跳到正文
原文
Nature 健康护理主题·· 27 天前精选AI 评分63

Age and Ethnicity Influence Serum Holotranscobalamin and Methylmalonic Acid in Vitamin B12 Status Assessment

Age and ethnicity influence serum holotranscobalamin (‘Active’ B12) and methylmalonic acid concentrations, two markers of vitamin B12 status assessment

AI 导读

一项回顾性研究分析了东南伦敦61,193例holoTC检测及13,160例MMA检测结果。研究发现,Black成人holoTC浓度最高,White成人最低,且Black成人中56.8%的holoTC结果高于128 pmol/L。

推荐理由

基于6万余例常规样本,结果可提示统一参考区间在解读holoTC和MMA时可能未充分考虑种族差异。

正文

Introduction

Serum total vitamin B12 (B12) and holotranscobalamin (holoTC; ‘Active’ B12) are both biomarkers used for the assessment of vitamin B12 status, either alone or in combination with functional biomarkers (confirmatory testing), i.e. serum methylmalonic acid (MMA) and/or homocysteine [1, 2]. B12 assays measure the concentration of vitamin B12 bound to two vitamin B12 binding proteins circulating in blood, transcobalamin and haptocorrin. Vitamin B12 bound to transcobalamin represents the biologically active fraction of vitamin B12 (holotranscobalamin, holoTC). Automated assays measuring holoTC are widely available [2]. Studies support holoTC measurement as a first-line diagnostic test and show that holoTC may be the earliest marker for B12 depletion [1, 3]. Although the diagnostic advantage of using holoTC over B12 testing may overall be modest, for infants, children, and pregnant women, holoTC testing offers timely and more accurate diagnostic outcomes than B12 [4,5,6].

Several factors, including age and ethnicity, affect serum B12 concentration independently of vitamin B12 intake and/or its absorption. The reasons for the influence of age and ethnicity have not been fully elucidated [7,8,9]. Suggestions include differences in the abundance of the vitamin B12 binding proteins, haptocorrin and transcobalamin, as well as more efficient hepatic storage of B12, all possibly caused by genetic factors [7, 9, 10]. A high intake of milk may be a significant contributor to the overall concentration of B12 in young children [8]. Whether age and ethnicity influence holoTC concentrations is currently insufficiently covered [11].

In humans, the activity of two enzymes, methionine synthase and (R)-methylmalonyl-CoA mutase, is dependent on vitamin B12 [12]. Vitamin B12 (specifically the adenosyl-cobalamin form) is a cofactor for the methylmalonyl-CoA mutase catalysed conversion of methylmalonyl-CoA to succinyl-CoA [13]. MMA is a by-product of this pathway with no known biological function. In vitamin B12 insufficiency, excess methylmalonyl-CoA is hydrolysed to MMA, causing the circulatory concentration of MMA to increase. Assays for the measurement of MMA remain analytically challenging. and relatively expensive to perform. It is also recognized that interpretation is particularly complex in elderly populations and those with impaired renal function. There is a dearth of information on the influence of ethnicity on MMA concentrations [14]. Despite these limitations, MMA is regarded as a sensitive marker of vitamin B12 deficiency, and this test is increasingly used as a confirmatory test to define status [2, 15].

The aim of this study was to evaluate whether age and ethnicity influence holoTC concentrations and their impact on vitamin B12 status using serum MMA as a status index test. We also aimed to demonstrate how the diagnostic algorithm used in this work performs in clinical practice and to highlight discrepancies that may arise when paired holoTC and MMA testing is used.

Material and methods

Participants

De-identified serum holoTC, together with methylmalonic acid (MMA) results from routine samples from an ethnically diverse hospital patient population, from south east London, UK, between Jan 2022 and Oct 2023 were collected. HoloTC testing, with MMA reflex testing performed for holoTC results between 25–70 pmol/L, is routinely used at St Thomas’ Hospital in London to support the diagnosis of vitamin B12 deficiency. This diagnostic approach is applied consistently across both inpatient and outpatient samples included in this dataset [1]. Age, ethnicity, and clinical specialty of origin were also gathered. Patients were categorized by self-identified ethnicity: White, Black, Asians, Mixed (Mixed Asians, Mixed Black-Asian, Mixed Black-White, Mixed Asian-White, Mixed-Other), Asians Far East, Latin American, Middle East or Unknown. The study was conducted in accordance with the Declaration of Helsinki.

