Screening of clinical samples

A total of 183 samples were included but only 147 were analyzed with the LAMP Alethia® The malaria method, the gold standard of positivity in this study, to determine if there was a Plasmodium infection or not (Supplementary File 1). The results showed 104 LAMP positive samples of which only 86 were positive by microscopy and 43 negative.

For the remaining 36 Plasmodium spp. positive samples included in this study, but not tested with the LAMP method, DNA extracts were still used to compare species diagnosis and to correlate cycle thresholds (Ct) and parasite density for the different methods of qPCR.

Average values ​​of the peaks of the melting curve (Tm) for Plasmodium spp. and P. falciparum

The internal HRM-qPCR test identified the presence of a Plasmodium spp. or one P. falciparum infection. Plasmodium The spp infections displayed a specific Pan melting curve (Tm1) peak value, which was reproducible regardless of species or species combination. In case of a P. falciparum infection, a second Tm value specific to P. falciparum (Tm2) was also present (Table 2). As shown in Fig. 1, the Tm of P. falciparum differs markedly from that of Plasmodium spp. (Stove).

Table 2 Mean values ​​of Tm and their standard deviations (SD) for P. falciparum and Plasmodium spp other than P. falciparum
Fig. 1

Peaks of the derived melting curve (Tm) for P. falciparum and Plasmodium spp. after the HRM phase. A Tm peaks for both P. falciparum and Plasmodium spp. targets in case of P. falciparum carriage. B Tm peak for the Plasmodium spp. target in case of Plasmodium spp. other than P. falciparum carriage. The X axis represents temperature (°C). Y axis represents fluorescence (derived melting curve)

Assessing the sensitivity and specificity of qPCR methods

The 147 isolates tested with LAMP were evaluated with the four different qPCR methods. All 43 LAMP negative samples were also negative by all four methods, giving all of them 100% specificity. Of the 104 positive samples, five were not detected by Plasmodium Type (Bio-Evolution), three by Ampliquick® Malaria and two by HRM-qPCR but all with internal TaqMan-qPCR (Table 3).

Table 3 Sensitivity and specificity of microscopy and the various qPCR methods studied compared to LAMP Alethia® Malaria

Compared to LAMP, internal qPCR TaqMan and HRM were the most sensitive (sensitivity = 100%, 95% CI [100–100] and 98.1%, 95% CI [93–99.8] respectively), followed by the two commercial kits: the Ampliquick® Malaria test (sensitivity = 97.2%, 95% CI [91.8–99.4]) and Bio-Evolution Plasmodium Type (sensitivity = 95.4%, 95% CI [89.1–98.4]). Microscopic techniques had a sensitivity of 85.2% (95% CI [74–89.4]). The Kappa coefficient shows that the qPCRs tested are very consistent with the LAMP method, in particular the in-house qPCRs (Table 3). All samples positive by microscopy (n=86) were also positive after analysis with the four qPCR methods tested.

Species diagnostic accuracy

On each sample included, a microscopic diagnosis and an identification of the species by qPCR were systematically carried out (Supplementary File 2). Bio-evolution Plasmodium Typage is a qPCR kit commonly used for malaria diagnosis at FNMRC. It allows the simultaneous identification of P. falciparum, P. oval, P. vivax, P. malariae and P. knowlesi and was therefore considered the reference method for the Plasmodium identification of the species when the sample was positive by this method.

Analysis with the Bio-Evolution kit showed 135 positive and 48 negative samples. Of the 135 positive samples: 73 were identified as P. falciparum, 36 were identified as P. oval spp, 8 have been identified as P. vivax, 15 have been identified as P. malariae, 2 mixed infections per P. falciparum and P. malariae and mixed infection P. falciparum, P. malariae and P. oval.

These results were compared to other techniques (Table 4). The results for P. falciparum, P. oval, P. vivax and mixed infections were correlated regardless of the qPCR method used. Moreover, of the 15 P. malariae samples positive by the Bio-Evolution kit, one turned out to be a mixed infection demonstrated by the in-house TaqMan since this test confirmed the presence of P. falciparum also (Table 4).

Table 4 Accuracy of Plasmodium identification of species with the various qPCR methods studied in relation to Plasmodium Type (Bio-Evolution)

The five negative isolates with Bio-Evolution Plasmodium typing but positive with LAMP Alethia® Malaria was positive (four P. falciparum and an Plasmodium spp. other than P. falciparum) after the internal TaqMan-qPCR analysis, which was correlated with the initial LAMP results (Table 5).

Table 5 Correlation between Alethia® Malaria (LAMP) and the different qPCR for discordant samples with Bio-Evolution

Tables 4 and 5 show that for internal TaqMan-qPCR, the 18S rRNA target (Pan) was positive in 96% (74/77) of P. falciparum mono-infections. The remaining three mono-infections were only positive for the var gene target (P. falciparum).

Correlation between parasite density and qPCR cycle threshold values ​​(Ct)

The median Pan-Plasmodium Ampliquick side® Malaria (median Ct = 23.40), the species Ct of Bio-Evolution Plasmodium typing (median Ct = 27) and undifferentiated Ct of internal HRM-qPCR (median Ct = 22.88) and internal TaqMan-qPCR (median Ct = 23.58) were calculated. Data shows that in-house qPCR and commercial Ampliquick® Malaria has lower Ct values ​​than Bio-Evolution Plasmodium Typing (pooled Ct for all species) (p var genes and 18S rRNA targets were not different (24.2 vs 23.5, p=0.31, Mann–Whitney U test).

As shown in Figure 2, the comparison of the parasite density determined by microscopy and the Ct values ​​of the 135 positive samples with the different qPCR methods shows a clear correlation between the four methods and the parasite density (all linear regression p-values ​​are

Figure 2
Figure 2

qPCR Ct values ​​compared to parasite density. Linear regression between parasite density and Pan-Plasmodium Ct of Biosynex Ampliquick® Malaria (red), the Ct species of Bio-Evolution Plasmodium typing (green), the undifferentiated Ct of internal qPCR-HRM (blue) and internal qPCR-TaqMan (purple). 95% confidence interval (grey) and coefficient of determination R2 are indicated