Reference spectrophotometric values for glucose-6-phosphate dehydrogenase activity in two-to six-month-old infants on the Thailand-Myanmar border

Background Glucose-6-phosphate dehydrogenase (G6PD) deficiency represents a barrier to the full deployment of anti-malarial drugs for vivax malaria elimination and of first-line antibiotics. Lack of established reference ranges for G6PD activity in breast-fed infants puts them at risk of drug-induced haemolysis and restricts access to safe treatment of their mothers. Methods The present work was undertaken to establish age-specific G6PD normal values using the gold standard spectrophotometric assay to support the future clinical use of tafenoquine in lactating women and safer antibiotic treatment in infants. Results Spectrophotometric results collected at the Thai-Myanmar border from 78 healthy infants between the ages of 2 and 6 months showed a trend of decreased enzymatic activity with increasing age (which did not reach statistical significance when comparing 2–3 months old against 4–6 months old infants) and provided a reference normal value of 100% activity for infants 2–6 months old of 10.18IU/gHb. Conclusions Normal reference G6PD activity in 2–6-month-old infants was approximately 140% of that observed in G6PD normal adults from the same population. Age specific G6PD activity thresholds should be used in paediatric populations to avoid drug-induced haemolysis.

from 78 healthy infants between the ages of 2 and 6 months showed a trend of decreased enzymatic activity with increasing age (which did not reach statistical significance when comparing 2-3 months old against 4-6 months old infants) and provided a reference normal value of 100% activity for infants 2-6 months old of 10.18IU/gHb.

