BACKGROUND: Immune checkpoint inhibitors (ICIs) have revolutionized the cancer therapy landscape in recent years. Despite the success of the checkpoint blockade strategies targeting cytotoxic T lymphocyte antigen 4 (CTLA-4) and programmed death receptor 1 (PD-1), many cancer patients cannot benefit from these therapies. T cell immunoglobulin and mucin domain 3 (TIM-3) has emerged as one of the next generation ICI targets, with potentially lower toxicity and higher safety compared to CTLA-4 and PD-1 blockades. TIM-3 is widely expressed in different immune lineages playing various roles such as mediating immune tolerance and regulating innate immune response. However, the mechanism of action of TIM-3 inhibition in different malignancies is not completely understood. Whether the expression and genomic status of TIM-3 and its ligands, Ceacam-1, galectin-9, HMGB1 and phosphatidyl serine (PtdSer) are associated with clinical outcome or any indication would be essential for patient selection for TIM-3 targeting therapies. METHODS: Human PD-1/TIM-3 double-knock-in mice (PD-1/TIM-3 dKI HuGEMM™) engrafted with CT26.WT tumor model were used to test human PD-1 antibody (Keytruda) and TIM-3 antibody (MBG453). Immune phenotyping of blood, tumor draining lymph node (TDLN) and spleen tissues were assessed by flow cytometry at different time points after dosing. Cytokine levels in serum were measured by MSD assays at 48 hours post the 3rd dose and at study termination. Patient genomic and clinical data for various cancer types such as colorectal adenocarcinoma and pancreatic adenocarcinoma were collected for prognostic biomarker analysis. RESULTS: In the in vivo efficacy study of single and combination treatment with Keytruda and MBG453, we observed that NK cells were induced by anti-TIM-3 treatment, alone and in combination, in both blood and TDLN. This indicates TIM-3 blockade may lead to NK cell proliferation in the TME to enhance tumor killing. Transcriptomic analysis on thousands of patients from TCGA showed that high expression of TIM-3 was highly associated with MSI-H and MSI/CIMP subtypes of colorectal adenocarcinoma, suggesting the potential of Tim-3 target therapy in combination with PD-1 blockade in colorectal cancers. By survival analysis, we observe that one of the TIM-3 ligands, HMGB1, expression is associated with patient OS and PFS with pancreatic adenocarcinoma, but not in colorectal cancers. Furthermore, TIM-3 expression was associated with many immune cell signatures, including macrophages, dendritic cells, CD8+ memory T cells, CD4+ memory T cells and Tregs in both colorectal and pancreatic adenocarcinoma. CONCLUSIONS: Evaluation in preclinical model demonstrated that TIM-3 blockade may cause NK cell proliferation to enhance anti-tumor immunity. In addition, the expression and genomic alteration of TIM-3 and its ligand have prognostic values for certain cancers. Citation Format: Jia Xue, Yu Zhang, Xianfei He, Henry Q. Li, Sheng Guo. Tim-3 as an immune therapy target, mechanism and action, and prognostic values with its ligands in patient stratification [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 615.

Cancers are collections of diverse diseases of genetic and immunological abnormalities. The heterogeneous tumor microenvironment (TME), including immune components, and their interactions with tumor cells play critical roles in tumor progression and response to pharmaceutics, particularly immuno-oncology (I/O) therapy. However, investigating TME-specific components is rather challenging for the difficulty to separate stroma from tumor cells, either physically via microdissection or in silico via bioinformatics. Patient derived xenograft (PDX) may be a new system to investigate TME1, where human and mouse content can readily be separated in silico2. We have transcriptome-sequenced ~1600 bulk tumor tissues from subcutaneous PDXs grown in athymic mice3. By aligning reads to human and mouse genomes, we found that the average mouse-to-human sequencing read ratio is around 11% (5~20%), consistent with the previous report2. After removal of the low-expressed and less-variable genes and by deconvolution analysis of gene expression data, we identified all types of TME components, including adaptive and innate immune cells. The corresponding fractions vary across cancer types and individual models. Co-regulation analysis identified a huge number of intra-species interactions and also, a smaller number of inter-species interactions that vary greatly among different cancer types. The cross-species interactions observed are likely implicated in the growth of these tumors, and their numbers may also reveal the degree of the dependence of tumor growth on TME, which should be reversely correlated to the transplantation take-rate of corresponding type of PDX. Indeed, we have demonstrated this reverse-correlations (# interactions: take-rate %) with statistical-significance (p-value = 0.034) across our PDX collections, including melanoma (406:27%), lung (146:50%), colorectal (CRC) (157:68%) and pancreatic cancer (32:80%). The cancer type with the highest take-rate and lowest # interactions is pancreatic cancer that also has highest KRAS mutation rate (>90%), hinting the role of KRAS mutation in tumor growth independency on TME. This is further confirmed in KRAS mutant CRC (1:96%) vs. wild-type (98:53%). Moreover, some putative cross-species co-regulations in specific cancers were also observed in human tumors (e.g. in TCGA dataset), indicating potential importance in TME-tumor interaction and tumor development in human. Further investigation of each of these interactions may reveal novel TME-related disease pathways and thus novel targeting strategy for cancer therapy. In conclusion, transcriptomic analysis of large number of bulk PDXs provides a novel and unique platform to study TME, likely to facilitate new discovery of disease pathways and strategy to treat cancers involving the TME mechanism, particularly I/O strategy. Citation Format: Jia Xue, Wubin Qian, Sheng Guo, Xiaoyu An, Xuesong Ouyang, Henry Q. Li. Transcriptomic analysis of bulk tissues of large PDX collection as a novel platform discovering new TME target/drug [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 1016.

