Tumor-Infiltrating Normal B Cells Revealed by Immunoglobulin Repertoire Clonotype Analysis Are Highly Prognostic and Crucial for Antitumor Immune Responses in DLBCL

PURPOSE

Tumor-infiltrating B lymphocytes (TIL-B) have demonstrated prognostic and predictive significance in solid cancers. In this study, we aimed to distinguish TIL-Bs from malignant B-cells in diffuse large B-cell lymphoma (DLBCL) and determine the clinical and biological significance.

EXPERIMENTAL DESIGN

A total of 269 patients with de novo DLBCL from the International DLBCL R-CHOP Consortium Program were studied. Ultra-deep sequencing of the immunoglobulin genes was performed to determine B-cell clonotypes. The frequencies and numbers of TIL-B clonotypes in individual repertoires were correlated with patient survival, gene expression profiling (GEP) data, and frequencies of DLBCL-infiltrating immune cells quantified by fluorescent multiplex IHC at single-cell resolution.

RESULTS

TIL-B abundance, evaluated by frequencies of normal B-cell clonotypes in the immunoglobulin repertoires, remarkably showed positive associations with significantly better survival of patients in our sequenced cohorts. DLBCLs with high versus low TIL-B abundance displayed distinct GEP signatures, increased pre-memory B-cell state and naïve CD4 T-cell state fractions, and higher CD4+ T-cell infiltration. TIL-B frequency, as a new biomarker in DLBCL, outperformed the germinal center (GC) B-cell-like/activated B-cell-like classification and TIL-T frequency. The identified TIL-B-high GEP signature, including genes upregulated during T-dependent B-cell activation and those highly expressed in normal GC B cells and T cells, showed significant favorable prognostic effects in several external validation cohorts.

CONCLUSIONS

TIL-B frequency is a significant prognostic factor in DLBCL and plays a crucial role in antitumor immune responses. This study provides novel insights into the prognostic determinants in DLBCL and TIL-B functions with important therapeutic implications.

Abstract 4688: Comprehensive analysis of natural killer cell-associated markers using MultiOmyxTM immunofluorescence assay

As a promising alternative platform for cellular immunotherapy, natural killer cells (NK) have recently gained attention as an important type of innate immune regulatory cell. NK cell immunotherapy approaches have been translated into clinical applications, and clinical trials of NK cell infusion in patients with hematological malignancies (HM) and solid tumors have thus far yielded many encouraging clinical results. Understanding the pattern of NK expression and the relationship to different states of NK cells may have direct relevance for immune responses in cancer. Approaches capable of simultaneously detecting NK cells with detailed information on NK differentiation state, however, remain limited. In this study, we used MultiOmyx hyperplexed immunofluorescence (IF) assay to classify and characterize the spatial arrangement of NK cell markers in a pan-cancer cohort including 9 tissue microarrays (TMAs) from breast, head and neck, prostrate, colon, bladder, lung, kidney, ovarian and melanoma indications. The panel includes CD3, CD4, CD8, CD16, CD45, CD56, CD57, CD137, FoxP3, Granzyme B, HLA-E, NKG2A, NKP46 and tumor segmentation marker. The panel enables the detection of NK cells expressing CD56 and/or NKP46 in the 9 different tumor indications. We also studied the expression of activating and inhibiting receptors, such as CD16 and NKG2A, in NK cell population. The active cytotoxic subsets of mature NK cells were evaluated using co-expression of NK cell surface markers with CD57 and the cytotoxic molecule expression in NK cells was assessed using co-expression with Granzyme B. Using proprietary deep-learning-based image analysis, we were able to quantify the densities of these different NK cell population and study the prevalence of these NK cells in different cancer indications included in this study. Many strategies have been developed for exploiting NK-mediated anti-tumor activities. CAR-NK cell therapy and antibodies that directly target NK cell inhibitory receptors such as NKG2A and TIGIT, are currently being evaluated in the clinical trials. The MultiOmyx NK cell panel reported in this study enables the comprehensive profiling of the NK population and can provide greater understanding of NK cell biology during cancer progression. The panel can be further used to explore the efficacy of the NK cell-based immune therapy.

Citation Format: Erinn A. Parnell, Jiong Fei, Harry Nunns, Eric Leones, Elaine Yeung, Blair Russell, Flora Sahafi, Qingyan Au. Comprehensive analysis of natural killer cell-associated markers using MultiOmyxTM immunofluorescence assay. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4688.

Abstract 4696: Characterizing CD39 and CD73 cell subtypes in the tumor microenvironment using MultiOmyx

CD39 and CD73 are membrane bound enzymes that function together to convert extracellular adenosine 5’-triphospate (ATP)/adenosine diphosphate (ADP) into adenosine. CD39 first catalyzes ATP/ADP into adenosine monophosphate (AMP), which is then converted by CD73 into adenosine. Accumulation of extracellular adenosine creates an immunosuppressive tumor environment and facilitates tumor growth and metastasis. Conversely, the presence of ATP promotes a pro-inflammatory, tumor-suppressive environment. Suppression of CD73/CD39 activity and reduction of extracellular adenosine has been shown to support an antitumor immune response. Therefore, the targeting of CD73 and CD39, both individually as well as in combination with immune checkpoint inhibitors for biomarkers such as PD-1 and CTLA-4, is an emerging strategy for cancer therapeutics. CD73/CD39 have been described in multiple cell types, including tumor cells, fibroblasts, endothelial cells, tumor infiltrating lymphocytes (TILs), myeloid cells, and natural killer (NK) cells. While known to be expressed on multiple cell types, the spatial characterization of CD39 and CD73 in the tumor microenvironment (TME) is still poorly understood. Characterizing the diversity of active CD73 and CD39 populations in the tumor environment will help improve development of targeted therapies for cancer treatment. The multiplex immunofluorescence (mIF) platform MultiOmyx will be used to define the distinct populations of cells expressing CD73 and/or CD39 in the TME. MuliOmyx is a proprietary mIF platform for the visualization and characterization of up to 60 protein biomarkers in a single FFPE section. Herein we report the design and use of a novel panel of commercially-available antibodies broad enough to characterize CD39 and CD73 subpopulations in the TME of a variety of tumor indications including lung, prostate, and colon cancer. Using this panel in combination with proprietary deep-learning based image analysis, CD39 and CD73 positive cells can be characterized into different immune and TME subtypes. Understanding of the variety and phenotype of CD39 and CD73 expressing cells in the TME is crucial to define the populations being targeted by therapies for cancer treatment.

Citation Format: Courtney Todorov, Kevin Gallagher, Juliana Wortman, Harry Nunns, Erinn Parnell, Eric Leones, Elaine Yeung, Blair Russell, Flora Sahafi, Qingyan Au. Characterizing CD39 and CD73 cell subtypes in the tumor microenvironment using MultiOmyx. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4696.

Abstract 1705: Spatial interrogation of tertiary lymphoid structures (TLS) in colorectal carcinoma (CRC) tumor microenvironment (TME) using the MultiOmyx࣪ assay

Immune checkpoint blockade (ICB) therapy has revolutionized the landscape of cancer therapy in multiple tumor types since Ipilimumab, the first ICB agent, was approved for the treatment of metastatic melanoma is 2011. Current predictive biomarkers for therapy response include PD-L1 expression, tumor mutational burden (TMB) and microsatellite instability (MSI) status. However, responses to ICB vary widely and these predictive markers have demonstrated imperfect predictive power to ICB therapy. For instance, half of microsatellite instability-high (MSI-H) colorectal carcinoma (CRC) patients do not respond to ICB immunotherapy. There is still a crucial need to identify and develop biomarkers predictive of outcome to ICB therapy. Tertiary lymphoid structures (TLS) have been observed in a variety of solid tumors in humans, and their presence is a favorable prognostic indicator for survival in a variety of solid cancers including CRC. Further, it was recently shown that the presence of mature TLS was associated with improved response rate to ICB and overall survival in patients with advanced tumors in a large retrospective study. In this study, we used a 17-plex MultiOmyx panel to detect TLS, explore TLS maturation stage and characterize the tumor microenvironment (TME) in 40 CRC patients with known MSI status. The 17-plex includes CD3, CD4, CD8, CD45RO, FOXP3, CD20, CD56, CD68, CTLA-4, PD1, PD-L1, CXCL13, PNAd, CD21, CD23, DC-LAMP and PanCK. Using this panel, we successfully detected different stages of TLS in these CRC samples. Using proprietary deep-learning-based image analysis algorithm (NeoLYTX), we quantified TLS’s in the CRC samples and classified them by maturation stage based on biomarker expression and spatial organization of immune cells. While there was no difference in the number of early TLS (E-TLS; characterized by CD3, CD20 and pNAd) between MSI-H and microsatellite-stable (MSS) CRC samples, more primary follicle-like TLS (P-TLS; defined by CD3, CD20 and CD21) and secondary follicle-like TLS (S-TLS; defined by CD3, CD20 and CD23) were observed in the MSI-H group than the MSS group. In particular, the abundance of mature S-TLS was significantly higher in MSI-H CRC samples. This 17-plex MultiOmyx assay provides a powerful tool to characterize the cellular composition and spatial organization of the tumor microenvironment. The panel enables quantification of TLS, PD-L1 expression and abundance of tumor infiltrating lymphocytes from one single FFPE slide. The rich datasets generated by the MultiOmyx assay can provide greater understanding of the immune contexture within the TME and deeper insights into the correlations between biomarkers. These findings may have broad application and help identify biomarker signatures with improved predictive performance to immune checkpoint inhibition efficacy in solid tumors.

