Nine-year distribution pattern of hepatitis C virus (HCV) genotypes in Southern Italy

Introduction It has been greatly described that different hepatitis C virus (HCV) genotypes are strictly correlated to various evolution, prognosis and response to therapy during the chronic liver disease. Aim of this study was to outline the changes in the epidemiology of Hepatitis C genotypes in Southern Italy regions from 2006 to 2014. Material/Methods Prevalence of HCV genotypes was analyzed in 535 HCV-RNA positive patients with chronic Hepatitis C infection, selected during the period 2012–2014, and compared with our previous data, referred to periods 2006–2008 and 2009–2011. Results In all the three periods analyzed, genotype 1b is predominant (51.8% in 2006–08, 48.3% in 2009–11 and 54.4% in 2012–14) while genotype 2 showed an increase in prevalence (27.9% in 2006–08, 31.7% in 2009–11 and 35.2% in 2012–14) and genotypes 3a and 1a a decrease during the same period (6.8% in 2006–08, 4.7% in 2009–11 and 3.2% in 2012–14 and 7.9% in 2006–08, 4.7% in 2009–11 and 2.6% in 2012–14, respectively). Subtype 1b seems to be equally distributed between males and females (52.7% vs 56.6%) and the prevalence in the age range 31–40 years is significantly higher in the 2012–14 period than in both previous periods (53.8% vs. 16.6% in 2009–11, p< 0.001 and 13.4% in 2006–08, p < 0.001). Conclusions Genotype 1b is still the most prevalent, even if shows a significantly increase in the under 40 years old population. Instead, genotype 3a seems to have a moderate increase among young people. Overall, the alarming finding is the “returning” role of the iatrogenic transmission as risk factor for the diffusion of Hepatitis C infection.


Introduction
It has been greatly described that different hepatitis C virus (HCV) genotypes are strictly correlated to various evolution, prognosis and response to therapy during the chronic liver disease. Aim of this study was to outline the changes in the epidemiology of Hepatitis C genotypes in Southern Italy regions from 2006 to 2014.

Material/Methods
Prevalence of HCV genotypes was analyzed in 535 HCV-RNA positive patients with chronic Hepatitis C infection, selected during the period 2012-2014, and compared with our previous data, referred to periods 2006-2008 and 2009-2011.

Conclusions
Genotype 1b is still the most prevalent, even if shows a significantly increase in the under 40 years old population. Instead, genotype 3a seems to have a moderate increase among young people. Overall, the alarming finding is the "returning" role of the iatrogenic transmission as risk factor for the diffusion of Hepatitis C infection. PLOS  Introduction trade in the 1700 [10,32] and in some regions of Italy [17][18][19]33]. Subtype 3a, which is very common among intravenous drug abusers, is common mainly in Europe, USA and South East Asia while genotype 4 prevails in North Africa and Middle East and genotypes 5 and 6 are endemic, respectively, in South Africa and in South China [10,[31][32][34][35].
Data recently published showed that in the overall Italian population the predominant genotypes is 1 (64,7%), followed by 2 (26.0%), while genotypes 3 and 4 are both estimated under 4% [31], although these data seems to change considerably in the PWID population where the genotype 3 is the most prevalent (41,3%) followed by subtypes 1a (23,1%) and 1b (20,6%) [36][37]. It is important to consider, anyway, that these data were extrapolated from studies restricted to limited populations.
Focus of our paper was to provide updated finding on the changing epidemiology of Hepatitis C infection and on the prevalence of the various HCV genotypes in Southern Italy, studying three different groups of HCV positive patients in the periods 2006-2008, 2009-2011 and 2012-2014.  Exclusion criteria for the study were: HCV-RNA negative samples or, as previously specified, insufficient or unavailable serum for HCV-RNA determination or for genotyping. No significant differences of gender or age were found between the analyzed subjects and the excluded patients.