Analytical methods

HoloTC was measured using the ARCHITECT® immunoassay by Abbott Diagnostics, Maidenhead, UK. The tests were performed by the Nutristasis Unit, London, UK, with a satisfactory analytical performance monitored using the UK National External Quality Assessment Scheme (NEQAS), Birmingham, UK External Quality Assurance (EQA) scheme. The manufacturer’s quantification range was 5-128 pmol/L with holoTC <25, 25–70 and >70 pmol/L assigned as deficient, indeterminate and replete B12 status, respectively [15]. HoloTC values < 5 pmol/L and >128 pmol/L (below and above the dynamic assay range) were reported as <5 pmol/L and >128 pmol/L, respectively. For patients with holoTC 25–70 pmol/L and corresponding normal renal function (eGFR ≥60 mL/min/1.73m2), where available within one week prior to holoTC measurement, MMA was quantified by LC–MS/MS. Thresholds of >280 nmol/L and >360 nmol/L were applied to indicate vitamin B12 deficiency in patients aged ≤65 years and >65 years, respectively [16]. A further threshold of 750 nmol/L was applied, as this has been proposed to indicate confirmed vitamin B12 deficiency [17]. This cut-off was developed for diagnosing vitamin B12 deficiency in individuals with indeterminate plasma total vitamin B12 concentrations (156-672 pmol/L) [17]. A Gerstel Multi Purpose Sampler (MPS) coupled directly to liquid chromatography tandem mass spectrometry (MPS-LC-MS/MS) was used for MMA analysis. The Danish External Quality Assurance Scheme (DEKS) EQA scheme was utilised to monitor MMA performance https://deks.dk/en/?s=mma.

Statistical analyses

Age was categorized into twelve groups (0–1, 2–5, 6–9, 10–13, 14–17, 18–29, 30–39, 40–49, 50–59, 60–69, 70–79, 80+ yrs). Since the holoTC quantification range was between 5 and 128 pmol/L, values < 5 pmol/L and >128 pmol/L were converted to 4 pmol/L and 129 pmol/L, respectively. HoloTC results were additionally categorized into the following groups: (A) holoTC concentration, <5, 5–15, 16–24, 25–49, 50–70, 71–95, 96–128 and >128 pmol/L, (B) holoTC defined B12 status: <5–24, 25–70 and >70 pmol/L, (C) incidence of elevated MMA concentrations as a function of holoTC, 25–30, 31–40, 41–50, 51–60, 61–70 pmol/L.

For highly elevated MMA results, we selected an arbitrary threshold of 5000 nmol/L as indicative of metabolic disease. These results were excluded from analyses, where continuous MMA data were presented.

Descriptive statistics were used to present the distribution of holoTC by sex, ethnicity, age, concentration and requesting location. Mann–Whitney, Kruskal-Wallis and chi-squared tests (as appropriate) were used to test for differences in holoTC and MMA between ethnicities and age. Binary logistic regression was used to assess the likelihood of elevated MMA within the holoTC 25–70 pmol/L range. Statistical significance was set at p < 0.05. Analyses were performed using IBM SPSS Statistics (version 32).

Results

Characteristics of the whole patient population

A total of 61,193 anonymized holoTC results were analyzed (54.2% female).

The characteristics of the patient population stratified by ethnicity, gender, age and holoTC concentration categories are shown in Table 1.

Table 1 Characteristics of the patient population.

Full size table

HoloTC results were available from 38.9%, 17.2%, 5.0%, 1.9%, 1%, 0.3%, 0.2% of patients from White, Black, Asian, Mixed (combined), Asians Far East, Latin American and Middle East ethnic groups respectively. Ethnicity could not be categorized for 35.5% of patients. 62.3% of samples were from people over the age of 50 years. Only 4.6% samples were analyzed for children. The frequency of holoTC measurements was higher for women of childbearing age (18-49 yrs old) compared to men of the same age, while more requests came from men >50 years of age compared to women. There were no overall sex differences in vitamin B12 status when holoTC concentrations were categorised (Table 1) and the same pattern was observed within the main ethnic groups.

There was also no overall difference in holoTC concentrations between females and males (p = 0.053). However, sex differences were identified within some ethnic and age subgroups. For example, although holoTC concentrations did not differ overall between White females (median = 97 pmol/L) and White males (median = 96 pmol/L; p = 0.092), White females aged >40 years had higher holoTC concentrations (median = 101 pmol/L) and lower MMA concentrations (median = 258 nmol/L) than White males (median holoTC = 94 pmol/L; median MMA = 280 nmol/L), with the difference in holoTC being statistically significant (p < 0.001). In contrast, no significant sex differences in holoTC concentrations were observed among Asian or Black participants aged >40 years.

