Expression Analysis on Archival Material: Comparison of 5 Commercially Available RNA Isolation Kits for FFPE Material Carolien Boeckx, MSc,* An Wouters, PhD,* Bea Pauwels, PhD,* Vanessa Deschoolmeester, PhD,* Pol Specenier, MD, PhD,*w Krzysztof Lukaszuk, MD, PhD,z Jan B. Vermorken, MD, PhD,*w Patrick Pauwels, MD, PhD,*y Marc Peeters, MD, PhD,*w Filip Lardon, PhD,* and Marc F.D. Baay, PhD* Background: Formalin-fixed paraffin-embedded (FFPE) tissue is the most common tissue specimen widely available. Moreover, long clinical follow-up is on hand. Therefore, FFPE material is a precious source of material for identifying predictive and/or prognostic biomarkers in cancer research on the basis of gene expression. However, the main drawback of FFPE tissue is the significant reduction in quantity and quality of the extracted RNA. The aim of this study is the comparison of different commercially available kits for the RNA isolation in FFPE tissue material. Methods: Five commercially available RNA isolation kits were tested and the concentration, purity, integrity, and raw cycle threshold values were determined. Results: The mean total RNA concentrations were as follows: Qiagen 25957 ± 19417 ng, Ambion 8249 ± 2898 ng, SA Bio- sciences 8070 ± 3700 ng, and Macherey-Nagel 622 ± 394 ng. The mean A 260 /A 280 ratios were as follows: Qiagen: 1.81, SA Biosciences: 0.66, Ambion: 1.03, and Macherey-Nagel: 1.04. The mean A 260 /A 230 ratios were as follows: Qiagen: 1.88, SA Biosciences: 1.61, Ambion: 1.54, and Macherey-Nagel: 1.88. The RNA extractions from Epicentre could not be measured by the Nanodrop and, therefore, were excluded from further analysis. The mean RNA integrity number (range, 2.09 to 2.47) and the mean raw cycle threshold values (range, 33.43 to 35.37) were more or less the same for all the tested RNA isolation kits. Conclusions: Altogether, on the basis of the number of adequate isolations, the kit from Qiagen seems to be the most appropriate kit to be used in our further studies that require RNA isolation from FFPE material. Key Words: RNA, FFPE, qRT-PCR (Diagn Mol Pathol 2011;20:203–211) G ene expression studies are highly informative to identify predictive and/or prognostic biomarkers in cancer research. Therefore, high-throughput molecular genetic techniques, such as quantitative reverse transcrip- tase polymerase chain reaction (qRT-PCR) and micro- array are very important tools in this area. 1 A potential complicating aspect in this is the reality that the most commonly available tissue specimen connected with long- term clinical follow-up is formalin-fixed paraffin-em- bedded (FFPE) material. Although formalin fixation preserves the integrity of the tissues (in essence, tissue architecture and proteins), its main limitation is the significant reduction in the quality and quantity of the isolated RNA. 2–5 RNA extraction from FFPE tissue is challenging due to several issues, such as degradation, cross-linking with proteins, and modifications to the RNA. 6 The degree of enzymatic and chemical degrada- tion of RNA is affected by storage conditions (tempera- ture and time), time between sample acquisition and fixation, fixation time, and sample thickness. 3,7 The chemical modification of RNA involves extensive cross- linking between nucleic acids and proteins. As a result, RNA becomes more resistant to extraction and the isolated RNA has an average size of r200 bases. 3,8 Therefore, amplification is generally limited to short fragments. Nevertheless, Zhang et al 9 were able to amplify RNA extracted from FFPE material up to fragments of 705 bp. Another problem is the formation of monomethylol adducts on adenine, which seems to be more susceptible than the other bases in RNA. Conse- quently, the poly A tail of mRNA in FFPE specimens is heavily modified. 3,6,10 All these processes are known to re- duce the efficiency of qRT-PCR reactions. The qRT-PCR Copyright r 2011 by Lippincott Williams & Wilkins From the *Center for Oncological Research (CORE) Antwerp, Laboratory of Cancer Research and Clinical Oncology, University of Antwerp, Wilrijk; wDepartment of Medical Oncology, University Hospital Antwerp (UA/UZA); yDepartment of Pathology, Univer- sity Hospital Antwerp (UA/UZA), Edegem, Belgium; and zDepartment of Gynaecologic Endocrinology, Medical University of Gdansk, Poland. C.B. was supported by a scholarship from the University of Antwerp and this study was supported by grants from the Special Research Fund (BOF) of the University of Antwerp to B.P and M.B. The other authors declare no conflict of interest. Reprints: Carolien Boeckx, MSc, University of Antwerp, Laboratory of Cancer Research and Clinical Oncology, Universiteitsplein 1 (T3.11), B2610 Wilrijk, Belgium (e-mail: carolien.boeckx@ua.ac.be). ORIGINAL ARTICLE Diagn Mol Pathol  Volume 20, Number 4, December 2011 www.molecularpathology.com | 203