Elevated Serum Vitamin B₁₂ Levels as a Prognostic Factor for Survival Time in Metastatic Cancer Patients: A Retrospective Study

ABSTRACT

Background: Serum vitamin B₁₂ levels have been proposed as one of the survival prediction factors, although no survival analysis in metastatic cancer patients has been conducted yet. This study examined whether serum vitamin B₁₂ levels could be a prognostic factor in metastatic cancer patients. Methods: Data from a retrospective chart review were used to perform Kaplan-Meier and multivariate analyses of the Cox proportional hazards. Subgroup analysis was performed on patients without a liver lesion (hepatocellular carcinoma or liver metastasis). Results: A total of 523 patients were included. The median survival time was 1.8 months (mo) in the high B₁₂ group (>911 pg/mL) and 5.1 mo in the normal B₁₂ group (211–911 pg/mL) (p < 0.001). In patients without a liver lesion, the median survival times were 2.1 and 6.1 mo in the high and normal B₁₂ groups, respectively (p < 0.001). Multivariate analysis revealed that serum vitamin B₁₂ level was an independent prognostic factor for overall survival (hazard ratio [HR]: 1.62; 95% confidence interval [CI]: 1.34–1.96, p < 0.001). Conclusion: Serum vitamin B₁₂ level can be used to predict survival time in metastatic cancer patients. Further large-scale cohort studies are required to confirm these findings.

Introduction

Metastatic cancer is incurable and its prognosis is poor. Medical teams have to make a choice between anticancer therapies to prolong the survival time and supportive care to manage patient symptoms according to individual circumstances. Although anticancer therapies could improve overall survival time in metastatic cancer patients (1), the toxic effects of aggressive treatment can be detrimental (2). Therefore, the risks and benefits should be considered for patients with a short life expectancy (3). According to a study on terminal cancer patients, medical teams tend to optimistically predict survival time (4), which can lead to the implementation of aggressive treatment rather than supportive care (3). However, serious adverse events have been reported due to the application of aggressive treatment regimens on vulnerable patients (5). Therefore, the accurate prediction of survival time is important for the welfare of terminal cancer patients (6).

Several prognostic tools have been developed to more accurately predict survival time in cancer patients. As their performance status worsen, poor prognosis is observed (7), and specific symptoms such as cancer-anorexia-cachexia syndrome (CACS), dyspnea, delirium, and cognitive failures have been suggested as prognostic factors of survival time (8–10). Biological factors such as leukocytosis, lymphocytopenia, and elevated C-reactive protein (CRP) levels have also been used as survival time prediction markers (11–13). Additionally, assessment tools such as palliative prognostic score (PaP) and palliative prognostic index (PPI), in which the symptoms listed above and biological factors are combined, have been developed (14,15).

Although studies on the significance of elevated vitamin B₁₂ levels in cancer patients are limited, an association has recently been reported (16). Furthermore, the diagnostic applicability of vitamin B₁₂ levels as a nonspecific cancer marker has been under discussion (17) and higher mortality has been observed in hepatocellular carcinoma patients with elevated serum vitamin B₁₂ levels compared to those with normal levels (18). In addition, there has been an effort to develop an index which reflects vitamin B₁₂ levels as an indicator of survival time (19), but survival time analysis research using serum vitamin B₁₂ levels in metastatic cancer patients is limited despite the fact that survival time prediction is most needed for this population.

This study investigated the characteristics of patients with elevated serum vitamin B₁₂ levels and the relationship between survival time and serum vitamin B₁₂ levels among metastatic cancer patients with various solid tumors. A retrospective chart review was conducted to identify factors that could be used as indicators for poor survival time in patients with metastatic cancer.

Materials and Methods

Subjects

Data were collected from cancer patients who visited the Kyung Hee University Hospital at Gangdong, Seoul between March 2006 and June 2015. Those above 18 years of age whose serum vitamin B₁₂ levels were measured were included. Patients who were not in the metastatic stage were excluded. Of the remaining included patients, except for those with leukemia or lymphoma, only patients with solid tumors were included. Patients who had undergone a procedure that could affect the absorption of vitamin B₁₂, had received vitamin B₁₂ therapy within the past two years, had gastric cancer, or had received a gastrectomy were excluded from the study. Patients with a low vitamin B₁₂ levels (<211 pg/mL) were also excluded.

The protocol was approved by the Kyung Hee University Hospital at Gangdong Institutional Review Board (IRB approval number KHNMCOH 2015-08-004).