Introduction
Malaria is a leading cause of morbidity and mortality in many developing countries with an estimated 300 to 500 million clinical cases worldwide every year.Tafenoquine is an 8-aminoquinoline antimalarial drug developed by GlaxoSmithKline (GSK) in collaboration with the Medicines for Malaria Venture and 60 Degrees Pharmaceuticals (60°P; (Chu & Freedman, 2019)), that provides radical cure of Plasmodium vivax (P.vivax) by eliminating dormant parasites in the liver and thereby preventing relapses (Lacerda et al., 2019;Llanos-Cuentas et al., 2019).
The 8-aminoquinoline class of drugs, including primaquine and tafenoquine, is haemolytic in subjects with glucose-6-phosphate dehydrogenase (G6PD) deficiency.G6PD deficiency is the most prevalent human enzymopathy, affecting nearly 400 million people (8% allelic frequency worldwide, (Howes et al., 2012)) and reaching prevalence over 20% in populations living in malaria endemic countries.Biochemical and genetic studies have identified over 200 genetic variants (Gomez-Manzo et al., 2016).The consequences of G6PD deficiency in humans are most apparent in red blood cells.Affected individuals are largely asymptomatic and otherwise healthy.However, a precipitating oxidative stress may cause haemolytic anaemia (Cappellini & Fiorelli, 2008).Common causes of this haemolytic reaction are infections, certain medicines (e.g.anti-malarial compounds such as 8-aminoquinolines and antibiotics), chemicals (e.g.naphthalene in newborns) or foods (e.g.fava beans).G6PD deficiency is also associated with increased risk of moderate and severe neonatal hyperbilirubinaemia by a mechanism not completely understood and only partly explained by haemolysis (Kaplan & Hammerman, 2002;Luzzatto et al., 2020).
The G6PD gene is located on the X-chromosome.Males with G6PD deficiency are hemizygotes and will only carry deficient red blood cells.Adult male hemizygotes and female homozygotes, with <30% of normal G6PD activity, can usually be detected by the inexpensive fluorescent spot test (FST, (Beutler & Mitchell, 1968)) or G6PD RDTs (Ley et al., 2019).However heterozygous females have both a normal and deficient population of red cells (genetic mosaics) resulting in a range of G6PD activity.Females with G6PD activity in the range of 30-70% of normal activity, while susceptible to drug-induced haemolysis (Chu et al., 2018), are generally miss-classified as normal by FST.Because of increased enzymatic activity at birth (Doherty et al., 2014), the FST also performs poorly in newborns, missing even fully deficient neonates (Thielemans et al., 2018).Performance of FST in infant blood is unknown but might be subject to the same shortcomings, especially during the first 6 months of life.
Normal G6PD activity reference values are established by gold standard spectrophotometric assay (Beutler et al., 1977) and the 30% and 70% thresholds are derived from the median activity obtained in normal males (Domingo et al., 2013).While standard 14-days primaquine regimen for radical cure of vivax malaria can safely be given to most adults with normal FST results (e.g.≥ 30% enzymatic activity), safe administration of shorter high-dose primaquine treatment and tafenoquine requires identification of patients with at least 70% of normal G6PD enzymatic activity (Rueangweerayut et al., 2017).Treatment with other haemolytic drugs, including first-line antibiotics, are generally considered safe in FST-normal subjects but data are inconsistent (Bancone & Chu, 2021).
There is a paucity of data on G6PD activity levels during the first year of life.Early literature indicates that G6PD activity declines from higher neonatal levels to "adult" levels during this period (Gross & Hurwitz, 1958;Stave & Pohl, 1960).Travis and colleagues described the changes in G6PD activities in 10 infants tested at 6 time points during the first year of life (Travis et al., 1980) showing median activities in the first week of life equivalent to 160% of adult values, decreasing to roughly 140% of adult values at 3 months and reaching near-adult values after 6 months of life.A study published in 2020 (Yang et al., 2020) showed a similar trend in decreasing enzymatic activity during the first 3 months of life in 410 infants aged 7 to 90 days.This data gap has implications for treatment, in particular for malaria, in both infants and their mothers.Lactating women are excluded from radical cure of P. vivax because of the theoretical risk of iatrogenic haemolysis in the breastfed infant (WHO, 2022).Lack of accurate diagnostics for G6PD deficiency in infants and lack of pharmacokinetic data in breast milk prevents appropriate treatment for women during the post-partum period.G6PD deficiency is highly prevalent in the Karen and Burman populations attending SMRU clinics along the Thailand-Myanmar border (Bancone et al., 2014;Bancone et al., 2017a), with an estimated prevalence of 9-18% in males; the main mutation found in this population is Mahidol (487G>A) which is associated to a residual enzymatic activity of <30% of normal in hemizygous mutated males.SMRU has recently demonstrated negligible exposure to primaquine in breastfed infants of mothers receiving primaquine (Gilder et al., 2018); these findings support radical curative treatment of P. vivax in breastfeeding women to enhance malaria elimination.Up-coming studies involving treatment of lactating women with tafenoquine will require the assessment of paediatric G6PD normal reference values in order to include breastfed infants <1 year of age.Furthermore, reference values in infants will be helpful to analyse haemolytic effects of other drugs commonly used in this portion of the population.

Amendments from Version 1
We have revised the abstract to include more information on where the study was carried out; we have included more information in the method section about the study setting and approach, a reference for anaemia definition and the significance level for statistical analyses.Results and figures have not been modified.The discussion has been slightly modified to include suggestions from the reviewers.

Any further responses from the reviewers can be found at the end of the article
The current work was undertaken to establish age-specific G6PD normal values in infants using the gold standard spectrophotometric assay to support clinical use of tafenoquine in lactating women and safer antibiotic treatment in infants.

Study design
This was a laboratory study to establish G6PD normal values in infants aged two to six months designed to collect a single blood sample in 80 consecutive healthy male infants attending the Shoklo Malaria Research Unit (SMRU) clinic.The study was conducted SMRU's Maw Ker Thai clinic situated along the Thailand-Myanmar border in Tak province (Thailand) starting in October 2020.In SMRU clinics free care is provided for migrant populations, especially pregnant women and patients with infectious diseases.The clinics do not have a pediatric program but provide vaccination through a standard Expanded Programme of Immunization (EPI).

