Collaborative study on influenza vaccine clinical trial serology - part 1: CHMP compliance study.

: The Quality of Medicines & HealthCare (EDQM, Council of Europe) and the European Union (EU) Commission to evaluate the reproducibility of clinical serology results for seasonal influenza vaccines and to assess the impact of technical differences between laboratories on the compliance with the Committee for Human Medicinal Products (CHMP) criteria set by the European Medicines Agency (EMA). The study was run in 2 phases. The present article reports the 1st phase of the study, which aimed at evaluating the variability of the results obtained by 11 laboratories (5 national control laboratories and 6 influenza vaccine manufacturers) using their routine haemagglutination inhibition (HI) assay to test a common panel of clinical trial sera. The results confirmed the limited inter-laboratory reproducibility of the HI testing of influenza vaccine clinical trial samples. In some cases a good agreement was found between laboratories, while a systematic bias or a random scatter of results was observed in other cases. Analysis of estimated systematic bias confirmed that differences between laboratories can be significant (up to 16-fold) in some cases. Correction for this bias resulted in limited improvement. Differences between laboratories were found to result in discrepant decisions on marketing acceptance of vaccines or to decisions based on compliance to different criteria. The study showed that the seroconversion (SC) and mean fold increase (MFI) criteria are more robust against systematic over- or under-estimation of titres whereas the protection rate (PR) is very sensitive to this effect. The fundamental issues with the PR criteria are discussed.