Vitamin D status and its relation with insulin resistance and VDR-FokI polymorphism in Iranian non-melanoma skin cancer (NMSC) patients: a case-control study

Background: Sunlight exposure, the main source of endogenous vitamin D synthesis, may increase the risk of non-melanoma skin cancer (NMSC) development. Vitamin D receptor (VDR) polymorphisms are associated with 25(OH)D levels, cancer development and insulin resistance. This study was aimed to examine the associations among vitamin D status, VDR FokI polymorphism, insulin resistance and NMSC. Methods: This case-control study included 73 diagnosed cases of NMSC and 72 healthy controls from dermatology clinics at Razi Hospital, Tehran, Iran. A questionnaire was used to assess sunlight exposure. The extracted DNA from whole blood samples were genotyped. Fasting serum 25-hydroxyvitaminD (25(OH)D)), lipid profile, glucose, and insulin were measured. To evaluate insulin resistance, HOMA-IR formula was used. Results: We found a significant higher duration of cumulative sunlight exposure in cases compared with controls (p<0.001). However, 25(OH)D concentrations were not significantly different between cases and controls (30±15 vs. 29±15 ng/mL, p=0.78). Higher levels of insulin (p = 0.004) and HOMA-IR score (p= 0.019) were observed in Ff and ff genotype of FokI. We did not observe any significant increased risk of NMSC due to f allele, as compared with FF (OR =2.33, 95% CI 0.81-6.75, p=0.12). The components of lipid profile, fasting serum glucose, iPTH and anthropometric measures did not differ significantly across VDR genotypes. Conclusion: In conclusion, sunlight exposure was associated with NMSC risk. VDR FokI polymorphisms appears to influence insulin resistance in the NMSC patients. In conclusion, our results suggest that VDR FokI polymorphisms are not significantly associated with the development of NMSC in Iranian subjects. However, this lack of association might be due to our study’s small sample size and low genetic power to detect disease susceptibility. The present study suggests an interaction between VDR FokI polymorphism and measures of insulin resistance and circulating 25(OH)D. FokI VDR polymorphisms may be linked to insulin resistance and might represent a genetic determinant for developing MetS in NMSC patients. These findings have clinical implications. Some studies revealed that components of MetS including raised blood pressure and deranged glucose and lipid metabolism may contribute in development of skin cancers (60) and also treatment of MetS can improve skin conditions (61). Therefore, weight control, having a healthy diet and improvement of vitamin D status may all help NMSC prevention. Further studies on large population are essential, besides the genome wide association studies to determine the direct effect of VDR polymorphisms on insulin resistance.


Introduction
Suboptimal vitamin D status reported from many countries in different age and sex subpopulations is now considered a serious global health problem (1). The importance of vitamin D deficiency (VDD) is not confined to the deleterious effects on musculoskeletal system. A growing body of evidence shows the predisposing effect of VDD in many human pathologies including cardiovascular disease, diabetes, autoimmune disorders (2) and certain cancers (3)(4). Skin cancers are among the most common types of human neoplasms (5). In Iran the incidence of skin cancers is increasing (6) comprising approximately 15% of all cancers (7).

Exploration of vitamin D receptor (VDR) on a vast variety of tissues and cells including
basal cell and squamous cell carcinomas indicated new roles for the vitamin (8). Though several lines of evidence indicate a protective role for vitamin D against colorectal, breast, prostate and pancreatic cancers (9) , the association of vitamin D and skin cancers is intriguing. While the main natural source of vitamin D is solar ultra violet beam (UVB), direct exposure to the same wavelength of UVB has been known as the major culprit of skin cancers (10). It is, therefore, speculated that subjects with skin cancers may have longer periods of sun exposure and hence higher vitamin D status than the general population (11).
The possible link between insulin resistance (IR) and human cancers, including skin malignancies, can be a new argument (12). The associations between IR and several malignancies including colon, liver, pancreas (13), endometrium (14), breast (15), lung (16) and thyroid cancers (17) have been reported. The ameliorating effect of vitamin D on IR has been shown by clinical trials (18). Whether this effect can mediate any anti-cancer property of vitamin D is still unknown.
Despite previous studies on relationship between 25(OH)D3 and IR, the molecular mechanisms of the role of VDR in insulin resistance remains to be determined (19-21).
The limited evidence available suggests that two common VDR gene polymorphisms (BsmI and FokI) might have an effect on BMI, insulin resistance and serum HDL-cholesterol (21-24). However, little information is available regarding this association in cancer patients.
On the other hand, the nutrigenetic effect of VDR polymorphisms on response to vitamin D intake has been already documented (25). VDR polymorphism is characterized by altered expression levels leading to decrease or increase in vitamin D activity (26-27). FokI (rs2228570) polymorphism is the only known polymorphism that generates an altered protein (28). VDR polymorphisms have been linked to several cancer risks including prostate, breast, bowel (29-30) and skin malignancies (31-32). However, the associations of VDR polymorphisms with skin cancer risk has not been understood clearly yet (33).
Data from a meta-analysis suggest that there might be a possible positive link between VDR FokI (rs2228570) and BsmI (rs1544410) polymorphisms and cutaneous malignant melanoma (CMM) and non-melanoma skin cancer (NMSC) risks (34). However, in a systematic review no significant association was observed between TaqI (rs731236) ,

