Josephine van Dongen

8 Summary and general discussion 191 Changing rotavirus epidemiology One of the possible explanations for the low vaccine effectiveness is the change in rotavirus epidemiology. Globally, the introduction of rotavirus vaccines has substantially changed the epidemiology. 26–28 Fur thermore, in the Netherlands and other non-vaccinating countries epidemiology of rotavirus was altered as well. In general, rotavirus epidemiology is influenced by vaccination, climate and temperature changes, bir thrates, previous rotavirus seasons and size of susceptible populations. 29–31 Changes observed in recent years in both rotavirus vaccinating and non-vaccinating countries include a shift from annual high endemic peaks towards biennial, lower and later peaks. 28,32 Strain diversity changed from primarily G1P8 genotype dominance, to alternating high prevalence of G2P4, G3P8, G4P8 and G9P8 in the more recent years. 19–21 This change in circulating genotypes is thought to be the result of vaccine-induced selective pressure. 33 Rotavirus vaccination is less well effective towards (par tly) heterotypic strains, that were circulating more after 2014. During the RIVAR study period less than 14% of rotavirus detections by national surveillance sentinel laboratories were tested as G1P8 strain, while the fully heterotypic G2P4 strain became more prevalent with 12%. 19,34,35 This altered landscape of rotavirus strain distribution may have contributed to our low vaccine effectiveness estimates ( table 1 ). While early on, rotavirus vaccine effectiveness estimates against par tly heterotypic strains were 74-87% and 85-94% against fully heterotypic strains (compared to 89-95% against homotypic strains). 36–38 Recent estimates from the era of altered strain diversity are not available. In general, there is a lack of research on rotavirus vaccine performance or effectiveness in more recent years. Within five years after the first rotavirus vaccines were globally introduced, 24 primary studies repor ted estimates of vaccine effectiveness in different geographic regions. In the window between five to ten years after vaccine introduction only eight studies repor ted on vaccine effectiveness and no studies have yet been published that cover the period beyond ten years post-introduction. However, in light of the substantial changes in rotavirus strain distribution, continuedmonitoring of vaccine effectiveness is warranted. For influenza, a European Influenza Surveillance Scheme (EISS) exists. Strain-specific and overall vaccine effectiveness repor ts are published for each influenza season based on the available clinical and epidemiological data. 39–41 This aids in monitoring the virus, mismatch with vaccine strains and circulating strains in Europe. For rotavirus, the National Respiratory and Enteric Virus Surveillance System (NREVSS) and New Vaccine Surveillance Network (NVSN) in Nor th America provided repor ts on clinical and epidemiological data of rotavirus until 2016. 25,42 In Europe, EuroRotanet (a network of collaborating laboratories) 43 already collects information on rotavirus types co-circulating in Europe. Integration of these data in vaccine performance studies seems feasible. Active surveillance of rotavirus epidemiology should be combined with monitoring vaccine effectiveness to better understand the impact of strain replacement on vaccine performance and burden of disease.