Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Sep 28.
Published in final edited form as: J Invest Dermatol. 2023 Feb 20;143(8):1538–1547. doi: 10.1016/j.jid.2023.01.028

Significant association of Poly-A and Fok1 polymorphic alleles of the vitamin D receptor with vitamin D serum levels and incidence of squamous cutaneous neoplasia

Taylor A Bullock 1,*, Judith A Mack 2,*, Jeffrey Negrey 4, Urvashi Kaw 1, Bo Hu 3, Sanjay Anand 1,2,5, Tayyaba Hasan 6, Christine B Warren 1,5, Edward V Maytin 1,2,5,6
PMCID: PMC10439970  NIHMSID: NIHMS1913173  PMID: 36813159

Abstract

Vitamin D3, a prohormone, is converted to circulating calcidiol and then to calcitriol, the hormone that binds to the Vitamin D receptor (VDR, a nuclear transcription factor). Polymorphic genetic variants of the VDR are associated with increased risk of breast cancer and melanoma. However, the relationship between VDR allelic variants and risk of squamous cell carcinoma (SCC) and actinic keratoses (AK) remains unclear. We examined associations between two VDR polymorphic sites, Fok1 and Poly-A, versus serum calcidiol levels, AK lesion incidence, and history of cutaneous SCC in 137 serially enrolled patients. By evaluating Fok1 (F) and (f) alleles and the Poly-A long (L) and short (S) alleles together, a strong association between genotypes FFSS or FfSS and high calcidiol serum levels (50.0 ng/mL) was found; conversely, ffLL patients showed very low calcidiol levels (29.1 ng/mL). Interestingly, the FFSS and FfSS genotypes were also associated with reduced AK incidence. For Poly-A, additive modeling demonstrated that Poly-A (L) is a risk allele for SCC, with an odds ratio of 1.55 per copy of L allele. We conclude that AK and SCC should be added to the list of squamous neoplasias that are differentially regulated by the VDR Poly-A allele.

INTRODUCTION:

Non-melanoma skin cancers (NMSC), comprising basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), account for ~80% and ~16% of all cutaneous cancers; malignant melanoma represents ~4% (Guy et al., 2015, Ridky, 2007, Rogers et al., 2015). SCC, the second most common skin cancer, carries a significant risk for metastasis (Burton et al., 2016). Actinic keratoses (AK) are common precancerous skin lesions (Siegel et al., 2017) that can progress to invasive SCC (Ratushny et al., 2012). Mechanisms of ultraviolet light (UV)-induced carcinogenesis, involving oncogenic DNA mutations that initiate neoplasia and progression to SCC, have been heavily studied for years (Cleaver and Crowley, 2002, Shah and He, 2015, Yu et al., 2014, Ziegler et al., 1994), yet a complete understanding to facilitate prevention and cure remains elusive.

Vitamin D signaling is an important determinant of cutaneous carcinogenesis (Bikle, 2020, Tang et al., 2012b). Vitamin D3 (D3; cholecalciferol) is a natural vitamin and prohormone, obtainable through the diet or produced in the skin by a reaction catalyzed by sunlight. To exert biological effects, D3 must first be converted via hydroxylation in the liver to 25(OH)D3 (calcidiol), and then via a second hydroxylation step to 1,25(diOH)D3 (calcitriol) in the kidney (Bikle, 2000) or in skin keratinocytes (Reichrath et al., 2017). Calcitriol binds to the vitamin D receptor (VDR), which then translocates to the nucleus and binds to regulatory regions of genes whose expressed proteins govern calcium metabolism, cell growth, differentiation, and other processes crucial to cancer development (Bikle, 2000, Reichrath et al., 2017). Evidence suggests that endogenous levels of vitamin D might affect the initiation of AK and its progression to cutaneous SCC (Song et al., 2013, Tang et al., 2012b). Likewise, there is evidence that vitamin D levels can affect the response of cutaneous neoplasia to therapy. For example, 25(OH)D3 can serve as a predictor of AK responsiveness to photodynamic therapy (PDT) (Bullock et al., 2022, Moreno et al., 2020). Administration of high dose oral D3 prior to PDT, which raises the 25(OH)D3 and 1,25(diOH)D3 levels (Anand et al., 2014), facilitates an enhanced therapeutic response (Bullock et al., 2022).

Nearly all actions of vitamin D are mediated by binding to VDR, a protein of 427 amino acids encoded by exons 2–9 of the human VDR gene (Pike and Shevde, 2005, Whitfield et al., 2005); see Fig. 1. Activity of the VDR is dependent upon binding to its ligand, calcitriol. While some ligand-bound VDR migrates to the plasma membrane and mediates calcium fluxes and cell signaling (non-genomic actions), most of VDR’s actions are mediated by its translocation to the nucleus where it binds to promoter/enhancer regions of genes that regulate calcium metabolism, growth, differentiation, and other functions which impact skin carcinogenesis (Reichrath et al., 2017, Whitfield et al., 2005). When considering VDR activity, it is important to consider genetic polymorphisms within the VDR gene that could affect VDR expression and functionality. The human VDR gene is very large, with over sixty single-nucleotide polymorphisms (SNPs) identified within or near exons 2 to 9 (Denzer et al., 2011, Morrison et al., 1992, Uitterlinden et al., 2004). The most commonly studied SNPs are Fok1, Bsm1, Apa1, and Taq1 (see Fig. 1). Bsm1 and Apa1 are located within the last intron of the gene, and Taq1 is located in exon 9. Due to their close proximity, Bsm1, Apa1, and Taq1 are in linkage disequilibrium and can be regarded as genetically linked (Morrison et al., 1994, Verbeek et al., 1997). Fok1, on the other hand, is located at the 5’ end of the gene within exon 2. Fok1 is the only variant known to alter the physical structure of the VDR protein, because it affects the translation start site (Arai et al., 1997). The f variant (associated with Fok1 enzyme restriction cleavage) introduces a new ATG residue (encoding methionine) located 9 nucleotides upstream of the original start site, yielding a VDR protein 3 amino acids longer than the protein associated with the F variant.

