Despite a characteristic indolent course, a substantial subset of follicular lymphoma (FL) patients has an early relapse with a poor outcome. Cells in the microenvironment may be a key contributor to treatment failure. We used a discovery and validation study design to identify microenvironmental determinants of early failure and then integrated these results into the FLIPI. In total, 496 newly diagnosed FL grade 1–3 A patients who were prospectively enrolled into the MER cohort from 2002 to 2012 were evaluated. Tissue microarrays were stained for CD4, CD8, FOXP3, CD32b, CD14, CD68, CD70, SIRP-α, TIM3, PD-1, and PD-L1. Early failure was defined as failing to achieve event-free survival at 24 months (EFS24) in immunochemotherapy-treated patients and EFS12 in all others. CyTOF and CODEX analysis were performed to characterize intratumoral immunophenotypes. Lack of intrafollicular CD4 expression was the only predictor of early failure that replicated with a pooled OR 2.37 (95%CI 1.48–3.79). We next developed a bio-clinical risk model (BioFLIPI), where lack of CD4 intrafollicular expression moved patients up one FLIPI risk group, adding a new fourth high-risk group. Compared with BioFLIPI score of 1, patients with a score of 2 (OR 2.17; 95% CI 1.08–4.69), 3 (OR 3.53; 95% CI 1.78–7.54), and 4 (OR 8.92; 95% CI 4.00–21.1) had increasing risk of early failure. The favorable intrafollicular CD4 T cells were identified as activated central memory T cells, whose prognostic value was independent from genetic features. In conclusion, lack of intrafollicular CD4 expression predicts early failure in FL and combined with FLIPI improves identification of high-risk patients; however, independent validation is warranted.
Reference atlases, molecular and spatial maps of mammalian tissues, are critical resources for discovery efforts and translational research. Their utility is dependent on operationalizing the resulting data by identifying cell types, histological patterns, and predictive biomarkers underlying health and disease. The human lymph node (LN) offers a compelling use case because of its importance in immunity, structural and cellular diversity, and neoplastic involvement. One hematological malignancy, follicular lymphoma (FL), evolves from developmentally blocked germinal center B cells residing in and trafficking through these tissues. To promote survival and immune escape, tumor B cells undergo significant genetic changes and extensively remodel the lymphoid microenvironment. Here, we present an integrated portrait of healthy and FL LNs using multiple genomic and advanced imaging technologies. By leveraging the strengths of each platform, we identified several tumor-specific features and microenvironmental patterns enriched in individuals who experience early relapse, the most high-risk of FL patients.
BackgroundPD-1 checkpoint blockade therapy (CBT) has greatly benefited patients with select solid tumors and lymphomas but has limited efficacy against diffuse large B-cell lymphoma (DLBCL). Because numerous inhibitory checkpoint receptors have been implicated in driving tumor-specific T cell dysfunction, we hypothesized that combinatorial CBT would enhance the activity of anti-PD-1-based therapy in DLBCL. T cell immunoglobulin and immunoreceptor tyrosine-based inhibitory motif domain (TIGIT) is a coinhibitory receptor expressed on dysfunctional tumor-infiltrating T cells, and TIGIT blockade has demonstrated encouraging activity in combination with PD-1 blockade in murine tumor models and in clinical studies. However, the degree to which TIGIT mediates T cell dysfunction in DLBCL has not been fully explored.ResultsHere, we demonstrate that TIGIT is broadly expressed on lymphoma-infiltrating T cells (LITs) across a variety of human lymphomas and is frequently coexpressed with PD-1. TIGIT expression is particularly common on LITs in DLBCL, where TIGIT+LITs often form distinct cellular communities and exhibit significant contact with malignant B cells. TIGIT+/PD-1+LITs from human DLBCL and murine lymphomas exhibit hypofunctional cytokine production on ex vivo restimulation. In mice with established, syngeneic A20 B-cell lymphomas, TIGIT or PD-1 mono-blockade leads to modest delays in tumor outgrowth, whereas PD-1 and TIGIT co-blockade results in complete rejection of A20 lymphomas in most mice and significantly prolongs survival compared with mice treated with monoblockade therapy.ConclusionsThese results provide rationale for clinical investigation of TIGIT and PD-1 blockade in lymphomas, including DLBCL.