Citation Format: Harry Nunns, Erinn A. Parnell, Michael Lazare, Judy Kuo, Eric Leones, Flora Sahafi, Josette William, Qingyan Au. Spatial interrogation of tertiary lymphoid structures (TLS) in colorectal carcinoma (CRC) tumor microenvironment (TME) using the MultiOmyx࣪ assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1705.

Abstract 627: Custom Vectra® Polaris™ fluorescent multiplex IHC panel identifies mature tertiary lymphoid structures in colorectal cancer

Introduction: Tertiary lymphoid structures (TLS) are promising prognostic indicators of positive outcomes for patients with solid tumors including colorectal cancer (CRC) [1]. Large-scale retrospective analysis shows patients with mature TLS in particular respond to PD-1/PD-L1 antibody treatment with improved objective response, progression-free and overall survival [2]. Since not all patients respond to PD-1/PD-L1 antibody treatment, identifying patients with mature TLS is clinically relevant as it enables selection of patients likely to respond to immune checkpoint blockade. Mature TLS are composed of T cells (CD3+), follicular B cells (CD20+), germinal center B cells (CD23+), and follicular dendritic cells (CD21+) [3]. The ability to identify and evaluate TLS is limited by detection methods which traditionally employ hematoxylin and eosin (H&E) staining for visual quantification of immune aggregates, potentially significantly underestimating their quantity [4]. Multiplexed fluorescent immunohistochemistry (IHC) detection assays have the capability to precisely quantify mature TLS within the TME. Colorectal cancer specimens expressing a dynamic range of TLS are stained with a novel custom Vectra® Polaris™ multiplex immunofluorescence panel detecting CD20, CD21, CD23, CD3 and pan-cytokeratin. Identification of mature TLS with CD3+CD20+CD21+ and CD3+CD20+CD23+ expression in the TME is reported.

Results: Vectra® Polaris™ TLS multiplex immunofluorescence panel successfully identified mature TLS in the TME of colorectal cancer patient samples. TLS were quantified via custom analytics algorithms generated with Indica HALO software.

Conclusion: Characteristics of TME immunity in colorectal cancer differentially impact an individual patient’s odds of survival [5]. The novel Vectra® Polaris™ multiplex assay in this study shows a detailed picture of mature TLS in patient cancer samples. Future applications of this panel include investigations of TLS associated with successful anti-tumor immune development and therapeutic response to treatment. [1] Oncoimmunology. 2020; 9(1): 1724763 [2] Nat Cancer. 2, 794-802 (2021) [3] Oncoimmunology. 2021; 10(1): 1900508. [4] Mod Pathol 2017 Sep;30(9):1204-1212. doi: 10.1038/modpathol.2017.43. [5] Cell. 2020 Sep 3;182(5):1341-1359.e19. doi: 10.1016/j.cell.2020.07.005

Citation Format: Sara G. Pollan, Arezoo Hanifi, James Hargrove, Erinn A. Parnell, Jessica Lin, Josette William Ragheb, Qingyan Au. Custom Vectra® Polaris™ fluorescent multiplex IHC panel identifies mature tertiary lymphoid structures in colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 627.

Genetic Subtyping and Phenotypic Characterization of the Immune Microenvironment and MYC/BCL2 Double Expression Reveal Heterogeneity in Diffuse Large B-cell Lymphoma

PURPOSE

Diffuse large B-cell lymphoma (DLBCL) is molecularly and clinically heterogeneous, and can be subtyped according to genetic alterations, cell-of-origin, or microenvironmental signatures using high-throughput genomic data at the DNA or RNA level. Although high-throughput proteomic profiling has not been available for DLBCL subtyping, MYC/BCL2 protein double expression (DE) is an established prognostic biomarker in DLBCL. The purpose of this study is to reveal the relative prognostic roles of DLBCL genetic, phenotypic, and microenvironmental biomarkers.

EXPERIMENTAL DESIGN

We performed targeted next-generation sequencing; IHC for MYC, BCL2, and FN1; and fluorescent multiplex IHC for microenvironmental markers in a large cohort of DLBCL. We performed correlative and prognostic analyses within and across DLBCL genetic subtypes and MYC/BCL2 double expressors.

RESULTS

We found that MYC/BCL2 double-high-expression (DhE) had significant adverse prognostic impact within the EZB genetic subtype and LymphGen-unclassified DLBCL cases but not within MCD and ST2 genetic subtypes. Conversely, KMT2D mutations significantly stratified DhE but not non-DhE DLBCL. T-cell infiltration showed favorable prognostic effects within BN2, MCD, and DhE but unfavorable effects within ST2 and LymphGen-unclassified cases. FN1 and PD-1-high expression had significant adverse prognostic effects within multiple DLBCL genetic/phenotypic subgroups. The prognostic effects of DhE and immune biomarkers within DLBCL genetic subtypes were independent although DhE and high Ki-67 were significantly associated with lower T-cell infiltration in LymphGen-unclassified cases.

CONCLUSIONS

Together, these results demonstrated independent and additive prognostic effects of phenotypic MYC/BCL2 and microenvironment biomarkers and genetic subtyping in DLBCL prognostication, important for improving DLBCL classification and identifying prognostic determinants and therapeutic targets.

51 A novel cross-site analysis of Vectra® Polaris™ multiplex fluorescence PD-1/PD-L1 immunohistochemistry on colorectal cancer with high and low microsatellite instability

Background

Colorectal cancer (CRC) is the third most diagnosed cancer in the United States with a projected 52,980 deaths in 2021.1 Microsatellite instability-high (MSI-H) CRCs with deficiencies in mismatch repair (MMR) are significantly associated with positive response to immunotherapy and improved outcomes when treated with immune checkpoint inhibitors. Programmed cell death ligand-1 (PD-L1) is an effective biomarker of MSI-H status to identify CRC patients who will respond to treatment, however, reproducible quantification of programmed cell death receptor-1 (PD-1)/PD-L1 in the tumor microenvironment (TME) across laboratory sites has been under-reported.2–3 In this study, our group directly addressed this issue by interrogating PD-1/PD-L1 cross-site at Akoya Biosciences and NeoGenomics Laboratories by employing the MOTiF™ PD-1/PD-L1 Panel kit along with the Vectra Polaris imaging system.

Methods

Serial sections from 40 CRC samples with known MSI status were stained at Akoya and NeoGenomics Laboratories using a modified MOTiF PD-1/PD-L1 Lung Panel Kit on the Leica BOND RX. Sections were scanned using the Vectra Polaris imaging system at both sites. Inter-site staining reproducibility was assessed using image analysis algorithms developed with inForm tissue analysis software. Cell counts and densities were calculated using the R-script package PhenoptrReports and correlations were plotted per marker.

Results

The average signal intensity for all markers/Opal fluorophores was within the recommended ranges of 10–30 normalized counts, with the exception of Polaris 780, which has an advised range of 1–10. This indicates the protocol stained successfully and reproducibly across all serial sections at both sites. Inter-site concordance analysis of cell densities for each marker yielded an average R2 value of ≥0.70. H-Score of PD-L1 quantified at the cell membrane trended with MSI status (stable/high).

Conclusions

This study demonstrated that the MOTiF PD-1/PD-L1 Panel kit imaged in conjunction with the Vectra Polaris is not only a reliable assay that can be run across different sites, based on the concordant cross-site data, but that re-optimization of the kit allows for the assay panel to be successfully adapted to other cancers, such as CRC, which can then capture biological differences across a multitude of samples.

References

American Cancer Society https://www.cancer.org/cancer/colon-rectal-cancer/about/key-statistics.html

Yi M, Jiao D, Xu H, Liu Q, Zhao W, Han X, et al. Biomarkers for predicting efficacy of PD-1/PD-L1 inhibitors. Mol Cancer 2018;17(1):129

Lemery S, Keegan P, Pazdur R. First FDA approval agnostic of cancer site - when a biomarker defines the indication. N Engl J Med 2017;377(15):1409–12.

