Application potential of spatial transcriptome technology in tumor immunotherapy

Heterogeneity in tumor brings great challenges to the accurate diagnosis of cancer patients and the formulation of individualized treatment strategies. In addition, this heterogeneity may be the basis of treatment of drug resistance, disease progression and cancer recurrence. Although immunotherapy can achieve a high success rate, selective pressure and dynamic evolution within the tumor promote the emergence of drug-resistant clones, making the tumor persist in some patients. In order to improve the efficacy of immunotherapy, researchers use spatial transcriptome technology to identify and block the source of tumor heterogeneity.

In situ hybridization (ISH) is a molecular technique to visualize specific DNA or RNA molecules in cells or tissues. ISH hybridizes the labeled nucleic acid probe to the target in situ based on the complementarity of DNA/DNA or DNA/RNA double strands. In this way, we can get useful spatial information.

FISH is an effective clinical tool for detecting microorganisms, diagnosing solid tumors and hematological tumors, and guiding cancer treatment. For example, FISH is commonly used to detect BCR-ABL1t (9; 22) Many fusion genes in translocation and various cancers. FISH is also used to confirm the amplification of HER2 gene in breast cancer, so as to determine the patients who are most likely to benefit from the anti-HER2 monoclonal antibody trastuzumab. Another important example is the detection of EML4-ALK fusion gene in non-small cell lung cancer. As more and more immunotherapy is developed and approved, researchers try to use FISH to predict the responsiveness of cancer immunotherapy. In order to expand the effectiveness of FISH, we can combine FISH with IHC or IF, and simultaneously detect RNA and protein of different cell types, so as to better characterize tumor microenvironment (TME).

In order to solve the limitation of traditional FISH, researchers turned from studying DNA to studying single-molecule RNA, and adopted Qualcomm's method, thus producing smFISH technology, which can visualize and quantify single mRNA molecules and characterize the spatial pattern of endogenous gene expression. SmFISH has become a powerful tool to evaluate the transcriptional heterogeneity of tumors by targeting cellular mRNA instead of DNA molecules.

RNAscope is a commercial technology based on ISH, which can detect as many as 12 different RNA targets, and can be easily combined with IHC and/or IF to study RNA and protein simultaneously in an automated way. Compared with other fish-based technologies, RNAscope has designed more than 13000 RNA probes, which have been verified by commercial processes. Therefore, it is a time-saving and friendly method for basic research and clinical experiments. RNAscope has been widely used in various disciplines, including infectious diseases, cancer, immunotherapy, inflammation and neuroscience. In particular, it is a powerful substitute for IHC, and can evaluate the expression of immune checkpoints in various solid tumors, such as PD-L 1. By detecting specific RNA, RNAscope clarified TME, immune escape mechanism and new biomarkers for predicting and predicting cancer.

Under the background of immunotherapy, RNAscope plays a valuable role in understanding CAR-T cell therapy. RNAscope has been used to evaluate the specificity of target gene expression and track the distribution of CAR-T cells in xenotransplantation mouse models. Extending to human samples, studies have confirmed that the expression of BCMA is the target of CAR-T cell immunotherapy for multiple myeloma.

Although RNAscope and other techniques can achieve higher sensitivity and specificity, fFISHh-based techniques are needed to allow Qualcomm transcriptome analysis to better characterize rare cell populations and cell types showing unique gene expression profiles. MERFISH and seqFISH not only provide improved RNA quantification, signal amplification and detection, but also provide image-based transcriptome analysis.

MERFISH is improved from smFISH, which adopts the combined labeling method based on bar code, and then carries out multiple rounds of hybridization to ensure the high brightness of fluorescence signals and a large number of RNA that can be detected at one time.

SeqFISH is another multiplex smFISH technology based on a series of bar code hybridization. For example, seqFISH is used to detect mouse embryonic stem cells and brain tissues >: Imaging with 10000 mRNA has high accuracy and resolution. Related studies have proved that seqFISH is a powerful tool to study and obtain the dynamic expression of regulatory genes during T cell maturation. Another study combined microfluidic technology with multiplication Smfish technology to study tumor heterogeneity in breast cancer, which proved that multiplication Smfish can be further optimized from different angles.

Although smFISH technology has broad prospects, due to the complexity of probe design, verification, image analysis and decoding, the composite technology based on smFISH has not been widely used in transformation research or clinical application. It is usually more convenient to study the expression of a single gene at mRNA or protein level by using non-multiplex FISH, quantitative PCR, IHC and IF, especially when the number of genes studied is small, such as a group of prognostic markers. Another limitation is that due to the nature of sequence hybridization, the total imaging time is at least 18 hours, excluding the initial probe hybridization time of 36 ~ 48 hours, so the overall throughput is lower than other technologies (such as DSP and Visium). Besides multiplexing? SmFISH technology can only evaluate one analyte in fresh frozen tissue, such as RNA. Emerging technologies such as DSP can evaluate protein and RNA in fresh frozen tissues and standard formalin-fixed paraffin-embedded (FFPE) tissues commonly used in pathology.

