If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Internal Medicine 1, University Hospital, Goethe University, Frankfurt, GermanyMedizinische Klinik II, St. Josefs-Hospital, Wiesbaden, Germany
Recently acquired HCV infections primarily occur in HIV-coinfected MSM and PWID
•
Over 10 years, an overall decline of recently acquired HCV infections and a certain HCV genotype pattern were observed
•
Country-mixing MSM-specific clustering with a shift towards HCV GT4d infections in recent years could be identified
•
Ongoing risk of repeated HCV transmission in MSM due to increase of reinfection incidence and low spontaneous clearance rates
•
Our findings underscore the importance of constructive preventive strategies and early treatment initiation in MSM to achieve HCV elimination
Abstract
Background and aims
Ongoing transmission of hepatitis C virus (HCV) infections is associated with risk factors like drug injection, needle stick injuries and men who have sex with men (MSM). Ways of transmission, the course of acute infection, changes of virologic features, and incidence over time are not well known.
Methods
Over a period of 10 years, n=161 patients with recently acquired HCV infection (RAHC) (median follow-up 6.8 years) were prospectively enrolled. NS5B sequencing was performed to re-evaluate the HCV genotype and for phylogenetic analyses.
Results
Patients with RAHC were mainly male (92.5%), MSM (90.1%) and HIV-coinfected (86.3%). Transmission risk factors for MSM and non-MSM were sexual risk behaviour (100 and 6.3%, respectively), injection drug use (9.7 and 37.5%, respectively) and nasal drug use (15.2 and 0%, respectively). Spontaneous and interferon- or direct-acting antiviral (DAA)-based clearance rates were 13.6%, 84.3% and 93.4%, respectively. Mean RAHC declined from 19.8 in the first to 13.2 in the past 5 study years. While the majority of infections was caused by HCV-1a, the frequency of HCV-4d and slightly HCV-3a increased over time. No relevant clustering of HCV isolates was observed in non-MSM. However, 45% of HCV-1a and 100% of HCV-4d MSM-cases clustered with MSM-isolates from other countries. Travel-associated infections were supported by personal data in a MSM subgroup. No international clustering was detected in MSM with HCV-1b or HCV-3a.
Conclusions
RAHC were mainly diagnosed in MSM-HIV-coinfected patients and were associated with sexual risk behaviour. Spontaneous clearance rates were low and phylogenetic clusters were observed in the majority of patients.
Lay summary
We evaluated the occurrence and transmission of recently acquired HCV infections (RAHC) over a period of ten years. Our data demonstrate that the presence of RAHC was mainly found in HIV-coinfected MSM with internationally connected transmission networks being observed in the majority of patients. Spontaneous clearance rates were low and reinfection rates increased mainly driven by a small subset of MSM patients with high-risk behaviour.
CS and JD designed the study concept. Data collection was performed by JD, CS, CG, TL, PG, CS, GK, MB, KHP, FF, GD, AM, NW and SZ. Analysis and interpretation of data was performed by CG, JD, CS, LF, NB, NF and EH. Draft of the manuscript was performed by CG, JD, CS, LF and NB. CS, SZ, NW and NB provided the critical revision of the manuscript.
Conflicts of Interest
Christiana Graf: Speaking and/or consulting fees from AbbVie and Gilead.
Lara Fuhrmann: no conflicts to disclose.
Thomas Lutz: no conflicts to disclose.
Christoph Stephan: no conflicts to disclose.
Gaby Knecht: no conflicts to disclose.
Peter Gute: no conflicts to disclose.
Markus Bickel: no conflicts to disclose.
Kai-Henrik Peiffer: no conflicts to disclose.
Fabian Finkelmeier: received travel support from Abbvie and Novartis, and speaker fees from Abbvie and MSD. Consulting Fees from Fresenius and Ibsen.
Georg Dultz: Speaking and/or consulting fees from AbbVie and Gilead.
Antonia Mondorf: no conflicts to disclose.
Nils Wetzstein: no conflicts to disclose.
Natalie Filmann: no conflicts to disclose.
Eva Herrmann: no conflicts to disclose.
Stefan Zeuzem: Speaking and/or consulting fees: Abbvie, BMS, Gilead, Janssen,
Merck/MSD.
Niko Beerenwinkel: no conflicts to disclose.
Julia Dietz: Research support from Gilead.
Christoph Sarrazin: Speaking and/or consulting fees: Abbvie, BMS, Gilead, Merck/MSD. Research support: Abbvie, Gilead.