Because exact holoTC values above 128 pmol/L were not available, limiting comparisons between males and females at higher concentrations, more detailed sex-stratified analyses were not undertaken.

The distribution of holoTC results by location

HoloTC results were requested at 364 clinical locations served by Guy’s and St. Thomas’ NHS Trust. The locations were further grouped into 44 specialities/locations. The prevalence of results was the highest (N = 10,418, 17%) from the renal specialty, followed by adult assessment 8.3%, gastroenterology 7.5%, geriatrics 5.8%, and cardiology 5.7%. N = 3406 (5.6%) of samples could not be categorized by specialty (unknown). Samples were referred from haematology (5.2%), neurology (3.8%), orthopaedics (3.6%), dental (3.6%), diabetes (3.4%), dermatology (2.7%), children (2.6%), antenatal (2.4%), inherited metabolic disease (1.8%), general medicine (1.7%), cardiovascular (1.7%), accident and emergency (AE), urology (1.1%) and oncology (1%) departments. Samples from neonatal, foetal, postnatal, gynaecology, ophthalmology, endocrinology, endoscopy, ear, nose and throat (ENT), rheumatology, intensive care unit (ICU), lupus, amputee, chemotherapy, breast unit, virology, allergy, anticoagulation specialities/locations consisted of less than 1% of total referrals each.

The proportion of samples originating from the renal specialty was significantly higher among Black (30.7%) and Asian (23.7%) patients than among White patients (16.8%). While the distribution of age and ethnic groups varied naturally across these 44 specialties, with expected demographic clustering in departments such as geriatrics or antenatal, the inclusion of such a broad range of clinical locations ensures that the observed cohort reflects a diverse, hospital-wide population rather than the specific case mix of a single department.

The distribution of holoTC concentrations by age group and holoTC concentration categories within the most abundant ethnic groups for females and males combined

Significant variations in holoTC concentrations were seen between the three most prevalent ethnic patient groups: White, Black and Asians, Fig. 1. The sample sizes for patients of Mixed, Asian Far East, Latin American, and Middle Eastern backgrounds were too small to allow age‑based comparisons within these groups.

Fig. 1: Mean serum holoTC concentration by age and ethnicity.

HoloTC values < 5 pmol/L and >128 pmol/L were assigned concentrations of 4 pmol/L and 129 pmol/L, respectively. Error bars represent 95% confidence intervals. Group sizes: n = 101, 241, 205, 228, 306, 1371, 2298, 2597, 3417, 4190, 4982, 3840 (White patients); n = 26, 93, 69, 100, 178, 862, 1166, 1393, 2289, 2001, 1238, 1110 (Black patients); n = 20, 67, 59, 58, 88, 303, 483, 443, 355, 554, 348, 275 (Asian patients) and Kruskal-Wallis p-values for differences between groups: 0.425, 0.005, 0.066, 0.438, 0.029, <0.001, <0.001, <0.001, <0.001, <0.001, <0.001, <0.001 for 0–1, 2–5, 6–9, 10–13, 14–17, 18–29, 30–39, 40–49, 50–59, 60–69, 70–79, 80+ years old, respectively.

Black adults ( > 30 years) had the highest holoTC concentrations, followed by Asians adults, with the lowest values seen in White adults (p < 0.001 between each group). In children, differences in holoTC concentrations between ethnic groups were not statistically significant in the 0–1 years (p = 0.425), 6–9 years (p = 0.066), or 10–13 years (p = 0.438) age categories. In the 2–5 years age group, Asian children had significantly higher holoTC concentrations than both Black (p = 0.002) and White children (p = 0.003). A significant difference was also observed in the 14–17 years age category, where Asian children had higher holoTC.

The removal of results from renal (17.0% of samples), antenatal (2.4%), metabolic disease (1.8%) and oncology (1%) specialties had no significant impact on differences in holoTC concentrations between these three ethnic groups. Given the large proportion of samples with holoTC concentrations exceeding the assay quantification limit (128 pmol/L), mean holoTC values may be biased across age and ethnicity (Fig. 1). It should be noted as well that the sample sizes within some paediatric age categories were relatively small, which may have limited the statistical power of the analyses and reduced the ability to detect differences between ethnic groups.