2 Quantitation of CD137 and Nectin-4 expression across multiple tumor types to support indication selection for BT7480, a Bicycle tumor-targeted immune cell agonist™ (Bicycle TICA™)

Background

Bicycles are fully synthetic constrained peptides with antibody-like affinities that target selectively, readily penetrate tumor tissue, have relatively short half-lives, and can be chemically linked together to generate multifunctional molecules. BT7480 is a Bicycle TICA™ that binds both CD137 on immune cells and Nectin-4 on cancer cells to deliver a potent anti-tumor immune signal in Nectin-4 expressing tumors. Nectin-4 has been reported to be highly expressed in a wide range of human solid tumors, however the expression of CD137, abundance and localization of CD137+ immune cells in Nectin-4+ tumors are unknowns. A translational and informatics pipeline was established to interrogate the human tumor microenvironment to identify patient populations most likely to benefit from BT7480, which is being developed as a potential first-in-class molecule for the treatment of high unmet need cancers associated with Nectin-4 expression.

Methods

TCGA RNAseq data for Nectin-4 and CD137 were analyzed from ~10,000 samples across 36 human cancers. Using a proprietary Nectin-4 mAb and MultiOmyx™ technology, a 19-plexed immunofluorescence assay was developed to simultaneously quantify the presence of Nectin-4+ and CD137+ cells, identify immune cell subsets and their spatial topography in 43 human tumor FFPE samples from HNSCC, lung, bladder, and breast cancers. Each FFPE slide was presented to a pathologist for tissue annotation and selection of regions of interest for image analysis. Proprietary deep learning-based workflows were applied to identify stroma and tumor regions, individual cells and perform cell classification for phenotypes of interest.

Results

RNA expression analysis indicated co-expression of Nectin-4 and CD137 in several tumor types with >50% tumors within NSCLC, HNSCC, breast, esophageal, and ovarian cancers expressing high levels of both targets. Spatial proteomic studies in HNSCC, lung, breast and bladder cancer samples demonstrated that Nectin-4 and CD137 co-expression at the protein level (>1% positive cells) was detected in 74% samples tested. CD137+ cells in Nectin-4+ tumors were identified as CD4+ T cells (37.6%), CD8+ T cells (16.8%) and CD68+ macrophages (5.9%). A subset of CD137+ cells (32.7%) were found to be deeply tumor penetrant and within close proximity of Nectin-4+ tumor cells across all indications tested.

Conclusions

Results from this study support prioritization of indications for BT7480 clinical development and the utility of the MultiOmyx™ assay to monitor Nectin-4 and CD137 expression and to demonstrate proof-of-mechanism for the BT7480 FIH clinical trial expected to start in 2H-2021.

Antitumor immune effects of preoperative sitravatinib and nivolumab in oral cavity cancer: SNOW window-of-opportunity study

BACKGROUND

Sitravatinib, a tyrosine kinase inhibitor that targets TYRO3, AXL, MERTK and the VEGF receptor family, is predicted to increase the M1 to M2-polarized tumor-associated macrophages ratio in the tumor microenvironment and have synergistic antitumor activity in combination with anti-programmed death-1/ligand-1 agents. SNOW is a window-of-opportunity study designed to evaluate the immune and molecular effects of preoperative sitravatinib and nivolumab in patients with oral cavity squamous cell carcinoma.

METHODS

Patients with newly-diagnosed untreated T2-4a, N0-2 or T1 >1 cm-N2 oral cavity carcinomas were eligible. All patients received sitravatinib 120 mg daily from day 1 up to 48 hours pre-surgery and one dose of nivolumab 240 mg on day 15. Surgery was planned between day 23 and 30. Standard of care adjuvant radiotherapy was given based on clinical stage. Tumor photographs, fresh tumor biopsies and blood samples were collected at baseline, at day 15 after sitravatinib alone, and at surgery after sitravatinib-nivolumab combination. Tumor flow cytometry, multiplex immunofluorescence staining and single-cell RNA sequencing (scRNAseq) were performed on tumor biopsies to study changes in immune-cell populations. Tumor whole-exome sequencing and circulating tumor DNA and cell-free DNA were evaluated at each time point.

RESULTS

Ten patients were included. Grade 3 toxicity occurred in one patient (hypertension); one patient required sitravatinib dose reduction, and one patient required discontinuation and surgery delay due to G2 thrombocytopenia. Nine patients had clinical-to-pathological downstaging, with one complete response. Independent pathological treatment response (PTR) assessment confirmed a complete PTR and two major PTRs. With a median follow-up of 21 months, all patients are alive with no recurrence. Circulating tumor DNA and cell-free DNA dynamics correlated with clinical and pathological response and distinguished two patient groups with different tumor biological behavior after sitravatinib alone (1A) versus sitravatinib-nivolumab (1B). Tumor immunophenotyping and scRNAseq analyses revealed differential changes in the expression of immune cell populations and sitravatinib-targeted and hypoxia-related genes in group 1A vs 1B patients.

CONCLUSIONS

The SNOW study shows sitravatinib plus nivolumab is safe and leads to deep clinical and pathological responses in oral cavity carcinomas. Multi-omic biomarker analyses dissect the differential molecular effects of sitravatinib versus the sitravatinib-nivolumab and revealed patients with distinct tumor biology behavior.

TRIAL REGISTRATION NUMBER

NCT03575598.

Genomic complexity is associated with epigenetic regulator mutations and poor prognosis in diffuse large B-cell lymphoma

Diffuse large B-cell lymphoma (DLBCL) is the most common type of lymphoma with high mutation burdens but a low response rate to immune checkpoint inhibitors. In this study, we performed targeted next-generation sequencing and fluorescent multiplex immunohistochemistry, and investigated the clinical significance and immunological effect of mutation numbers in 424 DLBCL patients treated with standard immunochemotherapy. We found that KMT2D and TP53 nonsynonymous mutations (MUT) were significantly associated with increased nonsynonymous mutation numbers, and that high mutation numbers (MUT^high) were associated with significantly poorer clinical outcome in germinal center B-cell-like DLBCL with wild-type TP53. To understand the underlying mechanisms, we identified a gene-expression profiling signature and the association of MUT^high with decreased T cells in DLBCL patients with wild-type TP53. On the other hand, in overall cohort, MUT^high was associated with lower PD-1 expression in T cells and PD-L1 expression in macrophages, suggesting a positive role of MUT^high in immune responses. Analysis in a whole-exome sequencing dataset of 304 patients deposited by Chapuy et al. validated the correlation of MUT-KMT2D with genomic complexity and the significantly poorer survival associated with higher numbers of genomic single nucleotide variants in activated B-cell-like DLBCL with wild-type TP53. Together, these results suggest that KMT2D inactivation or epigenetic dysregulation has a role in driving DLBCL genomic instability, and that genomic complexity has adverse impact on clinical outcome in DLBCL patients with wild-type TP53 treated with standard immunochemotherapy. The oncoimmune data in this study have important implications for biomarker and therapeutic studies in DLBCL.

Abstract 154: Improving MultiOmyxTM Analytics cell classification workflow efficiency by Invariant Information Clustering on historical data

Understanding the tumor composition and microenvironment provides insight into the efficacy and viability of immunotherapy treatments. Obtaining cell-level co-expression data (biomarker patterns expressed on cells indicating their phenotype) provides insight into where and how various cell types are distributed throughout the tumor. The MultiOmyx technology platform provides a detailed quantitative output including co-expression data through a multiplexed fluorescence microscopy approach, followed by imaging analysis using a proprietary analytics pipeline.The MultiOmyx analytics pipeline involves training a neural network for every biomarker of interest. Each biomarker's neural network will classify all cells as positive or negative for the marker. A major time-sink when constructing these neural network models is manually collecting and annotating training data. Moreover, new models must be built for the same biomarkers across different studies due to variability between samples. However, previously-built models can be applied on new images provided that their training image data and the new images follow the same distribution. This eliminates the bottleneck step of training a new model. In order to find ideal existing models for a new batch of samples, Invariant Information Clustering can be used to identify similarity of the new batch to previous images. Invariant Information Clustering (IIC) is used to group historical images by visual appearance. Image data is collected from the internal database, which stores all images from past studies. These images are split into image tiles (to avoid memory overload) and are then clustered with IIC. IIC works by generating image pairs that consist of the original image and an augmented version. It then maximizes the mutual information between the image and its transformed version using a neural network with a Resnet-34 architecture. This results in extracting the meaningful parts out of the original image while discarding the instance specific details. Augmentations including rotations, flipping, cropping, and noise are used to guide the network to learn proper features and be invariant to others. Visually similar tiles form distinct and tightly-bound cluster groups, each of which has a model assigned to it. When running inference, new image tiles are passed through the network and are assigned to the closest cluster. Cells from each tile are then classified using the model which corresponds to that tile cluster. If a new tile falls far from an existing cluster, cells from that tile will be processed with a new, manually-annotated model. As a result, this method eliminates the need for building new models and collecting new training data for every new set of patient samples, except for extreme outliers. This expedites the analytics process, reduces turnaround time, and allows for a greater volume of samples to be processed.

Citation Format: Vivek Reddy, Nicholas Stavrou, Mate Nagy, Qingyan Au. Improving MultiOmyxTM Analytics cell classification workflow efficiency by Invariant Information Clustering on historical data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 154.

