The study cohort included all MSM patients who were diagnosed with acute HCV infection at the Infectious Disease Department of the Hospices Civils de Lyon, France, and for whom NS5B sequencing was performed between January 2014 and December 2016. Data on drug use patterns and risk behaviours were collected. HCV isolated from 42 non-MSM, HIV-negative, male patients of similar age (genotype 1a: 30; genotype 4d: 12) were analysed by NS5B sequencing at the same time for phylogenetic analysis. This study was conducted in accordance with French ethics regulations. All patients gave their written inform consent to allow the use of their personal clinical data. The study was approved by the Ethics Committee of Hospices Civils de Lyon.

HCV infections were diagnosed during: i) regular follow-up visits for HIV-positive patients and ii) during visits for PrEP or for sexually transmitted infections (STI) for HIV-negative patients. The time of acute HCV infections was based on the first occurrence of: i) seroconversion for anti-HCV antibodies or ii) detection of HCV RNA in plasma before seroconversion or iii) concomitant detection of anti-HCV antibodies and HCV RNA in a context of elevated transaminases with a history of recent HCV exposure and negative previous serology and normal transaminases.

The period of contagiousness for HCV was defined by the presence of detectable HCV RNA in plasma during follow-up.

Bacterial sexually transmitted infections (syphilis, gonorrhoea, chlamydia infection) at the time of HCV diagnosis, the use of IV or nasal recreational drugs and rough sex practices such as sex in group, fisting or sharing sex toys were recorded from patients file.

HCV RNA was detected and quantified using the Abbott RealTime HCV assay (Abbott Molecular, Rungis, France).

The NS5B fragment of HCV was amplified between nucleotides 8256 and 8644 by RT-PCR as previously described [22] and sequenced using the Sanger method. Electrophoresis and data collection were performed on a GenomeLab^™ GeXP Genetic Analyzer (Beckman Coulter). Consensus sequences were assembled and analysed using the GenomeLab^™ sequence analysis software. The genotype of each sample was determined by comparing its sequence with HCV reference sequences [23] obtained from GenBank.

Phylogenetic trees for genotype 1a and genotype 4d sequences were constructed using MetaPIGA v.3.100 software. The Maximum Likelihood criterion was applied using the Hasegawa-Kishino-Yano (HKY) substitution model [24]. MetaPIGA calculations were stopped when the mean relative error of 10 consecutive consensus trees stayed below 5% using trees sampled every five generations. Consensus trees were visualized in Figtree v.1.4.3 (http://tree.bio.ed.ac.uk/software/figtree/). Reference strains for genotype 1a (Genbank accession numbers AF009606, EF407457, HQ850279, M62321 and M67463) and for genotype 4d (Genbank accession numbers DQ418786, FJ462437 and EU392172) were chosen for comparison.

Identification of transmission clusters was performed using Cluster Picker software with a support threshold of 0.9 and a genetic distance threshold of 0.045 [25]. Transmission “clusters” and “pairs” were defined as phylogenetic clades with n≥3 and n = 2 sequences respectively.

Considering the limited number of HIV-negative patients infected with HCV in the study, no statistical methods were applied to compare characteristics between HIV-positive and HIV-negative patients.

All HCV NS5B sequences isolated in MSM and non-MSM patients reported in this study were submitted to the GenBank database. The list of Genbank accession numbers for all sequences is provided in S1 Table.

The study population included 43 HIV-positive MSM and 6 HIV-negative MSM infected with HCV. Patient characteristics are described in Table 1. Whilst no comparison could be done due to the limited number of patients, HIV-negative patients appeared younger than HIV-positive patients (39.5 years vs 48 years). Most of the HIV-positive patients were on antiretroviral treatment, with an undetectable HIV-RNA and CD4 cells count over 500/mm³. Four out of 6 HIV-negative MSM were PrEP users but only 1 patient was receiving PrEP at the time of HCV diagnosis. For each group, reported rates of risk behaviour were high. Indeed, at least one risk factor (STI, rough sex or chemsex including nasal or intravenous drug administrations) was identified in 72% of HIV-positive individuals and 83% of HIV-negative individuals. The use of intravenous drugs was reported by 5 out of 6 HIV-negative patients (83%), but in only 40% of HIV-positive patients. Thirteen reinfections were observed, all in HIV-positive patients (eradication occurred following spontaneous cure in 2 patients and post-treatment in 11).

During the study period, 47 out of the 49 patients infected with HCV were treated with direct-acting antivirals, from which 42 achieved sustained virologic response at week 12, 4 achieved end of treatment response and 1 was reinfected before week 12. One was lost to follow-up and 1 patient is awaiting treatment.

Based on NS5B sequencing, HCV strains identified in MSM patients during the study period belonged to either genotype 1a (n = 29) or genotype 4d (n = 20). In 78% of cases, sequencing was performed on a sample collected within 6 months after the estimated date of infection, i.e. during or near the acute phase of hepatitis C. In 4% of cases, the sample analysed was collected between 6 months and one year after the estimated date of infection and in 18% of cases, more than one year after the estimated date of infection.

Phylogenetic trees were constructed with sequences from MSM and non-MSM patients (Fig 1). For genotype 1a, the 29 sequences from MSM clustered together with sequences from non-MSM patients in different branches of the phylogenetic tree. For genotype 4d, all 20 sequences from MSM clustered separately from sequences of non-MSM origin.

For genotype 1a, we identified three transmission clusters in MSM with eight isolates each and one pair. Notably, in cluster 2, seven of eight NS5B sequences had 100% homology. For genotype 4d, we identified one transmission cluster in MSM with 13 isolates and two pairs. Among all four clusters of more than 2 cases identified, clusters 1, 2 and 4 included isolates from both HIV-positive and HIV-negative, patients. Among the three pairs, pair 2 (genotype 4d) also included one HIV-positive and one HIV-negative patient.

Based on the estimated time of HCV infection and the time of the last detectable HCV viral load during follow-up, we evaluated the periods of contagiousness for each member of the transmission clusters (Fig 2). In each cluster, individual periods of contagiousness overlapped, supporting the cluster identification. Each cluster mixing HIV-positive and HIV-negative MSM, started with several HIV-positive patients infected with HCV before the first HIV-negative patient was identified.

Characteristics of the patients included in the 4 clusters identified are presented in Table 2. Cluster 2 was characterized by a shorter interval between infections, the 8 acute HCV infections being diagnosed in less than 10 months (Fig 2). Additionally, patients in cluster 2 appeared younger than patients from the other clusters, and 5 out of 8 where using drugs, a higher proportion than in the other clusters. This period of contagiousness, together with the very high homology between the NS5B sequences within this cluster (Fig 1A), and with demographic and behavioural characteristics, suggests that direct transmission of HCV between members of this cluster may have occurred, presumably through drug use. In cluster 4, one HIV-positive patient was infected and remained viremic several years before other grouped cases were identified in a very short period, suggesting that this patient could be the source of the following cases.