Impact of HCV genotype on treatment regimens and drug ... · REVIEW Impact of HCV genotype on treatment regimens and drug resistance: a snapshot in time Lize Cuypers1*, Francesca
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
R E V I E WImpact of HCV genotype on treatment regimensand drug resistance: a snapshot in timeLize Cuypers1*, Francesca Ceccherini-Silberstein2, Kristel Van Laethem1,Guangdi Li1,3, Anne-Mieke Vandamme1,4 and Jürgen Kurt Rockstroh51KU Leuven – University of Leuven, Department of Microbiology and Immunology, Rega Institute forMedical Research, Clinical and Epidemiological Virology, Leuven, Belgium2Department of Experimental Medicine and Surgery, University of Rome Tor Vergata, Rome, Italy3Department of Metabolism and Endocrinology, Metabolic Syndrome Research Center, Key Laboratory ofDiabetes Immunology, Ministry of Education, National Clinical Research Center for Metabolic Diseases, TheSecond Xiangya Hospital, Central South University, Changsha, Hunan, China4Center for Global Health and Tropical Medicine, Microbiology Unit, Institute for Hygiene and TropicalMedicine, University Nova de Lisboa, Lisbon, Portugal5Department of Medicine I, University Hospital Bonn, Bonn, Germany
Received: 16 March 2016; Revised: 11 June 2016; Accepted: 15 June 2016
BACKGROUNDSince the discovery of HCV [1], a preventive vac-cine remains elusive, resulting each year into twomillion new infections [2]. Because of its high ge-netic variability, HCV manifests into seven geno-types (GTs) and more than 50 subtypes [3], allvarying in geographical distribution, prevalence,level of genetic diversity [4] and pre-existing DAAresistance variants [5]. HCV GTs 1–3 circulateworldwide, whereas GTs 4–6 are more restrictedto specific geographical areas (Figure 1). Globally,GT1 accounts for almost half of all infections,followed by the second most prevalent GT3 [6,7].
Based on the presence of HCV RNA [8],approximately 80 (64–103) million people arechronically infected with HCV [9]. HCV infected
*Correspondence to: L. Cuypers, KU Leuven – University of Leuven,Department of Microbiology and Immunology, Rega Institute forMedical Research, Clinical and Epidemiological Virology, Leuven,Belgium.E-mail: [email protected]
patients are at risk to develop cirrhosis, end-stageliver diseases and hepatocellular carcinoma(HCC), with increasing numbers of mortality casesreported in the last years. They are also the sourceof continuing new infections. The HCV healthcareburden, for the four to five million people co-infected with HIV [10], is even higher because of ahigher prevalence of cirrhosis and HCC cases[11,12]. The primary goal of HCV treatment issustained virological response (SVR), which is de-fined as an undetectable viral load 12 or 24weeksafter end of therapy. The secondary goal is preven-tion of related liver complications, because viralcure is associated with a lower risk for morbidityand mortality, albeit to a lesser extent forHIV/HCV co-infected patients [13–16]. However,when therapy is initiated at a late stage and evolu-tion to cirrhosis has already started, risk reductionfor morbidity and mortality is smaller but not ab-sent, warranting continued HCC screening, evenafter achieving SVR. All HCV mono- and co-infected patients, treatment-naïve or -experiencedwith chronic liver disease, willing to be treatedand without contraindications for treatment,should be considered for therapy [17,18]. However,certain patient groups should be prioritized andregimens should be chosen with consideration ofhost and viral factors.
Genotype-dependent treatment regimens
Before 2011, the only therapeutic option for HCVinfected patients was the combination of pegylatedinterferon-α (pegIFN-α) and ribavirin (RBV) for24–72weeks, however, associated with severeadverse effects and varying effectiveness in differ-ent HCV GTs (Figure 2). HCV GTs 1, 4, 5 and 6showed SVR rates of ~50% in HCV mono-infectedpatients and lower than 30% in HIV/HCVco-infected patients [19], whereas higher SVR rateswere achieved for GTs 2 and 3. The HCV genotypewas therefore the most important baseline predic-tor for response to antiviral therapy based onpegIFN-α and RBV [20]. The advent of DAAs,which specifically target the NS3/4A protease,NS5A or NS5B polymerase [21], dramatically im-proved the efficacy of treatment strategies. Addingfirst generation NS3/4A protease inhibitors, suchas boceprevir (BOC) and telaprevir (TVR), topegIFN-α and RBV, increased SVR rates to morethan 70% in HCV GT1-infected patients [22].However, these drugs were also associated withlimited pan-genotypic activity, severe side effectsand rapid emergence of drug resistance variants[23]. More efficacious viral suppression is currentlyachieved by oral DAA-only combination therapywith SVR rates higher than 90%, broader antiviralactivity, less viral escape variants and less adverse
Figure 1. World map of the predominant HCV genotype in each country. This choropleth map shows the most prevalent HCV genotypeper country, using the Robinson’s projection, with the visualization software QGIS version 2.8.5-Wien (http://qgis.org/en/site/). HCV ge-notypes 1a and 1b are shown in the same color as HCV genotype 1 and for countries where prevalence of HCV1a and HCV1b are distin-guished, hatching is used to indicate the prevalent subtype. In light grey, countries are visualized for which no data or conflicting datawere reported. Data to construct this map were obtained through extensive literature search [6,7], with the respective references indicatedin the Supporting Information
events [22–30] (Figure 2). Table 1 summarizesgenotype-dependent SVR rates for the main clinicaltrials, approved regimens or experimental inhibi-tors in late clinical stages. These clinical trials arefocusing on treatment-naïve and -experiencedpatients, while some also include HIV/HCVco-infected and cirrhotic populations. Table 2 listsall currently approved drugs for the three differentDAA classes.
HCV genotype 1Despite high SVR rates, interferon-based regimensas mentioned in Table 1 (section A) are no longerrecommended for HCV GT1 infected patients[17,18], given the good performance of fiveapproved IFN-free regimens and one combination
licensed only in Japan. The regimen SOF+VEL isexpected to be approved soon.
SOF+SMVResults of four trials (1–4) (Table 1 section B) andthree real-life cohorts (5) for the regimen ofNS3/4A protease inhibitor simeprevir (SMV) andNS5B polymerase inhibitor sofosbuvir (SOF), withor without RBV, have been reported. (1) This regi-men resulted in SVR12 rates of 93% for therapy-naïve and –experienced patients in the COSMOSstudy [31]. (2) During the OPTIMIST-1 trial lowerSVR rates were demonstrated when treatmentduration was shortened from 12 to 8weeks innon-cirrhotic patients, either therapy-naïve (85%vs 97%) or -experienced (77% vs 95%) [32].
