INTRODUCTION
Malaria is caused by infection with one or more of five parasite species: Plasmodium falciparum (Pf), Plasmodium malariae (Pm), Plasmodium ovale (Po), Plasmodium vivax (Pv), and Plasmodium knowlesi. Of these, the greatest burden of severe disease and death is caused by P. falciparum. 1 According to the 2020 World Malaria Report, it is estimated that more than half of the world’s population is at risk for malaria every year. In 2019, malaria deaths were slightly more than 409,000. 2 The burden of morbidity and lasting damage from clinical, nonfatal malaria, especially in children is high, with 67% of malaria deaths (274,000) occurring in children under age 5. 3 Sub-Saharan Africa suffers disproportionately from this burden and remains the region at greatest risk. 2 Individuals who frequently suffer from malaria eventually develop partial immunity, and, in areas where malaria transmission remains high, older children and adults often remain asymptomatic during parasite infections. 4 Residents of endemic areas occasionally have clinical episodes but likely carry parasites much of the time, with variable parasitemias that are too low to be detected by microscopy, also known as submicroscopic parasitemias. 5 Infection with a novel parasite strain or alteration in general immune status can increase the risk of these semi-immune individuals for more frequent episodes of clinical malaria 6–8 with significant clinical interactions between HIV-1 and malaria (reviewed in Hewitt et al. 9 ).
In previous studies in sub-Saharan Africa, elevated parasitemia as well as more frequent clinical episodes of malarial fever have been associated with decreased CD4+ T-cell counts in HIV-1 symptomatic individuals, trends not observed in individuals earlier in the course of HIV-1 infection. 7,8,10 In addition, HIV-1-infected adults from a study in Uganda presented with more than twice as many episodes of clinical malarial relative to HIV-1-uninfected adults, with the risk of symptomatic disease rising with falling CD4+ T-cell counts. 7 Although numerous studies have examined one or more of the aspects of HIV-1 coinfection and symptomatic malaria, it is also important to determine the epidemiological impact of HIV-1 coinfection on asymptomatic malaria in regions characterized by high transmission and antimalarial drug resistance.
In an attempt to describe the burden of asymptomatic malarial parasitemia and to better define the interactions between HIV-1 coinfection and asymptomatic parasitemia, we focused on the Lake Victoria region of Western Kenya, Kombewa subcounty. This region has the highest prevalence of malaria and HIV-1 in Kenya. Malaria transmission in this region occurs throughout the year. The most recent survey in Kombewa in Western Kenya, conducted in 2003 and 2004, estimated the entomological inoculation rate for P. falciparum to be 31.1 infectious bites per person per year. 11 However, apart from one study of the distribution of malaria within the Lake Victoria Islands in 2016, 12 no study has evaluated malaria parasite species diversity in Kenyan mainland since 2009, with the period since defined by significant changes in population demographics and HIV-1 epidemiology. In the 2016 study, Idris et al. 12 reported high levels of asymptomatic and submicroscopic infections. Among the 3,867 malaria infections described in the Idris study, 92% of cases were asymptomatic, 30% were mixed infections and 50% were submicroscopic, of which 94% were also asymptomatic. 12 These numbers highlight concerns about the magnitude of asymptomatic and submicroscopic reservoirs of malaria in Kenya. However, these patterns may be island-specific and could be very different from the infection dynamics in the mainland of Lake Victoria basin of Western Kenya. Further, the need remains to quantify the impacts of HIV-1 coinfection on these important reservoirs of parasite carriage and on patterns of single and mixed parasite infections in Kenya. Given that prolonged, low-level parasitemia has also been associated with intermittent and unpredictable gametocytemia, 13–15 detailed coinfection data for asymptomatic parasitemia could be critical to designing synergistic, rather than inadvertently antagonistic, control programs for both diseases. In this context, the goal of our study was to determine the prevalence of asymptomatic infections with single and multiple species of malaria parasites in HIV-1 coinfected study volunteers and study volunteers with malaria only in the Lake Victoria mainland region using sensitive and specific molecular techniques.
METHODS
Sample collection.
This study was a noninterventional cross-sectional study with a single sampling event from volunteers enrolled between January 2015 and July 2018. Average enrollment rate was 49 samples per month during the study period. A total of 1,762 samples were collected from a population of apparently healthy adult Kenyans (> 18 years old) who self-presented for voluntary HIV-1 testing and counseling at the HIV-1 Testing and Counseling (HTC) Center, Kisumu West Hospital, Kombewa subcounty, Nyanza Province, Kenya, or at an HTC Center in the Kisumu West District associated with the Kenya Medical Research Institute (KEMRI)/Walter Reed Project (WRP) PEPFAR program. Individuals entering the HTC for HIV-1 testing were given an opportunity to provide informed consent based on study details and blood sampling required for malaria testing in addition to their HIV-1 testing. HIV testing in this study was performed according to the approved national testing algorithm and in accordance with Kenyan government HIV testing standard operating procedures. Briefly, the Determine HIV-1/2 (Abbott, Chicago, IL) rapid diagnostic test (RDT) was performed for initial screening. For positive reactions, the First Response HIV-1-2 RDT test was performed for confirmation. Inconclusive tests were referred to the national laboratories for molecular testing. After consent was obtained, the Parascreen Pan/Pf RDT (Zephyr Biomedicals, Verna, Goa, India) was administered to detect the presence or absence of infection in accord with Kenya Ministry of Health (MoH) guidelines. No specific malaria screening was performed, but if the malaria RDT was positive, antimalarial treatment was made available at the time of enrollment. Two to five 50-μL blood samples were collected on Whatman 903 Protein Saver filter paper cards (GE Healthcare Life Sciences, Chicago, IL), immediately air dried, and stored at –80°C for subsequent molecular analysis. The observational trial design and protocol details are published elsewhere (J. Oyieko, manuscript in preparation).
