| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We previously reported that parasite mutations that confer low-level resistance to SP may contribute to the potential for transmission of Plasmodium falciparum and the spread of resistance.1 In a setting with an SP treatment failure rate less than 4%, point mutations in parasite dihydrofolate reductase (DHFR) that confer in vitro resistance to pyrimethamine were associated with longer parasite clearance time (PCT) and the presence of gametocytes (the sexual form of the parasite responsible for transmission by the mosquito vector) after SP treatment. This suggested that even before clinical SP resistance is apparent, drug treatment that eradicates the asexual parasites that cause disease may still lead to the spread of resistance by selecting for resistant parasites that survive and propagate in the form of gametocytes.
More than 50 years ago, in vitro studies showed that strains of P. gallinaceum that became resistant to antifolates by growth under drug pressure exhibited a higher count of gametocytes.3 More recently, chloroquine treatment of P. falciparum has also been shown not only to lead to an increase in gametocyte numbers, but also to increased numbers of oocysts in mosquitoes fed on patients infected with chloroquine-resistant parasites.4,5 The transmission potential of SP-resistant parasites is unknown. Our recent data highlight the importance of investigating the role of low-level drug resistance mutations on the selection and the spread of drug resistance in its earliest stages.
In this study, we assessed the association between the occurrence of mutations in P. falciparum DHFR and dihydropteroate synthase (DHPS) and transmission capacity of parasites to Anopheles mosquitoes. In particular, we compared the prevalence of DHFR and DHPS point mutations before (in asexual parasites) and after (in gametocytes) SP treatment and determined parasite infectivity to Anopheles mosquitoes of post-treatment gametocytes with and without DHFR and DHPS mutations.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study population. Subjects were recruited from outpatient clinics and at a field research station in Buenaventura. Malaria patients who fulfilled entry criteria were invited to participate in the study. Inclusion criteria were an age more than 5 years, a positive blood smear for P. falciparum malaria, informed consent from participant or parent, and intention to remain in the study area for at least four weeks from the time of enrollment. Patients with mixed species infections and those with clinical symptoms or laboratory results confirming or suggesting complicated malaria were excluded.
Ethical issues. The protocol and the informed consent document were reviewed and approved by the Universidad del Valle Ethical Review Board in Cali, Colombia, before commencement of the study. The informed consent document was read to the subject, parent, or guardian of prospective participants and a copy was given to the subject.
Treatment and assessment of parasite and clinical responses. Individuals received a single oral dose of SP equivalent to 1.25 mg/kg of pyrimethamine 25 mg/kg of sulfadoxine up to a maximum dose of 75 mg of pyrimethamine and 1,500 mg of sulfadoxine. To monitor for adverse reactions and to make sure the medicine was well tolerated, all subjects were observed for at least 60 minutes. If vomiting occurred within 30 minutes, the full dose was repeated, and if vomiting occurred between 30 and 60 minutes, half of the dose was repeated. All patients also received an insecticide-impregnated bed net to decrease the likelihood of reinfection.
On enrollment and prior to treatment, symptoms and findings of a physical examination were recorded. We used a slightly modified version of the 2003 World Health Organization (WHO) protocol for measuring antimalarial drug efficacy in areas with low transmission.10 Follow-up was by active surveillance visits on days 1, 2, 3, 7, 14, 21, and 28, with an additional evaluation on day 4 if the blood smear remained positive on day 3. Every visit included physical examination, determination of blood hemoglobin concentration, and blood smear for malaria diagnosis.
Thick smears were stained with Field’s stain and read immediately for initial parasitemia screening. An experienced laboratory technician measured asexual parasitemia (asexual parasites/microliter of blood) by reading 200 high-power fields on a second Giemsa-stained thick blood film. Thin smears were preserved for Giemsa staining and parasite identification. We defined PCT as the number of days to the first negative parasitemia after treatment.
Polymerase chain reaction (PCR) detection of DHFR and DHPS mutations. At days 0, 7, 14, 21 and 28 after treatment, finger stick blood from infected individuals was blotted onto strips of filter paper, air-dried, and stored at room temperature in separate plastic envelopes for PCR analysis of DHFR and DHPS mutations. Parasite DNA was extracted using a simple methanol fixation heat extraction method that relied on chelating resin extraction only for occasional samples that did not amplify well after methanol extraction. The PCR methods to assess parasite mutations were applied according to protocols described elsewhere11 (available from http://medschool.umaryland.edu/CVD/plowe.html).
