Malaria continues to be a major global health problem exposing over 2000 million of the world’s population to varying degrees of malaria risk. Of the four parasite species,
Worldwide, the focus of research in malaria has centered around
Thus, the primary objective of the present clinical study was to assess the efficacy of various PQ regimens as antirelapse treatment along with no PQ regimen for
The Institutional Ethics Committee approval and patient’s informed consent were obtained prior to enrollment in the study.
It was an open label, randomized, controlled, parallel group, assessor blind study comparing antirelapse efficacy of 3 regimens of PQ, 15 mg/day × 14 days (group B), 30 mg/day × 7 days (group C), and 30 mg/day × 14 days (group D) along with a group of
Patients suspected to be suffering from malaria were referred to Malaria Out-Patient Department (OPD), for microscopic diagnosis of malaria using peripheral blood smear stained using Giemsa stain. Patients diagnosed as positive for
Inclusion criteria were as follows: adult patients, male and female (equal or over the age of 18 years); peripheral blood smear diagnosis of willing to undergo hospitalization for the entire duration of primaquine treatment; willing to provide informed consent; willing to undergo investigations and come for regular followup; normal G6PD; hemoglobin greater or equal to 10 gm/dL.
Exclusion criteria were as follows: mixed infection with pregnancy and lactation; evidence of significant hepatic, renal, or cardiac disease as diagnosed by history, clinical examination, and laboratory tests whenever necessary; any other condition which would interfere with patient’s participation in the study or compliance with the treatment.
All the patients enrolled in the study were given the standard WHO recommended treatment regimen of Chloroquine (CQ) [
The primary efficacy outcome of the study was the number of patients showing recurrence, relapse, and reinfection of
The recurrence was monitored by evaluating the efficacy of the CQ and PQ. CQ response was monitored as per the criteria given by Wernsdorfer et al. [
The PQ effect was assessed by regular followup once in a month, after the first 28 days for CQ sensitivity monitoring, by examination of peripheral blood smear until next 6 months. The peripheral smear examination was also done whenever patients developed fever within the 6-month follow-up period. Patients showing reappearance of
The recurrence was classified as relapse if it occurred in the period between January and June (low transmission season) and reinfection if it occurred in the period between July and December (active transmission season) [
Our previous study has shown that patients not treated with PQ show the recurrence of
A simple, computer generated randomization scheme was used for the randomization of patients into the three PQ regimen groups.
This was an open label study and no concealment of treatment allocation was followed.
The patients fulfilling the inclusion exclusion criteria and consenting to participate in the study were enrolled in the study. The blood spots were blotted on the filter paper (pretreatment sample) for genotyping analysis. All the patients enrolled in the study were hospitalized either for 10 or 17 days depending on the PQ group they were randomized to and then treated with standard WHO recommended treatment regimen of CQ for 3 days, followed by PQ as per randomization. All the patients were given follow-up card at the time of discharge and were asked to follow up once a month for subsequent six months and also as when they get fever. In order to get maximum follow-up data and reduce drop outs due to lost to followup, post cards were sent to remind them about their follow-up visits. Active followup was also done when the laboratory assistant visited the homes of the patients and made blood smears, thus avoiding the need for the patient to visit the hospital. Any of these patients diagnosed to be positive for
Flow chart of the study.
Although the study was not blinded in terms of treatment administration, the person seeing the slides and carrying out other outcome assessments were blinded to the treatment group by coding the samples.
Protocol method of analysis was used to analyze the data. Chi-square test was applied to find the differences between the 4 groups for baseline characteristics.
The total number and percentages of patients showing recurrence of
The concordance between three methods was checked by comparing the results of the methods. The data of PCR sequencing were used to depict the diversity of different genotypes prevalent in the samples studied.
The number of patients in each group who had adverse events or discontinued medication due to adverse events was expressed as a percent of the total. The percentages were compared using the Chi-square or Fisher’s test as appropriate. For the purpose of the study, any adverse events (irrespective of cause-effect relationship) which were not present at baseline or worsened after baseline were taken for statistical analysis.
