DOI: 10.1634/THEONCOLOGIST.12-3-301
关键词:
摘要: Molecular profiling, the classification of tissue or other specimens for diagnostic, prognostic, and predictive purposes based on multiple gene expression, is a technology that holds major promise optimizing management patients with cancer. However, use these tests clinical decision making presents many challenges to overcome. Assay development data analysis in this field have been largely exploratory, leavenumerouspossibilitiesfortheintroductionofbias. Standardization profiles remains exception. Classifier performance usually overinterpreted by presenting results as p-values multiplicative effects (e.g., relative risks), while absolute sensitivity specificity remain modest at best, especially when tested large validation samples. Validation has often done suboptimal attention methodology protection from bias. The postulated classifier may be inflated compared what can achieve. With exception breast cancer, we little evidence about incremental discrimination molecular provide versus classic risk factors alone. Clinical trials started evaluate utility using cancer management. Until obtain trials, impact net benefit under real-life settings unknown. Optimal incorporation into practice not straightforward. Finally, cost-effectiveness difficult appreciate until are addressed. Overall, profiling fascinating promising technology, but its requires careful planning robust evidence. Oncologist 2007;12: 301–311