Correlation of several forms of folic acid with endometrial cancer: cross-sectional data from the National Health and Nutrition Examination Surveys (NHANES) 2011–2018

Endometrial cancer (EC) is a common malignancy of the female reproductive system and although most patients have a good prognosis, 20–30% of patients with advanced disease have a poor prognosis. There are currently no reliable biomarkers for early diagnosis and effective prognostic improvement of the disease. The purpose of this study was to explore the correlation between different forms of folic acid and endometrial cancer. This study included 8809 female subjects aged ≥ 20 years in the NHANES database from 2011 to 2018, including 8738 non-oncology patients and 71 EC patients. Selection bias was reduced using 1:1 propensity score matching (PSM) method. Restricted cubic spline (RCS) was plotted to explore the non-linear relationship between different forms of folic acid and EC. Using data from the NHANES database from 2011 to 2018, the association between folic acid and the risk of developing EC was assessed. The results of the 1:1 ratio propensity score matching (PSM) showed 68 each for EC patients and non-oncology participants. Total serum folate, 5-methyltetrahydrofolate (5-methylTHF), 5-formyltetrahydrofolate (5-formylTHF), tetrahydrofolate (THF) and 5,10-methylenetetrahydrofolate (5,10-methenylTHF) were significantly correlated with EC (p < 0.05). In addition, the RCS showed a significant non-linear correlation between THF and 5,10-formyl THF and the risk of developing EC. The results of this study showed that changes in serum total folate, 5-methylTHF, 5-formylTHF, THF and 5,10-methenylTHF were related to EC.


Introduction
Endometrial cancer (EC), a common malignancy of the female reproductive system, is increasing in incidence and mortality every year (Kalampokas et al. 2022;Shen et al. 2023).In 2020, 417,367 newly diagnosed cases were reported globally, accounting for 2.2% of all newly diagnosed malignant tumors.EC causes 97,370 deaths worldwide, accounting for 1% of all deaths from malignancies (Koppikar et al. 2023).Most EC patients are in the early stages of diagnosis (International Federation of Gynecology and Obstetrics FIGO stage I or II) and have a good prognosis after surgery only, with a 5 year survival rate of over 90% (Crosbie et al. 2022;Braun et al. 2016).For patients in FIGO stages III and IV, it is difficult to apply effective individualized treatment regimens, with 5 year overall survival rates falling to 47-69% and 15-17% (Wang et al. 2021), and 5 year survival rates for recurrent patients being even lower than 20% (Clarke et al. 2018;Murali et al. 2014).The identification of risk factors for EC and subsequent interventions targeting the risk factors, particularly in high-risk groups, is essential to improve survival in EC patients.
Folic acid (FA) is a water-soluble B vitamin consisting of pterin, p-aminobenzoic acid and glutamic acid (Wang et al. 2022;Ratajczak et al. 2021).As the body does not have the ability to synthesize folic acid, it must be obtained through diet or supplements (Pieroth et al. 2018).Serum folate levels reflect recent intake, with red blood cell (RBC) folate representing folic acid status over several months.The WHO recommends a threshold value of 340 nmol/L for erythrocyte folate and 10 nmol/L for serum folate (Gool et al. 2018).Elevated urinary formyl glutamate excretion and deoxyuridine inhibition tests are also used to assess folic acid status.Elevated homocysteine is a functional indicator of folic acid status (Menezo et al. 2022).Physiologically folic acid enters cells by endocytosis using cell membrane-associated proteins or the folate receptor (FOLR) (Narmani et al. 2019).Folic acid is a component and catalyst of fundamental biochemical reactions and plays an important role, particularly in DNA methylation, synthesis and repair (Oaks et al. 2010).As a result, rapidly proliferating cells such as intestinal cells, haematopoietic cells and tumor cells consume high levels of folic acid to meet their needs for newly synthesized nucleotides and for DNA replication and gene expression (Kanarek et al. 2020).Folic acid deficiency can lead to DNA hypomethylation, as well as hypermethylation of the promoter region of oncogenes, which in turn activates proto-oncogenes and suppresses the expression of oncogenes.Secondly, folic acid deficiency leads to DNA strand breaks, enhanced mutation rates and impaired DNA repair mechanisms.Folic acid deficiency can also inhibit the proliferation of CD8 + T lymphocytes and reduce the body's ability to clear tumor cells (Smith et al. 2008).However, excess folic acid can also promote the synthesis of DNA in rapidly proliferating cells, thereby promoting the growth and progression of tumors in situ (Field and Stover 2018).One study showed that people with high serum folate concentrations (> 23.61 ng/mL) had 4.86 times the risk of colorectal cancer than those with low folate levels (≤ 8.86 ng/mL) (Crider et al. 2012).Therefore, folic acid may play a dual role in the occurrence and development of tumors, which can prevent the occurrence of tumors in the early stage, but promote cancer transformation after the establishment of precancerous lesions, which may be related to time and folic acid dosage (Wang et al. 2020).There is a lack of evidence for a correlation between high or low levels of folic acid and EC tumor growth.

