(P-457) Systematic Review of Prospective Pregnancy Studies in Europe: Scope for Improvement

Background: Regulatory agencies request post-approval studies (PASS) to evaluate adverse effects on pregnancy, birth and childhood development from exposure in pregnancy of newly approved drugs. These studies are almost invariably observational and may involve primary data collection and/or secondary data sources.

Objectives: To identify the methodology and assess quality and performance of pregnancy primary data collection prospective studies in Europe.

Methods: A systematic literature review of completed prospective pregnancy studies between January 2011 and November 2019 conducted in at least one European country with outcomes recorded was conducted using the European Union electronic Post-Authorisation Studies Register. Information from EMA for studies with absent or partial study reports publicly available was sought.

Results: From a total of 30 pregnancy studies, a total of 8 eligible prospective pregnancy studies were identified.
The eight studies were conducted through 18 different European countries. Seven studies (87.5%) were multi-country studies.
Half the studies (4/8) were conducted as active prospective data collection using a ´wide-net all comers´ approach and half used an established registry database to recruit patients. The ´wide-net all registrants´ approach did not reach the target sample size in 75% (3/4) of studies. In established registry database studies, the target sample size was not reached in 25% (1/4) of studies.
Of the eight studies, data were available for both dose and trimester exposure in 25.0% (2/8), confounder information in 62.5% (5/8) and loss of outcomes in 87.5% (7/8). Only half (4/8) reported loss of follow up, which ranged from 1% to 28.6% and three studies (37.5%) were population-based. The median number of patients in the studies was 1,186, with a range from 2 to 197,948 patients. The two largest studies included 197,948 and 11,010 patients. The median duration of studies from start to end was 36 months, with a range from 10 months to 89 months.

Conclusions: Up to almost the end of 2019, only small proportion of studies in Europe registered in the EU PAS Register were prospective pregnancy studies.
Many prospective cohort studies had deficiencies in providing important information and raise questions about their validity. Few were population-based and make comparators difficult to construct. Many performed poorly in terms of time and target sample size. Therefore, there is a need to address these deficiencies in future prospective pregnancy studies. The ´wide-net all registrants´ approach needs to be adapted with novel more efficient and methodologically more robust designs.