Study population. A total of 1,140 children were recruited across the three sites studied. However, only microscopically positive samples were used to extract parasite DNA from a total of 377 samples from the sites for the molecular marker of resistance study. The baseline characteristics of the study population by the site are summarised in Table 1. Significant differences in mean age and temperature were observed between the study populations at the 3 sites, as shown in Table 1.
Table 1
Baseline characteristics of the study individual by site.
|
Sites
|
All sites
|
p-value
|
|
Banfora
|
Gaoua
|
Orodara
|
No. of patients
|
75
|
244
|
58
|
377
|
-
|
Males, %
|
52%
|
48%
|
47%
|
49%
|
0.8
|
Age (years)
|
|
|
|
|
-
|
mean (SD*)
|
5.7 (2.3)
|
5.60 (2.5)
|
6.7 (1.6)
|
5.8 (2.4)
|
< 0.001
|
range (min-max)
|
0.6–10
|
0.5–10
|
2.2–9.6
|
0.5–10
|
-
|
Temperature (°C), day 0
|
|
|
|
|
-
|
mean (SD*),
|
36.9 (0.5)
|
36.7 (0.6)
|
36.5 (0.4)
|
36.7 (0.6)
|
< 0.001
|
Parasitaemia (µl), day 0
|
|
|
|
|
-
|
median (95% CI)
|
827
(509-1,346)
|
940
(680–1240)
|
597
(276–1024)
|
800
(662–1059)
|
0.1
|
range (min-max)
|
24–544,000
|
16–207,273
|
16–180,870
|
16–544,000
|
-
|
SD : Standard deviation
Mutations associated with resistance to antimalarial drugs. Of the 377 samples analyzed, 346 (91.7%), 369 (97.9%), 368 (97.6%), and 374 (99.2%) were successfully sequenced for Pfcrt, Pfmdr-1, dhfr, and dhps. There was no significant difference between the sites, except for Pfcrt, where the proportion of missing data was lower in Gaoua (3.7% in Gaoua vs. 18.6% and 16.0% in Banfora and Orodara, respectively, p < 0.0001).
Mutations in Pfcrt genes. Pfcrt mutations were observed at 5 codons (C72S, M74I, N75E, K76T, and I356T/L). The highest frequency was observed at codon 76T (9.8%), followed by M74I (9.5%), N75E (9.5%), and I356T (4.0%). The frequency of the other two non-synonymous mutations were below 1% (I356L, 0.9%, and C72S, 0.3%). Pfcrt mutations at codons 74, 75, 76, and 356 were significantly more frequent in Orodara (Table 2).
Table 2
Distribution of the Pfcrt mutations by site.
Codons
|
amino acid change
|
Site
|
All sites
|
p-value
|
Banfora
|
Gaoua
|
Orodara
|
72
|
C > S
|
0.0%
|
0.4%
|
0.0%
|
0.3%
|
0.7
|
74
|
M > I
|
8.2%
|
7.7%
|
20.0%
|
9.5%
|
0.02
|
75
|
N > E
|
8.2%
|
7.7%
|
20.0%
|
9.5%
|
0.02
|
76
|
K > T
|
8.2%
|
8.1%
|
20.0%
|
9.8%
|
0.03
|
93
|
T > S
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
97
|
H > Y
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
145
|
F > I
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
218
|
I > F
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
343
|
M > L
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
350
|
C > R
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
353
|
G > V
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
356
|
I > T
|
4.9%
|
2.1%
|
12.0%
|
4.0%
|
0.02
|
I > L
|
0.0%
|
1.3%
|
0.0%
|
0.9%
|
A total of seven Pfcrt genotypes were identified. The wild-type genotype (CVMNKTHFIMCGI) predominated (89.3%), followed by the triple mutant (CVIETTHFIMCGI, 5.8%) and the quadruple mutant (CVIETTHFIMCGT, 3.5%). The remaining genotypes were found at frequencies below 1% (Table 3). The distribution of Pfcrt genotypes was similar between sites but differed significantly by sex (p = 0.035) with a 5-fold higher proportion of triple mutant genotype (CVIETTHFIMCGI) in females than males (Table 3).
