Positive Association of Serum Vitamin B6 Levels with Intrapulmonary Lymph Node and/or Localized Pleural Metastases in Non-Small Cell Lung Cancer: A Retrospective Study

The relationship between vitamin B levels and the development and progression of lung cancer remains inconclusive. We aimed to investigate the relationship between B vitamins and intrapulmonary lymph nodes as well as localized pleural metastases in patients with non-small cell lung cancer (NSCLC). This was a retrospective study including patients who underwent lung surgery for suspected NSCLC at our institution from January 2016 to December 2018. Logistic regression models were used to evaluate the associations between serum B vitamin levels and intrapulmonary lymph node and/or localized pleural metastases. Stratified analysis was performed according to different clinical characteristics and tumor types. A total of 1498 patients were included in the analyses. Serum vitamin B6 levels showed a positive association with intrapulmonary metastasis in a multivariate logistic regression (odds ratio (OR) of 1.016, 95% confidence interval (CI) of 1.002–1.031, p = 0.021). After multivariable adjustment, we found a high risk of intrapulmonary metastasis in patients with high serum vitamin B6 levels (fourth quartile (Q4) vs. Q1, OR of 1.676, 95%CI of 1.092 to 2.574, p = 0.018, p for trend of 0.030). Stratified analyses showed that the positive association between serum vitamin B6 and lymph node metastasis appeared to be stronger in females, current smokers, current drinkers, and those with a family history of cancer, squamous cell carcinoma, a tumor of 1–3 cm in diameter, or a solitary tumor. Even though serum vitamin B6 levels were associated with preoperative NSCLC upstaging, B6 did not qualify as a useful biomarker due to weak association and wide confidence intervals. Thus, it would be appropriate to prospectively investigate the relationship between serum vitamin B6 levels and lung cancer further.


Introduction
Lung cancer is one of the most prevalent malignancies and remains the leading cause of cancer death worldwide [1]. Non-small cell lung cancer (NSCLC) accounts for approximately 85% of lung cancers and represents a significant economic and social burden [2,3]. Tumor-nodes-metastasis (TNM) staging is commonly used to clarify the staging of NSCLC and plays an important role in the treatment and prognosis [4]. However, it is difficult L-5,6,7,8-tetrahydrofolic acid, and cobalamin, respectively. The analyzer enabled the use of an electrochemiluminescence method to deposit the measured substance on the sensor surface; then, the reverse voltage was applied to dissolve the substance accumulated on the sensor and the content was analyzed based on the polarization curve of the dissolution process. The normal thresholds were as follows: vitamin B 1 , 50-150 nmol/L, vitamin B 2 , 4.26-18.42 mg/L; vitamin B 6 , 14.6-72.9 nmol/L; vitamin B 9 , 6.8-36.3 nmol/L; and vitamin B 12 , 200-900 pg/mL.

Intrapulmonary Metastasis Status Classification
We evaluated intrapulmonary lymph nodes and pleural metastases using surgical records and pathology reports. The location and number of metastases were recorded. All the metastatic findings were pathologically confirmed according to the World Health Organization Classification of Thoracic Tumors 2021 [29] and the 8th edition of the Pathological Tumor-Node-Metastasis (pTNM) Staging from the Union for International Cancer Control [4]. Patients were then divided into two groups based on the presence or absence of intrapulmonary metastases.

Covariates
Based on previous experience and studies [18,23,27,30], clinical characteristics were evaluated as potential confounders of the association between serum vitamin levels and intrapulmonary metastasis status, including age, sex, BMI, family history of cancer, smoking history, alcohol consumption, the season of blood collection for vitamin testing, educational level, and tumor type (pathology, diameter, number, and density).

