No treatments have yet been shown to slow the progression of the inherited neurodegenerative disorder Huntington's disease (HD) in humans, but several attempts at disease modification in animal models have been successful. Human clinical trials based on present clinical measures would require unfeasibly large subject numbers, particularly in premanifest HD mutation carriers, and the main aim of biomarker research in HD is to enable the development of surrogate end points to enable such trials to be conducted. In this article, the imaging, biofluid, quantitative motor and cognitive measures that have been shown to reflect the progression of HD are reviewed. A conceptual framework and pipeline for evaluating the large number of potential HD biomarkers is presented, and the need for systematic head-to-head comparison of candidate markers is highlighted.