Mutations in Pfmdr-1 genes. Pfmdr-1 variants were observed at 4 codons (N86Y, Y184F, S1034C, and N1042D). Codon 184F had the highest frequency (30.9%), followed by 86Y (1.9%). The proportion of the two other mutants were below 1% (1042D, 0.8%, and 1034C, 0.3%). The distribution of the mutations was similar between the different sites (Table 3).
Table 3
Distribution of the Pfmdr-1 mutations according to the site.
Codons
|
amino acid change
|
Sites
|
All sites
|
p-value
|
Banfora
|
Gaoua
|
Orodara
|
86
|
N > Y
|
1.4%
|
2.5%
|
0.0%
|
1.9%
|
0.4
|
184
|
Y > F
|
35.6%
|
26.9%
|
41.4%
|
30.9%
|
0.06
|
1034
|
S > C
|
0.0%
|
0.4%
|
0.0%
|
0.3%
|
0.7
|
1042
|
N > D
|
0.0%
|
1.3%
|
0.0%
|
0.8%
|
0.4
|
1246
|
D > Y
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
In total, six Pfmdr-1 genotypes were found. The single mutant genotype (NFSND) predominated (66.7%), followed by the wild-type genotype (NYSND, 30.4%) and the double mutant genotype (YFSND, 1.4%). The 3 remaining genotypes were detected at frequencies less than 1% (Table 5). The distribution of Pfmdr-1 genotypes was similar between sites. However, it differed significantly according to parasite density (p = 0.0002). The frequency of the single mutant genotype (NFSND) increased with increasing parasite density (49% for < 500 p/µL isolates, 75% for 501-5,000 p/µL isolates, 81.8% for 501 − 50,000 p/µL isolates and 85% for > 50,000 p/µL isolates), while the frequency of the wild type (NYSND) decreased as parasite density increased (47.6% for < 500 p/µL isolates, 22.9% for 501-5,000 p/µL isolates, 13.6% for 5,001–50,000 p/µL isolates and 15.0% for > 50,000 p/µL isolates).
Mutations in the dhfr gene. Mutations in the dhfr gene were observed at 3 codons (N51I, C59R, and S108N). The highest frequency was observed at codon 59R (69.8%). This was followed by codons 51I (66.6%) and 108N (14.7%). The 108N dhfr mutation was found to be significantly more frequent in Gaoua (Table 6).
Table 6
Distribution of the dhfr mutations by site.
Codons
|
amino acid change
|
Sites
|
All sites
|
p-value
|
Banfora
|
Gaoua
|
Orodara
|
16
|
A > V
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
50
|
C > R
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
51
|
N > I
|
60.8%
|
68.2%
|
67.2%
|
66.6%
|
0.5
|
59
|
C > R
|
70.3%
|
69.0%
|
72.4%
|
69.8%
|
0.9
|
108
|
S > N
|
6.8%
|
19.0%
|
6.9%
|
14.7%
|
0.006
|
164
|
I > L
|
0.0%
|
0.0%
|
0.0%
|
0.0%
|
-
|
A total of seven different dhfr genotypes were found. The double mutant genotype (ACIRSI) predominated (52.4%). This was followed by the wild-type genotype (ACNCSI, 29.4%), the triple mutant genotype (ACIRNI, 13.6%), and the single mutant genotype (ACNRSI, 2.9%). The remaining 3 genotypes were detected at proportions below 1% (Table 7). The distribution of the dhfr genotypes differed significantly between the sites (p = 0.02): the double mutant genotype (ACIRSI) had a higher proportion in Orodora (62.1%) compared to Banfora (55.4%) and Gaoua (49. 2%), the triple mutant genotype (ACIRNI) was higher in Gaoua (18.2%) than in Banfora (5.4%) and Oradara (5.2%), and the single mutant genotype (ACNRSI) was higher in Banfora (8.1%) than in Oradara (3.4%) and Gaoua (1.2%). In addition, the distribution of the dhfr genotypes was significantly different according to the parasite density (p = 0.0001) and the axillary temperature (p = 0.01). The frequency of the triple mutant genotype (ACIRNI) increased with increasing parasite density (2.7% for < 500 p/µL isolates, 9.9% for 501-5,000 p/µL isolates, 37.9% for 501 − 50,000 p/µL isolates and 40.0% for > 50,000 p/µL isolates), while the frequency of the wild-type (ACNCS) decreased as parasite density increased (45.9% for < 500 p/µL isolates, 26.8% for 501-5,000 p/µL isolates, 7.6% for 5,001–50,000 p/µL isolates and 0% for > 50,000 p/µL isolates). Similarly, the frequency of the triple mutant genotype (ACIRNI) increased with increasing body temperature (12.4% for isolates from individuals with < 37.5°C and 25.0% for isolates from individuals with ≥ 37.5°C), whereas the frequency of the wild type (ACNCSI) decreased with increasing body temperature (30.5% for isolates from individuals with < 37.5°C and 19.4% for isolates from individuals with ≥ 37.5°C).
