Authors' Affiliations: Department of Investigational Cancer Therapeutics (Phase I Program), University of Texas MD Anderson Cancer Center, Houston, Texas;
Division of Medical Oncology, University of Ottawa, Ottawa, Canada.
Clin Cancer Res. 2014 Jul 1;20(13):3364-70. doi: 10.1158/1078-0432.CCR-13-3206. Epub 2014 Apr 15.
The U.S. Food and Drug Administration (FDA) Center for Drug Evaluation and Research is responsible for evaluating drug safety and efficacy, including oversight of clinical trials and principal investigators. The FDA Clinical Investigator Inspection List (CIIL) contains online, detailed, relevant information of all FDA inspections. We reviewed FDA inspections of clinical investigators to ascertain their outcome and included all inspections on the list (July 1977 through December 31, 2009; n = 9,481 inspections). Eighty-eight percent of inspections were "data audit" (primary purpose = verification of data), and the rest (12%) were "for cause." The number of inspections each year significantly increased over time (P < 0.0001) and averaged 350 per year in the past decade. No deficiencies were found in only 11.2% of all "data audit" and 5% of all "for cause" inspections. Only 31% of inspections resulted in "no action indicated." About two thirds of inspections resulted in some finding, requiring either voluntary investigator action (61.3% of inspections) or official FDA action (3.9%). The most frequently cited deficiencies were failure to follow investigational plan (34%), inadequate informed consent form (28%), and inadequate/inaccurate records (27%). In conclusion, over the past decade, the FDA has performed approximately 350 inspections per year, with the number increasing over time. The vast majority of FDA inspections yield deficiency findings and, as a result, only about one third of inspections have an outcome of "no action indicated."
美国食品和药物管理局（FDA）药品评估和研究中心负责评估药物的安全性和有效性，包括对临床试验和主要研究者的监督。FDA 临床研究者检查清单（CIIL）包含所有 FDA 检查的在线、详细、相关信息。我们审查了 FDA 对临床研究者的检查，以确定其结果，并包括清单上的所有检查（1977 年 7 月至 2009 年 12 月 31 日；n = 9481 次检查）。88%的检查是“数据审核”（主要目的是验证数据），其余 12%是“有因检查”。每年的检查数量随着时间的推移显著增加（P < 0.0001），在过去十年中平均每年有 350 次检查。只有 11.2%的“数据审核”和 5%的“有因检查”没有发现缺陷。只有 31%的检查结果为“无需采取行动”。大约三分之二的检查结果发现了一些问题，需要调查员自愿采取行动（61.3%的检查）或 FDA 采取官方行动（3.9%）。最常被引用的缺陷是未遵循研究计划（34%）、知情同意书不充分（28%）和记录不充分/不准确（27%）。总之，在过去十年中，FDA 每年大约进行 350 次检查，且数量随着时间的推移而增加。绝大多数 FDA 检查都发现了缺陷，因此只有大约三分之一的检查结果为“无需采取行动”。
