1 reviews current technologies where multiplexed analyses have been applied to clinical samples, focusing on examples used within clinical workflows. Current limitations of these technologies are examined, including sample type accessibility and multiplexity challenges. Chapters 2 and 3 describe distinct clinical evaluations within latent tuberculosis infection (LTBI) diagnostics, an asymptomatic form of TB. Through multiplexed cytokine measurements in cell supernatant, I examined immunoprofiles of interest and was able to make accurate predictions regarding risk of reactivation. This study evaluates a panel of thirteen cytokines across six stimulation conditions, including a peptide pool with 300 Mtb-derived T cell epitopes, utilizing normalization strategies to account for basal cytokine levels. Random Forest feature selection identified correlative cytokine signatures from normalized stimulation conditions. Receiver Operator Characteristic curves revealed predictive accuracies of greater than 80% for both LTBI+ and High Risk designations. Chapter 3 uses this LTBI analysis workflow to incorporate the current clinical pipeline, wherein QuantiFERON testing (QFT) stimulated plasma samples for TB evaluation. Our results show predictive accuracy of 90% for detection of LTBI and >80% for risk of reactivation. I successfully designed a combinatorial technique that can be implemented into the clinical workflow to predict LTBI and risk of reactivation using patient QFT samples. Chapter 4 describes a multiplexed panel for neonates, evaluating basal immune signatures at different gestational ages and the effects of chorioamnionitis. This analysis shows important changes in immature immune responses for unstimulated serum and indicates that there are distinctive profiles for healthy and chorio-exposed preterm infants. Statistical analyses revealed significance for CCL2, TNF-α, IL-1β, IL-6, IL-8, and IL-10. Biomarkers of interest were identified across 23-36 weeks of corrected gestational age. These findings indicate that exposure to chorioamnionitis has long-lasting immune consequences, which may alter their ability to respond to infections. Through this rapid (<1hr) immunoassay, relevant measurements can be made to directly affect treatment options and clinical outcomes. Chapter 5 establishes a diverse panel for immunoassay implementation, with a total of 15 cytokines for SARS-CoV-2 related analyses. Following the workflow established in Chapter 2, stimulated conditions are utilized to understand how COVID infection produces and (dys)regulates host response. Statistical analyses were utilized to distinguish biomarkers of importance for COVID+ individuals, which include absolute and normalized levels of CCL3, GM-CSF, IFN-λ1, IL-1β, IL-2, and IL-7. Although currently sample limited, multiple cytokines show promise for continued evaluation and potential clinical utility. Finally, Chapter 6 summarizes the current diagnostic space created, and the future implementations in longitudinal LTBI diagnostics, co-infections in neonatal subjects, and the potential for vaccine immunity evaluation in COVID.