Accordingly, holoTC results were also analyzed by concentration category (Table 2). These analyses supported the findings presented in Fig. 1 and showed that the prevalence of replete B12 status (holoTC >70 pmol/L) was 83%, 78% and 68% in Black, Asian and White patients, respectively (Table 2). Moreover, the prevalence of results >128 pmol/L was highest for patients with a Black ethnic background, 56.8% (Table 2).

Table 2 The prevalence of holoTC concentrations by concentration categories <5, 5–15, 16–24, 25–49, 50–70, 71–95, 96–128 pmol/L within White, Black and Asian patients.

Full size table

The prevalence of vitamin B12 status (deficient, indeterminate, replete) based on holoTC concentrations

Overall, the prevalence of holoTC <25, 25–70 and >70 pmol/L was 1.7%, 25.3% and 73.0%, respectively. When stratified by age, the highest prevalence of holoTC <25 pmol/L was observed in the 0–1 year age group (6.2%), followed by those aged 14–17 years (2.5%), 18–29 years (2.4%), and ≥80 years (2.0%). Within the three most prevalent ethnic groups, the highest incidence of low and indeterminate (requiring MMA confirmation) holoTC concentrations was for White patients, 1.9% and 29.6%, respectively. Only 15.9% of Black patients had holoTC between 25 and 70 pmol/L, with 56.8% of holoTC >128 pmol/L.

Patients who had MMA testing performed

The total number of MMA tests performed for this sample cohort was 13,160 which included N = 45 for samples with holoTC <25 pmol/L, N = 12,233 for samples with holoTC between 25–70 pmol/L and N = 882 for samples with holoTC >70 pmol/L. For patients within the holoTC 25–70 pmol/L range, MMA test is automatically reflexed to confirm B12 status. A small proportion of patients outside these thresholds, as indicated above, also had MMA measured upon request.

Of the 45 samples with holoTC <25 pmol/L, 32 (71%) had elevated MMA which is concordant with B12 deficiency. Thirteen cases (29%) had normal MMA despite holoTC <25 pmol/L. These included a 4-year-old child, likely receiving treatment for presumed transcobalamin deficiency (holoTC <5 pmol/L). In this metabolic condition, holoTC is consistently undetectable, even in patients receiving vitamin B12 injections, because the assay antibody relies on binding to transcobalamin-B12 complex [18, 19], which is either absent or non-functional. Other cases with holoTC <25 pmol/L and normal MMA included; a 13-year-old from the surgery specialty (holoTC = 14 pmol/L); and a 27-year-old from neurology with MMA = 274 nmol/L, close to the deficiency cut-off. The remaining ten cases had holoTC between 20 and 23 pmol/L, with MMA values ranging from 111 to 300 nmol/L.

A total of 882 (6.7%) of MMA tests were ordered and performed for samples with a holoTC >70 pmol/L, of which 707 (80.2%) had normal MMA. There were 71 (41%) results within those with elevated MMA (N = 175 (19.8%)) which had a holoTC >128 pmol/L, suggesting that most of these cases had either a metabolic disorder or renal impairment. Sixteen of these cases with holoTC >128 pmol/L had MMA > 5000 nmol/L.

Overall, 15,475 (25.3% of all results) patients had holoTC between 25 and 70 pmol/L. Of these 12,233 (79%) had MMA testing performed. Of the 21% of samples with holoTC between 25 and 70 pmol/L for which MMA was not measured, most likely included patients with impaired renal function.

The distribution of MMA concentration by age and ethnicity for samples with holoTC 25–70 pmol/L

The median (IQR) MMA for samples with holoTC values between 25–70 pmol/L was 238 (184–330) nmol/L. N = 17 had highly elevated MMA > 5000 nmol/L, which included one infant, and eight patients in the 17–30 years age bracket.

The distribution of MMA concentrations by ethnicity showed that, across all age groups and within holoTC 25–70 pmol/L range, Black patients had the lowest MMA concentrations compared with Asian and White patients (Fig. 2). Overall, there was no difference in MMA concentrations between Asians and White patients. However, when stratified by age, White patients had higher MMA up to 60 years of age, while Asians had higher MMA for 60+ age groups when compared with Whites.

Fig. 2: Median MMA (nmol/L) concentrations by age and ethnicity for patients with holoTC 25–70 pmol/L.

*N = 17 cases with MMA > 5000 nmol/L were excluded. Group sizes: n = 8, 20, 31, 36, 123, 484, 676, 637, 728, 926, 986, 567 (White patients); n = 6, 12, 8, 21, 69, 242, 268, 203, 193, 145, 69, 79 (Black patients); n = 0, 3, 6, 6, 24, 80, 132, 74, 43, 51, 30, 22 (Asian patients); for 0–1, 2–5, 6–9, 10–13, 14–17, 18–29, 30–39, 40–49, 50–59, 60–69, 70–79, 80+ years old, respectively.