Participants
Mothers of male infants were approached to enter the study by SMRU clinical staff at EPI visits scheduled for 2, 4 and 6 months or study follow-up visits.Staff explained to the mothers of infants the rationale of the study and the blood sampling procedures in their local language, and the mothers were asked to sign a consent form to participate in the research.
Infants were included if they were in the appropriate age range (2-3 months and 4-6 months old) and were healthy (no severe illness, no fever, no diarrhoea, no vomiting).Hospitalization in the previous 3 months, antibiotic use in the last 2 weeks and prematurity (EGA< 37 weeks' gestation at birth) were further exclusion criteria.

Ethical approval and consent
The study was approved by Oxford Tropical Research Ethics Committee, UK (OxTREC 543-19), the Mahidol University Faculty of Tropical Medicine Ethics Committee, Thailand (TMEC 19-047, MUTM 2019-079-01) and the Tak Province Border Community Ethics Advisory Board (TCAB201911).Written informed consent was obtained from literate mothers; a thumbprint was obtained in the presence of a literate witness for illiterate mothers.

Data collection
A single 100µl (2-3 drops of blood) capillary blood sample was collected by heel-prick into an EDTA-treated Safe-T-fill ® capillary tube.Samples were stored at 4°C at the clinics until transported refrigerated (within 8 hours) to the central haematology lab of SMRU.At the central haematology laboratory, blood was analysed by gold standard spectrophotometric assay and complete blood count (CBC) or Hemocue 301 + system within 24 hours from collection.The Pointe Scientific spectrophotometric assay was used (assay kit # G7583-180, Lysis Buffer # G7583-LysSB).Kinetic determination of G6PD activity at 340 nm was performed using a SHIMAZU UV-1800 spectrophotometer with temperature-controlled cuvette compartment (30°C).Samples were analysed in duplicate and mean activity was expressed in IU/gHb using the Hb concentration obtained by CBC or Hemocue 301+.The final result was calculated using manufacturer's Temperature Control Factor of 1.37.Two controls (Normal, Intermediate or Deficient; Analytic Control Systems, Inc. USA) were analysed at every run and results compared to expected ranges provided by manufacturer.
CBC was performed using a CeltacF MEK-8222K haematology analyser (Nihon Kohden, Japan).Three-level quality controls were run every day and device maintenance and calibration were performed regularly.Hemocue 301+ was run according to manufacturer's instructions.

Sample size and statistical analyses
The proposed sample size was calculated from the following formula to estimate the mean of a population with a 95% CI of [mean± δ]: Where z=1.96 and σ is the expected standard deviation.
We used a σ=2 and δ=0.75, so our minimum sample was 28.To account for exclusions from analysis of G6PD deficient subjects (around 15-18% of the male population) and a 10% invalid results, a total of 40 subjects was planned to ensure a total of 30 evaluable subjects.
To maximize inclusion of healthy infants attending the clinic at EPI visits, two age groups were evaluated, one group included infants aged from 2 months + 0 days to 3 months + 14 days (herein called "2-3 months old group") and a second group included infants aged from 4 months + 0 days to 6 months + 14 days (i.e."4-6 month old groups").
Anaemia was defined as haemoglobin concentration <10.0 g/dL following WHO recommendations.
Male median equivalent to reference 100% activity was calculated as the median of activity in G6PD normal males.Medians were compared by Mood's test and Mann-Whitney test.Statistical significance was assessed at the 5% level.
Comparison of data from the current study with data on male infants in published literature was performed on studies that used different laboratory methodologies to assess G6PD enzymatic activity (Gross & Hurwitz, 1958;Konrad et al., 1972;Stave & Pohl, 1960;Tang et al., 1992;Travis et al., 1980;Yang et al., 2020).In order to compare results, percentage enzymatic activity was obtained by normalising newborns and infant data from each study to their reported normal value in adults.