Figure 1.

Figure 1.

Open in a new tab

Genomic structure of the human VDR. (Adapted from (Whitfield et al., 2005)).

Another polymorphic site called “Poly-A,” a microsatellite of repeated adenosines, is located within the 3’ untranslated region (3’ UTR); it actually has no relationship to mRNA polyadenyation) (Whitfield et al., 2001). Increasing evidence continues to support a strong association of Poly-A alleles with carcinogenesis, in several tumor types (see Discussion). The Poly-A and the Bsm1 sites are in almost perfect linkage disequilibrium (Bretherton-Watt et al., 2001, Guy et al., 2004), implying that Poly-A must also be linked to Apa1 and Taq1. This allows one to make comparative correlations between published studies of different 3’ VDR SNPs. However, examining the VDR genotype at only one SNP site is unlikely to accurately predict bioactivity of the VDR produced from that allele, and may yield inconsistent or contradictory results (Whitfield et al., 2005). Therefore, we decided to examine two sites, the Poly-A (3’ VDR) and the Fok1 locus (5’ VDR), simultaneously to ask whether these allelic biomarkers together reflect the VDR’s ability to influence patient susceptibility to developing squamous neoplasia. Here, we describe results of a clinical study that compared the status of the two alleles to the incidence of AK and cutaneous SCC, and also to circulating levels of 25(OH)D3 in a serial cohort of 137 patients enrolled and studied under controlled conditions.

RESULTS

Demographics of the study patients

Two clinical registry-based studies, the “Biomarker registry” and “VDAK study” respectively, were conducted to investigate effects of vitamin D status upon PDT treatment outcomes in patients with AK lesions. Results for therapeutic PDT outcomes have already been reported (Bullock et al., 2022, Heusinkveld et al., 2022). In this manuscript, blood samples from the patients were evaluated for possible associations between VDR allelic status, serum 25(OH)D3 levels, AK burden, and skin cancer history. Subjects from the two studies were combined (see flowchart, Fig. 2), providing a cohort of 137 patients. Study patients were Caucasian, average age of 70 ± 8.2 years, Fitzpatrick skin type 1 or 2, and predominantly male (89%). Patients had an average of 51 ± 33 AK lesions (mean ± SD) at baseline, and 74.5% had experienced one or more skin cancers during their lifetime (54.0% had SCC, 55.5% had BCC). Detailed demographics for each patient, along with results of genotypic analyses of their Fok1 and Poly-A alleles, are provided in Supplemental Table 1.

Figure 2.

Figure 2.

Open in a new tab

Flowchart for the two studies from which patient samples for VDR analysis were obtained. ClinicalTrials.gov identifiers: NCT03319251 and NCT04140292.

The prevalence of Fok1 alleles in our patient cohort (42.3% FF, 44.6% Ff, 13.1% ff), and of long and short Poly-A alleles (33.6% LL, 52.5% LS, 13.9% SS), closely resembles prevalence data reported in the general population. A detailed comparison of our allelic data with prior data from the literature is provided in Supplemental Table 2.

Serum 25(OH)D3 levels are weakly associated with the Fok1 “F” allele and strongly associated with the Poly-A “short” allele.

To analyze the relationship between individual VDR alleles and 25(OH)D3 serum levels, we employed the Fok1 score and Poly-A score described by Whitfield (Whitfield et al., 2001). Scores for the Fok1 allele (Table 1a) indicate a weak but significant relationship between presence of homozygous (F) alleles and a higher-than average serum 25(OH)D3 level. Table 1b presents data for the Poly-A allele using the original Whitfield score, in which the long (L) allele is given a score of 1. The data suggest a relationship in which the homozygous short (S) alleles is strongly associated with the highest serum 25(OH)D3 levels. However, examination of Fok1 and Poly-A alleles in isolation might miss possible interactive effects. To see whether Fok1 and Poly-A sites might be interacting in terms of their influence upon serum 25(OH)D3 levels, we tried combining the Fok1 and Poly-A scores into a single VDR score. Combining Fok1 and Poly-A scores as originally formulated by Whitfield et al. showed no meaningful relationship (Supplemental Table 3). However, combining an “inverted” Poly-A score with the Fok1 score resulted in a strong, continuous, monotonic correlation between the combination score and serum 25(OH)D3 levels (Table 1c). Thus, patients with homozygous FF and SS alleles had high mean serum 25(OH)D3 levels (50.0 ng/mL) whereas patients with homozygous ff and LL alleles were vitamin D deficient, with levels of only 29.1 ng/mL.

Table 1. Association of serum 25(OH)D3 levels with Fok1 and Poly-A alleles of the VDR.

For this analysis, 25(OH)D3 levels were compared to the following allelic scores: (a) Fok1 alone; (b) Poly-A alone, or (c) Fok1 and Poly-A together as a dual score that combines a Fok1 score with an inverted Poly-A score.

(a) Data sorted according to the Fok1 score described in (Whitfield et al., 2001).
Each F allele has a value of 1; each f allele has a value of 0. Genotypes: ff = 0, Ff = 1, FF = 2
Fok1 score Serum 25(OH)D3 level (ng/mL) # of patients P-value *
Mean SEM** (n)
ff = 0 33.3 10.7 (18) ---
Ff = 1 36.9 2.0 (61) ---
FF = 2 40.6 1.7 (58) 0.029
(b) Data sorted according to the original Poly-A score described in (Whitfield et al., 2001).
Each long allele (L) has a value of 1; each short allele (S) has a value of 0. Genotypes: SS = 0, LS = 1, LL = 2
Poly-A score Serum 25(OH)D3 level (ng/mL) # of patients P-value *
Mean SEM (n)
SS = 0 48.4 4.3 (19) ---
LS = 1 37.1 1.6 (72) 0.003
LL = 2 35.1 1.59 (46) 0.0006
(c) Data sorted according to a dual score combining Fok1 and “inverted” Poly-A scores.
The big F allele and the short Poly-A allele receive the highest credit: f = 0, F = 1; and L = 0, S = 1.
“Inverted” Fok1 and Poly-A scores Serum 25(OH)D3 level (ng/mL) ** # of patients P-value *
Mean SEM (n)
ff/LL = 0 29.1 9.3 (5) 0.003
ff/LS, or Ff/LL = 1 34.6 10.4 (32) 0.001
ff/SS, Ff/LS, or FF/LL = 2 35.7 13.3 (53) 0.008
FF/LS, or Ff/SS = 3 42.5 15.8 (39) 0.094 (trend)
FF/SS = 4 50.0 13.5 (8) ---
Open in a new tab
*

2-sided T-test relative to first row (ff = 0)

**

SEM, standard error of the mean

*

2-sided T-test relative to first row (SS = 0)

*

From 2-sided T-test, relative to the last row (FF/SS =4).

