Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection.

Neonatal jaundice is a common and severe disease in premature infants with Glucose-6-Phosphate Dehydrogenase (G-6-PD) deficiency. The World Health Organization (WHO) has recommended screening for G-6-PD deficiency in newborns for early recognition as well as to prevent unwanted outcomes in a timely manner. The present study aimed to assess a point-of-care, careSTARTTM G6PD biosensor as a quantitative method for the diagnosis of G-6-PD deficiency. Factors influencing the evaluation of G-6-PD enzyme activity were examined in 40 adults, including ethylenediaminetetraacetic acid (EDTA) anticoagulant, hematocrit concentration, storage temperature and time. Analytic performance of the careSTARTTM G6PD biosensor was evaluated in 216 newborns and compared with fluorescent spot test (FST) and standard quantitative G-6-PD enzyme activity (SGT) assay. The results of factors affecting the G-6-PD enzyme activity showed that the activity determined from finger-prick was not statistically different from venous blood (p = 0.152). The G-6-PD value was highly dependent on the hematocrit and rose with increasing hematocrit concentration. Its activity was stable at 4°C for 3 days. Reliability analysis between the careSTARTTM G6PD biosensor and SGT assay showed a strong correlation with a Pearson's correlation coefficient of 0.82 and perfect agreement by intraclass correlation coefficient (ICC) of 0.90. Analysis of the area under the Receiver Operating Curve (AUC) illustrated that the careSTARTTM G6PD biosensor had 100% sensitivity, 96% specificity, 73% positive predictive value (PPV), 100% negative predictive value (NPV) and 97% accuracy at 30% of residual activity. While the diagnostic ability for identifying G-6-PD deficiency had 78% sensitivity, 89% specificity, 56% positive predictive value (PPV), 96% negative predictive value (NPV) and 88% accuracy when stratified by gender. The careSTARTTM G6PD biosensor is an attractive option as a point-of-care quantitative method for G-6-PD activity detection. Quantification of G-6-PD enzyme activity in newborns is the most effective approach for the management of G-6-PD deficiency to prevent severe jaundice and acute hemolysis.

Introduction

Glucose-6-phosphate dehydrogenase (G-6-PD) deficiency affects millions of people worldwide and comprises the most common inherited blood disorder [1-3]. Defective G-6-PD enzymes cause the increased susceptibility of red blood cells to reactive oxygen species, leading to hemolysis [1, 4]. Its clinical manifestations vary from asymptomatic to severe hemolytic anemia depending on the residual G-6-PD enzyme activity [1, 5]. Affected individuals generally present with acute hemolytic anemia, favism, or chronic non-spherocytic hemolytic anemia [1]. The clinical symptoms tend to increase as the red blood cells undergo oxidative stress triggered by agents such as anti-malarial drugs (primaquine), mothballs, infection or the ingestion of fava beans [1, 3–9]. Several studies have reported that diabetes, myocardial infarction and strenuous physical exercise can stimulate hemolysis in individuals with G-6-PD deficiency [1, 7]. Notably, affected neonates frequently show severe jaundice. Lack of awareness concerning G-6-PD deficiency in newborns could cause extreme hyperbilirubinemia, bilirubin neurotoxicity, kernicterus and, eventually, mental retardation [8, 10]. Accordingly, the World Health Organization (WHO) has recommended screening for G-6-PD deficiency in newborns to promote early diagnosis and prevent unwanted outcomes in a timely manner [8, 10–13]. Moreover, individuals suffering from severe G-6-PD deficiency with residual activity less than 30% and 70% should be excluded from primaquine and tafenoquine administration, respectively [14, 15]. Thus, the measurement of G-6-PD activity is a crucial step before starting malaria treatment. Although the fluorescent spot test (FST) is widely used for qualitative screening of G-6-PD deficiency, its discriminative power appears limited, requiring UV visualization and highly specific skills for interpretation [6, 8, 16–19]. The reference method for G-6-PD deficiency detection is based on a quantitative G-6-PD enzymatic assay. However, it is time-consuming, laborious and requires a spectrophotometer, which may not be suitable for the field or large-scale testing [17, 18]. Quantitative point-of-care G-6-PD tests are an essential tool in low-resource settings. Recently, the point-of-care quantitative careSTARTTM G6PD biosensor was developed as an alternative to the labor-intensive standard G-6-PD enzymatic method.

The careSTARTTM G6PD biosensor is an electrochemical biosensor that measures the electron transfer from the change of nicotinamide adenine dinucleotide phosphate (NADPH) into its reduced form by using G-6-PD enzymes. It is proposed for the quantitative measurement of G-6-PD enzyme activity in whole blood. Therefore, the present study aimed to examine the performance of the careSTARTTM G6PD biosensor compared with the gold standard enzymatic assay. Moreover, the most widely used FST method was evaluated. Measurement interference effects were also investigated, including ethylenediaminetetraacetic acid (EDTA), hematocrit concentration and enzymatic stability.

Materials and methods

Sample collection

Ethical approval for this study was obtained from the Third Ethics Committee of Thammasat University, Pathum Thani Province, Thailand (COA No. 150/2561). Drawing blood from a newborn can be difficult, potentially harmful and may yield limited sampling blood volume. Therefore, the factors affecting the G-6-PD enzyme activity measurement were evaluated in 40 adults. Prior to enrolment, all volunteers provided informed written consent to participation in the study as well as publication of the results. The volunteers included 20 males and 20 females ranging in age between 18 and 24 years old. The blood samples were collected from both finger-prick (5 μL) and venous puncture (3 mL in K2EDTA vacutainer tubes), after which the samples were analyzed by standard quantitative G-6-PD enzymatic assay (OSMMR2000-D G-6-PD kit, R&D Diagnostics, Ltd., Greece)

To evaluate the performance of the careSTARTTM G6PD biosensor (WELLS BIO, INC., Korea), 216 neonatal blood samples ranging in age between 1 to 35 years old were collected and kept in a microtainer tube with K2EDTA anticoagulant. The ethics committee provided a waiver of informed consent due to the neonatal blood samples used in this study being leftover samples following routine analysis. All data samples were de-identified before access. Complete blood count (CBC) (DxH 800 hematology analyzer, Beckman Coulter, Inc., USA) and fluorescent spot test (FST) (SQMMR500, R&D Diagnostics, Ltd., Greece) were assessed to screen for G-6-PD deficiency. G-6-PD enzyme activity was measured by the careSTARTTM G6PD biosensor compared with standard quantitative G-6-PD enzymatic assay (SGT).

All blood samples were carried out a single time following standard operating procedures under controlled time and temperature.

Assessment of factors influencing G-6-PD enzyme activity

Capillary blood samples were immediately assessed for G-6-PD enzyme activity by the standard SGT. Meanwhile, venous blood was aliquoted into 3 parts for assessment of the factors that affect G-6-PD enzyme activity, including anticoagulant (EDTA), hematocrit (Hct) concentration and enzymatic stability.

The effect of EDTA anticoagulant collected from venous blood was investigated by measuring G-6-PD enzyme activity and comparing with the sample from finger-prick.

Due to the activity of G-6-PD enzyme depending on the storage time and temperature, EDTA blood was aliquoted into a sealed, dark 1.5 mL microcentrifuge tube and kept at room temperature (25°C) and 4°C for 3 days. G-6-PD enzyme activity was measured at 1, 2, 6, 12, 24, 48 and 72 hours.

The proportion of red blood cells to total blood volume (Hct) varies substantially depending on race, age and sex, which are commonly used to determine the presence of anemia. In this study, the Hct levels of each of the 40 samples were adjusted to 30%, 40%, 50% and 60% by using their own plasma. Each Hct level was assessed for G-6-PD enzyme activity and compared to each other.

The study of factors influencing G-6-PD enzyme activity was performed using the SGT assay according to the manufacturer’s instructions. Briefly, five microliters of blood were mixed with the Elution Buffer (75 μL) in a U-bottom microtiter plate for 10 min at room temperature (25±2°C). The eluted (15 μL) blood was transferred to 75 μL of the Reagent Mixture contained in a new flat-bottom microtiter plate and mixed thoroughly. Then, 100 μL of Color Reagent Mixture was added to each well and G-6-PD activity reaction (kinetic mode) was read with an ELISA reader (800TS Microplate Reader, BioTek, USA) at 550 nm for 15 min at 1 min intervals. After the final reading was taken, the plate was read at wavelength 405 nm to get the hemoglobin (Hb) content of each sample. G-6-PD enzyme activity was calculated by comparing the rates of a blood sample to the rate of normal control with known G-6-PD activity. The reported G-6-PD activity was normalized to Hb and the results expressed directly into international units per gram of hemoglobin (IU/gHb).

