Abstract
Histidine-rich protein 2- (HRP2-) based rapid diagnostic tests (RDTs) are widely used to detect Plasmodium falciparum in sub-Saharan Africa. Reports of parasites with pfhrp2 and/or pfhrp3 (pfhrp2/3) gene deletions in Africa raise concerns about the long-term viability of HRP2-based RDTs. We evaluated changes in pfhrp2/3 deletion prevalence over time using a 2018-2021 longitudinal study of 1,635 enrolled individuals in Kinshasa Province, Democratic Republic of the Congo (DRC). Samples collected during biannual household visits with ≥ 100 parasites/μL by quantitative real-time PCR were genotyped using a multiplex real-time PCR assay. Among 2,726 P. falciparum PCR-positive samples collected from 993 participants during the study period, 1,267 (46.5%) were genotyped. No pfhrp2/3 deletions or mixed pfhrp2/3-intact and -deleted infections were identified in our study. Pfhrp2/3-deleted parasites were not detected in Kinshasa Province; ongoing use of HRP2-based RDTs is appropriate.
Body
Progress toward malaria control and elimination in Africa requires prompt diagnosis and treatment with effective antimalarial drugs. Rapid diagnostic tests (RDTs) are widely used to identify individuals infected with Plasmodium; their deployment enabled significant improvements in malaria diagnostic testing across the continent over the past 15 years. The majority of RDTs used to diagnose falciparum malaria in sub-Saharan Africa detect P. falciparum histidine-rich protein 2 (HRP2) and its paralog HRP3, encoded by the pfhrp2 and pfhrp3 genes, respectively. HRP2-based RDTs are generally more sensitive and heat-stable than RDTs detecting other antigens.1 However, test-and-treat strategies that rely upon HRP2-based RDTs are threatened by the emergence of P. falciparum strains that escape detection due to deletion of the pfhrp2 and/or pfhrp3 (pfhrp2/3) genes.2,3 High prevalence of pfhrp2/3- deleted parasites in Eritrea,4 Ethiopia,5,6 Djibouti,7,8 and surrounding countries recently prompted changes in malaria diagnostic testing policies. Reports from sub-Saharan countries outside of the Horn of Africa, however, indicate lower prevalence.9
The Democratic Republic of the Congo (DRC) has one of the highest malaria burdens in the world, accounting for 12% of global malaria cases and deaths.2 In the DRC, we previously reported 6.4% pfhrp2 deletion prevalence in samples from a 2013-2014 nationally representative cross-sectional survey of asymptomatic children under five years of age.10 However, no pfhrp2/3 deletions were observed in our 2017 cross-sectional study of symptomatic children and adults across three DRC provinces.11 These studies were both cross-sectional and did not provide information about how the prevalence of deletions may be changing over time. This study aims to estimate the pfhrp2/3 deletion prevalence and changes over time in Kinshasa Province, DRC.
This study includes samples collected as part of a 2018-2021 longitudinal study of malaria conducted at seven sites across Kinshasa Province, DRC. A total of 1,635 participants were enrolled in one urban neighborhood, three peri-urban villages, and three rural villages (Figure 1). Study visits were conducted as part of twice-yearly household surveys in the dry and rainy seasons (active surveillance) and as part of routine care at local health centers (passive surveillance) as previously described.12,13 At each visit, a comprehensive questionnaire on malaria symptoms and treatment, and bed net usage was administered. A finger- or heel-prick sample was obtained at each visit for RDT (SD Bioline Ag P.f./Pan RDT [05FK60], Alere, Gyeonggi-do, Republic of Korea, or CareStart, Access Bio, Somerset, U.S.A.) and dried blood spot (DBS) preparation for future molecular investigation. RDT-positive patients were treated according to national guidelines.
Study sites in Kinshasa Province, DRC
DNA was extracted from DBS using Chelex-100 and saponin or tween as previously described. 14,15 Quantitative real-time PCR (qPCR) targeting the P. falciparum lactate dehydrogenase (pfldh) gene was used to estimate P. falciparum parasitemia, using serial dilutions of DNA extracted from a mock DBS made with cultured P. falciparum 3D7 or FCR3 strain parasites at known parasite density. Samples with ≥ 100 parasites/μL were selected for pfhrp2/3 deletion identification using a multiplex real-time PCR assay that detects pfldh, pfhrp2, pfhrp3, and human beta-tubulin (HumTuBB).16 We used this parasite density threshold to reduce the risk of unintentional misclassification of deletions in the setting of low DNA concentrations.17,18 Positive calls required cycle threshold (Ct) values < 35. Samples positive for HumTuBB and pfldh but negative for pfhrp2 or pfhrp3 were subjected to a confirmatory real-time PCR targeting the P. falciparum beta-tubulin (PfBtubulin) gene. Deletion calls were limited to samples positive for HumTuBB and both single-copy P. falciparum genes (pfldh and pftubulin), but negative for pfhrp2 and/or pfhrp3. Mixed infections of pfhrp2/3-intact and -deleted strains were defined conservatively as samples in which (pfhrp2 Ct - pfldh Ct) >3 or (pfhrp3 Ct - pfldh Ct) > 3. All assays included P. falciparum 3D7, DD2, and HB3 strain DNA as wildtype, pfhrp2-deleted, and pfhrp3-deleted controls, respectively. All participants provided informed consent. Ethical approval for this study was granted by the Institutional Review Boards of the University of North Carolina-Chapel Hill and the Kinshasa School of Public Health.