Clinical Data

Patient information including gender, age, cancer origin, liver metastasis, performance status, use of parenteral nutrition, infection status, vitamin B₁₂ level, and creatinine level was collected during the chart review. If vitamin B₁₂ measurements had been performed multiple times, the last measurement was used. High-risk groups (absolute neutrophil count <100 cells/μL, for >7 days) with chemotherapy-induced neutropenia that required the prophylactic use of antibiotics (20) were not included in this study; so, it was assumed that individuals receiving antibiotic treatment had an infection. Therefore, infection status was defined as a patient who received antibiotic treatment. Patient survival was evaluated through December 31, 2015. Patients were divided into two groups according to serum vitamin B₁₂ levels: normal (211–911 pg/mL) and high (>911 pg/mL). The clinical characteristics and survival times were compared between the two groups.

Tumor origin was defined as follows: non-small cell lung cancer and small cell lung cancer were included in the lung cancer group; the hepatobiliary cancer group included hepatocellular carcinoma, gallbladder cancer, and cholangiocarcinoma; the gynaecological cancer group included endometrial carcinoma, cervical cancer, and ovarian cancer; the urological cancer group included bladder cancer and renal cell carcinoma; and the other tumor group included esophageal cancer, ampulla of vater cancer, pancreatic cancer, prostate cancer, thyroid cancer, and unknown primary cancer.

Statistical Analysis

The results are presented as n (%) or mean ± standard deviation (SD). Continuous and categorical variables were compared between the two groups using Student's t-test and chi-squared test, respectively. The overall survival time analysis was performed using the Kaplan-Meier method, and the differences between the two groups were compared using log-rank test. We analyzed the overall survival time for all patients together and for each tumor group. Subgroup analysis was performed on data from patients without infection or liver lesion (hepatocellular carcinoma or liver metastasis). We investigated 30-day and 90-day mortality rates with corresponding 95% confidence intervals (CIs) according to vitamin B₁₂ levels. To identify the predictors of survival, univariate and multivariate analyses were performed using the Cox proportional hazards model. Data were analyzed using PASW Statistics for Windows, version 18.0, and statistical significance was defined as p-values less than 0.05.

Results

Patient Characteristics

This study included 523 of 1,093 patients except for those who did not meet the inclusion criteria (four patients with low vitamin B₁₂ levels were excluded); of the included patients, 269 (51.4%) were men and 254 (48.6%) were women. As shown in Table 1, 302 patients (57.7%) were in the normal B₁₂ group and 221 patients (42.3%) were in the high B₁₂ group. The mean vitamin B₁₂ levels were 1360.8 pg/mL (range: 215–18,540) in total patients, 567.8 pg/mL (range: 215–906) in the normal B₁₂ group, and 2,444.5 pg/mL (range: 923–18,540) in the high B₁₂ group.

Table 1. Patient characteristics according to vitamin B₁₂ levels.

		Total (n = 523)	Normal B12Group (n = 302)	High B12Group (n = 221)
		n(%) or Mean ± SD	n(%) or Mean ± SD	n(%) or Mean ± SD	p value
Gender	Male	269 (51.4%)	157 (52.0%)	112 (50.7%)	0.768
	Female	254 (48.6%)	145 (48.0%)	109 (49.3%)
Age		59.9 ± 12.7	59.2 ± 13.2	60.9 ± 9.8	0.127
Tumor Origin	Lung	138 (26.4%)	83 (27.5%)	55 (24.9%)	0.003
	Hepatobiliary	124 (23.7%)	49 (16.2%)	75 (33.9%)
	Colorectal	96 (18.4%)	65 (21.5%)	31 (14.0%)
	Gynaecological	31 (5.9%)	19 (6.3%)	12 (5.4%)
	Breast	29 (5.5%)	18 (6.0%)	11 (5.0%)
	Head and neck	25 (4.8%)	18 (6.0%)	7 (3.2%)
	Urological	16 (3.2%)	9 (3.0%)	7 (3.2%)
	Sarcoma	14 (2.7%)	9 (3.0%)	5 (2.3%)
	Neurological	6 (1.1%)	3 (1.0%)	3 (1.4%)
	Other tumor groups	44 (8.4%)	29 (9.6%)	15 (6.8%)
Liver Lesion	Hepatocellular carcinoma	24 (4.6%)	7 (2.3%)	17 (7.7%)	<0.001
	Metastasis	198 (37.9%)	93 (30.8%)	105 (47.5%)
	None	301 (57.6%)	202 (66.9%)	99 (44.8%)
ECOG PS	0∼1	156 (29.8%)	112 (37.1%)	44 (19.9%)	<0.001
	2∼4	367 (70.2%)	190 (62.9%)	177 (80.1%)
Chemotherapy	Yes	160 (30.6%)	93 (30.8%)	67 (30.3%)	0.907
	No	363 (69.4%)	209 (69.2%)	154 (69.7%)
Radiotherapy	Yes	43 (8.2%)	28 (9.3%)	15 (6.8%)	0.307
	No	480 (91.8%)	274 (90.7%)	206 (93.2%)
Parenteral Nutrition	Yes	134 (25.6%)	60 (19.9%)	74 (33.5%)	<0.001
	No	389 (74.4%)	242 (80.1%)	147 (66.5%)
Infection	Yes	191 (36.5%)	87 (28.8%)	104 (47.1%)	<0.001
	No	332 (63.5%)	215 (71.2%)	117 (53.0%)
Creatinine	≤1.2 (mg/dL)	455 (87.0%)	263 (87.1%)	192 (86.9)	0.944
	> 1.2 (mg/dL)	68 (13.0%)	39 (12.9%)	29 (13.1%)