3020 Background: Understanding the underlying heterogeneity of the tumor microenvironment (TME) on a single-cell level is becoming increasingly important to predict a patient’s response to immunotherapy. Conventional imaging methods can help reveal tissue heterogeneity, but are not optimal for identifying multiple cellular subpopulations or cellular interactions from a single slide image, limiting their use in clinical settings. Here, we present a clinical artificial intelligence (AI)-driven multiplex immunofluorescence (MxIF) imaging pipeline based on novel cell segmentation and cell typing methods to evaluate tumor cellular heterogeneity, immune cell composition, and cell-to-cell interactions. Methods: A machine learning (ML)-based cell segmentation algorithm was trained on a manually annotated dataset created from 219 different regions of interest (ROIs) that contained 85,991 cells from various tissues (colon, kidney, lung, lymph node, tonsil, and ureter). A dataset containing 58,676 cells from 146 ROIs was used for validation and accuracy was determined between automated and manually annotated images; accuracy was further evaluated by calculating the f1-score using available methods (DeepCell and Stardist). Marker stains with a low signal-to-noise ratio were automatically enhanced, allowing for adequate cell-to-cell interaction analysis. Results: An automated MxIF image processing workflow was developed. Validation of the trained cell segmentation model showed high accuracy (0.80 f1-score), demonstrating superior performance compared to other methods (DeepCell and Stardist - 0.55 and 0.78 f1-score, respectively). The pathologist-determined accuracy (0.84 mean f1-score) indicated a near-human performance of the developed method. Normalized expression values obtained from the cell typing model allowed automated cell recognition. We analyzed cellular heterogeneity across 3 regions of colorectal cancer (CRC), gastric cancer (GC), and non-small cell lung cancer (NSCLC) samples. While proportions of immune cells varied, proportions of malignant epithelial cells were stable across all regions of each sample, as concordant percentages of Ki67+ cells were identified (CRC-19%; GC-21%; NSCLC-5%). Analysis of cell-to-cell interactions and immune communities identified tumor-, immune-, and stromal-enriched communities in all tumor samples that were stable across regions. Conclusions: By analyzing complex tumor tissue at single-cell resolution with high accuracy, this AI-driven MxIF imaging technology is able to characterize tumor and microenvironment heterogeneity across cancer types. This novel AI-based tool is currently being integrated into several ongoing prospective clinical studies to aid in the development of predictive and prognostic biomarkers.
Diagnosis of low-grade myelodysplastic syndromes (LG-MDS) is one of the most challenging in hematopathology as it relies predominantly on morphologic assessment of dysplasia. Prior studies have demonstrated poor interobserver agreement among pathologists. Histomorphological evaluation of bone marrow core biopsy samples remains the gold standard for diagnostic workup of LG-MDS, including myelodysplastic syndromes (MDS) and other myeloid neoplasms. However, this approach may be subjective, and cannot quantitatively assess subtle differences in marrow topography and the cellular microenvironment. Multiparametric in situ imaging (MISI) through various techniques enables multiple biomarker detection in a single tissue. BostonGene has developed an AI-based image analysis platform to reveal spatial information and subtle histomorphologic features in an objective, quantitative fashion. Here, we demonstrate the potential for automated AI-based imaging analysis of MISI to assist in the differentiation of LG-MDS samples from normal marrow tissues (NBM). Decalcified human bone marrow core biopsy tissues from LG-MDS (n=6) and uninvolved staging marrows (NBM, n=4) were first prepared by immunofluorescence-based MISI via staining with DAPI and CD34, CD38, CD117, CD71, CD15, and CD61 antibodies. BostonGene analyzed the resulting images (fig.1) using a proprietary AI-based imaging platform to identify cells by segmentation performed with the pre-trained instance segmentation neural network. Cell types were identified with mean marker expression values using an accelerated version of BostonGene's phenograph clustering algorithm. Pathologists manually masked fat and bone trabeculae. Using a combination of cell size/shape parameters and antigen expression levels, the following unique cell types were identified: myeloblasts, proerythroblasts, erythroid normoblasts, maturing granulocytes, megakaryocytes, mast cells, plasma cells, and B-cell precursors (hematogones). Data revealed