Improving detection of adverse effects of marketed drugs.
Improving detection of adverse effects of marketed drugs.
复制标题
改进上市药物不良反应的检测。
DOI：
10.1001/jama.298.3.333
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
C. O'brien
中科院分区：
文献类型：
--
作者：
D. Klein;C. O'brien
公众对美国食品和药物管理局 (FDA) 批准的上市药物的严重毒性的担忧引起了制药行业和 FDA 的怀疑。这导致 FDA 对抗抑郁药的使用发出了严厉的黑框警告；然而，采取这一行动的理由可能不充分。核心问题是，当前有缺陷的系统甚至无法检测罕见但严重的毒性，也无法检测药物使用后期出现的毒性或超说明书使用期间发生的毒性。此外，上市前临床试验并未解决与药物相互作用或共存疾病可能产生的毒性相关的问题。人们常常没有认识到，在临床试验中获得的原始适应症信息没有任何基础，导致其他适应症的合法超说明书使用，药品申办者也不负责检查此类使用。 FDA 的警告（例如黑框）是否建议加强监测（主要是假阳性）或表明停药实际上可以改善公众健康，目前仍不得而知。一个常见的建议是，FDA 应要求药品申办者在上市后进行临床试验，首先仅延长有条件的上市批准。尽管这种策略可能是合理的，但它并没有充分解决这样一个事实：即使在扩展的临床试验中，获得足够的能力来正确评估罕见事件也是不切实际的。更长、更大规模的上市后试验会带来重大的实际问题。患者退出，使得研究的评估变得越来越困难。大型临床试验需要多中心方案，并伴随着管理和临床问题。招募大量相关患者样本，考虑每种新药的年龄、性别、合并症和药物相互作用，显然是不可行的。 FDA 上市后监测系统的不足是显而易见的。对它的结构、笨拙或易受行业影响的指责转移了人们对这个核心问题的注意力。一些国际领先的精神药理学研究专家认为，只有前瞻性的交联计算机医学数据库才能前瞻性地提供足够的临床数据以进行理性判断。
PUBLIC CONCERN ABOUT SERIOUS TOXICITIES FROM MARketed drugs approved by the US Food and Drug Administration (FDA) has led to suspicion of both the pharmaceutical industry and the FDA. This has caused the FDA to issue severe, black-box warnings about antidepressant use; however, this action may have been taken based on inadequate grounds. The central problem is that the current faulty system cannot even detect rare but severe toxicities nor can it detect those that appear late in the use of medication or those that occur during off-label use. Furthermore, premarketing clinical trials do not address issues related to possible toxicities due to drug interactions or comorbid illnesses. It is often not recognized that there is no basis from information gained in clinical trials for the original indications that lead to legal off-label use for other indications, nor is the pharmaceutical sponsor responsible for checking on such use. Whether FDA warnings, such as black boxes, that recommend increased monitoring (largely of false-positives) or suggest that drug withdrawals actually improve the public health remains unknown. A frequent suggestion is that the FDA should require pharmaceutical sponsors to conduct clinical trials after marketing by first extending only conditional marketing approval. Although this tactic may be reasonable, it does not adequately address the fact that attaining adequate power to properly evaluate rare events is simply not practical, even in extended clinical trials. Longer and larger postmarketing trials incur major practical problems. Patients drop out, making studies progressively more difficult to evaluate. Large clinical trials require multisite protocols with attendant administrative and clinical problems. The recruitment of large samples of relevant patients, accounting for age, sex, comorbidity, and drug interactions for each new agent, is obviously unfeasible. The inadequacy of the FDA’s postmarketing surveillance system is evident. Accusations about its structure, bungling, or vulnerability to industry influence deflect attention from this central issue. Some leading international experts in psychopharmacological research have argued that only a prospective crosslinked computerized medical database can prospectively furnish sufficient clinical data to allow for rational judgment.