Biotechnology Healthcare - June 2008 - (Page 39) “Biologists and clinical investigators want to use technology like this because it’s very powerful, but it’s a real challenge for them because the data analysis is complicated,” says Richard M. Simon, DSc, of the National Cancer Institute. “It’s an enormous amount of work,” Hayes says. “These aren’t just opinion-based guidelines. We tried to make them as evidencebased as we could, to review all the available literature relative to the specific topic. Then we tried to use fair, balanced, and objective criteria for our recommendations.” One critical tool in this laborintensive process was the Tumor Marker Utility Grading System, de- vised by ASCO panelists, including Hayes, when ASCO first decided to include tumor markers in its practice guidelines in 1995. The system established three Levels of Evidence to define the data quality on a given marker: • Level of Evidence I, which is the most credible, designates data from prospective, randomized, controlled trials specifically designed to test the utility of the marker, or meta-analyses of well-designed studies. • Level of Evidence II designates data from prospective therapeutic trials in which marker utility is a secondary study objective. • Level of Evidence III designates data from large but retrospective studies. It is this methodology, along with the biostatistical and clinical trial “dos and don’ts” that Simon writes about, that ultimately caused the ASCO panel to recommend Oncotype DX, but not the Breast Cancer Gene Expression Ratio, MammaPrint, or the Rotterdam Signature. Marketed by AviaraDx, in Carlsbad, Calif., the Breast Cancer Gene Expression Ratio quantifies the ratio of the HOXB6 and IL17BR genes in tumor tissue. Developmental studies were done with microarrays and frozen tissue, but validated by realtime quantitative polymerase chain reaction (RT-PCR) using the same tumor samples. The ASCO Update Committee reports that the Breast Cancer Gene Expression Ratio “is significantly and independently associated with poorer disease-free survival in two studies of lymph node-negative, ER-positive, tamoxifen-treated patients with breast cancer.” However, there are no published studies demonstrating that the assay is an improvement over conventional methods of classifying patients by recurrence outcomes, or of predicting chemotherapy benefit. “Most of us just don’t know how to use this [MammaPrint] assay,” says Hayes. “It’s been shown that this assay does indeed provide prognostic information, but it’s not clear in what situation or how to apply that information to a specific patient. They have not actually addressed a specific clinical question.” MAY/JUNE 2008 · BIOTECHNOLOGY HEALTHCARE 39 Rob Crandall
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.