Overview of FFPE RNA-Seq

FFPE (formalin-fixed paraffin embedding) is one of the most commonly used methods of tissue preservation in clinical practice, which is mainly used in pathology, clinical diagnosis and scientific research. By collecting fresh samples fixed with formalin and embedded in paraffin for long-term preservation, this method can preserve the complete tissue structure, and tissue sections can help us to carry out histomorphological and immunohistochemical studies.

In addition, the FFPE preservation method can also retain RNA molecules or DNA molecules in tissues or cells. However, formalin may cause damage to RNA molecules and DNA molecules during sample fixation, and paraffin embedding uses high temperature infiltration to accelerate nucleic acid degradation. As a result, the quality of RNA and DNA in FFPE organization block is poor, and low-quality nucleic acid samples also pose challenges to RNA / DNA sequencing. However, with the development of technology, FFPE RNA-seq has become an effective method to obtain valuable transcriptome information from such samples.

What is FFPE RNA-seq

FFPE RNA-seq (Formalin-Fixed Paraffin-Embedded RNA Sequencing) is a research method based on RNA-Seq for extracting and analyzing RNA from FFPE tissue samples. FFPE RNA-seq allows researchers to obtain transcriptomic data from historically archived FFPE tissues such as pathological samples, clinical tissue specimens, and so on. FFPE tissues to obtain transcriptome data, which in turn reveals information about long-term genetic changes, gene mutations, transcript structure, and more.

Technique of FFPE RNA-seq

The first step in FFPE sample preservation is to fix the sample in formalin, but formalin fixation causes damage to the RNA molecules in three main ways:

(1) degradation of the RNA fractions (especially long-stranded RNA molecules), which usually results in shorter RNA fragments (100-500 nt) in the final FFPE organization block and impairs the integrity and availability of the RNA molecules.

(2) Formalin reacts with RNA and proteins by cross-linking, causing the RNA molecules to form covalent bonds with other molecules in the tissue, and the cross-linking makes it much less efficient for us to extract RNA from the sample.

(3) Chemical modifications such as methylation, uracil modification, etc. may be introduced, thus interfering with subsequent steps such as reverse transcription and amplification.

Therefore FFPE RNA-seq is different from general RNA-seq and requires special steps to ensure that usable RNA molecules are extracted.

The following are the main steps of FFPE RNA-seq:

Sample Preparation

• FFPE Tissue Sections: FFPE tissue is cut into 5-10 micron thick slices. For efficient RNA extraction, it is often necessary to prepare multiple slices to increase the amount of RNA extracted.
• Removal of paraffin: Paraffin in the outer surface layer interferes with RNA extraction, so we need to remove paraffin from the tissue using an organic solvent (e.g. xylene), followed by rehydration through a series of alcoholic solutions in order to restore the hydration of the tissue.
• RNA extraction: After paraffin removal, as RNA in FFPE organization block tends to be degraded and fragmented, a gentle extraction method is required to maximize RNA recovery, which can be performed using a specialized RNA extraction kit for FFPE organization blocks (e.g. Qiagen's RNeasy FFPE Kit).
• RNA quality assessment: Since RNA quality in FFPE organization blocks is poor and RNA is often short, RNA quality is assessed using equipment such as the Agilent Bioanalyzer or NanoDrop. If RNA from FFPE organization blocks shows a low RNA integrity (RIN) score, this indicates more severe degradation.

RNA Repair and De-Crosslinking

• De-cross-linking: formalin fixation will induce cross-linking of RNA with proteins, so specialized de-cross-linking reagents (e.g., Tris, EDTA, etc.) or heat are needed to release cross-links and break the chemical bonds between RNA and proteins after RNA extraction.
• RNA repair: RNA degradation in FFPE organization blocks is usually manifested as fragmentation, so before reverse transcription reaction, RNA repair enzymes (e.g., T4 RNA Ligase) are generally needed to repair RNA strand breaks and restore intact RNA fragments as much as possible.