Our analyses were limited to mutations that have been shown to be of primary prevalence in the area: DHFR mutations at codons 108 (serine to asparagine) and 51 (asparagine to isoleucine) and DHPS mutation at codon 437 (ala-nine to glycine). Evaluation of other mutations was initiated and discontinued if a mutation was not found in the first 50 samples tested during follow-up: DHFR-59 (cysteine to argi-nine), DHFR-164 (isoleucine to leucine) DHPS-581 (alanine to glycine) and DHPS-540 (lysine to glutamate). For all assays, infections were determined to be wild-type, mutant, or mixed with respect to each mutation site, based on agarose gel electrophoresis of diagnostic PCR or restriction digestion products. Samples found mutant at DHFR-51 that were not amplified for DHFR-108 were considered double mutant because this last point mutation always precedes subsequent mutations. In addition, samples found with no mutant at DHFR-51 and not amplified for DHFR-108 were grouped as wild or DHFR-108 only mutant.
Gametocytemia (transmission potential). From samples taken to evaluate parasite response from days 0 to 28, we also measured sexual stage (gametocyte) production. Using the same technique described above to count asexual forms, we quantified the number of gametocytes/microliter.
Parasite infectivity to Anopheles mosquitoes. Blood samples were collected by venipuncture of patients on days 0, 7, 14, 21, and 28 after treatment. Five milliliters of whole blood were collected in tubes containing EDTA as anticoagulant. Blood cells were separated from plasma by centrifugation at 3,000 rpm for 5 minutes and plasma was replaced by a similar volume of normal human plasma known to sustain malaria transmission. This step was performed to avoid the possible interference (blocking or enhancing) antibodies to gametocytes in the donor’s plasma. Laboratory-reared An. albimanus mosquitoes of the Buenaventura strain were used for the evaluation of transmission capacity by using membrane feeding assays (MFAs) at the field site laboratory as described elsewhere.12 Briefly, batches of 200–300 mosquitoes per cage were used 2–3 days after emergence. All experiments were conducted at room temperature (approximately 25°C) within one hour after patient bleeding. A total of 2–3 mL of the test blood was injected into glass feeders maintained at 40°C and mosquitoes were allowed to feed for approximately 15 minutes. After feeding was completed, mosquitoes were maintained at standard rearing conditions (temperature = 27°C and relative humidity = 80%). Seven to eight days after feeding, surviving mosquitoes were dissected and analyzed to determine the presence of oocysts in the midgut. The proportion of infected mosquitoes and the oocysts count were determined for each mosquito batch.
Statistical analysis. In evaluation of the selection of the mutant parasites, the exposure of interest was the status of mutations in asexual parasite genes before therapy and the endpoint of interest was the status of the mutations in the sexual forms after treatment. We used the McNemar test for this paired analysis. To control for potential confounders (e.g., level of parasitemia), we used conditional logistic regression whereby each participant contributed two records for the analysis: one before treatment (i.e., unexposed) and one after treatment (i.e., exposed).
We also performed an analysis based on marginal probabilities by determining whether the 95% confidence interval (CI) for the overall probability of having two mutations after treatment (p) included the probability of having two mutations before treatment (q). It was expected that if the lower bound of the 95% CI of p was greater than q, we could conclude that there was proliferation of mutations linked to increasing resistance.
We determined infectivity before and after SP treatment. The proportion of infected mosquitoes and the mean number of oocysts among infected mosquitoes were calculated. We determined the significance of the infectivity by using exact methods to calculate the 95% CI of the average proportion of infectious mosquitoes. To test for the potential heterogeneity of this proportion according to the type of mutations in the sexual form, we developed an empirical logistic regression with the observed log odds of infected mosquitoes as the outcome. We allowed proportions equal to zero to be included in this analysis by replacing with a constant (i.e., 0.5) all MFA results with zero infected mosquitoes. In addition, we carried out simple linear regression methods to similarly test for heterogeneity in the mean number of oocysts counts among infected mosquitoes.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Table 1
summarizes treatment responses and follow-up periods among all study participants. According to the standard criteria (WHO test) for parasitologic and clinical evolution after SP therapy, seven patients developed an early treatment failure (i.e., during the first three days after treatment). All received rescue treatment with quinine and were followed-up until complete recovery. Adequate clinical and parasitologic response was observed in 16 patients with follow-up periods of 7–21 days (16 of 155, 10.3%) and in 132 patients with complete follow-up until day 28. The per-protocol treatment efficacy (including only patients with complete follow-up) was 95.0% (132 of 139).
