Many cancer biomarkers were put into clinical use without a proper sequence of “level of evidence” evaluation (e.g., PSA for prostate cancer screening). Furthermore, most cancer biomarkers failed on their journey to clinic use. False positive findings are rampant. Better strategies are necessary for structured phase of development parallel to the phases for drug development, and rigorous study design standards to enhance the chance of producing biomarkers with clinical utility.

As the Coordinating Center for the Early Detection Research Network (EDRN), we have proposed a 5-phase biomarker development guideline and PRoBE study design standards to address the above issues. The key elements of PRoBE principles are: (i) the Clinical Context; (ii) Performance Criteria; (iii) the Biomarker test; and (iv) Study power and termination.

Two EDRN liver cancer biomarker validation studies, a completed Phase-2 and an on-going Phase-3, were used to illustrate how these concepts can guide a more structured and systematically development of cancer biomarkers with clinical goal in mind. The Phase-2 study identified promising candidate biomarkers and produced a specimen reference set to systematically evaluate new biomarkers. The Phase-3 study is set to address the clinical question of using biomarkers to improve the early detection of HCC and to reduce unnecessary CT/MRI for patients with liver cirrhosis.

Citation Format: Ziding Feng. Biomarker statistical design. [abstract]. In: Proceedings of the Eleventh Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2012 Oct 16-19; Anaheim, CA. Philadelphia (PA): AACR; Cancer Prev Res 2012;5(11 Suppl):Abstract nr ED03-02.