RNA-Seq is currently the most prevailing method for measuring transcriptional activities in cells and tissues. It relies on high-quality RNA in order to yield reliable and reproducible results, which is often challenging due to RNA degradation during sample collection and processing. Agilent’s RNA Integrity Number (RIN) is a commonly adopted standard for evaluating RNA quality in NGS workflows. However, while most RNA-Seq experiments are geared towards the quantification of mRNA, the RIN metric heavily relies on the amount of 18S and 28S ribosomal RNA and does not directly measure the integrity of mRNA. To overcome this limitation, researchers have proposed several post-alignment measures of transcript integrity, such as mRIN (mRNA Integrity Number), TIN (Transcript Integrity Number, from RSeQC package), and DI (Degradation Index, from DegNorm package), but so far there is no consensus as to which works best. It is also unclear to what extent RNA degradation impacts the results of downstream analysis when samples with suboptimal RNA quality are included. To answer these questions in the context of cancer research, we analyzed 198 RNA-Seq samples from 7 syngeneic mouse tumor models of different cancer types (4T1, CT26, EL4, E.G7-OVA, H22, Hepa1-6, and KLN205) with a wide range of RIN values (2.3 to 9.8). Interestingly, we found a high concordance between RIN and medTIN (median TIN score) in 4T1, E.G7-OVA, Hepa1-6, and KLN205 samples, but a surprisingly low concordance in EL4 and H22 samples. A tentative interpretation is that, depending on the tissue/cell type, it is possible for an RNA sample to have heavily degraded ribosomal RNA (hence a low RIN value), while still retaining relatively intact mRNA (resulting in a decent medTIN value). Principal component analysis (PCA) revealed that both RIN and medTIN are strongly correlated with the strongest explanatory variable (PC1) of the transcriptome across all library types, confirming that RNA degradation can heavily bias the results of transcript-level DE (differential expression) analysis. We then evaluated the performance of TIN correction, a method proposed in conjunction with the TIN metric to correct for RNA-degradation bias, and found it to be largely ineffective on our dataset. Apparently, there is a need for better normalization/correction methods when a dataset consists of samples with wildly varying RNA quality. On the other hand, the impact of RNA degradation on gene-level DE analysis is much smaller. In PCA on gene-level data, the first 4 PCs, which altogether explain > 70% of the variation in the transcriptome, showed weak or no correlation with RIN or medTIN, and the samples cluster according to their respective tissue groups as expected. Our study calls for cautious analysis and interpretation of gene expression data from degraded RNA samples, and highlights a need for more suitable RNA quality metrics and bias correction methods. Citation Format: Yanghui Sheng, Wubin Qian, Xiaobo Chen, Henry Q. Li, Sheng Guo. Impact of RNA degradation on transcriptomic profiling in tumors samples from syngeneic mouse models [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 3371.