All experiments were conducted for the quality control of deficient blood (2.0 IU/g Hb with the specification of 1.0–3.0 IU/gHb) and normal blood (14.6 IU/gHb with the specification of 9.5–19.7IU/gHb in duplicate.

Performance of the careSTARTTM G6PD biosensor for G-6-PD enzyme activity detection

The diagnostic ability of the quantitative careSTARTTM G6PD biosensor was tested using 216 neonatal blood samples. According to the manufacturer’s instructions, ten microliters of blood pricked by lancet were applied to the end of the test strip, which was inserted into the analyzer. After 4 mins, the G-6-PD enzyme activity result was displayed on the analyzer’s monitor in terms of units per deciliter (IU/dL). The activity derived from the biosensor measured the G-6-PD enzymes from RBC in one deciliter, which should be divided by Hb (IU/gHb). The EDTA blood was also used for identifying G-6-PD deficiency by qualitative FST assay. Five microliters of EDTA blood was added to 100 μL of G-6-PD substrate reagent, mixed thoroughly and incubated at 25°C for 10 min. After that, the mixture (100 μL) was applied to filter paper and observed under UV light. The results acquired by the careSTARTTM G6PD biosensor and FST were compared with standard quantitative G-6-PD assay (SGT) as a reference assay.

Statistical analysis

Statistical analysis was performed using MedCal software (version 18.2.1; MedCalc, Mariakerke, Belgium) and IBM SPSS statistics (IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp software). Since all data were normally distributed, the parametric statistics test was used throughout the study. The mean, median, standard deviation (SD) and ranges were determined for all G-6-PD enzyme activity values with a 95% confidence interval (CI). Normal G-6-PD activity or 100% activity can be defined by the median value of hemizygous males with a G-6-PD normal allele, called adjusted male median (AMM), which is calculated according to the description of Domingo GJ et al. [20]. Any males and females who have G-6-PD activity less than to 30% of the AMM must be regarded G-6-PD deficient [20-22]. Meanwhile, males with G-6-PD activity of 30% or more of the AMM can be identified as G-6-PD normal. Heterozygous females can have intermediate status with G-6-PD activity levels ranging from 30% to 80% of AMM. A G-6-PD activity value of 80% or more of the AMM is considered G-6-PD normal [20-22]. Also, p values less than 0.05 indicate a statistically significant difference between the groups. A paired t-test was used to evaluate the difference of G-6-PD enzyme activity measured from finger-prick and venous blood keeping in EDTA vacutainer tube. The effect of Hct concentration and storage time was assessed by one-way ANOVA running with a post hoc test. A linear regression model was used to predict the time at which G-6-PD enzymatic activity fell below 90% of the median initial value. Pearson correlation (r) and Bland-Altman plots were performed to evaluate the relationship, the degree of agreement between careSTARTTM G6PD biosensor and the reference SGT assay. Analysis of the area under the Receiver Operating Curve (AUC) under the receiver operating characteristic (ROC) curve was performed between the careSTARTTM G6PD biosensor and a quantitative SGT assay at 30%, 70% and 80% of AMM. The performance of the careSTARTTM G6PD biosensor included sensitivity, specificity, positive likelihood ratio (+LR), negative likelihood ratio (-LR), positive predictive value (PPV), negative predictive value (NPV) and disease prevalence was tested applying standard formulas [23]. The diagnostic potential of the careSTARTTM G6PD biosensor and FST analyzed in clinical samples was determined between genders, assuming the SGT as the reference method.

Results

Study population and distribution of G-6-PD enzyme activity

Study population characteristics including mean, median and range of G-6-PD activity were stratified by gender (Table 1). G-6-PD enzyme activity measured by quantitative SGT was taken for normal distribution. Forty adult samples from males and females with an average age of 21 (21 ± 1 year) and Hct level of 42% (42 ± 3.7%) were studied. The results demonstrated that there were no significant differences in enzyme activity between females (5.8 ± 1.9 IU/gHb) and males (6.4 ± 2.1 IU/gHb) (p = 0.328). The adult adjusted male median was 6.6 IU/gHb. In newborns with an average age of 4 (4 ± 4 day), complete blood count (CBC) and G-6-PD enzyme activity were investigated. The CBC results showed that the RBC count, hemoglobin (Hb) and hematocrit (Hct) had statistically significant differences between males and females (p<0.05) (S1 Table). As shown in Table 1, no significant differences in G-6-PD enzyme activity were observed between females (7.6 ± 1.9 IU/gHb) and males (7.1 ± 3.4 IU/gHb) (p = 0.235). The AMM of newborns was 8.1 IU/gHb, representing 100% G-6-PD activity in the study population.

Table 1

Proposed reference values for G-6-PD enzyme activity in this study population.

Reference values	Total	Female	Male	Adjusted male
Adult
Number of cases	40	20	20	20
Mean (95% CI; IU/gHb)	6.1 (5.5–6.7)	5.8 (4.9–6.7)	6.4 (5.4–7.4)	6.4 (5.4–7.4)
Standard deviation	1.9	1.9	2.1	2.1
Median (95% CI; IU/gHb)	6.4 (5.0–6.7)	5.6 (4.9–6.8)	6.6 (4.8–7.4)	6.6 (4.8–7.4)
Range (IU/gHb)	2.5–10.7	2.5–9.4	2.7–10.7	2.7–10.7
Newborns
Number of cases	216	86	130	121
Mean (95% CI; IU/gHb)	7.3 (6.9–7.7)	7.6 (7.1–8.0)	7.1 (6.5–7.7)	7.6 (7.0–8.1)
Standard deviation	2.9	1.9	3.4	2.9
Median (95% CI; IU/gHb)	7.8 (7.5–8.1)	7.5 (7.1–8.1)	8.0 (7.8–8.2)	8.1 (7.8–8.5)
Range	0.1–13.3	1.7–10.8	0.1–13.3	0.9–13.3

The effect of certain factors on G-6-PD activity measurement was assessed in 40 adults by quantitative SGT. G-6-PD enzymatic activity collected by finger-prick was compared to venous blood keeping in EDTA vacutainer tube. The results demonstrated that the mean of G-6-PD activity from venous blood (6.1 ± 1.9 IU/gHb, 95%CI: 5.5 to 6.7) was slightly higher than the finger-prick (5.6 ± 1.6 IU/gHb, 95%CI: 5.1 to 6.1) with no statistically significant difference (p = 0.257) (Fig 1A). Activity of G-6-PD enzyme rose significantly with increasing hematocrit (Hct) concentration, accounting for 4.5 ± 0.3, 6.4 ± 0.2, 7.9 ± 0.3 and 8.7 ± 0.2 IU/gHb for 30%, 40%, 50% and 60%, respectively (Fig 1B). A Tukey post hoc test revealed that G-6-PD activity showed a statistically significant difference among groups (p<0.05) (S2 Table). The results implied that hematocrit concentration is a dependent factor influencing G-6-PD activity detection.

Fig 1

Effect of G-6-PD enzyme activity detection on type of blood collection (A) and hematocrit concentration (B). * indicates a significant difference among groups.

The stability of G-6-PD enzymatic activity was evaluated after storage for 3 days at room temperature (temperature range from 25°C to 30°C) and 4°C (Fig 2). G-6-PD activity was measured immediately (0), 1, 2, 6, 12, 24, 48, and 72 hours after blood collection. For the room temperature condition, the mean of G-6-PD enzyme activity was decreased gradually over time and fell to 5.1 ± 0.2 IU/gHb (95%CI: 4.7 to 5.5) at 72 hours (a mean fractional fall of 1.1 IU/gHb) (Fig 2A). On the other hand, there was no statistically significant difference when the blood was kept at 4°C for 72 hours with mean activity of 6.1 ± 0.2 IU/gHb (95%CI: 5.7 to 6.4) and a mean fractional fall of 0.2 IU/gHb) (Fig 2B). A drop in G-6-PD activity was statistically correlated with its activity at baseline within 72 hours at room temperature and 4°C (p = 0.001 and p = 1.000) (S3 Table). The regression analysis indicated that G-6-PD activity would decrease to 5.5 IU/gHb (10% from median 6.1 IU/gHb) for 2 days at room temperature and 182 days at 4°C (Fig 2C and 2D, respectively). The results suggest that G-6-PD enzyme activity can remain stable at 4°C for 3 days.

Fig 2

Variation of G-6-PD activity detection after storage for 72 hours at room temperature (A) and 4°C (B). The ●, ■, ▲, and ♦ symbols indicate outlier data with less or greater than the interquartile range (<5th or > 95th percentile). Regression analysis of G-6-PD activity detection after storage for 72 hours under room temperature (C) and 4°C (D).