Participants and Clinical Samples
This case-control study included 65 participants with a BCC or SCC within 3 months of diagnosis who were visited at the dermatology clinics at Razi Hospital of Tehran University of Medical Sciences in a period between September 2016 and April 2018. The controls were composed of 65 unrelated healthy volunteers who were matched for age and sex to the patients with NMSC (Table I). All participants completed a questionnaire including information on marital status, education level, medication and supplement use history, disease history, sunscreen use and mean hours of daily sun exposure. Participants who were taking any nutritional supplement including vitamin D, calcium, omega-3 and antioxidant, or medications that modify vitamin D metabolism (corticosteroids, estrogens and calcitonin) for at least 3 months preceding the study, history of any other cancers and renal or liver diseases were excluded from the study. Fasting blood samples were obtained from all participants for DNA genotyping and biochemical analyses. The study procedures were approved by the Research Council and the Ethical Committee of the National Nutrition and Food Technology Research Institute (NNFTRI), respectively. All participants signed a written informed consent form. Hip circumference (HC) and waist circumference (WC) were both measured by a measuring tape to the nearest 0.1 cm. WC was measured at the approximate midpoint between the lowest rib and iliac crest after a normal expiration. Hip circumference (HC) was also measured at the level of the greater trochanters. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured using a digital sphygmomanometer (BC08; Beurer GmbH, Ulm, Germany).

Glycemic status and lipid profile
Fasting serum glucose, total cholesterol, triglycerides (TG), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) were determined using enzymatic colorimetric methods (all from Pars Azmoon, Tehran, Iran). Serum insulin was measured by means of an enzyme immunoassay (EIA) kit (Demeditec Diagnostics GmbH, Kiel, Germany). Homeostasis model assessment of insulin resistance (HOMA-IR was used as an index of insulin resistance and was calculated by the following formula (37): HOMA-IR= ((fasting insulin (mU/mL) x fasting glucose (mg/dL))/405)

Serum calcidiol and iPTH measurements
Serum 25(OH)D and iPTH were assayed using EIA kits (both from Euroimmun, Medizinische Labordiagnostika AG, Germany). Subjects were categorized as vitamin D deficient if 25(OH)D concentration was below 20 ng/ml, insufficient with concentrations 20-29 ng/ml and sufficient with concentrations of at least 30 ng/ml (38-41).

DNA Extraction and Genotyping
Genomic DNA was isolated from whole blood samples using PrimePrep Genomic DNA isolation kit (GeNet Bio, Daejeon, South Korea) according to the manufacturer's protocol.
For VDR FokI polymorphism (rs2228570) the forward primer was 5′-GTCAAAGTCTCCAGGGTCAG -3′, and the reverse primer used was 5′-GCCTGCTTGCTGTTCTTAC -3′. Genotyping was done by high-resolution melting (HRM) assay initial denaturation-activation step at 95°C for 15 min, followed by a 40-cycle program (denaturation at 95°C for 15 s, annealing at 61°C for 20 seconds, 72°C for 20 seconds) and HRM step from 60 to 95°C rising at 0.1°C per second. Curves for each duplicate were checked on the shape and peak height to meet reproducibility. Normalized and temperature-shifted melting curves from HRM, suggestive of SNP, were distinguished, and direct Sanger sequencing was used to confirm genotyping results from the samples.

Statistical analyses
Data were expressed as mean±SD for continuous variables and frequencies for categorical variables. Normality of data was checked using Kolmogrov-Smirnov test. For betweengroup comparison of variables, independent sample t test, Mann-Whitney, or χ 2 tests were used when appropriate. Means of variables were compared among the different polymorphism groups using ANOVA or Kruskal-Wallis tests for data with normal or nonnormal distribution, respectively. Tukey's HD correction for multiple comparisons was applied, as required. The associations between VDR FokI polymorphism and risk of NMSC were estimated by computing odds ratios (ORs) and 95% confidence intervals (CIs) using logistic regression analyses. The Hardy-Weinberg equilibrium (HWE) was tested by a goodness-of-fit chi-squared test to compare the observed genotype frequencies with the expected frequencies among controls. All statistical analyses were done with SPSS software IBM SPSS Statistics version 23.A two-tailed value of less than 0.05 was considered statistically significant.