Figure 2. Sustained virological response (SVR) rates of HCV antiviral treatment through time. In the last years, HCV antiviral treatmenthas evolved from an IFN-based treatment to several IFN-free treatment options, characterized by differences in antiviral activity towardsthe six main HCV genotypes. SVR rates are defined as an undetectable viral load 12 or 24 weeks after stop of treatment. These SVR resultsare visualized through time (for details see Table 1), for the different regimens approved, split up for the different genotypes (GTs). Asspecified in the legend, the different categories of SVR rates are colored from red to green and with a white box indicating that this reg-imen was not approved for this particular HCV genotype or no in vivo data is available. All regimens are indicated by their abbreviations,more particularly boceprevir (BOC), daclatasvir (DCV), elbasvir (EBR), grazoprevir (GZR), ledipasvir (LDV), paritaprevir boosted with ri-tonavir, ombitasvir and dasabuvir (3DAA), ribavirin (RBV), simeprevir (SMV), sofosbuvir (SOF), telaprevir (TVR) and velpatasvir (VEL)
Although similar SVR rates were observed for pa-tients with a baseline HCV RNA level below 4 mil-lion IU/ml, shortening of therapy duration basedon baseline viral load is not yet recommended. (3)For cirrhotic patients, therapy duration should beextended to 24weeks, with or without RBV, to de-crease the risk of relapse [33]. (4) Higher SVR rateswere observed for SMV+SOF compared to SOF+pegIFN-α+RBV in HCV GT1a cirrhotic patients[34]. (5) Large-scale real-life cohorts reported simi-lar high SVR rates [35,36], although lower for cir-rhotic HCV mono-infections [37].
SOF+DCVHigh SVR rates were reported for SOF and NS5Ainhibitor daclatasvir (DCV) in three clinical trials(1–3) (Table 1 section C), and in compassionate useprograms (4). (1) Therapy-naïve or -experienced pa-tients were randomly assigned to treatment armscontaining SOF+DCV, resulting into SVR rates of98% for GT1 [38]. (2) In the ALLY-2 trial, includingHIV/HCV GT1 co-infected patients, SVR rates of96% and 76%were reported for treatment-naïve pa-tients, treated for 12 or 8weeks, similar to the resultsof therapy-experienced patients treated for12weeks [39]. (3) SVR rates of ALLY-1 resulted intothe recommendation to extend treatment durationto 24weeks for all GT1a infected patients, with orwithout RBV [40]. Despite limited evidence, thesame approach was applied for GT1b. (4) Large co-horts in compassionate use programs suggest thatcirrhotic patients may benefit from a longer therapyof 24weeks [41–43]. Nevertheless, no clinical benefitin the context of disease complications and mortal-ity was found for patients with severe recurrentHCV after liver transplantation [44].
SOF+LDVThe fixed-dose combination of SOF and NS5A in-hibitor ledipasvir (LDV) was studied during eightclinical trials (Table 1 section D). (1)–(2)–(3) In thethree first ION trials, high SVR rates were achieved,irrespective of treatment duration or addition ofRBV, including (non-)cirrhotic therapy-naïve and -experienced patients [45–47]. SOF+LDV for8weeks may be considered in treatment-naïve,non-cirrhotic patients with a baseline viral load(VL) below 6 million IU/ml, as determined by theRoche Cobas TaqmanHCVassay [48], although thiscut-off remains debatable [49]. (4) All HIV/HCV co-Ta
infected patients except for one, cured theirinfection during the ERADICATE trial [50]. (5) Sim-ilar SVR rates were obtained for HIV/HCV co-infected patients (ION-4), regardless of cirrhoticstatus or prior treatment [51]. Nevertheless, inclu-sion criteria of cirrhotic patients in trials have beenreported to be discordant with real-life cohorts[52]. (6) For this difficult-to-treat cirrhotic group,an increase in SVR for treatment-experienced pa-tients was observed when RBV was added or ther-apy was extended to 24weeks [53]. (7) In theSIRIUS study, prior PI-experienced cirrhotic pa-tients yielded SVR12 rates of 96%, when treatedfor 24weeks or 12weeks with RBV [54]. (8) Patientswith advanced liver disease, were treated for 12 or24weeks in the SOLAR studies, showing high SVRrates [55,56].
PTV/r+OBV+DSV+RBVThe triple DAA regimen of NS3/4A protease inhib-itor paritaprevir (PTV) boosted with ritonavir (/r),NS5A inhibitor ombitasvir (OBV) and NS5B poly-merase inhibitor dasabuvir (DSV), was evaluatedin eight clinical trials (1–8) (Table 1 section E), andhigh SVR rates were confirmed in the TRIO net-work and in a German study [57,58]. (1–2) In theSAPPHIRE trials, this regimen was efficacious inboth therapy-naïve and -experienced patients, al-though in cirrhotic GT1a infected patients a longertreatment period of 24weeks instead of 12weekswas required [59,60]. (3–5) All GT1a and GT1bpreviously untreated patients achieved high SVRrates during the PEARL studies [61,62]. Rates of vi-rological failurewere higher without RBV thanwithRBV among HCV GT1a but not among GT1b [62].
(6) HIV/HCV co-infected patients in TURQUOISE-I,including (non-)cirrhotic therapy-naïve and -experienced patients, obtained SVR rates of 91–94%,regardless of treatment duration or time of firstvirological response [63]. (7) Additionally, highSVR rates were reported for cirrhotic patients[64]. (8) Recently, GT1b-infected patients withcompensated cirrhosis and prior therapy-failure,were able to achieve 100% SVR using this 12-weekregimen without RBV, suggesting that RBV andlonger treatment durations are only beneficial forGT1a infected patients [65].
ASV+DCVFor GT1b therapy-experienced patients, treatedwith NS3/4A PI asunaprevir (ASV) and DCV(Table 1 section F), SVR rates of 77% and 95% wereachieved, the latter when combined with pegIFN-α+RBV [66,67]. Overall, more viral breakthroughswere observed for GT1a infected patients, evenwhen treated for 24weeks [68]. For HIV/HCV co-infected and cirrhotic patients, promising resultswere reported [69,70].