After sample collection, a standard protocol was used to extract both DNA and RNA from a single dried blood spots (DBS), followed by specific primer-driven reverse transcription of RNA for an 18S quantitative polymerase chain reaction (qPCR) assay (described subsequently) to detect any human Plasmodium spp. Briefly, DBS lysate was prepared and split in half for RNA and DNA extraction using the Qiagen RNeasy Mini kit and QIAmp DNA Mini Kit (Qiagen, Hilden, Germany), with elution in 30 µL and 50 µL of water, respectively. This was followed by cDNA preparation for 18S using QuantiTect Reverse Transcription Kit (Qiagen) and the malaria-conserved, genus-specific 18S Outer Reverse Primer: 5′-GTT CCT CTA AGA AGC TTT-3′ in place of random hexamers to enhance amplification of the target gene. This improvement to the published protocol of Kamau et al. 16 enabled amplification from both cDNA and genomic DNA, reducing background and increasing assay sensitivity and specificity (unpublished data).
qPCR assays.
In addition to amplification from both cDNA and genomic DNA, the 18S qPCR assay was improved to include a probe targeting the genus-conserved sequences of the parasite 18S ribosomal gene from Kamau et al. 16 (Table 1). Each reaction contained 1X Qiagen Taqman universal mix, 0.25-µM probe, 0.25 µM of each genus-specific forward and reverse primers, and 1 µL of template cDNA and genomic DNA in a reaction volume of 20 µL. Human DNA from a parasite-free donor (hDNA) and no-template controls (DNase/RNase free water) were included in each assay as negative controls. The positive controls were six serial 10-fold dilutions of a stock plasmid. The positive control plasmid was prepared by cloning the 100 bp 18S PCR product into pCR 2.1 TOPO TA vector (Life Technologies, Carlsbad, CA) following manufacturer’s guidelines. 18S amplicon copy number per microliter was estimated from quantification by NanoDrop 2000C (ThermoFisher Scientific, Waltham, MA). The 18S plasmid stock was diluted in water to generate a 10-fold dilution series from 100,000 18S copies per microliter to 0.1 18S copies per microliter. Assay validation was completed in a 7500 Fast real-time PCR machine (Applied Biosystems, Life Technologies) to determine the linear dynamic range, specificity, repeatability, reproducibility, and limit of detection. Amplification conditions were initial denaturation at 95°C for 20 seconds followed by 45 cycles of denaturation at 95°C for 3 seconds and annealing at 60°C for 30 seconds.
Target genes and primer and probe sequences for the 18S genus-specific and species-specific qualitative polymerase chain reaction assay
| Target gene | Primer/probe name | Primer/probe sequence | Reference |
|---|---|---|---|
| Small subunit rRNA (ssrRNA) | Genus18S F | 5′ GCTCTTTCTTGATTTCTTGGATG 3′ | Kamau et al., 2011 |
| Genus18S R | 5′ AGCAGGTTAAGATCTCGTTCG 3′ | ||
| Genus18S Probe | 6-FAM-ATGGCCGTTTTTAGTTCGTG-TAMRA | ||
| Pf var acidic terminal sequence (varATS) | var_ ATS F | 5′ CCCATACACAACCAAYTGGA 3′ | Hofmann et al., 2015 |
| var_ATS R | 5′ TTCGCACATATCTCTATGTCTATCT 3′ | ||
| var_ATS Probe | 6-FAM-TRTTCCATAAATGGT-MGB | ||
| Pm circumsporozoite (csp) | PmVo F | 5′ CTCAAATTCCACCAAGTCAAGAAA 3′ | Modified from Vo et al., 2006 |
| PmVo R | 5′ GATTCGTGCTATATCTGACTTCTAACTCA 3′ | ||
| PmVo Probe | 6-FAM-AGTGAGTTGTGTTACAATAA-MGB | ||
| Po reticulocyte binding protein 2 (rbp2) | Porbp2 F | 5′ CCACAGATAAGAAGTCTCAAGTACGA ATT 3′ | Miller et al., 2015 |
| Porbp2 R | 5′ TTG GAG CAC TTT TGT TTG CAA 3′ | ||
| Porbp2 Probe | 6-FAM-TGAATTGCTAAGCGATATC-MGB | ||
| Pv enoyl-acyl carrier protein reductase (ecpr) | PvVo F | 5′ CAAGCGGAAGGGATAAATGG 3′ | Modified from Vo et al., 2006 |
| PvVo R | 5′ CCGCGATGAAGCAGATGTCT 3′ | ||
| PvVo Probe | 6-FAM-AAGGGAGAACCCC-MGB |
If a sample contained detectable Plasmodium-specific 18S copies, a panel of highly species-specific DNA-based qPCR assays were used to determine the presence and quantity of single and mixed species infections of P. falciparum, P. ovale, P. malariae, and P. vivax. Species-specific primers for P. malariae and P. vivax were redesigned 17 to amplify targets that lacked homologous sequences in the other three species, whereas P. ovale primers were previously described 18 (Table 1). The P. falciparum assay was based on var gene acidic terminal sequence (varATS), a target with 59 copies per haploid genome. 19 All assays were optimized for fluorescent probe-based detection. The optimization and assay performance parameters for P. falciparum, P. malariae, and P. vivax assays are reported elsewhere (D. Rockabrand et al., in preparation). Assay performance for P. ovale was described by Miller et al. 18 The assay standards for P. malariae, P. ovale, and P. vivax were prepared from 10-fold dilution series of species-specific plasmids (100,000 to 0.01 copies/μL), and the P. falciparum standards were prepared from a 10-fold dilution of 3D7 parasites (20,000 parasites to 0.02 parasites per μL). Microscopy analysis on a subset of samples was done by expert microscopists at the Malaria diagnostic center in Kisumu, Kenya.
