Elevated Vitamin B12 Levels and Cancer Risk in UK Primary Care: A THIN Database Cohort Study

Abstract

Background: Elevated vitamin B12 levels (B12) are associated with increased short-term cancer risk. However, the implications for early cancer detection in primary care have not been assessed. Methods: Individuals with plasma B12 measurements were sampled from The Health Improvement Network primary care database, UK. Persons with low B12 levels were excluded together with persons with cancer or B12 treatment before date of B12 measurement. Incident cancer was the outcome of interest and was identified through Read codes. Individuals were disaggregated according to plasma B12 levels (unit: pmol/L): 150-600 (reference range values), 601-800, 801-1,000, and >1,000. Results: Among the 757,185 persons who met the inclusion criteria, we identified 33,367 incident cancers during 2,874,059 years of follow-up. We found a higher 1-year cancer risk among the 25,783 (3.4%) persons with elevated B12 levels compared with those with normal B12 levels. After multivariable adjustment for lifestyle factors and social deprivation, persons with B12 >1,000 pmol/L had a 1-year incidence rate ratio of 4.72 (95% confidence interval: 3.99-5.58). The association showed a nonlinear dose-response pattern, and it remained robust in stratified analyses, including when reducing the risk of confounding by indication in subanalyses. The risks were particularly elevated for liver cancer, pancreas cancer, and myeloid malignancies among persons with elevated B12 levels. Conclusions: Elevated plasma B12 levels were associated with a higher 1-year cancer risk than normal B12 levels among persons seen in UK primary care, suggesting that some cancers may affect B12 metabolism. Impact: Elevated B12 may mark occult cancer.

Strength of the Association of Elevated Vitamin B12 and Solid Cancers: An Adjusted Case-Control Study

“…In our study, the aOR was 2.0 [95%CI: 1.2-3.2] for nonmetastatic cancers and 4.2 [95%CI: 2.7-6.6] for metastatic cancers with a threshold of 1000 ng/L defining the elevated B12. Danish and British registries reported higher SIRs, i.e., 6.3 [95%CI: 5.7-6.9] and 4.7 [95%CI: 4.0-5.6], respectively [14,15]. There may be two explanations for these differences: the thresholds defining elevated B12 differed between the two registries (1084 and 1355 ng/L, respectively), and were higher than that of our study.…”

“…This finding was slightly lower than that reported in previous studies, including among inpatients with frequencies ranging from 10 to 15% [3,16]. Lower prevalence of elevated B12 was observed in the general population: 3.5% of the population had a B12 > 813 ng/L in the British registry [15], and 6.6% of the population had a B12 > 1084 ng/L in the Danish registry [14]. The differences between these studies may be explained by variations in the recruitment (mean age, inpatients or outpatients, studied departments) and the threshold defining the status of elevated B12.…”

“…To our knowledge, the highest threshold evaluated was 1000 pmol/L (1355 ng/L) in the British registry. In this study, the association between solid cancers and elevated B12 appeared to be stronger in people with a B12 > 1000 pmol/L than those with levels between 800 and 999 pmol/L [15]. Higher B12 levels can be observed and raise the question about a growing association with solid cancers.…”

“…A link between elevated vitamin B12 level and solid cancers has been discussed since the 1970s in case series, raising the possibility of using B12 as a biomarker of solid cancers [6]. Recently, two retrospective cohorts based on health registries have shown an association between elevated B12 and a newly-diagnosed solid cancer [14,15]. A B12 > 800 pmol/L (1084 ng/L) in the Danish registry published in 2013 and a B12 between 800 and 1000 pmol/L (1084-1355 ng/L) in the British registry published in 2019 were associated with the occurrence of a cancer within one year after the assay (Standardized Incidence Ration (SIR) respectively 6.3 [95%CI: 5.7-6.9] and 2.9 [95%CI: 2.4-3.5]) [15].…”

“…Recently, two retrospective cohorts based on health registries have shown an association between elevated B12 and a newly-diagnosed solid cancer [14,15]. A B12 > 800 pmol/L (1084 ng/L) in the Danish registry published in 2013 and a B12 between 800 and 1000 pmol/L (1084-1355 ng/L) in the British registry published in 2019 were associated with the occurrence of a cancer within one year after the assay (Standardized Incidence Ration (SIR) respectively 6.3 [95%CI: 5.7-6.9] and 2.9 [95%CI: 2.4-3.5]) [15]. Nevertheless, in these registries, the association between elevated B12 and solid cancers was not adjusted for all elevated B12-related causes, except for excessive vitamin B12 supplementation in the Danish registry and alcohol misuse in the British registry.…”