Grantsupport: LF was funded by European Union’s Horizon 2020 research and innovation program, under the Marie Skłodowska-Curie Actions Innovative Training Networks grant agreement no. 955974 (VIROINF).
Introduction
Ongoing transmission of hepatitis C virus (HCV) infection is associated with high numbers of patients with acute hepatitis C globally[
]. However, over the last two decades, a persistent epidemic of recently acquired HCV infections (RAHC) has also been observed in HIV-positive men who have sex with men (MSM) in several metropolitan areas worldwide[
]. In patients, who initiated antiviral treatment with direct-acting antivirals (DAAs) during the acute HCV infection phase, high sustained virological response (SVR) rates have been achieved, even after using a shortened treatment duration[
Ledipasvir plus sofosbuvir fixed-dose combination for 6 weeks in patients with acute hepatitis C virus genotype 1 monoinfection (HepNet Acute HCV IV): an open-label, single-arm, phase 2 study.
Reinfection With the Hepatitis C Virus in Men Who Have Sex With Men After Successful Treatment With Direct-acting Antivirals in Germany: Current Incidence Rates, Compared With Rates During the Interferon Era.
]. Comprehensive data on HCV reinfection incidence from the DAA era are still scarce, but first longitudinal reports provide evidence that despite high treatment uptake and excellent treatment outcomes, HCV reinfection rates in HIV-positive MSM continue to be comparably high[
Reinfection With the Hepatitis C Virus in Men Who Have Sex With Men After Successful Treatment With Direct-acting Antivirals in Germany: Current Incidence Rates, Compared With Rates During the Interferon Era.
]. Therefore, questions remain about what impact DAA availability will have on the epidemic and whether elimination will be achieved in the short term as intended by the World Health Organization (WHO) agenda. One of the major reasons for the ongoing epidemic in MSM may be their maintained risk behaviour consisting of traumatic sexual practices, recreational intravenous and non-intravenous drug use, recently referred to as chemsex [
Reinfection With the Hepatitis C Virus in Men Who Have Sex With Men After Successful Treatment With Direct-acting Antivirals in Germany: Current Incidence Rates, Compared With Rates During the Interferon Era.
]. Several studies showed a separate phylogenetic clustering of HCV strains in IVDA patients, sexual networks and in HIV-coinfected patients, implicating different transmission networks[
]. Interestingly, MSM-specific transmission networks were not restricted to single countries, but largely overlapped in an interconnected international transmission network [
Overall, there is a lack of data concerning the investigation of patients with RAHC over extended time periods. In the present study, we analysed patients with recently acuired hepatitis C presented at large liver and infectiology centres in Frankfurt am Main, Germany, over a period of up to 10 years. In addition to epidemiological and clinical parameters, virologic outcome and detailed phylogenetic analyses were conducted in order to identify possible HCV transmission clusters.
Methods
Study population
All patients aged 18 and over, who were diagnosed with an acute hepatitis C infection at the participating centers (Liver- and HIV-Centers at Goethe-University Hospital and Infektiologikum, Frankfurt, Germany), were prospectively enrolled between January 2009 and December 2019 and followed up until October 2020. Acute hepatitis C was diagnosed if one of the following criteria was met: (I) documented HCV antibody seroconversion within 12 months and a positive HCV RNA, (II) positive HCV RNA and a documented prior negative HCV antibody or HCV RNA test within 12 months, (III) positive HCV RNA and an increase in liver transaminases >5 times the upper limit of normal (ULN) and normal transaminases the year before. Sustained virologic response (SVR) was defined by a negative HCV PCR at least 12 weeks after the end of treatment. The definition of a spontaneous hepatitis C clearance is described elsewhere[
]. HCV reinfection was defined by a detectable positive HCV RNA level after documented HCV RNA negativity, by a switch in HCV genotype or by phylogenetic analysis suggesting HCV reinfection following SVR. Serum and EDTA samples were collected at the time of diagnosis of an acute hepatitis C infection (baseline) and longitudinally at monthly intervals during the follow-up routine medical visits. Collection of samples was stopped as soon as an SVR or a spontaneous clearance were documented. Sera from patients with chronic hepatitis C infection, which were used as a control cohort, were obtained from the European resistance database, which was established based on a non-interventional study [
]. Further screens were conducted based on individual risk behaviour and in cases of a suspected HCV coinfection. Monitoring of transaminases in HIV-positive individuals was performed every 3-12 months. Demographic and clinical data were recorded at baseline and at each follow-up visit. Moreover, the history of sexually transmitted diseases within 6 months prior to the onset of RAHC as well as the history of a prior hepatitis C infection were assessed by medical chart review. Patients were followed up to the last medical visit with available biochemical parameters at their treatment centre. All participants provided written informed consent before inclusion in the study. Ethical approval of this study as well as for usage of patient blood samples for research purposes were obtained from the local ethics committee at the University Hospital Frankfurt, Germany (ethics committee reference number: 58-09). All investigations were performed in accordance with the Declaration of Helsinki.
