Existing clinical biomarkers do not reliably predict treatment response or disease progression in patients with advanced intrahepatic cholangiocarcinoma (ICC). Circulating neoplastic-immune hybrid cells (CHCs) have great promise as a blood-based biomarker for patients with advanced ICC. Peripheral blood specimens were longitudinally collected from patients with advanced ICC enrolled in the HELIX-1 phase II clinical trial (NCT04251715). CHCs were identified by co-expression of pan-cytokeratin (CK) and CD45, and levels were correlated to patient clinical disease course. Unsupervised machine learning was then performed to extract their morphological features to compare them across disease courses. Five patients were included in this study, with a median of nine specimens collected per patient. A median of 13.5 CHCs per 50,000 peripheral blood mononuclear cells were identified at baseline, and levels decreased to zero following the initiation of treatment in all patients. Counts remained undetectable in three patients who demonstrated end-of-trial clinical treatment response and conversely increased in two patients with evidence of therapeutic resistance. In the post-trial surveillance period, interval counts increased prior to or at the time of clinical progression in three patients and remain undetectable in one patient with continued long-term disease stability. Using our machine learning platform, treatment-resistant CHCs exhibited upregulation of CK and downregulation of CD45 relative to treatment-responsive CHCs. CHCs represent a promising blood-based biomarker to supplement traditional radiographic and biochemical measures.

Multiplexed tissue imaging (MTI) and other spatial profiling technologies commonly utilize serial tissue sectioning to comprehensively profile samples by imaging each section with unique biomarker panels or assays. The dependence on serial sections is attributed to technological limitations of MTI panel size or incompatible multi-assay protocols. Although image registration can align serially sectioned MTIs, integration at the single-cell level poses a challenge due to inherent biological heterogeneity. Existing computational methods overlook both cell population heterogeneity across modalities and spatial information, which are critical for effectively completing this task. To address this problem, we first use Monte-Carlo simulations to estimate the overlap between serial 5μm-thick sections. We then introduce COEXIST, a novel algorithm that synergistically combines shared molecular profiles with spatial information to seamlessly integrate serial sections at the single-cell level. We demonstrate COEXIST necessity and performance across several applications. These include combining MTI panels for improved spatial single-cell profiling, rectification of miscalled cell phenotypes using a single MTI panel, and the comparison of MTI platforms at single-cell resolution. COEXIST not only elevates MTI platform validation but also overcomes the constraints of MTI’s panel size and the limitation of full nuclei on a single slide, capturing more intact nuclei in consecutive sections and thus enabling deeper profiling of cell lineages and functional states.