Data analysis.
Data analysis was performed using GraphPad Prism version 8.4.1 (GraphPad, San Diego, CA) and Stata version 16 (StataCorp, College Station, TX). Variables were summarized as frequencies and percentages, ranges, and mean ranges as appropriate. Parasitemia, age, and gender were compared by Fischer’s exact test or χ2 test. Mann-Whitney test was used to compare continuous variables between groups. Kruskal-Wallis test of trend was used to analyze parasitemias over time. The level of significance was set at α = 0.05.
Ethical considerations.
Ethical approval for human use was granted by the Ethical Review Committee of KEMRI, Nairobi, Kenya (SSC # 2600), the Walter Reed Army Institute of Research Institutional Review Board, Silver Spring, MD (WRAIR # 2033), and the Uniformed Services University of the Health Sciences Institutional Review Board (USUHS # G18753).
RESULTS
Patient demographics.
Study volunteers were from the Seme subcounty and a portion of Kisumu West subcounty of Kisumu County, Kenya (formerly the Kombewa and Maseno Divisions of Nyanza Province). This region is within the Kombewa Health and Demographic Surveillance System (Kombewa HDSS) run by KEMRI/WRP. Of the 1,762 adult subjects enrolled in the study, 57.3% (1,010/1,762) were female and 42.7% (752/1,762) were male, with an average age of 27 (range 18–56) for all participants. The prevalence of malaria by RDT was 19.9% (352/1,762; Table 2), whereas that of HIV-1 was 10.6% (186/1,755) (Table 3). Seven subjects (0.4%) had inconclusive HIV-1 RDT results that required further molecular characterization as per the Kenyan MoH procedures; these seven samples were omitted from HIV-1 status comparisons (Table 3).
Frequency of detection by malaria rapid diagnostic test (RDT) and 18S qualitative polymerase chain reaction assay (qPCR)
| 18S qPCR negative | 18S qPCR positive | Total | |
|---|---|---|---|
| RDT negative | 586 (33.3%) | 824 (46.8%) | 1410 |
| RDT positive | 42 (2.4%) | 310 (17.6%) | 352 |
| Total | 628 | 1134 | 1762 |
Chi-square test showed an association between the two detection methods (P = 0.0001) and tests of proportions between and within the groups showed significant differences between the two detection methods. Each cell contains the sample size and the proportion (%) of the total sample size.
Malaria positivity (18S qPCR) by HIV-1 status
| 18S qPCR negative | 18S qPCR positive | Total | |
|---|---|---|---|
| HIV-1 uninfected | 571 (32.5%) | 998 (56.9%) | 1569 |
| HIV-1 infected | 54 (3.1%) | 132 (7.5%) | 186 |
| Total | 625 | 1130 | 1755 |
qPCR = qualitative polymerase chain reaction assay. Seven subjects (0.4%) had inconclusive HIV-1 rapid diagnostic test results that required further molecular characterization as per the Kenyan Ministry of Health procedures. These seven samples were omitted for these analyses, reducing total volunteers from 1,762 (Table 2) to 1,755. Chi-square test showed an association between HIV-1 and asymptomatic parasitemia (P = 0.047). Each cell contains the sample size for the specific combination and the proportion (%) of the total sample size.
Parasitemia prevalence by 18S qPCR assay.
The overall prevalence of asymptomatic parasitemia as detected by 18S qPCR was 64.3% (1,134/1,762; Table 2). There were no significant differences in prevalence of parasitemia over time during the 2015–2018 study period (Kruskal Wallis, P = 0.1123) (Supplemental Figure 1). Our 18S qPCR assay detected infections with ≥ 1 parasite per 50 μL sample. As expected, parasitemia prevalence as determined by this molecular assay was much higher than that determined by RDT (19.9%, 352/1,762). Contingency analysis (Table 2) comparing the proportions of positive samples by 18S qPCR and RDT followed by a test of proportions showed significant differences between and within groups (P < 0.0001). Concordance rate by the two methods was 17.6% (310/1,762), whereas discordance was 46.8% with 824 samples being positive by qPCR and negative by RDT (Table 2). An analysis of copy numbers between these two categories showed that the concordant samples had significantly higher (P < 0.0001) mean 18S copy numbers (geometric mean 4,649 copies/μL; 95% confidence interval [CI] 3,185–6,787) than the discordant samples (geometric mean 178 copies/μL; 95% CI 143–222), suggesting that concordance and discordance between the two methods were largely driven by parasite density.
