mIF Panel Research Articles

Abstract 2756: A toolbox enabling a data-driven selection of regions of interest in tumor tissue sections for imaging mass cytometry analysis

Abstract Highly multiplexed imaging technologies are increasingly applied in biomedical research to identify underlying patterns of disease. As part of an IMI-funded project, the IMMUcan (Integrated immunoprofiling of large adaptive cancer patients cohorts) consortium will generate single-cell resolved molecular profiles for thousands of cancer patients from hospitals across Europe. Patient tissues will be profiled using two imaging modalities, multispectral immunofluorescence (mIF) and Imaging Mass Cytometry (IMC), in junction with data from different omics techniques. IMC enables highly multiplexed tissue imaging, but its current throughput only enables to image a fraction of large/whole-slide tumor tissue sections. The selection of regions of interest for IMC analysis is of critical importance with respect to the cells present in an area and the antibody panel used for the analysis of such an area. To target areas of interest in tumor tissue sections, e.g. with immune cell infiltration or tertiary lymphoid structures, we have developed a data driven approach based on mIF imaging analyses from consecutive sections to select such regions of interest for analysis with IMC. A mIF panel was designed to stain T cells, B cells, macrophages, neutrophils, dendritic cells and tumor cells to have an overview of immune cell populations within the whole tissue. After quantification, mIF images and data are hosted on a webserver and regions that contain cell-types of interest are annotated with the help of a web-based tool. mIF images are used for alignment with brightfield whole slide scans of the consecutive FFPE sections which are used for IMC. After alignment, the annotated regions of interest are acquired in IMC. The mean positional offset from IF to IMC for regions of interest was roughly 120 µm and the overall cell-type numbers across both technologies were well correlated. The approach presented here has the great advantage to allow us to reproducibly target similar areas for IMC across thousands of samples based on mIF within the IMMUcan consortium and will enable single-cell analysis for whole tumor sections based on mIF and IMC to gain a deeper understanding of the underlying tumor biology. Funding: IMI2 JU grant agreement 821558, supported by EU's Horizon 2020 and EFPIA. Citation Format: Daniel Schulz, Stephanie Renaud-Tissot, Robin Liechti, Nils Eling, Michelle Daniel, Jonas Windhager, Yervand Karapetyan, Thomas Mrowiec, Sylvie Rusakiewicz, Henoch S. Hong, Marie Morfouace, Bernd Bodenmiller. A toolbox enabling a data-driven selection of regions of interest in tumor tissue sections for imaging mass cytometry analysis [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 2756.

Multi-institutional TSA-amplified Multiplexed Immunofluorescence Reproducibility Evaluation (MITRE) Study

BackgroundEmerging data suggest predictive biomarkers based on the spatial arrangement of cells or coexpression patterns in tissue sections will play an important role in precision immuno-oncology. Multiplexed immunofluorescence (mIF) is...

Methods to Determine and Analyze the Cellular Spatial Distribution Extracted From Multiplex Immunofluorescence Data to Understand the Tumor Microenvironment.

Image analysis using multiplex immunofluorescence (mIF) to detect different proteins in a single tissue section has revolutionized immunohistochemical methods in recent years. With mIF, individual cell phenotypes, as well as different cell subpopulations and even rare cell populations, can be identified with extraordinary fidelity according to the expression of antibodies in an mIF panel. This technology therefore has an important role in translational oncology studies and probably will be incorporated in the clinic. The expression of different biomarkers of interest can be examined at the tissue or individual cell level using mIF, providing information about cell phenotypes, distribution of cells, and cell biological processes in tumor samples. At present, the main challenge in spatial analysis is choosing the most appropriate method for extracting meaningful information about cell distribution from mIF images for analysis. Thus, knowing how the spatial interaction between cells in the tumor encodes clinical information is important. Exploratory analysis of the location of the cell phenotypes using point patterns of distribution is used to calculate metrics summarizing the distances at which cells are processed and the interpretation of those distances. Various methods can be used to analyze cellular distribution in an mIF image, and several mathematical functions can be applied to identify the most elemental relationships between the spatial analysis of cells in the image and established patterns of cellular distribution in tumor samples. The aim of this review is to describe the characteristics of mIF image analysis at different levels, including spatial distribution of cell populations and cellular distribution patterns, that can increase understanding of the tumor microenvironment.

Profiling of immune checkpoint biomarkers by multiplex immunofluorescence in breast cancer brain metastases.