G-6-PD status was assessed in all 216 neonates by CBC, FST, careSTARTTM G6PD biosensor and quantitative SGT. The mean delay between blood collection and laboratory processing was 22.4 hours (range from 12 to 26 hours). Efficacy evaluation involving the correlation analysis, sensitivity, specificity, positive likelihood ratio (+LR), negative likelihood ratio (-LR), positive predictive value (PPV) and negative predictive value (NPV), the hemoglobin (Hb) data from CBC results were applied to normalize the G-6-PD enzyme activity. Correlation analysis for G-6-PD enzyme activity measurement between the careSTARTTM G6PD biosensor and SGT assay was very strong with Pearson’s correlation coefficient of 0.82 (95%CI: 0.78 to 0.86, p = 0.000) (Fig 3A). Intraclass correlation analysis between the two assays showed good agreement with the coefficient of 0.90 (95%CI: 0.87 to 0.92). The mean difference between the SGT and careSTARTTM G6PD biosensor was 2.9 U/g Hb with 95% limit of agreement ranging from 6.6 to -0.8 U/g Hb (Fig 3B). The results indicated that 95% of the differences between the two assays were within this range.

Fig 3

Correlation analysis between the careSTARTTM G6PD biosensor and standard G-6-PD enzymatic test (SGT) analyzed by Pearson’s correlation analysis (A) and Bland-Altman plot (B).

The classification of G-6-PD deficiency relies on the guide of G-6-PD deficiency rapid diagnostic testing to support the curative treatment of malaria [14, 22] and WHO prequalification [24]. The receiver operator curve (ROC) analysis between the careSTARTTM G6PD biosensor and SGT was considered at 30%, 70% and 80% G-6-PD activity (Fig 4). Analysis of the area under the ROC curve (AUC) was 0.982 (95%CI: 0.954 to 0.995), 0.993 (95%CI: 0.971 to 1.000) and 0.978 (95%CI: 0.948 to 0.993) in 30%, 70% and 80% activity, respectively. The comparison analysis of AUC showed a statistical difference between areas (p < 0.001) at all thresholds (Fig 4). The performance of the careSTARTTM G6PD biosensor included sensitivity, specificity, +LR, -LR, PPV, NPV and disease prevalence, as illustrated in Table 2. A suitable cut-off was observed at 30% residual activity (≤ 2.4 IU/gHb) with 100% sensitivity, 96% specificity, 28 of +LR, 73% PPV, 100% NPV, 97% accuracy and 8.8% disease prevalence.

Fig 4

Receiver Operator Characteristic (ROC) analysis of careSTART G6PD biosensor. (biosensor) with standard G-6-PD enzymatic test (SGT) at 30% (A), 70% (B) and 80% (C) cut-off. AUC means the area under the ROC curve.

Table 2

Performance of the careSTARTTM G6PD biosensor for different residual G-6-PD activity.

Cut-off	G-6-PD activity cut-off value (IU/gHb)	Sensitivity(95%CI)	Specificity (95%CI)	Positive likelihood ratio(95%CI)	Negative likelihood ratio(95%CI)	Positive predictive value(95%CI)	Negative predictive value(95%CI)	Accuracy(95%CI)	Disease prevalence(95%CI)
30%	<2.4	100%(82–100)	96%(93–99)	28(14–58)	0	73%(57–85)	100%	97%(93–99)	8.8%(5–13)
70%	≤5.7	100%(89–100)	93%(89–97)	15(9–26)	0	73%(61–83)	100%	94%(91–97)	15%(11–21)
80%	≤6.5	100%(92–100)	91%(85–95)	11(7–17)	0	73%(63–81)	100%	93%(88–96)	20%(15–26)

Assessment for the clinical diagnostic ability of the careSTARTTM G6PD biosensor was analyzed with standard SGT by gender (Table 3). Normal G-6-PD enzyme activity was identified from AMM in the study population. The cut-off value was assigned at 30% and 80% according to the above mentioned. Individuals with G-6-PD enzyme activity less than 30% of AMM (< 2.4 IU/gHb) were identified as having G-6-PD deficiency. Considering female heterozygous having G-6-PD activity at 30% to 80% of AMM (2.4–6.5 IU/gHb) was defined as G-6-PD intermediate. Females with G-6-PD activity more than 80% of AMM (> 6.5 IU/gHb) were deemed normal. Sensitivity, specificity, accuracy, PPV and NPV analyzed by the careSTARTTM G6PD biosensor were 89%, 93%, 92%, 67% and 98% in males, respectively, 64%, 83%, 80%, 43% and 92% in females, respectively, and 78%, 89%, 87%, 56% and 96% in total, respectively (Table 3). The results were considered with a commonly used qualitative FST. As shown in Table 4, the overall efficiency of FST was 18% sensitivity, 100% specificity, 88% accuracy, 100% PPV and 88% NPV. These findings suggested that the careSTARTTM G6PD biosensor possessed higher sensitivity than FST.

Table 3

Diagnostic performance of the careSTARTTM G6PD biosensor classified by gender.

careSTARTTM G6PD biosensor	Standard quantitative G-6-PD test (SGT)
	Deficiency(<3.2 IU/gHb)	Intermediate(3.3–8.8 IU/gHb)	Normal(>8.8 IU/gHb)	Total	Performance	% (95% CI)
Male					Sensitivity	88.9 (65.3–98.6)
Deficiencya	16	*NA	8	24	Specificity	92.9 (86.4–96.9)
Normalc	2	*NA	104	106	Accuracy	92.3 (86.3–96.3)
Total	18	*NA	112	130	PPV	66.7 (50.1–79.9)
Prevalence of G-6-PD deficiency (%)				13.9	NPV	98.1 (93.4–99.5)
Female
Deficiencya	1	1	0	2	Sensitivity	64.3 (35.1–87.2)
Intermediateb	0	7	12	19	Specificity	83.3 (72.7–91.1)
Normalc	0	5	60	65	Accuracy	80.2 (70.3–88.0)
Total	1	13	72	86	PPV	42.9 (28.2–58.9)
Prevalence of G-6-PD deficiency (%)				16.3	NPV	92.3 (85.5–96.1)
Total
Deficiencya	17	1	8	26	Sensitivity	78.1 (60.0–90.7)
Intermediateb	0	7	12	19	Specificity	89.1 (83.7–93.2)
Normalc	2	5	164	171	Accuracy	87.5 (82.3–91.6)
Total	19	13	184	216	PPV	55.6 (44.3–66.3)
Prevalence of G-6-PD deficiency (%)				14.8	NPV	95.9 (92.4–97.8)

*Not available (NA)

The threshold of G6PD biosensor for identifying subjects as deficient (a) is less than < 2.4 IU/gHb, intermediate (b) is 2.4 to 6.5 IU/gHb and normal (c) is more than 6.5 IU/gHb.

Table 4

Diagnostic performance of the FST classified by gender.

careSTARTTM G6PD biosensor	Standard quantitative G-6-PD test (SGT)
	Deficiency(<3.2 IU/gHb)	Intermediate(3.3–8.8 IU/gHb)	Normal(>8.8 IU/gHb)	Total	Performance	% (95% CI)
Male					Sensitivity	27.8 (9.7–53)
Deficiency	5	*NA	0	5	Specificity	100 (96.8–100.0)
Normal	13	*NA	112	125	Accuracy	90.0 (83.5–94.6)
Total	18	*NA	112	130	PPV	100
Prevalence of G-6-PD deficiency (%)				13.9	NPV	89.6 (86.6–92.0)
Female
Deficiency	1	0	0	1	Sensitivity	7.1 (0.2–33.9)
Intermediate	*NA	*NA	*NA	*NA	Specificity	100 (95.0–100.0)
Normal	0	13	72	85	Accuracy	84.9 (75.5–91.7)
Total	1	13	72	86	PPV	100
Prevalence of G-6-PD deficiency (%)				16.3	NPV	84.7 (82.7–86.5)
Total
Deficiency	6	0	0	6	Sensitivity	18.8 (7.2–36.4)
Intermediate	*NA	*NA	*NA	*NA	Specificity	100 (98.0–100.0)
Normal	13	13	184	210	Accuracy	88.0 (82.9–92.0)
Total	19	13	184	216	PPV	100
Prevalence of G-6-PD deficiency (%)				14.8	NPV	87.6 (85.7–89.3)

*Not available (NA)