differences in the cellular content of NBM and LG-MDS samples, and separation of LG-MDS samples with the del(5q) subtype (n=2). While linear slender islet-like small clusters of erythroid normoblasts were detected in NBM, we observed a chaotic arrangement of them in LG-MDS samples. In LG-MDS samples, we found an increase in the total number of erythroid normoblasts from 17% to 31%, LG-MDS-del(5q) had 14%. The ratio of maturing granulocytes to erythroid normoblasts (M:E ratio) was significantly lower in LG-MDS (0.63) than in NBM (1.95). The M:E ratio generated by MISI strongly correlated to the M:E ratio produced by manual differential count of bone marrow aspirate samples (R=0.83, p < 0.003). Additionally, fewer hematogones were identified in LG-MDS marrows compared to NBM samples, as reported by others using orthogonal methods. Topographic analysis showed the fat to cellular tissue area ratio was higher in NBM (0.73) than LG-MDS (0.41), but the ratio of trabecular area to total tissue area was higher in LG-MDS (1.67) than NMB (0.74). Spatially, myeloblasts and megakaryocytes were found closer to trabeculae in NBM than LG-MDS;12 different cell communities were identified;2 of them (cluster 3 - erythroid normoblasts enriched, cluster 5 - erythroid normoblasts contacting proerythroblasts) were distributed statistically significantly differently in NBM and LG-MDS samples, indicating the use of MISI with AI-based imaging to distinguish LG-MDS from NBM. Patients of MDS-del(5q) subtype differ significantly from other MDS samples and are more similar to NBM. AI-based image analysis applied to MISI of bone marrow tissue revealed multiple cell types in single tissue sections, along with histologically subtle differences in topography between NBM and LG-MDS samples. These results highlight the importance of integrating in situ tissue analysis with techniques that examine single cell characteristics for a more comprehensive picture of the differences between normal tissue and tumor samples. Coupling sophisticated imaging analytics with this imaging method may provide a more powerful tool for novel biomarker discovery of prognostic and therapeutic significance in the management of MDS and other marrow-based disorders. Figure 1 Figure 1. Disclosures Svekolkin: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Varlamova: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Galkin: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Akaeva: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Smirnova: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Ovcharov: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Polyakova: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties. Tabakov: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company. Postovalova: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties: BostonGene. Koltakova: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company. Gunn: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company. Bagaev: BostonGene Corp.: Current Employment, Current holder of stock options in a privately-held company, Patents & Royalties: BostonGene.
The spatial anatomy of hematopoiesis in bone marrow has been extensively studied in mice and other preclinical models but technical challenges have precluded a commensurate exploration in humans. Institutional pathology archives contain many thousands of paraffinized bone marrow core biopsy tissue specimens, providing a rich resource for studying the intact human bone marrow topography in a variety of physiologic states. Thus, we developed an end-to-end pipeline involving multiparameter whole tissue staining, in situ imaging at single cell resolution, and AI-based digital image analysis, and then applied it to a cohort of disease-free samples to survey alterations in the hematopoietic topography associated with aging. Our data indicate heterogeneity in marrow adipose tissue (MAT) content within each age group, and an inverse correlation between MAT content and proportions of early myeloid and erythroid precursors, irrespective of age. We identify consistent endosteal and perivascular positioning of hematopoietic stem and progenitor cells (HSPCs) with medullary localization of more differentiated elements and, importantly, uncover new evidence of aging-associated changes in cellular and vascular morphologies, microarchitectural alterations suggestive of inflammaging, and diminution of a potentially active megakaryocytic niche. Our findings overall suggest that there is topographic remodeling of human hematopoiesis associated with aging. More generally, we demonstrate the potential to deeply unravel the spatial biology of normal and pathologic human bone marrow states using intact archival tissue specimens.
