A8 The epidemiology and evolution of influenza A/H1N1 and A/H3N2 virus from 2010 to 2015, in Ho Chi Minh City, Vietnam

waves have been detected in China. Live poultry market (LPMs) exposure is regarded as a major risk of H7N9 virus infection. However, despite strict interventions implemented in the epicenter cities during each outbreak, reports indicate a gradual nationwide spread of the virus. In this study, the impact of LPMs interventions in virus persistence and transmission in the province of Guangdong, the epicenter of the second and third epidemic waves, was assessed by genetic and spatial analyses. We analyzed the temporal and spatial distribution of all of H7N9 clinical cases reported in Guangdong from August 2013 to March 2015. Viral isolation and genome sequencing were performed for 81 of all 182 clinical H7N9 infection cases as well as for 65 H9N2 viruses from live poultry and LPMs environments, from 16 prefecture-level cities during the three epidemic waves. Molecular clock and spatial phylogenetic analyses were applied to trace virus persistence and transmission across epidemic waves, genetic segments and geographic regions. Temporary LPMs closure in epicenter cities reduced by 35% the number of clinical cases from 110 in the second wave to 72 during the third wave. However, eastern Guangdong, which reported few cases of H7N9 infection in the second wave, became the new epicenter of H7N9 outbreak during the third wave. Genetic analyses of the virus external genes showed, with strong support, that the third wave outbreaks in central and eastern Guangdong are the result of virus persistence rather than virus importation from elsewhere. Analyses of the internal genes of H7N9 virus from the third wave sampled in Guangdong indicate additional reassortment events with virus lineages from central and eastern China. Guangdong province reported the highest number of H7N9 cases during the second epidemic wave. In response to the outbreak, LPMs closure in epicenter Guangdong cities has been implemented. However, our study shows that the LPM closure in epicenter Guangdong cities during the second wave was insufficient. We find that the viruses responsible for the third wave outbreak in Guangdong descend from the virus circulating in the second wave in the same region. In addition, the newly identified reassortment events between Guangdong H7N9 internal genes and those from strains circulating in other provinces suggest that LPMs closure without a ban of poultry trade may contribute to the increase of H7N9 virus genetic diversity through reassortment with imported strains. Importantly, that a new outbreak in eastern Guangdong during the third wave was caused by a rarely detected virus that was circulating in the same region during the second wave, raising an alarm that the cities outside of H7N9 outbreak epicenters should not be neglected from the list of LPM interventions.