Reverse Transcription and Library Construction

• Reverse transcription: Reverse transcription is a key step in FFPE RNA-Seq. Since the RNA in FFPE organization blocks is usually fragmented, reverse transcription is performed using either random primers or specific priming systems (e.g., small fragment primers). At the same time, the choice of reverse transcriptase is also very important, and usually an enzyme that can efficiently transcribe fragmented RNA is used.
• Library construction: The purpose of library construction is to fragment the reverse transcribed cDNA and add sequencing junctions to facilitate sequencing on a high-throughput sequencing platform. Since RNAs in FFPE organization blocks are usually short fragments, the short fragment library construction method is usually used for library construction.
• Junction ligation: Junction ligation of cDNA for subsequent PCR amplification.
• PCR amplification: Due to the small amount of RNA extracted from FFPE organization blocks, the PCR amplification step is particularly critical. To avoid amplification bias, low cycle count amplification is usually used.
• Library quality check: The quality of the library is assessed by techniques such as Bioanalyzer or qPCR to ensure that the library is suitable for high-throughput sequencing.

High-throughput Sequencing

• Sequencing platform selection: The constructed library can be sequenced using Illumina, PacBio, Nanopore and other high-throughput sequencing platforms, which are often used for FFPE RNA-Seq due to their high sensitivity and low cost.
• Data analysis: The analysis of FFPE RNA-Seq data includes gene expression quantification, differential expression analysis, splice variant analysis, mutation detection, etc. Special attention is paid to the fact that due to the poor quality of RNA in FFPE organization blocks, some quality control and filtering is required for data analysis to remove low-quality sequence data.

Processing FFPE samples for the TempO-Seq assay (Trejo CL et al.,2019)

For FFPE RNA-seq, there are generally two library construction methods: removal of rRNA library and exon capture method, the two library construction methods have their own advantages and disadvantages, we are introduced separately:

Removal of rRNA library Construction Method

Because rRNAs accounts for the majority of intracellular RNAs, and rRNAs are usually not involved in the regulation of gene expression, removing rRNAs can dramatically increase the proportion of mRNAs or other RNAs of interest for analysis. Especially true for FFPE samples, where the RNAs are usually already degraded and fragmented, efficient removal of rRNA is particularly important.

Principles of rRNA removal

• Hybridization-based depletion: by using specific probes for rRNA ( specific sequences for 16S and 18S ( in eukaryotic cells ) or other rRNA subunits ) to bind to rRNA molecules, it is removed from RNA samples.
• Capture-based depletion: using a specific capture probe (e.g., a DNA oligonucleotide probe) in combination with magnetic bead or column separation techniques and remove rRNA.

Specific methods for rRNA removal

For FFPE samples, the following methods are commonly used to remove rRNA:

• rRNA removal kits: commonly used rRNA removal kits such as Illumina's Ribo-Zero, NEBNext rRNA Depletion Kit, etc. These kits remove rRNA by designing specific probes.
• Removal of rRNA after reverse transcription: after RNA extraction, reverse transcription is performed using primers specific for rRNA. This process removes most of the non-mRNA portion, making the remaining RNA more inclined to mRNA or small RNA, thereby improving the efficiency of mRNA sequencing.
• Magnetic bead capture method: capture rRNA molecules by magnetic beads combined with rRNA-specific probes and remove them from the sample.

Advantages and challenges of rRNA removal

• Advantage: For FFPE organization blocks removing rRNA can significantly improve data quality and reduce the background noise of rRNA, thus improving the detection sensitivity of mRNA or other RNAs of interest.
• Challenges: RNA in FFPE organization blocks has been formalin-fixed, resulting in cross-linking and degradation of rRNA with other RNA molecules. The process of removing rRNA may result in some of the mRNA being mistakenly removed as well, or the overall recovery rate of the sample may be low, affecting the efficiency of library construction and the quality of the data.