DSP is a high-complexity spatial analysis method, which overcomes the main limitations of multi-channel smFISH technology. DSP uses oligonucleotide detection technology to quantify protein or RNA in FFPE tissue samples.

Different from sequence hybridization technology (such as MERFISH), DSP provides a more efficient workflow, which can produce results from 10~20 tissue sections or up to 384 target areas within 48 hours. In addition, compared with multi-channel smFISH which only analyzes RNA, DSP can simultaneously detect 96 kinds of protein or 1400 mRNA. This feature is particularly relevant to cancer immunotherapy, because the difference of mRNA and protein expression patterns can be used to clarify post-transcriptional regulation and post-translational modification, which leads to protein instability and affects prognosis and treatment response. At the same time, DSP also retains the integrity of tissue samples and can store precious samples for further analysis in the future.

DSP is widely used in the field of immunotherapy. For example, DSP has been used to evaluate the immune microenvironment of patients with diffuse large B-cell lymphoma receiving chemotherapy. DSP also has research on immune checkpoint blocking therapy, including anti-PD-L 1 and anti-PD- 1 therapy. DSP can be used as an auxiliary diagnostic tool to standardize, quantitatively and objectively evaluate the expression of PD-L 1 protein in the spatially limited region of TME. In another study, DSP successfully identified more than 20 biomarkers, which can predict the response of melanoma patients to immunotherapy.

In the process of single-cell RNA sequencing, because tissues are usually homogenized to obtain the average profile of transcriptome, spatial information is lost. Recently, the technology of spatial transcriptome was developed, and the quantitative visualization and analysis of transcriptome in complete tissue sections were realized by using spatial bar code oligodeoxythymine microarray.

This new technique was first proved on mouse olfactory bulb, and followed the following standard workflow: tissue sectioning, fixation, hematoxylin and eosin (H&; E) dyeing, bright field imaging, tissue infiltration, cDNA synthesis, tissue excision, probe release, library preparation, sequencing, data processing, data visualization and analysis.

The data analysis of breast cancer, prostate cancer and skin malignant melanoma by st shows that the heterogeneity within and between tumors has reached an unprecedented level. Through RNA sequencing analysis and/or standard morphological annotation, there are obvious differences in gene expression profiles between tumor areas and periphery. In addition, in vivo experiments using this technique found that IL-6 signal was induced by the regeneration of microglia, which may be valuable in treatment.

In order to take advantage of the potential of ST, researchers recently developed an analysis method called MIA, which integrates the data set generated by single-cell RNA sequencing and ST technology to locate cells in specific areas of tissue. As a proof of concept, MIA was carried out on the data set of pancreatic ductal adenocarcinoma, and revealed the enrichment of specific cell types and subsets in spatially restricted areas, which were previously unknown or undetectable.

Based on the concept of spatial transcription, 10× Genomics released a solution of Visum spatial gene expression. Compared with the first iteration of st technology, it has higher resolution and sensitivity. It can be used for in-depth study of diseases related to tissue structure and function, not only for cancer immunotherapy, but also for nervous system diseases.

Although the technology of transcription space analysis is relatively new, it has been widely explored in tumor immunotherapy. FISH and RNAscope are effective clinical tools for diagnosis and prediction of solid tumors and hematological tumors. Newer technologies, such as MERFISH and Visium, have achieved batch transcriptome analysis with unprecedented resolution and sensitivity. With the popularization of this technology, new biomarkers can be found to predict the response of immunotherapy, and personalized treatment can be carried out according to the heterogeneity of its unique TME. These spatial analysis techniques can also be combined with dimensionality reduction techniques. For example, UMAP is used to visualize the immune landscape of TME.

Looking ahead, DSP provides spatial analysis and digital characterization of mRNA expression, but it is still limited by the number of gene targets that can be studied at the same time. Although Visium is relatively new in the market, it has been continuously improved in a short time and has great potential in disease pathology research and clinical transformation.

Researchers can make use of various developing spatial transcriptome technologies. It is important to consider not only technical characteristics, including spatial resolution, sensitivity, specificity and organization type, but also practical factors, such as cost, compatibility with existing resources and turnaround time. Researchers must carefully consider their research problems and choose appropriate technologies closely related to their research and clinical goals.

First WeChat official account: National Gene Bank Big Data Platform

refer to

Zhang Chunlin, et al. Application of gene expression profile analysis in tumor immunotherapy [J]. China Cancer Science Press, 2002. Cancer, 2020, 12(9): 2572.