The Role of Vitamin D on the Prognosis and Incidence of Lung Cancer: A Systematic Review and Meta-Analysis

Background: The correlation between vitamin D intake and lung cancer development is controversial. This meta-analysis aims to evaluate the relationship between vitamin D and the prognosis and incidence of lung cancer. Methods: A comprehensive database search on Pubmed, Web of Science, EBSCO, and Cochrane Library was carried out from the beginning to November 2020. Long-term survival and the incidence rate of patients with lung cancer were the primary outcomes of the study. Results: Ten eligible studies were selected for the meta-analysis following specic inclusion and exclusion criteria. Four included studies, covering 5007 patients, compared the overall survival (OS) and relapse-free survival (RFS) of lung cancer patients among total vitamin D users with non-users. Signicantly, the estimated pooled hazard ratio (HR) revealed that vitamin D could improve OS and RFS of lung cancer patients [HR=0.83, 95% CI (0.72-0.95); HR=0.79, 95% CI (0.61-0.97), respectively]. Vitamin D intake was inversely associated with lung cancer incidence in six studies [OR=0.90, 95% CI (0.83-0.97)]. Conclusions: The present meta-analysis shows vitamin D not only improves the long-term survival of lung cancer patients but has a benecial effect on the incidence of lung cancer. Notwithstanding, more studies are needed to conrm the study results.


Background
Lung cancer, with distinct subtypes, including non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) [1], is a leading cause of cancer incidence and death [2]. In 2018, GLOBOCAN estimated 2.09 million new cases (11.6% of total cancer cases) and 1.76 million deaths (18.4% of total cancer deaths), much higher than reported rates in 2012 (1.8 million new cases and 1.6 million deaths) [1]. Hence, early prevention and diagnosis are incredibly vital to reduce the morbidity and mortality of the disease. Presently, the role of vitamin D has been recognized in several cancers [3,4], including lung cancer [5].
Vitamin D is principally obtained from three pathways in humans (diet, supplements, and ultraviolet radiation from the sun synthesized on the skin). Additionally, vitamin D supplements usually contain a much higher dose of vitamin D compared to foods [6]. Vitamin D is metabolized to 25-hydroxy vitamin D (25(OH)D) in the liver and further activated by 1-hydroxylase to 1,25-dihydroxyvitamin D (1,25(OH) 2 D) in the kidneys. It also accelerates calcium absorption and is connected tightly with bone health [7]. More speci cally, the vitamin D receptor (VDR) is a steroid hormone nuclear receptor that regulates a variety of genes within a cell [8]. Vitamin D is hypothesized to prevent cancer relapse by inhibiting cell proliferation, angiogenesis, and metastasis while inducing apoptosis and differentiation [9][10][11][12][13]. Previous studies have reported that vitamin D intake is associated with a decreased risk of different cancers, including prostate [14] and breast [3] cancers, whereas there is no association between vitamin D and cancer types such as esophageal [15], pancreatic [16], skin [17], and gastric [18] cancers. Several observational studies [5,19] have also investigated the correlation between vitamin D and lung cancer prognosis, but results are not conclusive. However, the effect of vitamin D intake on the incidence of lung cancer remains elusive. For instance, vitamin D was inversely associated with lung cancer incidence in the Carotene and Retinol E cacy Trial (CARET) [20], but not in the Shanghai Women's Health Study (SWHS) [21]. Therefore, to address these issues and guide physicians, we conducted a comprehensive meta-analysis using the available data to evaluate the relationship between vitamin D and the prognosis and incidence of lung cancer.

Search strategy
This meta-analysis was conducted per the guidelines of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statement [22]. Two investigators independently assessed the observational studies and the quality of the randomized controlled trials (RCTs) via the Newcastle-Ottawa Quality Assessment Scale (NOS) checklist and the Cochrane Collaborations tool,, respectively. In the meta-analysis, systematic retrieval of databases (Pubmed, Web of Science, EBSCO, and Cochrane Library) was conducted from the beginning to November 2020. The subject search was bonded with free text terms and medical subject headings, including ((('lung' OR 'pulmonary') AND ('cancer' OR 'carcinoma' OR 'adenocarcinoma' OR 'squamous carcinoma' OR 'neoplasms')) OR ('non-small cell lung cancer' OR 'small cell lung cancer' OR 'NSCLC' OR 'SCLC')) AND ('vitamin D'). Also, literature references were manually retrieved to avoid omitting relevant studies.

