Reliable measurement of transcriptomics in FFPE material

Extracting gene-expression profiles from Biobanks

Biobanks harbor a wealth of gene-expression data of patients with a certain disease. The effects of treatment are documented and can be followed over time on molecular level. These biobanks generally consist of Formalin-Fixed Paraffin-Embedded (FFPE) tissue. High degree of RNA degradation of such FFPE cohorts is often hampering analyses by, for instance, leading to exclusion of key samples and/or essential time points. Factors that affect RNA quality are e.g. age of the blocks, treatment of the material before fixation, storage conditions. Also, formalin fixation damages RNA, causing extensive shredding and modification of nucleotides.

GenomeScan’s solution for FFPE RNA analysis

Our scientists have thoroughly studied the effects of RNA degradation and RNA yield on the accuracy and reliability of the transcriptomics data.
In validation studies performed with the library preparation chemistry of New England Biolabs (NEB) combined with Illumina sequencing, we have determined the parameters that are best suitable for monitoring the validity of the data. The figure, below on this page, shows that the mapping of FFPE material remains consistent, even with less input. Thus, gene-expression analysis of FFPE RNA is robust and reproducible in the range of 5 to 200 ng, for this cohort.

Our service includes:

• Experimental design optimization
• Dedicated Project Manager to give special attention to your project
• Contact points throughout the entire project

• Optimal RNA input: 100 ng / sample
• RNA quantity and quality check
• Library QC
• ISO/IEC 17025 accredited service

• rRNA depletion
• Stranded RNA library preparation
• Sequencing NovaSeq PE150

• FASTQ files (raw data)
• Count file table (expression levels (FPKM values) per gene for individual sample)
• Extended data-analysis

• Isolation of RNA from FFPE blocks
• Tumor microdissection by Pathologists from the Leiden University Medical Center (LUMC)

Getting the best transcriptomics results out of your FFPE cohort requires expertise. GenomeScan offers you a comprehensive solution for FFPE RNA analyis.

Gene expression analysis of FFPE RNA is robust and reproducible:

Gene expression analysis of FFPE RNA is robust and reproducible

Fragment Analyzer measurement of RNA in submitted samples. From left to right: good quality, intermediate and poor quality RNA (resp. RIN 8.8, 6.6, 2.3). Ribosomal RNA’s are clearly visible in samples with a high integrity score. Fixation damages RNA, resulting in degradation of the 16S and 18S peaks. In fixated tissue the RNA becomes highly fragmented (RIN 2,3). GenomeScan’s FFPE gene profiling service can handle even highly degraded RNA.

Conclusion

Every FFPE cohort is different. All new study starts with a pilot experiment to explore the quality of the your FFPE material. This will reduce the chance on suboptimal data quality and quantity. We will provide you data and guidance to make an informed decision before proceeding with the main study.
Over many projects, we have shown that our laboratory and data-analysis pipelines provide the optimal means for generating results that are biologically relevant and suitable for publication.

Gene-expression analysis on FFPE RNA is reproducible in a wide range of input amounts

Effect of the amount of FFPE RNA on mapping of the reads to exons, introns and intragenic regions. A high concordance between results is a measure of robustness and reliability. Note that the quality of FFPE material is generally diverse and results will vary depending on the extent and type of RNA damage. Sample: RNA was obtained from a pool of multiple archival FFPE blocks. Average insert size 318 ±22 bp. Isolation method: Qiagen AllPrep DNA/RNA FFPE.

Conclusion

In the figure on the left, the number of detected genes and percentage of reads that are successfully mapped are shown. These parameters decline rapidly when the samples perform suboptimal. Despite that the location to which the reads are mapped remain stable until the lower input range, these and other quality metrics show that sufficient input RNA is necessary to reach the highest level of reproducibility. Thus, especially with FFPE material, optimal results are obtained with higher amounts of RNA input. Generally, our labexperts observe optimal results with 100 to 200 ng input, depending on the extent of degradation.

Gene-expression analysis with higher amounts of RNA input leads to higher quality data

Effect of FFPE RNA input on number of detected genes and percentage mapping.

Conclusion

Our scientists have thoroughly studied the effects of RNA degradation and RNA yield on the accuracy and reliability of the transcriptomics data.
In validation studies performed with the library preparation chemistry of New England Biolabs (NEB) combined with Illumina sequencing, GenomeScan determined the parameters that are best suitable for monitoring the validity of the data. The figure on the left shows that the mapping of FFPE material remains consistent, even with less input. For this cohort, gene-expression analysis of FFPE RNA is robust and reproducible in the range of 200 to 5 ng, although below 10 ng the number of presence genes are significantly deceasing.

Frequently Asked Question

How does FFPE compare to Fresh Frozen material?

Studies generally include a comparison between Fresh Frozen (FF) material and FFPE. The goal is to show that data derived from various sources, will generate similar results. [show_more more=’…View more’ less=’View less’ color=#2ac4ea list=”>”]As elegantly shown by our partners of New England Biolabs (NEB), fixation will always lead to a certain change of the transcripts that you retrieve (see figure below). Factors that affect RNA quality are e.g. age of the blocks, treatment of the material before fixation, storage conditions. The GenomeScan entry QC will measure the extent of fragmentation of the RNA. The effect of other factors such as depurination will be measured in the small pilot study in which we perform the whole workflow. This will provide you direct and sample-specific information on the quality of your samples.

Effect of sample fixation on the read distribution*

*Figure adapted from the NEB rRNA depletion kit. Data generated using the Universal Human RNA Reference (UHR).

[/show_more]