About 180 million people are infected worldwide with the hepatitis C virus (HCV) [1] . In Africa in general, HCV remains a public health problem; its prevalence varies from one region of the continent to another. A recent study on the prevalence of HCV among migrants from North and West Africa to Europe gives an average prevalence of 7.6% with a West African prevalence between 2.1% and 14.1% [2] . The HCV infection has long existed in West Africa [3] . Sub-Saharan Africa has about 20% of global infections [4] . In Burkina Faso, its prevalence is around 1% [5] . Compared to HBV (14.7%), this prevalence appears low. However, the absence of a vaccine and the high cost of treatment make HCV a major health problem in the country. It is responsible for over 900 deaths per year in Burkina [6] . In general, because of the lag between HCV infection and complications that ensue (cirrhosis, hepatocellular carcinoma, decompensating hepatitis), morbidity related to HCV may increase in the next decades [7] .

Until recently, HCV has had six identified genotypes. A new genotype was recently identified, which brings the number of genotypes of the virus to 7, with 67 subtypes [8] - [10] . This genetic diversity is a challenge for vaccine development. Also, the dosage and duration of treatment depend on the genotype implicated in the infection [11] . For example, sofosbuvir (SOF) and ribavirin (RBV) are safe and effective drugs for the treatment of patients with HCV genotype 2 infection [12] , while the direct-acting antivirals (DAAs) with pan-genotypic activities (Simeprevir, sofosbuvir, and Daclatasvir) are recommended in triple regimens with PEG-IFN/RBV for the treatment of HCV genotype 4 [13] . Identification of the genotype implicated in an infection is of great importance.

The objective of this study was to determine the current genotypes of HCV in Burkina Faso and to compare the results of rapid tests to those of molecular methods.

Previous studies have reported that HCV genotypes circulate among blood donors in Burkina Faso. Genotypes 2 and 3 were found to be the most common [6] . Apart from blood donors, HCV genotypes in Burkina Faso have not been well investigated. This study sought to identify genotypes in first-time blood donors and in patients found to be HCV seropositive at the clinics, and take a critical look at the rapid tests used extensively in sub-Saharan Africa by comparing their results with those of molecular methods. The results of confirmation tests by Monolisa gave

22.35% of confirmed HCV antibody-positive status ( Table 1 ), with over 77% of not confirmed and hence considered as false positives. A big difference between the number of HBV positives by ELISA and by PCR (200 versus 41) has already been reported by Tao et al. among blood donors in Burkina Faso in 2015. In 2010, in Uganda, for HCV, significant variation between the results of rapid tests and ELISA (14/48) has been reported [14] . This high proportion of false positives could be linked to the performance of the serological tests used. These test kits are imported from the global North and may be degraded if the conditions of transport and / or storage are far from what they should be. It is also known that the general diagnostic performance of these antibody tests is low, and false positives are more likely to occur due to previous HCV infection in the individuals. A total of 47.36% of HCV antibody-positive individuals were confirmed RNA positive ( Table 2 ), a disagreement of about 52% between HCV sero-positive by Monolisa and the results of molecular analyses. This discordance is, however, a big surprise. Indeed, Zeba et al., in 2013, had found a 1.5% positive RNA, which is a difference of nearly 99% between serological and molecular results. This could be explained by the fact that most patients have not yet detectable viral RNA. Mullis and colleagues reported in Uganda in 2013 that 7.6% of undetectable viremia in 1000 HCV-infected people [15] . However, it is not possible to consider this as a fact because NTAGIRABIRI and colleagues in 2014 achieved 100% positive RNA from HCV antibodies positive in Burundi [16] . Response to treatment, long-term prognosis, and seroconversion profile are influenced by genotype. This is why it is very important to know the genotype of an infected person. In this study, we report genotypes 1 and 2 ( Table 3 ) as those encountered in Burkina Faso. Among blood donors, the identified genotype is genotype 2. Zeba et al. have also found this, mainly (56.3%) among blood donors, in addition to genotypes 3 and 4. Genotypes 2 and 3 are the most virulent and account for the majority of infections worldwide. Our results corroborate those of other studies. Indeed, genotype 2 is considered to be from West Africa [17] . Candotti et al. reported in 2003 that, in West Africa, genotype 2 (2a) is responsible for the majority of infections [18] . For samples for which the subtypes could not be clearly determined, sequencing of the core region may be able to remove the ambiguity.

Viral loads that range from 10,542 to 2,057,042 IU/ml are high. For treatment, we suggest the use of the combination pegylated interferon alfa 2a and ribavirin, which gave a sustained virological response (SVR) of 92.3% for genotype 2 and 81.2% for all genotypes combined in patients in Burkina Faso in 2010 [19] . Nevertheless, further studies on the efficacy of treatment for genotype 2 remain indicated [20] .

Limitations of the study

The small number of patients does not allow for the study of the correlation between socio-demographic characteristics and HCV infection. This small number constitutes a limit to the statistical strength of the study.

Genotype 2 is the most frequent both in blood donors and patients. The identified Subtypes of genotype 2 are subtypes 2f and 2d; subtype 1d for genotype 1. We can conclude that genotype 2 is the most common genotype in Burkina Faso. The genotype 1, although rare in West Africa, is present in our country (11.11%). There is a high percentage of false positives by RDT; the causes of this situation need to be known to avoid harmful consequences. In the meantime, we suggest to the health authorities that particular attention be paid to the conditions of transport and storage of rapid diagnostic tests in particularly hot countries.

The Ethics Committee for Health Research (CERS) of Burkina Faso has approved this study (Reference No. 2013-7-065 of July 11, 2013). All participants provided a written informed consent to participate in the study. Anonymity and confidentiality with respect to the information collected were scrupulously respected.

All relevant data supporting the findings are contained within the manuscript. The dataset is available from the corresponding author upon a reasonable request.

IT and EY collected samples; IT and DC performed molecular analysis and data analysis; IT and ATY drafted the manuscript; FWD and CB contributed to the study design and revised the manuscript; OAK revised the manuscript; DOY and JS designed the study and revised the manuscript for important intellectual content.

The authors are grateful to the Ministère de l’Enseignement Supérieur, de la Recherche et de l’Innovation of Burkina Faso for their financial support and to the Embassy of France in Burkina Faso for organizing the concours MT180. They also thank the staff of the National Blood Transfusion Centre (CNTS), Burkina Faso, of HOSCO, and the staff of the former National Blood Transfusion Institute (INTS) of Paris.