ECOG PS: Eastern Cooperative Oncology Group Performance status; p values for continuous variables were obtained by Student independent t-tests; p values for categorical variables were obtained by chi-squared tests; infection was defined as whether receiving antibiotic treatment at that time.

There were no differences between the two groups in gender and age distribution, the proportion of patients who were currently receiving chemotherapy and radiotherapy, and creatinine levels, but, tumor origin differed significantly (p = 0.003); in the normal B₁₂ group, the number of lung cancer patients was the highest, while in the high B₁₂ group, the number of hepatobiliary cancer patients was the highest. There were also differences in liver lesion, administration of parenteral nutrition, and infection status between the two groups: the percentages of patients with a liver lesion were 33.1% and 55.2% in the normal and high B₁₂ groups, respectively (p < 0.001); the percentages of patients who received parenteral nutrition were 19.9% and 33.5% in the normal and high B₁₂ groups, respectively (p < 0.001); and the percentages of patients with infection were 28.8% and 47.1% in the normal and high B₁₂ groups, respectively (p < 0.001). In addition, there were significant differences in Eastern Cooperative Oncology Group (ECOG) performance status scores between the two groups (p < 0.001); in the normal B₁₂ group, 37.1% of patients had good performance status (ECOG <2), compared with 19.9% of patients in the high B₁₂ group.

The mean vitamin B₁₂ level of the patients differed according to the cancer origin. Hepatobiliary cancer patients had the highest mean vitamin B₁₂ level (2,025.8 ± 2724.7 pg/mL), followed by neurological cancer (1,896 ± 2,527.8 pg/mL), colorectal cancer (1,314.3 ± 2,287.5 pg/mL), and lung cancer (1,205.9 ± 1,470.2 pg/mL). However, other breast cancer (803.6 ± 401.7 pg/mL), head and neck cancer (823.1 ± 531.9 pg/mL), and sarcoma (739.1 ± 361.8 pg/mL) had mean vitamin B₁₂ levels within the normal range (Figure 1).

Figure 1. Vitamin B₁₂ levels according to tumor origin.

Overall Survival Time

The median follow-up period, defined as the time from the test date to the survival time analysis period, was 57.9 ± 31.0 months (mo).

The 30-day mortality differed significantly between the two groups: 10.6% (95% CI: 7.3–14.6) in the normal B₁₂ group and 38.9% (95% CI: 32.6–44.8) in the high B₁₂ group (p < 0.001). The 90-day mortality also differed significantly between the two groups: 34.1% (95% CI: 29.5–40.1) in the normal B₁₂ group and 63.8% (95% CI: 57.5–70.6) in the high B₁₂ group (p < 0.001), as shown in Table 2.

The median survival times for each primary cancer are shown in Table 3. Significant differences in the median survival times between the normal and high B₁₂ groups were observed in lung, hepatobiliary, colorectal, breast, and head and neck cancer patients, In lung cancer patients, the median survival times were 4.4 and 2.1 mo in the normal and high B₁₂ groups, respectively (p = 0.001); in hepatobiliary cancer patients, the median survival times were 4.4 mo and 1.2 mo, respectively (p = 0.007); in colorectal cancer patients, the median survival times were 5.3 mo and 1.8 mo (p < 0.001), respectively; in breast cancer patients, the median survival times were 8.3 mo and 2.5 mo (p = 0.019), respectively; and in head and neck cancer patients, the median survival times were 29.7 mo and 1.7 mo (p = 0.001), respectively. The other groups, including those with gynaecological, urological, and neurological cancer and sarcoma patients, did not show significant differences in median survival times.

