Hemoglobins F, A2, and E levels in Laotian children aged 6‐23 months with Hb E disorders: Effect of age, sex, and thalassemia types

Abstract Introduction Determination of hemoglobins (Hbs) F, A2, and E is crucial for diagnosis of thalassemia. This study determined the levels of Hbs F, A2, and E in children aged 6‐23 months and investigated the effect of age, sex, and types of thalassemia on the expression of these Hbs. Methods A total of 698 blood samples of Laotian children including 272 non‐Hb E, 271 Hb E heterozygotes, and 155 Hb E homozygotes were collected. Hb profiles were determined using the capillary zone electrophoresis. Coinheritance of α‐thalassemia and the homozygosity for Hb E mutation were checked by PCR‐based assay. Results Children heterozygous and homozygous for Hb E had significantly higher Hb F and A2 levels than non‐Hb E children (median Hb F = 1.1% for non‐Hb E group, 2.7% for Hb E heterozygotes, and 9.4% for Hb E homozygotes; median Hb A2 = 2.6% for non‐Hb E group, 3.8% for Hb E heterozygotes, and 5.2% for Hb E homozygotes). The median Hb E levels were 21.9% for Hb E heterozygotes and 85.3% for Hb E homozygotes. Comparing within group, there was a statistically significant difference between children with and without an α‐gene defect for Hb A2 and E, but not Hb F. Based on a multiple regression analysis, age and sex were significantly associated with the expression of Hb F and A2 but not Hb E. Conclusions Our findings can guide the development of a diagnostic approach to thalassemia in children aged 6‐23 months.


| INTRODUC TI ON
Hemoglobin F (Hb F; α 2 γ 2 ) is a major hemoglobin during the fetal period. In neonates, Hb F normally declines gradually and reaches normal levels of less than 1% of total hemoglobin at the age of around 10 months. 1 Meanwhile, adult hemoglobin A (Hb A; α 2 β 2 ) increases gradually in replacement of Hb F. Small amounts of minor adult hemoglobin, hemoglobin A 2 (Hb A 2 ; α 2 δ 2 ), can be detected after birth.
Its expression increases gradually and has been shown to reach a normal adult level at age 5-6 months. 2 Inherited disorders of hemoglobin could result in varying levels of Hb F and A 2 depending mainly on the types of the affected gene. In adults heterozygous for β-thalassemia (β-thal), the level of Hb A 2 is usually higher than 3.5% with varying levels of Hb F. A persistently high Hb F level is found in patients with β-thal as well as in those who carry high Hb F determinants. 3,4 Hemoglobin E (Hb E; α 2 E 2 ), a structural Hb variant commonly found in South East Asia, results from a single nucleotide base substitution (G to A) at codon 26 of the β-globin gene. The mutation activates a cryptic splice site in the mRNA and leads to a reduced expression of the affected β-globin gene. 5 Adult individuals heterozygous for Hb E usually have no clinical symptoms with Hb E ranging from 25% to 35% of the total hemoglobin. 6,7 In an individual with Hb E homozygote, the so-called Hb EE disease, mild microcytic anemia can be observed.
Without Hb A production, Hb E constitutes up to 80%-90% of total hemoglobin. 8 Evidence from previous studies in adults showed that the complex interactions of Hb E with α-and β-thal could alter Hb E, F, and A 2 levels, [9][10][11] and this could result in difficulty making an accurate diagnosis. In young children, little information is available. In addition to thalassemia types, it is questionable for children under 24 months whether age and sex have a significant effect on the expression of these hemoglobins.
We used data from the "Lao Zinc study" a community-based intervention trial among young Laotian children 6-23 months of age at enrollment to evaluate the effect of age, sex, and different types of thalassemias on the expression of Hb F, A 2, and E in children with and without Hb E.

| Subjects and samples
The Lao Zinc Study was a randomized placebo-controlled, double blind, community-based trial implemented from September 2015 to April 2017 in Khammouane province, Lao PDR. The primary objective of the Lao Zinc study was to determine the effects of two forms of daily preventive zinc supplementation vs therapeutic zinc supplementation for diarrhea on young children's physical growth and other health outcomes. 12 The study protocol and consenting procedure were approved by the National Ethics Committee for Health Research, Lao PDR; the Institutional Review Board of the University of California Davis, USA; and the Khon Kaen University, Thailand. The study procedures are described in detail elsewhere. 13 Briefly, children were considered eligible if they were 6-23 months of age, and their families accepted weekly home visits, planned to remain within the study area for the duration of the study and signed the informed consent document. Children were ineligible if they met one of the following criteria: severe anemia (Hb < 70 g/L), weight-for-length z-score <−3 standard deviation, 14  This add-on study was approved by the Ethics Committees of Khon Kaen University (HE592006). All children who had complete laboratory analyses for Hb typing (n = 698) were recruited.