|
. Mutations at DHFR codons 59 and 164 and at DHPS codons 581 and 540 were not found in this study after processing all samples at enrollment. During follow-up, after processing at least half of the available samples, these mutations were also found to be absent and assays for them were discontinued.
|
). Marginal probability of finding any DHFR mutation (108 mutant only or 108 and 51 mutants) increased from 86.3% (95% CI = 79.8–91.3) on day 0 to 93.6% (95% CI = 82.5–98.7) and 100% (95% CI = 90.3–100) on days 21 and 28, respectively. Double-mutant parasites were less frequently observed at enrollment (i.e., 58.8% among asexual parasites) than on day 7 (i.e., 70.8% among gametocytes), but this difference was not significant (P = 0.17).
|
). Of the 19 individuals with discordant mutation status in days 0 and 7, 15 individuals whose infections on day 0 were not mutant or 108 single-mutant had double-mutant infections by day 7 (P = 0.019). The magnitude of the association (odds ratio [OR] = 3.8, 95% CI = 1.2–15.5) confirmed a substantial force of selection of mutant parasites seven days after SP treatment. This association remained significant and was even stronger after adjusting by the level of parasitemia (OR = 4.9, 95% CI = 1.2–20.0). The corresponding estimates (i.e., OR) for matched comparisons of mutant occurrence between days 0 and 14 and between days 0 and 28 were 2.3 (95% CI = 0.8–7.4) and 7.0 (95% CI = 0.9–315.5), respectively.
|
Transmission potential (gametocytemia after treatment). At enrollment, 13 (8.4%) of 155 individuals were carrying gametocytes. Gametocytemia peaked by day 7 after treatment when 112 (75.7%) of 148 individuals had detectable gametocytes, and the mean log10 gametocytemia was 2.56, which corresponded to 367 gametocytes/µL (n = 112). On days 14, 21, and 28, the corresponding statistics were 70.4% (100 of 142) and 282 gametocytes/µL, 58.8% (80 of 136) and 184 gametocytes/µL, and 40.2% (53 of 132) and 128 gametocytes/µL, respectively.
Infectivity. At enrollment, most the MFAs showed no infection of mosquitoes because only 17 (11.0%) of 154 mosquito batches were positive for oocysts after mosquito feeding. Positivity of MFA peaked by day 7 after SP treatment when 73 (50.3%) of 145 batches showed positive mosquitoes, and decreased thereafter to 40.7% (57 of 140), 32.2% (29 of 90), and 14.1% (9 of 64) on days 14, 21, and 28, respectively. The average proportion of mosquitoes with oocysts per each batch was 2.9% at enrollment, peaked by day 7 (13.5%), and decreased to 9.7% (day 14), 4.1% (day 21), and 0.5% by the end of follow-up (day 28). Mean numbers of oocysts among positive assays were 4.8, 11.2, 6.8, 2.0, and 1.8 oocysts/mosquito by days 0, 7, 14, 21, and 28 after treatment, respectively. These results not only show that gametocytes after SP treatment are infectious but also suggest that infectivity reaches its peak 7 days after treatment with a higher probability of infected mosquitoes and a higher number of oocysts.
Mutations and infectivity.
Based on the observed selection of mutant parasites after treatment (i.e., by day 7 OR = 3.8 for selection of double-mutant 108 and 51 DHFR codons), and the concurrent higher probability of transmission in mosquitoes, we evaluated the relationship between occurrence of resistance-conferring mutations in parasites and infectivity to mosquitoes. Figure 1 shows the proportions of positive mosquitoes according to DHFR status and day of follow-up. Parasites found with point mutations at 108 and 51 DHFR codons showed higher proportions of infective mosquitoes before and after treatment. The OR of finding a mosquito infected reached statistical significance (P < 0.05) 7–14 days after treatment when point mutations at 108 and 51 were compared with no mutant infections (Table 5). During these days, DHFR double-mutant infections had 7–10 times the odds of infecting mosquitoes than infections with no mutations, and infectivity tended to decrease after that. In addition, among positive mosquitoes, the mean number of oocysts showed no differences (P > 0.10) between single- and double-mutant infections.
|
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We also measured parasite infectivity to Anopheles mosquitoes of post-treatment gametocytes with and without DHFR and DHPS mutations. Results of MFAs of An. albimanus mosquitoes with blood obtained after SP treatment demonstrated that gametocytes carrying DHFR and DHPS mutations are infective to mosquitoes. In addition, our data show that double-mutant infections are associated with a higher probability of finding infected mosquitoes. We demonstrated oocysts are more frequent in mutant infections, but certainly it would be of interest in another next study to look at transmission of sporozoites.