To further understand this discrepancy, a subset of 50 samples that had been tested by 18S qPCR assay were selected for quantification by microscopy. Of these, 20% (10 of 50) had detectable parasitemia by expert microscopy compared with 92% (46 of 50) that were positive by 18S qPCR (Supplemental Figure 2). All 10 samples detected by microscopy had high 18S copy numbers, suggesting that the majority of asymptomatic infections were below the detection limit of microscopy. Accordingly, we believe that our 18S qPCR assay was suitable for estimating the prevalence of asymptomatic parasitemia in our volunteers.
Interestingly, 42 or 1,762 (2.4%) volunteers were RDT-positive but failed detection by 18S qPCR (Table 2). A review of blood film slides for these 42 samples by expert microscopist showed that 95% of these samples (40 of 42) indeed appeared malaria negative. The positive RDT test could, therefore, be due to persistent PfHRP2 soluble proteins detectable by RDT in the absence of active parasite infection. 20 Of the remaining two samples, one had low parasitemia (63 parasites/μL), and the other had relatively high parasitemia (438 parasites/μL). Additional testing (sequencing) could help to resolve the reason that two samples out of 42 failed detection by 18S qPCR.
Malaria parasite speciation.
All 1,134 18S qPCR assay-positive samples were tested using a panel of four species-specific DNA-based assays. Of these, 671 were positive for malaria species either as a single infection (628/1,134), double infection (37/1,134), or triple infection (3/1,134; Table 4). The proportions of P. falciparum, P. malariae, and P. ovale single infections were 53.4% (N = 606), 0.8% (N = 9), and 1.2% (N = 13), respectively. Mixed infections with P. falciparum and P. malariae were 2.5% (N = 28), whereas those with P. falciparum and P. ovale were 0.8% (N = 9) of 18S-positive samples. Three volunteers had mixed infections with P. falciparum, P. malariae, and P. ovale (0.3%). As expected, P. vivax was not detected in this population. A total of 463 (40.8%) of 18S-positive samples did not yield data in any of the species assays. A comparison of 18S copy numbers between speciated and undetermined samples (Supplemental Figure 3) showed that undetermined samples had significantly fewer 18S copy numbers (mean of 4,792 copies/μL; 95% CI 249–9,334 copies/μL) than speciated samples (mean 160,803 copies/μL; 95% CI 93,127–228,479 copies/μL). Lack of species determination in 40.8% (463/1,134) of previously 18S positive samples therefore likely resulted from the fact that the 18S qPCR assay is more sensitive 16 than the species-specific assays that detect only DNA targets with fewer copies per infected cell.
Prevalence of infection by malaria parasite species
| Species identification | n (% 18S positive) |
|---|---|
| Plasmodium falciparum | 606 (53.4) |
| Plasmodium malariae | 9 (0.8) |
| Plasmodium ovale | 13 (1.2) |
| Plasmodium vivax | 0 (0.0) |
| P. falciparum and P. malariae | 28 (2.5) |
| P. falciparum and P. ovale | 9 (0.8) |
| P. falciparum, P. malariae, and P. ovale | 3 (0.3) |
| No species determination | 463 (40.8) |
| Total | 1,134 (100) |
Impacts of HIV-1 coinfection, age, and gender on asymptomatic malarial parasitemia.
We first examined the age distribution of volunteers by HIV-1 status and gender. HIV-1-infected volunteers were, on average, significantly older than HIV-1-uninfected volunteers (P < 0.0001; Figure 1A) with a mean age of 32 versus 27, respectively, and male volunteers, regardless of their HIV-1 status, were slightly older than the female volunteers (P < 0.003; Figure 1A) with a mean age of 28 versus 27, respectively. When HIV-1 status and gender were considered together, these groups showed significant differences by age (Figure 1B). Given that age is strongly associated with immunity to malaria and parasitemia, 4,21 and observations of an association between age and malaria in endemic areas, 4 we examined the association between asymptomatic parasitemia (18S copy number) and age as continuous variables. There was a weak but significant negative correlation between parasitemia and age by Spearman’s Rho (r = –0.1580; P < 0.0001; Figure 2).