Discussion

The detection of G-6-PD deficiency has been promoted to identify newborns in many countries [5, 6, 10, 25]. Early diagnosis is the most effective management strategy not only for correcting the problem of jaundice in newborns, but also making patients more aware of their deficiency. Boonpeng et al. reported that newborns in Thailand with more than 4.4 mg/dL of microbilirubin had a higher risk of G-6-PD deficiency [26]. This finding created a significant impact on the health policy and clinical management of newborns for improving the quality of life and protecting the activities associated with life expectancy. Most G-6-PD deficient individuals should avoid the oxidative stress triggered by agents such as antimalarial drugs (primaquine, dapsone or tafenoquine), infection, or the ingestion of fava beans [1, 4, 7]. The FST assay is an easy, rapid and cost-effective qualitative method recommended by the International Committee for Standardization in Hematology (ICSH) for screening G-6-PD deficiency [5, 19]. However, the quantitative G-6-PD enzymatic activity assay (SGT) measured by spectrophotometric and automated UV enzymatic methods remains the reference method [5, 16, 25, 27]. The threshold of G-6-PD activity has also been used as criteria for malarial drug execution [7, 8, 11–13]. The WHO recommended that patients having less than 30% G-6-PD enzymatic activity and 70% normal activity should be excluded from primaquine and tafenoquine treatment [14, 22]. The SGT method requires a source of electricity, refrigeration for reagents and substrates, information on patient hemoglobin levels, a spectrophotometer to measure the absorbance and personal expertise to perform the test. These limitations restrict the SGT for routine testing and field application, especially in endemic countries with malaria infection [18]. The careSTARTTM G6PD biosensor has been created due to the current trend of developing point-of-care diagnostics. It meets not only all of the point-of-care testing (POCT) requirements, but is also an interesting quantitative tool to identify G-6-PD deficiency in newborns. The careSTARTTM G6PD biosensor used in the present research is a new model (version 2017), which has improved reliability for results by adding the reference electrode. The performance and factors influencing G-6-PD activity measurement were elucidated in this investigation. Basic information about G-6-PD activity in the study population was analyzed by gender (Table 1). There were no significant differences in G-6-PD activity between males and females. Moreover, the mean and AMM of G-6-PD enzyme activity in newborns were significantly higher than adults (p < 0.05), corresponding with previous study [27]. The factors affecting G-6-PD activity measurement were examined in 40 adults involving the type of blood collection, Hct concentration, storage time and temperature. The results found that the venous blood preserved in EDTA anticoagulant did not interfere with G-6-PD enzyme activity measurement compared to finger blood (Fig 1A). This finding confirmed the report by Roca-Feltrer et al.[28], which solved the question of von Fricken ME’s studies [25]. Accordingly, the hematocrit in newborns has a higher concentration than adults [29, 30], corresponding to this study (42±3.7% in adults and 49±7.7% in newborns). This means that newborns should have higher G-6-PD activity than adults, which is consistent with the results in Fig 1B. The results suggest that the reference value of newborns should be differentiated from adults [31]. G-6-PD activity drops gradually as time progresses [6]. Because the acceptance decreasing rate of G-6-PD activity should be no more than 10% [6, 32], G-6-PD activity was reduced over acceptance value at room temperature for 50 hours (Fig 2A) and remained stable at 4°C for 3 days (Fig 2B). The regression analysis found that blood samples should be stored at 4°C for 3 days (Fig 2D). Similar to a previous report, the G-6-PD activity did not fall by 10% until after 9 days of storage [6]. The performance evaluation of the careSTARTTM G6PD biosensor was analyzed using 216 neonates. The results demonstrated a strong positive correlation between the careSTARTTM G6PD biosensor and SGT (Fig 3). Thus, it can be described that both methods directly determine the kinetics of G-6-PD enzymatic activity reactions [18]. Bland-Altman plot analysis showed the absolute activity values of the same samples obtained from SGT, which were slightly higher than the careSTARTTM G6PD biosensor (+2.9 IU/gHb). Contrary to the automated UV enzymatic assay, G-6-PD activity was significantly higher than that of the SGT method [16]. This results from the different detectors; the careSTARTTM G6PD biosensor measures the electrons of Fe3+ from the change of NADP+ to NADPH reaction upon G-6-PD enzyme, meaning the electrons may be lost during operation. In this study, we set the cut-off point for G-6-PD deficiency at 30%, 70% and 80% of AMM based on previous suggestions [14, 20, 22]. The current thresholds used as exclusion criteria for primaquine treatment (<30% activity) and tafenoquine (<70% activity) [5, 28]. The diagnostic ability of the developed careSTARTTM G6PD biosensor represented an ideal method compared to a reference SGT assay at all thresholds, especially at 30% cut-off (Fig 4 and Table 2). With ROC analysis, the results suggested the optimal cut-off value to provide appropriate diagnostic performance (Table 2). The indication is that individuals with G-6-PD enzyme activity less than 2.4 IU/gHb can be identified as having G-6-PD deficiency, while those with G-6-PD enzyme activity between 2.5 to 6.4 IU/gHb can be identified as having intermediate G-6-PD deficiency. As presented in this study, the main advantage for clinical use in quantitative point-of-care diagnostics utilizing the careSTARTTM G6PD biosensor is to provide a more accessible way of defining G-6-PD activity at bedside and identifying individuals with intermediate G-6-PD deficiency, especially in terms of differentiating heterozygous females. The study also illustrates the sensitivity, specificity, PPV, NPV and accuracy of the careSTARTTM G6PD biosensor and FST compared to standard SGT (Table 3 and Table 4). The diagnostic efficacy of the careSTARTTM G6PD biosensor was found to be slightly higher than reported in the past [6]. Such variance in the results may be caused by the small-sized group and different population (216 neonates in the present study versus 900 adults in the previous study). Table 4 illustrates the low sensitive achievement by FST assay at all thresholds. This result confirmed that all current qualitative tests perform poorly [6, 17, 25, 28]. However, it has been recommended by ICSH as the appropriate qualitative method for screening G-6-PD deficiency in the field [5, 19]. The careSTARTTM G6PD biosensor showed a high potential for identifying G-6-PD deficiency in newborns and partial G-6-PD deficiency in females, imposing a risk for primaquine the same as tafenoquine administration. Moreover, quantitative G-6-PD activity has more alternative, attractive methods than molecular analysis because some cases of intermediate or G-6-PD deficiency could not be identified in the mutation [1, 3, 33, 34]. More than 400 different mutations have been found in individuals with G-6-PD deficiency [1, 3, 33, 34]. Some cases contain the mutations in cis-acting regulatory sequences or in the non-coding region of the G-6-PD gene, which may interfere with its expression [16]. However, the results by the careSTARTTM G6PD biosensor should be normalized with Hb concentration, especially in hemolytic anemia patients, because false positive results may occur.

Conclusion

The careSTARTTM G6PD biosensor has a high diagnostic ability to quantify G-6-PD enzymatic activity. It meets the needs for point-of-care testing and an appropriate setting for health promotion. Particularly, the careSTARTTM G6PD biosensor is suitable for early identification of intermediate and G-6-PD deficiency in newborns, which is the most effective strategy for the prevention of severe hemolytic anemia and the initiation of corrective treatment in a timely manner.

The CBC results of 216 neonates.

Results of RBC count, Hb and Hct in males were significantly higher than females.

(DOCX)

Click here for additional data file.

Multiple comparisons of G-6-PD activity among hematocrit (Hct) concentration in 40 adults.

The G-6-PD enzyme activity at certain Hct was subtracted with another getting the different of G-6-PD activity. The mean difference of G-6-PD activity measures the absolute difference between the mean values in two groups.

(DOCX)

Click here for additional data file.

Multiple comparisons of G-6-PD enzyme storage stability at room temperature and 4°C for 72 hours.

The mean difference measures the absolute difference between the mean value in two groups.

(DOCX)

Click here for additional data file.

2 Sep 2019

PONE-D-19-22851

Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection

PLOS ONE

Dear Dr Chomean,

Thank you for submitting your manuscript to PLoS ONE. After careful consideration, we felt that your manuscript requires revision, following which it can possibly be reconsidered. As quoted by the reviewers, a number of methodological concerns should be clarified otherwise it my compromise the manuscript. The results needs to be revised, and the MS should be revised by a native English speaker or professional language editing service. For your guidance, a copy of the reviewers' comments was included below.

We would appreciate receiving your revised manuscript by September 30. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Luzia Helena Carvalho, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal). We noted your study included minors, please state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

3. Please remove your figures from within your manuscript file, leaving only the individual TIFF/EPS image files, uploaded separately.  These will be automatically included in the reviewers’ PDF.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The article from Pengboon and colleagues presents a very important topic with implications on the health and clinical management of newborns in Thailand and elsewhere. The manuscript in its current form presents a number of issues that needs to be addressed.

