Posted: July 3, 2014

Steps Towards Precision Medicine: Utilizing FFPE Specimens for Comprehensive Genomic Characterization

Image: Dr. Roy W. Tarnuzzer

Roy W. Tarnuzzer, Ph.D.
Biospecimen Core Resource Program Manager at The Cancer Genome Atlas Program Office

When completed, The Cancer Genome Atlas (TCGA) will have provided comprehensive characterization of cancer genomes from 24 of the most common as well as 9 of the most rare tumor types.  This work has already resulted in the identification of new drivers of tumor biology and new therapeutic targets, and the development of a molecular-based cancer taxonomy.  As these emerging classification schemas mature, there will be greater opportunity to apply these approaches to the clinic, thereby bringing precision medicine that much closer to becoming a clinical reality.

To date, TCGA has utilized DNA and RNA extracted from frozen tissue specimens, a preservation method that is consistent with that used in many seminal studies that have optimized current state-of-the-art genomics and transcriptomics platforms. However, as the scientific and clinical community move forward into an era of precision medicine, the reality is that the availability of frozen tissue samples for clinical studies is limited. Frozen tissue procurement poses many difficulties that render the use of most frozen samples impractical for diagnostic purposes.  

In fact, the vast majority of specimens available for diagnostic purposes are in formalin-fixed paraffin-embedded (FFPE) form.  Given that this preservation method introduces molecular artifacts, there is a well-recognized need to optimize pipelines that utilize state-of-the-art technologies for the isolation and analysis of analytes derived from FFPE samples. With this realization, the TCGA Program Office established the Formalin-fixed Paraffin Pilot Project (FPPP) to test and implement methods that could enable the use of DNA and RNA analytes extracted from FFPE specimens in conjunction with these technologies. The aims of FPPP were:

  1. To develop a method for optimal co-extraction of DNA and RNA from the same portion of an FFPE sample, maximizing the analyte integrity obtained and allowing for co-isolation of nucleic acids
  2. To assess the performance of state-of-the-art platforms for whole exome sequencing, copy number profiling, RNA and miRNA sequencing, and whole genome methylation analysis with FFPE derived analytes
  3. To define the best practices for the analysis of these platforms with data obtained from nucleic acids extracted from FFPE samples
  4. To report on the limitations imposed by the nature of FFPE samples and provide guidelines for the optimal use of current genomic and transcriptomic methods for the analysis of FFPE samples

In preparation for the FPPP, the Biospecimen Core Resource (BCR) was tasked with the development of an optimized DNA/RNA co-isolation protocol that improved the integrity of nucleic acids extracted from freshly prepared control FFPE blocks and scrolls compared to “off-the-shelf” commercially available methods. Then, this optimized method was tested on a panel of 13 FFPE biospecimens with existing frozen counterparts in TCGA.  These “Pilot 1” samples were subsequently characterized on existing platforms using protocols for frozen specimens. The participating Genome Sequencing Centers (GSCs) and Genome Characterization Centers (GCCs) compared the quality metrics obtained from the FFPE-derived analytes with those from case-matched frozen specimen-derived data to further optimize the characterization protocols for the FFPE-derived nucleic acids. The results were encouraging and indicated that FFPE derived analytes could in fact be effectively characterized using the TCGA model.  However, during the analysis phase, it was found that numerous variables due to the fact that the FFPE analytes were characterized long after the frozen analytes complicated the identification of molecular artifacts specifically introduced by the FFPE preservation method.

As a result, a second phase, “Pilot 2” of FPPP, was designed to further verify the suitability of FFPE-derived DNA and RNA for use in the TCGA characterization model. The FPPP included multiple cancers from 38 patients that were selected based on the availability of matched frozen and FFPE tumor specimens as well as germline DNA extracted from blood. The BCR utilized the newly developed and optimized co-isolation protocol to extract DNA and RNA from the 38 sets of biospecimens.  To ensure suitability for downstream characterizations, these nucleic acids were subjected to the same rigorous quality control standards as are all TCGA samples.  Upon qualification, these samples were sent to the GSCs and GCCs for comprehensive characterization, including whole exome sequencing, RNA sequencing, miRNA sequencing, array profiling (SNP6 and methylation), and analyzed using the best practice approaches implemented by the TCGA characterization centers. The FPPP study allowed direct comparison of results derived from FFPE and frozen tissues in a side-by-side design, thus controlling for any variability in methods and handling.

The FPPP is currently in its analysis phase; the FFPE Analysis Working Group (AWG) is finalizing its findings and preparing a manuscript.  The FFPE AWG consists of representatives from each of the GSCs, GCCs, and Genome Data Analysis Centers (GDACs) at Memorial Sloan Kettering Cancer Center and Oregon Health and Science University.  It is co-chaired by Erik Zmuda, Ph.D., from the BCR at Nationwide Children’s Hospital and Jorge Reis-Filho, M.D., Ph.D., from Memorial Sloan Kettering Cancer Center.  The major objectives are to further define best practices for the use of FPPE-derived nucleic acids in molecular diagnostics and genomic characterizations, to identify artifacts inherent in the use of these analytes, and to design informatic workarounds to compensate for any imposed limitations. These best practices, protocols, and informatic tools developed in the TCGA FPPP will lay the groundwork for future projects at the Center for Cancer Genomics.