Good Laboratory Practice - Criticism of GLP

Criticism of GLP

GLP studies require adequate and permanent documentation of everything involved in an experimental test: staff qualifications, valid study design, standard operating procedures (SOPs), training, performance, formulation and statistical analyses, and the retention of summary/individual data; so that there can be confidence in the study's design, performance and its results, and anyone (as public agencies have access to the GLP records) can subsequently fully reconstruct the study. GLP is by most regulatory authorities worldwide adopted as the lowest common standard for quality assurance. ISO 17025, GMP or GCP criteria are alternatives in some cases.

OECD Guideline test methods are recommended by regulatories as studyplan to follow for toxicology studies. These methods are all very standardized/extensively peer reviewed, and are adopted worldwide. Independent of the test guidelines, GLP is recommended by the authorities to assure the correct execution of those study plans. The correct execution of a GLP study is verified by an independent GLP monitoring authority on a regular basis (2-3 yearly). This verification means an in situ inspection of the whole test facility and his connected test sites worldwide. Audits of the studies registered with unrestricted access to all raw data produced during the whole study are a part of the inspection. In this sense it means a much deeper peer review of the study than done for an academic publication.

By contrast, academic scientists perform a wider range of basic/exploratory experimental research to: identify unknown potential hazards of chemicals, elucidate the mode/mechanism of action for known toxicants, and explore novel toxic endpoints. Accordingly, their experimental methods vary greatly in the delivery route of the test chemical, the number of test animals and the range of doses. These test methods are far more varied than the GLP test protocol is; and (at least before peer review) academics do not like to share their results or methods with laboratories competing for grant money or to give insight in raw data produced. These factors make it hard for regulatory agencies to use the results of academic researchers in chemical risk assessment.

The problem is, the regulatory agencies universal requirement that toxicity studies be performed according to OECD/GLP protocols automatically excludes the toxicity results of the independent researchers. The latter's methods, though variable, do test more realistic doses than the OECD protocols use. Thus if they find toxicity at lower doses, that important risk is not included in the risk assessment, due to the GLP requirement. Tens of thousands of published findings of toxicity from chronic toxicity have been excluded from risk assessment, a large fraction of which find toxicity at lower dose than OECD tests. Not all these independent results are high quality, but many are; and critically, they are financially disinterested.

Reviews of toxicity studies have confirmed that this false negative error (a finding of no risk when there is) is common: dozens of reviews have confirmed it for Guideline tests of pharmaceuticals; while for chemicals at least four reviews have found it. In one of those, the toxicity studies funded by the manufacturers of a high volume & well-studied chemical never found low-dose toxicity, but over 90% of its many government-funded studies did. The specific factors that lead to such false negative error by OECD/GLP studies have been analyzed.