1. English language needs an accurate revision throughout the manuscript.

2. The abstract does not seem to be coherent with the results and should be revised.

3. Methods: in the abstract and introduction only newborns are mentioned but then in methods the authors say they have analyzed adult samples and newborn samples. The rationale is not very clear in this section and I wonder whether this laboratory part needs to be included in the manuscript (see more comments below).

More details need to be provided about the Biosensor used (since CareStart has produced different models over the years and even made software updates within the same Biosensor model). Probably the dates of when the study was carried out would be helpful. Calculation of G6PD activity in Iu/gHb should be used always and authors need to specify here how they calculated for the Biosensor.

More details need to be provided about the G6PD reference test: how was it performed, using which reagents, how many replicates per sample, how many and how frequently control samples were used, which temperature was used, etc.

4. The WHO classification cannot be used to classify individual enzymatic activities. The classification was developed in the 80’s to categorize G6PD mutations based on the residual enzymatic activities found in hemizygous males. In order to classify subjects as G6PD deficient (<30% normal activity), normal (>80% normal activity) or intermediate (30-80% normal activity), an adjusted male population median needs to be calculated according to Domingo et al 2013. The authors need to revise and delete the use of WHO classification throughout the manuscript.

5. Results: for the analysis of factors influencing G6PD activity, samples should have been analyzed by the gold standard reference test while they were only analyzed with the Biosensor under validation making the results difficult to assess and possibly unreliable. Also, I don’t think 6-months storage stability ca be extrapolated from a 72hour experiment so this part should be cut.

The BA plot shows that the 95%CI of the difference between the 2 assay is +6.6 and -0.8 with the mean difference in the 2 assay is 2.9U/gHb, a rather large difference for a test that has a range of 0-15U; in fact the 95%CI span for 7.4U, corresponding to almost 50% of the entire activity range. I am therefore surprised by the results of the AUC for ROC analysis as it seems to show much better performances of Biosensor than what showed by the BA plot. The tables and text show results for CareStart in terms of U/dL but the whole manuscript should present the quantitative G6PD data in a comparable way, ie.e IU/gHb. The authors need to revise Table 1 and 2 and especially explain Table 2 because it is not clear what the table is showing.

The direct comparison of FST with Biosensor is not appropriate because the FST is a qualitative test, Nonetheless, it makes sense to analyze the performances of both tests, but I would keep the results separate.

Reviewer #2: There are no page numbers or page lines making the review of this article difficult.

The language needs significant editing.

Will just refer to sections and sentences:

1. Results: Assessment of factors affecting for G-6-PD enzyme activity analysis. In the paragraph starting “Due to the newborn was classified…” this whole paragraph requires language editing. Presumably the first sentence refers to the fact that variable were study in adults since more blood could be collected from them?

2. “The results suggested that the G-6-PD enzyme activity can be stable at 4°C for 5 months.” Should be removed. This claim should be based on realtime data. There is no precedent for this level of stability.

3. Results: Performance of the careSTARTTM G6PD biosensor for G-6-PD enzyme activity detection “For efficacy evaluation…”.. The authors should explain what this means.

4. In the same section “ According to the adjusted male median, the receiver operator curve (ROC)…” the authors should clarify whether U/dL were used or U/g Hb values were used.

5. Figure 4. The plots are Carestart in U/g Hb, so these have been normalized.

6. Discussion: ..” WHO recommended that the patients that have G-6-PD enzymatic activity less than 30% and 70% of normal activity should be excluded from primaquine and tafenoquine treatment, respectively [5, 14, 16].” This statement is simply false, and the references have nothing to do with this statement.

7. Discussion: …”It proved the doubt messages of von Fricken ME’s studies [22] and consistent

with the Roca-Feltrer A. and coworker performing by CareStart G6PD RDT [14]” the authors should clarify what this means.

8. Discussion: ..” that blood samples should be stored at 4°C for 6 months.” Again this claim should not be made only that it is stable for 72 hours.

9. Discussion:..” The current thresholds used as exclusion criteria for primaquine treatment (<30% activity) and tafenoquine (<70% activity) [5, 14]” This statement is correct but references are incorrect: the authors should refer to WHO malaria guidelines and Llanos et al article on tafenoquine.

10. Discussion: The authors should highlight that a limitation of the product is that, as their data suggests the Carestart product does not correct for hematocrit, and this may be a source of error in a larger study population.

11. Discussion: “…showed a high diagnostic potency..” a high potential?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

4 Oct 2019

Response to Referees’ comments and detail changed.

Reviewer #1

1.

Referees’ comments

The article from Pengboon and colleagues presents a very important topic with implications on the health and clinical management of newborns in Thailand and elsewhere. The manuscript in its current form presents a number of issues that needs to be addressed.

Revised statements

I would appreciate proper recommending of clinical management of newborns in Thailand. The mentioned paper has been quoted in the discussion part. “Boonpeng P. and colleague reported that newborns with microbilirubin more than 4.4 mg/dL having a higher risk of G-6-PD deficiency in Thailand [26]. This finding has valuable to define the policy on health and clinical management of newborns for improving the quality of life and protecting the activity affect life expectancy.”

Location

Manuscript in Discussion part (page 15 line 5 to 8 and reference # 26)

2.

Referees’ comments

English language needs an accurate revision throughout the manuscript.

Revised statements

We sincerely apologize for the grammar and typos. We have revised the manuscript.

Location

Throughout Manuscript

3.

Referees’ comments

The abstract does not seem to be coherent with the results and should be revised.

Revised statements

We have revised the abstract and mentioned below.

Location

Manuscript: Abstract part

4.

Referees’ comments

Methods: in the abstract and introduction only newborns are mentioned but then in methods the authors say they have analyzed adult samples and newborn samples. The rationale is not very clear in this section and I wonder whether this laboratory part needs to be included in the manuscript (see more comments below).

More details need to be provided about the Biosensor used (since CareStart has produced different models over the years and even made software updates within the same Biosensor model). Probably the dates of when the study was carried out would be helpful. Calculation of G6PD activity in Iu/gHb should be used always and authors need to specify here how they calculated for the Biosensor.

More details need to be provided about the G6PD reference test: how was it performed, using which reagents, how many replicates per sample, how many and how frequently control samples were used, which temperature was used, etc.

Revised statements

Methods: in the abstract and introduction only newborns are mentioned but then in methods the authors say they have analyzed adult samples and newborn samples.

Author’s responses

The present study aimed to evaluate the G-6-PD activity in newborns. However, the blood samples collected from newborns should be as little as possible. Therefore, the factors affecting the G-6-PD activity was determined in adult blood.

Anyway, the abstract was edited and also added the methods as well as results of adults and newborns.

Location

Manuscript in Abstract part (page 2 line 7 to 10)

Manuscript in Materials and Methods (page 4 line 8 to 10)

Revised statements

More details need to be provided about the Biosensor used (since CareStart has produced different models over the years and even made software updates within the same Biosensor model).

Author’s responses

The careSTARTTM G6PD biosensor used in this study is a new model (2017) which modified the reference electrode to improve the reliability of the test. This detail was appended in discussion part. As the text “A careSTARTTM G6PD biosensor used in the present study is a new model (version 2017) which improved the reliability of results by adding the reference electrode.”

Location

Manuscript in Discussion part (page 15 line 24 to 26)

Revised statements

Calculation of G6PD activity in Iu/gHb should be used always and authors need to specify here how they calculated for the Biosensor.

Author’s responses

We accepted the reviewer’s comment. The unit of G6PD activity was changed in IU/gHb throughout the manuscript. The activity derived from biosensor measured the G-6-PD enzyme from RBC in one deciliter which should be divided by Hb (IU/gHb).

Location

Throughout Manuscript

Manuscript in Materials and Methods (page 5 line 29 to 30)

Fig. 1, 2 and 3 were changed.

Revised statements

More details need to be provided about the G6PD reference test: how was it performed, using which reagents, how many replicates per sample, how many and how frequently control samples were used, which temperature was used, etc.

Author’s responses

The method of a reference assay (SGT) and FST were explained in materials and methods.

For SGT assay, All experiments were performed by using the SGT assay according to the manufacturer’s protocol. Briefly, five microliters of blood was mixed with the Elution Buffer (75 µL) in U-bottom microtiter plate for 10 min at room temperature. The eluted (15 µL) was transfer to 75 µL of the Reagent Mixture containing in a new flat-bottom microtiter plate and mix thoroughly. Then, adding 100 µL of Color Reagent Mixture to each well and read the G-6-PD activity reaction (kinetic mode) with ELISA reader (800TS Microplate Reader, BioTek, USA) at 550 nm for 15 min with 1 min intervals. After the final reading is taken, the plate was read at wavelength 405 nm to get the hemoglobin (Hb) content of each sample. The G-6-PD enzyme activity is calculated by comparing the rates of blood sample to rate of normal control with known G-6-PD activity. The reported G-6-PD activity has been normalized to Hb and expressed results directly into international units per gram of hemoglobin (IU/gHb). Quality control of deficient blood (2.0 IU/g Hb with specification of 1.0-3.0 IU/gHb) and normal blood (14.6 IU/gHb with specification of 9.5-19.7IU/gHb) was operated throughout the experiments.”