GZR+EBRThe fixed-dose combination of NS3/4A PIgrazoprevir (GZR) and NS5A inhibitor elbasvir(EBR), which was recently approved in the UnitedStates, showed promising results in seven trials(Table 1 section G). (1) Respectively 92% and 99%of the (non-)cirrhotic GT1a and GT1b treatment-naïve patients, were virologically cured in theC-EDGE trial [27]. (2) For treatment-experiencedpatients, treated either for 12 or 16weeks, SVR ratesof 92–94% and 92–97% respectively were achieved,
Table 2. Summary of direct-acting antivirals approved for clinical use. For all three drugclasses, NS3/4A protease inhibitors, NS5A inhibitors and NS5B polymerase inhibitors, thenames of the drugs currently approved for clinical use or described in this paper, andtheir respective abbreviation, are listed
depending on the addition of RBV [71]. (3) For pa-tients who previously failed a PI-based therapy,high SVR rates were achieved when RBV wasadded [72]. (4) In prior untreated HCV mono- andco-infected patients without cirrhosis, SVR12 ratesof 87–98% were reported, either without or withRBV [73]. (5) A large trial enrolling HIV-1therapy-naïve patients co-infected with HCV, withor without cirrhosis, reported overall SVR12 ratesof 96% [74]. (6) Both treatment-naïve and -experienced patients with chronic kidney diseasestages 4–5 were studied in the C-SURFER trial,resulting into overall SVR of 99% [75]. (7) An inte-grated analysis of compensated cirrhotic patientsshowed that for therapy-experienced patients in-fected with GT1b, a 12-week regimen is sufficientcompared to GT1a infected, which benefit from anextended treatment duration to 16 or 18weeks,and the addition of RBV [76].
Pipeline: SOF+VELThe first 12-week fix-dose combination of SOF andNS5A inhibitor velpatasvir (VEL) (Table 1 sectionH) was studied in ASTRAL-1, in prior untreatedand treated patients, including those with cirrhosis,demonstrating SVR rates of 99% [77]. Studying theregimen more in depth for patients with decom-pensated cirrhosis (ASTRAL-4) showed SVRs of83%, 94% and 86%, respectively for 12weeks,12weeks+RBV and 24weeks [78]. Pooled analysisresulted into therapy efficacy proven for all GTs[79]. Concerning HIV/HCV co-infected patients,SVR rates of 95% were achieved for all HCV geno-types, irrespective of cirrhosis status or treatmenthistory [80].
Pipeline: ABT-493+ABT-530High efficacy was demonstrated for the combina-tion of next-generation DAAs, NS3/4A protease in-hibitor ABT-493 and NS5A inhibitor ABT-530, forall HCV genotypes, irrespective of cirrhosis status[81,82]. Nevertheless, larger trials are needed toconfirm the very promising initial results for cir-rhotic patients. A shorter treatment duration of8weeks resulted into equal high SVR12 rates fornon-cirrhotic patients with HCV GT1 or 2 infec-tions (97–98%) [83]. For patients who previouslyfailed a DAA-containing regimen, the new combi-nation showed high efficacy, irrespective of RBV,in the MAGELLAN-I study [84].
HCV genotype 2SOF+RBV has become the gold standard to treatHCV GT2 infected patients. Other options to treatthese patients are SOF+RBV+pegIFN and SOF+DCV. Soon the dual DAA regimen SOF+VELwill be added to this list.
SOF+RBVThis regimen was tested during seven clinical trials(1–7) (Table 1 section I), and one real-life cohort (8).(1) Previously untreated patients were randomlyassigned to receive SOF+RBV for 12weeks, orpegIFN-α2a+RBV for 24weeks, resulting into anSVR of 95% for the first group (FISSION) [25]. (2)Patients for whom a therapy consisting of IFN isnot an option or who previously did not respondto IFN (FUSION), achieved SVR rates of 86% and94%, when treated for 12 or 16weeks [85]. (3) Dur-ing the POSITRON trial, 93% of the patients consid-ered as IFN-intolerant, virologically cured theirviruses [85]. (4) The VALENCE study obtainedhigh SVR rates, irrespective of previous treatmentor disease progression [86]. (5) In the BOSONstudy, SVR12 rates of 87%, 100% and 94% wereachieved for hard-to-treat patients, respectivelyfor SOF+RBV 16weeks or 24weeks and SOF+RBV+pegIFN 12weeks [87]. (6–7) In the PHO-TON studies, HIV/HCV co-infected patientsachieved high SVR rates irrespective of cirrhoticstatus [88,89]. (8) Real-world data confirmed thelower SVR rates for cirrhotic patients [36,90],whereas large trials are still needed to determinewhether 16weeks is the correct treatment durationfor these patients.
SOF+pegIFN-α+RBVAdding pegIFN-α to SOF+RBV (Table 1 section A)was studied in the LONESTAR-2 trial, resultinginto SVR rates of 96% [91], similar to the IFN-freevariant which achieved overall SVR rates of 95%(FISSION) [25].
SOF+DCVSVR rates of 92% were reported for the regimenSOF+DCV (Table 1 section C), independent oftherapy duration (AI444-040 and ALLY-1)[38,40,92]. Based on data of other GTs, 12weeks oftherapy is probably sufficient. Because of this lowersuccess rate (<95%) and the high cost associated to
the combination, this regimen should only be usedwhen other options are not available.
Pipeline: SOF+VELThe regimen SOF+VEL is a forthcoming combina-tion for GT2 (Table 1 section H), because high SVRrates were observed, even for cirrhotic patients(ASTRAL-1) [77,78]. Its efficacy was compared toSOF+RBV in the ASTRAL-2 and -3 studies, reveal-ing the superiority of this new regimen (SVR 99%vs 94%) [93,94]. Similar SVR rates were docu-mented in HIV/HCV co-infected patients [80].
Pipeline: ABT-493+ABT-530High SVR12 rates were achieved for cirrhotic andnon-cirrhotic patients treated with the combinationABT-493+ABT-530 [81,82], even when treated for ashorter period of 8weeks [83].
HCV genotype 3Standard treatment schemes have evolved fromIFN-based to IFN-free combinations. While GT1used to be the most difficult-to-treat HCV geno-type, this has now shifted to GT3, being the maingenotype where currently IFN-containing regimensare still an option. GT3 is also associated with ahigher prevalence of liver steatosis [95]. The regi-mens SOF+RBV, SOF+DCV, or SOF+pegIFN-α+RBV are used.