| Laboratory methods for detection of thalassemia
All samples were investigated for Hb types and the corresponding fraction using the capillary zone electrophoresis (Capillarys II; Sebia, Lisses, France). Identification of α 0 -thal mutations (SEA and THAI deletions) was performed using a multiplex PCR as described previously. 15 Identification of deletional and nondeletional mutations causing α + -thal, that is, 3.7 kb and 4.2 kb deletions, Hb CS and Hb Ps, was carried out using a multiplex gap-PCR and allele specific PCR (ASPCR). 16,17 Additional analysis of β-thal genes was carried out in cases suspected for β-thal carrier, as indicated by Hb A 2 > 3.5%. 18,19 Homozygosity for Hb E was tested in cases with either Hb EE or Hb EF phenotype using a previously described ASPCR. 9

| Statistical analysis
Data were tested for normality using the Shapiro-Wilk test. Descriptive statistics, median, and interquartile range (IQR) were used to describe the levels of Hb F, A 2 , and E. Statistically significant differences among three or more independent groups were tested using the Kruskal-Wallis test. The difference between two independent groups was compared with the Mann-Whitney U test. Box plots were constructed to demonstrate the trend of change. To demonstrate the effect of age on Hb F, A 2 , and E levels, children were categorized according to age as 6-12 months >12-18 months, and >18-23 months. Age group, sex, and thalassemia type were entered into the multiple regression model as independent variables. All graphics were constructed using the MedCalc Statistical Software version 18.11.6 (MedCalc Software bvba, Ostend, Belgium; https ://www.medca lc.org; 2019). The STATA statistical software version 10.0 (StataCorp.) was used for multiple regression analysis. Statistical significance was set at P < .05.

| Hb F, A 2, and E levels in children age 6-23 months
The mean age of all children was 15.3 (SD = 5.2) months, and 368 (52.7%) were boys. The levels of each Hb in children aged 6-23 months varied greatly, ranging from 0% to 43.7% for Hb F, 0.5 to 8.1 for Hb A 2 , 12.4%-28.6% for heterozygous Hb E, and 52.9%-94.1% for homozygous Hb E (Figure 1). The median values of Hb F, A 2, and E levels in children aged 6-23 months, categorized by thalassemia types, are shown in Table 1. Hb F levels in children heterozygous and homozygous for Hb E were significantly higher than that of non-Hb E groups, that is, 2.7% (IQR = 1.5%-4.5%) and 9.4% (IQR = 6.0%-13.5%) vs 1.1% (IQR = 0.5%-2.2%) (P < .001). Comparing within the non-Hb E group, Hb F levels in non-thal children and children with α-thal genes did not differ significantly. Similarly, in children heterozygous and homozygous for Hb E, Hb F levels were not different between those without α-thal genes and those with either one or two α-gene defects ( Figure S1). Similar to Hb F, significantly higher Hb A 2 levels were observed in children heterozygous and homozygous for Hb E, that is, 3.8% (IQR = 3.6%-4.1%) and 5.2% (IQR = 4.5%-5.8%) vs 2.6% (IQR = 2.4%-2.8%) (P < .001). Among non-Hb E children, a significantly reduced Hb A 2 was seen in children with α-thal (P < .001). However, in children heterozygous and homozygous for Hb E, the concomitance of α-thal (either one or two α-gene defect) did not result in significant changes in Hb A 2 ( Figure S2). in proportion to the number of α-gene defects (P < .001). In contrary, children homozygous for Hb E coinherited with α-thal (either one or two α-gene defect) showed significantly higher Hb E levels, as compared with those without α-thal (P = .017) (Supplementary Figure S3). Subsequent analysis comparing the levels of Hb F and A 2 between deletion and nondeletion α-thal was done within a group of one α-gene defect. Children with nondeletion showed significantly lower Hb A 2 level, as compared with those with deletion (median value = 2.4% vs 2.6%; P = .025). For Hb F level, no significant difference was observed ( Figure S4).