For FST assay, “Five microliter of EDTA blood was added to 100 µL of G-6-PD substrate reagent, mixed thoroughly and incubated at 25°C for 30 min. After that the mixture (100 µL) was applied to filter paper and observed under UV light. Results accomplished by careSTARTTM G6PD biosensor and FST were compared with standard quantitative G-6-PD assay (SGT) as a reference assay.”

Location

Manuscript in Materials and Methods (page 5 line 10 to 22 and page 6 line 2 to 6)

5.

Referees’ comments

The WHO classification cannot be used to classify individual enzymatic activities. The classification was developed in the 80’s to categorize G6PD mutations based on the residual enzymatic activities found in hemizygous males. In order to classify subjects as G6PD deficient (<30% normal activity), normal (>80% normal activity) or intermediate (30-80% normal activity), an adjusted male population median needs to be calculated according to Domingo et al 2013. The authors need to revise and delete the use of WHO classification throughout the manuscript.

Revised statements

We accepted the reviewer’s comments. So, the results were re-analyzed using the cut-off at 30% and 80% of residual G-6-PD enzyme activity. Moreover, the threshold at 70% was performed according to the suggestion of tafenoquine treatment. In addition, the adjusted male population median was calculated according to Domingo et al . (2013).

Location

Manuscript: Materials and Methods (statistical analysis) part. (page 6 line 11 to 16 and reference # 20 #21 #22)

6.

Referees’ comments

Results: for the analysis of factors influencing G6PD activity, samples should have been analyzed by the gold standard reference test while they were only analyzed with the Biosensor under validation making the results difficult to assess and possibly unreliable. Also, I don’t think 6-months storage stability ca be extrapolated from a 72hour experiment so this part should be cut.

Revised statements

We accepted the reviewer’s comments and demonstrated the results from a standard SGT as a reference method in this study. Moreover, the stability of G-6-PD activity was concluded at 4°C for 3 days.

Location

Manuscript: Materials and Methods part (page 4 line 24, page 5 line 10 to 22), Results part (page 9 line 12 to 13) and Discussion part (page 16 line 13 to 14)

7.

Referees’ comments

The BA plot shows that the 95%CI of the difference between the 2 assay is +6.6 and -0.8 with the mean difference in the 2 assay is 2.9U/gHb, a rather large difference for a test that has a range of 0-15U; in fact the 95%CI span for 7.4U, corresponding to almost 50% of the entire activity range. I am therefore surprised by the results of the AUC for ROC analysis as it seems to show much better performances of Biosensor than what showed by the BA plot. The tables and text show results for CareStart in terms of U/dL but the whole manuscript should present the quantitative G6PD data in a comparable way, ie.e IU/gHb. The authors need to revise Table 1 and 2 and especially explain Table 2 because it is not clear what the table is showing.

Revised statements

In order to evaluate whether the careSTARTTM G6PD biosensor and standard G-6-PD enzymatic test (SGT) can be used for classifying deficient, intermediate, and normal activity between the two assays, the Bland–Altman plot was used (Fig. 3). The measurements made by the SGT exceeded those obtained by careSTARTTM G6PD biosensor by an average of 2.9 IU/g Hb, indicates that value from careSTARTTM G6PD biosensor tends to be lower than those obtained by SGT. However, a significant correlation between the points on the Bland-Altman plot (Pearson’s correlation analysis, r=0.824; P<0.05), which provides evidence of a linear association. In conclusion, the Bland Altman plot provides only an agreement between the two methods.

To clarify “The BA plot shows that the 95%CI of the difference between the 2 assay is +6.6 and -0.8 with the mean difference in the 2 assay is 2.9U/gHb, a rather large difference for a test that has a range of 0-15U; in fact the 95%CI span for 7.4U, corresponding to almost 50% of the entire activity range” from reviewer’s comment. We would like to demonstrate some data from previous studies, Anantasomboon P. and coworkers [16] shows the 95%CI of the difference between the 2 assay is +3.21 and -7.28 with the mean difference in the 2 assay is -2.04U/gHb (spectrophotometric vs Mindray G6PD assay), the 95%CI span for 10.49 U/gHb. Conversely, the study from field evaluation [6] showed the mean differences: 0.83 U/gHb, 95%CI:-3.10-1.44 U/gHb (span for 4.54 U/gHb)

Due to the variation of the test, the 95%CI of the difference between the 2 assays may affect the sensitivity and specificity of the test. However, ROC analysis of careSTARTTM G6PD biosensor (biosensor) and fluorescent spot test (FST) with standard G-6-PD enzymatic test (SGT) indicate that AUC of careSTARTTM G6PD biosensor is greater than those fluorescent spot test (FST) Finally, and the optimal cut- off value can be determined using ROC curve analysis for careSTARTTM G6PD biosensor.

However, the ROC results were revised which analyzed between the biosensor and standard SGT (Fig. 4). Based on AUC analysis, the results also represented the optimal value for identifying the G-6-PD status at a certain cut-off (Table 2).

Location

Manuscript: Results part (page 10 line 7 to 19; Fig 4 and Table 2)

8.

Referees’ comments

The direct comparison of FST with Biosensor is not appropriate because the FST is a qualitative test, Nonetheless, it makes sense to analyze the performances of both tests, but I would keep the results separate.

Revised statements

We agreed with reviewer’s comments and the results were separately illustrated.

Location

Manuscript: Results part (page 12 line 1 to 13; Table 3 and Table 4)

Referee #2

1.

Referees’ comments

Results: Assessment of factors affecting for G-6-PD enzyme activity analysis. In the paragraph starting “Due to the newborn was classified…” this whole paragraph requires language editing. Presumably the first sentence refers to the fact that variable were study in adults since more blood could be collected from them?

Revised statements

We sincerely apologize for the grammar and typos.

The sentences were edited to “By the reason of blood drawing from a newborn can be difficult, potentially harmful and small volume, the factors affected to the G-6-PD enzyme activity measurement were evaluated in 40 adults” in Materials and Methods.

Location

Manuscript: Materials and Methods part (page 4 line 8 to 10)

2.

Referees’ comments

“The results suggested that the G-6-PD enzyme activity can be stable at 4°C for 5 months.” Should be removed. This claim should be based on realtime data. There is no precedent for this level of stability.

Revised statements

We accepted the reviewer’s comments and demonstrated the stability of G-6-PD activity for 3 days. As the sentence “The results suggested that the G-6-PD enzyme activity can be stable at 4°C for 3 days.”

Location

Manuscript: Results part (page 9 line 12 to 13) and Discussion part (page 16 line 13 to 14)

3.

Referees’ comments

Results: Performance of the careSTARTTM G6PD biosensor for G-6-PD enzyme activity detection “For efficacy evaluation…”.. The authors should explain what this means.

Revised statements

The efficacy evaluation involves assessing the performance of careSTARTTM G6PD biosensor comparable with standard SGT method. It included the correlation analysis, sensitivity, specificity, positive likehood ratio (+LR), negative likehood ratio (-LR), positive predictive value (PPV), negative predictive value (NPV) and disease prevalence.

Location

Manuscript: Results part (page 9 line 22 to 25)

4.

Referees’ comments

In the same section “ According to the adjusted male median, the receiver operator curve (ROC)…” the authors should clarify whether U/dL were used or U/g Hb values were used.

Revised statements

We re-analyzed and the results were demonstrated only the unit of IU/gHb.

Location

Throughout Manuscript

5.

Referees’ comments

Figure 4. The plots are Carestart in U/g Hb, so these have been normalized.

Revised statements

We re-analyzed and the results were demonstrated only the unit of IU/gHb.

Location

Figure 2.

6.

Referees’ comments

Discussion: ..” WHO recommended that the patients that have G-6-PD enzymatic activity less than 30% and 70% of normal activity should be excluded from primaquine and tafenoquine treatment, respectively [5, 14, 16].” This statement is simply false, and the references have nothing to do with this statement.

Revised statements

We sincerely apologize and changed the references to [14] WHO. Testing for G6PD deficiency for safe use of primaquine in radical cure of P. vivax and P. ovale (Policy brief). Geneva: World Health Organization. 2016:28.and [22] WHO. Guide to G6PD deficiency rapid diagnostic testing to support P. vivax radical cure. Geneva: World Health Organization. 2018:34.