Statistical Analyses
Values of demographic and clinical characteristics are expressed as mean (range) or median (interquartile range; IQR). Baseline comparisons between groups were performed using a t-test (for variables with an assumed Gaussian distribution) or a Mann-Whitney U test (for non-normally distributed variables).
Incidence rates of a first HCV infection and of a reinfection during follow-up (FU) of the study period were calculated using person time of observation. Details on the calculation of incidence rates and the corresponding denominators can be found in the Supplementary Material. Confidence intervals (CI) for incidence rates (reinfections and new infections) were assessed using a Poisson model. A p-value of ≤ 0.05 was considered statistically significant. Statistical analysis was performed using SPSS Version 22 and R (Version 4.0.4).
NS5B amplification and sequencing analyses
Extraction of HCV RNA and cDNA synthesis were conducted as described previously [
]. Further details on the PCRs and sequencing analyses can be found in the Supplementary Methods.
Phylogenetic analyses
We compared the NS5B sequences from 161 patients with RAHC in the current study to a set of 369 published sequences obtained from patients with RAHC and a sexual transmission mode (MSM). The latter sequences are available from GenBank (https://www.ncbi.nlm.nih.gov/genbank/) [
Changing Trends in International Versus Domestic HCV Transmission in HIV-Positive Men Who Have Sex With Men: A Perspective for the Direct-Acting Antiviral Scale-Up Era.
]. As a control, 61 sequences from patients with chronic hepatitis C infection and non-sexual transmission were included in the analysis. For each genotype (GT), the maximum likelihood phylogenetic tree with 100 bootstrap replicates was inferred using Iqtree version 2.0.3 [
]. The best-fit substitution models for each genotype were determined according to the BIC (GT1a: TIM2e+I+G4, GT1b: K2P+I+G4, GT3a: K2P+I+G4, GT4d: K2P+I+G4) by the built-in method ModelFinder [
]. Further details on the construction of phylogenetic trees for each genotype, alignment, derivation of phylogenetic clusters and accession numbers from published sequences are available in the Supplementary Information.
Phylogenetic analysis of reinfections
Phylogenetic analyses to detect potential reinfections were conducted for 13 patients with recently acquired HCV GT1a infection and 2 patients with HCV GT4d infection. Samples from patients with chronic hepatitis C and a documented relapse to a DAA-based treatment were used as controls. Relapse was defined as HCV RNA negativity at the end of DAA treatment (EOT) followed by detectable HCV RNA within 12 weeks after EOT. Control patient samples were obtained from the European Resistance Database and sequence data at baseline and at the time of relapse were used for phylogenetic analyses (n=4, HCV GT1a; n=1, HCV GT4d) [
]. The baseline and posttreatment consensus sequences for subtype 1a and 4d were aligned separately using BioEdit version 7.2.5. For each subtype, Iqtree version 2.0.3 was used to infer maximum likelihood phylogenetic trees with 100 bootstrap replicates [
Demographic and clinical characteristics of the study population are shown in Table 1. Between January 2009 and December 2019, a total of 161 patients diagnosed with RAHC were included in the study, of whom 145 (90.1%) were MSM. The majority of patients were Caucasian (93.2%), 1.9% were Asian and 2.5% were Hispanic. At the time of study inclusion, the median age was 43 (interquartile range (IQR), 38-48 years), and 149 (92.5%) participants were male. Symptomatic disease was observed in a total of 59 patients (36.7%); 20 patients had jaundice, and 9 had a bilirubin level >10mg/dl. Furthermore, 24 patients reported fatigue and 13 patients had abdominal pain at the time of HCV diagnosis. Median ALT level at baseline was 463 IU/ml. Median maximum of ALT and bilirubin values was 4754 IU/ml and 23 mg/dl respectively. Median HCV RNA viral load at baseline was 6.1 log10 IU/ml (range 1.2-9.6). No patient experienced fulminant disease with progression to acute liver failure. The overall prevalence of an active or previous HBV-coinfection was 1.9% and 31.7%, respectively. Among 34 patients with available information, none of them suffered from an active HEV-coinfection during follow-up period and 2 patients were tested positive for a resolved HEV infection. Moreover, 139 (86.3%) individuals were living with HIV, of whom 134 received HAART. The median CD4 cell count at the time of study inclusion was 590/μl (IQR, 432-766), and 63.3% of HIV-HCV coinfected patients had an HIV viral load less than 50 copies/ml. For 17 individuals, at least one prior infection, either effectively treated or spontaneously cleared, was documented before included in the study with the actual HCV episode. With regard to the last available laboratory measurements, the total follow-up time was 1035 patient-years and the median follow-up time for each patient was 82 months (range, 1-142).