Study population and sample collection
Information about risk factors for Hepatitis C were available for all the included patients. In case multiple risk factors the one most likely to associated with an increase risk of disease was assigned. Anyway, it is interesting to note that, while dental therapy is an isolated risk factor (over 95% of the selected patients), we detected a very close correlation between drug abuse and tattoos (56%). Instead, surgery, is generally equally associated to all the other risk factors. It is important to highlight, anyway, that even counting each risk factor separately, analysis of the results doesn't change significantly.
All patients enrolled were asymptomatic, negative for anti-HIV, HBsAg and anti-HDV and had no clinical symptoms or biochemical markers of other chronic liver disease (autoimmune disorder, non-alcoholic fatty liver disease, hemocromatosis, etc.). No patient admitted alcohol abuse defined as the consumption above 30g pure alcohol per day for females and 40g / day for males in the last 6 months.
No differences in pattern of sampling or testing used in the three considered periods. The Ethic Committee of Istituto Nazionale Tumori-Fondazione "G. Pascale"granted approval for the study that was conducted according the principles of ICG-GCP and declaration of Helsinki. Prior written informed consent was obtained from each patients and all data was deidentified using data collection.

Serology and molecular analysis
We detected the positivity of anti-HCV antibodies in all plasma samples by Vitros ECi test (Ortho Clinical Diagnostics), used according to the manufacturers' instructions.
The Ortho-Clinical Vitros ECi test is an immunoassay system in which a positive antibodyantigen reaction create a light signal that is directly proportional to the amount of antibody present. Results are reported as signal to cut off and values�1.00 are considered reactive for HCV antibodies. The test is highly specific and sensitive (99.97% and 100%respectively).
Only repeatedly anti-HCV positive samples were subsequently tested for the detection and quantification of viral genome by Polymerase Chain Reaction (PCR) in Real Time by means of COBAS Ampliprep/COBAS Taqman HCV (Roche Diagnostics System Inc.). Linear range of assay was 1.50 x 10 1 to 6.90 10 7 IU/ml, with the accuracy acceptance criterion of +/-0.3 log 10 . The test had a specificity of 100% and its limit of detection (LOD) was 15 IU/ml. HCV genotyping was assessed using the Versant HCV Genotype Assay 2.0 LiPA, a fully automated system (Siemens Healthcare Diagnostics). Viral genome is first amplified and then viral fragments are hybridized by means of genotype-specific probes bounded onto nitrocellulose strips. The test have a specificity and sensitivity of 96% and 99.4% respectively, and its LOD is 15 IU/ml. No mixed infections were described.

Statistical analysis
All statistical analysis was done through SPSS software, version 17. The χ 2 tests was used to analysed the frequency tables and the correlation between the categorical variables was assessed using the Pearson correlation. Student's t test was used to analyse differences in the mean ages in the genotypes and subtypes. In all tests, p-values< 0.05 indicates a strong evidence for statistical significance.
Multiple logistic regression analysis was used to estimate the odds ratio of genotype prevalence in relation to gender and age in the analysed periods.

Results
Prevalence of anti-HCV during the nine-year of analysis (from 2006 to 2014) is reported in Fig  1 and shows a marked decrease (-2.8%, p<0,00001), while the viraemic rate in the same period exhibits an evident increase (+ 13.0%, p<0.005).
The prevalence of the genotypes among the patients during the three analysed periods is illustrated in Fig 2. All subtype 3 genotype were subtype 3a, and no case of genotypes 5 and 6 In this paper, subtypes 1 � and 2 � (not further subclassified) constitute the 3rd and the 4th largest groups, respectively, in the 3-year period 2009-2011 (9.4 and 8.2%) and 2012-2014 (3.8% and 6.5%). During the nine years studied subtype 3a gradually decreased from 6.8 to 3.2%, as genotype 1a from 7.9 to 2.6% (p<0.05). It is important to point that the distribution of subtypes 1a and 1b during the nine years could significantly change since the exact classification of not-further-subtypable genotype 1 � is not completely clear. Genotype 4 during the analysed period proved to be the most constant with an overall prevalence of 1.0% (Fig 3).
Gender-related differences observed among the subtypes are reported in Table 1. Genotype 1b in the past more frequently observed among females than males (39.3% of 89 males vs.    To further compare what kind of HCV genotypes prevails in accordance with age patients, these were grouping into 5 age groups, as shown in Tables 2, 3 (Figs 4-6).
Instead, the prevalence of genotype 2 (including subtype 2a/2c and non-subtypable genotype 2) showed no difference among age groups throughout the analysed periods with the exception for the 51-      Multiple logistic regression analysis was used to assess the meaning of HCV genotypes prevalence during the nine-year periods. The different distribution of genotypes was independently associated with gender and age. Our analysis showed that all the differences in HCV genotype distribution were significant and not influenced by the variability of gender or age group.