SOF+RBVSVR rates for GT3 infected patients (Table 1 sectionI) (1) were lower (56%) compared to those infectedwith GT2 (95%), when treated for 12weeks (FIS-SION) [25]. Patients for whom IFN therapy wasnot an option, were included in the FUSION andPOSITRON trials, (3) resulting in SVR12 rates of61%. (2) When treatment duration was extendedto 16weeks among prior treated patients, SVR in-creased dramatically (62% vs 30%) [85]. (4) Extend-ing duration to 24weeks resulted in even higherSVR rates of 85–90%, both for prior treated and un-treated patients, although lower efficacy was re-ported for cirrhotic patients, especially intreatment-experienced patients (VALENCE) [86].Therefore, SOF+RBV for 24weeks is only recom-mended in non-cirrhotic patients, while it is consid-ered suboptimal in patients with cirrhosis.
SOF+DCVHigh SVR rates in therapy-naïve patients (AI444-040) [38] and -experienced patients (ALLY-3 study)[26] were confirmed during a multicenter compas-sionate use program, suggesting that cirrhoticGT3 infected patients may benefit from a treatmentof 24weeks [96] (Table 1 section C). Nevertheless,treatment of GT3 infected patients with decompen-sated cirrhosis for 12weeks with SOF+DCV+RBVresulted into SVR rates of over 70% [43]. Recently,SVR4 rates of 88% and 96% were obtained for the12- and 16-week arms in the ALLY-3+ study, in-cluding patients with advanced fibrosis and cirrho-sis; however, only when ribavirin was added [97].
SOF+LDVTo date, SOF+LDV is not recommended to treatGT3 infections [17,18], because all data from trialsand early access programs did not show highenough SVR rates, and little antiviral activity wasobserved in vitro for LDV [98]. Lower SVR rateswere observed for SOF+LDV compared to SOF+DCV in the ELECTRON-2 trial [43], with for theregimen SOF+LDV also lower rates were reportedfor cirrhotic versus non-cirrhotic patients (73% vs89%) [99].
SOF+pegIFN-α+RBVThis regimen remains a good option for HCV GT3(Table 1 section A), because high efficacy was re-ported for treatment-naïve and -experienced pa-tients, proving superiority compared to SOF+RBV for 12 or 24weeks (NCT01188772 and BO-SON) [90,91,100]. The LONESTAR-2 study ob-tained SVR rates of 83% in prior treated patients,suggesting its use in difficult-to-treat patients [91].
Pipeline: SOF+VELThis regimen will soon enter the antiviral drugmarket (Table 1 section H), because higher SVRwas reported compared to SOF+RBV, in theASTRAL-2 and -3 trials [93,101].
Pipeline: ABT-493+ABT-530HCV GT3 infected patients were separately studiedfor ABT-493+ABT-530 after obtaining high SVRrates in general [81,82]. So far, no virological failurehas been observed with this combination, of whichthe first study included only non-cirrhotic patientsand a second focused specifically on cirrhotic
patients [102,103]. Non-inferiority to SOF+DCVhas been shown in the ENDURANCE-3 trial [104]and results of cirrhotic treatment-experiencedHCV GT3 patients are expected soon.
HCV genotype 4HCV GT4 infections are increasing in prevalenceworldwide, represented by a high variety of sub-types. For these patients, SOF+pegIFN+RBV(NEUTRINO and NCT01565889) [25,105] and fourIFN-free regimens were approved.
SOF+RBVThis regimen was evaluated in two Egyptian trials,resulting in SVR rates of 68–77% or 90–93%, eitherfor a 12- or 24-week treatment [106,107] (Table 1section I). In the PHOTON-II study, a small groupof HIV/HCV co-infected patients were treated for24weeks, resulting into SVR rates of 84% [89].
PTV/r+OBV+/�RBVTable 1 section E describes a triple (adding DSV inGT1) or dual (GT4) DAA regimen. Because DSVshows exclusive antiviral activity towards GT1[28], in the PEARL-I study, non-cirrhotic GT4 in-fected patients were treated with a dual DAA regi-men, achieving high SVR rates, independent ofprior therapy-experience [108]. In the AGATE-Iand -II studies, this combination showed highSVR rates in cirrhotic patients after therapy for 12,16 and 24weeks [109,110].
SOF+LDVSVR rates of 95% were reported for therapy-naïvepatients, supporting the role of SOF+LDV in GT4infected patients (NCT01826981 and SYNERGY)[111,112] (Table 1 section D). Replacing LDV withDCV was tested in a multicenter compassionateuse program [42], with SVR rates of 100%.
GZR+EBR and in pipeline: SOF+VEL andABT-493+ABT-530For GZR+EBR (Table 1 section G), overall SVRrates of 95% were reported in treatment-naïve pa-tients, either mono- or co-infected patients (C-EDGE) [27,74]. Pooled analysis showed improvedSVR rates when RBV was added and durationwas extended to 16weeks in case of prior on-treatment virological failure [113]. Soon, SOF+VEL will be available, reported to have 99%
SVR rates [77] (Table 1 section H). SVR12 rates of100% have been reported for ABT-493+ABT-530in HCV GT4 [81,82,114].
HCV genotype 5–6Currently, only two regimens have been approvedto treat patients infected with HCV GTs 5 and 6, be-cause clinical studies are limited (Table 1 sections Cand D). Treatment with SOF+LDV for 12weeks intreatment-naïve and -experienced patients resultedin SVR rates of 95–96% (GT5: NCT01826981 [111]and GT6: preliminary data of the ELECTRON-2study [99]), however slightly lower in cirrhotic pa-tients. Patients can also be treated with SOF+DCV, however only based on extrapolation of re-sults obtained in other GTs. In the NEUTRINOtrial, SOF+pegIFN-α+RBV resulted into 100%SVR for GT5 [25] (Table 1 section A). Soon the reg-imens GZR+EBR [27] and SOF+VEL [77,78] willbe available as well (Table 1 section G and H), withSVR rates in the range of 95–99%. SVR12 rates of100% were also reported for HCV GTs 5-6, usingABT-493+ABT-530 [114].