Table 1Baseline characteristics of MSM and Non-MSM diagnosed with acute Hepatitis C virus infection between 2008 and 2019 (n=161).
Characteristics
MSM (n=145)
Non-MSM (n=16)
Median age (years)
43 (39 - 47)
32 (25 – 51)
Male sex, n (%)
145 (100)
4 (25)
Caucasian ethnicity
136 (93.8)
14 (87.5)
BMI (kg/m2), median (range)
24.0 (21 – 26)
20.8 (20 – 23)
HIV-coinfection, n (%)
138 (95.2)
1 (6.3)
CDC stage C, n (%)
15 (10.9)
0 (0)
HIV RNA viral load (log10) (IU/mL), median (range)
Twenty-two episodes (13.6%) resulted in a spontaneous clearance of RAHC. Six patients were lost to follow-up before treatment initiation. HCV treatment was initiated in 131 patients (81.4%) with a median time from recently acquired HCV diagnosis to treatment initiation of 5 months (range, 0-79). The overall ITT-SVR rate following interferon-based and DAA-based therapy was 84.3% (59/70) and 93.4% (57/61), respectively.
Annual recently acquired HCV diagnoses and prevalence of HCV genotypes in MSM and Non-MSM
Overall, the highest number of HCV infections was found in the years 2009 to 2014, which can be assigned to the interferon era (Figure 1A). With regard to the distribution of the HCV genotypes, we observed different frequencies during the observation period. In the years 2009 to 2017, HCV GT1a was always the most common (Figure 1B). However, GT1a decreased in 2018 and 2019, respectively. The proportion of GT4d cases increased over the years. While no GT4d cases were diagnosed at all in 2013, the proportion rose to 40% in 2018 and 2019. There was also a trend towards more GT3a cases in 2018 and 2019, while for GT1b and GT2 only individual cases were detected between 2009 and 2014.
Figure 1Characteristics of HCV-mono- and HCV/HIV-coinfected patients with recently acquired hepatitis C infection. Annual recently acquired hepatitis C cases (A) and prevalence of different HCV geno- and subtypes (B) over the study time period.
In total, 25 HCV reinfections occurred in 22 MSM resulting overall in 186 cases of HCV infection (Table 2). Of note, no HCV reinfection was observed in non-MSM patients with RAHC.
Reinfection incidence was 2.7 per 100 person-years (PY; 95% CI, 1.39-3.05) and the median duration to reinfection was 17 months (IQR, 0-130 months). Patients with reinfections were exclusively HIV-positive MSM and were significantly younger than patients without reinfection (p=0.03).
Of the 22 patients with an HCV reinfection, 14 (64%) had a different HCV GT at the time of reinfection compared to the first infection (Table 2). In the remaining 8 patients, no HCV GT switch could be detected (all HCV GT1a). However, HCV reinfection was suspected in these cases due to clinical data, since sexual risk behaviour was documented at the time of suspected reinfection. In additional 7 cases in HCV GT 1a and 4d with detectable HCV RNA after previous achievement of HCV RNA negativity 12 weeks after end of treatment (SVR 12), late virologic relapse was supposed, since the same HCV isolate could but be detected as before treatment. However, reinfection with a different HCV isolate could not be ruled out.
In order to distinguish HCV reinfections from a virologic relapse in patients with GT1a and GT4d, we conducted phylogenetic analyses and used samples of 13 patients with RAHC and patient samples from the European resistance database with a proven relapse to DAA treatment as a control for the genetic drift over time.