**

A 25(OH)D3 serum level > 31 ng/mL and < 80 ng/mL is considered normal.

To ask whether there might be a finer association between Fok1 and Poly-A allelic status, all possible combinations of the two allelic sites were plotted as in Fig. 3a. From this display it is clear that the presence of two short Poly-A alleles (SS) along with at least one Fok1 (F) allele is associated with significantly elevated 25(OH)D3 levels, relative to all other VDR genotypes.

Figure 3.

Figure 3.

Open in a new tab

Allelic VDR genotypes that contain homozygous Poly-A (SS) alleles show significant relationships with serum Vitamin D levels and with AK lesion burden. The figure shows each possible dual combination of Poly-A and Fok1 alleles, versus (a) 25(OH)D3 serum level, or (b) AK skin lesions at the first study visit. P-values are from t-test comparisons between means with a Benjamini-Hochberg correction. (n), number of patients per condition. Yellow bars, genotypes that contain homozygous Poly-A short alleles.

The presence of a homozygous Poly-A “short” allele is associated with fewer AK lesions.

Because many researchers have hypothesized that higher levels of vitamin D correlate with lower susceptibility to skin cancer (Tang et al., 2012a), we wondered whether the pattern in Figure 3a, wherein Poly-A (SS) alleles were associated with high 25(OH)D3 levels, might reflect a correlation with low AK burden in patients. To address this, all possible dual VDR genotypes were plotted against AK lesion counts for each allelic cohort (Fig. 3b). A clear pattern emerged. Whenever the Poly-A (SS) genotype was present, fewer AK lesions were seen, relative to AK counts with an LS or LL genotype within the same Fok1 genotypic group (Fig. 3b). The overriding association of homozygous Poly-A (SS) genotype with lower AK numbers, although statistically significant only for patients also harboring the homozygous Fok1 (FF) allele, bears a striking inverse relationship to the pattern observed for serum vitamin D concentrations and supports the hypothesis that high 25(OH)D3 levels correlate with a low burden of precancerous skin lesions.

A consideration of potentially confounding environmental factors that might affect serum 25(OH)D3 levels or AK development did not reveal any relationships to seasonal sun exposure or to the presence of immunosuppression (Supplemental Table 1, columns F and Q).

The presence of a homozygous Poly-A “short” allele is associated with reduced incidence of SCC.

Since it is well-established that AK are precursors to cutaneous SCC (Ratushny et al., 2012), we investigated associations between SCC incidence (defined as a history of SCC) and Fok1 or Poly-A allelic status. A dual-allele approach revealed no consistent relationships (data not shown). However when Poly-A and Fok1 alleles were evaluated separately, an interesting pattern suggesting possible association with SCC development became apparent (Fig. 4c,f), especially when displayed alongside the results for serum 25(OH)D3 (Fig. 4a,d) and AK burden (Fig. 4b,e). SCC incidence appeared directly related to number of Poly-A (L) alleles, or inversely related to number of short (S) alleles (Fig. 4c). A positive relationship between Fok1 (F) allele and SCC frequency was also observed (Fig. 4c). To address this further, we modeled each genotype using a risk allele approach (additive model) as shown in Fig. 4g. That analysis revealed a significant positive trend between the Poly-A (L) allele and cutaneous SCC risk, with an odds ratio of 1.55 per copy of the L allele. It was not possible to confirm a significant relationship between the Fok1 (F) allele and SCC risk, perhaps due to limited patient sample size or to a relatively weaker association between Fok1 alleles and serum 25(OH)D3 or AK burden. For basal cell carcinoma (BCC), no significant relationship between BCC risk and VDR alleles could be demonstrated (Supplemental Figure 3).

Figure 4.

Figure 4.

Open in a new tab

Comparison of associations between Poly-A or Fok1 VDR genotypes versus serum 25(OH)D3 levels, number of AK lesions, and history of SCC skin cancer. The graphs depict the observed relationships between serum 25(OH)D3 levels (a, d), number of AK lesions per patient (b, e), and percentage of patients with a history of skin cancer (c, f). Numbers in parenthesis, number of patients per cohort. P values, from t-test comparisons between means with a Benjamini-Hochberg correction. ns, not significant. (*), Dotted lines in c and f are trend lines that were tested statistically using risk-allele modeling in panel g. (g), Results of additive modeling in which the risk for developing SCC is calculated per copy of the hypothesized risk allele. The level of significance in this analysis is set at p < 0.1. (*), Gender was used as a covariate with female as the reference standard.

DISCUSSION

In this paper, we have identified correlations between two VDR gene polymorphisms (Fok1 and Poly-A) and three skin cancer-related variables frequently hypothesized to be associated in human populations, i.e. (1) serum vitamin D levels, (2) incidence of actinic keratoses (squamous precancer), and (3) incidence of cutaneous SCC. Our results showed a strong association between the Poly-A (S) allele and high serum 25(OH)D3 levels, and a weak association between the Fok1 (F) allele and high serum 25(OH)D3. An examination of both alleles together revealed a functional interaction between Fok1 and Poly-A sites, with increasing 25(OH)D3 concentrations associated with increasing Fok1 (F) alleles and decreasing Poly-A (L) alleles (Table 1c). Patients with genotype ffLL were Vitamin D deficient with a mean 25(OH)D3 level of 29.1 ng/dL, whereas FFSS genotype was associated with high 25(OH)D3 levels (50 ng/dL). Interestingly, the positive association between Poly-A (S) and high serum 25(OH)D3 was mirrored by an inverse relationship between Poly-A (S) and AK; in the latter case, Poly-A (S) was also associated with a low burden of AK lesions and a lower likelihood of cutaneous SCC. The correlations between Poly-A (S), higher 25(OH)D3, and lower AK and SCC incidence tend to support the hypothesis that high serum vitamin D3 levels are somehow protective against cancer.

