The Validity of Chemical Analytical Methods by the Example of the Heparin Product Recall

In the last months, 11 countries recalled heparin products because of severe side effects, which included 62 deaths reported to the FDA on or after January 1, 2008, in the USA. Active pharmaceutical ingredients have to fulfil quality requirements. The determination of the API's (Active Pharmaceutical Ingredients) quality relies on the results of analytical testing. The decisions, as to whether the quality is acceptable or not acceptable are risk-based decisions. The quality of analytical data is therefore necessary for the control and release of the batches. Selectivity, specificity and sensitivity are key factors according to the ISO standards, allowing the calculation of positive and negative predictive values. The FDA now postulates the application of two analytical test methods: capillary electrophoresis and H-NMR. To date, no data exist concerning the above mentioned analytical quality parameters. Therefore, in this study, the Bayes' theorem, a statistical tool, was applied to calculate risk probabilities. Furthermore it is discussed: Are the actual API audit and inspection controls effective? From a statistical point of view, are the FDA's proposed analytical methods suitable for the qualified person judgement for release of heparin and other APIs? What is the safety risk of the spot checks-based release practice? Are PAT (Process Analytical Technology) tools, in combination with a whole supply chain management, suitable for optimizing safety and minimizing risk?