Results
Rate of enrolment was heavily influenced by the COVID-19 ongoing pandemic and the travel restrictions imposed during the years 2020 and 2021.From 8 October 2020 to 24 June 2021 and from 16 September to 27 January 2022, a total of 92 male infants (49 in the 2-3 months old group and 43 in the 4-6 months old group) were enrolled in the study; blood samples from 9 infants in the 2-3 months old group and 5 infants in the 4-6 months old group were rejected because they were clotted.Analyses were performed on specimens from 78 infants (40 in 2-3 months old group and 38 in 4-6 months old group, Figure 1).Median (min-max) age was 65.0 (60-101) days in the 2-3 months old group and 130  days in the 4-5 months old group.The full dataset can be found under Underlying data (Bancone et al., 2022a, "GNB study database").

Haematological characteristics
White blood cells and platelets counts in infants of different age were similar; red blood cell (RBC) count was significantly higher in older infants and mean corpuscular volume (MCV) and MCH (mean corpuscular haemoglobin) significantly lower in older infants (Table 1) reflecting different distribution of RBC ages in the two groups.

G6PD activity
The distribution of G6PD activity showed a clear bimodal shape with 10.3% (8/78) infants in the left peak (G6PD deficient) with an activity less than 1.3IU/gHb (Figure 2).
The established male median from the whole sample allowed definition of the 30% and 70% activity threshold for infants of both sex 2-6 months of age corresponding to 3.05IU/gHb and 7.13 IU/gHb.G6PD activity is normalized using haemoglobin concentrations therefore results in anaemic subjects can be falsely elevated (Bancone et al., 2017b) as plotted in Figure 3.     4.
Data from the current study were also compared to published literature that reported reference G6PD enzymatic activity     3).
assessed in term newborns, infants up to one year of life and adults (where available) within the same study (Figure 5).

Discussion
This was the first study to assess quantitative G6PD activity in infants outside the newborn period in this population, and represented, to our knowledge, the largest published dataset of G6PD activity in 4-6-month-old infants.When compared to median G6PD activity at birth assessed in a larger cohort in the same population (13.3IU/gHb) and reference male median in adults (7.5IU/gHb), the male median observed in 2-3-month-old infants (10.5IU/gHb) and in 4-6-month-old infants (9.7IU/gHb) confirmed a trend of gradual decline of enzymatic activity during infancy.Enzymatic activity at birth was over 170% of the adult levels and declined to 140% of adult levels by 6 months of age.This trajectory was comparable with previously collected data and consistent with a higher proportion of younger RBCs in newborns and/or shorter lifespan in erythrocytes from younger infants (Kuruvilla et al., 2017).
Importantly, when observing increased levels of G6PD activity, thresholds for definition of "G6PD deficiency" increased proportionally; should adult thresholds be applied to infants, deficient and intermediate infants would be inappropriately classified as normal.Misclassification of deficient individuals as normal could put them at risk for drug-induced hemolysis if the G6PD normal misdiagnosis follows them through childhood and adulthood.It is unknown whether higher enzymatic activity observed in infancy would provide some protection from oxidative haemolysis in this phase of life.
Though much progress has occurred in the past decade towards P. falciparum elimination in the populations seeking care at SMRU's clinics, P. vivax has resisted elimination efforts due to the persistence of hypnozoites and operational barriers to broad implementation of radical cure.Following over 2 years of political unrest, Myanmar and the bordering countries have already seen increasing clinical cases of P. vivax, (World malaria report 2023 (who.int)) as similarly observed during armed conflicts in Africa (Sedda et al., 2015).Radical cure for vivax malaria using single-dose tafenoquine has an obvious advantage in treatment adherence compared to 14-days but also 7-day primaquine courses; this becomes even more relevant for mobile populations living in areas with decreased security and poor health systems.A sizeable proportion of the population which lacks access to radical vivax treatment is represented by women of reproductive age (Brummaier et al., 2020), especially when they experience multiple pregnancies in close succession; a safe radical cure should be provided during lactation (Watson et al., 2018) and the result of this study, together with previously established reference G6PD values in adults, will support clinical trials to assess safety and efficacy of using tafenoquine in breastfeeding mothers.
This work established reference normal values for G6PD activity in infants of 2 to 6 months of age, after which G6PD activity was expected to approach adult levels.However, median G6PD activity in 6-month-old infants from this cohort appears to still be well above the adult median (10.2IU/gHb corresponding to 140% of adult levels), suggesting that more data in late infancy (up to 12 months of life) and anaemic infants would be valuable.

