Table 1.
Observational studies on plasma/serum vitamin B12 concentrations (p-B12 or s-B12) or B12 intake and liver cancer.
| Study, Design | Exposure/Outcome | High B12 vs. Reference Category | Results Summary | Adjustments/Limitations |
|---|---|---|---|---|
| Lin et al., 2010 [7], Case-control |
P-B12 (and B12 intake) in relation to survival in patients with hepatocellular carcinoma 90 cases and 90 controls, “Taiwan” |
Tertile T1 < 699 ng/L * = reference T2; 699–1500 ng/L T3; >1500 ng/L B12 > 699 (cut-off) vs. < 699 ng/L B12 intake did not differ by tertiles of plasma B12: in T1 2.8 ± 2.0 µg/d; in T2 2.6 ± 2.0 µg/d; in T3 2.4 ± 1.5 µg/d |
HR (95%CI) of survival 2.95 (1.22–7.11) 3.24 (0.99–10.60) 2.88 (1.26–6.60) B12 intake: 3.0 ± 5.7 μg/d in patients vs. 4.3 ± 8.3 μg/d in the age and sex matched controls. |
Adjustments not clear for what. High B12 associated with low albumin, hemoglobin, erythrocytes count, and alanina amino transferase, but with high alpha-fetal protein and tumor size. High B12 concentrations associated with malnourishment, liver injuries, and tumor progression. |
| Cui et al., 2016 [8], Case-control | P-B12 among 312 patients with hepatic cancer and 325 controls “China” |
lowest Q1 (227–265) = reference Q2 (266–406) Q3 (407–589) Q4 (590–1478) pmol/L |
OR (95%CI) for liver cancer 1.43 (0.72–2.81) 0.63 (0.31–1.25) 2.01 (1.02–3.98) |
Adjusted for age, sex, smoking, and Hepatitis B surface antigen. Small sample size and uncontrolled confounders. Higher proportion of patients with hepatic cancer had low B12 concentrations than in the controls. |
| Simonsen et al., 2014 [9], Case-control | P-B12 in 120 hepatic cancers and 46 controls and 102 patients with chronic liver diseases “Denmarkand Australia” |
Median (range) of p-B12 = 500 (120–1480) pmol/L in patients with liver cancer vs. 330 (140–800) pmol/L in the controls | P < 0.001 | Also TCN1 was elevated in plasma of patients with chronic liver diseases and those with liver cancer. |
| Chang et al., 2015 [10], Case-control | P-B12 in 204 cases with liver cancer and 415 controls “China” | p-B12, pmol/L Q1 (<154) = reference Q2 (154–229) Q3 (229–324) Q4 (>324) |
OR (95%CI) = 1.00 1.37 (0.59–3.16) 4.27 (2.00–9.10) 9.90 (4.80–20.44) |
Adjusted for age, sex, BMI, education, smoking, H.Pylori (in stomach cancer), Hepatitis B infection and aflatoxin (in liver cancer), and other micronutrients. Found positive association between p-B12 and esophagus, stomach cancer for the Q4 vs. Q1. But the association with liver cancer was stronger. |
| Arendt et al., 2016 [11], Cohort | P-B12 measured in the previous year/30-day mortality post diagnosis. 327 liver cancers were identified in health registers data of B12 measurements “Denmark” | < 200 pmol/L excluded 200–600 pmol/L = reference 601–800 pmol/L > 800 pmol/L |
mortality risk ratio 1.0 1.2 (0.6–2.5) 3.0 (1.7–5.3) |
Adjusted for age, sex, calendar year, Charlson comorbidity score index, and cancer stage. Analyzing the data by cancer type do not consistently support that the mortality is higher in patients with high B12. Excluded all B12 levels < 200 pmol/L, possible confounding by indication, and underreported supplements. |
| He et al., 2022 [12], Cohort | B12 intake in relation to mortality among 905 newly diagnosed hepatic cancer patients were recruited in the Guangdong Liver Cancer Cohort “China” | Median (P25, P75) of B12 intake, µg/d Q1 0.4 (0.1, 0.7) = reference Q2 1.1 (1.0, 1.2) Q3 1.6 (1.5, 1.8) Q4 2.8 (2.3, 4.3) |
Median B12 (IQR) intake in 12 months pre diagnosis of cancer 1.3 (0.9, 2.0) μg/d. HR (95% CI) for all-cause and hepatic cancer-specific mortality during the follow up of 791 days according to intake quartiles 1.04 (0.76–1.42) 0.86 (0.61–1.20) 0.83 (0.61–1.13) For hepatic cancer specific mortality 1.04 (0.76–1.42) 0.86 (0.61–1.20) 0.83 (0.61–1.13) |
Adjusted for sex, age, BMI, energy intake, physical activity, and education level, smoking, alcohol drinking, presence of chronic diseases (hypertension, diabetes, dyslipidemia, fatty liver disease, and cirrhosis), Barcelona Clinic Liver Cancer stage (0, A, B, C), and treatment (surgery, other treatments). |
* 0.74 to convert from ng/L to pmol/L. BMI, body mass index; CI, confidence intervals; HR, hazard ratio; OR, odds ratio; P, plasma; Q, quartile or quintile; T1 through 3, Tertiles.