Overall 4215 (8.15%) cases of
Baseline demographic characteristics of the patients enrolled in the study.
PQ regimens | Group A |
Group B |
Group C |
Group D |
---|---|---|---|---|
Number of patients enrolled | 397 | 398 | 381 | 380 |
Total drop outs | 92 (23.1%) | 76 (19%) | 83 (21.7%) | 63 (16.5%) |
Number of patients who completed 6 months of followup | 305 | 322 | 298 | 317 |
Age in years |
18–76 (30 ± 14) | 18–76 (31 ± 12) | 18–74 (32 ± 13) | 18–70 (32 ± 11) |
Gender | ||||
Male | 374 | 378 | 364 | 365 |
Female | 23 | 20 | 17 | 15 |
Parasitemia | 120–16780/ |
160–17600/ |
175–15620/ |
105–19239/ |
All the 1556 patients (except 1 who was discontinued CQ due to AE) in 4 groups (A, B, C, and D) showed clearance of parasites by day 6 from the initiation of treatment, without subsequent reappearance until D28, showing that all
Differentiation of recurrences as relapse or reinfection.
PQ regimens | Group A |
Group B |
Group C |
Group D |
---|---|---|---|---|
Number of patients who completed 6 months of followup | 305 | 322 | 298 | 317 |
Number of patients showing recurrence (%)* | 50 (16.39) | 26 (8.07) | 30 (10.07) | 21 (6.62) |
|
||||
As per the month of recurrence | ||||
Number of relapses | 21 | 5 | 12 | 9 |
Number of reinfections | 29 | 21 | 18 | 12 |
Relapse rate** |
|
|
|
|
Reinfection rate |
|
|
|
|
|
||||
Genotyping (PCR-RFLP) | ||||
Paired primary and recurrence samples available for analysis | 24 | 15 | 18 | 16 |
Number of PCR-RFLP genotyping results available for the paired samples | 16 | 11 | 11 | 9 |
Number of samples showing same genotypes | 8 | 3 | 5 | 5 |
Number of samples showing different genotypes | 8 | 8 | 6 | 4 |
Relapse rate*** |
|
|
|
|
Reinfection rate |
|
|
|
|
|
||||
Genotyping (PCR sequencing) | ||||
Paired primary and recurrence samples available and analyzed | 24 | 15 | 18 | 16 |
Number of PCR sequencing results available for the paired samples | 12 | 11 | 13 | 13 |
Number of samples showing same genotypes | 2 | 2 | 2 | 3 |
Number of samples showing different genotypes | 10 | 9 | 11 | 10 |
Relapse rate |
|
|
|
|
Reinfection rate |
|
|
|
|
Out of 1556 patients, 9 (0.57%) patients (A = 1, B = 0, C = 4, and D = 4) reported adverse events during the CQ treatment. The number of AEs reported was significantly different in the 4 groups (
AEs of CQ and PQ in the four groups.
PQ regimens | Group A |
Group B |
Group C |
Group D |
---|---|---|---|---|
Number of patients enrolled | 397 | 398 | 381 | 380 |
AEs of CQ* | 1 | 0 | 4 | 4 |
Drop outs (discontinuation of CQ) due to CQ AEs | 0 | 0 | 0 | 1 |
AEs of PQ** | 3 | 5 | 10 | 13 |
Drop outs (discontinuation of PQ) due to PQ AEs*** | 0 | 2 | 2 | 7 |
Total drop outs due to AEs (CQ and PQ)**** | None | 2 | 2 | 8 |
Out of 1556 patients, 31 (1.99%) patients (A = 3, B = 5, C = 10, and D = 13) reported adverse events during PQ treatment (after completion of CQ treatment). The number of AEs was significantly different in the 4 groups (
Of these 31 patients, AEs in 11 patients (A = NA, B = 2, C = 2, and D = 7) lead to discontinuation of PQ treatment. The number of AEs leading to discontinuation of PQ treatment differed significantly in the 4 groups studied (
The number of drop outs due to AEs in 4 groups was significantly different (
There were 33 paired samples for which the results from all the three methods were available to assess the concordance. The concordance between all the three methods was 45% whereby all the three methods classified the case of recurrence as relapse or recurrence in the same way. While relatively high concordance was noted between PCR-RFLP and PCR sequencing (57.58%), good concordance was also noted between the month of recurrence and PCR-RFLP (48.48%) but little concordance between month of recurrence and PCR sequencing method (15.15%) (Table
Comparison of the methods for their concordance to classify the cases of recurrence as relapse of reinfection.