The principle of exon capture method

The exon trapping method is based on the design of specific probes that bind only to the exon portion of the gene, thereby selectively enriching mRNA. This method usually includes the following steps:

• cDNA synthesis: First, the RNA extracted from FFPE samples was reverse transcribed into cDNA ( usually using random primers or specific primers ). Because the RNA of FFPE organization blocks is usually shorter, enzymes that can effectively reverse short fragments are often selected for reverse transcription.
• Exon probe design and capture: Design specific probes ( usually oligonucleotide sequences ) for exons of human or other species genomes and add captureable markers. The probe was hybridized with the cDNA exons in the sample and captured by magnetic beads or other methods.
• Elution and amplification: After capture, the unbound probes were removed by a series of washing steps. The captured exon region cDNA was eluted and subjected to PCR amplification to construct a high-quality library.
• Sequencing and data analysis: The cDNA of the captured exon region will be sequenced, and the generated data will be used for gene expression analysis, mutation detection ( such as SNP, insertion / deletion mutation ), splicing analysis, etc.

Advantages and challenges of exon capture method

• Advantages: By enriching specific exon regions, the capture method can significantly improve the coverage and sensitivity of mRNA, especially in FFPE organization blocks, RNA degradation is more serious, and exon capture can effectively make up for this deficiency. And exon capture only enriches the sequence of the gene coding region, so it has strong pertinence and cost-effectiveness, avoiding the interference of a large number of non-coding sequences in sequencing.
• Challenge: Exon capture requires high-quality exon probes, and RNA in FFPE organization blocks is often degraded, resulting in less enrichment than fresh samples. Moreover, due to the serious RNA fragmentation of FFPE samples, the capture probe may not fully bind to the target area, resulting in insufficient data coverage or low quality.

Applications of FFPE RNA-seq

FFPE RNA-seq technology is particularly suitable for obtaining gene expression data from historically stored clinical pathology samples, and has a wide range of applications, especially in the fields of oncology research, disease diagnosis and precision medicine. With the deepening of medical research, clinical pathology samples (especially FFPE) have become a valuable resource library. These samples are usually kept in hospitals and research institutes for many years, and have unique historical value. FFPE RNA-seq technology is able to successfully extract sufficient amount and quality of RNA from these historical samples through advanced RNA extraction and library construction techniques, and then perform high-throughput gene expression sequencing. This not only enables researchers to analyze old samples that have been abandoned, but also provides new data support for many studies that have lost their fresh sample sources.

Especially in the field of tumor research, the use of FFPE samples is of great significance. Tumor occurrence and development is often a long-term process, and historical tissue samples from patients can help researchers retrace the early changes and progression mechanisms of tumors, providing key molecular markers and gene expression information for early diagnosis, staging, prognostic assessment, and targeted therapy.

In addition, this technology can help counteract the drug resistance mechanisms of patient tumors during treatment and promote the development of precision medicine, thus enabling personalized treatment strategies. In the field of disease diagnosis and precision medicine, FFPE RNA-seq can not only provide in-depth gene-level understanding of complex diseases (e.g., cancer, genetic diseases, neurodegenerative diseases, etc.), but also provide strong data support for biomarker discovery and clinical translation.

• Cancer research:

The most common method used in cancer research is to use fresh tissue for genome-wide or transcriptome identification. These fresh tissue samples provide high-quality RNA and DNA samples, which can be used to analyze the expression profiles of tumor molecules at a certain stage or period of time in the fresh tissues, however, for most of the cancers, only a few of them can provide enough fresh tissues samples to obtain high-quality RNA samples and accurate in vitro RNA expression.

However, for most cancers, only a small percentage of cancers can provide enough fresh tissue samples to obtain high-quality RNA samples and accurate in vitro RNA expression, and for cancers that develop over time, fresh tissues are not available, and long-term outcome data are not easy to obtain, making FFPE RNA-seq especially important.

