Building quality measures into the regulatory review process

Abstract

The Centre for Medicines Research (CMR) in the UK has produced a publication based on its 17th workshop that dealt with quality in the regulatory process [1], which was a logical sequel to the 16th workshop on the clinical dossier [2]. In some ways, it is a companion piece to the recent EMEA/IFAPP conference [3]. As stated in the foreword, the quality of the review process is inseparable from the quality of the submitted dossier and, in turn, the latter depends on the quality of the work performed. However, this simplistic linkage fails to differentiate between the factual basis of the work and of the dossier, on the one hand, and the opinions expressed by experts and the decision-making process in the regulatory review, on the other. After all, the same dossier is reviewed by different regulatory agencies, and the fact that each may identify and/or focus on one or two different issues from the others surprises people in the industry, who have seen this scenario many times. Possible reasons for any divergence was illustrated by the first contribution describing the findings of a CMR International Study. In 1999, the CMR sent a questionnaire to ten major regulatory authorities, of which nine responded. While most agencies provided training for their reviewers and collaborate with each other in training, there was a marked difference in training for external experts and advisory committee members, who make the decisions. Only one agency provided such training. Another difference in practice between authorities was the fact that, in half the cases, Advisory Committees reviewed only selected dossiers, while in the rest all were reviewed. In six authorities, the reviews only covered the assessment reports and only in one was the complete dossier reviewed. Five agencies have extensive contact with sponsors during development and four during the review process. One agency has no contact with the sponsor during development and another has no contact during the review process. A second paper presented an analysis of the outcomes of the European Union Centralized Procedure between 1995 and 1999. The main findings were the rise in the number of negative outcomes, primarily due to withdrawals of the submissions when the applicant received the list of questions, peaking in 1998 (41%) and falling slightly in 1999 (31%). The major negative decisions were based on clinical efficacy, mainly due to their design and the doubtful the robustness of pivotal data, and on clinical safety, where increased mortality and other serious adverse events activated those decisions. There was a positive concordance between the Food and Drug Administration (FDA) and the EMEA on positive decisions, but additionally the FDA approved 35% of applications that had a negative EMEA outcome. The FDA’s recent actions were presented in another paper that described the enhanced performance goals and the Good Review Practice (GRP) initiative. The latter was launched in 1994 and it has examined clinical holds and refusals to file and is setting guidelines in order to achieve consistency and uniformity. It has provided training for Advisory Committee members, although it is not mandatory. The syndicate groups and the workshop discussion addressed the differences in quality assurance methods employed by various national authorities and the impact of them. It also looked at the quality of the dossier on the quality of the review process. The key recommendations included global standards, clarification of questions during the review process, and regulators using review templates. The industry should check the quality of the dossier, provide a ‘road map’ in dossiers to orientate reviewers, and provide early label proposals and supporting studies for them.