Evaluation of a Novel Quantitative Test for Glucose-6-Phosphate Dehydrogenase Deficiency: Bringing Quantitative Testing for Glucose-6-Phosphate Dehydrogenase Deficiency Closer to the Patient

Abstract

Glucose-6-phosphate dehydrogenase (G6PD) deficiency, a common genetic blood condition, can result in kernicterus at birth, and later in life as severe hemolysis on exposure to certain infections, foods, and drugs. The unavailability of point-of-care tests for G6PD deficiency is a barrier to routine curative treatment of Plasmodium vivax malaria with 8-aminoquinolines, such as primaquine. Two quantitative reference tests (Trinity Biotech, Bray, Ireland and Pointe Scientific, Canton, MI; Cat No. G7583) and the point-of-care STANDARD™ G6PD test (SD Biosensor, Suwon, South Korea) were evaluated. The STANDARD G6PD test was evaluated at multiple temperatures, in anticoagulated venous and capillary samples, including 79 G6PD-deficient and 66 intermediate samples and across two laboratories, one in the United States and one in Thailand. The STANDARD test performed equivalently to a reference assay for its ability to diagnose G6PD deficiency (< 30% normal) with a sensitivity of 100% (0.95 confidence interval [CI]: 95.7-100) and specificity of 97% (0.95 CI: 94.5-98.5), and could reliably identify females with less than 70% normal G6PD activity with a sensitivity of 95.5% (0.95 CI: 89.7-98.5) and specificity of 97% (0.95 CI: 94.5-98.6). The STANDARD G6PD product represents an opportunity to diagnose G6PD deficiency equally for males and females in basic clinical laboratories in high- and low-resource settings. This quantitative point-of-care diagnostic test for G6PD deficiency can provide equal access to safe radical cure of P. vivax cases in high- and low-resource settings, for males and females and may support malaria elimination, in countries where P. vivax is endemic.

Conflict of interest statement

Conflict of Interest: PATH supports a portfolio of G6PD test development efforts including SD Biosensor. Neither PATH nor any of the authors participating on this article have financial interests in the commercialization of any resulting products.

Figures

Figure 1.

Correlation between two reference quantitative assays for glucose-6-phosphate dehydrogenase (G6PD) activity: Pointe Scientific and Trinity Biotech. (A) Linear regression comparing the G6PD values measured on the same specimens by both assays. Glucose-6-phosphate dehydrogenase values normalized for hemoglobin are shown (U/g Hb). (B) Bland–Altman plot comparing the two data sets. Lines indicating the mean, 1.96 times the standard deviation (SD), and a difference in 2 U/g Hb in red are shown. (C) Receiving operating characteristic curve analysis used to establish optimal threshold values for 30% (left) and (D) 70% (right). This figure appears in color at www.ajtmh.org.

Figure 2.

Temperature robustness for glucose-6-phosphate dehydrogenase (G6PD) activity (U/g Hb) and hemoglobin (g/dL) measurement of the STANDARD G6PD test. The STANDARD G6PD test performance was evaluated against the reference assay across multiple temperatures for 210 clinical specimens, including 25 G6PD-deficient and 13 G6PD-intermediate specimens. (A) Linear regression plot (left) and corresponding Bland–Altman plot (right) for matched specimens run at 22°C, 32°C 50% humidity, 37°C 50% humidity, and 37°C 75% humidity; a total of 414 data points. Lines indicating the mean, 1.96 times the standard deviation, and a difference in 2 U/g Hb in red are shown on the Bland–Altman plots. (B) Linear regression plot (left) and corresponding Bland–Altman plot (right) for the hemoglobin values (g/dL) for the same 414 data points shown in A on the STANDARD G6PD test compared with the Hemocue hemoglobin values (used to normalize the reference Pointe Scientific G6PD values). (C) Receiving operating characteristic curve analysis for the G6PD values used to establish optimal threshold values for 30% (left) and 70% (right). This figure appears in color at www.ajtmh.org.

Figure 3.

Performance of the STANDARD glucose-6-phosphate dehydrogenase (G6PD) test with frozen blood samples from a Thai–Myanmar border population. STANDARD G6PD test performance was evaluated against the reference assay with 150 clinical specimens: 54 G6PD-deficient (red), 53 G6PD-intermediate (orange), and 43 normal (green) specimens. (A) Linear regression plot (left) and corresponding Bland–Altman plot (right) for the G6PD values. (B) Linear regression plot (left) and corresponding Bland–Altman plot (right) for the hemoglobin values. Lines indicating the mean, 1.96 times the standard deviation (SD), and a difference in 2 U/g Hb in red are shown on the Bland–Altman plots. (C) Receiving operating characteristic curve analysis used to establish optimal threshold values for 30% (left) and 70% (right). This figure appears in color at www.ajtmh.org.