In breast cancer, ER+ breast cancer patients have been found to have a survival rate of more than 95% at 5 years of disease, but metastatic trends have been found in approximately 20%-40% of patients with 10-20 years of disease. To investigate the long-term nature of this cancer in relation to time, the researchers selected breast cancer tissue samples preserved in FFPE for 2-23 years (cervical TRB-approved breast specimens from non-menopausal women aged 20-45 yrs.) for RNA-seq. It was tested that one or two 10 μm slices of the breast FFPE specimen were sufficient to extract a sufficient amount of RNA sample for analysis. Comparison of ER+ and ER- data from fresh tissue with FFPE preserved samples resulted in the identification of more than 14,000 DEGs, and eight regulator networks differentially expressed in ER+ and ER- were identified by regulator analysis. There were two types of regulator networks: those defined by KDM4B associated with transcripts, and in particular KDM4B was strongly associated with ER+ breast cancer, but further evaluation of the association between KDM4B and ER+ breast cancer survival is needed. In conclusion FFPE RNA-seq provides long-term outcome data in breast cancer and helps us to develop new tumor markers(Pennock ND et al.,2019).

CRT is the standard treatment for SCCA, but 10%-26% of patients do not respond to CRT treatment or even after undergoing CRT treatment, about 25% of patients will experience recurrence. Moreover, no effective biomarkers have been detected in SCCA patients after receiving CRT. In order to characterize the response that occurs in SCCA patients in vivo and the therapeutic efficacy of CRT, the researchers chose to perform RNA-seq on the FFPE organization blocks of 9 non-relapsed and 3 relapsed patients to identify the differences in their transcripts. 449 DEGs, including mRNA, IncRNA, snRNA, etc., were identified in the two transcripts, and their PCA and GO analyses revealed that the core of up-regulated genes in the non-relapsed tissues were CD40LG, IL4, ICAM2,HLA-I, HLA-II, etc., and that the pathway in which they were collectively involved was the xenotransplantation rejection reaction, and that they were able to inhibit tumor The gene miR-4316, which inhibits tumor growth and proliferation, was significantly up-regulated in non-recurrent tissues, whereas the gene lncRNA-SOX21-AS1, which is associated with cancer development, was more significantly expressed in recurrent tissues. This suggests that identification of FFPE organization blocks from tumor patients by RNA-seq has found genes associated with SCCA recurrence and non-recurrence, which can provide effective information for treatment(Ye Y et al.,2023).

• Fusion Gene Detection:

Fusion gene is one of the most important causes of cancer, and the detection of transcripts of fusion genes is of great significance for the diagnosis and treatment of cancer. RNA-seq is recognized as a good method for extracting transcripts, but a large portion of the RNA-seq results are based on fresh tissues, which are not easy to carry around, are inconvenient, and have a low accessibility. So FFPE RNA-seq was developed to extract transcript data based on FFPE-preserved clinical tissue samples, and it was proven that FFPE RNA-seq can obtain high-quality and effective gene expression and fusion transcript assays.

The researchers selected FFPE organization blocks from the ethnic cohorts of two research teams: 136 patients with breast cancer and 76 patients with breast cancer (patients aged 8.5-13.4 years), and out of these 212 different samples, the researchers identified 118 fusion transcripts by FFPE RNA-seq (based on single-ended 50bp reads from gFuse), 100 unique fusion junction events, and approximately 61% of FFPE RNA-seq was accurate as analyzed by the qRT-PCR assay TaqMan. Application of gFuse demonstrated that single-ended RNA-seq data from aged FFPE tumor tissues also had the same fusion transcript events described above. Three fusion linkage events were also found to be present in all 212 patients. These fusion genes were associated with breast cancer prognosis and also showed that FFPE RNA-seq can detect fusion transcripts thus helping us to develop new biomarkers by studying samples of tissues from long-term clinical miles(Ma Y et al.,2014).