2021 Background: Despite potential clinical implications, the complexity of immune microenvironment in breast cancer (BC) brain metastases (BM) is still poorly understood. Multiplex immunofluorescence (mIF) allows simultaneous visualization of several IF labeled proteins while maintaining spatial information. This novel technique can be used to comprehensively describe BCBM immune microenvironment, potentially providing useful information to guide novel therapeutic approaches. Methods: Clinical data and archival BM samples from 60 BC patients undergoing neurosurgery (2003-2018) at three institutions were collected. BCBMs were characterized using a custom mIF panel, including immune checkpoint and co-inhibitory molecules (CD3, PD1, PD-L1, TIM3, LAG3, CD163) and localization (keratin for tumor recognition) markers. Mean marker density was determined by digital image analysis (positive cells/mm2) and classified in tumor and stroma areas. Associations between immune marker densities, BC subtype and overall survival from BM diagnosis (OS) were studied. Results: Sixty BCBM samples were analyzed; 32% HR+/HER2-, 38% HER2+, 30% HR-/HER2-. At a median follow-up of 43 months, the only clinical variable associated with OS was BC subtype (shortest for HR-/HER2- and longest for HER2+, p=0.02). In the total sample area and tumor area, no significant difference in marker density was observed according to BC subtype. In the stroma area, a significant difference in TIM3+ cell density was observed according to BC subtype (highest density in HR+/HER2- and lowest density in HER2+ tumors, Kruskal-Wallis p=0.017). Higher CD163 density (a marker of M2 macrophage polarization), both in the tumor and in the stroma area, was significantly associated with worse OS, even after correction by BC subtype. In the subgroup of patients with HR+/HER2- BCBM, high TIM3+ cell density in the stroma area was significantly associated with longer OS (median OS 54.1 versus 23 months respectively for TIM3+ density above and below median value; p=0.01). Conclusions: In BCBM, stromal TIM3+ immune infiltrate differs according to BC subtype. M2 macrophage polarization is consistently associated with worse OS across all BC subtypes and might represent a potential therapeutic target for these patients.[Table: see text]

Immuno-profiling and cellular spatial analysis using five immune oncology multiplex immunofluorescence panels for paraffin tumor tissue

Multiplex immunofluorescence (mIF) has arisen as an important tool for immuno-profiling tumor tissues. We updated our manual protocol into an automated protocol that allows the use of up to seven markers in five mIF panels to apply to formalin-fixed paraffin-embedded tumor tissues. Using a tyramide signal amplification system, we optimized five mIF panels that included cytokeratin to characterize malignant cells (MCs), immune checkpoint markers (i.e., PD-L1, B7-H3, B7-H4, IDO-1, VISTA, LAG3, ICOS, TIM3, and OX40), tumor-infiltrating lymphocytic markers (i.e., CD3, CD8, CD45RO, granzyme B, PD-1, and FOXP3), and markers to characterize myeloid-derived suppressor cells (i.e., CD68, CD66b, CD14, CD33, Arg-1, and CD11b). To determine analytical reproducibility and the impact of those panels for immuno-profiling tumor tissues, we performed an exploratory analysis in a set of non–small cell lung cancer (NSCLC) samples. The slides were scanned, and the different cell phenotypes were quantified by simultaneous co-localizations with the markers using image analysis software. Comparison between the time points of staining showed high analytical reproducibility. The analysis of NSCLC cases showed an immunosuppressive microenvironment with PD-L1/PD-1 expression as a predominant axis. Interestingly, high density of MCs expressing B7-H4 was correlated with recurrence. Unexpectedly, MCs expressing OX40 were also detected, and those cells were a closer distance to CD3+T-cells than were MCs expressing other immune checkpoints. Two different cellular patterns of spatial distribution were determined according the CD3 distribution, and the predominant pattern was related with active immunosuppressive interaction with MCs. Our study shows that these five mIF panels can identify multiple targets in a single cell with high reproducibility. The study of different cell populations and their spatial relationship can open new ideas for therapeutic approaches.

Identification of distinct immune landscapes using an automated nine-color multiplex immunofluorescence staining panel and image analysis in paraffin tumor tissues

Immune profiling is becoming a vital tool for identifying predictive and prognostic markers for translational studies. The study of the tumor microenvironment (TME) in paraffin tumor tissues such as malignant pleural mesothelioma (MPM) could yield insights to actionable targets to improve patient outcome. Here, we optimized and tested a new immune-profiling method to characterize immune cell phenotypes in paraffin tissues and explore the co-localization and spatial distribution between the immune cells within the TME and the stromal or tumor compartments. Tonsil tissues and tissue microarray (TMA) were used to optimize an automated nine-color multiplex immunofluorescence (mIF) panel to study the TME using eight antibodies: PD-L1, PD-1, CD3, CD8, Foxp3, CD68, KI67, and pancytokeratin. To explore the potential role of the cells into the TME with this mIF panel we applied this panel in twelve MPM cases to assess the multiple cell phenotypes obtained from the image analysis and well as their spatial distribution in this cohort. We successful optimized and applied an automated nine-color mIF panel to explore a small set of MPM cases. Image analysis showed a high degree of cell phenotype diversity with immunosuppression patterns in the TME of the MPM cases. Mapping the geographic cell phenotype distribution in the TME, we were able to identify two distinct, complex immune landscapes characterized by specific patterns of cellular distribution as well as cell phenotype interactions with malignant cells. Successful we showed the optimization and reproducibility of our mIF panel and their incorporation for comprehensive TME immune profiling into translational studies that could refine our ability to correlate immunologic phenotypes with specific patterns of cells distribution and distance analysis. Overall, this will improve our ability to understand the behavior of cells within the TME and predict new treatment strategies to improve patient outcome.