Two important aspects of our study should be noted. First, amongst 28 prior reports that looked at associations between VDR polymorphisms and cutaneous SCC (Burns et al., 2017, Gandini et al., 2009, Köstner et al., 2012, Morgado-Aguila et al., 2020), AK lesions (Carless et al., 2008), or melanoma (reviewed in (Birke et al., 2020), none had examined the VDR Poly-A allele. Second, we have demonstrated significant associations between the VDR Fok1 and Poly-A sites, 25(OH)D3 levels, precancer (AK) burden, and history of SCC all within the same patient cohort.

When considering VDR Poly-A alleles and malignancy risk, evidence for an association is strongest for breast cancer. Studies from India (Chakraborty et al., 2009), Iran (Colagar et al., 2015), the U.K. (Bretherton-Watt et al., 2001, Guy et al., 2004) and Sweden (Wedren et al., 2007) all found that the long (L) allele of Poly-A was significantly related to increased breast cancer risk. Additional evidence comes from studies that examined 3’ VDR polymorphisms (Bsm1, Apa1, Taq1) that are genetically linked to Poly-A. In one study from the UK, the Poly-A (L) allele and genetically-linked Bsm (b) allele were each significantly associated with breast cancer risk (Guy et al., 2004). Interestingly, Fok1 showed no association when analyzed in isolation, but did have an influence when analyzed together with bb and LL (Guy et al., 2004). Other studies showed that Bsm1 (bb), Apa1 (aa), and Poly-A (LL) genotypes were all associated with breast cancer risk (Iqbal and Khan, 2017), consistent with linkage of these alleles in a “b-a-L” haplotype (Uitterlinden et al., 2004). The Poly-A (L) allele has also been associated with increased risk of colon cancer (Sweeney et al., 2006). For prostate cancer, some studies showed increased risk with the Poly-A (L) allele (Ingles et al., 1997), whereas others did not (Andersson et al., 2006). For melanoma, 3’ UTR alleles linked to Poly-A (L) are associated with increased risk of malignancy (Birke et al., 2020). For example, the Taq1 (T) allele correlated with increased melanoma risk (Li et al., 2007); note that Poly-A (L) is linked to Taq1 (T) as part of the “b-a-T-L” haplotype (Uitterlinden et al., 2004). The risk for multiple primary melanoma was elevated in patients with the Bsm1 (b) allele (Mandelcorn-Monson et al., 2011). For nonmelanoma skin cancer, a case-control study (100 patients, 100 controls) found that individuals with the Bsm1 (b) allele were nearly twice as likely to develop SCC and BCC (Burns et al., 2017). For keratinocyte pre-cancer (AKs), a study of 190 AK patients and 190 controls found increased AK risk with Taq1 and Apa1 (both alleles linked to Poly-A) (Carless et al., 2008). For all cancer, a meta-analysis of 73 cancer-VDR association studies demonstrated that the BsmI (B) allele, linked to the Poly-A (S) allele, is protective against most cancers examined (Raimondi et al., 2014).

Regarding the Fok1 allele and risk of non-melanoma skin cancer, a meta-analysis of eight papers revealed that the Fok1 (ff) genotype conferred a heightened risk for both SCC and BCC (Wan et al., 2022). This was opposite to our findings for SCC, perhaps because the number of patients in our study was too small.

Our observed association between Poly-A (LL or LS) genotypes and low levels of serum 25(OH)D3 levels. is largely in agreement with the literature. In a breast cancer study, low 25(OH)D3 levels correlated with the Poly-A (L) allele (Colagar et al., 2015). The Bsm1 (b) allele, linked to Poly-A (L), was associated with low 25(OH)D3 levels in prostate cancer (Ma et al., 1998) and in two non-cancer studies (Ezhilarasi et al., 2018, Oliveira et al., 2018). How the length of the Poly-A (L) allele might affect serum 25(OH)D3 levels remain speculative. The Poly-A polymorphism does not affect VDR protein structure, but it might affect stability, localization, transcription, or translation of VDR mRNA, perhaps by interacting with a Poly-A binding protein (Whitfield et al., 2005). In a study analyzing the stability of globin mRNAs attached to the VDR 3’ UTR, different Poly-A alleles did not seem to affect mRNA stability (Durrin et al., 1999, Verbeek et al., 1997). Whitfield et al. examined the hypothesis that VDR alleles enhance the efficiency of VDR transcription; a vitamin D-responsive promoter construct was transfected into human fibroblast cell lines and reporter expression was lower when transfected into fibroblasts bearing a VDR Poly-A (SS) genotype, as compared to a Poly-A genotype (LL) (Whitfield et al., 2001). At the time, this was interpreted to mean that the Poly-A (S) variant yielded lower VDR protein levels. This result was later questioned, both because the sample size in Whitfield et al. was relatively small and because a large meta-analysis (Uitterlinden et al., 2004) indicated the opposite, namely that SS homozygotes (inferred by linkage to Bsm1, Apa1 and Taq1 polymorphisms) produced the more active VDR. Interestingly, the idea that Poly-A (SS) produces the more active VDR actually complements the findings of Jurutka et al. (2000), in which the VDR produced by the Fok1 (F) variant was more active than the (f) variant. If both findings were true, then this would agree nicely with our finding that the strongest phenotypic effects (reduced AK, elevated calcidiol) occur with the FFSS genotype, with SS and FF both favoring the more active VDR.

