Citations

Abstract Motivation Multiplex immunofluorescence (MxIF) enables the quantification of multiple protein markers at a single-cell level while preserving spatial information, offering a powerful tool for studying tissue microenvironments. However, the flexibility in MxIF panel design poses challenges in standardizing cell phenotyping. Results We present SpaFlow, an efficient, customizable pipeline for unsupervised clustering and classification of MxIF data, implemented using Nextflow. SpaFlow performs quality control, clustering, and postclustering analysis on segmented and quantified MxIF data, facilitating reproducible and scalable analyses across various computing platforms. The SpaFlow pipeline integrates three clustering and classification packages—Seurat, SCIMAP, and CELESTA—each providing unique methodologies for identifying cell types based on phenotypic markers. A novel “meta-clustering” approach condenses clusters across multiple regions of interest into common meta-clusters, streamlining the cell-type identification process in large datasets. SpaFlow’s robust quality control steps, including signal summation and cell density filtering, mitigate artifacts that may impact clustering accuracy. We demonstrate the utility of SpaFlow in a case study involving 297 ovarian tumor cores, where SpaFlow successfully identified biologically meaningful cell populations, including tumor-infiltrating lymphocytes, efficiently and rapidly. Additionally, SpaFlow’s reproducibility is validated using serial tonsil sections, confirming its capability to consistently identify distinctive cell populations across matched ROIs. Availability and implementation SpaFlow is freely available with detailed documentation and examples at https://github.com/dimi-lab/SpaFlow.

Multiplex immunofluorescence (MxIF) images provide detailed information of cell composition and spatial context for biomedical research. However, compromised data quality could lead to research biases. Comprehensive image quality checking (QC) is essential for reliable downstream analysis. As a reliable and specific staining of cell nuclei, 4′,6-diamidino-2-phenylindole (DAPI) signals were used as references for tissue localization and auto-focusing across MxIF staining–scanning–bleaching iterations and could potentially be reused for QC. To confirm the feasibility of using DAPI as QC reference, pixel-level DAPI values were extracted to calculate signal fluctuations and tissue content similarities in staining–scanning–bleaching iterations for identifying quality issues. Concordance between automatic quantification and human experts’ annotations were evaluated on a data set consisting of 348 fields of view (FOVs) with 45 immune and tumor cell markers. Cell distribution differences between subsets of QC-pass vs QC-failed FOVs were compared to investigate the downstream effects. Results showed that 87.3% FOVs with tissue damage and 73.4% of artifacts were identified. QC-failed FOVs showed elevated regional gathering in cellular feature space compared with the QC-pass FOVs. Our results supported that DAPI signals could be used as references for MxIF image QC, and low-quality FOVs identified by our method must be cautiously considered for downstream analyses.

Direct interactions between tumor and immune cells mediate the antitumor effect of all modern cancer immunotherapeutic agents. Simultaneously, tumor cells have evolved mechanisms of evasion, including the downregulation of HLA-I, potentially disrupting the mechanism of action employed by many immune checkpoint inhibitors. And yet, the in situ interplay between these cells within the tumor immune microenvironment (TIME) remains elusive. Recent advances in histologic multiplex bioimaging platforms have enabled in-depth molecular characterization of single cells within spatially preserved and clinically archived tumor tissues. Herein, we applied multiplex immunofluorescence to excisional lymph node biopsies from 14 patients with metastatic melanoma who experienced clear objective responses to immunotherapy (7 complete response; 7 progressive disease) to determine distinguishing features of the TIME in the pretreatment setting. Distinct regions of the TIME were evaluated using 35 proteins probing tumor, immune, and vasculature components across 323 fields of view. Single-cell compositional analysis confirmed established prognostic immune cell types including increased prevalence of cytotoxic T cells within the tumor core fields of view of responders. Integrating single-cell quantification with the spatial arrangement of cellular neighborhoods surrounding tumor cells revealed novel, spatial immune signatures capable of stratifying TIME based on clinical response. Our analysis revealed dynamic cellular composition of the tumor-centric cellular neighborhood (TCCN) based on anatomic subregion, functional expression of HLA-I by the index tumor cell and ultimately clinical response to immunotherapy. Overall, this study provides an analytic framework to resolve the cellular complexity of the TIME, increasingly relevant to the outcomes of modern cancer immunotherapy. Significance: Findings from this work propose a novel approach to resolving clinical heterogeneity of the TIME by objectively quantifying the cellular interactions occurring in metastatic melanoma lymph node tissue utilizing multiplex immunofluorescence. This study provides an analytic and biologically derived unit of measure, the TCCN which is customizable for studying critical paracrine interactions within spatially preserved tissue of various cancers and across the spectrum of multiplex imaging modalities.