Conclusions
G6PD reference ranges in infants 2 to 6 months of age were established locally to support expansion of radical curative treatments of vivax malaria in lactating mothers, starting with clinical trials to quantify the amount of drug present in breast milk.Reference ranges will also be useful to improve treatment with potentially haemolytic drugs (e.g.antibiotics) in pediatric patients.More work will need to be done to establish reference ranges for up to 1 year of life and to assess use of quantitative point-of-care G6PD tests in infants.

Vinod K Bhutani
Stanford University School of Medicine, Stanford, CA, USA General Comments Maternal-child-health perspectives of living safely with inherited Glucose-6-Phosphate Dehydrogenase (G6PD) deficiency remains a poorly investigated area in both medical and public health research disciplines.Bancone et al provide a valuable contribution to address the lack of established reference ranges for G6PD activity in breast-fed infants who are at risk of drug-induced and other environmental induced hemolysis.This condition not only has direct adverse impact on babies but also restricts access to safe intervention options of their mothers.
The authors could have highlighted a wider impact of their data such that future research could be directed at: a) newborn babies who are immediately initiated on breast feeding; b) on the safety among mothers who most likely have varying degree of enzyme deficiency because of the complex inheritance of this X-linked genetic condition; c) collection of potential adverse outcome data among newborns and infants who are inadvertently exposed to unsafe anti-malarial drugs; and, d) importantly, expand the study age range of the cohort to study both neonatal and infancy phases and trajectory of G6PD enzyme activity concurrent with the natural decline in hemoglobin concentrations.
Thus, this body of work is not just limited to the public health/infection disease community caring for malarial prophylaxis only.The authors are commended to help define the assessment of pediatric G6PD normal reference values in order to include breastfed infants <1 year of age.By this token, the report should extend from age 0 to 364 days.Since only male infants were studied, exclusion by gender is problematic.At this stage, limiting the cohort to term and latepreterm neonates is appropriate.However, future opportunities to study safety of breastfeeding premature infants should be sought.I do understand the malaria perspective but the MCH perspective in an endemic area is of much wider.significance This is a good start.I agree with most of the comments by Collins Ouma.

Specific Comments: I suggest additional considerations inclusive of clarifications:
Reference by age: The present work has been undertaken to establish age-specific-G6PD enzyme (specify by days, weeks or months) normal values using the gold standard spectrophotometric assay to support the future clinical applications to identify the enzyme deficiency status in lactating women and their infants.Do you have data for the mothers? 1.
Deficiency threshold: Spectrophotometric results from 78 healthy infants between the ages of 2 and 6 months showed a trend of decreased enzymatic activity with increasing age (in days, weeks or months).These data provide a "reference" normal (mean, median?) value of 100% activity for infants 2-6 months (as a single cohort) old of 10.18 IU/gHb (is this the mean or median, add SD, range?).Is there a statistical basis to offer a specific 'deficiency threshold?Could the authors suggest a statistical out-of-range value to guide a "deficiency" threshold?How is gender of the cohort handled?

2.
Timeliness: Can the authors highlight the total turn-around time (collection of blood to clinical action upon results) for enzyme assay results since a timed interval is critical to support the least interruption of lactational care?