PD-L1 Multiplex and Quantitative Image Analysis for Molecular Diagnostics

Simple SummaryAssessing some diagnostic tests can be extremely difficult, even for highly trained clinicians. We have shown in the past that by using an advanced computer software program (QuPath), applied to high resolution images of patient tissue samples, we can assist pathologists in their assessment of a routine test that determines immunotherapy treatment. We also showed that by using a different testing method in the laboratory, called multiplexing, which detects several proteins at once rather than just one alone, we are subjectively more confident in the patient’s reported score. Here, we show that multiplexing is comparable to the traditional method, and that we can also easily apply our computer software tools to extract very specific information from the patient samples, which we are unable to do using the traditional laboratory method. We believe these tools can support pathologists to triage patient cases for this important diagnostic test.Multiplex immunofluorescence (mIF) and digital image analysis (DIA) have transformed the ability to analyse multiple biomarkers. We aimed to validate a clinical workflow for quantifying PD-L1 in non-small cell lung cancer (NSCLC). NSCLC samples were stained with a validated mIF panel. Immunohistochemistry (IHC) was conducted and mIF slides were scanned on an Akoya Vectra Polaris. Scans underwent DIA using QuPath. Single channel immunofluorescence was concordant with single-plex IHC. DIA facilitated quantification of cell types expressing single or multiple phenotypic markers. Considerations for analysis included classifier accuracy, macrophage infiltration, spurious staining, threshold sensitivity by DIA, sensitivity of cell identification in the mIF. Alternative sequential detection of biomarkers by DIA potentially impacted final score. Strong concordance was observed between 3,3’-Diaminobenzidine (DAB) IHC slides and mIF slides (R2 = 0.7323). Comparatively, DIA on DAB IHC was seen to overestimate the PD-L1 score more frequently than on mIF slides. Overall, concordance between DIA on DAB IHC slides and mIF slides was 95%. DIA of mIF slides is rapid, highly comparable to DIA on DAB IHC slides, and enables comprehensive extraction of phenotypic data and specific microenvironmental detail intrinsic to the sample. Exploration of the clinical relevance of mIF in the context of immunotherapy treated cases is warranted.

Abstract 4248: Multispectral immunofluorescence identifies pS6 as a biomarker of intrinsic resistance to ruxolitinib in patients with relapsed/refractory T-cell lymphomas