All 3 methods | RFLP and PCR sequencing | Month of recurrence and PCR-RFLP | Month of recurrence and PCR sequencing |
---|---|---|---|
15/33 (45%) | 19/33 (57.58%) | 16/33 (48.48%) | 5/33 (15.15%) |
Comparison of three methods of classifying cases of recurrence: all groups.
PCR-RFLP restriction enzyme digestion banding pattern.
The rates of relapse and reinfection identified by the method of month of recurrence and PCR-RFLP were found to be in agreement. While PCR sequencing results were bit different than the results of both these methods of classification used (Table
Fifty-two genetic subtypes were identified in a total of 143 isolates (paired primary and recurrence samples) analyzed.
The present comparative study carried out to assess safety and efficacy of different PQ regimens as antirelapse treatment identified a higher recurrence rate in the patients who did not receive PQ treatment as compared to the patients who received different PQ regimens. When the cases of recurrence were subjected to the methods differentiating relapse and reinfection, majority of them were found to be reinfection cases.
The study did not find any major or unexpected adverse events. However, significantly higher adverse events in higher dose of PQ, group D (30 mg × 14 days), leading to significantly higher drop outs were noted.
Discordance was noted in the results of the methods used to differentiate cases of recurrence as relapse or reinfection. The study also documented a lot of genetic diversity in the types of genotype identified using the PCR sequencing method of genotyping indicating the diversity of type of
The low recurrence rate found in the patients receiving different PQ regimens as compared to patients not receiving PQ (16.3%) is indeed a good indicator of PQ effect and thus it is used as antirelapse treatment. The earlier study in the same population has documented recurrence rate of 9.2% in patients not treated with PQ [
The present study reported recurrence even in patients given PQ treatment. This also has been documented in earlier studies. For example, the study in the same population indicated recurrence rate of 4.6% for the 15 mg × 14 days PQ regimen [
In order to increase the effectiveness of PQ, use of higher dosages of PQ has been tried including a latest case report being successfully treated with high dose PQ as high as 45 mg once weekly for 8 weeks at 3rd instance; after the failure of 15 mg/d for 2 weeks at 1st instance and 7.5 mg 4 times daily for 2 weeks at 2nd instance for radical cure of vivax malaria [
The recurrence of
Extreme variation is noted in the recurrence rates reported in these studies. Several factors including the type of prevalent genetic strains of vivax as well as the varied length of follow-up period used in the studies might explain the variations observed in these studies. The period of followup is an important consideration for observing the radical cure of
The present study used 3 methods to differentiate the cases of recurrence as relapse or reinfection in order to better assess the antirelapse effect of PQ. It was noted that, although recurrence rates were high, when these cases of recurrence were classified as relapse or reinfection using different methods, relapse rates observed were much lower.
There is not much research involving genotyping of paired samples to compare with the present study. However, the PCR corrected rates of relapse observed in the current study were comparable to our earlier study [
The relapse and reinfection rates varied between the three methods used in the present study leading to discordance in the results of the three methods used to differentiate the cases of recurrence. There could be several reasons for this discordance ranging from the varying sensitivity and specificity of the method, the polymorphic markers used and certain unresolved issues like infection with multiple clones. For example, PCR sequencing technique could detect higher reinfection rate because it can differentiate the two isolates more precisely as compared to PCR-RFLP technique. The recent genotyping study done to find the genetic diversity of
As far as selection of marker for genotyping is concerned, it is the highly polymorphic gene that is preferred. However, if the region to be studied is highly polymorphic then it leads to amplification problem due to the polymorphism existing in the primer binding region. We used MSP-3 gene family for RFLP genotyping on account of the polymorphism documented with this gene. However, recent research [
Further studies in various other geographic areas are needed for taking decision on policy change, even if that has to be made.