3.
Gender considerations: Normal reference G6PD activity in 2-6-month-old male infants was approximately 140% of that observed in G6PD normal adults (just males?What about nonpregnant females, pregnant mothers, or lactating mothers from the same population?).Age specific G6PD activity thresholds should be used in both male and female pediatric populations so that the goal would be to achieve healthy outcomes and prevent environmental, diet and chemical induced hemolysis in the entire population.

Study Cohort:
The rationale for sub-dividing the cohort is unclear and needs explanation to 5. merit a statistical comparison (Table 2).Perhaps, the data could be presented of the two cohorts: deficient and "normal" [as per by the bi-modal differentiation {Figure 2}] in relation to post-natal age (in days) in format of mean/median and confidence interval or range.
Confounding factors: Since the authors have focused on a specific target population, is the emphasis related to ethnicity, ancestral heritage or related endemicity, geography or socioeconomic factors related to literacy as well integrity of self-care health practices?6.

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility?Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Paediatrics and Neonatology
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
significance.We would also like to direct the reviewer to a manuscript published by our group in which we evaluate a point-of-care G6PD test for use in neonates which is likely to improve neonatal clinical care for jaundice in LMIC (Technical evaluation and usability of a quantitative G6PD POC test in cord blood: a mixed-methods study in a low-resource setting -PubMed (nih.gov)) which addresses in part some of the issues raised by the reviewer."By this token, the report should extend from age 0 to 364 days."This would be very interesting and would provide useful data but, unfortunately, we do not have data available up to 12 months of life.As reported to the other reviewer, a) our main focus, stated throughout the manuscript, was to obtain reference ranges of G6PD activity in the specific population of infants up to 6 months of age; b) we did not know before obtaining the data that the G6PD activity in 6-months old infants had not reached levels comparable to adults; c) the clinical setting in which we work does not have a pediatric clinic and our capacity to include healthy children older than 6 months is very limited.We have clarified this last point in the methods section."Since only male infants were studied, exclusion by gender is problematic" This is indeed a very common approach when assessing normal reference values of G6PD enzymatic activity; due to the natural bimodal distribution in G6PD activitiy in a population of random males, it is easier to select G6PD normal males without genotyping and to establish normal reference value.This allows to keep a small sample size and obtain data that can be then used for the whole population (males and females).Please see G6PD testing in support of treatment and elimination of malaria: recommendations for evaluation of G6PD tests -PubMed (nih.gov)"At this stage, limiting the cohort to term and late-preterm neonates is appropriate.However, future opportunities to study safety of breastfeeding premature infants should be sought.I do understand the malaria perspective but the MCH perspective in an endemic area is of much wider significance This is a good start."We fully agree with the reviewer and hope to be able in future to investigate more in this direction.