Abstract Introduction: Peripheral and cutaneous T cell lymphomas (TCL) are highly heterogeneous diseases that in the relapsed/refractory (R/R) stage have poor outcome. JAK/STAT, Pi3K/AKT and MAP kinase signaling are commonly involved in the pathogenesis of these lymphomas, and the efficacy of small molecule inhibitors targeting these pathways is being assessed in phase II trials. However, biomarkers to define patients most likely to benefit from specific inhibitors are lacking and mechanisms of acquired resistance in TCLs have not been reported. Methods: We are conducting two early phase trials in R/R TCL: 1) NCT02974647 targets JAK1/JAK2 with ruxolitinib (RUX, a JAK 1-2 inhibitor) and has enrolled 53 patients and 2) NCT02783625 targets Pi3K/AKT with duvelisib (DUV, a Pi3K gamma-delta inhibitor) in combination with either romidepsin or bortezomib and has enrolled 92 patients. We analyzed pre-, on and post- treatment biopsies with multiplex immunofluorescence (mIF) using the Vectra platform, and analyzed images for marker expression and cellular location using HALO software. We designed multiple mIF panels with 6 markers per slide, 3 identifying TCL cells based on disease-specific phenotypes (e.g. CD4+PD1+CD8- for angioimmunoblastic TCL), pSTAT3,5, pS6, and a macrophage marker (CD68). Areas of TCL involvement within biopsy slides were manually defined during image analysis based on mIF and comparison to clinical immunohistochemistry. Differences in marker expressions were analyzed with the Wilcoxon test using the program R. Results: We analyzed 53 biopsies from 39 patients with high-quality mIF images, of which 17 were from lymph nodes and 36 were from extranodal sites. Of 39 patients, 13 were treated with RUX (9 biopsies pre-, 7 on, 4 post- treatment) and 26 with DUV (20 biopsies pre-, 6 on, 10 post- treatment). Median number of total TCL cells in pre-treatment biopsies was 1041 (range, 97-6463) and median TCL cell fraction within involved areas was 27.86% (range, 5.12-88.02%). Median macrophage fraction within involved areas was 5.92% (range, 2.89-11.88%). For either agent, response to treatment was not associated with: 1) quantity of macrophages (p=0.1 for responders vs non-responders to DUV, p=0.53 for RUX), 2) average distance between TCL cells and closest macrophage, 3) TCL cell fraction within involved areas, or 4) fraction of TCL cells expressing pSTAT3/5 (Figure). In contrast, pS6 expression in <25% of TCL cells within pre-treatment biopsy completely distinguished responders from non-responders (Figure; p=0.02). Neither agent induced consistent changes in expression of pSTAT3/5 or pS6 at the on or post-treatment timepoints. Conclusions: mIF is a feasible platform for biomarker discovery and translation in TCL, a set of lymphomas with heterogeneous immunophenotypes. pS6 expression in <25% of TCL cells may predict response to ruxolitinib, suggesting that activation of S6 kinase downstream of PI3K or MAP kinase, may overcome dependence on JAK/STAT signaling. This is also the first data suggesting that ruxolitinib responses involve a TCL cell-intrinsic mechanism of action. Based on these data, we are assessing the use of pS6 IHC as a clinically-validated biomarker in patients with TCL prior to ruxolitinib. Citation Format: Paola Ghione, Alison Moskowitz, Priadarshini Kumar, Andrea Knezevic, Venkatraman Seshan, Eric Jacobsen, Jia Ruan, Johnathan Schatz, Sarah Noor, Patricia Myskowski, Helen Hancock, Theresa Davey, Obadi Obadi, Carlissa Onwasigwe, Nivetha Ganesan, Lauren Pomerantz, Christine Jarjies, Allison Sigler, Mark Geyer, Ariela Noy, David Straus, Natasha Galasso, Giorgio Inghirami, Steven Horwitz, David Weinstock, Travis Hollmann, Ahmet Dogan. Multispectral immunofluorescence identifies pS6 as a biomarker of intrinsic resistance to ruxolitinib in patients with relapsed/refractory T-cell lymphomas [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 4248.

906 Early effector T-cell densities predict response in patients with advanced Merkel cell carcinoma treated with anti-PD-1

Background: We previously nominated PD-1+ cell densities in pre-treatment tumor specimens as a potential biomarker for response to pembrolizumab (anti-PD-1) in patients with advanced Merkel cell carcinoma (MCC). Here, we used multiplex immunofluorescence (mIF) in a separate cohort of MCC patients treated with anti-PD-1 to determine which immune cell subset(s) contribute PD-1 to the tumor microenvironment (TME). We also characterized PD-1 expression intensity on lymphocytes, as an indication of their functional state, and tested for an association with anti-PD-1 response. Methods: We performed mIF staining on pre-treatment formalin-fixed paraffin embedded tumor specimens from 23 unique patients. The mIF panel included PD-1, CD4, CD8, CD20, FoxP3, and CD56 (tumor marker). The density of each cell type and intensity of PD-1 expression (PD-1lovs. PD-1hi) were determined and correlated with response assessments by RECIST1.1. Results: Total PD-1 cell densities were higher in responders vs. non-responders, supporting our original findings. Additionally, median CD4, CD8 and CD20 cell densities were increased in those who responded to therapy. While CD8+ cells are often cited as contributing the majority of PD-1 to the TME, we found an equal proportion was contributed by CD4+ cells. Approximately 4% of the CD8+PD-1+ cells also expressed FoxP3. Of the immune cell phenotypes tested, CD8+FoxP3+PD-1lodensities showed the strongest association with response to anti-PD-1 (p=0.01, Mann-Whitney U test). Conclusions: The next generation of tissue-based biomarkers for predicting response to immunotherapy are likely to utilize multiplex IF/IHC technologies, which can quantify complex cell phenotypes in situ. Here, we highlight CD8+FoxP3+PD-1loas a key cell type associated with anti-PD-1 response. Representing early, tumor-specific effector T-cells with strong proliferative potential and IFN-gamma secretion, CD8+FoxP3+ cells warrant additional exploration as a predictive biomarker.

Immune microenvironment profiling of breast cancer brain metastases using multiplex immunofluorescence.