Research population
The NHANES database (https:// wwwn.cdc.gov/ nchs/ nhanes/) is a population-based cross-sectional survey designed to collect information on the health and nutritional status of adults and children in the U.S. The survey is unique in that it combines interviews and physical examinations.The database includes demographics data, dietary data, examination data, laboratory data, questionnaire data, limited access data.The survey approach uses a complex, multi-stage probability sampling design.The survey results will be used to determine the prevalence of major diseases and risk factors for the diseases.Using information from the NHANES database, this study aimed to investigate the association between erythrocyte folate and different serum folate forms and the risk of EC.
Data from participants in the NHANES database from 2011 to 2018 were selected for data collection in 2 year cycles: 2011-2012, 2013-2014, 2015-2016 and 2017-2018, including a total of 39,156 participants.19,848 women were included, excluding those with missing covariate data and folic acid concentration data, resulting in the inclusion of 8,809 participants, including 8,738 non tumor patients and 71 EC patients.A flow chart of the screening process is shown in Fig. 1.

Folic acid concentration determination
RBC folic acid was determined using a microbiological assay.Total serum folate was determined by liquid chromatography-tandem mass spectrometry, as well as the concentrations of different forms of folic acid in serum, namely 5-methylTHF, UMFA, 5-formylTHF, THF, 5,10-methe-nylTHF and MeFox.

Malignancy status
Malignancy status was assessed by two consecutive questions: first, if a participant answered "Yes" to the question "Did your doctor or other health professional tell you that you have a malignancy or any type of malignancy?" he/she was classified as having a malignancy.Next, participants will be asked "What kind of malignancy?" and the type of malignancy can be identified.The patients included in the study refer to individuals who were currently or previously diagnosed with EC.

Statistical methods
All analyses in this study were performed using State 26.0 (SPSS Inc., Chicago, IL, USA) and R 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria.http:// www.Rproje ct.org/).Categorical variables in the baseline characteristics were described using compositional ratios, and comparisons between groups were made using Chi-square tests or exact Fisher tests.Folic acid concentration was a continuous variable and was first tested for normal distribution and variance, with normality expressed as mean ± standard deviation ( x±s) and non-normality expressed as median (quartiles), and comparisons between groups were made by independent samples t-test or rank sum test.Propensity score matching (PSM) has been used extensively in observational studies to reduce selection bias for matching EC patients with non-oncology participants.A 1:1 ratio PSM analysis was used to balance differences between EC patients and non-oncology participants, adjusted for confounding variables including: age, race, marital status, education, BMI, hypertension, diabetes, smoking, alcohol consumption, age at menarche, age at last menstruation, oral contraceptive use and use of hormonal drugs.Propensity matching was performed by the R software, resulting in 68 successful matches.Correlation analysis of different serum folic acid forms and EC prevalence was performed using matched data.In addition, restricted cubic spline (RCS) with four sections was used to explore the dose-response relationship between different folic acid concentrations and EC.p < 0.05 indicates a statistically significant difference.

