Spatial distribution of G6PD deficiency variants across malaria-endemic regions

Abstract

Background: Primaquine is essential for malaria control and elimination since it is the only available drug preventing multiple clinical attacks by relapses of Plasmodium vivax. It is also the only therapy against the sexual stages of Plasmodium falciparum infectious to mosquitoes, and is thus useful in preventing malaria transmission. However, the difficulties of diagnosing glucose-6-phosphate dehydrogenase deficiency (G6PDd) greatly hinder primaquine's widespread use, as this common genetic disorder makes patients susceptible to potentially severe and fatal primaquine-induced haemolysis. The risk of such an outcome varies widely among G6PD gene variants.

Methods: A literature review was conducted to identify surveys of G6PD variant frequencies among representative population groups. Informative surveys were assembled into two map series: (1) those showing the relative proportions of the different variants among G6PDd individuals; and (2) those showing allele frequencies of G6PD variants based on population surveys without prior G6PDd screening.

Results: Variants showed conspicuous geographic patterns. A limited repertoire of variants was tested for across sub-Saharan Africa, which nevertheless indicated low genetic heterogeneity predominated by the G6PD A(-202A) mutation, though other mutations were common in western Africa. The severe G6PD Mediterranean variant was widespread across western Asia. Further east, a sharp shift in variants was identified, with high variant heterogeneity in the populations of China and the Asia-Pacific where no single variant dominated.

Conclusions: G6PD variants exhibited distinctive region-specific distributions with important primaquine policy implications. Relative homogeneity in the Americas, Africa, and western Asia contrasted sharply with the heterogeneity of variants in China, Southeast Asia and Oceania. These findings will inform rational risk assessments for primaquine in developing public health strategies for malaria control and elimination, and support the future development of regionally targeted policies. The major knowledge gaps highlighted here strongly advocate for further investigation of G6PD variant diversity and their primaquine-sensitivity phenotypes.

Figure 1

Survey inclusion criteria and G6PDd variant map outputs. Orange rectangles indicate the exclusion criteria, grey hexagons summarize the two final input data, and green rods represent the two map types.

Figure 2

G6PDd variant proportion map: Americas (map series 1). Map includes nine surveys with a mean sample size of 71 (range: 8-196); one survey was mapped at the national level. Pie charts represent individuals previously identified as G6PDd. Sample size is reflected in the size of the pie charts, which is transformed for clarity on a square-root scale. Surveys which could only be mapped to the country-level are indicated by a white star; this applied to seven surveys globally. Spatial duplicates from independent studies were mapped with a “jitter” of 0.5-1 decimal degrees in their latitude or longitude coordinates to allow visualization of the multiple charts. Malaria-endemic countries in the region mapped are shown with a yellow background; white backgrounds indicate endemic countries outside the region in focus; grey backgrounds represent malaria-free countries. G6PDd variants that could not be diagnosed were reported as “Other”.

Figure 3

G6PDd variant proportion map: Africa+ (map series 1). Map includes six surveys with a mean sample size of 43 (range: 5-110). One survey was mapped at the national level. (See Figure  2 for full legend).

Figure 4

G6PDd variant proportion map: Asia (map series 1). Map includes 87 surveys with a mean sample size of 53 (range: 1-532). Four surveys were mapped to the national level. (See Figure  2 for full legend).

Figure 5

G6PDd variant proportion map: West Asia (map series 1). A higher resolution map of Figure  4.

Figure 6

G6PDd variant proportion map: East Asia (map series 1). A higher resolution map of Figure  4.

Figure 7

G6PDd variant proportion map: Asia-Pacific (map series 1). Panel A show the full map, which includes 36 surveys with a mean sample size of 18 (range: 1-128). One survey was mapped at the national level. The East Nusa Tenggara province islands identified by the dotted box are shown at higher resolution in Panel B. (See Figure  2 for full legend).

Figure 8

G6PD A- variant SNP frequency map for Africa (map series 2). Bar charts representing the frequencies of G6PD variants. The variants that were tested for in each location are listed above the x-axis. Empty spaces along the x-axis indicate that the named variant was tested for but not identified in the population sample. Survey locations are mapped to closest approximation at the point of origin of the plots and exact locations shown in Additional file 4. Sample size is listed under each plot. Equivalent maps of the Americas and Asia are Additional files 3 and 5.