HCV genotyping assays as a prognostic tool:selection of treatmentIn the DAA era, the correct determination of theHCV genotype remains important to guide the se-lection of the most appropriate treatment schemefor each patient [17,18], as even the DAAs do notharbor equal antiviral activity across all GTs[115,116] (Table 1). Commercial assays are availablefor determining HCV genotype and subtype, alltargeting the highly conserved and best-characterized 5′ untranslated region. However, be-cause this region has been shown inappropriate todiscriminate certain HCV strains [117], the twomost used diagnostic assays, Abbott RealTimeHCV Genotype II and INNO-LiPA-HCV-2.0, alsotarget the NS5B or the core gene, providing addi-tional information to distinguish GT1a and 1b[118]. Nevertheless, they can still assign strains asGT1 without subtype, as ‘undetermined’ or ‘mixed’[118], making it necessary to use in-house sequenc-ing to correctly assign the HCV GT. Reports aboutthe concordance between subtyping results fromcommercial assays and sequence-based genotyp-ing, focus mainly on GT1 unresolved infectionsand are biased because in-house methods are oftengenotype- or subtype specific and target different
regions of the HCV genome which may potentiallydiffer if recombination has happened. However, incontrast to HIV which has a strong tendency to un-dergo intra- and inter-subtype recombination, thisphenomenon has only sporadically been describedfor HCV. Although the Abbott assay was able to re-solve 90% of the GTs, additional testing usingcore/E1, NS3, NS5A or NS5B assays, was requiredin 9–10% of the cases to fully resolve the GT [119–121]. Genotyping through sequencing can gatheradditional information about the presence of drugresistance variants; moreover, the HCV GT also im-pacts prevalence and development of resistance-associated variants (RAVs).Only few cases have been reported concerning
mixed HCV GT infections, mainly in personswho inject drugs (PWID) and patients onhemodialysis or multiple transfusions [122,123].In these infections, one of the GTs prevails,because they differ in replication efficacy or viralinterference. As commercial assays are not al-ways able to identify the minor genotype(s) thatexist(s) aside the dominant genotype [124], theirimpact on SVR rates with DAAs should beconsidered [125].
Known RAVs to DAAsBecause of the high error prone HCV RNA poly-merase coupled with a 100-fold higher virion pro-duction than HIV [126], HCV replicates as apopulation of closely related viral variants withina patient. It has been predicted that each nucleotidewithin the HCV genome theoretically can besubstituted every day, with most RAVs or nowa-days called resistance-associated substitutions(RASs) produced naturally during the replicationcycle [127,128]. The frequency of these RAVs de-pends on multiple factors, such as replication fit-ness, fitness cost and genetic barrier to resistance[129]. Theoretically it is possible to detect a singleRAV against any of the three DAA classes as mi-nority variants in all patients prior to treatment,while virological failure of combination therapieswould require multiple RAVs on multiple drug tar-gets [130]. Combinations of multiple RAVs to therecommended IFN-free regimens are howeverrarely detected in DAA-naïve patients [131].
Resistance to NS3/4A protease inhibitorsProtease inhibitors (PIs) interact with the enzymesubstrate binding site and prevent cleavage of the
HCV polyprotein into several non-structural pro-teins. Virological failure with first generation PIsis often associated with the emergence of RAVs[132], more specifically the most prevalentV36A/M, T54A/S, V55A, Q80R/K, R155K/T,A156S/T/V, I/V170A and D168A/E/K/T/V/Y[23] (Figure 3A), of which only A156V/T confershigh level of resistance [133]. Drug resistancestrains were found in more than 80% of the patientswho failed triple therapy with TVR or BOC. Cross-resistance between the first- and second-wave PIswas observed for variant R155K and for amino acidsubstitutions at residue D168, with the lattermainly known to confer resistance to second-wavePIs [134]. Prevalence of RAVs after therapy failurevaries according to genotype [5], for example, vari-ant R155K is mainly found in GT1a, while for GT1bA156T/V is more frequent, because two nucleotidesubstitutions are required for GT1b to developR155K. GZR retains potent antiviral activity evenin the presence of the key RAVs mentioned above,although viruses with substitutions at NS3 positionA156 and D168 display some reduced susceptibil-ity [135,136].
Resistance to NS5A inhibitorsThe exact function of NS5A is still obscure. Itregulates viral replication, participates in assem-bly and release of HCV particles and displaysseveral interactions with host proteins. NS5A in-hibitors interact with domain I of the NS5A di-mer, but the inhibitory mechanism remainsunclear [137]. Nevertheless, it has been recentlysuggested that the binding of inhibitors to adrug-resistant NS5A protein causes conforma-tional changes [138,139]. The most importantRAVs to NS5A inhibitors are M/L28T/V,Q/L30E/H/R/S, L31M/V, H58D and Y93C/H/N [98,127] (Figure 3B). DCV and LDV displaysimilar potencies in HCV GT1a and GT1b wild-type replicons during in vitro assays, althoughDCV proved to be superior against resistant vari-ant Y93H [98], which has a natural prevalence of>10% and displays high level resistance to bothLDV and DCV in GT1b replicon cells [140]. Alsofor HCV GTs 2–4, DCV has significantly higherpotency in vitro compared to LDV [98], with thehighest fold resistance values for variant F28S inGT2, Y93H in GT3, and RAVs on NS5A positions30 and 93 for GT4 [140–142]. Variants conferring
resistance towards EBR were studied for HCVGT1a, GT1b and GT3 replicon cells [143]. No con-sistent pattern of RAVs was observed for five re-lapsers treated with a therapy containing VEL
[144], and no impact on treatment outcome wasreported for the presence of NS3 and NS5A RAVsfor ABT-493+ABT-530 in GT3 [103] or treatment-experienced GT1 infected patients [83].
