Marker classification and evidence review

Marker classification framework 

Different types of evidence were considered to determine the association between specific genetic alterations and antimalarial drug resistance. Alterations that met the domain-specific thresholds were considered for inclusion, while those that did not meet minimum criteria in any domain were excluded. 

Based on a combined assessment of evidence across the three domains, each genetic alteration was classified as a potential, candidate, or validated marker of antimalarial drug resistance. The Venn diagram illustrates the classification framework, integrating laboratory, clinical, and genetic epidemiology data to define the three categories of molecular markers, reflecting the level of confidence in the evidence supporting their association with drug resistance. 

Evidence evaluation

Molecular markers were evaluated using three domains of evidence encompassing different types of data: laboratory (A1-A3), clinical (B1-B3), and genetic epidemiology (C1). The data represented in the evidence reviewed and their respective strengths and limitations are outlined below (for further details, please refer to the full document on Methods and thresholds).

• Laboratory evidence provides direct insight into how genetic alterations affect parasite susceptibility under controlled in vitro conditions. The strongest evidence comes from transfection experiments (A1), which introduce specific alterations into a defined genetic background to confirm their effect on drug susceptibility. Other data sources include drug susceptibility assays on field isolates or culture-adapted laboratory strains (A2), or on progeny from genetic crosses (A3). While less conclusive, these studies still contribute valuable information on resistance mechanisms and potential evolution. The investigation of molecular markers of drug resistance in P. vivax has been significantly hampered by an inability to sustain P. vivax parasites in continuous in vitro culture. Ex vivo schizont maturation assays have been used for antimalarial drug susceptibility testing in field isolates, but these assays are confounded by the stage specificity of antimalarial activity and limited drug exposure.  As a result, laboratory evidence has been generated from heterologous expression of mutant genes in P. falciparum, P. knowlesi, bacteria, or yeasts.

• Clinical evidence links genetic alterations to patient outcomes by analysing delayed parasite clearance (B1), treatment failure (recrudescence) (B2), or selection of variants after treatment in a patient (B3). Clinical evidence is central, as therapeutic efficacy in patients remains the foundation for guiding malaria treatment policy. However, interpreting clinical outcomes can be complex, and causal attribution to a specific genetic alteration remains challenging. This is particularly true for artemisinin-based combination therapies (ACTs), where the effects of both components must be considered. Factors such as drug absorption, metabolism, and host immunity also influence clinical outcomes. In relapsing malaria species such as P. vivax, interpretation is further complicated by difficulties distinguishing recrudescence due to drug failure from reinfection or relapse.

• Genetic epidemiological evidence reflects parasite population dynamics and indicates the transmission or spread potential of a genetic alteration. A genetic alteration observed at significant prevalence may indicate its potential to spread under current conditions, including existing treatment practices and drug pressure (C1). However, establishing causality is challenging, particularly for combination therapies, and genetic epidemiology should always be interpreted with caution, serving primarily as supporting evidence alongside laboratory and clinical data.

Evidence reviewed 

For P. falciparum, 3 441 publications published before May 2025 were identified in an initial search. Of these, 863 articles contained potentially relevant evidence and were included in the expert review. Laboratory, clinical, and genetic epidemiological evidence corresponding to reported genetic alterations was evaluated against established thresholds, and classifications were refined through expert assessment of study quality and methodology. Articles that did not meet the predefined criteria, including those with weak or non-significant associations, very small sample sizes, or insufficient supporting data, were excluded. The P. falciparum compendium references 283 articles, which form the evidence base for the markers summarized in the detailed tables.

For non-P. falciparum species, a separate systematic review was conducted to identify articles published in any language between 1 January 1960 and 31 July 2025 that reported evidence of resistance or molecular markers in P. vivax, P. malariae, P. ovale wallikeri, P. ovale curtisi, or P. knowlesi. The review focused on markers homologous to known resistance markers in P. falciparum. MEDLINE, Web of Science, and Embase were searched using combinations of terms related to Plasmodium species, resistance, and candidate resistance genes (plasmodium AND (vivax or ovale or malariae or knowlesi) AND resistan* AND (mdr* OR crt* OR mrp* OR dhfr OR dhps OR Cytochrome B OR Kelch OR K12 OR plasmepsin)). The search identified 309 articles. After title and abstract screening, 89 were considered informative and underwent full-text review. Evidence meeting the review criteria was identified only for P. vivax. No eligible evidence of drug resistance markers homologous to known P. falciparum resistance markers was identified for P. knowlesi, P. ovale curtisi, P. ovale wallikeri, or P. malariae.