Of course, one cannot directly extrapolate from in vitro studies to human patients, for many reasons. Transcriptional responses to a lower level of VDR protein may be completely opposite for different genes and different cell types. The VDR targets >1000 genes representing many different functional pathways, regulated by interacting factors and regulatory circuits that make results hard to predict (Maestro et al., 2016, Nurminen et al., 2019). In the absence of calcitriol, the VDR may suppress transcription of genes normally induced by VDR binding to the hormone (Lee et al., 2014, Pike et al., 2016). Transcriptional activation of VDR-response genes requires binding (heterodimerization) of the VDR with retinoid receptor RXR, adding another level of regulatory complexity (Whitfield et al., 2005).

Our initial hypothesis was that serum 25(OH)D3 levels may reflect a causal relationship between cancer susceptibility and some aspect of the vitamin D metabolic pathway. However, the idea that 25(OH)D3 serum levels are a good indicator of calcitriol (ligand)-dependent VDR activity is not supported by some ancillary data obtained in our trial. In a subset of 27 patients, we measured both calcidiol and calcitriol serum levels (Supplemental Table 4). There was an astonishing lack of correlation between the two parameters. Lack of a correlation between calcitriol levels and AK or SCC incidence suggests that ligand-driven actions of the VDR may not be directly responsible for observed epidemiological cancer patterns. Keratinocytes and keratinocyte-derived cancers produce their own 1,25(di-OH)D3, which makes VDR activity in keratinocytes less reliant on serum 1,25(di-OH)D3 levels (Bikle and Christakos, 2020). If calcitriol ligand levels are not critical for cancer outcomes, then perhaps other transcription factors or proteins that interact with VDR are key to the differential cancer-regulatory effects of the VDR allelic variants. For example, Jurutka showed that the VDR protein interacts with transcription factor IIB, and this interaction differs in Fok1 (F) versus (f) encoded versions of VDR based upon the presence of positively charged amino acid residues near the N-terminus (Jurutka et al., 2000). Mutagenesis of those residues to uncharged alanines confirmed their importance and provided a mechanistic explanation for the higher activity of VDR produced from the F allele (Jurutka et al., 2000).

Complex interactions may also explain why population-based, genome-wide association studies (GWAS) have so far failed to detect a role for VDR genetic variability as a contributor to 25(OH)D3 levels or incidence of skin cancer. To illustrate, large GWAS studies (samples sizes from several thousand to over 400,000 subjects) have identified ~35 SNPs that contribute to variability in 25(OH)D3 levels, consistently across studies (Hypponen et al., 2022, Jiang et al., 2018, Manousaki et al., 2020). Identified genes include several enzymes involved in Vitamin D synthesis (DHCR7, CYP2R) and in storage of Vitamin D in lipid-rich tissues. However, the overall estimate of heritability of 25(OH)D3 levels from common SNPs, detectable by GWAS, is only ~7.5% (Jiang et al., 2018). Thus, many genetic variations may reside in as-yet undetected rare SNPs that encode structural protein variants and/or affect gene-gene and/or protein-protein interactions. The latter idea is bolstered by this manuscript which illustrates how the full effect of two alleles (Poly-A and Fok1) are not fully recognizable until both alleles are analyzed together.

The major limitation of this study is its relatively small study population; larger future studies are clearly warranted. A second limitation is the low female-to-male ratio, so that possible gender-related differences could not be evaluated. Third, our study did not include a cancer-free control group (i.e., patients lacking any AK or SCC), so it is possible that reported relationships between 25(OH)D3 levels and VDR Poly-A alleles could be dependent upon the presence of skin cancer or pre-cancer. To address the latter question, an AK burden analysis was performed in which patients with relatively few AK lesion counts were compared to patients with intermediate or high AK lesion counts (Supplemental Table 5). The absence of any loss or change in the 25(OH)D3 elevating effect of Poly-A (SS) amongst the three groups argues that the relationship is not AK dose-dependent. Despite these caveats, the findings in this manuscript suggest that VDR Poly-A and Fok1 alleles together play an important role in regulating serum Vitamin D levels, and the development of cutaneous AK and SCC is influenced by the VDR Poly-A allele.

METHODS

Study design

Data were obtained as part of two clinical trials (ClinicalTrials.gov NCT03319251 “Biomarker registry”, and NCT04140292 “VDAK study”) in which patients with actinic keratoses (AK) of the face or scalp were enrolled at the Cleveland Clinic between October 2017 and October 2020. Each trial was approved by the Institutional Review Board of the Cleveland Clinic. Patients with at least 10 AK lesions were enrolled after providing written, informed consent. All patients had an initial blood draw to measure serum 25(OH)D3 levels and collect leukocyte DNA for VDR gene analysis. AK lesions were identified by visual inspection and palpation by a dermatologist who marked and photographed the lesions. A history of skin cancers experienced over a lifetime was obtained for each patient, focusing on squamous cell carcinomas (SCC) and basal cell carcinoma (BCC) because of the strong relationship between these tumor types and chronic UV exposure. Subsequently, patients received PDT under the various conditions outlined in Fig. 2. The analysis in this manuscript employs data obtained at the initial study visit.

Preparation of DNA from human blood samples

Blood was collected, blood cells lysed, and DNA isolated using a cell lysis/DNA extraction kit from Qiagen (Germantown, MD), as described in Supplemental Figure 1.

Analysis of polymorphisms in the human Vitamin D Receptor

Alleles of Fok1 (F or f) were analyzed via conventional PCR; primers and conditions are described in Supplemental Figure 1. The microsatellite site (Poly-A) was evaluated using PCR amplification of the 3’UTR followed by fragment size analysis using an ABI-3730 Genetic Analyzer, as described in Supplemental Figure 2.

Statistical Analysis

Continuous variables (serum vitamin D levels or AK counts) were summarized with appropriate descriptive statistics by Fok1 and Poly-A genotype, respectively, and their means were compared pairwise using the t-test. For each outcome (serum vitamin D levels or AK counts), the multiple pairwise results were corrected for multiple testing using the Benjamini-Hochberg procedure. For evaluation of the discrete variable of SCC cancer history, a logistic regression model was fit with each genotype as a predictor and gender as a covariate, and odds ratios were computed per risk allele under an additive genetics model (Fig. 4g).