Discussion
HCV genotypes are widely distributed throughout the world and their distribution varies mostly according to the geographical region [10]. It is well known that HCV subtypes play a Hepatitis C virus (HCV) genotypes distribution in Southern Italy crucial role in the therapeutic approach, since the severity and the prognosis of the disease may vary greatly according to the different genotype [3,[38][39].
Although findings recently published show a global decrease of HCV infection, particularly in resource-rich countries [10], likely due to the introduction of the new antiviral therapies, some prospective modelization studies suggest that HCV-related mortality will increase in the next years [20][21] since the use of classic therapeutic approach "genotype-dependent" is still largely common, especially in resource-poor countries. This implies that a deeply knowledge of HCV epidemiology is still necessary to create strategies needed to eradicate HCV pandemic.
As previously globally described [10], also data obtained in this regional study suggest a decrease in the prevalence of HCV in Southern Italy from 2006-2008 to 2012-2014 (-1.6%) and a contemporary increase of the viraemic rate (+7.4%) for sure related to the aging of infected population.
Our data seems to suggest some interesting changes in the epidemiology of HCV genotypes in Southern Italy over the nine years. Genotype 1b is historically the most prevalent, both in Southern Italy and throughout the whole Italy [18,[40][41] and still remains so (54.4%), followed by genotype 2a/2c (28.7%) that however shows a marked increase from 2006-2008 to 2012-2014 (+7.0%). It is likely that the migration fluxes from Balkan area to Italy may have caused an increase prevalence of genotype 2 in the Southern Italy [17]. As we recently described, comparing our data with those collected by The Polaris Observatory [42], G2, typical of Albania [43], seems to increase its prevalence in the last years only in Italy (+12.0%) without any significant change in the other Southern Europe countries [18][19]33,[44][45]. Although some hypothesis suggest that G2 was probably introduced in Italy as a consequence of Albanian campaign during Second World War [33], it is more likely that the migration fluxes from Albania to Italy in the 90s may have increased its prevalence in the Southern Italy [17].
On the other hand, genotypes 1a and 3a, considerably less common in our area (2.6 and 3.2%, respectively), compared to data from other regions of Italy [41] exhibit a drastic decrease during the nine studied years (-5.3% and -3.6%, respectively), even though these data could be modified by the "not further classified" genotype 1 � whose classification is still unclear. Regarding genotype 4, recently documented in some Mediterranean countries, such as Albania, Spain and Greece [46][47][48] and probably related to migratory flows from Middle Eastern  Hepatitis C virus (HCV) genotypes distribution in Southern Italy countries and Northern Africa, its presence do not seems to be relevant in our area, at least until 2014, showing a stable prevalence in the three periods studied. The gender distribution of HCV genotypes shows, as previously described [8][9][10][11], a marked prevalence of genotypes 1a and 3a among males (4.1% vs. 0.7% and 4.1 vs 2.0%, respectively), probably due to a higher prevalence of PWID among men, even if, especially for genotype 3a, the male/female ratio seems to be decreased from 2006 to 2014. Subtype 1a, before equally distributed between females and males (8.1% vs. 7.9% in 2006-2008, 4.4% vs. 5.1% in 2009-2011) appears to have now a statistically significant increase in males. Conversely, genotype 1b in the past more frequently observed among females (64.3% vs. 39.3% in 2006-2008 and 54.0% vs. 41.5% in 2009-2011) seems not show any significant differences now between females and males (56.6% vs. 52.7% in 2012-2014).
Considering genotypes distribution related to age, genotype 1b, as widely reported in literature [49][50][51][52][53][54][55], is considerably more frequently in the older patients, especially over 50 years, although it is interesting to highlight its increasing in the 31-40 age group in 2012-2014 period if compared to the two previous periods (+37.2% respect to 2009-2011 and + 40.4% respect to 2006-2008). Since this genotype is historically correlated to transfusion-related transmission, this new peak in younger patients might be related to an emerging new transmission route. Instead, evaluating age distribution of genotype 2 (including subtype 2a/2c and non-subtypable genotype 2) it is evident its statistically significant increase in patients over 50 years in the three-year period 2012-2014 when compared to both the previous periods (+7.2% and +7.6%, respectively). This finding seems to confirm the reduced risk of transfusion-associated transmission in the years and in the meantime highlights the increasing role of no-age related risk factors in diffusion of HCV infection. On the contrary, subtype 3a is greatly present in younger patients showing a moderate increase in the age group < 30 years old (+10% from 2006-2008 to 2012-2014), even if these data must be confirmed for the reduced number of patients belonging to genotype 3 examined.