Abstract 2681: Co-detection of a tumor-infiltrating lymphocyte immunofluorescence (IF) panel and cytokine RNA in-situ hybridization (ISH) markers in non-small cell lung cancer (NSCLC) tumor microenvironment using combined MultiOmyx and RNAscope platforms

Co-detection of RNA and protein can greatly expand the data output from a single specimen, providing critical information such as the source of secreted proteins (e.g. cytokines) or cell type specific transcript levels. MultiOmyx is a proprietary immunofluorescence (IF) platform for the visualization and characterization of up to 60 protein biomarkers in a single formalin-fixed paraffin-embedded (FFPE) section. RNAscope Multiplex is a highly sensitive fluorescent in-situ hybridization (ISH) assay that can detect up to 3 RNA markers in a single FFPE section. Combination of MultiOmyx IF with RNAscope Multiplex ISH would therefore provide a novel and powerful platform to co-detect multiple RNA and protein markers in a single slide/sample. However, the RNAscope Multiplex assay includes a protease pretreatment step, which may compromise downstream antibody-antigen interaction and thereby the IF signal. This study is a validation of the sensitivity, specificity, reproducibility, and repeatability of an integrated MultiOmyx IF and RNAscope ISH assay. Depending on the context, cytokines interleukin-10 (IL10) and interferon gamma (IFNγ) have both been shown to either induce immunosuppression and favor tumor growth or promote an anti-tumor response. The mechanisms and cues determining the pro- or anti-tumor activity of IL10 and IFNγ are still poorly understood. Spatiotemporal characterization of IFNG and IL10 is therefore critical to help define the dynamic relationship of cytokines and the immune system within the tumor microenvironment. Therefore, for validation of the integrated MultiOmyx-RNAscope platform, RNA ISH markers for IL10 and IFNγ were combined with the MultiOmyx 12 marker tumor infiltrating lymphocyte (TIL) panel (CD3, CD4, CD8, CD20, CD68, CD56, CD45RO, PD-1, PD-L1, CTLA4, FOXP3, and tumor marker PanCK) on FFPE human NSCLC samples. This combined MultiOmyx-RNAscope workflow was performed for three individual runs using triplicate NSCLC samples for each run. Intensity and cell classification for each ISH and TIL marker was quantified using the proprietary MultiOmyx Analytics pipeline. The results demonstrate that the integrated assay maintains sensitivity and specificity of the TIL IF markers in the integrated workflow when benchmarked to an IF alone workflow. Furthermore, these results can be used to characterize expression of IL10 and IFNγ within immune cell subsets represented by the TIL IF panel (e.g. T cells, NK cells, macrophages). This integrated approach can be used to spatially correlate the distribution of cytokine and immune cell expression within the tumor microenvironment. Therefore, the RNAscope-MultiOmyx IF assay provides a robust and powerful platform to simultaneously co-detect RNA with protein IF in a single specimen.

Citation Format: Courtney Todorov, Maricel Gozo, Harry Nunns, Erinn Parnell, Judy Kuo, Eric Leones, Mate Nagy, Qingyan Au. Co-detection of a tumor-infiltrating lymphocyte immunofluorescence (IF) panel and cytokine RNA in-situ hybridization (ISH) markers in non-small cell lung cancer (NSCLC) tumor microenvironment using combined MultiOmyx and RNAscope platforms [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2681.

Abstract 2771: Comprehensive analysis of immuno oncology markers in the tumor microenvironment of solid tumor samples using GeoMxTM digital spatial profiler (DSP) and MultiOmyxTM hyperplexed immunofluorescence (IF)

In recent years, novel immunotherapy strategies have shown remarkable results in treating various advanced cancers. The efficacy of immunotherapeutics has been dictated by the dynamic and spatially heterogeneous tumor microenvironment (TME) which includes immune tumor infiltrates and immune checkpoint protein expression, in addition to tumor molecular profiles, leading to widely variable clinical outcomes and responses in patients. The local milieu as well as complex tumor-TME interactions vary widely between tumors of the same type and also within the same tumor and therefore highlights the need for enhanced analysis of tissue-immune content and their spatial context for predicting response to immunotherapies and biomarker discovery. In this study, we will apply a multi-faceted, highly multiplexed tissue analysis approach to quantitate and better characterize the spatial arrangement of key immuno oncology protein markers in a pan-cancer cohort of up to 35 FFPE samples from breast, head and neck, prostrate, non-small cell lung cancer (NSCLC), endometrial and colorectal indications using NanoString GeoMxTM Digital Spatial Profiler (DSP) and MultiOmyxTM Hyperplexed Immunofluorescence (IF) assay. For DSP analysis, each sample will be profiled for up to 52 immuno oncology (IO) markers included on NanoString Human IO protein panels. Selection of regions of interest will be guided by H&E staining, fluorescent markers (CD45, pan-CK and Syto 13). Protein profiling of tumor and TME regions will be achieved through segmenting by pan-CK+/CD 45+ followed by collection of indexed oligonucleotides and digital counting using the NanoString nCounter system. Data normalization and unsupervised clustering of IO protein expression across the ROIs will be performed to understand the spatial distribution of these IO biomarkers within the TME. An adjacent section of each tumor sample will be also stained and analyzed using MultiOmyx technology. MultiOmyx is a proprietary IF platform for the visualization and characterization of up to 60 protein biomarkers in a single FFPE section. Using the MultiOmyx IF assay in combination with proprietary algorithms, we will study the expression and spatial distribution of 26 IO markers to characterize different subtypes of immune cells and immune checkpoint inhibitors within the TME in the 35 FFPE tumor samples. Combination of MultiOmyx IF with Nanostring DSP provides a complementary and powerful solution to study the IO markers within the TME. Using the DSP nCounter output along with MultiOmyx cell classification and spatial analysis results, the researchers can simultaneously profile the high-plex protein expression, characterize the immunophenotypes and visualize the spatial distribution of tumor infiltrating immune cells at single-cell resolution within the TME.

Citation Format: Lakshmi Chandramohan, Qingyan Au, Mate Nagy, Erinn Parnell, Tricia Peters. Comprehensive analysis of immuno oncology markers in the tumor microenvironment of solid tumor samples using GeoMxTM digital spatial profiler (DSP) and MultiOmyxTM hyperplexed immunofluorescence (IF) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2771.

Abstract 434: Characterizing viral mRNA and immuno-protein expression in head and neck squamous cell carcinoma using a novel automated RNAscope™/Polaris™ integrated assay

Background: Chronic viral infection can generate inflammatory microenvironments leading to neoplastic growth and cancer development. Cancer patients with chronic viral infection have been shown to exhibit worse outcomes compared to non-infected individuals. One such cohort, head and neck squamous cell carcinoma (HNSCC) patients infected with chronic human cytomegalovirus (CMV), have increased risk of death when receiving radiotherapy or radiochemotherapy. This is clinically significant as HNSCC already carries a high mortality rate and low response to surgery and chemotherapeutic treatment. The ability to detect and localize viral infection and immune response in the same patient tissues has been historically under-developed and may assist in stratifying patients for therapeutic intervention.

Methods: Immune infiltrate to tumor in CMV infected and non-infected samples was assessed in HNSCC patient tissue using RNAscope in-situ hybridization (ISH) probe to detect CMV mRNA and Vectra® Polaris™ automated multiplex protein detection of CD8, PD-L1, CD68 and panCK with OPAL dyes.

Results: We developed a novel automated RNAscope™/Vectra® Polaris™ integrated multiplex immunofluorescence (IF) assay with OPAL detection and quantification of signal using Indica HALO® algorithms. The results include quantitative and reproducible RNAscope fluorescent ISH counting, cell-by-cell expression profiles, multiplex protein quantification and whole-slide image analysis.

Conclusion: NeoGenomics Laboratories RNAscope™/Polaris™ integrated assay detects and quantifies CMV viral infection and protein expression of PD-L1 positive and negative cytotoxic T cells and macrophages within and adjacent to HNSCC tumor regions.

Citation Format: Sara Pollan, Arezoo Hanifi, Mate Nagy, Nicholas Stavrou, Erinn Parnell, Maricel Gozo, Nickolas Attanasio, Josette William, Qingyan Au. Characterizing viral mRNA and immuno-protein expression in head and neck squamous cell carcinoma using a novel automated RNAscope™/Polaris™ integrated assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 434.

Abstract 855: Spatial analytics of the tumor microenvironment on double stained immunohistochemistry images using deep learning

Spatial locations of immune cells in the tumor microenvironment (TME) have been shown to correlate with clinical outcome in different cancers. A worse patient outcome has been reported in oral squamous cancer for individuals with an increased number of Tregs within 30 µm of CD8+ cells [1]. Likewise, the spatial relationship between CD8+ and PD-L1+ cells has become an area of interest as a possible indication for the response to PD-L1-inhibition. It was recently shown in a retrospective study that the combined assessment of CD8 and PD-L1 in NSCLC tumors outperformed CD8 or PD-L1 alone as prognostic markers for predicting treatment with immune checkpoint inhibitors [2].