Baseline and relapse sequences of these controls were each located on a single branch in the phylogenetic tree (Figure 2A and 2B). In all 7 cases with suspected virologic relapse, minimal genetic changes between baseline and posttreatment NS5B consensus sequences could be detected, indicated by clustering and separation from other sequences by a supported branch (aHCV8, aHCV10, aHCV11, aHCV12 and aHCV13, aHCV14, aHCV15). Thus, infection with the same HCV isolate in these patients is plausible (relapse or intermediate HCV RNA negative course). In further 6 cases, phylogenetic analyses of NS5B sequences showed that the baseline sequence and posttreatment sequence were located on different branches and did not cluster with each other (aHCV1, aHCV3, aHCV5, aHCV6, aHCV7 and aHCV9). These results strongly suggest that patients were reinfected with a new HCV strain. In two further patients we assumed HCV reinfection due to the fact that the patients were sexual partners and an infection event was documented, although all sequences were located on the same branch (aHCV 2 and aHCV4).
Figure 2Differentiation of viral relapse versus HCV reinfection by phylogenetic trees in acute hepatitis C infections with HCV Genotype 1a (A) and HCV Genotype 4d (B). Baseline and posttreatment sequences are marked by a circle and triangle each in the same colour, respectively. Control samples are shown in white and marked as BL (Baseline) and REL (Relapse). Abbreviations: aHCV, acute hepatitis C infection; BL, baseline; Cont., control; REL, relapse.
In order to study the incidence of HCV infections due to different treatment levels over a period of ten years, we compared the cumulative incidence for primary HCV infections and HCV reinfections among MSM patients in the second generation DAA era (primary HCV infections: 2017-2019; HCV reinfections: 2017-2020) with the first generation DAA era (2013-2017) and interferon era (2009-2013; Table 3 and Table 4). The incidence rate for a first HCV infection substantially declined in the period between 2017 and 2019 (3.3/1000PY) compared to the DAA era between 2013 and 2017 (5.9/1000 PY) and to the interferon era (2009–2012; 8.3/1000 PY; Table 3). Conversely, the incidence of HCV reinfections remained a frequent finding among MSM during the whole study period and increased slightly from 1.9/100 PY to 2.8/100 PY over the 10 years study period (Table 4).
Table 3Incidence of first HCV infections over time.
Phylogenetic analyses to compare recently acquired hepatitis C in MSM versus non-MSM
In order to investigate whether larger HCV infection networks exist in patients with RAHC, we conducted phylogenetic analyses for each HCV genotype. HCV NS5B sequences of Frankfurt MSM patients and non-MSM patients with RAHC were compared with published NS5B reference sequences obtained from MSM. Moreover, we generated sequences from patients with chronic hepatitis C and other non-sexual transmission routes (e.g., blood transfusion, needle stick injury or IVDA) that served as controls (see Methods).
Phylogenetic trees were constructed for each HCV GT separately. Phylogenetic analyses revealed 44 monophyletic clusters of MSM-specific strains containing a total of 377 MSM isolates (Figure 3). Moreover, 139 singleton MSM sequences that were not closely related to any other strain in this study population were identified. The 44 MSM clusters ranged in size from 2 to 99 sequences and consisted predominantly of HCV GT1a und 4d.
Figure 3Phylogenetic trees based on HCV NS5B sequencing analyses conducted at the time point of acute hepatitis C diagnosis in patients with HCV Genotype 1a (A), HCV Genotype 1b (B) and HCV Genotype 3a (C), HCV Gentoype 4d (D). Clusters are illustrated by triangles, numbers in the triangles represent the number of sequences in each cluster. Country of origin is shaded (): non-MSM, Frankfurt, Germany (), MSM, Frankfurt, Germany (), MSM (published sequences), Germany (), MSM (published sequences), United Kingdom (), MSM (published sequences), Netherlands (), MSM (published sequences), Switzerland (), MSM (published sequences), Spain (), MSM (published sequences), USA (), MSM (published sequences), Canada (), MSM (published sequences), Australia (), MSM (published sequences), France, Abbreviations: GT, genotype; HCV, hepatitis C virus.
