Nearly ten years after the withdrawal of CQ in the first-line treatment of malaria in the DRC, CQR strains remain above the warning threshold established by WHO. However the prevalence of pfcrt 76 T mutants continues to decline as shown in previous studies performed in DR Congo where a slow reversion in sensitivity has been recorded. Indeed, the percentage of Pf resistant strains went from 100% in 2000 , to 93% in 2002 , 83.8% in 2008  and 73.2% in the present study. Finally, over the last ten years, the prevalence of resistant strains has decreased by 26.8%. In other countries however, the decrease was much more significant like in Malawi where CQR fell from 85% to 13% in 8 years  after replacing CQ by SP or in Gabon with a drop of 55% (100% to 45%) in 6 years . This fact could be explained by the large area covered by DRC that is more difficult to control, and by the relatively high cost of ACT compared to former monotherapies, resulting in a slower anti-malarial policy shift. Until 2007, CQ was still the most widely used anti-malarial drug in the DRC  although it was officially removed since 2002 and firstly replaced by SP. Aside from the efficiency with which new drug policies were implemented, malaria transmission intensity in the region and the use of many other anti-malarial drugs in addition to the national policy have been suggested to explain the difference in trend .
The CVIET haplotype was the only resistant haplotype identified in the present study. Indeed, it is the most prevalent haplotype among the five discovered to date in Africa and it seems that it has migrated from Southeast Asia to Africa via India [26, 27]. Some studies have described this haplotype as a necessary but insufficient condition to confer AQ resistance [28, 29]. Beshir et al. suggested that if the CVIET haplotype is highly prevalent, AQ can still be effective and further mutations on the P. falciparum multidrug resistance gene (pfmdr1) are required for the development of clinically significant AQ resistance .
The only existing results on 72-76 pfcrt haplotypes that are available for the DRC refer to samples collected in 2000. The study identified 100% of resistant strains and these were distributed as follows: 14/27 (51.8%) SVIET, 12/27 (44.4%) CVIET and 1/27 (3.7%) of CVMNT . Studies on a larger scale should be more representative of the high prevalence of the SVIET haplotype, which appears to be rare elsewhere in the world. In fact, it was found before only once, in a study performed in Papua New Guinea .
Our results showed that the SVMNT haplotype is not yet present in Kinshasa suggesting that AQ remains valid as a partner drug for ACT. However, continuous monitoring is necessary because, initially absent from Africa, the SVMNT haplotype suddenly appeared in Tanzania in 2004, where it was not present the year before . This fact shows how the emergence of a new haplotype can be spontaneous. This haplotype has been also found recently in Angola by Gama et al. who supposed that this haplotype may have been imported from Brazil . Frosch et al. showed that Angola and Tanzania were among African countries where AQ was more extensively used than in other countries during the last decade . This fact could have contributed to the selection of SVMNT resistant strains, as suggested by Sa and by others, who stipulated that AQ had a prominent role in the selection of parasites carrying the SVMNT haplotype [10, 13, 32, 33]. In fact, SVMNT haplotype is predominant in countries where AQ was early and widely used .
Malaria is a disease of great geographical diversity and, as DRC is a very large country, the results of this study concern only the region of Kinshasa and further studies are needed.