Location

Manuscript: Discussion part. (page 15 line 15 to 18 and reference #14 and #22)

7.

Referees’ comments

Discussion: …”It proved the doubt messages of von Fricken ME’s studies [22] and consistent

with the Roca-Feltrer A. and coworker performing by CareStart G6PD RDT [14]” the authors should clarify what this means.

Revised statements

The results were correlated to the Roca-Feltrer A study[28] which can answer the question from von Fricken ME’s report [25]. This statement was rewritten to “This finding was consistent with the Roca-Feltrer A [(28)] and coworker report which solved the question of von Fricken ME’s studies (25).”

Location

Manuscript: Discussion part (page 16 line 4 to 5)

8.

Referees’ comments

Discussion: ..” that blood samples should be stored at 4°C for 6 months.” Again this claim should not be made only that it is stable for 72 hours.

Revised statements

We agree with this statement and change the G-6-PD storage time for 3 days. As “The regression analysis found that that blood samples should be stored at 4°C for 3 days (Fig 2D).”

Location

Manuscript: Discussion part (page 16 line 13 to 14)

9.

Referees’ comments

Discussion:..” The current thresholds used as exclusion criteria for primaquine treatment (<30% activity) and tafenoquine (<70% activity) [5, 14]” This statement is correct but references are incorrect: the authors should refer to WHO malaria guidelines and Llanos et al article on tafenoquine.

Revised statements

My apologies for the mistakes. The references were changed in [14] WHO. Testing for G6PD deficiency for safe use of primaquine in radical cure of P. vivax and P. ovale (Policy brief). Geneva: World Health Organization. 2016:28.and [22] WHO. Guide to G6PD deficiency rapid diagnostic testing to support P. vivax radical cure. Geneva: World Health Organization. 2018:34..

Location

Manuscript: References part [14] and [22]

10.

Referees’ comments

Discussion: The authors should highlight that a limitation of the product is that, as their data suggests the Carestart product does not correct for hematocrit, and this may be a source of error in a larger study population.

Revised statements

The limitation of a point-of-care, careSTARTTM G6PD biosensor was added in the discussion as the sentence “However, the results by careSTARTTM G6PD biosensor should be normalized with Hb concentration especially in hemolytic anemia patients because the false positive result may occur.”

Location

Manuscript: Discussion part (page 17 line 21 to 23)

11.

Referees’ comments

Discussion: “…showed a high diagnostic potency..” a high potential?

Revised statements

We sincerely apologize for the grammar and typos. We have revised this sentence as “The careSTARTTM G6PD biosensor showed a high potential for identifying G-6-PD deficiency in newborns and partially G-6-PD deficiency in female and imposing a risk for primaquine as same as tafenoquine administration.”

Location

Manuscript: Discussion part (page 17 line 13 to 15)

Editor’s comments

1) Thank you for submitting your revised manuscript to PLOS ONE. Could you please make changes to the ethics statement that you have provided as follows:

- Please include the statement about consent in the Methods section of the manuscript (and not just on the online details form).

- In the Methods, please clarify whether the adults in the study provided written informed consent.

Revised statements

Prior to enrolment all volunteers provided written informed consent to participate and publication of results. This includes 20 males and 20 females with the age range of 18 to 24 years. After obtaining written informed consent, the blood was collected both finger-prick (5 µL) and venous puncture (3 mL in K2EDTA vacutainer tubes) and analyzed by standard quantitative G-6-PD enzymatic assay (R&D Diagnostics, Ltd., Greece).

Location

Manuscript: Materials and Methods part (page 4 line 10 to 15)

- In the Methods, please clarify who waived informed consent from the parents/guardians of the neonates who provided blood samples, was informed consent waived by the IRB, or did the authors make this decision?

Revised statements

To evaluate the performance of careSTARTTM G6PD biosensor (WELLS BIO, INC., Korea), the 216 neonatal blood samples remaining from a routine laboratory testing were collected and kept in microtainer tube with K2EDTA anticoagulant. The neonatal blood samples used in this study were leftover samples following routine analysis and all data were de-identified before access. For these reasons, the ethics committee provided an waiver of informed consent.

Location

Manuscript: Materials and Methods part (page 4 line 16 to 20)

23 Oct 2019

PONE-D-19-22851R1

Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection

PLOS ONE

Dear Dr Chomean,

Thank you for resubmitting your manuscript to PLoS ONE. After careful consideration, we felt that your study has the potential to be published if it is revised to address fundamental point raised now by the reviewer. As quoted by the reviewers, the authors should clarify some specific topics related to methods and tables. At this time, we strongly suggest that the manuscript should be revised by a native English-speaker or a professional language editing service.

We would appreciate receiving your revised manuscript by  November 10. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Luzia Helena Carvalho, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: No

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript has improved considerably from its first version. The authors have addressed most of my previous requests and comments. I still have some questions for the authors and they should clarify a number of issues.

The G6PD reference test: the authors report the protocol but they do not mention whether the test was performed at controlled temperature and whether samples were analyzed in duplicate. The authors should also report the manufacturer name.

The FST is not performed according to manufacturers’ instructions. Why the authors chose to incubate the blood and reagent mix for 30 minutes? I expect this to have a pretty big impact on the results of the test, in particular toward misclassification of deficient/intermediate samples as normal.

Table 2: I do not understand what this table is showing. The se and sp at the 3 different threshold is better using the values of the second column as compared to the “optimal cut-off value” of the 10th column. This needs to be explained in further details

The English language still needs a lot of work. There are many parts of the manuscript that are difficult to read; in some parts the actual meaning of what the authors wrote is not very clear. It is important that the authors improve the language throughout the manuscript to make it understandable.

Reviewer #2: The authors have been very diligent about responding to the reviewers comments and have actually presented the data in a clear manner.

The English in the manuscript still requires correcting. I have just focused on the abstract, but the rest of the manuscript probably needs further review:

1. line 12"...enzyme activity inspected that the activity determined from finger-prick.." replace "inspected" with "showed"

2. Line 13: " However G6PD activity has significantly increased in higher hematocrit concentration". maybe replace with "The G6PD activity value was highly dependent on the hematocrit, and increased with increasing hematocrit"

3. line 16: "...strongly correlated with Pearson’s " replace with "..a strong correlation with a Peasron's..."

4. line 16 replace "perfectly" with "perfect"

5.lines 17-19 define what G6PD range this was accurate for: G6Pd deficient?

The rest of the manuscript should also go further review for English language. Although the effort put by the authors is greatly appreciated.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

4 Nov 2019

(R2): Response to Referees’ comments and detail changed.

Reviewer #1

1.

Referees’ comments

The G6PD reference test: the authors report the protocol but they do not mention whether the test was performed at controlled temperature and whether samples were analyzed in duplicate. The authors should also report the manufacturer name.

Author’s responses

Thank you for reviewer’s suggestion. I have added more information about your comments. All clinical blood samples were studied by whether FST, careSTARTTM G6PD biosensor or SGT assay at single time. Whereas, the control blood samples including deficient and normal blood was determined by FST, careSTARTTM G6PD biosensor and SGT assay in duplicate. Moreover, the manufacturer name was also added according to R&D Diagnostics, Ltd., Greece.

Revised statements

All blood samples were carried out standard operating procedures under controlled time and temperature followed the manufacturer's instruction. All experiments were conducted the quality control of deficient blood (2.0 IU/g Hb with the specification of 1.0-3.0 IU/gHb) and normal blood (14.6 IU/gHb with the specification of 9.5-19.7IU/gHb in duplicate.

Location

Manuscript in Materials and Methods part (page 4 line 26 to 27 and page 5 line 26 to 28)

2.

Referees’ comments

The FST is not performed according to manufacturers’ instructions. Why the authors chose to incubate the blood and reagent mix for 30 minutes?

Author’s responses

I sincerely apologize for having this mistake. I confirmed the FST protocol that the EDTA blood and G-6-PD substrate reagent was incubated at 25°C for 10 minutes.

Revised statements

Five microliters of EDTA blood was added to 100 µL of G-6-PD substrate reagent, mixed thoroughly and incubated at 25°C for 10 min.

Location

Manuscript in Materials and Methods part (page 6 line 9)

3.

Referees’ comments

Table 2: I do not understand what this table is showing. The se and sp at the 3 different thresholds is better using the values of the second column as compared to the “optimal cut-off value” of the 10th column. This needs to be explained in further details

Author’s responses

The purpose in Table 2 needs to represent the sensitivity, specificity, PPV, NPV and accuracy at different cut-off value. The cut-off value and optimal cut-off value of G-6-PD activity was calculated by a certain residual G-6-PD activity and statistical SPSS analysis, respectively.