2526 Background: Despite clinical implications, the complexity of brain metastases (BM) immune microenvironment in breast cancer (BC) patients is poorly understood. Multiplex immunofluorescence (mIF), a novel imaging technique allowing simultaneous visualization and quantification of several IF labeled proteins while maintaining spatial information, holds promise to comprehensively describe BCBM immune microenvironment, potentially providing valuable information to improve treatment. Methods: Clinical data and archival BM samples were collected for 60 BC patients undergoing neurosurgery (2003-2018) at three institutions. BM immune contexture was characterized using a custom mIF panel, including cell subtyping (CD4, CD8, FOXP3, CD68), activation (Granzyme B) and localization (keratin for tumor recognition) markers. Mean immune cell density (cells/mm2) for each sample was determined by digital image analysis and classified in tumor and stroma areas. Associations between immune subpopulations, BC subtype and overall survival from BM diagnosis (OS) were studied. Results: Up to date, 30 BCBM samples have been analyzed; 33% HR+/HER2-, 20% HR-/HER2+, 10% HR+/HER2+, 37% HR-/HER2-. At a median follow-up of 46 months, BC subtype was the only clinical variable associated with OS (longest for HER2+ and shortest for HR-/HER2-, log-rank p = 0.002). In the total sample area, no significant difference in immune cell densities was observed according to BC subtype. In the tumor area, HR+/HER2- tumors showed higher densities of CD8+ and CD68+ cells compared to other subtypes (p = 0.036 and p = 0.016, respectively). In stroma, HR-/HER2- tumors presented numerically higher densities of CD4+ and FOXP3+ cells and higher ratio of CD4/CD8 and FOXP3/CD8 ratio (not statistically significant). Higher CD4/CD8 and FOXP3/CD8 ratio in the stroma was significantly associated with worse OS, even after correction by BC subtype (Table). Conclusions: In BCBM, immune infiltrate differs according to BC subtype. Preliminary results suggest that a more tolerogenic immune microenvironment is associated with worse OS and might represent a target for optimization of immunotherapy for these patients. Updated results for all 60 patients will be presented. [Table: see text]

Multiplex Immunofluorescence Image Cytometry Combined with Spatially-Resolved UMAP Defines Novel Immune Prognostic Biomarkers in Metastatic Melanoma

Abstract Multiplex IF (mIF) provides a detailed characterization of spatial relationships and complex cell phenotypes in the tumor microenvironment. However, the data-analysis and visualization is complex and time-consuming. Here, we developed a platform to analyze mIF data through flow cytometry workflows (image cytometry), while maintaining spatial information, and applied it to tissue microarrays of metastatic melanoma specimens (n=93; 6-plex mIF panel: PD-1, PD-L1, CD163, CD8, FoxP3, Sox10/S100). Then, we used a UMAP-based approach driven by cell-to-cell distances (rather than fluorescence intensity) to display and analyze geographic organization and cell interactions. Our pipeline provided equivalent results to the digital pathology gold standard with faster run times (5-fold reduction) and higher reproducibility. We identified key prognostic immune variables, including CD8PD1Low and CD8PD1Neg cells which associated with longer overall survival (OS, both p<0.01), and CD163PDL1Neg cells which associated with shorter OS (p=0.001). The spatial UMAPs showed that PD-L1 and PD-1 intensities were spatially encoded, and their expression on distinct cell subsets was organized in geographic clusters. Specifically, PD-L1Hi cells co-located to areas of CD8 cells, and PD-1Hi cells were observed near dense collections of tumor cells. Spatial UMAP subtraction analysis (survivors vs. non-survivors at 5 years) identified geographic and co-expression signatures associated with improved prognosis, i.e. CD8-driven PD-L1 expression and lacking CD163PDL1Neg macrophages. These data demonstrate the use of image cytometry and spatial UMAPs for improved visualization and interpretation of single-cell, spatially-resolved mIF data.

Abstract LB-318: Multi-institutional TSA-amplified Multiplexed Immunofluorescence Reproducibility Evaluation (MITRE study): Reproducibility assessment of an automated multiplexed immunofluorescence slide staining, imaging, and analysis workflow