Supplementary Material

supplement

Supplemental Figure 1. Measurement of Fok1 alleles: Methods and illustrative examples.

Supplemental Figure 2. Measurement of Poly-A alleles: Methods and illustrative examples.

Supplemental Figure 3. Relationship of BCC history to VDR Poly-A or Fok1 alleles.

Supplemental Table 1. Summary of demographics, 25(OH)D3 serum levels, and VDR allele data.

Supplemental Table 2. Fok1 and Poly-A alleles in our patient cohort, and comparison to the literature.

Supplemental Table 3. Original Fok1 + Poly-A combined score: Lack of correlation.

Supplemental Table 4. Comparison of 25(OH)D3 and 1,25(diOH)D3 serum levels in a patient subset.

Supplemental Table 5. Relationship between serum 25(OH)D3 levels and VDR alleles as a function of AK lesion burden.

ACKNOWLEDGEMENTS

We thank Dr. G. Kerr Whitfield for his helpful suggestions and input. This work was supported by grants (R01 CA204158, and P01 CA084203) from the National Cancer Institute of the National Institutes of Health (NCI/NIH).

Abbreviations:

AK

actinic keratosis

SCC

squamous cell carcinoma

BCC

basal cell carcinoma

D3

vitamin D3

GWAS

genome wide association study

NMSC

non-melanoma skin cancer

PDT

photodynamic therapy

SNP

single nucleotide polymorphism

UTR

untranslated region

mRNA

messenger RNA

UV

ultraviolet light

VDR

vitamin D receptor

25(OH)D3

25-hydroxy-vitamin D3

1,25(diOH)D3

1,25-dihydroxy-vitamin D3

Footnotes

CONFLICT OF INTEREST

The authors state no conflict of interest.

CReditT STATEMENT

Conceptualization: TAB, JAM, SA, TH, EVM; Formal analysis: TAB, BH, EVM; Funding acquisition: TH, EVM; Investigation: TAB, JAM, JN, UK, CBW; Methodology: TAB, JAM, BH, SA, EVM; Project administration: JN; Writing (lead): TAB; Writing (supporting): CBW, EVM

DATA AVAILABILITY STATEMENT

No genome-wide datasets were generated during this study.