Figure 2. Survival curves according to vitamin B₁₂ levels. A: All patients; B: Patients without infection; C: Patients without a liver lesion.

(A)	Overall survival time for all patients Table 2. 30-day and 90-day mortality according to vitamin B12 levels. 30-day Mortality (95% CI) 90-day Mortality (95% CI) Normal B12 Group (n = 302) 10.6% (7.3–14.6) 34.1% (29.5–40.1) High B12 Group (n = 221) 38.9% (32.6–44.8) 63.8% (57.5–70.6) According to the Kaplan-Meier survival time analysis, the median survival times were 5.1 mo (95% CI: 4.1–6.1) and 1.8 mo (95% CI: 1.4–2.2) in the normal and high B12 groups, respectively. The median survival time differed significantly between the two groups (p < 0.001). The HR of the high B12 group was significantly higher than that of the normal B12 group (HR: 2.03; 95% CI: 1.69–2.44, p < 0.001) (Figure 2A).
(B)	Overall survival times in patients without infection Table 3. Overall survival times of metastatic cancer patients based on tumor origin and vitamin B12 levels. Median Overall Survival (95% CI) (months) Tumor Origin n(%) Normal B12 group High B12 group p value Lung 138 (26.4%) 4.4 (2.7–6.1) 2.1 (0.5–3.6) 0.001 Hepatobiliary 124 (23.7%) 4.4 (3.5–5.4) 1.2 (0.5–1.9) 0.007 Colorectal 96 (18.4%) 5.3 (2.4–8.1) 1.8 (1.2–2.5) <0.001 Gynaecological 31 (5.9%) 11.3 (3.3–19.3) 4.3 (0.0–9.0) 0.078 Breast 29 (5.5%) 8.3 (0.0–16.9) 2.5 (0.7–4.4) 0.019 Head and neck 25 (4.8%) 29.7 (12.7–46.7) 1.7 (1.4–2.0) 0.001 Urological 16 (3.2%) 4.6 (0.0–12.3) 2.1 (1.7–2.4) 0.478 Sarcoma 14 (2.7%) 3.3 (0.2–6.4) 1.1 (1.0–1.6) 0.290 Neurological 6 (1.1%) 8.6 (1.4–15.8) 6.1 (0.0–14.9) 0.786 Other tumor groups 44 (8.4%) 3.3 (1.5–5.2) 1.5 (0.6–2.5) 0.018 Table 4. Factors influencing overall survival time in metastatic cancer patients. Univariate Analysis Multivariate Analysis Variable HR (95% CI) p value HR (95% CI) p value Gender Female 0.92 (0.76–1.10) 0.338 Age 1.01 (1.00–1.02) 0.003 Liver lesion Yes 1.53 (1.28–1.84) <0.001 1.45 (1.20–1.75) <0.001 ECOG PS Bad (ECOG 2,3,4) 2.10 (1.70–2.59) <0.001 1.81 (1.46–2.25) <0.001 Chemotherapy Yes 0.81 (0.66–0.99) 0.037 Radiotherapy Yes 1.10 (0.79–1.53) 0.561 Parenteral Nutrition Yes 2.48 (2.01–3.06) <0.001 1.69 (1.34–2.14) <0.001 Infection Yes 1.94 (1.60–2.34) <0.001 1.45 (1.17–1.80) 0.001 Creatinine >1.2 (mg/dL) 1.47 (1.13–1.92) 0.004 1.40 (1.08–1.83) 0.012 Vitamin B12 >911 (pg/mL) 2.03 (1.69–2.44) <0.001 1.62 (1.34–1.96) <0.001 ECOG PS: Eastern Cooperative Oncology Group Performance status; infection was defined as whether receiving antibiotic treatment at that time. The median survival times in the normal and high B12 groups were 6.8 mo (95% CI: 5.3–8.4) and 2.7 mo (95% CI: 1.7–3.7), respectively (p < 0.001). The HR of the high B12 group was significantly higher than that of the normal B12 group (HR: 1.97; 95% CI: 1.55–2.50, p < 0.001) (Figure 2B).
(C)	Overall survival time in patients without a liver lesion The median survival times in the normal and high B12 groups were 6.1 mo (95% CI: 4.7–7.6) and 2.1 mo (95% CI: 1.2–3.1) (p < 0.001), respectively. The HR of the high B12 group was significantly higher than that of the normal B12 group (HR: 2.03; 95% CI: 1.57–2.63, p < 0.001) (Figure 2C).