Selection Criteria
Each study retrieved for inclusion was independently assessed by two authors. The inclusion criteria included: (1) studies conducted on human populations, (2) studies including RCTs, cohort, or case-control studies, and (3) reports on the correlation between vitamin D and the prognosis and incidence of lung cancer. Exclusion criteria included: (1) non-original articles, (2) articles with insu cient detailed data and irrelevant outcomes, (3) non-cancer diseases such as lung nodule, and (4) overlapping publications. Signi cantly, only the most recent, relevant, and comprehensive study was included to avoid duplicates or studies reporting data from the same population in the analysis.

Data Extraction And Study Quality Assessment
Two authors independently completed data extraction and study quality assessment. Uncertainties encountered between the two authors were resolved by consensus or by a third author. Two databases were used to record available information of the literature: a database of the correlation between vitamin D and the prognosis of lung cancer and a database of the correlation between vitamin D and the incidence of lung cancer. These indicators included primarily author(s), year of publication, country, cohort characteristics, sample size, time to recruit, age, follow-up time, vitamin D dose, study design, and adjustment factors.

Statistical analysis
Hazard ratio (HR), including a 95% con dence interval (CI), was used to evaluate the relationship between vitamin D and lung cancer prognosis. Odds ratio (OR), including 95% CI was used to evaluate the correlation between vitamin D and lung cancer incidence. Statistical analyses were conducted via STATA 12.0 Software (Stata, College Station). Cochran's Q statistic and I 2 statistics were applied to evaluate heterogeneity in the included studies. A random-effects model was utilized for meta-analysis on account of the variability of vitamin D type and dose. Publication bias was evaluated by Begg's and Egger's test, and P < 0.05 was considered statistically signi cant. Publication bias and sensitivity analysis were not performed on analysis subgroup with less than ten studies because of the low sensitivity of qualitative and quantitative tests [23].

Eligible studies and characteristics
A total of 1086 studies were screened after a database search. Additionally, four studies were identi ed via other sources. After deleting 329 duplicates, 761 records were browsed through reading article titles and abstracts. Also, 25 studies were evaluated via full-text articles. Finally, ten eligible articles were included in the quantitative synthesis (four studies [5,19,24,25] included the correlation between vitamin D and lung cancer prognosis, while six studies [20,21,[26][27][28][29] included the correlation between vitamin D and the incidence of lung cancer) (Fig. 1). The quality of ten studies was assessed, as shown in Supplementary Table 1. The characteristics of four studies on the correlation between vitamin D and lung cancer prognosis conducted on lung cancer patients are shown in Table 1Vitamin D intake was from dietary vitamin D and vitamin D supplementation. Adjustment factors were primarily gender, age, alcohol, body mass index, and smoking. Similarly, the characteristics of six studies investigating the correlation between vitamin D and the incidence of lung cancer based on human populations are indicated in Table 2. Total vitamin D intake was from dietary or supplement or both. Adjustment factors were primarily age, education, energy, body mass index, and smoking.

Vitamin D And Incidence Of Lung Cancer
Six studies, covering 225894 people, reported OR of lung cancer among vitamin D high intake compared with low intake. Additionally, there was no heterogeneity (I 2 = 0%). Due to the variability of vitamin D type and dose, a random-effects model was applied to the meta-analysis. The estimated pooled OR of lung cancer for high intake versus low intake of total vitamin D was 0.90 (95%CI: 0.83-0.97), indicating an association between increased total vitamin D intake and a small reduction in the risk of lung cancer (Fig. 4).

Publication Bias And Sensitivity Analysis
Considering that these studies selected were less than ten, Begg's and Egger's tests for publication bias and sensitivity analysis were not used for the metaanalysis due to the low e ciency of qualitative and quantitative tests.