Abstract Background: Emerging data suggests that predictive biomarkers based on the spatial arrangement of multiple cell types in FFPE tissue sections will be an important component of precision medicine in immune-oncology. Multiplexed immunofluorescence (mIF) facilitates such assessments. If mIF is to play a translational role in research and ultimately clinical practice, it is vital to refine, standardize, and validate an end-to-end workflow that supports large scale multi-site trials and clinical laboratory processes. To this end, six institutions collaborated to develop an automated 6-plex assay focused on the PD-1/PD-L1 axis and assessed its inter- and intra-site reproducibility. Specific attention was paid to assessment of %PD-L1 expression by immune cells (ICs), as pathologists have poor concordance for this parameter as noted in the Blueprint 2 and multi-institutional NCCN studies on PD-L1 IHC. Methods: A 7-color mIF panel (PD-L1, PD-1, CD8, CD68, FoxP3, Cytokeratin, and DAPI) was optimized on a Leica Bond Rx autostainer. Serial sections of tonsil and a lung cancer tissue-microarray (TMA), antibodies and TSA-Opal detection reagents (Akoya Biosciences) were distributed. Cell pellet arrays were also distributed and used to normalize batch variation in intensity measurements. Tonsil and TMA sections were stained at each site and imaged at 20x using a Vectra Polaris. Cells were segmented and phenotyped using image analysis algorithms. In tonsil sections, the average intensity of the top quartile of cells positive for each marker was assessed to identify potential variation in staining intensity. In lung TMAs, cell densities and %PD-L1 expression in immune cells (CD68+ and CD8+ cells) was determined. Results: The average staining intensity coefficients of variation (CV) for all markers within sites was 10% in tonsil samples. Inter-site concordance for tumor cell and immune cell subset densities in TMAs, had an average R2 value of 0.86 and slope of 0.96. Inter-site concordance for %PD-L1+ ICs had an average R2 value of 0.81, and slope of 0.82, in contrast to ICC values of <0.3 in the NCCN and Blueprint 2 studies. Conclusions: We demonstrate a reproducible end-to-end process for mIF characterization of the PD-1/PD-L1 axis including automated staining, multispectral imaging, and machine-learning-trained image analysis algorithms. This approach improved reproducibility of %PD-L1 IC assessment and brought it in line with %PD-L1 tumor cell assessment by pathologists. Sources of variation were identified and will be discussed. The described approach may serve as a template for assessing reproducibility of emerging mIF panels for other investigative teams, with an eye toward translating such approaches into clinical trials and ultimately into the clinic. Citation Format: Clifford Hoyt, Kristin Roman, Liz Engle, Chichung Wang, Carmen Ballesteros-Merino, Shawn M. Jensen, John McGuire, Yi Zheng, Carla Coltharp, Mei Jiang, Justin Lucas, Edwin Parra, Ignacio Wistuba, Darren Locke, Bernard A. Fox, David L. Rimm, Janis Marie Taube. Multi-institutional TSA-amplified Multiplexed Immunofluorescence Reproducibility Evaluation (MITRE study): Reproducibility assessment of an automated multiplexed immunofluorescence slide staining, imaging, and analysis workflow [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 LB-318.

Abstract 1180: Impact of the spatial analysis of tumor-associated lymphocytes and tumor-associated macrophages on recurrence at early stage of non-small cell lung carcinoma

Abstract Introduction. The interaction between malignant cells (MCs), stromal cells, tumor-associated lymphocytes (TILs), and tumor-associated macrophages (TAMs) is relevant for non-small cell lung carcinoma (NSCLC) progression. The spatial distribution of those cells may affect the prognosis and can be related to genetic intra-tumor heterogeneity (ITH). The aim of this study was to characterize the immunologic ITH and the spatial distribution of immune cells to MCs in primary NSCLC tumors at early stages using multiplex immunofluorescence (mIF) and image analysis approaches. Material and methods. We studied 33 surgically resected NSCLC cases (adenocarcinomas=23; squamous-cell carcinomas=10) with a history of recurrence in a follow-up of at least 60 months (recurrence, N=13; non-recurrence, N=15). Consecutive FFPE tissue sections were stained with two mIF panels (panel 1: cytokeratin (AE1/AE3), PD-L1, PD-1, CD3, CD8, and CD68; panel 2: AE1/AE3, CD3, CD8, granzyme-B, CD45RO, and FOXP3). Three intra-tumor regions (3mm2 each) per case were selected after gridding the whole tumor section. A total of 99 intratumor regions were scanned and analyzed using Vectra Multispectral-Microscope and InForm-software. From each intratumor region, TILs and TAMs densities, as well as the coefficient of variation, were evaluated. The median distance and the G-Cross area under the curve (AUC) for specific radial distances (10µm, 20µm, and 40µm) were obtained between TILs and TAMs phenotypes to MCs. Results. Recurrence was associated with higher MCs density and TAMs/TILs ratio, and lower TIL densities. A high ITH of cytotoxic T-cells (CTLs) PD-L1+ was associated with worse survival. The distance of TAMs PD-L1+ to MCs PD-L1 negative (60µm vs 25µm) or to MCs PD-L1 positive (25µm vs 13µm) was higher in the non-recurrence group than in recurrence group. Close TAMs PD-L1+ to MCs was associated with worst survival. In a radial distance of 10µm, 20µm, and 40µm, a higher infiltration of CTLs PD-1+, was observed in the group of recurrence than non-recurrence group, surrounding MCs PD-L1 negative (AUC 0.49, 3.80, and 20.03; vs AUC 0.01, 0.16, and 1.29, respectively), and MCs PD-L1 positive (AUC 0.60, 4.35, and 19.90; vs AUC 0.01, 0.20, and 2.20, respectively). A high infiltration of CTLs PD-1+ surrounding MCs, with or without expression of PD-L1, was associated with worse survival. All the differences were statistically significant (P<0.05). Conclusion. Close spatial proximity of antigen-experienced CTLs and TAMs PD-L1+ to MCs are associated with recurrence and poor survival in early stages of NSCLC. We determined that ITH of immune cell densities is associated with recurrence of surgically resected NSCLC. Tumor-immune cell spatial modeling offers a deep understanding of tumor microenvironment that impacts on clinical outcomes. Supported by CPRITRP160668 and UT Lung SPORE grants Citation Format: Alejandro Francisco-Cruz, Edwin R. Parra, Santhoshi N. Krishnan, Souptik Barua, Mei Jiang, Junya Fujimoto, Christine B. Peterson, Priyam Das, Chi-Wan Chow, Jaime Rodriguez-Canales, Carmen Behrens, Neda Kalhor, Annikka Weissferdt, John Heymach, Stephen Swisher, Boris Sepesi, Arvind Rao, J. Jack Lee, Cesar Moran, Andrew Futreal, Jianjun Zhang, Ignacio I. Wistuba. Impact of the spatial analysis of tumor-associated lymphocytes and tumor-associated macrophages on recurrence at early stage of non-small cell lung carcinoma [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 1180.