Factors Influencing Overall Survival Times

In univariate analysis, the factors that influenced on overall survival time included age, liver lesion, performance status (ECOG ≥2), presently undergoing chemotherapy, administration of parenteral nutrition, infection status, serum creatinine level >1.2 mg/dL, and serum vitamin B₁₂ levels > 911 pg/mL (see Table 3). In multivariate analysis, performance status (HR: 1.81; 95% CI: 1.46–2.25, p < 0.001), administration of parenteral nutrition (HR: 1.69; 95% CI: 1.34–2.14, p < 0.001), serum vitamin B₁₂ level (HR: 1.62; 95% CI: 1.34–1.96, p < 0.001), infection status (HR: 1.45; 95% CI: 1.17–1.80, p = 0.001), liver lesion (HR: 1.45; 95% CI: 1.20–1.75, p < 0.001), and serum creatinine level (HR: 1.40; 95% CI: 1.08–1.83, p = 0.012) were identified as factors affecting the overall survival time.

Discussion

We compared the characteristics and survival times of metastatic cancer patients with normal and high B₁₂ levels and found that those with elevated serum vitamin B₁₂ levels had a shorter survival time than those with normal serum vitamin B₁₂ levels. Therefore, serum vitamin B₁₂ level may be a useful prognostic factor in metastatic cancer patients.

In this study, among metastatic cancer patients with high serum vitamin B₁₂ levels, infections, liver lesions, and parenteral nutrition support were more common, and their performance status tended to be lower than those with normal vitamin B₁₂ levels. In particular, comparison of vitamin B₁₂ levels for each tumor origin revealed the highest level among hepatobiliary cancer patients. The median survival time of the high B₁₂ group was significantly lower than that of the normal B₁₂ group and there was a significant difference in the 30-day and 90-day mortality rates between the groups. Subgroup analysis performed according to the cancer origin revealed significant differences in survival time according to serum vitamin B₁₂ levels among patients with lung, hepatobiliary, colorectal, breast, or head and neck cancers. In patients with gynaecological, urological, and neurological cancers and sarcoma, statistically significant differences in survival times according to vitamin B₁₂ levels were not observed, which might be due to the limited number of patients. Infection status and the presence of a liver lesion might affect serum vitamin B₁₂ levels and may also be associated with elevated levels (17). Therefore, subgroup analysis was performed after excluding patients with either infection or liver lesion in order to confirm whether vitamin B₁₂ level could be used as an independent factor for cancer survival. In each analysis, patients with elevated vitamin B₁₂ levels also had shorter survival times than those with normal vitamin B₁₂ levels, which suggest that elevated levels are independently related to poor cancer survival. According to the results of the Cox proportional hazard analysis, patient performance status, administration of parenteral nutrition, vitamin B₁₂ level, infection status, liver lesion, and creatinine level were independent prognostic factors of overall survival. In the institution in which this study was performed, the administration of parenteral nutrition is generally initiated when the patient's total intake is less than 30% of the daily nutritional requirements due to poor general condition and impaired oral intake. Therefore, the administration of parenteral nutrition could be considered an indicator of malnourishment instead of a prognostic factor. Furthermore, parenteral nutrition formulas contain a maximum of 5 μg vitamin B₁₂ per day, which suggests a negligible contribution to serum vitamin B₁₂ levels.

Generally, vitamin B₁₂ measurements are performed only when a deficiency is suspected (16). According to Stabler (21), vitamin B₁₂ deficiency is associated with megaloblastic anemia and neuropsychiatric disorders. However, elevated serum vitamin B₁₂ levels were more prevalent than vitamin B₁₂ deficiency among patients who underwent serum vitamin B₁₂ measurement and only 11% of patients had vitamin B₁₂ deficiency compared to 15% with high vitamin B₁₂ levels (16). Although elevated vitamin B₁₂ levels are not unusual among patients, clinical information on elevated vitamin B₁₂ is limited. Recent studies have shown that high serum vitamin B₁₂ levels are observed with diseases such as solid tumor, hematologic malignancy, autoimmune disease, liver disease, infectious disease, and kidney disease, which suggests the importance of the clinical interpretation of elevated vitamin B₁₂ levels (22,23). Moreover, in this study, the prevalence of vitamin B₁₂ deficiency among metastatic solid tumor cancer (except gastric) patients was 0.8% (4/527 patients), while 41.9% (221/527 patients) of the patients had high vitamin B₁₂ levels, which shows that, in particular, elevation of vitamin B₁₂ is frequently observed in metastatic cancer patients.