Discussion
In the study, we conducted a comprehensive meta-analysis on ten articles. Four articles evaluated the correlation between vitamin D and lung cancer prognosis, while six articles examined the correlation between vitamin D and lung cancer risk. Signi cantly, vitamin D intake was related to a decline in lung cancer risk and an improved long-term prognosis in lung cancer patients. The relatively few relevant studies that observed the relationship between vitamin D intake and lung cancer prognosis and lung cancer incidence limited our ability to conduct subgroup analyses.
Vitamin D intake is derived from diet, supplements, and ultraviolet radiation from the sun synthesized on the skin. 1,25(OH) 2 D may exert its anticarcinogenic effects via stimulating the secretion of protein glues, such as E-cadherin and catenin, which make cells more adherent to each other [19]. It reduces the possible mobilization of several malignant cells into the lymphatic or blood circulation, invariably delaying lung cancer progression. Besides, 1,25(OH) 2 D inhibits the expression of cyclooxygenase-2 and prostaglandin synthesis, which can contribute to cancer cell growth and angiogenesis [30].
Vitamin D enhances the transcription of cathelicidin antimicrobial peptide genes, and the translation of CD14, a co-receptor for recognizing bacterial lipopolysaccharides. Both immensely improve the host body defense and are essential for congenital immunity [31]. The anticarcinogenic role of 1,25(OH) 2 D in regulating cell proliferation and angiogenesis are relevant to lung tumorigenesis. 1,25(OH) 2 D inhibits lung cancer signaling pathways, including mutations in epidermal growth factor receptor, Wnt-β-catenin dysregulation, and vascular endothelial growth factor [9,32].
Moreover, vitamin D promotes G1 cell-cycle arrest through signaling cyclin-dependent kinase inhibitors p21 and p27 [33]. Both p21 and p27 proteins closely cofunction with the ras oncogene family [34,35], and K-ras often mutate in adenocarcinoma [36]. Further, human lung tumor cell lines and animal experiments have demonstrated that the active metabolite of 25(OH)D-1,25(OH) 2 D suppresses angiogenesis and cancer cell growth by inhibiting the response to vascular endothelial growth factor [37].
The meta-analysis indicated that vitamin D intake could improve the long-term survival of lung cancer patients. A previous meta-analysis reported no signi cant association between blood vitamin D levels and lung cancer survival [38]. The study results above might be related to the vitamin D concentrations, which were too low (average value = 17.7 ng/ml) to observe any signi cant impact on the prognosis. Clinically, vitamin D levels below 20 ng/ml are considered de cient [39]. Besides, the disease stage was likely too advanced among patients selected for vitamin D levels to in uence prognosis.
It is worth noting that these studies focusing on serum levels of vitamin D on lung cancer progression might not be directly comparable to vitamin D intake. Vitamin D intake might not correlate with vitamin D levels, since sun exposure, dietary vitamin D, and supplementation source affect vitamin D serum levels. Nevertheless, vitamin D from sun exposure is limited in some countries situated above the latitude. Moreover, older patients have reduced dermal capacity to synthesize vitamin D and spend relatively large amounts of time indoors, while younger adults are more likely to use sun creams outdoors. In comparison, the present meta-analysis is the rst to investigate the correlation between vitamin D and lung cancer prognosis.
Notably, two cohort studies 5, [19] reported that vitamin D intake could signi cantly improve lung cancer patients' OS. However, the Women' s Health Initiative indicated that vitamin D has little effort on the OS of lung cancer patients [24]. Indeed, there is evidence that estrogens might play a role in lung adenocarcinoma in women [40]. Findings suggesting that estrogen receptors are present in lung tumors; however, playing a role in the activation of vitamin D and its receptor expression are provocative and warrant further exploration [41].
Simultaneously, two studies by Zhou et al. [19] and Akiba et al. [25] were con icting because the former highlighted that vitamin D intake could signi cantly improve the RFS of lung cancer patients, whereas the latter did not corroborate the result. Collectively, the present meta-analysis concludes that increased vitamin D intake could improve the prognosis of lung cancer patients. Thus, further research and more evidence are needed to prove if vitamin D intake can in uence lung cancer prognosis. A meta-analysis showed the inverse correlation between dietary vitamin D intake and lung cancer hazard, which was not investigated further if total vitamin D intake has a protective effect on lung cancer hazard [42]. Although another meta-analysis, which included only three RCTs, suggested that vitamin D intake cannot play a signi cant role in lung cancer [43]. In contrast, the present study showed that total vitamin D intake can protect against lung cancer occurrence. Therefore, it is pertinent to investigate further the correlation between total vitamin D intake and lung cancer risk.