Abstract 491: Tumor heterogeneity and its impact on immunoprofiling data: Whole tumor vs regions of interest (ROI) analysis of multiplex immunofluorescence in four carcinoma types

Abstract Background: Multiplex immunofluorescence (mIF) is a key tool for cancer immunoprofiling in tissue samples. These techniques typically require an investigator to select regions of interest (ROI) within the tumor for digital image analysis. However, tumor heterogeneity and sampling bias may result in data that may not be representative of the whole tumor. Our goal was to compare immunoprofiling data from ROI analysis versus whole tumor in lung, colorectal, and bladder carcinomas using mIF and multispectral image analysis. Methods: 9 lung adenocarcinomas, 11 lung squamous cell carcinomas, 10 colorectal adenocarcinomas (CRC), and 10 bladder urothelial carcinomas (BC) were stained with a 6-marker mIF panel (PDL1, CD8, Ki67, CD68, AE1/AE3, PD1) using the Opal technique, and imaged using a Polaris multispectral scanner. Two pathologists independently selected 5 ROI (0.64 mm2) within the tumor for each case. To address the possibility of human sampling bias, 2 additional sets of 5 ROI were randomly generated in the tumor by a computer. HALO software was used to analyze the ROIs and the whole tumor area. The data was compared using Spearman’s ranked correlation coefficients. Results: 6 cell populations were assessed in the tumor epithelium and stroma: cytotoxic T-lymphocytes (CTL, CD8+), antigen-experienced CTL (CD8+/PD1+), proliferating CTL (CD8+/Ki67+), macrophages (CD68+), PDL1+ tumor cells (AE1AE3+/PDL1+), and proliferating tumor cells (AE1AE3+/Ki67+). These immunophenotypes were assessed in each set of 5 ROI and compared with data from the whole tumor. Spearman’s correlation coefficients (r) ranged from -0.25 to 1.00 depending on markers and tumor type. The best correlations across all tumor types were percentage of PDL1+ tumor cells (r=0.82 to 0.89), while cell densities of CTL and PD1+CTL showed the lowest correlations (r=0.35). Among tumor types, lung tumors and BC showed overall good correlation (r=0.82 to 0.83), depending on the marker. CRC showed the highest variability (average r=0.68), with correlation coefficients as low as 0.35 (PD1+CTL) and 0.46 (PDL1+macrophages). Computer selected ROI were not significantly more or less concordant with whole tumor analysis as compared to ROI selected by pathologists. Conclusion: Our results suggest that for immunoprofiling data, particularly proportion-based endpoints such as percent of PDL1+ tumor cells in lung and BC, ROI analysis has acceptable correlation with data from the whole tumor. However, CRC showed high variability, particularly in PD1+CTL, Ki67+CTL, and PDL1+macrophages, suggesting that CRC has higher tumor immunological heterogeneity compared with lung cancer and BC. These results suggest that investigators should analyze the whole tumor areas, otherwise, special attention should be paid to the image analysis strategy and its validation. Citation Format: Jaime Rodriguez-Canales, Michael Surace, Jennifer Cann. Tumor heterogeneity and its impact on immunoprofiling data: Whole tumor vs regions of interest (ROI) analysis of multiplex immunofluorescence in four carcinoma types [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 491.