Characteristics of the study participants
Overall, 145 participants including 73 NMSC patients and 72 healthy controls were enrolled ( Table 1). The mean age of the patients and controls was 56±9 and 58±8 years, respectively, men consisted the greater proportion of both study groups. Compared with the healthy controls, NMSC subjects were significantly less educated (p=0.006).
Anthropometric measures including waist circumference (p=0.016), hip circumference (p=0.02), percentage of visceral fat (p=0.008) and blood pressure (p=0.001) were all significantly greater in cases than in controls. However, BMI did not statistically differ between two groups (p=0.8).
Despite significantly longer duration of sun exposure (including occupational exposure) in cases than in controls (p<0.001), there was no significant between-group difference in concentrations of circulating 25(OH)D and distribution of vitamin D status (Table 2).
Though fasting serum glucose showed no significant between-group difference, serum concentrations of insulin as well as HOMA-IR were both higher in NMSC group than in controls. Other variables did not show any significant between-group difference.
Our analyses revealed that the frequencies of VDR variants in our study population were in HW equilibrium (chi 2 value = 5.094, p=0.07 for control). The genotypes of FF, Ff and ff were 57%, 26%, 16% in NMSC cases and 68.1%, 23%, 8% in healthy controls, respectively.
The results of the evaluation of the association between VDR FokI polymorphism (rs2228570) and NMSC risk are demonstrated in Table 3. Logistic regression analysis did not show any significant increased risk of NMSC due to f allele (ACG codon), as compared with FF (ATG codon).
On comparison of different variables among VDR FokI variants, the measures of serum insulin and HOMA-IR were significantly higher in Ff and ff than in the FF genotype (Table   4). WHR values were higher in ff than in Ff and FF variants (p=0.032). There was no significant between-FokI-variant difference in other variables.

Discussion
We found a significant association between cumulative (including occupational) sunlight exposure and increased risk of NMSC. However, no significant difference in circulating concentrations of calcidiol between NMSC subjects and their healthy counterparts was detected. A great body of evidence indicates that chronic sunlight exposure induces most non-melanoma skin cancers (42). It is roughly estimated that only 25% of life span UV exposure occurs before the age of 18 (43). We found no significant association between VDR FokI polymorphisms (rs2228570) and NMSC risk. Accordingly, in a nested case-control study within Nurses' Health Study, there was no significant association between FokI ff genotype and skin cancers whereas BsmI BB variant was associated with increased risk of squamous cell carcinoma (SCC) (51). In contrast with this report, a meta-analysis suggested a possible role for polymorphisms of VDR FokI and BsmI in relation to skin cancer risks including malignant melanoma and NMSC (34). Altogether, previous studies investigating the association of VDR polymorphisms with NMSC risk have generated controversial results which can be explained by genetic variations in populations and types of data collected. Several studies have examined the possible interactions between VDR polymorphisms and vitamin D status (25, 52). However, we found no significant difference in circulating calcidiol concentrations among VDR FokI variants. But interestingly, Fok-I Ff and ff SNPs were related to increased insulin resistance in the whole study population. Furthermore, these genotypes were related to higher blood pressure, BMI, WC, total cholesterol and LDL-C in both NMSC and control participants. Our findings are consistent with previous reports (53).
VDR FokI polymorphism has been previously found to be associated with anthropometric and biochemical components of metabolic syndrome (MetS), as in a study VDR BsmI polymorphism was associated with BMI whereas FokI VDR polymorphism was related to insulin sensitivity and serum HDL cholesterol in men (22). In a case-control study, FokI To the best of our knowledge, this is the first report of a possible association between Fok-I VDR gene polymorphism and IR in a NMSC population. As IR and type 2 diabetes are caused by a set of interplays between genetic and lifestyle, large-scale, population-based studies are essential to further explore the relationship between this SNP and IR and its association with skin cancer development.
Some limitations of this study are acknowledged. Participants in a hospital-based casecontrol study may not be representative of the whole community. Secondly, comparison of the results from different studies for association between sunlight exposure and skin cancer could be very difficult because of the various methods used to estimate sunlight exposure. In case-control studies recall bias occurs because cases tend to recall past exposures more accurately than controls. Moreover, self-reporting of sunlight exposures in the past may result in measurement error due to difficulty in recalling sun exposure habits. Like some other studies, some important NMSC risk factors, including family history and skin type of white subjects were not evaluated. However, some studies reported that Iranian skin types, based on Fitzpatrick scale, can be classified as "lightly pigmented", i.e. skin types I to IV from north to south, respectively, based on the Fitzpatrick scale (57)(58) and there is no relationship between skin type and minimal erythermal dose (59). We did not examine any possible association between NMSC and other VDR polymorphisms, including BsmI.

Conclusion
In conclusion, our results suggest that VDR FokI polymorphisms are not significantly associated with the development of NMSC in Iranian subjects. However, this lack of association might be due to our study's small sample size and low genetic power to detect disease susceptibility. All participants signed a written informed consent form.

Availability of data and materials
The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

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

Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.

Authors' contributions
TN and FR designed the study. FR, AK, NS and MZ contributed to sample collection and performed the experiments. BN, TN and FR analysed the data. FR and TN wrote the paper with input from all authors.