Participant baseline characteristics
The NHANES database was searched and a total of 8809 eligible participants were included, including 8738 nononcologists and 71 EC patients.Of the EC patients, 91.5% were 44 years of age or older, 32 (45.1%) were non-Hispanic white, and 39.4% were married.There were 47 (66.2%) with hypertensive disorders, 16 (22.5%)with diabetic disorders, 28 (39.4%)under the age of 40 years at menopause, 34 between 40 and 55 years, and 26 (36.6%) on hormonal drugs (Table 1).The NHANES database uses complex multi-stage sampling, with varying probabilities of each person being selected, and the sampling data at each stage are not independent.Following NHANES' analysis and reporting guidelines, they were weighted.Some of the variables included in this study were collected in the MEC and were therefore analyzed using the MEC examination weights (WTMEC2YR).The sample weight used in the final analysis was equal to one quarter of the "WTMEC2YR" value, corresponding to the MEC weights for the 8 survey years (Zhang et al. 2023).The final weighted sample of 8809 participants representing 94,380,076 was analyzed and the weighted results showed statistically significant differences (p < 0.05) between the two groups for age, race, marital status, hypertension, age at menopause and use of hormonal medication (Table 2).

Propensity score matching
To eliminate the effects of confounding factors, a 1:1 match was performed using PSM, as shown in Fig. 2. Sixty-eight each of EC patients and non-oncology participants were included, and the clinical baseline characteristics of all subjects after PSM are shown in Table 3. Subsequently, comparison of the different forms of folate with EC showed that total serum folate, 5-methylTHF, 5-formylTHF, THF and 5,10-methenylTHF were significantly correlated with EC (p < 0.05) (Table 4).

Restricted cubic spline
To further test the existence of a correlation between different folic acid concentrations and EC, an RCS visual depiction was used (Fig. 3).The results showed that total serum folate, 5-methylTHF, 5-formylTHF, THF and 5,10-methe-nylTHF correlated with EC (p < 0.05).Among them,5-formylTHF, THF and 5,10-methenylTHF had significant non-linear correlation with EC with P values of 0.005,0.045and 0.005 for non-linear test respectively.There was no significant non-linear relationship between total serum folate and 5-methylTHF and EC, with P values of 0.524 and 0.562 for the non-linear test respectively.However, the 5-formylTHF data could not be curve plotted.For THF, EC prevalence increased with increasing concentration at concentrations less than 4.172 nmol/L; at concentrations equal to 4.172 nmol/L, the cut-off point (Odds ratio, OR) ≈ 1; at concentrations greater than 4.172 nmol/L, EC prevalence decreased significantly with increasing concentration.For 5,10-methenylTHF, EC prevalence increased with increasing concentrations at concentrations less than 0.219 nmol/L; at concentrations equal to 0.219 nmol/L, the OR of the cut-off point was ≈ 1; after concentrations greater than 0.219 nmol/L, EC prevalence decreased significantly with increasing concentrations.