Consistent with our findings, a previous study conducted in The Gambia demonstrated that the presence of a multidrug-resistant haplotype was associated with significantly higher oocysts burdens after treatment with the combination of chloroquine and SP.13 Our study confirms that low-level drug resistance mutations have the potential to enhance transmission of P. falciparum and the spread of resistant parasites. Our findings highlight the need to anticipate drug resistance before it manifests itself as clinical treatment failure. Neither the DHFR triple-mutant nor the DHFP/DHPS quintuple-mutant parasites most strongly associated with clinical SP treatment failure14 have been detected on the Pacific coast of Colombia. Based on the rapid appearance of resistance after introduction of antifolates15,16 and the ability of a single point mutation to cause some degree of resistance to DHFR inhibitors,17 it was thought that antifolate resistance arises frequently through independent mutational events. However, the most highly antifolate-resistant forms of P. falciparum appear to have spread through other regions of South America in genetic sweeps rather than arising de novo on the background of single- and double-mutant DHFR and single-mutant DHPS that we see in our setting.18 Microsatellite analyses have shown that DHFR triple-mutant parasites from diverse geographic regions share common ancestry,19 which further supports the idea that the threat to SP efficacy in our region may be from the importation of highly resistant parasites from outside the area. To reconcile theory and data, it has been hypothesized that new, independently arising DHFR triple-resistance parasites are unlikely to survive or are actually killed by drug therapy.20
Whether SP resistance arises locally or is imported, anti-malarial drugs and drug combinations that eliminate both asexual and sexual parasites deserve priority because they will reduce transmission of drug-resistant parasites that are highly infectious to mosquitoes. High rates of SP resistance and highly mutated DHFR and DHPS have been found in the Amazonian eastern region of Colombia (F. Mendez, unpublished data), and the use of gametocytocidal treatments should be considered now as part of a strategy to preserve SP as an effective component of combination therapies. Combination therapies including drugs with both short- and long-acting parasite killing action would decrease PCT, reduce gametocytemia, and decrease the probability that mutant parasites will be selected and transmitted to mosquitoes. The evaluation of combination therapies would benefit from determining whether, after treatment, mutant parasites are present in sexual forms and if they are infective to mosquitoes.
Received April 11, 2007. Accepted for publication June 13, 2007.
Financial support: This study was supported in part by a grant AI055687-02 from the National Institutes of Health and the Universidad del Valle, Cali, Colombia.
* Address correspondence to Fabián Méndez, Escuela de Salud Pública, Universidad del Valle, San Fernando, Calle 4B No. 36-140, Edificio 118, Cali, Colombia. E-mail: famendez{at}univalle.edu.co 
Authors’ addresses: Fabián Méndez, Escuela de Salud Pública, Universidad del Valle, San Fernando, Calle 4B No. 36-140, Edificio 118, Cali, Colombia, Fax: 57-2-557 0425, E-mail: famendez{at}univalle.edu.co. Sócrates Herrera, Bermans Murrain, Andrés Gutiérrez, and Luz A. Moreno, Malaria Vaccine and Drug Testing Center, Carrera 35 No. 4A-53, A.A. 25574, Cali, Colombia, Telephone: 57-2-5583937, Fax: 57-2-5560141, E-mail: sherrera{at}inmuno.org. María Manzano, Universidad Nacional de Colombia, Sede Palmira, Colombia. Álvaro Muñoz, Department of Epidemiology. The Johns Hopkins Bloomberg School of Public Health, 615 North Wolfe Street, E-7648, Baltimore, MD 21205, E-mail: jvaldez{at}jhsph.edu. Christopher V. Plowe, Center for Vaccine Development, University of Maryland School of Medicine, 685 West Baltimore Street, HSF 480, Baltimore, MD 21201, E-mail: cplowe{at}medicine.umaryland.edu.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
the 75th percentile + 1.5 IQR.