The selection of which cut-off value depends on the purpose of the G-6-PD detection. If the careSTARTTM G6PD biosensor users require to confirm G-6-PD deficiency, they should be chosen the cut-off value at 30% (<2.4 U/gHb) due to the performance of biosensor at this point gave a highest specificity (96%) compared to another. However, the performance of “optimal cut-off value” was lower than that of the “cut-off value” as reviewer’s comment; I deleted the content in “optimal cut-off value”.

Location

Manuscript: Results part and Table 2 (page 11)

4.

Referees’ comments

The English language still needs a lot of work. There are many parts of the manuscript that are difficult to read; in some parts the actual meaning of what the authors wrote is not very clear. It is important that the authors improve the language throughout the manuscript to make it understandable.

Author’s responses

I would like to apologize for making you uncomfortable. I take full responsibility to improve this manuscript. The English was proofread and edited by native English teacher.

Location

Throughout Manuscript

Referee #2

1.

Referees’ comments

The English in the manuscript still requires correcting.

Author’s responses

I'm very thankful for your suggestion. Also, I would like to apologize for making you uncomfortable. I take full responsibility to improve this manuscript. The English was proofread and edited by native English teacher.

Location

Throughout Manuscript

19 Nov 2019

PONE-D-19-22851R2

Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection

PLOS ONE

Dear Dr  Chomean,

Thank you for resubmitting your manuscript to PLoS ONE. After careful consideration, we felt that your manuscript still requires substantial revision, following which it can possibly be reconsidered. While the subject of the MS was of interest of the reviewers, relevant topics remain to be addressed.  More specifically, the authors should clarify about discrepant results between table 2 and 3 and include in the methods the information about single replicate to the reference G6PD test. As quoted by the reviewer, the manuscript language has only marginally improved At this time, we strongly recommend that the authors include the modifications requested by the reviewer, and the MS should be revised by a native English-speaker  or a professional language editing service.

We would appreciate receiving your revised manuscript by December 10. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Luzia Helena Carvalho, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: 1. After the modifications in the latest version there are still some important issues to address:

- I do not see how the results on sensitivity, specificity, PPV and NPV presented in Table 2 (and abstract) and Table 3 are compatible. The 2 tables give different results and it is unclear which ones are correct.

- In the results it is mentioned that newborn age ranged from 1 to 35 days, this should really be clarified.

- The fact that the reference G6PD test was run only in single replicate needs to be made explicit in the methods, not only in the response to the reviewers questions.

2. The English language in the manuscript has only marginally improved. There are still many sentences that are difficult to understand.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Nov 2019

(R3): Response to Referees’ comments and detail changed.

Reviewer #1

1.

Referees’ comments

I do not see how the results on sensitivity, specificity, PPV and NPV presented in Table 2 (and abstract) and Table 3 are compatible. The 2 tables give different results and it is unclear which ones are correct.

Author’s responses

The results in table 2 and 3 presented in different purpose.

The results depicted in Table 2 were analyzed by the receiver operating characteristic (ROC) curve which performed by using the 30%, 70% and 80% of G-6-PD residual activity. The results obtained from the area under the Receiver Operating Curve (AUC) suggested the optimal cut-off value indicating the proper sensitivity, specificity, positive likelihood ratio (+LR), negative likelihood ratio (-LR), positive predictive value (PPV) and negative predictive value (NPV). Also, it was analyzed by G-6-PD residual activity with no consideration about gender.

For table 3 and 4, the results illustrated the diagnostic potency of the careSTARTTM G6PD biosensor and FST by using 2x2 table. All data were indicated the G-6-PD deficiently status according to the description of Domingo GJ et al. [20]. So, we enter the number of cases in the diseased group that test positive (a) and negative (b); and the number of cases in the non-diseased group that test positive (c) and negative (d) [23].

For example, Sensitivity= a / (a+b)

Specificity= d / (c+d)

Positive likelihood ratio= True positive rate / False positive rate = Sensitivity / (1-Specificity)

Negative likelihood ratio= False negative rate / True negative rate = (1-Sensitivity) / Specificity

Revised statements

Abstract: Analysis of the area under the Receiver Operating Curve (AUC) illustrated that the careSTARTTM G6PD biosensor had 100% sensitivity, 96% specificity, 73% positive predictive value (PPV), 100% negative predictive value (NPV) and 97% accuracy at 30% of residual activity. While the diagnostic ability for identifying G-6-PD deficiency had 78% sensitivity, 89% specificity, 56% positive predictive value (PPV), 96% negative predictive value (NPV) and 88% accuracy when stratified by gender.

Table 3: Diagnostic performance of the careSTARTTM G6PD biosensor classified by gender

Table 4. Diagnostic performance of the FST classified by gender

Location

Manuscript in Abstract part (page 2 line 17 to 22) and results part (page 13 line 1 and page 14 line 1)

2.

Referees’ comments

In the results it is mentioned that newborn age ranged from 1 to 35 days, this should really be clarified.

Author’s responses

We need to explain the descriptive data analysis in newborns.

Revised statements

Materials and Methods: To evaluate the performance of the careSTARTTM G6PD biosensor (WELLS BIO, INC., Korea), 216 neonatal blood samples ranging in age between 1 to 35 years old were collected and kept in a microtainer tube with K2EDTA anticoagulant.

Results: In newborns with an average age of 4 (4 ± 4 day), complete blood count (CBC) and G-6-PD enzyme activity were investigated.

Location

Manuscript in Materials and Methods parts (page 4 line 19) and Results part (page 7 line 20 and 22)

3.

Referees’ comments

The fact that the reference G6PD test was run only in single replicate needs to be made explicit in the methods, not only in the response to the reviewers questions.

Author’s responses

We explained this comment in the Materials and Methods parts as following.

Revised statements

All blood samples were carried out a single time following standard operating procedures under controlled time and temperature.

Location

Manuscript in Materials and Methods part (page 4 line 27 and 28)

4.

Referees’ comments

The English language in the manuscript has only marginally improved. There are still many sentences that are difficult to understand.

Author’s responses

The English language entire the manuscript was proofread and edited by native English speaker.

Location

Throughout the manuscript.

4 Dec 2019

PONE-D-19-22851R3

Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection

PLOS ONE

Dear Dr Chomean,

Thank you for submitting your manuscript for review to PLoS ONE. After careful consideration, we feel that your manuscript will likely be suitable for publication if it is revised to address a few point raised by the reviewer. Basically, the authors should adjust Table to include the threshold used for the Biosensor.

We would appreciate receiving your revised manuscript by December 10. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Luzia Helena Carvalho, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I would like to thank the authors for addressing my requests. I wanted to specify further my question about Table 3; Since both the Biosensor and the reference test are quantitative tests (with continuous data results), in order to identify subjects as "deficient" or "normal" one needs to establish a threshold under which a sample is considered deficient and over which a sample is considered normal. So I would invite the authors to just specify somewhere in the Table or legend, what is the threshold they have used for the reference test and for the Biosensor.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

5 Dec 2019

(R4): Response to Referees’ comments and detail changed.

Reviewer #1

1.

Referees’ comments

I wanted to specify further my question about Table 3; Since both the Biosensor and the reference test are quantitative tests (with continuous data results), in order to identify subjects as "deficient" or "normal" one needs to establish a threshold under which a sample is considered deficient and over which a sample is considered normal. So I would invite the authors to just specify somewhere in the Table or legend, what is the threshold they have used for the reference test and for the Biosensor.

Author’s responses

Thank you for reviewer’s comments. To clarify the data, we specify the threshold of G6PD biosensor with the legend (a, b and c). Whereas the reference value by SGT assay was mentioned in the Table 3 and 4.

Revised statements

For SGT assay, we addressed as “Deficiency” (<3.2 IU/gHb), “Intermediate” (3.3–8.8 IU/gHb) and “Normal” (>8.8 IU/gHb) in Table 3 and 4.

For G6PD biosensor, we addressed as “The threshold of G6PD biosensor for identifying subjects as deficient (a) is less than < 2.4 IU/gHb, intermediate (b) is 2.4 to 6.5 IU/gHb and normal (c) is more than 6.5 IU/gHb.”

Location

Manuscript: Results part Table (page 13-15).

10 Dec 2019

Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection

PONE-D-19-22851R4

Dear Dr. Chomean,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Luzia Helena Carvalho, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

12 Dec 2019

PONE-D-19-22851R4

Evaluation of quantitative biosensor for glucose-6-phosphate dehydrogenase activity detection

Dear Dr. Chomean:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Luzia Helena Carvalho

Academic Editor

PLOS ONE