Background The importance of the immune system in modulating the trajectory of lymphoma outcomes has been increasingly recognized. We recently showed that CD4+ cells are associated with clinical outcomes in a prospective cohort of almost 500 patients with follicular lymphoma (FL). Specifically, we showed that the absence of CD4+ cells inside follicles was independently associated with increased risk of early clinical failure. These data suggest that the composition, as well as the spatial distribution of immune cells within the tumor microenvironment (TME), play an important role in FL. To further define the architecture of the TME in FL we analyzed a FL tumor section using the Co-Detection by Indexing (CODEX) multiplex immunofluorescence system. Methods An 8-micron section from a formalin-fixed paraffin-embedded block containing a lymph node specimen from a patient with FL was stained with a cocktail of 15 CODEX antibodies. Five regions of interest (ROIs) were imaged using a 20X air objective. Images underwent single-cell segmentation using a Unet neural network, trained on manually segmented cells (Fig 1A). Cell type assignment was done after scaling marker expression and clustering using Phenograph. Each ROI was manually masked to indicate areas inside follicles (IF) and outside follicles (OF). Relative and absolute frequencies of cell types were calculated for each region. Cellular contacts were measured as number and types of cell-cell contacts within two cellular diameters. To identify proximity communities, we clustered cells based on number and type of neighboring masks using Phenograph. The number of cell types and cellular communities were calculated inside and outside follicles after adjustment for total IF and OF areas. The significance of cell contact was measured using a random permutation test. Results We identified 13 unique cell subsets (11 immune, 1 endothelial, 1 unclassified) in the TME of our FL section (Fig. 1A). The unique phenotype of each subset was confirmed using a dimensionality reduction tool (t-SNE). The global composition of the TME varied minimally across ROIs and consisted primarily of B cells, T cells, and macrophages subsets - in decreasing order of frequency. Higher spatial heterogeneity across ROIs was observed in the frequency of T cell subsets in comparison to B cells subsets. Inspecting the spatial distribution of T cell subsets (Fig. 1B), we observed that cytotoxic T cells were primarily located in OF areas, whereas CD4+ T cells were found in both IF and OF areas. Notably, the majority of CD4+ T cells inside the follicles expressed CD45RO (memory phenotype), while most of the CD4+ T cells outside the follicles did not. Statistical analysis of the spatial distribution of CD4+ memory T cell subsets confirmed a significant increase in their frequency inside follicles compared to outside (20.4% vs 11.2%, p < 0.001; Fig. 1D). Cell-cell contact analysis (Fig 1C) showed increased homotypic contact for all cell types. We also found a higher frequency of heterotypic contact between Ki-67+CD4+ memory T cells and Ki-67+ B cells. Pairwise analysis showed these findings were statistically significant, indicating these cells are organized in niches rather than randomly distributed across image. Analysis of cellular communities (Fig. 1C) identified 13 niches, named according to the most frequent type of cell-cell contact. All CD4+ memory T cell subsets were found to belong to the same neighborhood (CD4 Memory community). Analysis of the spatial distribution of this community confirmed that these niches were more frequently located inside follicles rather than outside (26.3±4% vs 0.004%, p < 0.001, Fig. 1D). Conclusions Analysis of the TME using CODEX provides insights on the complex composition and unique architecture of this FL case. Cells were organized in a pattern characterized by (1) high degree of homotypic contact and (2) increased heterotypic interaction between activated B cells and activated CD4+ memory T cells. Spatial analysis of both individual cell subsets and cellular neighborhoods demonstrate a statistically significant increase in CD4+ memory T cells inside malignant follicles. This emerging knowledge about the specific immune-architecture of FL adds mechanistic details to our initial observation around the prognostic value of the TME in this disease. These data support future studies using modulation of the TME as a therapeutic target in FL. Figure 1 Disclosures Galkin: BostonGene: Current Employment, Patents & Royalties. Svekolkin:BostonGene: Current Employment, Current equity holder in private company, Patents & Royalties. Postovalova:BostonGene: Current Employment, Current equity holder in private company. Bagaev:BostonGene: Current Employment, Current equity holder in private company, Patents & Royalties. Ovcharov:BostonGene: Current Employment, Current equity holder in private company, Patents & Royalties. Varlamova:BostonGene: Current Employment, Current equity holder in private company, Patents & Royalties. Novak:Celgene/BMS: Research Funding. Witzig:AbbVie: Consultancy; MorphSys: Consultancy; Incyte: Consultancy; Acerta: Research Funding; Karyopharm Therapeutics: Research Funding; Immune Design: Research Funding; Spectrum: Consultancy; Celgene: Consultancy, Research Funding. Nowakowski:Nanostrings: Research Funding; Seattle Genetics: Consultancy; Curis: Consultancy; Ryvu: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other; Kymera: Consultancy; Denovo: Consultancy; Kite: Consultancy; Celgene/BMS: Consultancy, Research Funding; Roche: Consultancy, Research Funding; MorphoSys: Consultancy, Research Funding. Cerhan:BMS/Celgene: Research Funding; NanoString: Research Funding. Ansell:Trillium: Research Funding; Takeda: Research Funding; Regeneron: Research Funding; Affimed: Research Funding; Seattle Genetics: Research Funding; Bristol Myers Squibb: Research Funding; AI Therapeutics: Research Funding; ADC Therapeutics: Research Funding.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.