There is a consensus that fusion genes contribute to hematologic diseases and solid tumors, and the need to detect fusion genes is an essential part of clinical practice. The technology for detecting fusion genes has been evolving, and currently, more than 33,000 different fusion events involving 14,000 unique genes have been identified in cancer. The most common method of tissue preservation in the clinic is the FFPE method, and RNA extraction from FFPE organization blocks is indispensable for the identification of fusion genes. The researchers used mRNA capture sequencing to identify fusion genes in FFPE organization blocks, extracting FFPE biomaterial from six patients for fusion transcript analysis. It was shown that mRNA capture sequencing could identify to all demonstrated chromosomal rearrangement events. Application of RNA exome sequencing to identify FFPE tumor samples from 17 patients with ARMS and URCS revealed gene fusion events not identified by FISH analysis.RT-qPCR results confirmed the confidence of the FFPE RNA-seq data. Detection of this fusion event in clinicopathology indicates the presence of a clinicopathologic mystery sarcoma. This suggests that FFPE RNA-seq can not only identify known fusion genes but also detect specific gene fusion events not detected by other methods(Decock A et al.,2022).

• Expression profiling:

Whether it is healthy or diseased tissues, obtaining their transcriptome data is crucial for us to study diseases, and a good method of tissue preservation is FFPE preservation. However, FFPE preservation can damage the ribonucleic acid in the tissues, so it is especially important to analyze their transcriptomes to see if the transcriptomic data of the tissues can be accurately mapped.

The researchers selected human (FFPE-preserved clinical samples from the indicated dates through 2016), mouse (wild-type adult mouse samples of the C57BL/6 genetic strain preserved for 24 h by FFPE), and rat (obtained from the University of Arizona Cancer Center) for TempO-Seq, and performed transcriptome analyses on a wide variety of tissues (human prostate, pancreatic, and colorectal carcinomas, mouse breasts, hindlimb muscles, non rat liver, brain, and kidney FFPE organization blocks) and found that FFPE TempO-Seq data results had reproducible and high quality attributes. And the preservation time did not affect the RNA-seq results of FFPE organization blocks, and the results of both long-term preserved samples and short-term preserved samples had highly accurate and reproducible RNA expression levels. It was also found that the FFPE preservation method was equivalent to the results of fresh tissues, i.e., the expression profile data of FFPE organization blocks were comparable to those of fresh tissues, which proved that FFPE samples could obtain highly accurate RNA-seq data(Trejo CL et al.,2019).

The pathogenesis of idiopathic pulmonary fibrosis (IPF) is unclear, but the mortality rate is high and increases with age. Much of the research on IPF has utilized fresh frozen tissue, but this has drawbacks such as being available only in places with highly sophisticated instrumentation or not being able to map the histology in its entirety. Therefore, the researchers used RNA-seq to study FFPE-preserved IPF tissue to investigate whether this method is feasible. The researchers selected FFPE organization blocks (preserved for an average of 6 years) from 7 IPF patients and 5 controls for RNA-seq (50 bp end sequencing), identified multiple DEGs, and compared the expression profiling data obtained from FFPE organization blocks with that from fresh frozen tissue (which was clearly representative of IPF transcripts), demonstrating that the FFPE RNA-seq method is feasible and the results are comparable to the microarray results of FF tissues.The FFPE RNA-seq data identified DEGs associated with IPF and there were also some discrepancies in the results of FF tissues, but after NanoString nCounter analysis, the genes that were not the same in the two datasets were probably due to tissue heterogeneity, but this also indicates that FFPE RNA-seq can be used as a research method for resolving IPF(Vukmirovic M et al.,2017).

• Development of a biomarker:

BCa is currently the most common type of cancer in the urinary tract of both men and women.EVs can transport cargo to specific target cells and thus play a communication role, thus it is involved in many physiological and pathological processes.The presence of EVs can be detected in almost all types of biological fluids, thus it can be used as a biomarker in liquid biopsies. However, the EVs found in these fluids could not be identified as belonging specifically to which specific tissue such as blood, urine, etc. In order to identify EVs specific to blood and urine, the researchers performed Total RNA-seq using 16 FFPE tumor tissues and more than 20 samples each of exosomes, blood, and urine from patients with BCa, with the RNA from FFPE tumor tissues of higher quality. Molecules from blood versus urine were categorized according to the corresponding FFPE tumor tissue analysis, and by comparing the differential transcripts, it was found that KRT6A isoforms EVs were more abundant in fluid sEVs from FFPE tissue-based Ba/Sq isoforms patients' urine. Comparative analysis of differential transcripts in FFPE tumor tissue samples, urine sEVs, and plasma sEVs found 10 DEGs in different types from NMIBC and MIBC. and comparing sEVs from the three sources mentioned above with sEVs from adjacent sources of tumor tissue, four BCa-specific mRNA biomarkers were found: OR4K5, FAM138F, FAM71E2, and KRTAP26-1, and the researchers demonstrated that sEVs obtained from tissues are more responsive to a wider range of tissue- or disease-specific biological features than sEVs obtained in vivo, and that urinary sEVs are the sEVs most likely to yield specific biomarkers in vivo(Dong L et al.,2024).