Progress has been made regarding the significance of elevated vitamin B₁₂ levels in cancer patients, with a reported 73% of patients diagnosed with cancer having elevated serum vitamin B₁₂ levels (22). Cancer risk increases with elevated vitamin B₁₂ level, mostly within the first year of the follow-up period, suggesting that vitamin B₁₂ level could be used as a cancer diagnostic marker (24). In addition, the relationship between elevated vitamin B₁₂ level and poor cancer survival time has been reported. Hepatocellular carcinoma patients with elevated vitamin B₁₂ levels have a shorter median survival time in a dose-dependent manner (18). According to a recent cohort study, cancer patients with an elevated vitamin B₁₂ level prior to cancer diagnosis tended to have poorer one-year survival and higher 30-day mortality (25). In addition, the vitamin B₁₂/CRP index (BCI), an index generated from vitamin B₁₂ and CRP levels, has been developed as a survival time predictor for cancer patients. In advanced cancer patients, the median survival time of patients with a BCI level over 40,000 is around 29 days (19,26,27); unfortunately, 29 days is too short for the implementation of an appropriate intervention. However, in addition to being a prognostic factor, elevated CRP can also reflect transient, curable infection (28). Therefore, measurement of vitamin B₁₂ levels for survival time analysis of cancer patients instead of BCI has the advantage of not being influenced by confounding factors such as infection. Other survival time prognostic factors such as PaP and PPI have a survival prediction time of 30 days, which is fairly short (14,15), while that using vitamin B₁₂ is around two months. In addition, vitamin B₁₂ level is an objective value and not a symptom with subjective characteristics; thus, it is expected to have a wider range of application.

Vitamin B₁₂ is metabolized and stored in the liver and a specific binding protein, transcobalamin (TC), is involved in the blood circulation and cellular uptake of vitamin B₁₂ (29). It is unclear why serum vitamin B₁₂ levels are elevated, although the following mechanisms have been suggested. First, increased serum vitamin B₁₂ levels may occur due to the direct disintegration or release of vitamin B₁₂ from the liver due to liver disease. Second, increased TC levels may result in increased serum vitamin B₁₂ levels because vitamin B₁₂ only circulates in the blood when it is bound to TC; free vitamin B₁₂ is excreted in the urine. TC levels may increase from their decreased clearance from circulation due to impaired liver or kidney function or by excessive TC synthesis induced by cancer or inflammation (30). Third, serum vitamin B₁₂ levels might increase because of decreased cellular uptake of vitamin B₁₂ from the blood. In this case, excess vitamin B₁₂ in the blood cannot be fully utilized by the cells, which leads to a vitamin B₁₂ deficiency-like state, including problems in methionine synthesis, as indicated by elevated levels of homocysteine, an intermediate in methionine metabolism (17,23,31). Of the three mechanisms mentioned above, the mechanism associated with elevated levels of vitamin B₁₂ in cancer patients may be related to the excessive synthesis of TC by the cancer cells; in the case of cancer patients with a liver lesion, it is also related to direct dissolution of vitamin B₁₂ from the liver.

This study has several limitations. It is a retrospective study based on chart reviews and includes various types of cancers; however, cancer-specific survival time analyses were performed to overcome this limitation. For several cancer origins, significant differences were found in survival times according to serum vitamin B₁₂ levels. However, no significant differences in survival times were observed in others, which might be due to the limited number of patients in these groups. Additional large-scale cohort or experimental studies are required to elucidate the exact mechanism and to confirm the significance of vitamin B₁₂ as a predictor of cancer patient survival time.

Conclusion

In conclusion, the results of this study suggest that vitamin B₁₂ level is a useful independent prognostic factor in metastatic cancer patients. Therefore, serum vitamin B₁₂ levels may be useful for the prediction of life expectancy in metastatic solid cancer patients. Prediction based on vitamin B₁₂ levels could help clinician make better decisions regarding the appropriateness of aggressive anticancer therapy.

Funding

This study was supported by the Traditional Korean Medicine R&D program funded by the Ministry of Health & Welfare through the Korea Health Industry Development Institute (KHIDI) (grant no. HB16C0067) and Research Institute of Human Ecology at Seoul National University.