AbstractThrough our involvement in KEYNOTE‐059, we unexpectedly observed durable responses in two patients with metastatic gastroesophageal adenocarcinoma (mGEA) who received ramucirumab (anti‐VEGFR‐2)/paclitaxel after immune checkpoint inhibition (ICI). To assess the reproducibility of this observation, we piloted an approach to administer ramucirumab/paclitaxel after ICI in more patients, and explored changes in the immune microenvironment. Nineteen consecutive patients with mGEA received ICI followed by ramucirumab/paclitaxel. Most (95%) did not respond to ICI, yet after irRECIST‐defined progression on ICI, all patients experienced tumor size reduction on ramucirumab/paclitaxel. The objective response rate (ORR) and progression‐free survival (PFS) on ramucirumab/paclitaxel after ICI were higher than on the last chemotherapy before ICI in the same group of patients (ORR, 58.8% vs 11.8%; PFS 12.2 vs 3.0 months; respectively). Paired tumor biopsies examined by imaging mass cytometry showed a median 5.5‐fold (range 4‐121) lower frequency of immunosuppressive forkhead box P3+ regulatory T cells with relatively preserved CD8+ T cells, post‐treatment versus pre‐treatment (n = 5 pairs). We then compared the outcomes of these 19 patients with a separate group who received ramucirumab/paclitaxel without preceding ICI (n = 68). Median overall survival on ramucirumab/paclitaxel was longer with (vs without) immediately preceding ICI (14.8 vs 7.4 months) including after multivariate analysis, as was PFS. In our small clinical series, outcomes appeared improved on anti‐VEGFR‐2/paclitaxel treatment when preceded by ICI, in association with alterations in the immune microenvironment. However, further investigation is needed to determine the generalizability of these data. Prospective clinical trials to evaluate sequential treatment with ICI followed by anti‐VEGF(R)/taxane are underway.

e15184 Background: Clinical responses to anti-PD1 immunotherapy in patients with metastatic melanoma (MM) remain challenging to predict. This clinical heterogeneity is also reflected in the tumor-immune microenvironment among patients and within a single tumor lesion. With the emergence of multiplex imaging platforms, defining complex phenotypes at single cell resolution has become possible. Here, we sought to objectively quantify paracrine tumor-immune interactions contributing to the variable clinical responses observed in patients receiving anti-PD1 therapy. Methods: Excisional lymph node (LN) biopsies were obtained from treatment-naïve patients with MM who underwent subsequent anti-PD1 therapy. A single 5µm section of LN tissue was used to assess a 42 analyte panel by multiplex immunofluorescence. From 30 LN samples, 418 fields of view (FOVs) were selected resulting in 14,360 high-resolution images of 4 anatomical subregions: tumor core, tumor-immune interface, tumor infiltrate and adjacent immune stroma. Following image processing, we developed an adaptive classification for tumor-centric cellular neighborhoods (TCCN) to identify and quantify critical paracrine interactions within the tumor-immune microenvironment. Results: Stratification based on responsiveness to anti-PD1 therapy resulted in 4 complete responders (CR) and 12 non-responders (NR) at 12-week follow-up. From 126 FOVs, we defined the cellular composition of 197,865 TCCN across patients based on clinical response and LN subregions. Overall, the percentage of TCCN devoid of any T cells, B cells or macrophages was significantly higher in NR compared to CR irrespective of subregion. However, other markers differentiated TCCN based on subregion. Specifically in CR, tumor core regions were enriched for CD8 T cells, while enrichment for B cells and endothelial cells was demonstrated at the tumor-immune interface. Strikingly, tumor infiltrate regions demonstrate robust immune reactivity with enrichment for M1 polarized macrophages, NK cells and B cells in CR compared to NR. Complete data from the 30 patient cohort across 418 FOVs will be presented. Conclusions: Taken together, this data suggests cellular composition of TCCN across subregions of the LN is dynamic within a single metastatic site. In this small cohort, we introduce a formalized stratification to quantify and classify critical paracrine interactions within the immune-tumor microenvironment with the potential to inform clinical responsiveness to therapy.