Specific RNA-seq techniques and their uses can be found in the following pages: "Overview of RNA Sequencing Techniques".

References:

1. Pennock ND, Jindal S, Horton W, Sun D, Narasimhan J, Carbone L, Fei SS, Searles R, Harrington CA, Burchard J, Weinmann S, Schedin P, Xia Z."RNA-seq from archival FFPE breast cancer samples: molecular pathway fidelity and novel discovery."BMC Med Genomics. 2019;12(1):195.doi: 10.1186/s12920-019-0643-z
2. Ye Y, Maroney KJ, Wiener HW, Mamaeva OA, Junkins AD, Burkholder GA, Sudenga SL, Khushman M, Al Diffalha S, Bansal A, Shrestha S."RNA-seq analysis identifies transcriptomic profiles associated with anal cancer recurrence among people living with HIV. "Ann Med. 2023;55(1):2199366. doi:10.1080/07853890.2023.2199366
3. Ma Y, Ambannavar R, Stephans J, Jeong J, Dei Rossi A, Liu ML, Friedman AJ, Londry JJ, Abramson R, Beasley EM, Baker J, Levy S, Qu K."Fusion transcript discovery in formalin-fixed paraffin-embedded human breast cancer tissues reveals a link to tumor progression." PLoS One. 2014 Apr 11;9(4):e94202. doi:10.1371
4. Decock A, Creytens D, Lefever S, Van der Meulen J, Anckaert J, De Ganck A, Deleu J, De Wilde B, Fierro C, Kuersten S, Luypaert M, Rottiers I, Schroth GP, Steyaert S, Vanderheyden K, Vanden Eynde E, Verniers K, Verreth J, Van Dorpe J, Vandesompele J."mRNA Capture Sequencing and RT-qPCR for the Detection of Pathognomonic, Novel, and Secondary Fusion Transcripts in FFPE Tissue: A Sarcoma Showcase. "Int J Mol Sci. 2022;23(19):11007. doi:10.3390/ijms231911007
5. Trejo CL, Babić M, Imler E, Gonzalez M, Bibikov SI, Shepard PJ, VanSteenhouse HC, Yeakley JM, Seligmann BE."Extraction-free whole transcriptome gene expression analysis of FFPE sections and histology-directed subareas of tissue." PLoS One. 2019;14(2):e0212031. doi: 10.1371/journal.pone.0212031
6. Vukmirovic M, Herazo-Maya JD, Blackmon J, Skodric-Trifunovic V, Jovanovic D, Pavlovic S, Stojsic J, Zeljkovic V, Yan X, Homer R, Stefanovic B, Kaminski N."Identification and validation of differentially expressed transcripts by RNA-sequencing of formalin-fixed, paraffin-embedded (FFPE) lung tissue from patients with Idiopathic Pulmonary Fibrosis." BMC Pulm Med. 2017;17(1):15. doi: 10.1186/s12890-016-0356-4
7. Dong L, Feng M, Kuczler MD, Horie K, Kim CJ, Ma Z, Lombardo K, Lyons H, Amend SR, Kates M, Bivalacqua TJ, McConkey D, Xue W, Choi W, Pienta KJ. "Tumour tissue-derived small extracellular vesicles reflect molecular subtypes of bladder cancer." J Extracell Vesicles. 2024 ;13(2):e12402. doi:10.1002/jev2.12402