In the present study, the DELI-test was used to determine the profile of in vitro sensitivity of P. falciparum and P. vivax field isolates in Manaus, in the Brazilian Amazon. Despite some initial technical difficulties in the beginning of the study (2004), the performance of the test in 2005–2007 was similar to that observed in previous studies in Africa, which reported test success rates of 81% for P. falciparum.
Overall, a high prevalence (78.1%) of P. falciparum isolates resistant or with decreased sensitivity to chloroquine was observed. This was expected since earlier studies conducted in Amazonia suggested that 100% of the P. falciparum isolates circulating in the area were chloroquine-resistant . In addition, most of the “sensitive” isolates had IC50s close to the threshold for sensitivity (100 nM) and therefore represent rather borderline isolates. Only 2/64 (3.1%) of isolates showed clearly low IC50 to chloroquine. Any evidence of the re-emergence of chloroquine-sensitive parasites in Brazil through the documentation of a wild pfcrt P. falciparum haplotype (CVMNK) in Amazonian region was already reported .
In Brazil, reduced in vitro and in vivo sensitivity to mefloquine has been reported since 1981 [19–23] and this might explain the 21.2% resistant phenotypes in the present study. In fact, even before the official introduction of mefloquine in Brazilian endemic areas in 1987, P. falciparum isolates with profile of resistance to mefloquine had already been reported .
In relation to quinine, the prevalence of resistant isolates (IC50 above 500 nM) was 12.7%. Some isolates showed IC50s very close to the threshold. In any case, this scenario is worrisome because a continuous decreasing in parasite sensitivity has been noted in the Brazilian Amazon [17, 21, 24–27] and quinine plus doxycycline is actually recommended as P. falciparum second-line treatment by the Brazilian Malaria Control Programme.
Since 2007 the Brazilian Malaria Control Programme defined artemisinin-based combined therapy (ACT) as the first-line treatment for uncomplicated falciparum malaria. In this study, 11.7% of the isolates showed IC50 above the threshold of 10 nM (11.7%) for artesunate. This is surprising as these samples were collected in a period before the introduction of ACT in the area; therefore, no selection pressure is to be accounted for. A study performed in another area in the Amazonian region in the same period, showed no evidence of decreased sensitivity of local isolates to artesunate or artemether . However, a recent study showed that 23.9% of the studied isolates from Gabon had a reduced susceptibility to dihydroartemisinin . Decreased sensitivity of P. falciparum parasites to artemether in vitro in French Guiana and Senegal  as well as in vivo in Cambodia  has also been described. Given the immense value of artemisinin derivatives as alternative drugs for multidrug-resistant parasites, authorities must be alert to these data indicating a trend for emergence of isolates with decreased sensitivity to this drug.
The sensitivity of P. vivax parasites to anti-malarial drugs has not been widely monitored in vitro due the difficulties in cultivating P. vivax. Here, the in vitro resistance of P. vivax fresh isolates to chloroquine and mefloquine was evaluated for the first time in Brazil. Although the threshold of IC50 to define a sample as resistant to chloroquine is not well established for P. vivax, it has been proposed that the same threshold used for P. falciparum should be used for P. vivax. In this case, a considerable frequency (9.8%) of P. vivax isolates was above the adopted 100 nM threshold of chloroquine. This frequency was not surprising, because P. vivax resistance to chloroquine has been reported in Brazil [31, 32], as well as increasing numbers of severe P. vivax malaria cases [33, 34]. Consequently, the use of DELI test can be an important tool for backing decisions of the Brazilian health authorities in relation to the treatment policy of P. vivax.
Mefloquine is usually an alternative treatment for P. vivax infections. Interestingly, the prevalence of P. vivax isolates with resistant profile to mefloquine was greater (28%) than to P. falciparum parasites (21.2%), corroborating a study performed in Thailand where P. vivax isolates were more resistant to mefloquine when compared with P. falciparum isolates from the same area .
It should be highlighted that the in vivo outcome depends on a number of factors that cannot be evaluated in vitro, including the level of innate and acquired immunity. However, in vitro assays act as a preliminary warning system indicating a trend, since in vitro resistance may be indicative of clinical resistance.