Age, gender, and HIV-1 status were significantly different among study volunteers. (A) Male volunteers were slightly older than female volunteers (P = 0.003) with a mean age of 28 versus 27 for females. In addition, HIV-1-infected volunteers were significantly older than HIV-1-uninfected volunteers (P = 0.0001) with a mean age of 27 versus 32. (B) HIV-1-infected males were significantly older than HIV-1-infected female volunteers (P = 0.0001). HIV-1-infected males were significantly older than HIV-1-uninfected males (P = 0.0001) as well as HIV-1-uninfected females (P = 0.0001). However, mean ages of HIV-1-uninfected males and HIV-1-uninfected females were not significantly different (nonsignificant; P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
Age, gender, and HIV-1 status were significantly different among study volunteers. (A) Male volunteers were slightly older than female volunteers (P = 0.003) with a mean age of 28 versus 27 for females. In addition, HIV-1-infected volunteers were significantly older than HIV-1-uninfected volunteers (P = 0.0001) with a mean age of 27 versus 32. (B) HIV-1-infected males were significantly older than HIV-1-infected female volunteers (P = 0.0001). HIV-1-infected males were significantly older than HIV-1-uninfected males (P = 0.0001) as well as HIV-1-uninfected females (P = 0.0001). However, mean ages of HIV-1-uninfected males and HIV-1-uninfected females were not significantly different (nonsignificant; P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
Age, gender, and HIV-1 status were significantly different among study volunteers. (A) Male volunteers were slightly older than female volunteers (P = 0.003) with a mean age of 28 versus 27 for females. In addition, HIV-1-infected volunteers were significantly older than HIV-1-uninfected volunteers (P = 0.0001) with a mean age of 27 versus 32. (B) HIV-1-infected males were significantly older than HIV-1-infected female volunteers (P = 0.0001). HIV-1-infected males were significantly older than HIV-1-uninfected males (P = 0.0001) as well as HIV-1-uninfected females (P = 0.0001). However, mean ages of HIV-1-uninfected males and HIV-1-uninfected females were not significantly different (nonsignificant; P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
18S copy numbers were significantly but weakly negatively correlated to age of study volunteers. Parasitemia (expressed as 18S copy numbers) by age was weakly negatively correlated with age. Spearman’s rho nonparametric test for correlation (r = –0.1580; P < 0.0001).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
18S copy numbers were significantly but weakly negatively correlated to age of study volunteers. Parasitemia (expressed as 18S copy numbers) by age was weakly negatively correlated with age. Spearman’s rho nonparametric test for correlation (r = –0.1580; P < 0.0001).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
18S copy numbers were significantly but weakly negatively correlated to age of study volunteers. Parasitemia (expressed as 18S copy numbers) by age was weakly negatively correlated with age. Spearman’s rho nonparametric test for correlation (r = –0.1580; P < 0.0001).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
We next asked whether age, HIV-1 status, or gender were associated with the presence or absence of asymptomatic parasitemia. On the basis of findings from a previous study in which HIV-1-infected adults had nearly twice as many episodes of asymptomatic and symptomatic parasitemia relative to HIV-1-uninfected adults, 7 we hypothesized that HIV-1-infected volunteers might not only have higher prevalence of malaria parasitemia than uninfected volunteers but also higher parasite densities in this highly endemic area. Of 1,134 samples that were 18S qPCR-positive, four had inconclusive HIV-1 RDT results and were excluded, resulting in 1,130 samples that were available for analysis of association between asymptomatic malarial parasitemia and HIV-1 coinfection (Table 3). Of 1,130 18S qPCR-positive samples, 132 (11.7%) were HIV-1-infected. Analysis of the proportions of HIV-1 and 18S qPCR positivity showed an association between HIV-1 and asymptomatic parasitemia as detected by 18S qPCR (chi-square P = 0.0475; Table 3). Notably, nearly 2.5 times as many HIV-1-infected volunteers were 18S positive (132/186; 71%) relative to HIV-1-infected volunteers who were 18S negative (54/186; 29%). In comparison, the proportion of HIV-1-uninfected volunteers who were 18S positive was 64% (998/1,569) versus 36% (571/1,569) who were 18S negative. Further, the prevalence of HIV-1 infection was higher among those volunteers who were 18S positive (132/1,130; 11.7%) compared with those who were 18S negative (54/625; 8.6%).
In contrast to overall malaria positivity, two by two contingency analysis of HIV-1 status and species positivity for at least one of the four malaria parasite species showed no association between these variables (Fisher’s exact test, P > 0.510). Further, there were no significant associations between positivity for any single infecting parasite species and HIV-1 status (Fisher’s exact test, P > 0.05) and no associations between mixed-species infections with HIV-1 status (Fisher’s exact test, P > 0.05). Among those volunteers who were 18S positive, 18S copy numbers were not different between HIV-1-infected and uninfected volunteers (P = 0.2236) (Figure 3A), and there were no differences in 18S copy numbers among individuals when gender was included as a variable (P > 0.05) (Figure 3B). Finally, there were no differences between species-specific target copy numbers for each infecting parasite species (Pmcsp, Porbp2, Pfvar) and HIV-1 status (P > 0.05; Figure 4).