The incidence of an elevated MMA concentration in the indeterminate holoTC range by ethnicity

Overall, the incidence of an elevated MMA  > 280 and >360 nmol/L ( ≤ 65 and >65 yrs old, respectively) concentration in patients with holoTC between 25 and 70 pmol/L, suggesting B12 deficiency, was 30% (N = 3720). The overall suggested B12 deficiency based on these MMA cut-offs and holoTC <25 pmol/L was 7.8% (N = 4772) in the whole cohort of 61,193 results. The prevalence of elevated MMA decreased as holoTC concentrations increased, from 50% in patients with a holoTC 25–30 pmol/L to 21% in those with holoTC between 61–70 pmol/L (Table 3) when MMA cut-offs of >280 and >360 nmol/L were used.

Table 3 The prevalence of elevated MMA concentrations using two sets of cut-offs: (1) ( > 280 and >360 nmol/L ( ≤ 65 and >65 years old, respectively))* (2) > 750 nmol/L** for patients with holoTC between 25 and 70 pmol/L.

Full size table

When a 750 nmol/L MMA cut-off was used to define ‘definite’ vitamin B12 deficiency in patients with holoTC concentrations between 25 and 70 pmol/L, 3.6% of results (N = 441) exceeded this threshold (Table 3). Based on this MMA threshold, together with holoTC concentrations <25 pmol/L, the overall prevalence of suggested vitamin B12 deficiency was 2.4% (N = 1493) in the entire cohort of 61,193 results. The prevalence of elevated MMA decreased as holoTC concentrations increased, from 8.2% in patients with holoTC concentrations of 25–30 pmol/L to 1.5% in those with holoTC concentrations of 61–70 pmol/L when the 750 nmol/L MMA cut-off was applied (Table 3).

The proportion of samples within holoTC ranges, as specified in Table 3 was consistent within individual ethnic groups. However, when the prevalence of elevated MMA concentrations was compared by ethnicity among samples with holoTC concentrations of 25–70 pmol/L, substantial differences were observed. Using MMA cut-offs of >280/ > 360 nmol/L and >750 nmol/L, Black patients had only 15% and 1.2% of samples with elevated MMA concentrations, compared with 34% and 4.0% in White patients and 32% and 4.7% in Asian patients, respectively (Table 3 and Fig. 3A, B). Median MMA concentrations were 257, 185 and 226 nmol/L; p < 0.001 for White, Black and Asian patients, respectively. Binary logistic regression revealed that patients were more likely to have elevated ( > 280/ > 360 nmol/L) vs. normal MMA in holoTC 25–70 pmol/L range, if they were older and had lower holoTC, p < 0.001, but Asian ethnicity did not influence B12 status when the most abundant ethnicity (White) was used as a reference group, p = 0.823, Exp(B) = 0.977. However, Black patients were 34.6% less likely (p < 0.001) to have elevated MMA in holoTC 25–70 pmol/L range than White patients.

Fig. 3: Incidence of elevated methylmalonic acid (MMA) concentrations within the indeterminate holotranscobalamin (holoTC) range, stratified by ethnicity.

A Elevated MMA concentrations defined as >280 nmol/L in individuals aged ≤65 years and 360 nmol/L in individuals aged 65 years. B Elevated MMA concentrations defined as >750 nmol/L.

Among patients with MMA concentrations between >280/ > 360 nmol/L (age-dependent) and ≤750 nmol/L, the female-to-male ratio was 50.4–49.6%. Of these samples, 9.1% were from the gastroenterology specialty, followed by adult assessment (8.8%), geriatrics (7.8%), cardiology (7.1%), and surgery (7.1%). Overall, 41.9% of patients were aged >60 years, and the ethnic distribution was 74.8% White, 8.3% Black, and 6.0% Asian.

Among patients with MMA concentrations >750 nmol/L, the female-to-male ratio was 43.6–56.4%. Of these patients, 13.2% were from the geriatrics specialty, followed by unknown specialty (11.6%), surgery (9.0%), adult assessment (7.5%), and gastroenterology (7.3%). Overall, 56.1% of patients were aged >60 years, and the ethnic distribution was 73.7% White, 5.5% Black, and 7.6% Asian.