Data obtained by analysis of risk factors in the three studied periods show clearly that blood transfusions contribute minimally on the onset of new infections because of the more and more safer screening procedures globally adopted. Dental therapy, described in the first decade of new millennium as one of the major risk factor for the acquisition of HCV infection in Southern Italy [18], on the contrary, seems to have now a secondary role, as also documented by the increased prevalence of genotype 2 in the aged population. While PWID seems constantly to play a minor role in introducing new HCV infections in Southern Italian population, the increased incidence of surgery practices as risk factor and the increasing prevalence of genotype 1b in the 31- Genotype 1b is the most dominant among elderly patients, probably for a cohort effect. It was supposed that these patients, infected several years ago when genotype 1b was probably the only one existing in the area with an high incidence, have maintained their old infection, among a pool of younger carriers initially infected by genotype 1a and then by genotype 3a. In fact, a high prevalence of genotypes 3a in under-40-year-old subjects and of genotype 1a in 30-50 years old patients was observed in the period 2006-2008. This was not confirmed in the three-year periods 2009-2011 and 2012-2014 in which there was a reduction in the genotype 3a frequency in subjects under 30 years of age and in the meantime a moving of genotype 1a infected patients toward old age groups. The decreasing role of dental therapy as risk factor might explain the increasing of genotype 2 in the elder population. Multivariate analysis indicated that the different distribution of genotypes 3a and 1b during the analysed periods was independently associated with any variability in gender and/or age group.
By the analysis of our data it can be speculated that genotype 1b was the native "resident" one in Southern Italy, widespread in the seventies, especially through iatrogenic infections and the use of non-sterile syringes. The introduction of HCV screening programmes in the last decades of twentieth century have drastically reduced the infection risk, thereby generating the cohort effect we now describe. Genotype 2, the second "resident" genotype in Italy, whose circulation has been initially implemented through unsafe dental practices and then for immigration fluxes from Balkan area, especially from Albania, at the end of 90's, has become recently more and more common in the elder population. This effect could have been caused by the more and more safer dental procedures in our area as demonstrated by the reduced percentage of dental-related infections observed in the 2012-2014 period. Regarding intravenous drug abuse might have led to two different waves of HCV infection especially among young population, introducing before genotype 1a (now more common in aged patients) and then 3a [30].
In conclusion, the epidemiological framework of Hepatitis C infection in Southern Italy, particularly interesting for the high prevalence of this virus in the general population, seems to highlight the "returning" role of the iatrogenic transmission as risk factor for the diffusion of HCV infection. Furthermore, the small increase of genotype 3a among young people should be more investigated, with a support of a phylogenetic analysis.
At support of our hypothesis, some studies report small HCV outbreaks in Europe due to breaches in standards of health and safety practices among health-care workers [56]. Indeed, an interesting case-control study highlighted some unconventional routes of diffusion of Hepatitis C infection such as digestive endoscopy, beauty treatments and professional pedicure/ manicure [57]. This suggest not only a necessary evaluation of the safety practices in surgery, but the fundamental importance of not lowering the safety levels, especially among all healthcare professionals.