As reported in these cases, quantifying spatial relationships between two biomarkers can provide clinical and/or biological insights. Spatial analytics of the TME requires accurate cell segmentation and classification. To that end, NeoGenomics have developed an IHC assay for combined CD8 and 22C3/PD-L1 staining, as well as a deep learning pipeline to automatically identify, segment, and classify CD8+ and PD-L1+ cells from whole slide IHC images. We see a 95% concordance of the number of CD8+ and PD-L1+ cells detected in our double stained IHC assay with serial sections stained either for CD8 or PD-L1 alone.

We integrated our previously-reported cell segmentation and classification workflow used for MultiOmyx data (multiplexed IF images) [3] with Indica HALO to analyze whole slide IHC images double stained using two distinct chromogens. In this study, we performed IHC double staining and cell classification analysis on a CD8 - PD-L1 assay in NSCLC tumors. In addition to cell segmentation and classification image outputs, we also generate cell-level and slide-level tables with various cell morphological information, phenotype counts and densities, and biomarker intensity values that can be used to automatically define H-score measures for each biomarker. Additionally, advanced spatial analytics is performed to calculate the average distance between cells of various phenotypes, and spatial clustering patterns of different phenotypes in the TME (i.e. CD8+ and PD-L1+ cells). These analyses enable investigation of numerous complex cell interactions in TMEs.

NeoGenomics quantitative double stained IHC assay is compatible with any two biomarkers of interest even if they are expressed on the same cell as long as the sub-cellular localization of the markers is different. The combination of double staining for CD8 and PD-L1, and quantifying spatial relationships between the two biomarkers is capable of providing the needed context to guide treatment decisions.

References

1. Feng Z et al., JCI Insight. 2017 Jul 20:e93652

2. Steele KE et al., J Immunother Cancer 2018;6:20

3. Nagy ML et al., AACR Annual Meeting 2018; April 14-18, 2018; Chicago, IL

Citation Format: Mate L. Nagy, Anna Juncker-Jensen, Harry Nunns, Ben Ovadia, Ainura Kyshtoobayeva, Karen Yamamoto, Qingyan Au, Josette William Ragheb. Spatial analytics of the tumor microenvironment on double stained immunohistochemistry images using deep learning [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 855.

Abstract 3881: Distinguishing dendritic cell subtypes in the tumor microenvironment using MultiOmyxTM

Dendritic cells (DCs) are key initiators and regulators of the innate and adaptive immune responses. An emerging interest in cancer therapies is the capability to activate endogenous DCs to induce antigen specific T cell responses and thereby generate DC-based immunotherapies. Understanding the function and diversity of DC subsets in the tumor environment will help improve therapies developed for cancer treatment. DC subpopulations have been recognized in humans and categorized based on their phenotype and functional criteria. These DC subsets are classified based on biomarker expressions and include CD123+ plasmacytoid dendritic cells (pDCs), two types of classical dendritic cells CD141+Clec9A+CD11c+ HLADR+ conventional type 1 dendritic cells (cDC1), CD1c+CD11c+HLADR+ conventional type 2 dendritic cells (cDC2) and CD14+CD11c+CD209+ monocyte derived dendritic cells (Mo-DCs). To help understand the complexity of distinct subsets of DCs, their spatial distribution within the tumor microenvironment (TME), and correlation with other immune cells, multiplex immunohistochemistry using a panel of antibodies broad enough to differentiate and characterize multiple DC subsets and T cell populations will be used.

MultiOmyxTM, a novel hyperplexed multi ”omic” technology, enables visualization and characterization of multiple biomarkers across multiple assays on a single 4μm tissue section. MultiOmyx protein immunofluorescence (IF) assays utilize a pair of directly conjugated Cyanine dye-labeled (Cy3, Cy5) antibodies per round of staining. Each round of staining is imaged and followed by novel dye inactivation chemistry, enabling repeated rounds of staining and deactivation for up to 60 protein biomarkers. In this study, a MultiOmyx hyperplexed IF assay will be utilized to distinguish different DC subset within a tumor. Biomarkers including CD11c, CD123, CD141, CleC9A, CD1c, DC-Lamp, DC-sign, HLADR, CD14, CD68, CD163, CD3, CD4, CD8, FOXP3 and PanCK protein expression from a single 4 µm FFPE section in order to identify different subsets of DCs in tumor tissue from patients with Melanoma, a cancer type in which immunotherapeutic treatment has had a transformative effect and become the dominant therapeutic approach. Hopefully, a greater understanding of the phenotypes and functions of dendritic cells subsets will result in new cancer immunotherapy strategies.

Citation Format: Maricel C. Gozo, Vivek Reddy, Mate Nagy, Nickolas Attanasio, Naiyun Zhou, Sara Pollan, Erinn Parnell, Eric Leones, Judy Kuo, Anna Juncker-Jensen, Josette William Ragheb, Qingyan Au. Distinguishing dendritic cell subtypes in the tumor microenvironment using MultiOmyxTM [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3881.

Abstract 2143: Profiling exhausted T cells using Vectra® Polaris™multiplex immunofluorescence assay in HNSCC

Background: Head and neck squamous cell carcinoma (HNSCC) is a cancer with the ability to modulate the immune system to evade detection. It is the sixth most frequently diagnosed cancer with 550,000 new cases and 300,000 lives lost worldwide per year. New treatments for HNSCC are urgently needed as patients continue to experience a high mortality rate and low response to surgery and chemotherapeutic treatments. Part of the reason why HNSCC is difficult to treat is it upregulates the expression of immune-checkpoint signaling molecule TIGIT (T cell immunoreceptor with Ig and ITIM domains) to inhibit T cell activation in vivo. Emerging evidence shows TIGIT overexpression in the CD8+ and CD4+ T cells that infiltrate the tumor cells of HNSCC patients. TIGIT expression is also associated with up-regulation of immune-checkpoint ligands PD-1 (programmed cell death protein 1) and LAG-3 (lymphocyte-activation gene 3 aka CD223), markers of T-cell exhaustion. Altogether, activation of the TIGIT/PD-1/LAG-3 axis correlates with an immunosuppressive microenvironment as well as cancer development and progression. Although there is ample evidence that the upregulation of TIGIT decreases the immune response in HNSCC, only limited studies have been published that address the location, expression and co-expression of TIGIT, LAG-3 and PD-1 in the HNSCC microenvironment.

Methods: In this study, we sought to establish a robust report of immune cells in the tissue of patients with HNSCC. Using Vectra Polaris multiplex immunofluorescence (IF) assays, we studied T-cell exhaustion and T-cell expression in HNSCC patient tissue using a total of 9 markers essential in cancer immunology. Sequential tissue sections were stained in two panels of 6, an exhausted T cell panel comprised of TIGIT, PD-1, LAG-3, panCK, CD4 and CD8 and a T cell panel including CD3, FOXP3, CD45RO, panCK, CD4 and CD8.

Results: Multiplexing IF staining revealed a HNSCC histologic landscape characteristic of immune suppression in this study. The data demonstrated abundant T cells with TIGIT overexpression in the tissue microenvironment of HNSCC samples. Using Indica Halo algorithms, we quantified exhausted T cells (TIGIT+PD1+LAG3+CD4+CD3+, TIGIT+PD1+LAG3+CD8+CD3+), T helper cells (CD3+CD4+), T cytotoxic cells (CD3+CD8+), T regulatory cells (CD3+CD4+FoxP3), memory T-cells (CD3+CD4+CD45RO) and anergic T-cells (PD1+CD8+) within the tumor and the stromal regions.

Conclusion: Currently AB154, a fully humanized immunoglobulin G1 monoclonal antibody targeting human TIGIT is in phase I clinical trials in HNSCC patients and BGB-A1217, an anti-TIGIT monoclonal antibody in combination with anti-PD-1 monoclonal antibody Tislelizumab is in a Phase 1/1b clinical trial in patients with advanced solid tumors. The Vectra Polaris imaging reported in this study identifies T cell composition in the tumor microenvironment of patients facing high mortality.

Citation Format: Sara Pollan, Arezoo Hanifi, Mate Nagy, Nicholas Stavrou, Erinn Parnell, Maricel Gozo, Nickolas Attanasio, Josette William, Qingyan Au. Profiling exhausted T cells using Vectra® Polaris™multiplex immunofluorescence assay in HNSCC [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2143.