REFERENCES

  1. Anand S, Rollakanti KR, Horst RL, Hasan T, Maytin EV. Combination of oral vitamin D3 with photodynamic therapy enhances tumor cell death in a murine model of cutaneous squamous cell carcinoma. Photochem Photobiol 2014;90(5):1126–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Andersson P, Varenhorst E, Soderkvist P. Androgen receptor and vitamin D receptor gene polymorphisms and prostate cancer risk. Eur J Cancer 2006;42(16):2833–7. [DOI] [PubMed] [Google Scholar]
  3. Arai H, Miyamoto K, Taketani Y, Yamamoto H, Iemori Y, Morita K, et al. A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 1997;12(6):915–21. [DOI] [PubMed] [Google Scholar]
  4. Bikle D, Christakos S. New aspects of vitamin D metabolism and action - addressing the skin as source and target. Nat Rev Endocrinol 2020;16(4):234–52. [DOI] [PubMed] [Google Scholar]
  5. Bikle DD. Vitamin D: Production, Metabolism and Mechanisms of Action. In: Feingold KR, Anawalt B, Boyce A, Chrousos G, de Herder WW, Dhatariya K, et al. , editors. Endotext. South Dartmouth (MA); 2000. [Google Scholar]
  6. Bikle DD. The Vitamin D Receptor as Tumor Suppressor in Skin. Adv Exp Med Biol 2020;1268:285–306. [DOI] [PubMed] [Google Scholar]
  7. Birke M, Schöpe J, Wagenpfeil S, Vogt T, Reichrath J. Association of Vitamin D Receptor Gene Polymorphisms With Melanoma Risk: A Meta-analysis and Systematic Review. Anticancer Res 2020;40(2):583–95. [DOI] [PubMed] [Google Scholar]
  8. Bretherton-Watt D, Given-Wilson R, Mansi JL, Thomas V, Carter N, Colston KW. Vitamin D receptor gene polymorphisms are associated with breast cancer risk in a UK Caucasian population. Br J Cancer 2001;85(2):171–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bullock TA, Negrey J, Hu B, Warren CB, Hasan T, Maytin EV. Significant improvement of facial actinic keratoses after blue light photodynamic therapy with oral vitamin D pretreatment: An interventional cohort-controlled trial. J Am Acad Dermatol 2022;87(1):80–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Burns EM, Guroji P, Ahmad I, Nasr HM, Wang Y, Tamimi IA, et al. Association of Vitamin D Receptor Polymorphisms With the Risk of Nonmelanoma Skin Cancer in Adults. JAMA Dermatol 2017;153(10):983–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Burton KA, Ashack KA, Khachemoune A. Cutaneous Squamous Cell Carcinoma: A Review of High-Risk and Metastatic Disease. Am J Clin Dermatol 2016;17(5):491–508. [DOI] [PubMed] [Google Scholar]
  12. Carless MA, Kraska T, Lintell N, Neale RE, Green AC, Griffiths LR. Polymorphisms of the VDR gene are associated with presence of solar keratoses on the skin. Br J Dermatol 2008;159(4):804–10. [DOI] [PubMed] [Google Scholar]
  13. Chakraborty A, Mishra AK, Soni A, Regina T, Mohil R, Bhatnagar D, et al. Vitamin D receptor gene polymorphism(s) and breast cancer risk in north Indians. Cancer Detect Prev 2009;32(5–6):386–94. [DOI] [PubMed] [Google Scholar]
  14. Cleaver JE, Crowley E. UV damage, DNA repair and skin carcinogenesis. Front Biosci 2002;7:d1024–43. [DOI] [PubMed] [Google Scholar]
  15. Colagar AH, Firouzjah HM, Halalkhor S. Vitamin D Receptor Poly(A) Microsatellite Polymorphism and 25-Hydroxyvitamin D Serum Levels: Association with Susceptibility to Breast Cancer. J Breast Cancer 2015;18(2):119–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Denzer N, Vogt T, Reichrath J. Vitamin D receptor (VDR) polymorphisms and skin cancer: A systematic review. Dermatoendocrinol 2011;3(3):205–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Durrin LK, Haile RW, Ingles SA, Coetzee GA. Vitamin D receptor 3’-untranslated region polymorphisms: lack of effect on mRNA stability. Biochim Biophys Acta 1999;1453(3):311–20. [DOI] [PubMed] [Google Scholar]
  18. Ezhilarasi K, Dhamodharan U, Vijay V. BSMI single nucleotide polymorphism in vitamin D receptor gene is associated with decreased circulatory levels of serum 25-hydroxyvitamin D among micro and macrovascular complications of type 2 diabetes mellitus. Int J Biol Macromol 2018;116:346–53. [DOI] [PubMed] [Google Scholar]
  19. Gandini S, Raimondi S, Gnagnarella P, Dore JF, Maisonneuve P, Testori A. Vitamin D and skin cancer: a meta-analysis. Eur J Cancer 2009;45(4):634–41. [DOI] [PubMed] [Google Scholar]
  20. Guy GP Jr., Machlin SR, Ekwueme DU, Yabroff KR. Prevalence and costs of skin cancer treatment in the U.S., 2002–2006 and 2007–2011. Am J Prev Med 2015;48(2):183–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Guy M, Lowe LC, Bretherton-Watt D, Mansi JL, Peckitt C, Bliss J, et al. Vitamin D receptor gene polymorphisms and breast cancer risk. Clin Cancer Res 2004;10(16):5472–81. [DOI] [PubMed] [Google Scholar]
  22. Heusinkveld LE, Bullock TA, Negrey J, Warren CB, Maytin EV. Sandpaper curettage: A simple method to improve PDT outcomes for actinic keratosis. Photodiagnosis Photodyn Ther 2022;40:103050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hypponen E, Vimaleswaran KS, Zhou A. Genetic Determinants of 25-Hydroxyvitamin D Concentrations and Their Relevance to Public Health. Nutrients 2022;14(20). [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Ingles SA, Ross RK, Yu MC, Irvine RA, La Pera G, Haile RW, et al. Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst 1997;89(2):166–70. [DOI] [PubMed] [Google Scholar]
  25. Iqbal MUN, Khan TA. Association between Vitamin D receptor (Cdx2, Fok1, Bsm1, Apa1, Bgl1, Taq1, and Poly (A)) gene polymorphism and breast cancer: A systematic review and meta-analysis. Tumour Biol 2017;39(10):1010428317731280. [DOI] [PubMed] [Google Scholar]
  26. Jiang X, O’Reilly PF, Aschard H, Hsu YH, Richards JB, Dupuis J, et al. Genome-wide association study in 79,366 European-ancestry individuals informs the genetic architecture of 25-hydroxyvitamin D levels. Nat Commun 2018;9(1):260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, et al. The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 2000;14(3):401–20. [DOI] [PubMed] [Google Scholar]
  28. Köstner K, Denzer N, Koreng M, Reichrath S, Gräber S, Klein R, et al. Association of genetic variants of the vitamin D receptor (VDR) with cutaneous squamous cell carcinomas (SCC) and basal cell carcinomas (BCC): a pilot study in a German population. Anticancer Res 2012;32(1):327–33. [PubMed] [Google Scholar]
  29. Lee SM, Goellner JJ, O’Brien CA, Pike JW. A humanized mouse model of hereditary 1,25-dihydroxyvitamin D-resistant rickets without alopecia. Endocrinology 2014;155(11):4137–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Li C, Liu Z, Zhang Z, Strom SS, Gershenwald JE, Prieto VG, et al. Genetic variants of the vitamin D receptor gene alter risk of cutaneous melanoma. J Invest Dermatol 2007;127(2):276–80. [DOI] [PubMed] [Google Scholar]