Abstract Background Immunosuppression in glioblastoma (GBM) is an obstacle to effective immunotherapy. GBM-derived immunosuppressive monocytes are central to this. Programmed cell death ligand 1 (PD-L1) is an immune checkpoint molecule, expressed by GBM cells and GBM extracellular vesicles (EVs). We sought to determine the role of EV-associated PD-L1 in the formation of immunosuppressive monocytes. Methods Monocytes collected from healthy donors were conditioned with GBM-derived EVs to induce the formation of immunosuppressive monocytes, which were quantified via flow cytometry. Donor-matched T cells were subsequently co-cultured with EV-conditioned monocytes in order to assess effects on T-cell proliferation. PD-L1 constitutive overexpression or short hairpin RNA–mediated knockdown was used to determined the role of altered PD-L1 expression. Results GBM EVs interact with both T cells and monocytes but do not directly inhibit T-cell activation. However, GBM EVs induce immunosuppressive monocytes, including myeloid-derived suppressor cells (MDSCs) and nonclassical monocytes (NCMs). MDSCs and NCMs inhibit T-cell proliferation in vitro and are found within GBM in situ. EV PD-L1 expression induces NCMs but not MDSCs, and does not affect EV-conditioned monocytes T-cell inhibition. Conclusion These findings indicate that GBM EV-mediated immunosuppression occurs through induction of immunosuppressive monocytes rather than direct T-cell inhibition and that, while PD-L1 expression is important for the induction of specific immunosuppressive monocyte populations, immunosuppressive signaling mechanisms through EVs are complex and not limited to PD-L1.

Abstract Glioblastoma (GBM) is the most common and lethal primary brain tumor, and novel therapeutic strategies that make a substantial impact on outcomes are sorely needed. Immunotherapies have shown great promise in the treatment of a number of cancers in recent year, and a concerted effort is being made to apply these treatment paradigms to GBM. However, many GBM patients exhibit profound immunosuppression, likely limiting the efficacy of such approaches. The mechanisms of this immunosuppression are poorly understood, but tumor-derived extracellular vesicles (EVs), may play a role. We demonstrate that GBM-derived EVs induce the development of myeloid-derived suppressor cells (MDSCs) and non-classical monocytes (NCMs). We further demonstrate that these EV-induced monocytic cells are immunosuppressive, resulting in impaired T cell proliferation upon co-culture. We found that the immunosuppressive effects of tumor-derived EVs appear to be driven by the induction of these immunosuppressive monocytes, as EV treatment of T cells did not significantly impact T cell proliferation. Further, we sought to characterize the important of programmed death ligand 1 (PD-L1) in the induction of these immunosuppressive monocyte populations. We found PD-L1 to be expressed in the EVs from GBM cell lines, and that modulation in PD-L1 expression via either constitutive overexpression or shRNA-mediated knockdown resulted in concordant changes in expression in tumor-derived EVs. We demonstrate that PD-L1 is important for the induction of NCM but not for MDSCs. Taken together, these findings point to a significant role for tumor-derived EVs in the induction of immunosuppressive monocytes in GBM, and that these cells may be a driving force of systemic immunosuppression. PD-L1 is one factor expressed in EVs that has immunomodulatory properties, but additional EV cargo likely plays a major role in the induction of immunosuppressive cells.