Statistical Analyses
Continuous variables are presented as mean ± standard deviation (normal distribution) or median [interquartile range] (non-normal distribution). Categorical variables are presented as numbers and percentages. Quantitative variables were compared between groups using Student's t-test, analysis of variance, or non-parametric tests, while comparisons of categorical variables between groups were made using chi-squared tests.
Univariate and multivariate logistic regression analyses were used to evaluate the associations between serum vitamin levels and the presence of lung cancer or intrapulmonary metastases. Patients were then divided into four groups by quartiles of the serum vitamins selected by multivariate logistic regression analysis. A stratified analysis was used to examine the vitamin-metastasis association for different levels of factors including: age (<40, 40-60, or >60 years), sex, BMI (<18.5, 18.5-24, or >24 kg/m 2 ), family history of tumors (yes or no), smoking history (never a smoker, ex-smoker, or current smoker), alcohol consumption (yes or no), and the pathology (adenocarcinoma or squamous cell carcinoma), diameter (<1, 1-3, or >3 cm), number (solitary or multiple), and density (solid or subsolid) of the tumor(s). Estimated effects were reported as odds ratios (OR) and 95% confidence intervals (95% CI). The last three quartiles of serum vitamins were each compared to the first quartile with the lowest concentration. Tests for linear trends across serum vitamin quartiles were performed by assigning medians to each quartile and calculating coefficients for the quartile variables. p-values for interactions were also assessed by likelihood ratio tests comparing regression models with and without the cross-product terms for each assessed factor and serum vitamin levels (quartiles).
All analyses were performed in R software (version 4.2.2, The Free Software Foundation, Boston, MA, USA) and IBM SPSS Statistics for Windows, version 26.0 (IBM Corp., Armonk, NY, USA). A two-sided p-value < 0.05 was considered statistically significant for all tests.

Association of Serum B Vitamin Levels with Presence of Lung Cancer
None of the serum B vitamins showed a significant association with the presence of lung cancer in the univariate logistic regression (Table 3). In multivariate analysis, age and a subsolid tumor were positively associated with the presence of lung cancer.

Association of Serum B Vitamin Levels with Intrapulmonary Metastases
Among the serum B vitamins, only vitamin B 6 levels showed a positive association with intrapulmonary metastases in a multivariate logistic regression (OR: 1.016; 95%CI: 1.002-1.031) ( Table 4). Other significantly associated factors included: female gender, the diameter and number of the tumors as risk factors, and the density of the tumor as a protective factor. Meanwhile, there were no significant differences in the demographic, clinical, or tumor characteristics of patients in different quartiles for serum vitamin B 6 , except for those with different smoking histories (p < 0.032) ( Table 5).  After multivariable adjustment, the highest quartile (Q4) of serum vitamin B 6 was significantly associated with a 67.6% higher risk of metastasis compared with those in the lowest quartile (Q1) (OR = 1.676, 95% CI = 1.092-2.574, p = 0.018), with a p for trend of 0.030 (Table 6). We also evaluated the effect modification of the association between serum vitamin B 6 and metastases by age, sex, BMI, family history of cancer, smoking history, alcohol consumption, and the pathology, diameter, number, and density of the tumor(s) ( Table 6). The risk association appeared stronger for women (Q4 vs. Q1: OR of 1.968, 95% CI of 1.144 to 3.386, p = 0.014, p for trend = 0.014). Patients with a family history of cancer also appeared to have experienced a stronger association between vitamin B 6 and metastasis (Q4 vs. Q1: OR of 5.337, 95% CI of 1.492 to 19.093, p = 0.010, p for trend = 0.010). The association between serum vitamin B 6 and intrapulmonary metastasis also appeared to be stronger for current smokers, current drinkers, and patients with squamous cell carcinoma, a tumor of 1-3 cm in diameter, or a solitary tumor (Table 6). Meanwhile, age, sex, smoking history, alcohol consumption, and tumor type (pathology, diameter, number, and density) showed significant interactions with vitamin B 6 .  , and density), unless the factor is used for stratification. † Q2-Q4 of serum vitamin B 6 are each compared with Q1. ‡ p for trend is based on the statistical significance of the coefficient of the quartile variable (median value within each quartile). § p for interaction is based on the statistical significance of the cross-product term added to multivariable models. ¶ not enough data for stratified analysis.

