Received on December 7, 2004
Accepted on February 4, 2005

Automation and Analytical Techniques

Technology for Automated, Rapid, and Quantitative PCR or Reverse Transcription-PCR Clinical Testing

¹ Departments of Surgery, Biostatistics, and Pathology, Hillman Cancer Center, University of Pittsburgh, Pittsburgh, PA 15213
² Departments of Surgery, Biostatistics, and Pathology, Hillman Cancer Center, University of Pittsburgh, Pittsburgh, PA 15213, and Cepheid, Sunnyvale, CA 94089
³ Departments of Surgery, Biostatistics, and Pathology, Hillman Cancer Center, University of Pittsburgh, Pittsburgh, PA 15213, and Current affiliation: Department of Medicine, Mount Sinai School of Medicine, New York, NY 10029
⁴ Departments of Surgery, Biostatistics, and Pathology, Hillman Cancer Center, University of Pittsburgh, Pittsburgh, PA 15213, Hillman Cancer Center, University of Pittsburgh, Pittsburgh, PA 15213

^* To whom correspondence should be addressed. E-mail: tony.godfrey{at}mssm.edu.

Background: PCR-based assays can improve clinical care, but they remain technically demanding and labor-intensive. We describe a new instrument, the GeneXpert^®, that performs automated nucleic acid isolation, reverse transcription, and fluorescence-based quantitative PCR in 35 min.

Methods: Yield and integrity of RNA isolated on the GeneXpert were compared with Qiagen-based extraction for parallel samples (5-µm frozen tissue sections). The reproducibility of automated RNA isolation, reverse transcription, and quantitative PCR was determined by replicate (n = 10) analysis of 10 tissues, using duplex (target and endogenous control) reverse transcription-PCR reactions for two gene combinations. The GeneXpert was then used to perform rapid analysis of lymph nodes from melanoma, breast cancer, and lung cancer patients and analysis of melanoma metastatic to the lung, primary lung adenocarcinoma, and healthy lung tissue.

Results: On the GeneXpert, RNA was recovered in slightly over 6 min, and the yield was 70% of that from parallel Qiagen reactions. The RNA integrity was comparable to that of Qiagen-isolated RNA as determined by gel electrophoresis. The standard error of measurement for the Ct in the reproducibility analyses was 0.79 for one assay and 0.71 for the other. GeneXpert assays successfully detected the presence of metastatic melanoma, breast cancer, and lung cancer in lymph nodes and also differentiated among metastatic melanoma, lung adenocarcinoma, and healthy lung.

Conclusions: RNA yield and integrity on the GeneXpert are comparable to benchtop methods. Reproducibility of the GeneXpert data is similar to that seen with manual methods in our hands but may need improvement for some applications. The GeneXpert can perform RNA isolation, reverse transcription, and quantitative PCR in 35 min and could therefore be used for intraoperative testing when applicable.