Reference by age:
The present work has been undertaken to establish age-specific-G6PD enzyme (specify by days, weeks or months) normal values using the gold standard spectrophotometric assay to support the future clinical applications to identify the enzyme deficiency status in lactating women and their infants.Do you have data for the mothers?We already have reference G6PD values established in the adult population and will use them for safe treatment of lactating mothers.This study was undertaken to establish normal values in infants only.Deficiency threshold: Spectrophotometric results from 78 healthy infants between the ages of 2 and 6 months showed a trend of decreased enzymatic activity with increasing age (in days, weeks or months).These data provide a "reference" normal (mean, median?) value of 100% activity for infants 2-6 months (as a single cohort) old of 10.18 IU/gHb (is this the mean or median, add SD, range?).Is there a statistical basis to offer a specific 'deficiency threshold?Could the authors suggest a statistical out-of-range value to guide a "deficiency" threshold?How is gender of the cohort handled?The 10.18IU/gHb is the median value among normal males and represent the 100% normal reference value G6PD testing in support of treatment and elimination of malaria: recommendations for evaluation of G6PD tests -PubMed (nih.gov).The deficiency threshold of 30% reported in the manuscript follow WHO recommendation for adults of both sex Testing for G6PD deficiency for safe use of primaquine in radical cure of P. vivax and P. ovale (who.int) and what we also used for neonates Technical evaluation and usability of a quantitative G6PD POC test in cord blood: a mixed-methods study in a low-resource setting -PubMed (nih.gov); the 70% threshold is used for safe treatment using Tafenoquine ( Hemolytic Potential of Tafenoquine in Female Volunteers Heterozygous for Glucose-6-Phosphate Dehydrogenase (G6PD) Deficiency ( G6PD Mahidol Variant) versus G6PD-Normal Volunteers -PubMed (nih.gov)Timeliness: Can the authors highlight the total turn-around time (collection of blood to clinical action upon results) for enzyme assay results since a timed interval is critical to support the least interruption of lactational care?In the planned clinical studies, we will be testing G6PD by in-house spectrophotometry with results available within 24 hours.Primaquine and tafenoquine treatment can be delayed for 1-2 days without adverse clinical effects so this is acceptable for the purpose of the clinical studies.However, we agree with the reviewer's implication that a rapid test (such as a G6PD biosensor) would be useful in future when treatment of lactating women might become common practice and quantitative testing for both mother and infant will be needed at the point-of-care.Evaluation of G6PD Biosensor tests in infants will be needed.Gender considerations: Normal reference G6PD activity in 2-6-month-old male infants was approximately 140% of that observed in G6PD normal adults (just males?What about nonpregnant females, pregnant mothers, or lactating mothers from the same population?).Age specific G6PD activity thresholds should be used in both male and female pediatric populations so that the goal would be to achieve healthy outcomes and prevent environmental, diet and chemical induced hemolysis in the entire population.Yes, the normal threshold is intended to be used by the whole pediatric population, males and females.We have clarified this.The comparison was done with the reference adult levels used for the whole adult population.

Study Cohort:
The rationale for sub-dividing the cohort is unclear and needs explanation to merit a statistical comparison (Table 2).Perhaps, the data could be presented of the two cohorts: deficient and "normal" [as per by the bi-modal differentiation {Figure 2}] in relation to post-natal age (in days) in format of mean/median and confidence interval or range.The rationale for subdividing the cohort by age came from an effort to maximize participation of healthy infants in the study while attending EPI (this has been clarified in the methods) and the lack of knowledge on what G6PD activity level we could expect at the different ages.We believe Table 2 is already presenting these data.Confounding factors: Since the authors have focused on a specific target population, is the emphasis related to ethnicity, ancestral heritage or related endemicity, geography or socioeconomic factors related to literacy as well integrity of self-care health practices?The target population represents a collection of communities with high prevalence of G6PD deficiency that has been impacted by malaria for generations.SMRU has been working for over 3 decades with this population, providing free health care and conducting research that addresses the clinical needs of the local population.In this regard, evaluation of G6PD tests and establishment of relevant normal reference has become an important effort to improve local health by providing safe radical cure for vivax malaria and by supporting neonatal care for jaundice.We confirm, however, that the data apply to this specific population and pediatric reference laboratory ranges might need to be established in different settings.

Collins Ouma
University of New Mexico-Kenyan Global Health Programs Laboratories, New Mexico, Kenya Bancone et al presents findings on a study undertaken to establish age-specific G6PD normal values using the gold standard spectrophotometric assay to support the future clinical use of tafenoquine in lactating women and safer antibiotic treatment in infants.The manuscript can benefit from the following improvements: The Abstract presents data that are not statistically analysed.Statistics that lead to the current conclusions would be critical on the abstract. 1.
As much as the study is carried out on children in Thailand, the study site is not mentioned on the abstract.It takes the reader to get back on the title and possibly read through Methods to figure out where the study was performed.

2.
Introduction: Authors state that there is paucity on data on G6PD activity levels during the first year of life.This in my understanding is within the first 12 months of life.The rationale for narrowing down on children aged between 2-6 months need to be provided.If not provided adequately, I would suggest that data be presented for children within the first 12 months.

