Shi Leming, Tong Weida, Goodsaid Federico, Frueh Felix W, Fang Hong, Han Tao, Fuscoe James C, Casciano Daniel A
US Food and Drug Administration, Center for Toxicoinformatics, Division of Systems Toxicology, National Center for Toxicological Research, HFT-020, 3900 NCTR Road, Jefferson, AR 72079, USA.
Expert Rev Mol Diagn. 2004 Nov;4(6):761-77. doi: 10.1586/14737159.4.6.761.
The scientific community has been enthusiastic about DNA microarray technology for pharmacogenomic and toxicogenomic studies in the hope of advancing personalized medicine and drug development. The US Food and Drug Administration has been proactive in promoting the use of pharmacogenomic data in drug development and has issued a draft guidance for the pharmaceutical industry on data submissions. However, many challenges and pitfalls are facing the microarray community and regulatory agencies before microarray data can be reliably applied to support regulatory decision making. Four types of factors (i.e., technical, instrumental, computational and interpretative) affect the outcome of a microarray study, and a major concern about microarray studies has been the lack of reproducibility and accuracy. Intralaboratory data consistency is the foundation of reliable knowledge extraction and meaningful crosslaboratory or crossplatform comparisons; unfortunately, it has not been seriously evaluated and demonstrated in every study. Profound problems in data quality have been observed from analyzing published data sets, and many laboratories have been struggling with technical troubleshooting rather than generating reliable data of scientific significance. The microarray community and regulatory agencies must work together to establish a set of consensus quality assurance and quality control criteria for assessing and ensuring data quality, to identify critical factors affecting data quality, and to optimize and standardize microarray procedures so that biologic interpretation and decision-making are not based on unreliable data. These fundamental issues must be adequately addressed before microarray technology can be transformed from a research tool to clinical practices.
科学界对用于药物基因组学和毒理基因组学研究的DNA微阵列技术满怀热情，期望借此推动个性化医疗和药物研发。美国食品药品监督管理局积极推动在药物研发中使用药物基因组学数据，并已发布针对制药行业数据提交的指导草案。然而，在微阵列数据能够可靠地用于支持监管决策之前，微阵列领域和监管机构面临着诸多挑战与陷阱。四类因素（即技术、仪器、计算和解释因素）会影响微阵列研究的结果，而微阵列研究的一个主要问题一直是缺乏可重复性和准确性。实验室内的数据一致性是可靠知识提取以及有意义的跨实验室或跨平台比较的基础；不幸的是，并非每项研究都对此进行过认真评估和论证。通过分析已发表的数据集发现了数据质量方面的严重问题，许多实验室一直在忙于技术故障排除，而非生成具有科学意义的可靠数据。微阵列领域和监管机构必须共同努力，建立一套用于评估和确保数据质量的共识性质量保证和质量控制标准，识别影响数据质量的关键因素，并优化和规范微阵列程序，以便生物学解释和决策并非基于不可靠的数据。在微阵列技术从研究工具转变为临床实践之前，必须充分解决这些基本问题。