Discussion
In this retrospective study, we found that serum vitamin B 6 levels might be positively associated with intrapulmonary lymph nodes and/or localized pleural metastases in NSCLC patients, and the risk association was stronger for women, current smokers, current drinkers, and patients with squamous cell carcinoma, tumors 1-3 cm in diameter, or a solitary tumor. However, even though serum vitamin B 6 levels were associated with preoperative NSCLC upstaging, B 6 did not qualify as a useful biomarker due to weak associations and wide confidence intervals. Preoperative vitamin B 6 levels might have some value in the preoperative evaluation of intrapulmonary lymph nodes and/or localized pleural metastases in NSCLC; thus, it would be appropriate to prospectively investigate the relationship between serum vitamin B 6 levels and lung cancer further.
Vitamin B 6 promotes cell growth, differentiation, proliferation, and metastasis [31] and has adjuvant anti-inflammatory and antioxidant effects [32]. Although large cohort studies have been conducted, the association between vitamin B 6 intake, blood levels, and catabolic levels with lung cancer remains inconclusive. Vitamin B 6 has been implicated as a tumor promoter [15,[23][24][25][26], tumor suppressor [19], or being unrelated to carcinogenesis [27,28] in various studies. This contradiction may be attributed to the significant variation in vitamin B 6 among populations of different races, diets, lifestyles, and vitamin supplementation habits [33,34]. In our study, we found that patients with lung cancer had higher serum vitamin B 6 levels than patients with benign lung nodules, while patients with intrapulmonary metastases had higher serum vitamin B 6 levels than non-metastatic patients, i.e., serum vitamin B 6 levels increased with increasing tumor status. Theoretically, vitamin B 6 could promote tumor cell growth, differentiation, and proliferation, as well as be a risk factor for lung carcinogenesis [35,36]. On the other hand, it could act as an antitumor factor with regard to the inflammatory response caused by tumor cells [37][38][39]. Several studies have shown that vitamin B 6 can act as a promotor in tumors by affecting DNA stability [40,41] and the activation of antioxidant enzymes [42][43][44][45], which are involved in the regulation of post-translational modifications of activated proteins. The relationship of vitamin B 6 with post-translational modifications of key proteins such as NF-κB [46,47], as well as inflammasomes such as NLRP3 [48,49] and receptor-interacting protein 140 (RIP140) [47,50], has received significant attention. Although there are no specific findings related to the mechanism of vitamin B 6 and lung cancer metastasis, we speculated that these two aspects mentioned above together determine the effect of vitamin B 6 on tumors. Factors such as demographics and tumor type may play different roles in both aspects, leading to different conclusions. In our study, the positive relationship between vitamin B 6 and intrapulmonary metastasis suggests that vitamin B 6 may contribute to tumor metastasis in NSCLC patients.
We measured serum PLP as an assessment of vitamin B 6 levels. Vitamin B 6 consists of pyridoxal (PN), pyridoxamine (PM), pyridoxal (PL), and their phosphorylated derivatives pyridoxal 5'-phosphate (PNP), pyridoxamine 5'-phosphate (PMP), and PLP [51]. PLP is the bioactive form of vitamin B 6 in vivo and has been the common measure of vitamin B 6 status testing in previous studies [19,38,[51][52][53][54]. However, PLP is considered unstable because it may be influenced by inflammatory status, alkaline phosphatase levels, serum albumin, and inorganic phosphate [55]. Recent studies have used plasma vitamer ratios (PAr), calculated as PA:(PLP + PL), to eliminate the influence of the above factors [55,56]. However, PAr requires the simultaneous measurement of three indices, and mainly reflects the catabolism of vitamin B 6 in the inflammatory state. In contrast, PLP serves as a representation of the true vitamin B 6 levels and can directly reflect the relationship between vitamin B 6 and lung cancer. PLP is also more compatible with clinical applications because it requires less blood sampling and is less expensive.
A stratified analysis was performed to clarify the association of vitamin B 6 with intrapulmonary metastasis in different populations. We found that the risk association was much stronger in some situations. Vitamin B 6 served as a strong risk factor in females, current smokers, current drinkers, and those with a family history of cancer. The study by Fanidi et al. also found a positive trend for elevated PLP levels and lung cancer risk in their cohort of Asian women [19]. In addition, an elevated serum vitamin B 6 level, even within the normal range, was a risk factor for those with a family history of cancer, suggesting a possible genetic link might exist between them. Meanwhile, smoking and alcohol consumption are well-known risk factors for lung cancer [15,19]. Vitamin B 6 may be associated with smoking-induced inflammation and immune activation, which was also one of the mechanisms related to smoking in lung cancer development [30,57,58].