Figure 3. Drug resistant variants near the binding pocket of DAAs in HCV protein structures: (A) NS3/4A protease (NS3: pink, NS4A:blue) in complex with simeprevir, (B) NS5A dimer in complex with daclatasvir and (C) NS5B polymerase in complex with sofosbuvir andbeclabuvir. NS3/4A protease inhibitors bind to the catalytic triad of the NS3 serine protease, which consists of the three amino acids H57,D81 and S139. The mechanism of action of the NS5A inhibitors is not entirely understood, although it is known that they interact with theNS5A domain I. With different mechanisms of action, nucleotide inhibitors (e.g. sofosbuvir) and non-nucleoside inhibitors (e.g.beclabuvir) target the catalytic site and the allosteric site, respectively. Near the binding pocket, amino acid positions associated with drugresistance towards the respective inhibitors are visualized in colored spheres (see legend). For the visualization, PDB data of HCV proteinstructures were obtained from literature (NS5A [139]) and the Protein Data Bank (NS3/4A: 3KEE and 4B76, NS5B: 4NLD and 4WTG), usingvisualization software: PyMOL V1.7 (http://www.pymol.org/). Interactive movies are available on http://www.virusface.com/HCV/HCV_DrugResistance2016.html
Resistance to NS5B polymerase inhibitorsThe NS5B RNA polymerase of the membrane-associated HCV replication complex is structurallyorganized in a ‘right hand motif’ containing palmand thumb domains [28]. Nucleos(t)ide inhibitorsmimic natural substrates that are incorporated intothe nascent RNA chain and result in chain termina-tion, while non-nucleoside inhibitors (NNI) bindoutside the polymerase active site to allostericbinding sites, resulting in no cross-resistance be-tween the subclasses (Figure 3C). For the nucleo-tide analog SOF, resistant replicon cells with asingle NS5B S282T variant were selected, confer-ring decreased susceptibility to SOF [145]; how-ever, this variant is rarely identified in clinicalcases [99,146]. In a pooled analysis of SOF, NS5Bsubstitutions L159F and V321A were selectedpost-baseline in several infected subjects who didnot achieve SVR, with the highest proportion offailures detected in HCV GT1a, GT2 and GT3 in-fected patients [147]. Nevertheless, these RAVsconferred only 1.2- to 1.6-fold reduced phenotypicsusceptibility to SOF in vitro [147]. NS5B variantC316N/H/F was present at baseline in six GT1b in-fected subjects who virologically failed and in oneGT1a relapsing patient [148]. The rare NS5B RAVL320F was identified under therapy with SOF, pos-sibly contributing to drug resistance [148]. ForNNI, commonly observed NS5B substitutions areM414T and S556G [149], or A421V and P495L/S[150].
Pre-existing drug resistant variantsIn addition to RAVs emerging under DAA therapyor acquired at infection by transmission from aDAA-failing patient with resistance, they can alsopre-exist before treatment initiation as naturally oc-curring variants within the viral population of aninfected patient, prior to drug selective pressure.The NS3 RAV Q80K, associated with signifi-
cantly lower SVR rates for treatment with SMV+pegIFN-α+RBV [24], exists as a natural polymor-phism mainly in HCV GT1a [5]. In general, NS3RAVs were found in 19–31% of NS3 sequencesoriginating from all HCV genotypes [151,152], withfor the most prevalent variant Q80K a higher fre-quency in GT1a (20–52%) compared to GT1b(<1%) [5,134,153,154]. The presence of Q80K is es-pecially problematic in cirrhotic patients, becausefor GT1a infected patients treated with SMV
+SOF, lower SVR rates of 74% were observed inthe presence of Q80K versus 92% in the absenceof Q80K [33]. Irrespective of Q80K, all non-cirrhoticpatients responded well [32]. Therefore, monitoringof Q80K prior to therapy is recommended in allHCV GT1a infected patients starting treatmentwith SMV+pegIFN-α+RBV, while for therapywith SMV+SOF, testing is needed only for cirrhoticpatients [17].
A large prevalence study of natural NS5A RAVsacross different countries showed substantial re-gional differences [151,155], with a broad range of6–25%. The most common NS5A RAVs wereL31M, Q54H and Y93H [146]. For the combinationASV+DCV, in one study the NS5A variant Y93Hwas observed in half of the failing patients priorto treatment, all classified as HCV GT1b priornull-responders to pegIFN-α+RBV. In a differentstudy, the UNITY-1 study, despite the higher rateof NS5A RAVs at baseline detected in GT1b com-pared to GT1a infected patients (16% vs 11%), allGT1b infected patients achieved SVR in contrastto only 74% for GT1a [156]. Higher SVR rates wereobserved for patients lacking Y93H when treatedwith SOF+DCV [157].
For SOF+LDV, natural RAVs were observed in ahigher proportion in HCV GT1b compared to GT1asequences [158]; however this was not associatedwith lower SVR rates. Lower SVR rates for this reg-imen were only reported for GT1a, in therapy-experienced patients with RAVs conferring morethan 100-fold resistance [136,159]. Upon investigat-ing the original baseline sequence in a study on pa-tients failing 8 or 12weeks of SOF+LDV basedregimens, which were scheduled for retreatmentwith the same regimen for 24weeks, a link was re-vealed between the number of natural NS5A RAVsand the observed SVR. Only 50% of the patientsthat had two or more baseline resistance-relatedvariants cleared the virus, with the lowest SVRrates observed with variant Y93H/N [160]. A re-cent study showed that a longer duration of treat-ment with SOF+LDV and addition of RBV canreduce or even eliminate the impact of baselineNS5A RAVs [161].
In prior null-responders to pegIFN-α and RBV,the impact of natural NS5A variants on the efficacyof the GZR+EBR regimen in HCV GT1 was stud-ied by next-generation sequencing (NGS). Espe-cially HCV GT1a infected patients were affected,because only 52% that harbored NS5A variants
with >5-fold shift to EBR were able to achieve SVR[71,162]. For GT1a infected patients who initiatetreatment with GZR+EBR, it has recently becomerecommended to monitor high fold-change NS5ARAVs for EBR at baseline [17]. Naturally occurringRAVs in NS5A seem to have little effect on SOF+VEL or ABT-493+ABT-530, despite a high preva-lence of such variants, 97–100% achieved SVR[77,102,103].
Based on eight SOF monotherapy and five SOF+LDV trials, baseline sequences of 408 patientswho virologically failed, were evaluated usingNGS [163]. NS5B variant L159F was detected in1% of the GT1 infected patients and was only as-sociated with increased virological failure in pa-tients treated for short durations with SOF+RBV, but did not affect treatment outcome withLDV+SOF [163]. A Russian study focused on thecomparison of SVR12 rates achieved in patientswith and without variant L159F at baseline andtreated for 16weeks with SOF+RBV [164]. RAVL159F was mainly observed in GT1b (34% preva-lence) and was associated with decreased SVRrates of 25% compared to 65% in patients withoutthis variant [164]. Other variants conferring resis-tance towards NS5B polymerase inhibitors didnot have an impact on treatment outcome, for ex-ample the highly prevalent RAV C316N (48%) inHCV GT1b infected Japanese patients who initi-ated therapy with SOF+LDV [146]. Also in theAVIATOR trial, evaluating the regimenPTV/r+OBV+DSV, the most prevalent NS5BRAV (>3%) S556G was not associated with treat-ment response [130]. Nevertheless, in general nu-cleoside inhibitor based regimens have a lowprevalence of natural RAVs [131].
Sequencing as prognostic tool: drugresistance testingResistance testing is not routinely performed inHCV clinical practice, in contrast to HIV where itis recommended both prior to start of treatmentand during follow-up [165], in order to preventtherapy failure. While in HIV patients, any resistantvariant remains archived in the proviral DNA, thisis not the case for HCV, with time, the virus turn-over eliminates resistant variants that are often lessfit. There is no need to compile historical resistanceinformation for the individual HCV patient to findthe best treatment.