18S copy numbers by HIV-1 status and by gender were not different among study volunteers. (A) 18S copy numbers of HIV-1-infected and uninfected volunteers were not significantly different by Mann-Whitney nonparametric test (P = 0.2236). (B) Further, there were no significant differences in 18S copy numbers among HIV-1-uninfected females, HIV-1-infected females, HIV-1-uninfected males, and HIV-1-infected males by Mann-Whitney nonparametric test (P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
18S copy numbers by HIV-1 status and by gender were not different among study volunteers. (A) 18S copy numbers of HIV-1-infected and uninfected volunteers were not significantly different by Mann-Whitney nonparametric test (P = 0.2236). (B) Further, there were no significant differences in 18S copy numbers among HIV-1-uninfected females, HIV-1-infected females, HIV-1-uninfected males, and HIV-1-infected males by Mann-Whitney nonparametric test (P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
18S copy numbers by HIV-1 status and by gender were not different among study volunteers. (A) 18S copy numbers of HIV-1-infected and uninfected volunteers were not significantly different by Mann-Whitney nonparametric test (P = 0.2236). (B) Further, there were no significant differences in 18S copy numbers among HIV-1-uninfected females, HIV-1-infected females, HIV-1-uninfected males, and HIV-1-infected males by Mann-Whitney nonparametric test (P > 0.05).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
By species-specific qualitative polymerase chain reaction, there were no differences in copy numbers for Plasmodium malariae or Plasmodium ovale or parasite density for Plasmodium falciparum between HIV-1-uninfected and HIV-1-infected volunteers. Species-specific targets included P. malariae circumsporozoite gene, P. ovale reticulocyte binding protein 2 gene and P. falciparum var acidic terminal sequence gene (see primer sequence details in Table 1).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
By species-specific qualitative polymerase chain reaction, there were no differences in copy numbers for Plasmodium malariae or Plasmodium ovale or parasite density for Plasmodium falciparum between HIV-1-uninfected and HIV-1-infected volunteers. Species-specific targets included P. malariae circumsporozoite gene, P. ovale reticulocyte binding protein 2 gene and P. falciparum var acidic terminal sequence gene (see primer sequence details in Table 1).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
By species-specific qualitative polymerase chain reaction, there were no differences in copy numbers for Plasmodium malariae or Plasmodium ovale or parasite density for Plasmodium falciparum between HIV-1-uninfected and HIV-1-infected volunteers. Species-specific targets included P. malariae circumsporozoite gene, P. ovale reticulocyte binding protein 2 gene and P. falciparum var acidic terminal sequence gene (see primer sequence details in Table 1).
Citation: The American Journal of Tropical Medicine and Hygiene 105, 1; 10.4269/ajtmh.21-0012
DISCUSSION
In this study, we examined associations between asymptomatic malarial parasitemia and HIV-1 coinfection in an area of high malaria transmission in Kenya. Of 1,762 adults seeking HIV-1 counseling and testing who enrolled in our study from 2015 to 2018, 10.6% were HIV-1-infected, which was slightly lower than regional estimates for Kisumu as reported by the National AIDS Control Council of Kenya in 2015 (19.9%) 22 and 2018 (16.3%). 23 Among these volunteers, 1,134 were 18S qPCR positive (64.3%), and 352 were malaria RDT positive (19.9%). Most volunteers with asymptomatic malaria were infected with P. falciparum, with few volunteers presenting with P. ovale and P. malariae as single or mixed infections. As expected, P. vivax was not detected in the study population, consistent with majority of sub-Saharan populations that lack Duffy blood group antigens used by P. vivax for invasion. 24
A previous study by Jenkins et al. in proximity to our study area reported a malaria prevalence of 28% in adults by microscopy. 25 Idris et al. reported a malaria prevalence of 75.2% on the Lake Victoria Islands by nested, conventional PCR. 12 Although the latter prevalence is more similar to ours, malaria on the Lake Victoria Islands is characterized by greater parasite homogeneity, altitude and vegetational homogeneity, and relatively stable perennial transmission compared with the mainland. 12 Despite this variation in reported prevalence, our study and previous studies affirm a high burden of asymptomatic malaria in Western Kenya.
Although significant patterns for infecting parasite species and parasite density were not detected by age, gender, or HIV-1 status among our volunteers, we did confirm a significant association between HIV-1 infection and asymptomatic parasitemia as determined by 18S qPCR (P = 0.047; Table 3). These findings highlight substantial interactions between HIV-1 and presence of asymptomatic parasitemia in this region in Kenya.
Although microscopy remains the gold standard for detection of malaria parasites in clinical samples, its performance in asymptomatic parasitemia is poor mainly due to low sensitivity. 26–28 We used the Parascreen Pan/Pf malaria RDT in accord with Kenya MoH guidelines for patient care and detection of asymptomatic malaria. As expected, qPCR was more sensitive than RDT in the detection of asymptomatic parasitemia (Table 2). A further analysis of a subset of 50 samples confirmed the limited sensitivity of microscopy (Supplemental Figure 2) and the value of molecular tools in detection of asymptomatic parasitemia. Although we acknowledge that 18S copy numbers are not directly correlated to parasitemia, higher copy numbers generally reflected higher asymptomatic parasite load. Notably, samples that could not be speciated by the less sensitive species assays had low 18S copy numbers (Supplemental Figure 3), suggesting an effect of the parasite load. Thus, our data, together with other studies, 5,29,30 support the use of molecular diagnostic tools in malaria epidemiology for the detection of asymptomatic, submicroscopic malaria infections that are missed by conventional diagnostics.