Interestingly, pathology, tumor size, tumor number, and morphology all seemed to influence the strength of vitamin B 6 as a risk factor for intrapulmonary metastasis. Previous studies showed that increased vitamin B 6 catabolism was positively associated with the risk of lung carcinogenesis, especially in squamous cell carcinoma [15]. This finding was confirmed in our study. In addition, we found a stronger risk association in patients with a solitary tumor or a tumor 1-3 cm in diameter. A consistent trend toward an increased risk of metastasis with increasing vitamin B 6 levels was also found in the subsolid tumor population. These associations have not been previously studied, and the theoretical basis for them remains unclear. In clinical practice, patients with a solitary, subsolid tumor or a tumor 1-3 cm in diameter without metastases are good candidates for surgery. If the vitamin-metastasis association could be confirmed by future studies, vitamin B 6 could be of great value in supporting surgical treatment as a preoperative predictor for metastasis assessment.
In this study, we found that age, sex, smoking history, alcohol consumption, and several characteristics of tumors had significant interactions with vitamin B 6 , respectively. That is, different degrees of the above factors could significantly affect the strength of the association between vitamin B 6 and intrapulmonary metastasis. The results of the interaction and stratified analysis further confirmed that women, current smokers, current drinkers, patients with squamous cell carcinoma, those with tumors 1-3 cm in size, and those with a solitary tumor may be a part of the special population whose vitamin B 6 levels should be of great concern. In addition, although there was a significant interaction between age and vitamin B 6 , the p for trend was found to be statistically insignificant (ranging from 0.112 to 0.119), which might be caused by the sample size. Furthermore, there was no significant interaction between family history and vitamin B 6 , suggesting that the effect of family history on the association should be further investigated.
Tumor staging is crucial in cancer treatment and determines the outcome. The pTNM staging results could not be obtained comprehensively before surgery. Clinical TNM staging might be relatively accurate in evaluating the tumor stage, but are not precise in the evaluation of the nodule and metastasis stage, especially in the early stage of lung cancer. In our study, we found 53 patients (53/1283) who underwent surgery had postoperative confirmation of localized pleural metastases, suggesting that surgical treatment may not be appropriate for them. Vitamin B 6 testing may have the potential to improve preoperative TNM staging. Further studies with larger sample sizes are needed to verify the value of vitamin B 6 in surgical management.
Notably, the results obtained in this study should be treated with caution. On one hand, the association between lung cancer and vitamins is very complex. Many factors may affect serum vitamin levels, including dietary habits, vitamin supplements, body metabolism, and a variety of complications. In this study, we collected as much nutritional information as possible and excluded conditions that could affect serum vitamins. However, more rigorous prospective studies are necessary to properly research the pathogenetic relationship between vitamin levels in the blood and the occurrence and progression of lung cancer. On the other hand, even though vitamin B 6 levels showed positive association with intrapulmonary metastases in NSCLC patients, the relationship was weak, and the confidence intervals overlapped hugely. Thus, the statistically significant difference between groups of patients could not be translated into a predictor on the level of individual patients based on the results of this study. Vitamin B 6 levels might have some value in the preoperative evaluation of intrapulmonary metastases in NSCLC, but more research is needed before conclusions can be drawn. This study had several limitations. First, the sample size of this study was limited. Studies with larger sample sizes are needed for further validation. Second, there may be an interaction between vitamin B 6 and tumor metastasis status. The study identified a positive association between vitamin B 6 and intrapulmonary metastases, but the causeand-effect relationship between them was not clear. Third, this study was retrospective; therefore, prospective cohort studies are needed to explore the mechanism of action of vitamin B 6 levels and lung cancer. Fourth, we collected as much information as possible about patients' dietary intake, nutritional status, and nutritional supplementation, but as a retrospective study, nutrition-related clinical information was not sufficiently detailed and might influence the results.

Conclusions
In conclusion, our study demonstrated a possible positive association between serum vitamin B 6 levels and intrapulmonary lymph nodes and/or localized pleural metastases in NSCLC patients, and the association was stronger in specific populations including women, current smokers, current drinkers, patients with squamous cell carcinoma, those with tumors 1-3 cm in diameter, and those with a solitary tumor. However, B 6 did not qualify as a useful biomarker due to weak associations and wide confidence intervals. Preoperative vitamin B 6 levels might have some value in the preoperative evaluation of intrapulmonary lymph nodes and/or localized pleural metastases in NSCLC, and thus the relationship between serum vitamin B 6 levels and lung cancer should be prospectively investigated in the future.