Phylogenetic analyses in HCV GT1a-infected patients
The GT 1a phylogeny is illustrated in Figure 3A and reveals a total of 35 transmission clusters, of which 34 were MSM-specific. Infection with GT 1a was the most prevalent among MSM as well as among non-MSM patients. Interestingly, all but one of the GT1a sequences obtained from non-MSM patients appeared to be unrelated to any of the clusters. In contrast, the majority of GT1a infections in the Frankfurt MSM individuals (n=86/126, 68%) could be assigned to 18 different MSM-specific clusters. Ten of these 18 clusters exclusively contained sequences from the Frankfurt MSM patients and German reference MSM sequences from Bonn, but no sequences from other countries (clusters 1, 3, 4, 9, 10, 13, 19, 21, 27, 32; Figure 3A; Appendix Figure 1, Appendix Table 1.1 and 1.2). The remaining eight transmission clusters contained published MSM sequences from different European countries in addition to Frankfurt MSM sequences and are therefore considered as country-mixing clusters (clusters 5, 6, 15, 17, 18, 20, 31 and 34; Appendix Table 1.1 and 1.2). Overall, 45% (n=57/126) of the Frankfurt MSM GT1a sequences were assigned to these 8 international clusters.
Four of these international clusters contained sequences from Germany and one further European country (clusters 5 and 31, the Netherlands; cluster 20, United Kingdom; cluster 24, Switzerland). The remaining four clusters consisted of sequences obtained from German MSM as well as sequences from further two (cluster 6, the Netherlands and Switzerland; cluster 15, the Netherlands and United Kingdom) or three European countries (clusters 17 and 18: the Netherlands; United Kingdom and Switzerland). All country-mixing clusters contained only sequences obtained from European MSM. Australian MSM isolates clustered separately and were never part of an HCV GT1a cluster with European sequences (Figure 3A).
Of the fifteen sequences of patients with chronic hepatitis C and other HCV transmission routes, only one control sequence of a patient with IVDA transmission clustered with one MSM sequence obtained from the Netherlands (cHCV.2679.2015.IVDA, cluster 11). Besides that, no clustering of chronic HCV sequences with those of patients with RAHC was observed (Figure 3A).
Phylogenetic analyses of HCV GT1b-infected cases
Neither the one Frankfurt MSM sequence nor the three non-MSM sequences clustered with any reference sequence. In contrast, a clustering of published MSM sequences obtained from the Netherlands and from the UK was observed (clusters 1-3). Control sequences of our study obtained from patients with chronic hepatitis C and other transmission routes such as blood transfusion or IVDA could also not be assigned to any cluster (Appendix Table 2.1 and 2.2).
Phylogenetic analyses in HCV GT3a-infected patients
Overall, 83% (n=5/6) of Frankfurt HCV-GT3a MSM cases were assigned to a transmission cluster. Interestingly, Frankfurt MSM sequences showed a separate clustering in one single cluster, which did not contain reference sequences obtained from other European countries (cluster 2; Figure 3C; Appendix Figure 3, Appendix Table 3.1 and 3.2). In addition to the Frankfurt MSM cluster, we detected three additional HCV GT3a clusters, which exclusively consisted of published MSM sequences obtained from other European countries (Appendix Table 3.2). Sequences from non-MSM patients with RAHC (aHCV.1.2009.a, aHCV.242.2011.a, aHCV.394.2012.a) as well as sequences obtained from chronic hepatitis C patients were not assigned to any cluster (Figure 3C).
Phylogenetic analyses in HCV GT4d-infected patients
Strikingly, all HCV GT4d sequences (100%; n=27/27) obtained from Frankfurt MSM patients clustered with published MSM-isolates from other countries (Switzerland, The Netherlands, UK, Canada).
The Frankfurt MSM sequences were part of two transmission clusters (Figure 3D; Appendix Figure 4; Appendix Table 4.1 and Table 4.2): sequences from two Frankfurt MSM patients represented cluster 1, and further 25 Frankfurt MSM sequences (93%, n=25/27) were part of a very large country-mixing cluster together with 73 European (the Netherlands, Spain, Canada, France, Germany (Bonn) and the United Kingdom) and 13 Canadian sequences (cluster 2). The third cluster consisted exclusively of English MSM reference strains (cluster 3) (Appendix 4.1 and 4.2).
As previously observed for the other HCV GTs, control sequences obtained from patients with chronic hepatitis C without a sexual transmission route were not assigned to any cluster, except for one (cHCV.6036.2017.IVDA, cluster 2).
Clinical data supporting phylogenetic analysis results
Phylogenetic analysis results could be confirmed in a subgroup of patients with additional information on social, transmission and travel history (Appendix Tables 5-7). In 15 cases, partnerships or sexual contacts within our MSM cohort could be identified within the clusters. In country-mixing clusters, 8 Frankfurt MSM reported to have frequent stays abroad in order to carry out business activities (such as flight attendants) and 2 MSM had a partner working abroad in countries they clustered with. Moreover, 10 MSM could be identified, who lived or stayed in a foreign country or national region that was part of the respective cluster. Concerning transmission histories, in country-mixing clusters 7 cases could be identified, in which sexual contacts of our MSM cohort with MSM from clustering countries were documented. Finally, in 8 HCV cases, at least two infection events within one cluster corresponded in terms of time, transmission route and geographic region.