Other specialties of note in which the proportion of patients with MMA concentrations between >280/ > 360 nmol/L (age-dependent) and ≤750 nmol/L was higher than among those with MMA concentrations >750 nmol/L included diabetes (3.4% vs. 1.7%), haematology (6.2% vs. 3.3%), and neurology (5.0% vs. 3.1%). In contrast, the proportion of patients from the renal specialty was higher among those with MMA concentrations >750 nmol/L than among those with MMA concentrations >280/ > 360 nmol/L (age-dependent) and ≤750 nmol/L (5.9% vs. 2.1%).

Discussion

In this large hospital-based cohort, holoTC concentrations differed significantly across age and ethnic groups, with the highest values observed among Black adults, intermediate values among Asian adults, and the lowest values among White adults (p < 0.001). In patients with holoTC levels of 25–70 pmol/L, MMA was assessed as a confirmatory indicator of vitamin B12 deficiency. The prevalence of elevated MMA likewise varied by age and ethnicity, and Black patients consistently demonstrated the lowest MMA concentrations within this holoTC range across all age groups.

Consistent with our previous findings for the serum B12 test [7, 8], we observed age-related changes in holoTC concentrations across all ethnic groups. HoloTC levels rose after infancy, peaked at approximately six years of age, and then declined through adolescence into adulthood. From around 30 years of age, concentrations steadily increased until approximately 60–70 years, after which they began to fall. This trajectory closely mirrored the age-related pattern previously reported for total serum B12 [8]. In adulthood, holoTC concentrations were substantially higher among individuals of Black ethnic origin, paralleling the differences seen for serum B12.

Variations in the production of vitamin B12-binding proteins (haptocorrin and transcobalamin) across the lifespan and between ethnic groups may help explain the parallel age- and ethnicity-related patterns observed for both serum B12 and holoTC [7, 8, 20].

Supporting this interpretation, higher transcobalamin concentrations have been reported in Black compared with White individuals [9].

Interestingly, these ethnic differences were not evident in children across most age groups in our study. This suggests that early life may represent a period of relative metabolic homogeneity, potentially influenced by common dietary exposures such as breastfeeding or formula feeding, among children living in southeast London, as well as the immaturity of genetic expression involved in B12 metabolism. These factors may attenuate ethnic differences in childhood, with divergence emerging later due to genetic, dietary, or lifestyle influences.

Our study also showed that the highest prevalence of holoTC <25 pmol/L, indicating B12 deficiency when interpreted using reference intervals derived from adult cohorts, occurred in the 0–1 year age group, followed by individuals aged 14-29 years and those aged 80 years and older. These findings align with previous research [7, 20]. Several mechanisms may explain this pattern across the life course. In infants, low holoTC is often attributable to maternal B12 status, particularly among exclusively breastfed infants [21], or to congenital disorders affecting cobalamin absorption or metabolism [22]. Among adolescents, dietary insufficiency combined with increased physiological demands during rapid growth may contribute to lower holoTC concentrations [23]. In older adults, age-related reductions in gastric acid secretion and intrinsic factor production, together with increased comorbidities and polypharmacy, such as use of metformin or proton pump inhibitors, are well-established risk factors [24, 25].

Contrary to our serum B12 findings, significant differences in holoTC concentrations were observed between adult Asian and White patients, with Asians exhibiting higher concentrations across all adult age groups. Factors that may explain this discrepancy include higher rates of vitamin B12 supplementation among Asian patients [8] than among White patients, combined with the more rapid response of holoTC to recent B12 intake compared with total serum B12. The latter may make elevated holoTC levels more readily detectable in populations with higher supplementation rates [8].

In addition, holoTC is more strongly influenced by renal status than serum B12 [14], with concentrations positively associated with rising creatinine and/or declining eGFR [1], possibly because of a lower cell uptake of cobalamin in patients with renal impairment [26]. Although renal function data were not directly analysed in this study, the proportion of samples originating from the renal specialty was significantly higher among Asian (23.7%) and Black (30.7%) patients than among White patients (16.8%). This finding further supports this interpretation and is consistent with previous studies demonstrating ethnic differences in both the prevalence and management of renal disease [27, 28].

Further support comes from the observed differences in MMA concentrations between Asian and White patients (Fig. 2), including higher MMA levels in Asians over 60 years of age. This suggests that, despite higher holoTC concentrations, vitamin B12 utilization in Asian patients may be influenced by factors unrelated to B12 supply, or that the metabolic pathway leading to MMA production is affected by mechanisms independent of vitamin B12 status.