Based on the strata of cohort studies, a study by Cheng et al. [20] compared high and low vitamin D intake found a negative association between vitamin D intake and lung cancer. However, it was discrepant with a report [28] by Redaniel et al.. The contradicting results between Redaniel et al. [28] and other studies [20,27] probably could be due to the different vitamin D intake sources, with the latter studies, including the dietary intake of vitamin D [20,27].
Moreover, the present trial showed that the total vitamin D intake has no protective effect on lung cancer incidence, which was likely because the present trial did not separate calcium from vitamin D in uence on lung cancer, as these agents were simultaneously used in the study [29]. An RCT by Lappe et al.
comparing Calcium (Ca) plus vitamin D supplementation and Ca supplementation suggested that vitamin D intake was associated with a decreased lung cancer risk. Herein, the present study result demonstrated that vitamin D intake protects against the risk of lung cancer [26].
Dietary supplements should be supplemented with foods containing vitamin D, such as deep-sea sh and animal liver. However, it is not enough for individuals vulnerable to vitamin D de ciency to only get vitamin D from their daily diet. Therefore, it is recommended to consider vitamin D nutritional supplementation. The Institute of Medicine (IOM) recommended 400 IU daily for children under one year of age, 600 IU daily for children one year and older and older adults between the ages of 70 years, and 800 IU daily for adults over 70 years of age, to achieve a blood level of at least 20ng/mL.
The Endocrine Society recommends a daily dose of 400-1000 IU for children under one year of age, 600-1000 IU for children one year and older, and a 1500-2000 IU daily dose for all adults to sustain a circulating serum level of 25(OH)D > 30ng/mL [39]. Nonetheless, more studies are needed to con rm these suggestions to regulate the safety range better. Though the safe dose range of vitamin D is vast, intoxication, or other harmful events following vitamin D intake must be considered. Notably, the dose-response relation was not consistent. Thus, the optimal dose of vitamin D usage still needs to be evaluated further via large-scale studies. Therefore, to delay cancer progression and reduce lung cancer risk, individuals are advised to take vitamin D supplements in conjunction with a healthy lifestyle.
Study strengths: To our knowledge, the study herein is the rst comprehensive meta-analysis investigating the role between vitamin D intake and the prognosis of lung cancer patients. Besides, the present meta-analysis contains relatively more comprehensive studies for investigating vitamin D's effect on the hazard of lung cancer, thereby updating and extending previous results to provide more robust support.
Signi cantly, the meta-analysis herein has few inherent potential limitations that should be taken into account. Firstly, vitamin D dose categories varied among the included studies, which was an inherent bias in the quantitative assessment. Secondly, even though we attempted to carry out a literature search as comprehensively as possible, we could not acquire unpublished articles, and the language was merely con ned to English. Finally, the included studies were nonetheless limited. Hence, we could not carry out sensitivity analysis on possible publication bias, which might have in uenced the study conclusions.
Therefore, the present study results should be interpreted with caution.

Conclusion
The meta-analysis herein shows that vitamin D intake can improve the prognosis of lung cancer patients. Additionally, vitamin D intake is inversely linked with the incidence of lung cancer. Notwithstanding, the meta-analysis herein still requires well-designed large-scale observational studies and RCTs to con rm the study results. Meanwhile, there is a need to lengthen the duration of follow-up and evaluate the optimal dosing of vitamin D usage.

Consent for publication
Not applicable.

Funding
The authors have no nancial support to declare.

Competing interests
The authors declare that they have no con ict of interest.

Availability of data and materials
The datasets supporting the conclusions of this article are included within the article.
Authors' contributions LLQ designed the research process. DG and MXQ searched the database for corresponding articles. SPQ and XJF extracted useful information from the articles above. RRF used statistical software for analysis. LLY drafted the meta-analysis. DG and MXQ polished this article. All authors had read and approved the manuscript and ensured that this was the case.