Discussion
EC is a common malignancy of the female reproductive system, predominantly adenocarcinoma, and the prognosis is influenced by a number of factors (Raglan et al. 2019).
Most patients can be treated by early diagnosis surgery and the overall prognosis is more positive (Reijntjes et al. 2022).However, in patients with advanced disease, where treatment options are limited and prognosis is poor, it is important for early identification of high-risk factors associated with poor outcomes (Dizon 2010).In recent years, studies have found that folic acid plays a key role in tumor growth (Gangopadhyay et al. 2021).The folate receptor mediates the entry of folate into the cytoplasm of human eukaryotic cells and is expressed in elevated amounts in many types of tumors (Han et al. 2020).Folic acid, a small molecule ligand with high affinity for the folate receptor, can specifically recognize folate receptors highly expressed on tumor cells for targeted delivery (Koirala et al. 2019;Yu et al. 2022;Abbasi et al. 2023).Thus, folic acid can be used not only as a predictive marker for tumors, but also as a target for treatment and visualization.
It has been shown that low folate status leads to genomic instability and DNA mutation rates, and affects the methylation patterns of oncogenes (Du et al. 2022).In contrast, increased levels of folic acid accelerate the growth of tumors at the primary site (Cao et al. 2023).There is a lack of evidence for a correlation between high or low levels of folic acid and EC tumor growth (Peng et al. 2016;Oliai et al. 2019).This study explored the potential relationship between RBC folate and different serum folate forms and EC by combining and analyzing NHANES data from 2011 to 2018, including 8809 participants.Fan et al. showed by a dose-response analysis that each 100 ug/d increase in folic acid intake was associated with a 4.3% reduction in the risk of head and neck squamous cell cancer (Fan et al. 2017).Li et al. included nine case-control studies and five cohort studies for a meta-analysis and showed that increased folic acid intake led to a decreased risk of EC (OR = 0.89, 95% CI 0.76-1.05)(Hanley et al. 2021).The serum total folate concentration is usually considered to be an indicator of recent folate intake, while RBC folate concentration is considered to be an indicator of long-term folate status.In this study, no correlation was found between RBC folate and EC (P > 0.05) and serum total folate showed a negative correlation with EC prevalence.For different subtypes of folic acid, the concentration of 5-methylTHF is the largest and constant contributor to total serum folic acid (Pfeiffer et al. 2015a), and it directly participates in the metabolism of a carbon unit as the main cycling form of folic acid (Kaye et al. 2020).This study also demonstrated a negative correlation between 5-methylTHF and the incidence of EC.UMFA may occur in persons consuming folic acid-fortified foods or supplements and impair folic acid metabolism by inhibiting dihydrofolate reductase (Pfeiffer et al. 2015) and methylenetetrahydrofolate reductase (Rees et al. 2017;Obeid and Herrmann 2012;Pfeiffer et al. 2015c;Williams et al. 2021).No correlation between UMFA and EC was found in this study.Wei et al. also found that higher levels of 5-methylTHF were associated with a lower risk of developing lung cancer (Wei et al. 2023).Yang et al. investigated the association of different serum folate forms (total serum folate, 5-methylTHF, and UMFA) with the prevalence of non-alcoholic fatty liver disease and advanced fibrosis by including NHANES data from 2011 to 2018.The results showed that total serum folic acid and 5-methylTHF were negatively correlated with the prevalence of nonalcoholic fatty liver disease and advanced fibrosis (p < 0.05).The concentration of UMFA was positively correlated with the prevalence of non-alcoholic fatty liver disease and advanced fibrosis (p < 0.05) (Yang et al. 2023).THF metabolic pathway has long been a target for anti-tumor therapy (Ragsdale 2008;He et al. 2019), which is consistent with our research results.MeFox is an oxidation product of 5-methyltetrahydrofolate, and there may be a correlation between these two forms.This may indicate that factors exceeding the amount of circulating 5-methylTHF may affect the production of MeFox.However, this study found no correlation between the increase in serum MeFox concentration and EC (Zhang et al. 2015).In conclusion, using data from a large national representative cohort of American adults, the study found that high levels of folic acid reduced the incidence rate of EC, and that non methyl folic acid affected the incidence of EC in a non-linear mode.This study may provide guidance for the study of folic acid in EC.
The advantage of this study was to use the NHANES database, using standardized measurement methods, to better study the association between RBC folate and different serum folate forms (serum total folate, 5-methyl-THF, UMFA, 5-formylTHF, THF, 5,10-methenylTHF and MeFox) and EC by adjusting for potential confounding factors.However, this study also had certain limitations.First, cross-sectional design limited the ability to assess causal relationships; Second, the study was conducted in American adults receiving folic acid fortification; the dietary habits and appetite of cancer patients have undergone significant changes due to their condition, which may affect the folate levels they consume from their diet.Third, because the study was exploratory, multiple tests were not considered.

Conclusion
Analysis of NHANES data from 2011 to 2018 showed that serum total folate, 5-methylTHF, 5-formylTHF, THF and 5,10-methenylTHF in hemorrhage serum were related to the occurrence of EC.The incidence of EC decreased with the increase of serum total folate and 5-methylTHF concentrations.THF and 5,10-methenylTHF had significant non-linear correlations with EC.It was helpful for clinicians to better

Table 2
Baseline characteristics of NHANES participants over the period2011-2018 (weighted)

Table 3
conduct quantitative treatment for EC patients and improve their prognosis.