Declined persistence rates of RAVs post-treatment were reported with differences for thethree DAA classes, indicating indeed that there isa fitness cost to the development of RAVs. Whilefor first-generation PIs TVR and BOC, NS3 variantswere still detectable after one-year post-therapy[23], a long-term follow-up of patients who failedon BOC revealed that 73% of all NS3 RAVsreverted to the wild-type within three years post-therapy [166]. The one-year persistence rate ofNS3 RAVs for second wave or second generationPIs was much lower (9%) [133]. NS5A and NS5BRAVs persisted much longer, with respectively96% and 57% of the variants still present 48weeksafter therapy with PTV/r+OBV+DSV [167]. ForNS5A inhibitors EBR and LDV, the majority of thepatients still carried detectable RAVs 93weeks aftertreatment [168,169].The only drugs for which a resistance test is re-
quired before therapy initiation are SMV and EBR(Table 3). For combination regimens of SMV+pegIFN-α+RBV, or SMV+SOF (in case of cirrho-sis) resistance testing should be considered in HCVGT1a infected patients, because lower SVR rateswere reported in the presence of NS3 variantQ80K, which has a high prevalence in this subtype[17,24,32]. Nevertheless, the regimen SMV+pegIFN-α+RBV is no longer recommended touse in GT1 infected patients. Recently, treatmentguidelines changed for HCV GT1a infected pa-tients, because drug resistance testing is now alsorecommended when treatment with GZR+EBR isinitiated [17]. When high fold-change NS5A RAVsfor EBR (M28A/G/T, Q30D/E/H/G/K/L/R,L31F/M/V and Y93C/H/N/S) are detected atbaseline, treatment duration needs to be extendedfrom 12 to 16weeks and RBV needs to be addedto the regimen [17].Nevertheless, also for other NS5A and NS5B
RAVs, monitoring RAVs before start of treatmentcould be considered, even though the influence ofthese variants on clinical outcome is not suffi-ciently known yet. For instance, NS5A variantY93H may be monitored in HCV GT1 beforetreatment is started with ASV+DCV, becausethe presence of this variant was associated withtherapy outcome in both GT1a and GT1b infectedpatients [156]. For HCV GT1b infected patientswho want to start treatment with SOF, monitor-ing of NS5B variant L159F could be consideredbecause decreased SVR rates were reported for
Table 3. Treatment indications [17,18] and genotyping or sequencing requirements for HCVmono-infected or HCV/HIV co-infected patients with chronic HCV without or withcompensated cirrhosis, including treatment-naïve patients and patients who failed treatmentbased on pegylated interferon-α (pegIFN-α) and ribavirin (RBV). Interferon-free and -basedregimens containing direct-acting antivirals asunaprevir (ASV), daclatasvir (DCV), dasabuvir(DSV), elbasvir (EBR), grazoprevir (GZR), ledipasvir (LDV), ombitasvir (OBV), paritaprevir(PTV) boosted with ritonavir (/r), simeprevir (SMV), sofosbuvir (SOF) and velpatasvir (VEL)are summarized. Treatment schemes with and without cirrhosis (c) are listed, includinginformation about the weeks (w) of treatment and in case of cirrhotic patients the duration oftreatment with (+ RBV) and without ribavirin (� RBV). For the IFN-based regimen pegIFN-α+RBV+SMV, after 12weeks, treatment is continued without SMV for an additional 12 of24weeks (+12w or 24w). Genotyping refers to determining the genotype and subtype, it isrecommended before starting the indicated therapy, using assays designed for this purpose, orusing genetic sequencing. Sequencing refers to determining the nucleotide sequence of drugtarget genes for resistance testing purposes [17,33,156,160,162]. The RAVs to the respectivetreatments that are advised to be monitored are listed [5]. In the presence of these RAVs, adifferent treatment may be chosen, either recommended (bold), or it could be considered toadapt the regimen (plain text). Note that genetic sequencing can be used for both purposessimultaneously
ASV+DCV* 1 Yes NS5A: Y93H 24w 24wSOF+VEL† All No 12w 12wPegIFN-α+RBV+SOF
3+4+5+6 Yes 12w 12w
*Treatment regimen only approved in Japan.†In the pipeline, will soon become available.‡NS3 variant Q80K is indicated in bold, because it is a RAV for which testing is recommended in GT1a: if Q80K isdetected, SMV+SOF should be avoided for cirrhotic patients [33].§High fold-change NS5A RAVs for elbasvir which are recommended for testing in HCV GT1a infected patients [17] are28A/G/T, 30D/E/H/G/K/L/R, 31F/M/V and 93C/H/N/S.
patients harboring this variant compared to pa-tients lacking it [164].
Even when RAVs persist after failure of treat-ment, the large number of therapies available andthe lack of cross-resistance among different classesof DAAs imply that most HCV patients who failedto achieve SVR with a specific DAA-based regimenwill be able to be retreated with other DAA thera-pies [170], however with conflicting results whenRAVs are present [171,172]. However, HCV has alarger genetic variability than HIV, and the preva-lence of naturally occurring RAVs is much higher.Therefore, viral sequencing can play a role as prog-nostic tool to select the most appropriate secondline regimen for retreatment. Nowadays, expertswithin the virology field advise drug resistancetesting for all three target genes (NS3, NS5A andNS5B), for all failing regimens, to guide the selec-tion of a second line regimen (Table 4). This is notonly to detect drug resistance variants thatemerged under the failing therapy, but also to mon-itor RAVs to other drug targets that are present asnatural occurring variants. It is too early to makesolid recommendations for retreatment based onresistance testing, because studies assessing treat-ment success in the presence or absence of particu-lar resistance profiles are not available yet.However, we do have information about resistanceprofiles appearing in patients that failed a particu-lar regimen. Therefore, therapeutic decisions canbe made based on HCV genotype, detected resis-tance profiles, number of drugs used, use of RBVand treatment duration.
