P36-17. Phylogenetic reconstruction of hepatitis C virus transmission in Southern Sweden (2007-2015)

Malik Sallam[1], Birgitta G Holmgren[2] Anna Söderlund Strand[2] Gülşen Özkaya Şahin[1,2] Mattias Waldeck[3] Niclas Winqvist[3] Marianne Alanko Blomé[4] Mikael Ingman[1] Joakim Esbjörnsson[5] Anders Widell[1] and Patrik Medstrand[1]
Affiliates: [1]Lund University, Faculty of Medicine, Department of Translational Medicine, Malmö, Sweden. [2]Department of Clinical Microbiology, Laboratory Medicine, Skåne University Hospital, Lund, Sweden. [3]Regional Office of Communicable Disease Control and Prevention, Skåne County, Malmö, Sweden. [4]Infectious Disease Unit, Division of Clinical Sciences, Lund University, Malmö, Sweden. [5]Lund University, Faculty of Medicine, Department of Laboratory Medicine, Lund, Sweden.

Background
Hepatitis C virus (HCV) infection is a major cause of hepatic-related morbidity and mortality worldwide. In Sweden, the sero-prevalence of HCV is estimated to be less than 0.5%, with genotypes 1 and 3 being the most prevalent and injection drug use (IDU) as the most common route of HCV acquisition. Molecular epidemiology studies of HCV are useful to gain in-depth knowledge of timing and patterns of virus spread. The aim of the current study was to characterize the transmission dynamics of HCV in Southern Sweden.

Methods
We analysed 3395 partial HCV NS5B sequences (339 bases) collected in Southern Sweden (2007-2015) together with 1056 similar GenBank sequences. Subtyping was done using the COMET-HCV online tool. Transmission clusters were determined and dated by maximum likelihood and Bayesian phylogenetics. Variable association with phylogenetic clustering was tested using multinomial regression analysis.

Results
The most common subtypes in the study were: 1a (n=1306, 39%), 3a (n=1286, 38%), 1b (n=342, 10%) and 2b (n=340, 10%). Infection with subtypes 1a and 3a was associated with younger age compared to all other subtypes (p<0.001, linear-by-linear test for association). Subtypes 1a, 2b and 3a were associated with IDU compared to subtype 1b (p<0.001 for all comparisons, Fisher’s exact test [FET]), whereas subtype 1b was associated with history of blood transfusion (p 45, odds ratio [OR] = 1.57, 95% CI: 1.35-1.83%, p<0.001). Clustering was less associated with infection by subtype 1a (compared to subtype 3a, OR = 0.69, 95% CI: 0.58-0.81%, p<0.001) and subtype 1b (compared to subtype 3a, OR = 0.45, 95% CI: 0.34-0.60%, p<0.001), while clustering among subtype 2b was not different compared to subtype 3a (p=0.301). For the risk group categories, higher clustering was associated with IDU (compared to missing risk factor data, OR = 1.23, 95% CI: 1.02-1.47%, p=0.028) and sexual (compared to missing risk factor data, OR = 1.65, 95% CI: 1.06-2.57%, p=0.028). The time to most recent common ancestor (tMRCA) of the oldest subtype 1a cluster dated back to 1980 (95% HPD: 1960-1994). The tMRCA of the oldest subtype 3a cluster dated back to 1951 (95% HPD: 1930-1976). Five out of the fifteen large clusters (33%) had a median tMRCA between 1967 and 1975.

Conclusions
We characterized the molecular epidemiology of HCV in Southern Sweden with identification of subtypes 1a and 3a as the major HCV subtypes predominating the epidemic. IDU remains the major risk factor for HCV acquisition with substantial proportion of infections taking place as a result of local spread. The results of our study can be helpful to provide a well-informed framework for focused infection control measures.