3.
Recruitment: what happened to children who are HIV-positive but have no stated conditions for exclusions? 4.
The definition of anaemia provided as haemoglobin concentration <10.0g/dL is not referenced.Is this according to WHO definition or regional definition?

5.
Categorisation of enzymatic activities is stated to be arbitrary.I think this is not the first time such a study is being done.I would suggest a use of a more standard categorisation.

6.
No statistical analyses approaches provided in the methods, not even the level of assessing statistical significance.

7.
Results, Discussion and Conclusions: if my comments above are not addressed, the conclusions may sound spurious and as such need to be re-looked. 8.

Is the work clearly and accurately presented and does it cite the current literature? Partly
Is the study design appropriate and is the work technically sound?Yes Are sufficient details of methods and analysis provided to allow replication by others?
activity.Furthermore, HIV prevalence in this population is extremely low (< 1% Audit of antenatal screening for syphilis and HIV in migrant and refugee women on the Thai-Myanmar border: a descriptive study -PubMed (nih.gov)).
The definition of anaemia provided as haemoglobin concentration <10.0g/dL is not referenced.Is this according to WHO definition or regional definition?This is WHO definition; we have added the reference in the methods.

2.
Categorisation of enzymatic activities is stated to be arbitrary.I think this is not the first time such a study is being done.I would suggest a use of a more standard categorisation.
Categorization in this context had to be arbitrary because we presented data from neonates, infants and adults who have very different normal G6PD activities; a standard categorization used in adults would not be suitable for neonates and viceversa.It is worth noting that this categorization is only used for the purpose of graphical representation and it is not analysed statistically.

3.
No statistical analyses approaches provided in the methods, not even the level of assessing statistical significance.
Statistics for comparison of medians were already reported.We have added that the significance was assessed at 5% level.

4.
Results, Discussion and Conclusions: if my comments above are not addressed, the conclusions may sound spurious and as such need to be re-looked 5.
We hope we have addressed comments in a satisfactory way and that the conclusions are fully sustained when considering this additional information.

Figure 4 .
Figure 4. Distribution of G6PD activity (IU/gHb) categories according to age groups.Data in newborns were collected during a recent study in SMRU clinics involving 125 G6PD normal male neonates; data in adults were collected during a recent study in SMRU clinics involving 95 G6PD normal healthy males.

Figure 5 .
Figure 5. G6PD enzymatic activity at different ages: comparison of current data with published data.G6PD enzymatic activity was normalized by levels assessed at 1 year of age or in adults.The continuous pink line represents the median calculated from all published and current data (Table3).
G6PD point-of-care quantitative tests have been recently evaluated for use in newborns {Bancone, 2022 #1768} and their evaluation in infants would provide additional indication for use in non-adult patients.Use of the same G6PD point-of-care device across different settings(Ley et al., 2022) would bypass the need for locally established paediatric reference ranges for G6PD activity which presents an added barrier to diagnosis in children.

Competing Interests :
No competing interests were disclosed.Reviewer Report 18 August 2023 https://doi.org/10.21956/wellcomeopenres.20422.r63174© 2023 Ouma C.This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Table 2 . Median (min-max) enzymatic activity in G6PD deficient and normal infants in the two age groups.
G6PD activity in neonates, infants and adultsIn G6PD normal infants, G6PD activity classified in 5 categories (< 6.0 IU/gHb, 6.0-7.9IU/gHb,8.0-9.9IU/gHb,10.0-11.9IU/gHband ≥12IU/gHb) showed a trend of different distribution by age group as compared to samples previously collected in the same setting among newborns {Bancone, 2022 #1768} and healthy adults (unpublished data) as shown in Figure

Table 3 . G6PD enzymatic activity at different ages: comparison of current data with published data. Age (months) birth
*The same 10 infants were followed-up over time.NR= non reported