The identification of RDT-positive and 18S qPCR-negative samples (42 of 1,762 samples or 2.4%; Table 2) likely resulted from PfHRP2 soluble proteins that can persist in circulation for up to 28 days postinfection. 20,31,32 With expert microscopy, 40 of these 42 samples (95%) were confirmed to be parasite negative. One sample had low parasitemia (63 parasites/μL), and the other had relatively high parasitemia (483 parasites/μL) by expert microscopy. Sequence variation has been shown to result from recombination and mosaicism common in many Plasmodium multicopy genes, 33 suggesting that deep sequencing, for example, could be used to resolve these rare discrepancies. Notably, 18S rRNA gene variation was reported to interfere with species-specific probe hybridization-based detection of P. ovale parasites in field samples. 34
In other studies of our volunteers published elsewhere, we observed that HIV-1-infected individuals had an increased risk of carrying gametocytes relative to HIV-1-uninfected individuals. 35 Furthermore, we observed that P. falciparum mutation haplotypes for the folate pathway genes dihydrofolate reductase (dhfr) and dihydropteroate synthase (dhps) were significantly associated with volunteer HIV-1 status, 36 observations that collectively suggest that asymptomatic coinfected reservoirs in Western Kenya could be fueling transmission of drug-resistant parasites.
Overall, these results suggest that the geographic overlap between malaria and HIV-1 has important biological, epidemiological, and clinical implications. A follow-up longitudinal survey is being carried out by our group to further examine these associations, including evaluating the impact of antiretroviral therapy on asymptomatic malarial parasitemia and on patterns of antifolate resistance mutations in infecting parasites. With these overlapping endemicities in many regions of sub-Saharan Africa, we and others contend that malaria control efforts should adopt a focused approach to identify and target the HIV-1 coinfected population as an important reservoir of disease.
Acknowledgments:
We thank the Kombewa Clinical Research Center including counselors and clinical laboratory staff for their help in sample collection and storage. Special thanks to the Clinical Research Coordinator Mr. Solomon Otieno for coordinating the study activities. We are also grateful to the staff of Kombewa HIV-1 Testing and Counseling Center and the Kisumu West District WRP/KEMRI PEPFAR Program for their considerable help with this study. We also thank Mr. Michael Odero Ayaya for his expert malaria microscopy effort.
REFERENCES
- 1.↑
White NJ , Pukrittayakamee S , Hien TT , Faiz MA , Mokuolu OA , Dondorp AM , 2014. Malaria. Lancet 383: 723–735.
- 2.↑
Snow RW , Guerra CA , Noor AM , Myint HY , Hay SI , 2005. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 434: 214–217.
- 3.↑
World Health Organization Global Malaria Program , 2020. World Malaria Report2020. Available at: https://www.who.int/teams/global-malaria-programme/reports/world-malaria-report-2020.
- 4.↑
Langhorne J , Ndungu FM , Sponaas AM , Marsh K , 2008. Immunity to malaria: more questions than answers. Nat Immunol 9: 725–732.
- 5.↑
Bousema T , Okell L , Felger I , Drakeley C , 2014. Asymptomatic malaria infections: detectability, transmissibility and public health relevance. Nat Rev Microbiol 12: 833–840.
- 6.↑
Ofosu-Okyere A , Mackinnon MJ , Sowa MPK , Koram KA , Nkrumah F , Osei YD , Hill WG , Wilson MD , Arnot DE , 2001. Novel Plasmodium falciparum clones and rising clone multiplicities are associated with the increase in malaria morbidity in Ghanaian children during the transition into the high transmission season. Parasitology 123: 113–123.
- 7.↑
Whitworth J et al. 2000. Effect of HIV-1 and increasing immunosuppression on malaria parasitaemia and clinical episodes in adults in rural Uganda: a cohort study. Lancet 356: 1051–1056.
- 8.↑
French N , Nakiyingi J , Lugada E , Watera C , Whitworth JAG , Gilks CF , 2001. Increasing rates of malarial fever with deteriorating immune status in HIV-1-infected Ugandan adults. AIDS 15: 899–906.
- 9.↑
Hewitt K , Steketee R , Mwapasa V , Whitworth J , French N , 2006. Interactions between HIV and malaria in non-pregnant adults: evidence and implications. AIDS 20: 1993–2004.
- 10.↑
Patnaik P et al. 2005. Effects of HIV‐1 serostatus, HIV‐1 RNA Concentration, and CD4 cell count on the incidence of malaria infection in a cohort of adults in rural Malawi. J Infect Dis 192: 984–991.
- 11.↑
Ndenga B et al. 2006. Population dynamics of malaria vectors in Western Kenya highlands. J Med Entomol 43: 200–206.
- 12.↑
Idris ZM et al. 2016. High and heterogeneous prevalence of asymptomatic and sub-microscopic malaria infections on islands in Lake Victoria, Kenya. Sci Rep 6: 1–13.
- 13.↑
Tadesse FG et al. 2018. The relative contribution of symptomatic and asymptomatic Plasmodium vivax and Plasmodium falciparum infections to the infectious reservoir in a low-endemic setting in Ethiopia. Clin Infect Dis 66.
- 14.
Schneider P , Bousema JT , Gouagna LC , Otieno S , Van De Vegte-Bolmer M , Omar SA , Sauerwein RW , 2007. Submicroscopic Plasmodium falciparum gametocyte densities frequently result in mosquito infection. Am J Trop Med Hyg 76: 470–474.
- 15.↑
Ouédraogo AL et al. 2009. Substantial contribution of submicroscopical Plasmodium falciparum gametocyte carriage to the infectious reservoir in an area of seasonal transmission. PLoS One 4: e8410.