Background: Immunotherapies have made a significant contribution in the treatment of cancer; however, the interaction between tumor, treatment and immune response remains complex. To fully understand the interactions between cancer and immunity in a tumor microenvironment (TME), experimental murine models, derived from a variety of biological technologies, are essential preclinical systems utilized before clinical studies. Next-generation sequencing (NGS) technologies have been demonstrated to be superior to microarray-based methods in assessing molecular features on omics levels. Here we describe a robust and efficient NGS-based gene expression panel to characterize murine tumor-immune interactions. Methods: We developed a murine gene expression panel (NGSmIO) with ~1100 immune-associated marker genes and immuno-oncology (I/O) signaling genes. Oligonucleotide-based hybridization/capture techniques analogous to exome sequencing, and targeting specific regions, are well established on Illumina NovaSeq and BGI MGISEQ platforms. These methods were applied to the murine I/O panel assay, with accuracy and reproducibility verified across these two sequencing platforms. Tumor, blood and spleen tissues derived from MC38 and Hepa 1-6 tumor models with anti-PD-1 and anti-CD4 treatment were collected for both our NGS panel assay and the NanoString PanCancer Mouse IO 360™ Panel (NanoStringIO). The results were compared in parallel with historical cell cytometry, RNA-Seq (transcriptomics) and proteomics data in the same experimental conditions. Results: By systematic comparison of NGSmIO and NanoStringIO panels, we observed high correlation (mean R=0.856) and high consistency in common gene signature expression across 12 paired samples derived from different tissues and therapeutic conditions. To determine the performance of both panels, we used FACS, RNA-Seq and proteomics data in the same therapeutic conditions as previous studies for reference, and demonstrated that NanoStringIO systematically over-detects low expressed genes in different samples/conditions. Furthermore, NGSmIO detected gene expression changes under treatment, whereas NanoStringIO was unable to differentiate expression between control and treatment. Conclusions: We have established a murine-I/O NGS panel to efficiently characterize tumor-immune interactions in a robust manner for preclinical studies. Compared to the NanoStringIO panel, NGSmIO provides enhanced insight into the TME, and has the potential to both translate preclinical studies more effectively to the clinic as well as guide the design of immunotherapy companion diagnostics. Citation Format: Jia Xue, Xiaobo Chen, Henry Q. Li, Sheng Guo. NGS-based murine pan-cancer gene expression panel reveals high resolution for immuno-oncology and tumor microenvironment studies [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 2707.

Background: Genomic Structural variation (SV) refer to abnormalities in chromosome structure, is a normal part of variation in the human genome. Now Genomics SVs are recognized as the largest source of interindividual genetic variation and are closely associated with oncogenesis. The ability to identify constitutive and low-allelic fraction SVs is crucial. Standard SV detection method include chromosome banding, fluorescence in situ hybridization (FISH), and array comparative genome hybridization (CGH), which are manually intensive and have trouble finding low-frequency mutations. In recent year, genome physical mapping technologies have received increasing attention and optical mapping-based methods can well-accomplish sufficient to detect large size SVs or SVs within repetitive regions. Lately, Bionano optical mapping technology rapidly expanded its applications in the detection of structural variations. As well known, Acute Leukemia show distinct patterns of genetic aberrations including chromosome translocations, mutations, and aneuploidies in genes responsible for cell cycle regulation and lymphoid cell development. Method: We use our liver cancer model LI6671 to establish the Bionano Saphyr Gen2 genome imaging platform. DNA >100kbp is extracted, “CTTAAG” sequences were labelled across the entire genome by using Direct-Label-Enzyme (DLE) and linearized through chips for visualizations. Saphyr was loaded, linearized, and imaged labeled DNA in repeated cycles. Assembly Algorithms covert collected DNA images into constructing consensus genome maps. And this assay platform will be used to comprehensively identify SVs for studying our 21 HuKemia models. Results: By using this fluorescence labels distribution patterns, we find different structure variants among genome. We used filtered data for genome map assembly and compared with human genome hg38 for SV detection. In LI6671, we found out 3952 SVs in LI667 including 1166 deletion, 2497 insertion, 101 duplication, 78 inversion breakpoints, 66 interchr. translocation breakpoints and intrachr. translocation. Oncogenesis related SVs, such as a ~13Kb heterozygous deletion at KCNQ1 region, EPC1::SP1 gene fusions have be founded. Conclusion: Saphyr is a comprehensive platform, can discover a broad range of Genomic SVs and further improves our understanding of Acute Lymphocytic Leukemia (ALL) and Acute Myelocytic Leukemia (AML). Citation Format: Xiaobo Chen, Huan Tian, Wubin Qian, Henry Q. Li, Sheng Guo. Detection of structural variation and analysis on HuKemia models by Using Bionano optical mapping [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 2234.