Offner, Gwynneth D.Gupta, ShrutiAlikhan, Firasat Mir2026-02-182026-02-182025https://hdl.handle.net/2144/523352025The inability to reliably distinguish between immune checkpoint inhibitor-associated acute kidney injury (ICI-AKI) and AKI originating from non-ICI etiologies (non-ICI-AKI) has been associated with suboptimal clinical decisions and inferior outcomes among cancer patients. Utilizing the Olink proteomics immunoassay, we measured 140 immune and inflammatory proteins across three ICI-treated patient cohorts: AKI associated with AIN (acute interstitial nephritis – a common histopathologic feature of ICI-AKI observed on renal biopsy), termed AIN-AKI (1), non-AIN-AKI (2), and non-AKI (3). In addition, we retrospectively collected eight preserved kidney biopsy specimens of ICI-treated patients (4 with AIN-AKI and 4 with non-AIN-AKI). We analyzed the biopsy samples, using spatial transcriptomics to investigate the cellular and immune landscape associated with each condition. Our proteomics analysis yielded three interferon-gamma (IFN-γ) induced chemokines with strong correlation to AIN-AKI, the strongest being for CXC-motif ligand 9 (CXCL9). In addition, our spatial transcriptomic profiling uncovered a specific T cell and macrophage niche expanded specifically in the AIN-AKI cohort and exhibiting elevated CXCL9 expression.en-USAttribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/Molecular biologyAcute kidney injuryCXCL9ICI-AKIImmune checkpoint inhibitorProteomicsTranscriptomicsThesis/Dissertation2026-02-18