Abstract 497: Characterization of TIGIT expression using MultiOmyxTM hyperplexed immunofluorescence assay in solid tumors

TIGIT (T cell immunoreceptor with Ig and ITIM domains) is a recently identified immune-receptor that is expressed on T cells, natural killer (NK) cells and NKT cells. TIGIT has emerged as an important coinhibitory receptor. TIGIT expression on CD8+ tumor infiltrating lymphocytes (TILs) has been shown to be upregulated in solid cancers such as melanoma, colon cancer, and NSCLC and have been associated with a dysfunctional phenotype in TILs. TIGIT can also be activated on a subset of Regulatory T cells (Tregs) and may be critical in driving CD8+ T cell dysfunction. A second mechanism in which TIGIT inhibits immunosurveillance is through both competition and direct inhibition of CD226, impairing its ability to activate immunosurveillance. TIGIT has also been reported to inhibit T cell responses indirectly by triggering CD155 expression in dendritic cells (DCs), thereby preventing DC maturation. These findings render the TIGIT pathway as an attractive candidate for cancer immunotherapy In this study, we seek to use MultiOmyx hyperplexed immunofluorescence (IF) assay to exploit this new pathway and characterize the TIGIT expression in a total of 20 melanoma and NSCLC samples. The cancer FFPE slides will be stained with a 13-marker panel including TIGIT, CD226, CD155, CD3, CD4, CD8, FOXP3, CD56, CD45, CD11b, CD11c, PD1, LAG3, TIM3 and PanCK. This panel will enable the detection of TIGIT, CD226 and CD115 expression in the Melanoma and NSCLC samples. The TIGIT expression will be further characterized on different TILs including CD4+ helper T cells, CD8+ cytotoxic T cells, Tregs and NK cells. Using the MultiOmyx proprietary algorithm, we can quantify different subtypes of TIGIT expressing cells and measure the distance of different TIGIT expressing cells to the tumor. Increased expression of TIGIT and PD1 has been demonstrated in NSCLC and melanoma. Moreover, TIGIT+ Tregs have been reported to upregulate TIM-3 expression in mice tumor models. In this study, we will also evaluate the expression of TIGIT in conjunction with other coinhibitory receptors such as PD1, LAG3 and TIM3. The percentage of TILs that co-express TIGIT/PD1, TIGIT/LAG3 and TIGIT/TIM3 will be quantified and analyzed. Leveraging TIGIT in combination with other immune therapy may achieve more robust clinical outcomes. There are currently multiple phase I clinical trials using TIGIT monoclonal antibody (for instance, BMS-986207 by Bristol-Myers Squibb and MTIG7192A by Genetech) in combination with anti PD1/PDL1 antibodies in solid tumors. Our data can help provide more insight into how TIGIT modulate antitumor immunity in melanoma and NSCLC tumors. And the findings in this study can also be used to understand the synergistic effects between TIGIT and other coinhibitory receptors and help identify the additional opportunity for combination immunotherapy using checkpoint inhibitors.

Citation Format: Qingyan Au, Arezoo Hanifi, Erinn Parnell, Judy Kuo, Eric Leones, Flora Sahafi, Kathy Pham, RaghavKrishna Padmanabhan, Nicholas Hoe, Josette William. Characterization of TIGIT expression using MultiOmyxTM hyperplexed immunofluorescence assay in solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 497.

Immune Profiling and Quantitative Analysis Decipher the Clinical Role of Immune-Checkpoint Expression in the Tumor Immune Microenvironment of DLBCL

PD-1/L1 and CTLA-4 blockade immunotherapies have been approved for 13 types of cancers and are being studied in diffuse large B-cell lymphoma (DLBCL), the most common aggressive B-cell lymphoma. However, whether both PD-1 and CTLA-4 checkpoints are active and clinically significant in DLBCL is unknown. Whether PD-1 ligands expressed by tumor cells or by the microenvironment of DLBCL are critical for the PD-1 immune checkpoint is unclear. We performed immunophenotypic profiling for 405 patients with de novo DLBCL using a MultiOmyx immunofluorescence platform and simultaneously quantitated expression/coexpression of 13 immune markers to identify prognostic determinants. In both training and validation cohorts, results demonstrated a central role of the tumor immune microenvironment, and when its functionality was impaired by deficiency in tumor-infiltrating T cells and/or natural killer cells, high PD-1 expression (but not CTLA-4) on CD8⁺ T cells, or PD-L1 expression on T cells and macrophages, patients had significantly poorer survival after rituximab-CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) immunochemotherapy. In contrast, tumor-cell PD-L2 expression was associated with superior survival, as well as PD-L1⁺CD20⁺ cells proximal (indicates interaction) to PD-1 ⁺ CD8⁺ T cells in patients with low PD-1 ⁺ percentage of CD8⁺ T cells. Gene-expression profiling results suggested the reversibility of T-cell exhaustion in PD-1⁺/PD-L1⁺ patients with unfavorable prognosis and implication of LILRA/B, IDO1, CHI3L1, and SOD2 upregulation in the microenvironment dysfunction with PD-L1 expression. This study comprehensively characterized the DLBCL immune landscape, deciphered the differential roles of various checkpoint components in rituximab-CHOP resistance in DLBCL patients, and suggests targets for PD-1/PD-L1 blockade and combination immunotherapies.

Abstract 2256: Efficient large-scale cell classification and analysis for MultiOmyxTMassays: A deep learning approach

Traditional immunohistochemistry (IHC) techniques utilize one slide per biomarker. When clinical samples are precious and the number of serial sections is limited, comprehensive biomarker profiling becomes difficult with IHC. MultiOmyx is a proprietary, multiplexing methodology capable of staining up to 60 biomarkers on the same slide. The output of the assay enables quantitative profiling of tissues at a single cell level. The assay generates data for millions of cells with billions of queryable data points. To detect and classify cells efficiently at this large scale, we developed an image analysis framework using Deep Learning. Our framework consists of seven major steps: (1) manual annotation of a small subset of the nuclear staining channel (DAPI); (2) training of a fully convolutional neural network [1] on this annotation-set to generate a feature map identifying cell centers; (3) application of the trained network in (2) on the nuclear stain (DAPI) of the entire dataset to delineate individual cells; (4) manual annotation of a small subset of each of the other biomarker channels; (5) training of a convolutional neural network on these annotation-sets for binary classification of each of the biomarkers; (6) application of these classifiers to the entire dataset; (7) combination of the binary classification results to identify phenotypes of interest. For our output, we provide both visual label maps and classification summary tables for individual and co-localized biomarkers at the region of interest level and the entire slide level. In addition, combining the phenotype and location information allows us to visualize complex spatial relationships in the tissue.

The benefits of using this Deep Learning framework are greatly felt through increased time efficiency without a loss in accuracy, when compared to more traditional computer vision methods requiring high levels of parameter fine-tuning. For future work, we plan on fully automating the approach as more manual annotation-sets are generated.

Citation Format: Mate L. Nagy, Arezoo Hanifi, Ahalya Tirupsur, Geoffrey Wong, Jun Fang, Nicholas Hoe, Qingyan Au, Raghav K. Padmanabhan. Efficient large-scale cell classification and analysis for MultiOmyxTMassays: A deep learning approach [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2256.

Abstract 2092: Tumor-infiltrating myeloid cells: Using MultiOmyx to distinguish between TAMs, TANs, and MDSCs in the pancreatic tumor microenvironment

Myeloid cells are the primary recruited effector cells during inflammation. A subset of these, consisting primarily of tumor-associated M1/M2 macrophages (TAMs), tumor-associated neutrophils (TANs), and myeloid-derived suppressor cells (MDSCs), accumulate in tumors where they establish an inflammatory tumor microenvironment (TME) that is favorable for tumor progression (1-3). While these tumor-infiltrating myeloid cells (TIMs) are known to be major regulators of tumor-associated immune-suppression, angiogenesis, metastasis, and resistance to anti-cancer therapy much still remains to be understood about how they function within the TME and how they co-operate with tumor-resident lymphocytes to regulate antitumor immunity. Targeting TIMs to either eliminate them or to convert them from their immune-suppressive to an immune-stimulatory state has emerged as a new strategy complementing current cancer immunotherapy strategies. However, a major impediment to understanding the complexity of the distinct functions of subsets of TIMs and their spatial distribution within the TME is the ability to phenotypically characterize TIMs in FFPE tissues by standard immunohistochemistry, as most IHC studies do not utilize a panel of antibodies broad enough to characterize multiple myeloid cell subsets in the same sample.

We will be using MultiOmyx, a proprietary, multiplexing assay with similar staining characteristics as standard IHC stains but with the significant advantage that up to 60 protein biomarkers can be interrogated from a single FFPE section (4). MultiOmyx protein immunofluorescence (IF) assays utilize a pair of directly conjugated Cyanine dye-labeled (Cy3, Cy5) antibodies per round of staining. Each round of staining is imaged and followed by novel dye inactivation chemistry, enabling repeated rounds of staining and deactivation. Using this MultiOmyx multiplexing assay in combination with a proprietary algorithm that takes into account the staining pattern for each specific biomarker, we will seek to identify different subsets of TIMs and their spatial relationship in tumor tissue from patients with pancreatic ductal adenocarcinoma (PDAC), characterized by an excessive amount of desmoplastic stroma seeded with inflammatory cells.

TIM Panel: CD11b, CD14, CD15, CD16, CD33, CD68, CD163, HLA-DR, Arginase, PANCK.

