Automated Extraction of Microbial DNA From Formalin-Fixed, Paraffin-Embedded Tissue Using the Siemens Tissue Preparation System

• Published in: American journal of clinical pathology Vol. 150; no. suppl_1; pp. S151 - S152
• Main Authors: Saadalla, Abdulrahman, Thorvilson, Jill, Berry, Jonathan, Buckwalter, Seanne, Lynne, Sloan, Rucinski, Stefanea, Koelsch, Jolene, Pritt, Bobbi, Wengenack, Nancy, Schuetz, Audrey
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 21.09.2018
• Subjects: Automation; Bartonella; Clinical microbiology; Deoxyribonucleic acid; DNA; Efficiency; Endopeptidase K; Ethanol; Formaldehyde; Human error; Laboratories; Lysis; Mechanical unloading; Paraffin; Polymerase chain reaction; Proteinase; Sectioning; Tropheryma whipplei; Tuberculosis; Unloading; Xylene
• ISSN: 0002-9173; 1943-7722
• DOI: 10.1093/ajcp/aqy112.357

Abstract We describe the adoption of an automated nucleic acid extraction platform, Siemens Tissue Preparation System (TPS), for use with formalin-fixed, paraffin-embedded (FFPE) tissues intended for PCR microbial testing. The TPS uses a liquid handling robot to automate deparaffinization, followed by extraction and purification of nucleic acids using magnetic particle-based isolation. While it has been used previously for human DNA extraction, this is the first incorporation of the TPS into clinical microbiology practice for microorganism detection to our knowledge. Our laboratory transitioned to use of the TPS from manual deparaffinization of FFPE tissue with subsequent DNA extraction using Roche MagNA Pure (MP) instruments (manual/MP method). Unpublished data from our group demonstrated that the TPS performed better than the manual/MP method for FFPE tissue extraction (27% higher positivity in 86 FFPE specimens tested by both methods). We tracked total testing volumes, laboratory efficiency, TPS errors, and workflow. Additionally, we added a microtome instrument to our clinical microbiology laboratory to section FFPE blocks in order to increase efficiency by avoiding reliance on the histopathology laboratory. Since implementation, 710 FFPE specimens were processed on the TPS, with 724 downstream PCR tests run on extracts for eight laboratory-developed assays: Mycobacterium tuberculosis (MTB), Bartonella spp, Tropheryma whipplei, Coxiella burnetii, Kingella kingae, Coccidioides spp, Acanthamoeba spp, and Balamuthia mandrillaris/Naegleria fowleri. The most common tests run after TPS extraction included 468 (64.6%) MTB, 156 (21.5%) Bartonella, and 70 (9.7%) T whipplei; positivity rates were 4.9%, 6.4%, and 4.3%, respectively. Regarding workflow, use of the TPS decreased total processing time by approximately 70%. The TPS workflow from tissue block sectioning to DNA extraction averaged a total of 3.2 hours, of which 0.7 hours (21.8%) were hands-on time (microtome sectioning and TPS specimen loading/unloading). Comparatively, our manual/MP method averaged 10.6 hours, of which 1.1 hours (10.5%) involved hands-on time (microtome sectioning, manual deparaffinization using xylene and 95% ethanol, and MP loading/unloading). Approximately 7 hours of the manual/MP method were devoted to proteinase K lysis. The TPS accommodates 48 specimens per run, with each run lasting 2.5 hours. Our laboratory currently performs one run daily with an average of four specimens per run. Eleven (6.25%) failure events were noted in 176 runs over the past 9 months. Nine (82% of the errors) were attributed to robotic mechanical problems and two (18%) to human errors (reagent misplacement). The system software alerts the user when errors occur so the problem can be fixed and/or the run repeated. In conclusion, the automated TPS extraction system for FFPE is rapid and limits exposure to hazardous chemicals with minimal run failures. Transition of FFPE sectioning to the microbiology laboratory from the histology laboratory also increased efficiency.