Depending on the type of resistance detected andthe urgency of treatment, therapy could be post-poned until more evidence is available to betterguide retreatment decisions. For example, in theabsence of cirrhosis, it is advised to either wait formore active regimens or to administer at least twofully active drugs, with a preferential use of onedrug with high genetic barrier to resistance,and/or with extended treatment duration and ad-dition of RBV (Table 4). A longer treatment ofGZR+EBR for 16weeks and addition of RBV wasrecently recommended in patients who previouslyfailed the same regimen [17]. In patients failingNS5A based therapies, retreatment regimens in-cluding NS5A inhibitors are not advised, unless re-sistance testing showed absence of NS5A RAVs orpresence of minor NS5A RAVs which do not neces-sarily confer cross-resistance to the entire drug
class. When resistance information is absent, thesepatients could be treated by shifting drug class toa NS3 containing regimen, like SOF+SMV. Simi-larly, patients failing NS3 based therapies can stillbe treated with NS5A based regimens such asSOF+DCV or SOF+LDV, in case resistance infor-mation is absent or in the presence of high-fold re-sistant NS3 RAVs. The most difficult situation iswhen designing a therapeutic approach for patientswho harbor RAVs to multiple DAA classes. Thesepatients currently have few retreatment optionswith commercially available IFN-free combinationsand might be helped with multiple DAA combina-tions targeting nearly all replication steps. This ap-proach is currently under evaluation in someclinical trials [173,174].All therapy regimens with indications regarding
RAVs monitoring are listed in Tables 3 and 4. How-ever, it is not clear yet what the best strategy is tomeasure the presence of RAVs [175], Sanger popu-lation sequencing which can detect variants downto 20% of the population or NGS for which detec-tion limits down to 1% have been reported [176].Because of the higher intra-patient genetic variabil-ity, minority variants are deemed more importantin HCVresistance development than for HIV. How-ever, knowledge on the clinical relevance of detect-ing variants at low levels is still scarce, the mostrecent reports suggest 20% as a sufficient thresholdto detect the most impactful RAVs [162]. Othertechnical issues make HCV resistance testing quitechallenging, such as the design of genotype- andsubtype-specific PCR primers, error-rates and highcosts.
Sequencing as an epidemiological tool:transmission investigationDespite the high SVR rates associated with DAAregimens, and the limited need for extensive drugresistance testing compared to HIV, viral eradica-tion of HCV on a global scale is still hampered, be-cause of a vast majority of the HCV infectedpopulation that is not aware of their status, thehigh costs associated with these drugs, the un-known impact of acquired drug resistance [177],and the high re-infection rates in risk populations(13% for PWID and 22% for HIV/HCV co-infectedpatients) [178–180].Therefore, genetic sequences are highly valuable,
Table 4. Treatment indications [17,18] and genotyping requirements for HCV mono-infectedor HCV/HIV co-infected patients with chronic HCV who failed to achieve an SVR on priorantiviral therapy containing one or more direct-acting antivirals (DAA’s). Patients whopreviously failed treatment regimens can be retreated with several treatment schemes,including daclatasvir (DCV), dasabuvir (DSV), ledipasvir (LDV), ombitasvir (OBV),paritaprevir (PTV) boosted with ritonavir (/r), pegylated interferon-α (pegIFN-α), ribavirin(RBV), simeprevir (SMV) and sofosbuvir (SOF). New therapies are administered to HCVinfected patients for a different number of weeks (w) in non-cirrhotic and cirrhotic (c)patients. For all failing regimens, drug resistance testing of all three genes (NS3, NS5A andNS5B) is advised before retreatment. Depending on the type of resistance, if treatment isnot urgent, therapy should be postponed. In case of absence of cirrhosis, it is advised toeither wait for more active regimens or to administer at least two fully active drugs, with apreferential use of one drug with high genetic barrier to resistance, and with extendedtreatment durations and addition of RBV. Depending on the outcome of drug resistancetesting, retreatment strategies contain drugs belonging to the same DAA class as the failingtreatment or they need to be shifted towards other DAA classes (*)
Failed treatment Genotype New treatment Treatment period
epidemiological investigations. Together thesechallenges force the continued search for newpan-genotypic DAAs.
CONCLUSIONSA correct determination of the HCV genotype in-fecting a patient remains important to guide the se-lection of the most appropriate antiviral regimen.This is because treatment response rates, and theprevalence and development of drug resistance var-iants, differ for each DAA regimen, even for theones with broader genotypic antiviral activity. Base-line HCV sequencing can provide important viro-logical information for a correct genotype/subtypeassignment and for the detection of genetic variantsthat can potentially affect therapy response. Evenwith the pan-genotypic regimen SOF+VEL, andother combinations in phase II clinical trials forth-coming, HCV sequencing can still assist in the selec-tion of the most appropriate second line regimen, inpatients who need to be retreated after DAA failure.In the future, even when drug resistance will be-come a minor issue, HCV viral eradication will stillbe hampered because of low diagnosis rates, highassociated costs and high re-infection rates in cer-tain risk populations.
CONFLICT OF INTERESTThe authors declare no conflict of interest, otherthan the financial disclosures described above.
ACKNOWLEDGEMENTSThe authors are very grateful to Fossie Ferreira,who helped visualizing the predominant HCV ge-notypes in each country across the globe. LizeCuypers was supported by a PhD grant of theFWO (Fonds Wetenschappelijk Onderzoek—Vlaanderen, Asp/12), and sponsored by an FWOgrant (G.A029.11N). Guangdi Li was sponsoredby the National Nature Science Foundation ofChina (31571368) and the Project of Innovation-driven Plan of Central South University(2016CX031).
TRANSPARENCY DECLARATIONSF.C.-S. has received financial support for attendingsymposia, speaking, organizing educational activi-ties, consultancy or advisory board membership,or grant research support from AbbVie, Abbott Mo-lecular, Bristol-Myers Squibb, Gilead Sciences,Janssen-Cilag, Merck Sharp and Dohme, RocheDiagnostics, and ViiV Healthcare. A.-M. V. hasreceived financial support for organizing educa-tional activities, consultancy or advisory boardmembership, from AbbVie. J.R. has received finan-cial support for attending symposia, speaking, or-ganizing educational activities, consultancy oradvisory board membership from Abbott, AbbVie,BMS, Bionor Cipla, Gilead, Hexal, Janssen, Merck,Shinogi and Viiv.
REFERENCES1. Choo QL, Kuo G, Weiner AJ, et al. Isolation
Additional supporting information may be found in the online version of this article at the publisher’sweb site:
Table 1. References for each country visualized on the world map with the predominant HCV genotypes(Figure 1). For a large proportion of countries, data was based on two main publications [6–7],complemented with studies conducted on national or regional levels. Literature was not systematicallyreviewed, so not all studies conducted regarding the prevalence of the HCV genotypes, are reported here.