Human Myeloid Cell Subset Markers: M1 TAMs (CD68+HLA-DR+CD163-), M2 TAMs (CD68+CD163+HLA-DR+), TANs (CD11b+CD15+CD16+HLA-DR+Arginase+), M-MDSCs (CD11b+CD14+CD33+HLA-DRLoCD15-), G-MDSC (CD11b+CD15+CD33+Arginase+HLA-DRLoCD14-).

References

1.De Palma M, Lewis CE. Cancer Cell. 2013;23(3):277-286.

2. Rivera LB et al. Cell Rep. 2015;11(4):577-591.

3. Klemm F, Joyce JA. Trends Cell Biol. 2015;25(4):198-213.

4. Gerdes MJ et al. PNAS. 2013;110:11982-11987.

Citation Format: Anna Juncker-Jensen, Nicholas Hoe, Judy Kuo, Qingyan Au, Shijun Zhu, Eric Leones, Flora Sahafi. Tumor-infiltrating myeloid cells: Using MultiOmyx to distinguish between TAMs, TANs, and MDSCs in the pancreatic tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2092.

ERN1 and ALPK1 inhibit differentiation of bi-potential tumor-initiating cells in human breast cancer

Cancers are heterogeneous by nature. While traditional oncology screens commonly use a single endpoint of cell viability, altering the phenotype of tumor-initiating cells may reveal alternative targets that regulate cellular growth by processes other than apoptosis or cell division. We evaluated the impact of knocking down expression of 420 kinases in bi-lineage triple-negative breast cancer (TNBC) cells that express characteristics of both myoepithelial and luminal cells. Knockdown of ERN1 or ALPK1 induces bi-lineage MDA-MB-468 cells to lose the myoepithelial marker keratin 5 but not the luminal markers keratin 8 and GATA3. In addition, these cells exhibit increased β-casein production. These changes are associated with decreased proliferation and clonogenicity in spheroid cultures and anchorage-independent growth assays. Confirmation of these assays was completed in vivo, where ERN1- or ALPK1-deficient TNBC cells are less tumorigenic. Finally, treatment with K252a, a kinase inhibitor active on ERN1, similarly impairs anchorage-independent growth of multiple breast cancer cell lines. This study supports the strategy to identify new molecular targets for types of cancer driven by cells that retain some capacity for normal differentiation to a non-tumorigenic phenotype. ERN1 and ALPK1 are potential targets for therapeutic development.

Abstract 1469: Integrated analysis of microRNA, mRNA, and protein expression utilizing MultiOmyxTM and NanoStringTM from formalin-fixed paraffin-embedded, lung, head and neck, breast, and melanoma tumors

Cancer is characterized as a loss of normal cellular regulation, due to accumulation of mutations and disruption of complex biological pathways. MicroRNAs (miRNAs) regulation of co-stimulatory and immune checkpoint pathways have been implicated as one of the potential mechanisms for cancer evasion in immuno-oncology. It is estimated that 30% of all mRNA expression may be regulated by miRNAs, and some are either oncogenic or tumor suppressive. Complexity of miRNA regulation highlights the need for integrated assays, providing direct correlation between miRNA and mRNA, and protein expression. From a single 4 µm FFPE section, MultiOmyxTM hyperplexed immunofluorescent assay (demonstrated to stain up to 60 protein biomarkers) is utilized to measure CD3, CD4, CD8, CD16, CD56, Granzyme B, FoxP3, ICOS, OX40, OX40L, PD1, PDL1, HLA-DR, and Ki67 protein expression. From an adjacent 10 µm section, NanoStringTM nCounter PanCancer Immune Profiling Panel and Human v3 miRNA expression panel were utilized to comprehensively profile the expression of 770 mRNA and 800 miRNA.

Integrating MultiOmyx and NanoString technologies, the current study measured miRNA, mRNA, and protein expression in lung, head and neck, breast, and melanoma samples. For each indication, three samples were selected from a larger sample set, based on high protein expression of lymphocytes and macrophage markers (CD3, CD4, CD8, CD56, CD16), co-stimulator markers (ICOS, OX40), and immune checkpoint markers (PD1, PD-L1). Protein expression results indicate positive correlation between expression of ICOS and OX40 with higher infiltration of Thelper (CD3+CD4+), Tcytotoxic (CD3+CD8+), and effector T cells (CD3+CD8+Granzyme B). NanoString normalized mRNA counts for the protein biomarkers profiled indicate that all markers except for HLA-DR belong to low expressers group with counts ranging from 20-700. Comparison of protein expression to mRNA counts revealed inconsistencies in modulated markers (PD1, PD-L1, ICOS, OX40) which are attributed to differences in population of cells between the two sections. Direct assessment of up regulation of miRNA and down regulation of target mRNA could not be made for miRNAs reported in literature as negative regulator of PD-L1 (miR-34a, 34b, 34c), PD1 (miR-28, miR-107), FoxP3 (miR-210, miR-24, miR-31), and ICOS (miR-101). Analysis of multiple miRNAs (combinatorial targeting) mapped in context to mRNA, and their respective protein expression will be presented from lung, head and neck, breast, and melanoma cancer samples.

Citation Format: Qingyan Au, RaghavKrishna Padmanabhan, Nam Tran, Lakshmi Chandramohan, Nicholas Hoe. Integrated analysis of microRNA, mRNA, and protein expression utilizing MultiOmyxTM and NanoStringTM from formalin-fixed paraffin-embedded, lung, head and neck, breast, and melanoma tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1469. doi:10.1158/1538-7445.AM2017-1469

Abstract 5135: Detection of IFNγ induced PDL1 expression by combined in situ RNA analysis and protein profiling from a single FFPE slide

PD1 ligands (PDL1) are often upregulated on the cell surface of many different tumors. The primary role of PDL1 in cancer is to inhibit T-cell mediated immune response. Two general mechanisms for PDL1 expression on tumor cells have been proposed. Innate immune resistance, in which PDL1 expression is induced by the constitutive oncogenic signaling, and adaptive immune resistance, in which PDL1 expression is induced by T-cells releasing interferon-γ (IFNγ) and activating the STAT signaling pathway. In order to differentiate between these two mechanisms, IFNγ mRNA expression is measured as an effective alternative to detecting IFNγ protein. Detection of cytokines by IHC is challenging as secreted proteins are widely diffused and the associated staining pattern appears to lack cellular specificity. RNAscope RNA in situ hybridization (ISH) assay is utilized to measure Interferon-γ (IFNγ) mRNA expression, and MultiOmyxTM multiplexed assay (demonstrated to stain up to 60 protein biomarkers) is utilized to measure CD3, CD4, CD8, CD56, CD68, PD1, and PDL1 protein expression. In this study, combined MO and RNAscope ISH assays, enabled identification of individual cells with characteristic mRNA and protein expression profile.

The MultiOmyx assay utilizes a pair of directly conjugated Cyanine dye-labeled (Cy3, Cy5) antibodies per round of staining. Each Cy-dye conjugated antibody recognizes different target proteins. Each round of staining is imaged and followed by novel dye inactivation chemistry, enabling repeated rounds of staining. RNAscope is a novel RNA ISH assay capable of single-molecule detection in individual cells, utilizing hybridization mediated signal amplification. The assay utilizes a pair of RNA target specific oligonucleotide probes, which sequentially hybridize to preamplifier, amplifier, and fluorophore label probes.

Utilizing MultiOmyx and RNAscope assays, this study proposed to profile both RNA and protein expression in lung, breast, melanoma, colorectal, esophageal, and prostate cancer samples.

Differentiating PDL1 expression induced in response to inflammatory signals produced by an activated T-cell, from PDL1 expression induced by constitutive oncogenic signaling, has potential implications in effectiveness of PD1 blockade therapy. According to the proposed mechanisms, PD1 blockade as a mono therapy may only benefit individuals with strong endogenous immune response. In individuals with weak endogenous immune response, combinational therapies consisting of both immune activation and PD1-pathway blockade may be more effective than either mono therapy alone.

Citation Format: Qingyan Au, Kathy Nguyen, Michael S. Lazare, Edward J. Moler, Nicholas Hoe. Detection of IFNγ induced PDL1 expression by combined in situ RNA analysis and protein profiling from a single FFPE slide. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5135.

Abstract 4146: MultiOmyx multiplexed tumor infiltrating lymphocyte panel provides comprehensive immunophenotyping from a single FFPE slide

Immune checkpoint therapies target immune regulatory pathways to enhance anti-tumor immune response. These therapies have contributed to important clinical advances, and are a promising approach to combat cancer. Development of effective immune checkpoint therapies requires an understanding of the host immune response within the tumor microenvironment. GE Healthcare, through its affiliate Clarient Diagnostics Service Inc., has developed a multiplexed Tumor Infiltrating Lymphocyte (TIL) Panel* consisting of 12 key cancer immune markers: CD3, CD4, CD8, CD20, CD45RO, CD56, CD68, CTLA4, FOXP3, PD1, PD-L1 and Pan-CK. This MultiOmyx TIL panel identifies individual Thelper (CD3+CD4+), Tcytotoxic (CD3+CD8+), Tregulatory (CD3+CD4+FoxP3), memory T-cells (CD3+CD4+CD45RO), anergic T-cells (PD1+CD8+), natural killer cells (CD56+), macrophages (CD68+), and B-cells (CD20+) within the tumor and the stromal regions.