Discussion
Excellent treatment results in recently acquired and chronic HCV infection have prompted the WHO to incorporate HCV elimination until 2030 in its agenda. However, the burden of disease, especially attributed to RAHC, is still high among key at-risk populations such as PWID and HIV-infected MSM. Thus, we aimed to investigate whether high-risk networks exist that facilitate an ongoing HCV transmission over a long time period of ten years and what impact DAA availability may have on the HCV epidemic among MSM.
In our ten years observational study, we found a steady and substantial decline in RAHC incidence rates among HIV- positive MSM from 8.3 per 1000 PY in the interferon to 3.3 per 1000 PY in the DAA era which is in line with another study[
Decline in Hepatitis C Virus (HCV) Incidence in Men Who Have Sex With Men Living With Human Immunodeficiency Virus: Progress to HCV Microelimination in the United Kingdom?.
]. Our study cohort was predominantly characterized by HIV-coinfected MSM, who reported high-risk sexual behavior and recreational drug use. The minority of patients were HIV-negative non-MSM individuals with RAHC with one third of them reporting IVDA followed by other transmission routes such as transfusion or needlestick injury.
The decline in incidence rates in conjunction with the high SVR rates observed in this cohort suggests that scale-up of DAAs together with behavioral interventions may have the potential to reduce HCV transmission[
A Treatment-as-Prevention Trial to Eliminate Hepatitis C Among Men Who Have Sex With Men Living With Human Immunodeficiency Virus (HIV) in the Swiss HIV Cohort Study.
]. However, our findings also demonstrate an ongoing risk of repeated HCV transmission among MSM individuals: in line with previous findings in the German hepatitis C cohort (GECCO) study, 15.2% of MSM were diagnosed with an HCV reinfection during the observation period[
Reinfection With the Hepatitis C Virus in Men Who Have Sex With Men After Successful Treatment With Direct-acting Antivirals in Germany: Current Incidence Rates, Compared With Rates During the Interferon Era.
]. Moreover, the resulting reinfection incidence rate slightly increased in the DAA era compared to the interferon era, suggesting that a small group of patients with high-risk behavior mainly contributes to an ongoing HCV epidemic. Accordingly, higher HCV reinfection rates have been reported in recently acquired HCV cohorts, which mainly consist of PWID and HIV-positive MSM, compared to individuals treated for chronic HCV infection[
]. Aggravated are these observations by low spontaneous clearance rates detected in HIV-positive individuals of our study cohort (12%). These findings might have been biased by the relatively early treatment uptake in our study population. However, our results of lower spontaneous resolution rates in HIV-infected patients are consistent with several previous findings and might be due to the compromised immune response in HIV-coinfections, which limits the ability to clear an acute HCV infection in HIV-coinfected patients.
Regarding phylogenetic analysis, our study results suggest the existence of a large MSM-specific network connecting national outbreaks of RAHC in different European, Australian and Canadian countries with each other. The phylogenetic contrast between largely clustering sequences of MSM populations and unrelated sequences obtained from patients with other transmission routes reflects an expanding epidemic of HCV among MSM, transmitted permucosally. Moreover, all identified clusters in our study proved to be MSM-specific, which strengthens the hypothesis that transmission of HCV within MSM and IVDA communities occur in distinct clusters. The results of our phylogenetic analysis revealed that most of the HCV strains have rapidly spread among MSM populations in different countries via a joint international HCV transmission network. Interestingly, travel-associated infections could be supported in several of these cases, highlighting the existence of a well-connected and rapidly spreading network. Growing MSM tourism, cheaper international travel and increasing digitization seem to extend HCV transmission networks and increase the connectivity of high-risk MSM communities. We observed clustering of Frankfurt MSM sequences for GT 1a, 3a and 4d. However, overlap of Frankfurt MSM cases with networks from different countries was exclusively detected in GT 1a and 4d: 45% and 100% of Frankfurt MSM sequences were part of international clusters, respectively. Interestingly, these clusters contained published sequences from more than 10 years ago as well as more recent sequences from patients in our study. This suggests that large international transmission networks still exist in GT1a and GT4d. Our analysis over an extended period of 10 years reveals that
especially GT4d appears to play a major role in the changes of HCV genotype distributions over time, as GT4d considerably increased during the last 5 study years. Some studies have shown a phylogenetic clustering of GT4 sequences within one country but also between countries such as France, the Netherlands, UK and Germany for MSM patients with GT4d[
Phylogenetic analysis of acute hepatitis C virus genotype 4 infections among human immunodeficiency virus-positive men who have sex with men in Germany.