Mutations in the dhps genes. Polymorphisms at codons 431, 436, 437, 540, 581, and 613 were observed. The highest frequency was observed at codon 437K (89.3%). This was followed by codons 436A (61.2%), 613S (14.4%), and 431V (1.1%). The other four mutations were below 1% (436F 0.3%, 436Y 0.3%, 540E 0.3%, and 581G 0.8%). The 436A dhps mutation was found to be significantly more frequent in Gaoua (Table 8).
Table 8
Distribution of the dhps mutations according to the site.
Codons
|
amino acid change
|
Sites
|
All sites
|
p-value
|
Banfora
|
Gaoua
|
Orodara
|
431
|
I > V
|
0.0%
|
1.2%
|
1.7%
|
1.1%
|
0.6
|
436
|
S > A
|
44.6%
|
68.6%
|
51.7%
|
61.2%
|
0.0008
|
S > F
|
1.4%
|
0.0%
|
0.0%
|
0.3%
|
S > Y
|
1.4%
|
0.0%
|
0.0%
|
0.3%
|
437
|
A > K
|
91.9%
|
88.4%
|
89.7%
|
89.3%
|
0.7
|
540
|
K > E
|
0.0%
|
0.4%
|
0.0%
|
0.3%
|
0.7
|
581
|
A > G
|
0.0%
|
1.2%
|
0.0%
|
0.8%
|
0.4
|
613
|
A > S
|
10.8%
|
16.9%
|
8.6%
|
14.4%
|
0.2
|
In total, fourteen dhps genotypes were found. The double mutant genotype (IAKKAA) and the single mutant genotype (ISKKAA) were predominant (37.7% and 37.2%, respectively). They were followed by the triple mutant genotype (IAKKAS, 12.3%), the single mutant genotype (IAAKAA, 8.8%), and the double mutant genotype (IAAKAS, 1.1%). The rest of the 9 genotypes had a frequency of less than 1% (Table 9). The distribution of dhps genotypes was similar between sites but differed significantly by parasite density (p = 0.006). The frequency of the single 436A mutant genotype (IAAKAA) increased with increasing parasite density (6.2% at < 500 p/µL, 9.2% at 501 to 5,000 p/µL, 12.1% at 501 to 50,000 p/µL, and 15. The frequency of the single 437K mutant genotype (ISKKAA) decreased with increasing parasite density (54.1% at < 500 p/µL isolates, 27.5% at 501–5000 p/µL isolates, 27.3% at 5 001–50 000 p/µL isolates and 15.0% at > 50 000 p/µL isolates).
Table 9. Distribution of the fourteen dhps genotypes detected per location.
Distribution of dhfr/dhps haplotypes. The most frequent dhfr/dhps haplotypes were the triple mutant ACIRSI/IAKKAA (23%), followed by the wild-type ACNCSI/ISKKAA (19%) and the double mutant ACIRSI/ISKKAA (14%). The quintuple mutant haplotype N51I/C59R/S108N + G437A/540E (ACIRNI/ISAEAA) responsible for SP treatment failures in adults and children (but still effective for use in intermittent preventive treatment in pregnancy) was detected in one isolate from Gaoua. A septuple mutant ACIRNI/VAKKGA (triple mutant 51I/59R/108N and quadruple mutant 431V/436A/437K/581G) was observed in 2 isolates from Gaoua (0.5%).