MMA is formed from D-methylmalonyl-CoA by the enzyme D-methylmalonyl-CoA hydrolase. Under normal physiological conditions, D-methylmalonyl-CoA, derived from propionate, itself produced during amino acid and odd-chain fatty acid catabolism, is converted to succinyl-CoA through sequential actions of methylmalonyl-CoA racemase and methylmalonyl-CoA mutase. The latter reaction requires 5’-deoxyadenosylcobalamin as a cofactor. When cobalamin status is sub-optimal, methylmalonyl-CoA mutase cannot function effectively, leading to accumulation of its substrate and subsequent conversion to MMA [29].

Although vitamin B12 status and renal function are the primary determinants of elevated MMA concentrations, other influences exist, including odd-chain fatty acids, cholesterol, and the gut microbiota [13, 16, 30]. Thus, additional metabolic factors that differ between ethnic groups may also contribute to the variation in MMA concentrations observed in our study.

Differences in MMA concentrations between ethnic groups may also have genetic origins. Polymorphisms in genes encoding enzymes not directly involved in vitamin B12 metabolism have been reported. Most notably, a single polymorphism in 3-hydroxyisobutyryl-CoA hydrolase (HIBCH) has been associated with elevated MMA concentrations in an Irish White population [29]. Additional variants in ACSF3, SUCLG1, and SUCLA2 have also been described [31].

Our finding of lower MMA concentrations in individuals of Black ethnic origin is consistent with previous studies [14, 32] which reported that non-Hispanic White individuals had the highest prevalence of elevated serum MMA ( ≥ 300 nmol/L) compared with non-Hispanic Black and Hispanic/Mexican populations across all age groups. Notably, among those aged ≥50 years, the prevalence of elevated MMA in non-Hispanic Whites was nearly double that of non-Hispanic Blacks [32].

In line with Ganji and Kafai [32], serum MMA levels in all major ethnic groups began to increase after age 40. Their study reported a 63% higher MMA concentration in individuals over 70 years compared with those aged 18–20 years. In our cohort, the age-related increase in MMA among patients with holoTC 25–70 pmol/L was more pronounced in White and Asian patients compared with Black patients.

These age-related rises in MMA are likely driven by decreased renal function, reduced B12 absorption with age, and the use of medications that interfere with vitamin B12 absorption. However, the observed ethnic differences in MMA concentrations cannot be attributed to poorer kidney function in Whites; on the contrary, Black individuals had worse renal function than non-Hispanic Whites in the study by Ganji and Kafai [32]. In our study, the proportion of samples originating from the renal specialty was the highest among the Black patients.

The decrease in the percentage of elevated MMA ( > 280/ > 360 nmol/L, age-dependent) with increasing holoTC concentrations was expected and is consistent with prior studies [15, 33]. However, within the holoTC range of 50–70 pmol/L, the reduction in elevated MMA was minimal (Table 3, Fig. 3A). Among Asian patients, a slight increase in the prevalence of elevated MMA was observed, rising from 23% to 24% across the holoTC ranges of 51–60 pmol/L and 61–70 pmol/L, respectively. These findings suggest that within the 50–70 pmol/L holoTC range, MMA ( > 280 nmol/L and >360 nmol/L, aged ≤65 years and >65 years respectively) has limited ability to discriminate between vitamin B12 deficiency and sufficiency, may be influenced by other factors, or that vitamin B12 status can generally be considered normal. The age-dependent MMA thresholds ( > 280/ > 360 nmol/L) were derived from the Hordaland Homocysteine Study in Norway and represent the 2.5th–97.5th percentiles in a selected middle-aged and elderly population with serum vitamin B12 concentrations ≥400 pmol/L (the vitamin B12-replete group) [16]. In the same study, the corresponding 2.5th–97.5th percentiles for an unselected middle-aged and elderly population were 320 nmol/L and 490 nmol/L, respectively [16]. The authors also acknowledged that a serum vitamin B12 concentration <400 pmol/L cannot be used to define vitamin B12 deficiency, as more than 60% of their population with apparently adequate vitamin B12 status would otherwise have been classified as vitamin B12 deficient [16].