Validation of multiplex immunofluorescence panels using multispectral microscopy for immune-profiling of formalin-fixed and paraffin-embedded human tumor tissues

Immune-profiling is becoming an important tool to identify predictive markers for the response to immunotherapy. Our goal was to validate multiplex immunofluorescence (mIF) panels to apply to formalin-fixed and paraffin-embedded tissues using a set of immune marker antibodies, with the Opal™ 7 color Kit (PerkinElmer) in the same tissue section. We validated and we described two panels aiming to characterize the expression of PD-L1, PD-1, and subsets of tumor associated immune cells. Panel 1 included pancytokeratin (AE1/AE3), PD-L1, CD4, CD8, CD3, CD68, and DAPI, and Panel 2 included pancytokeratin, PD-1, CD45RO, granzyme B, CD57, FOXP3, and DAPI. After all primary antibodies were tested in positive and negative controls by immunohistochemistry and uniplex IF, panels were developed and simultaneous marker expressions were quantified using the Vectra 3.0™ multispectral microscopy and image analysis InForm™ 2.2.1 software (PerkinElmer).These two mIF panels demonstrated specific co-localization in different cells that can identify the expression of PD-L1 in malignant cells and macrophages, and different T-cell subpopulations. This mIF methodology can be an invaluable tool for tumor tissue immune-profiling to allow multiple targets in the same tissue section and we provide that is accurate and reproducible method when is performed carefully under pathologist supervision.

OA20.05 The Influence of Neoadjuvant Chemotherapy, on Immune Response Profile in Non-Small Cell Lung Carcinomas

The clinical efficacy observed with PD-1/PD-L1 inhibitors in non-small cell lung carcinoma (NSCLC) has prompted to characterize the immune response in lung tumors treated with chemotherapy. Our goal was to determine the characteristics of immune microenvironment of localized, surgically resected, NSCLCs from patients who received and did not receive neo-adjuvant chemotherapy. Using multiplex immunofluorescence (mIF) and image analysis, we investigated PD-1/PD-L1 expression, and quantified tumor infiltrating lymphocytes (TILs) and tumor associated macrophages (TAMs). We studied formalin-fixed and paraffin embedded (FFPE) tumor tissues from 111 stage II and III resected NSCLC, including 61 chemonaïve (adenocarcinoma, ADC=33; squamous cell carcinoma, SCC=28) and 50 chemotherapy-treated (ADC=30; SCC=20) tumors. mIF was performed using the Opal 7-color fIHC Kit™ and analyzed using the Vectra™ multispectral microscope and inForm™ Cell Analysis software (Perkin Elmer, Waltham, MA). The markers studied were grouped in two 6-antibody panels: Panel 1, AE1/AE3 pancytokeratins, PD-L1 (clone E1L3N), CD3, CD4, CD8 and CD68; and Panel 2, AE1/AE3, PD1, Granzyme B, FOXP3, CD45RO and CD57. Positive PD-L1 expression (>5%) in malignant cells (MCs) was detected in 48% (n=53/111) of NSCLCs. Overall, chemotherapy-treated tumors showed significantly higher percentages of MCs expressing PD-L1 (median, 18.2%) than chemo-naïve cases (median, 1.8%; P=0.033). Higher densities of inflammatory cells expressing granzyme B (P=0.036), CD57 (P=0.001) and PD-1 (P=0.016) were detected in chemotherapy-treated NSCLCs compared with chemo-naïve tumors. In contrast, lower densities of FOXP3-positive regulatory T cells were detected in chemotherapy-treated tumors when compared with chemo-naïve cases (P=0.032). Following chemotherapy ADCs exhibited significantly higher levels of CD57-positive cells (P<0.0001) and lower density of FOXP3-positive cells (P=0.002) than chemo-naïve tumors. Chemotherapy-treated SCCs demonstrated higher density of PD-1-positive cells than chemo-naïve tumors (P=0.004). In chemotherapy-treated cancers, lower levels of CD4 helper T positive cells and tumor associated macrophages (TAMs) CD68-positive cells were associated with worse overall survival (OS; P=0.04 and P=0.005, respectively) in univariate analysis. In chemotherapy-treated ADC patients, lower levels of CD68-positive (P=0.010) and higher levels of FOXP3-positive cells correlated with worse OS (P=0.044). We developed a robust mIF panel of 10 markers to study inflammatory cells infiltrates in FFPE NSCLC tumor tissues. Chemotherapy-treated NSCLCs exhibited higher levels of PD-L1 expression and T cell subsets compared to chemo-naïve tumors, suggesting that chemotherapy activates specific immune response mechanisms in lung cancer. (Supported by CPRIT MIRA and UT Lung SPORE grants, and MD Anderson Moon Shot Program).