Improved Fecal DNA Test for Colorectal Cancer Screening

• Published in: Clinical gastroenterology and hepatology Vol. 5; no. 1; pp. 111 - 117
• Main Authors: Itzkowitz, Steven H., Jandorf, Lina, Brand, Randall, Rabeneck, Linda, Schroy, Paul C., Sontag, Stephen, Johnson, David, Skoletsky, Joel, Durkee, Kris, Markowitz, Sanford, Shuber, Anthony
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 2007
• Subjects: Age Factors; Aged; Colonoscopy; Colorectal Neoplasms - diagnosis; Colorectal Neoplasms - genetics; DNA Fragmentation; DNA Methylation; DNA, Neoplasm - analysis; DNA-Binding Proteins - genetics; Feces - chemistry; Female; Gastroenterology and Hepatology; Humans; Male; Mass Screening - methods; Middle Aged; Patient Satisfaction; Polymerase Chain Reaction; Sensitivity and Specificity; Transcription Factors - genetics; Vimentin - genetics; CRC; DY; NC; CI; HLTF; MSP; DIA; PCR; colorectal cancer; DNA integrity assay; Helicase-like Transcription Factor; locus D (5p21) and locus Y (LOC91199); methylation-specific polymerase chain reaction; confidence interval; normal colonoscopy; polymerase chain reaction
• DOI: 10.1016/j.cgh.2006.10.006
• ISSN: 1542-3565; 1542-7714

Background & Aims: Fecal DNA testing has shown greater sensitivity than guaiac-based occult blood tests for noninvasive colorectal cancer (CRC) screening. The prototype assay (version 1), which analyzed 22 gene mutations and DNA integrity assay (DIA), showed a sensitivity of 52% for CRC detection and a specificity of 94% in average-risk individuals. The present study was conducted to determine the sensitivity and specificity of a second-generation assay (version 2) that uses improved DNA stabilization/isolation techniques and a new promoter methylation marker. Methods: Forty patients with CRC and 122 subjects with normal colonoscopy provided stool samples to which DNA preservation buffer was added immediately. DNA was purified using gel-based capture, and analyzed for the original panel of 22 mutations, DIA, and 2 new promoter methylation markers. Results: By using DNA that was optimally preserved and purified from stool, the sensitivity of the prototype version 1 assay increased to 72.5% because of enhanced performance of DIA. Vimentin gene methylation alone provided sensitivity and specificity of 72.5% and 86.9%, respectively. The optimal combination of vimentin methylation plus DIA resulted in 87.5% sensitivity and 82% specificity; cancers were detected regardless of stage or location. False-positive vimentin methylation was associated with older age. Conclusions: An improved fecal DNA test that incorporates only 2 markers shows much higher sensitivity for CRC. The new assay is easier to perform and should be less costly, thereby facilitating its use for noninvasive CRC screening.