- 16.↑
Kamau E , Tolbert LDS , Kortepeter L , Pratt M , Nyakoe N , Muringo L , Ogutu B , Waitumbi JN , Ockenhouse CF , 2011. Development of a highly sensitive genus-specific quantitative reverse transcriptase real-time PCR assay for detection and quantitation of Plasmodium by amplifying RNA and DNA of the 18S rRNA genes. J Clin Microbiol 49: 2946–2953.
- 17.↑
Vo TKD , Bigot P , Gazin P , Sinou V , De Pina JJ , Huynh DC , Fumoux F , Parzy D , 2007. Evaluation of a real-time PCR assay for malaria diagnosis in patients from Vietnam and in returned travellers. Trans R Soc Trop Med Hyg 101: 422–428.
- 18.↑
Miller RH , Obuya CO , Wanja EW , Ogutu B , Waitumbi J , Luckhart S , Stewart VA , 2015. Characterization of Plasmodium ovale curtisi and P. ovale wallikeri in Western Kenya utilizing a novel species-specific real-time PCR assay. PLoS Negl Trop Dis 9: 1–19.
- 19.↑
Hofmann N , Mwingira F , Shekalaghe S , Robinson LJ , Mueller I , Felger I , 2015. Ultra-sensitive detection of Plasmodium falciparum by amplification of multi-copy subtelomeric targets. PLoS Med 12: 1–21.
- 20.↑
Kattenberg JH , Tahita CM , Versteeg IAJ , Tinto H , Traoré-Coulibaly M , Schallig HDFH , Mens PF , 2012. Antigen persistence of rapid diagnostic tests in pregnant women in Nanoro, Burkina Faso, and the implications for the diagnosis of malaria in pregnancy. Trop Med Int Health 17: 550–557.
- 21.↑
Doolan DL , Dobaño C , Baird JK , 2009. Acquired immunity to malaria. Clin Microbiol Rev 22: 13–36.
- 23.↑
National Aids and STI Control Program, 2018. Kenya HIV Estimates2018. Available at: https://nacc.or.ke/wp-content/uploads/2018/11/HIV.
- 24.↑
Miller LH , Mason SJ , Clyde DF , Mcginniss MH , 1976. The resistance factor to Plasmodium vivax in Blacks: the Duffy-Blood-Group genotype, FyFy. N Engl J Med 295: 302–304.
- 25.↑
Jenkins R , Omollo R , Ongecha M , Sifuna P , Othieno C , Ongeri L , Kingora J , Ogutu B , 2015. Prevalence of malaria parasites in adults and its determinants in malaria endemic area of Kisumu County, Kenya. Malar J 14: 1–6.
- 26.↑
Hänscheid T , 2003. Current strategies to avoid misdiagnosis of malaria. Clin Microbiol Infect 9: 497–504.
- 27.
Ohrt C , Purnomo , Sutamihardja MA , Tang D , Kain KC , 2002. Impact of microscopy error on estimates of protective efficacy in malaria‐prevention trials. J Infect Dis 186: 540–546.
- 28.↑
Payne D , 1988. Use and limitations of light microscopy for diagnosing malaria at the primary health care level. Bull World Health Organ 66: 621–626.
- 29.↑
Leski TA et al. 2020. Use of real-time multiplex PCR, malaria rapid diagnostic test and microscopy to investigate the prevalence of Plasmodium species among febrile hospital patients in Sierra Leone. Malar J 19: 84.
- 30.↑
Zhao Y et al. 2017. Comparison of methods for detecting asymptomatic malaria infections in the China-Myanmar border area. Malar J 16: 159.
- 31.↑
Mayxay M , Pukrittayakamee S , Chotivanich K , Looareesuwan S , White NJ , 2001. Persistence of Plasmodium falciparum HRP-2 in successfully treated acute falciparum malaria. Trans R Soc Trop Med Hyg 95: 179–182.
- 32.↑
Tjitra E , Suprianto S , McBroom J , Currie BJ , Anstey NM , 2001. Persistent ICT malaria P.f/P.v Panmalarial and HRP2 antigen reactivity after treatment of Plasmodium falciparum malaria is associated with gametocytemia and results in false-positive diagnoses of Plasmodium vivax in convalescence. J Clin Microbiol 39: 1025–1031.
- 33.↑
Bull PC , Buckee CO , Kyes S , Kortok MM , Thathy V , Guyah B , Stoute JA , Newbold CI , Marsh K , 2008. Plasmodium falciparum antigenic variation. Mapping mosaic var gene sequences onto a network of shared, highly polymorphic sequence blocks. Mol Microbiol 68: 1519–1534.
- 34.↑
Kawamoto F et al. 1996. Sequence variation in the 18S rRNA gene, a target for PCR-based malaria diagnosis, in Plasmodium ovale from southern Vietnam. J Clin Microbiol 34: 2287–2289.
- 35.↑
Stiffler DM , Oyieko J , Kifude CM , Rockabrand DM , Luckhart S , Stewart VA , 2021. HIV-1 infection is associated with increased prevalence and abundance of Plasmodium falciparum gametocyte-specific transcripts in asymptomatic adults in western Kenya. Front Cell Infect Microbiol 10: 600106.
- 36.↑
Torrevillas BK et al. 2020. Plasmodium falciparum DHFR and DHPS mutations are associated with HIV-1 co-infection and a novel DHPS mutation I504T is identified in western Kenya. Front Cell Infect Microbiol 10: 600112.