In contrast, when the >750 nmol/L MMA threshold was applied to our data (Table 3 and Fig. 3B), the overall prevalence of elevated MMA was only 3.6%, compared with 30% using the age-dependent thresholds ( > 280/ > 360 nmol/L). The decline in the prevalence of elevated MMA with increasing holoTC concentrations was also consistent with this observation. Using the >750 nmol/L threshold, there was more than a two-fold reduction in the proportion of patients with elevated MMA between the holoTC categories of 41–50 pmol/L and 51–60 pmol/L, whereas only a 0.4 percentage point difference was observed between the 51 and 60 pmol/L and 61–70 pmol/L categories (Table 3). Furthermore, among Black patients with holoTC concentrations of 61–70 pmol/L, only 0.3% had an MMA concentration >750 nmol/L. A threshold of >750 nmol/L has been proposed for the diagnosis of vitamin B12 deficiency when plasma vitamin B12 and MMA are used in combination [17]. In this diagnostic algorithm, an MMA ‘grey zone’ of 290–750 nmol/L is defined, and if the MMA concentration falls within this range, the authors recommend repeating the MMA measurement after one year unless clinical signs or symptoms of vitamin B12 deficiency are present [17]. The 750 nmol/L cut-off is supported by a randomized placebo-controlled study demonstrating only limited clinical benefit of vitamin B12 treatment in patients with moderately elevated MMA concentrations [17, 34].

These findings, together with the observed age- and ethnicity-related differences, suggest that a single set of reference ranges for holoTC and MMA may not be appropriate when these biomarkers are used to guide clinical decision-making. Furthermore, our findings suggest that a holoTC concentration >50 pmol/L is likely to be indicative of a vitamin B12-replete state. Above this holoTC concentration, elevated MMA is possibly more likely to reflect factors other than vitamin B12 deficiency, including age, renal function, increased catabolism of MMA precursors, genetic variation (including polymorphisms in enzymes involved in MMA metabolism), and increased production of bacterial propionic acid [16]. Some of these factors may also contribute to the observed differences in holoTC and MMA concentrations between ethnic groups in this study.

This is particularly relevant in light of the recently published guidance from the National Institute for Health and Care Excellence (NICE) on the diagnosis and management of vitamin B12 deficiency in individuals over 16 years [NG239] [35, 36]. The guideline recommends using either serum B12 or holoTC as the initial or sole test to identify B12 deficiency. It also notes that people of Black ethnicity may have a higher reference range for serum B12 concentrations compared with White or Asian individuals. However, this adjustment is not applied to holoTC concentrations. According to NG239, holoTC <25 pmol/L confirms B12 deficiency, 25–70 pmol/L is considered indeterminate, and >70 pmol/L indicates B12 sufficiency for all patients [35].

Given our findings of clear ethnic differences in holoTC concentrations, applying a single reference range to all populations may not be optimal. Furthermore, NG239 recommends considering a further test to measure MMA in individuals with indeterminate holoTC who exhibit symptoms or signs of B12 deficiency. Our results suggest that a unified reference interval for MMA may also be inappropriate, as Black individuals have lower MMA concentrations compared with Asian and White patients. These observations underscore the need to consider age- and ethnicity-specific reference ranges when interpreting holoTC and MMA for clinical decision-making.

Limitations and proposed further work

This study has several limitations. All holoTC analyses were conducted using the ARCHITECT® immunoassay by Abbott Diagnostics, which has a relatively narrow dynamic range. The upper measurable concentration without sample dilution is 128 pmol/L. A substantial proportion of samples exceeded this limit; therefore, for data handling purposes, values of 129 pmol/L were assigned for analysis. As a result, the true ethnic differences in holoTC concentrations are likely greater than those reported. Nonetheless, categorising holoTC values helped mitigate this limitation during statistical analyses.

A further limitation of this retrospective study of de-identified laboratory data is the lack of access to comprehensive clinical records. Consequently, the specific clinical indications or underlying symptoms that prompted the requests for holoTC and MMA testing could not be ascertained. We were unable to comment on ethnic differences in smaller patient subgroups, such as individuals of mixed-race backgrounds or those from the Middle East, because these groups were under-represented in our South East London cohort.

Conclusions

Significant differences in holoTC concentrations were observed among Asian, Black, and White patients across adulthood. In line with serum B12 concentrations, Black adults had the highest holoTC levels, with nearly half recording values above 128 pmol/L. MMA concentrations, as well as the frequency of elevated MMA among individuals with holoTC values between 25–70 pmol/L, were also influenced by both age and ethnicity. Notably, Black patients had the lowest MMA concentrations across all age groups within this holoTC range, suggesting better B12 status, more efficient utilisation of holoTC, or potentially reflecting factors unrelated to cobalamin status. These findings support the consideration of age- and ethnicity-specific cut-offs for both holoTC and MMA.

来源:Nature 健康护理主题 · nature.com