| Effect of age, sex, and thalassemia types on the expression of Hb F, A 2 , and E
The results of multiple regression analysis are shown in Tables 2-4. Compared with children aged 6-12 months, children aged >12-18 months and >18-23 months had a significant decline in Hb F expression (coefficient = −2.44 and −3.17). Females had 0.8% higher Hb F than males (P = .006). Similar to the finding in section 1 above, multiple regression analysis indicated that all forms of Hb E resulted in an increased Hb F levels with a coefficient range from 1.7 to 8.5 (Table 2).
There was a trend toward an increased Hb A 2 expression with advancing age in which a statistical significance was attained in children age >18-23 months (coefficient = 0.102 for age >12-18 months and 0.24 for age >18-23 months). Expression of Hb A 2 was significantly lower in female than in male (coefficient = −0.12). Similar to that observed in the Hb F analysis, all forms of Hb E resulted in increased Hb A 2 levels with a coefficient range from 1.13 to 2.69 (Table 3).  c Including heterozygous states for α 0 -thal, homozygous states for α + -thal, Hb CS, and Hb Ps, and compound heterozygous states for α + -thal/Hb CS.
TA B L E 1 Hb F, A 2, and E levels in children aged 6-23 mo, categorized by thalassemia types [data presented as median (IQR)] In Hb E heterozygotes, age and sex had no significant effect on Hb E levels (Table 4). However, a significantly increased Hb E with increasing age was observed in homozygous states (coefficient = 5.71 for age > 12−18 months and 7.11 for age >18−23 months). Based on a multiple regression analysis, the concomitance of α-thal resulted in a reduction in Hb E of 1.32% for one α-gene defect and 6.96% for two α-gene defect, as compared with those without α-thal gene (P < .0001).
In contrast to Hb E heterozygotes, the coinheritance of α-thal among Hb E homozygotes resulted in an increment of Hb E of 3.31% for a group with one α-gene defect and 5.92% for a group with two α-gene defects.  21 The explanation for this phenomenon relies on the electrostatic interaction of hemoglobin assembly. In Hb E heterozygotes, there are two types of β-globin chains, the β A and the more positively charged β E . When α-globin is insufficiently produced, the α-globin chain preferentially combines to β Arather than β Echains. Hence, Hb E level is reduced in proportion to the number of α-globin defects.

| D ISCUSS I ON
Unlike Hb E heterozygotes, a significant increment of Hb E was observed in children homozygous for Hb E coinherited with α-thal (coefficient = 3.31 for one α-gene defect and 5.92 for two α-gene defects), and this was also the consequence of reduced Hb F. As observed in adults homozygous for Hb E, 8,9 the concomitance of α-thal could result in a significant reduction in Hb F due to the preferential binding of α-globin chain, which is limited, to β Eglobin rather than the γ-globin chains.
Little information is known about the effect of sex on the expression of Hb F, A 2, and E. Previous studies reported that females had higher Hb F but lower Hb A 2 levels, as compared with males. [24][25][26] This is also the case for our study population. These data confirm that there might be some X-linked factors involved in the expression of these two Hbs.
This study has limitation that iron deficiency (ID) was not excluded because of incomplete data on iron status. Whether ID has significant effect on the levels of Hb F, A 2, and E in young children requires further studies. Nonetheless, our study demonstrates a variation in Hb F, A 2 , and E expression in Laotian children 6-23 months of age. The key factors contributing to the expression of Hb F and A 2 appeared to be age and types of thalassemia, specifically Hb E and its interaction with α-thal. Sex also had a significant effect on Hb F and A 2 expression although the effect size was small. For Hb E, its expression appeared to rely mainly on the concomitance of α-thal. It is of immense importance to note that a wide variation in Hb F levels could lead to the difficulty in thalassemia diagnosis at this particular age. This is especially true for children homozygous for Hb E as we found many cases with remarkably high Hb F levels.
Without DNA analysis, these cases could be misdiagnosed as Hb E-β-thal disease. To differentially diagnose these two diseases, family study and DNA analysis are needed. Alternatively, the EE score recently described is probably useful for this differentiation. 27,28  Dr Pongdet Sarakarn, a biostatistician, for his kind advice on statistical analysis.

CO N FLI C T O F I NTE R E S T S
The authors have no competing interests.