MultiOmyx multiplexed immunofluorescence technology enables qualitative and quantitative analysis of these 12 biomarkers’ expression and co-localization from a single formalin-fixed, paraffin-embedded (FFPE) slide. MultiOmyx assay utilizes a pair of directly conjugated Cyanine dye-labeled (Cy3, Cy5) antibodies per round of staining. Each Cy-dye conjugated antibody recognizes different target proteins. Each round of staining is imaged and followed by novel dye inactivation chemistry, enabling repeated rounds of staining.

Herein, we report an analysis of immune response in the tumor microenvironment within solid tumors including breast cancer, lung cancer, colorectal cancer, esophageal cancer, prostate cancer, and melanoma utilizing the MultiOmyx TIL Panel. The results revealed two distinct immunologic phenotypes, high TIL tumors and Low TIL tumors. The high TIL tumors showed enhanced T cell population within the tumor and in the peritumoral stroma including CD8+ cytotoxic T cells, CD4+ helper T cells and CD45RO+ memory T cells. Increased expression of immune checkpoints markers such as CTLA4 and PD-1 were also observed. PD-1 was predominantly expressed in CD8+ cytotoxic T cells while CTLA4 was mostly found on CD4+ FOXP3+ regulatory T cells. PD-L1 expression was also induced, mainly on Pan-CK+ tumor cells and CD68+ macrophages. High density of PD-L1+ |CD68+ macrophages was localized in the stroma surrounding the tumor region. Conversely, many of the immune markers were not expressed in low-TIL tumors.

Immunophenotyping analysis offered by the MultiOmyx TIL panel may facilitate the identification of appropriate immunotherapeutic approach. Tumor with high TIL profiling may be effectively treated with single-agent immune checkpoint therapy, while tumor with low TIL profiling may require combination therapy with an agent that enhances endogenous antitumor response.

*MultiOmyx TIL panel is currently research use only tool.

Citation Format: Qingyan Au, Kathy Nguyen, Raghav Padmanabhan, Anne Kuller, Eddie Moler, Nicholas Hoe. MultiOmyx multiplexed tumor infiltrating lymphocyte panel provides comprehensive immunophenotyping from a single FFPE slide. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4146.

Abstract 2014: MultiOmyxTM: A multiplexed immunofluorescent assay capable of profiling protein expression and phosphorylation, in combination with next-generation sequencing from a single FFPE tissue section

A comprehensive signaling pathway profile in combination with mutational analysis may be a critical guide for selecting effective clinical strategies for targeted drugs in combinations or in sequential regimens. In the United States, colorectal cancer (CRC) is the third most common cancer and the third leading cause of tumor associated death in men and women. CRC is a heterogeneous disease defined by different receptor tyrosine kinase (RTK) activation, signaling through phosphatidylinositol 3-kinase (PI3K)/AKT and RAS/MAP2K pathways. In addition, activating mutations in RTK and/or activating or loss-of-function mutations in downstream intracellular signaling proteins, can alter the efficacy of targeted drugs resulting in an ineffective treatment.

GE Healthcare, through its affiliate Clarient Diagnostic Services Inc., has developed a novel hyperplexed multi-omic technology, MultiOmyx, to enable visualization and characterization of multiple biomarkers across multiple assays on a single 4μm tissue section. MultiOmyx protein immunofluorescence (IF) assays utilize a pair of directly conjugated Cyanine dye-labeled (Cy3, Cy5) antibodies per round of staining. Each round of staining is imaged and followed by novel dye inactivation chemistry, enabling repeated rounds of staining and deactivation for up to 60 protein biomarkers. The same protein IF processed slide is then used to perform DNA FISH assay, followed by DNA extraction using laser capture microdissection (LCM) from region(s) of interest for next generation sequencing.

Herein, we report an analysis of the key receptor tyrosine kinases (EGFR, HER2, HER3, and cMET) and their downstream signaling proteins (PI3K, phospho AKT, and phospho ERK1/2) in 10 colon tumor samples using the MultiOmyx technology. Mutational analysis was performed on LCM extracted tumor and tumor adjacent regions, and sequenced using the Ion AmpliSeq cancer panel, consisting of 50 targeted genes. Protein IF staining revealed heterogeneous expression and activation across different samples. With high EGFR and HER3 expression correlating with activation of AKT, possibly through EGFR:HER3 dimer, to high expression of EGFR correlating with positive staining for phospho ERK1/2. Additionally, intra-tumor heterogeneity was observed within the same tumor tissue, with varied expression of EGFR, HER2, and HER3. Preliminary mutational analysis also detected BRAF G469A known to increase kinase activity of BRAF and downstream activation of pERK1/2 in tumor but not in tumor-adjacent region.

The MultiOmyx assay can be utilized for clinical samples testing with limited size and availability. A comprehensive mutational and signaling protein analysis can guide oncologists tailored either a single or combinational targeted therapy to patient's individual tumor profile.

Citation Format: Qingyan Au, Maoyong Fu, Alexander Bordwell, Tripathi Pinky, Michael Lazare, Nam Tran, Nicholas Hoe. MultiOmyxTM: A multiplexed immunofluorescent assay capable of profiling protein expression and phosphorylation, in combination with next-generation sequencing from a single FFPE tissue section. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2014. doi:10.1158/1538-7445.AM2015-2014

Development of a high-throughput screening assay for cytoprotective agents in rotenone-induced cell death

Parkinson's disease (PD) is a neurodegenerative disease featured by selective loss of substantia nigra neurons. Rotenone administration in animals induces neurodegeneration accompanied by α-synuclein-positive Lewy body-like inclusions, recapturing typical histopathological features of PD. In an effort to screen for small-molecule agents to reverse rotenone-induced cytotoxicity, we developed and validated a sensitive and robust assay with neuroblastoma SK-N-SH cells. This assay was amenable to a high-throughput screening format with Z' factor of 0.56. Robotic screening of a bioactive compound library led to the identification of carnosic acid that can effectively protect cells from rotenone treatment. Using a high-content image-based assay and Western blot analysis, we demonstrated that carnosic acid protects cells from rotenone stress by significant induction of HSP70 expression. Therefore, the assay reported here can be used to identify novel cytoprotective agents for clinical therapeutics of PD.

Identification of inhibitors of HSF1 functional activity by high-content target-based screening

Cancer cells are known to experience a high level of stress and may require constant repair for survival and proliferation. Recent studies showed that inhibition of heat shock factor 1 (HSF1), the key regulator for the stress-activated transcription of heat shock protein (HSP), can reduce the tumorigenic potential of cancer cells. Such a "nononcogene addiction" phenomenon makes HSF1 an attractive cancer drug target. Here, the authors report an image-based high-content screening (HCS) assay for HSF1 functional inhibitors. A heat shock-based methodology was used to stimulate the stress response followed by quantitative measurement of HSF1/HSP70 granules for compound-induced inhibitory effects. The authors discovered a small molecule from a compound library that inhibits HSF1 granule formation substantially in heat-shocked HeLa cells with IC(50) at 80 nM. Electorphoretic mobility shift of HSF1 by this compound suggested significant inhibition of HSF1 phosphorylation, accompanied by reduced expression levels of HSP70 and HSP90 after heat induction. Importantly, HeLa cells stably transfected with HSF1 shRNA were more resistant to the compound treatment under lethal temperature than cells containing HSF1, further validating an HSF1-dependent mechanism of action. The HCS assay the authors developed was robust with a Z' factor of 0.65 in a 384-well plate format, providing a valuable method for identifying small-molecule functional inhibitors of HSF1 for potential cancer treatment.

Identification of a small molecule SIRT2 inhibitor with selective tumor cytotoxicity

As a member of the class III histone deacetylases, Sirtuin-2 (SIRT2) is critical in cell cycle regulation which makes it a potential target for cancer therapeutics. In this study, we identified a novel SIRT2 inhibitor, AC-93253, with IC(50) of 6 microM in vitro. The compound is selective, inhibiting SIRT2 7.5- and 4-fold more potently than the closely related SIRT1 and SIRT3, respectively. AC-93253 significantly enhanced acetylation of tubulin, p53, and histone H4, confirming SIRT2 and SIRT1 as its cellular targets. AC-93253 as a single agent exhibited submicromolar selective cytotoxicity towards all four tumor cell lines tested with a therapeutic window up to 200-fold, comparing to any of the three normal cell types tested. Results from high content analysis suggested that AC-93253 significantly triggered apoptosis. Taken together, SIRT2 selective inhibitor AC-93253 may serve as a novel chemical scaffold for structure-activity relationship study and future lead development.