A total of 1,267 samples collected from 649 individuals in 179 households between 2018 and 2021 were included in this study (Figure 2). Among these, the median number of P. falciparum infections with ≥ 100 parasites/μL detected per participant was 2.0 (IQR: 1-3). The median age at enrollment in 2018 was 9 years old (IQR: 5-15 years old); 48.8% reported female gender. At the time of enrollment, 48.7% reported malaria in the preceding 6 months, and 42.7% of those reported more than one episode in that 6-month period. The median household size was 8 (IQR: 6-10), with high bed net coverage across the study population (90.2%). The baseline characteristics of the study population are summarized in Table 1.
Baseline characteristics of the study population included in this analysis (1,267 samples with ≥ 100 Plasmodium falciparum parasites/μL from 649 participants), during enrollment in the Kinshasa longitudinal study.
Flow diagram for the detection of pfhrp2/3 deletion in 1,267 samples from 649 participants in phase two of the longitudinal study (2018-2021) in Kinshasa Province, DRC
All 1,267 samples had detectable human DNA as indicated by amplification of HumTuBB with Ct < 35. The multiplex PCR confirmed P. falciparum parasitemia in all but 2 samples which failed to amplify pfldh. Both pfhrp2 and pfhrp3 were negative in one sample (Ct > 35 for both gene targets). However, pfldh was negative in the multiplex assay and Pftubulin negative in follow-up testing, indicating that the original pfldh qPCR result was a false-positive. Thus, this sample did not meet criteria to be considered a pfhrp2/3-deleted parasite. There were no mixed infections of pfhrp2/3-intact and -deleted strains identified in our study population.
Overall, we did not find evidence of pfhrp2/3 deletion in parasites sampled during our longitudinal cohort in Kinshasa Province. Our conservative approach to calling pfhrp2/3 deletions prevents us from detecting low-density infections and thus could underestimate the true prevalence of these strains. These study results differ from our previous DRC pfhrp2/3 deletion prevalence estimates; this discrepancy could be attributed to spatial heterogeneity in prevalence, incorrect deletion calls due to laboratory artifact, or different approaches used to identify pfhrp2/3 deletions. However, our finding of no deletions is in line with our more recent study of symptomatic individuals and a similar study of asymptomatic and symptomatic school-aged children that showed no to little evidence of pfhrp2/3 deletion in the DRC.11,19 Together, these results support the continued use of HRP2-based RDTs for the diagnosis of malaria in Kinshasa Province, DRC.
Data Availability
All data produced in the present study are available upon reasonable request to the authors.
Financial support
This study was funded by R01AI132547 to JJJ and RRD, with partial support from R01AI129812 to AKT, R01AI139520 to JAB and a supplement to RF, T32AI070114 to KB, K24AI134990 to JJJ, and an ASTMH/Burroughs-Wellcome Fund award to JBP.
Disclosures
JBP reports research support from Gilead Sciences and non-financial support from Abbott Laboratories, outside the scope of this manuscript.
Co-author contact information
Ruthly François: University of North Carolina at Chapel Hill, Chapel Hill, NC, United States; ruthly_francois{at}med.unc.edu
Melchior Mwandagalirwa Kashamuka2: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; mkashamuka{at}yahoo.com
Kristin Banek1: University of North Carolina at Chapel Hill, Chapel Hill, NC, United States; kristin_banek{at}med.unc.edu
Joseph A. Bala2: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; jalexandrebala{at}yahoo.fr
Marthe Nkalani2: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; marthenkalani{at}gmail.com
Georges Kihuma2: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; georgeskihuma{at}gmail.com
Joseph Atibu2: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; fejef576{at}gmail.com
Georges E. Mahilu2: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; emomahilu{at}gmail.com
Kyaw L. Thwai1: University of North Carolina at Chapel Hill, Chapel Hill, NC, United States; thwai{at}email.unc.edu
Ashenafi Assefa1: University of North Carolina at Chapel Hill, Chapel Hill, NC, United States; ashenafi_assefa{at}med.unc.edu
Jeffrey A. Bailey3: Brown University, Providence, RI, United States; jeffrey_bailey{at}brown.edu
Rhoel R. Dinglasan4: University of Florida, Gainesville, FL, United States; rdinglasan{at}epi.ufl.edu
Jonathan J. Juliano: University of North Carolina at Chapel Hill, Chapel Hill, NC, United States; jonathan_juliano{at}med.unc.edu
Antoinette Tshefu2*: Kinshasa School of Public Health, Kinshasa, Democratic Republic of the Congo; antotshe{at}yahoo.com
Jonathan B. Parr: University of North Carolina at Chapel Hill, Chapel Hill, NC, United States; jonathan_parr{at}med.unc.edu
Acknowledgements
The authors thank the study participants and field teams who conducted study visits. They also wish to express their gratitude to the late Prof. Steven Meshnick for mentorship and his role in the longitudinal study upon which this analysis was based. The following reagents were obtained through BEI Resources, NIAID, NIH: Genomic DNA from P. falciparum strain 3D7, MRA-102G, contributed by Daniel J. Carucci; P. falciparum strain HB3, MRA-155G, contributed by Thomas E. Wellems; P. falciparum strain Dd2, MRA-150G, contributed by David Walliker.
Footnotes
↵# Co-senior authors