In the present study large number of samples showed nonamplification for genotyping methods used. This was also observed in the study by Maestre et al. [
The study also determined the extent of polymorphism in
From this study it is evident that, although the incidence of probable resistance to various PQ regimens, used in the present study, is fairly low at this point (relapse rate ranging from 1.47% to 4.58% using different methods of classification of recurrence), it is anticipated that resistance is likely to increase and therefore, continuous monitoring of the therapy with PQ is warranted. One should keep in mind the propagation of the resistant strain which may lead to increased relapse rate in near future. Also, this is an indication of declining efficacy of PQ which could be due to increasing resistance to the drug.
There are several methodological limitations of the study. One of the major limitation of the study is the nonrandomization of the No PQ group due to ethical reasons which forced us to select the patients for this group based on those who did not receive PQ because of the clinical reasons such as Hg < 10 gm/dL; G6PD deficiency, pregnancy, lactation, and so forth. However, we do not believe these variables might have had any effect on the recurrence rate. Additionally, even though we used this group to compare with rest of the PQ regimens tested, the comparative number of recurrence within 3 PQ groups presented in the table are self-explanatory. The patients in this study were followed up only for 6 months. The number thus presented in the study may be an underestimation as the late relapses may not have been adequately captured. The longer follow-up period pose challenge of more drop outs on account of data lost to followup. Additionally, because this was a comparative study it can be assumed that the longer followup may not influence the comparison even with increase in the number of patients showing recurrence. The study used month of recurrence as one of the method of classifying recurrence as relapse or reinfection. We certainly understand that this method of classifications of recurrences is deeply problematic. While a case may be made for relapses during low-transmission season, no case at all may be made for classifying recurrences as reinfection during the transmission season. Even low probability transmission may suffice to seriously confound this classification system. However, typically a trend in increase in cases of malaria is noted during the monsoon season in Mumbai, the geographic area where the study was carried out. In view of this scenario this method was considered. Our objective of including this method in the current study was to basically compile the possible methods of differentiation and present an analysis of the data even though it may not be an accurate method.
We did not use body weight of the patients enrolled in the study. This variable might have helped in calculating per kg dose of the PQ that patients received. The low total PQ dose per kilogram of body weight [
The study documented statistically significant higher recurrence rate in patients not given PQ as compared to patients receiving different PQ regimens. There was discordance in the methods (namely, month of recurrences, RFLP, or sequencing) used to differentiate relapse from reinfection. There could be several reasons for discordance. None of the methods are suitable given their limitation. The lack of concordance in the methods used has implications on the interpretation of the results and therefore warrants caution.
Using the available methods, the relapse rate ranging from 1.47% to 4.58% was noted in the patients receiving different PQ regimens. Considering that the paired samples classified as “reinfection” by RFLP or sequencing could be the cases of “relapse,” the probable resistance to PQ could be even higher indicating the decreased effect of PQ to eradicate liver forms of
The study found that PQ regimens used to be safe without any major or unexpected adverse events, although statistically significantly higher adverse events were noted in the higher dose PQ (30 mg × 14 days) leading to significantly higher drop outs due to discontinuation of the treatment.
The study also documented lots of genetic diversity in the types of genotype identified using the PCR sequencing method of genotyping, indicating diversity of type of
The authors declare that there is no conflict of interests regarding the publication of this paper.
This project was supported by the funding from Indian Council of Medical Research, New Delhi, India. The authors would also like to thank Mihir Gandhi for his help with statistical analysis.