]. Remarkably, we were also able to confirm this overlap with sequences from these countries for current sequences obtained from 2019. In addition, we demonstrate phylogenetic clusters between the Frankfurt MSM sequences with Spanish and even Canadian MSM sequences, which has not been shown to this extent before. This emphasizes the rapid spread of GT 4d among MSM as well as the immense influence of international joint networks on HCV transmission.
In contrast, we did not observe any country mixing or clustering of GT1b and GT3a which confirms observations from another study[
One major limitation of our study is the lack of a standardized protocol consisting of regular testing intervals in order to trace reinfections after cure of a first HCV infection. Reinfection cases were detected by the treating physicians, so we are not able to rule out differences in testing frequencies and missing cases if patients did not stay in care after DAA treatment. In addition, we did not collect behavioral data systematically by standardized questionnaires. Thus, potential transmission chains within clusters could only be drawn carefully and we were not able to identify specific risk behaviors as possible predictors for HCV infections. Moreover, we focused on HIV-positive MSM while the proportion of other high-at risk patients such as PWID with RAHC was small, limiting the results to the MSM population.
In conclusion, our data demonstrate the presence of RAHC mainly in association with HIV-coinfected MSM and injection drug use with an overall decline of cases and a certain HCV genotype pattern over the observation period of 10 years. In addition, the existence of large and internationally connected transmission networks among MSM, which unite the majority of RAHCs in form of clusters and are rapidly spreading throughout different countries could be demonstrated. Spontaneous clearance rates were observed to be low and reinfection rates increased mainly driven by a small subset of MSM patients with high-risk behavior. Thus, these data highlight the importance of implementing regular screening for HCV in the clinical routine care of HIV-positive MSM. Beyond that, further constructive preventive strategies such as scaling-up of DAAs in high-risk groups, early treatment initiation in order to avert secondary infections, raising awareness among MSM, PWID and clinicians as well as behaviour interventions are inevitable in order to achieve HCV elimination.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.
Acknowledgements
We thank Babithra Yoganathan-Kugarajan and Haruko Gölz Carnero for expert technical assistance.
Appendix A. Supplementary data
The following is/are the supplementary data to this article:
Ledipasvir plus sofosbuvir fixed-dose combination for 6 weeks in patients with acute hepatitis C virus genotype 1 monoinfection (HepNet Acute HCV IV): an open-label, single-arm, phase 2 study.
Reinfection With the Hepatitis C Virus in Men Who Have Sex With Men After Successful Treatment With Direct-acting Antivirals in Germany: Current Incidence Rates, Compared With Rates During the Interferon Era.
Changing Trends in International Versus Domestic HCV Transmission in HIV-Positive Men Who Have Sex With Men: A Perspective for the Direct-Acting Antiviral Scale-Up Era.
Decline in Hepatitis C Virus (HCV) Incidence in Men Who Have Sex With Men Living With Human Immunodeficiency Virus: Progress to HCV Microelimination in the United Kingdom?.
A Treatment-as-Prevention Trial to Eliminate Hepatitis C Among Men Who Have Sex With Men Living With Human Immunodeficiency Virus (HIV) in the Swiss HIV Cohort Study.
Phylogenetic analysis of acute hepatitis C virus genotype 4 infections among human immunodeficiency virus-positive men who have sex with men in Germany.
): non-MSM, Frankfurt, Germany (
), MSM, Frankfurt, Germany (
), MSM (published sequences), Germany (
), MSM (published sequences), United Kingdom (
), MSM (published sequences), Netherlands (
), MSM (published sequences), Switzerland (
), MSM (published sequences), Spain (
), MSM (published sequences), USA (
), MSM (published sequences), Canada (
), MSM (published sequences), Australia (
), MSM (published sequences), France, Abbreviations: GT, genotype; HCV, hepatitis C virus.
00032-0&id=ga1.jpg){kind=link}
00032-0&id=gr1.jpg){kind=link}
00032-0&id=gr2.jpg){kind=link}
00032-0&id=gr3.jpg){kind=link}