  31. Ma J, Stampfer MJ, Gann PH, Hough HL, Giovannucci E, Kelsey KT, et al. Vitamin D receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev 1998;7(5):385–90. [PubMed] [Google Scholar]
  32. Maestro MA, Molnar F, Mourino A, Carlberg C. Vitamin D receptor 2016: novel ligands and structural insights. Expert Opin Ther Pat 2016;26(11):1291–306. [DOI] [PubMed] [Google Scholar]
  33. Mandelcorn-Monson R, Marrett L, Kricker A, Armstrong BK, Orlow I, Goumas C, et al. Sun exposure, vitamin D receptor polymorphisms FokI and BsmI and risk of multiple primary melanoma. Cancer Epidemiol 2011;35(6):e105–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Manousaki D, Mitchell R, Dudding T, Haworth S, Harroud A, Forgetta V, et al. Genome-wide Association Study for Vitamin D Levels Reveals 69 Independent Loci. Am J Hum Genet 2020;106(3):327–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Moreno R, Najera L, Mascaraque M, Juarranz A, Gonzalez S, Gilaberte Y. Influence of Serum Vitamin D Level in the Response of Actinic Keratosis to Photodynamic Therapy with Methylaminolevulinate. J Clin Med 2020;9(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Morgado-Aguila C, Rey-Sanchez P, Gil-Fernandez G, Costa-Fernandez MC, Rodriguez-Velasco FJ. Vitamin D Receptor Polymorphisms and Non-Melanoma Skin Cancer Risk: A Case-Control Study. J Clin Med 2020;9(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, et al. Prediction of bone density from vitamin D receptor alleles. Nature 1994;367(6460):284–7. [DOI] [PubMed] [Google Scholar]
  38. Morrison NA, Yeoman R, Kelly PJ, Eisman JA. Contribution of trans-acting factor alleles to normal physiological variability: vitamin D receptor gene polymorphism and circulating osteocalcin. Proc Natl Acad Sci U S A 1992;89(15):6665–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Nurminen V, Seuter S, Carlberg C. Primary Vitamin D Target Genes of Human Monocytes. Front Physiol 2019;10:194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Oliveira ACR, Magalhaes CA, Loures CMG, Fraga VG, de Souza LC, Guimaraes HC, et al. BsmI polymorphism in the vitamin D receptor gene is associated with 25-hydroxy vitamin D levels in individuals with cognitive decline. Arq Neuropsiquiatr 2018;76(11):760–6. [DOI] [PubMed] [Google Scholar]
  41. Pike JW, Meyer MB, Benkusky NA, Lee SM, St John H, Carlson A, et al. Genomic Determinants of Vitamin D-Regulated Gene Expression. Vitam Horm 2016;100:21–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pike JW, Shevde NK. The Vitamin D Receptor. In: Feldman D, Pike JW, Glorieux FH, editors. Vitamin D. 1: Elsevier Academic Press; 2005. p. 167–91. [Google Scholar]
  43. Raimondi S, Pasquali E, Gnagnarella P, Serrano D, Disalvatore D, Johansson HA, et al. BsmI polymorphism of vitamin D receptor gene and cancer risk: a comprehensive meta-analysis. Mutat Res 2014;769:17–34. [DOI] [PubMed] [Google Scholar]
  44. Ratushny V, Gober MD, Hick R, Ridky TW, Seykora JT. From keratinocyte to cancer: the pathogenesis and modeling of cutaneous squamous cell carcinoma. J Clin Invest 2012;122(2):464–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Reichrath J, Saternus R, Vogt T. Endocrine actions of vitamin D in skin: Relevance for photocarcinogenesis of non-melanoma skin cancer, and beyond. Mol Cell Endocrinol 2017;453:96–102. [DOI] [PubMed] [Google Scholar]
  46. Ridky TW. Nonmelanoma skin cancer. J Am Acad Dermatol 2007;57(3):484–501. [DOI] [PubMed] [Google Scholar]
  47. Rogers HW, Weinstock MA, Feldman SR, Coldiron BM. Incidence Estimate of Nonmelanoma Skin Cancer (Keratinocyte Carcinomas) in the U.S. Population, 2012. JAMA Dermatol 2015;151(10):1081–6. [DOI] [PubMed] [Google Scholar]
  48. Shah P, He YY. Molecular regulation of UV-induced DNA repair. Photochem Photobiol 2015;91(2):254–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Siegel JA, Korgavkar K, Weinstock MA. Current perspective on actinic keratosis: a review. Br J Dermatol 2017;177(2):350–8. [DOI] [PubMed] [Google Scholar]
  50. Song EJ, Gordon-Thomson C, Cole L, Stern H, Halliday GM, Damian DL, et al. 1alpha,25-Dihydroxyvitamin D3 reduces several types of UV-induced DNA damage and contributes to photoprotection. J Steroid Biochem Mol Biol 2013;136:131–8. [DOI] [PubMed] [Google Scholar]
  51. Sweeney C, Curtin K, Murtaugh MA, Caan BJ, Potter JD, Slattery ML. Haplotype analysis of common vitamin D receptor variants and colon and rectal cancers. Cancer Epidemiol Biomarkers Prev 2006;15(4):744–9. [DOI] [PubMed] [Google Scholar]
  52. Tang JY, Fu T, Lau C, Oh DH, Bikle DD, Asgari MM. Vitamin D in cutaneous carcinogenesis: part I. J Am Acad Dermatol 2012a;67(5):803 e1–12, quiz 15–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Tang JY, Fu T, Lau C, Oh DH, Bikle DD, Asgari MM. Vitamin D in cutaneous carcinogenesis: part II. J Am Acad Dermatol 2012b;67(5):817 e1–11; quiz 27–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Uitterlinden AG, Fang Y, Van Meurs JB, Pols HA, Van Leeuwen JP. Genetics and biology of vitamin D receptor polymorphisms. Gene 2004;338(2):143–56. [DOI] [PubMed] [Google Scholar]
  55. Verbeek W, Gombart AF, Shiohara M, Campbell M, Koeffler HP. Vitamin D receptor: no evidence for allele-specific mRNA stability in cells which are heterozygous for the Taq I restriction enzyme polymorphism. Biochem Biophys Res Commun 1997;238(1):77–80. [DOI] [PubMed] [Google Scholar]
  56. Wan Y, Lyu Y, Xu Y, Huang P. The relationship between VDR polymorphisms and keratinocyte carcinomas: a systematic review and meta-analysis. Future Oncol 2022;18(23):2613–26. [DOI] [PubMed] [Google Scholar]
  57. Wedren S, Magnusson C, Humphreys K, Melhus H, Kindmark A, Stiger F, et al. Associations between androgen and Vitamin D receptor microsatellites and postmenopausal breast cancer. Cancer Epidemiol Biomarkers Prev 2007;16(9):1775–83. [DOI] [PubMed] [Google Scholar]
  58. Whitfield GK, Jurutka PW, Haussler CA, Hsieh J-C, Barthel TK, Jacobs ET, et al. Nuclear Vitamin D Receptor: Sturcture-Function, Molecular Control of Gene Transcription, and Novel Bioactions. In: Feldman D, Pike JW, Glorieux FH, editors. Vitamin D. 1. Second ed. London: Elsevier Academic PRess; 2005. p. 219–61. [Google Scholar]
  59. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, et al. Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 2001;177(1–2):145–59. [DOI] [PubMed] [Google Scholar]
  60. Yu SH, Bordeaux JS, Baron ED. The immune system and skin cancer. Adv Exp Med Biol 2014;810:182–91. [DOI] [PubMed] [Google Scholar]
  61. Ziegler A, Jonason AS, Leffell DJ, Simon JA, Sharma HW, Kimmelman J, et al. Sunburn and p53 in the onset of skin cancer. Nature 1994;372(6508):773–6. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement

Supplemental Figure 1. Measurement of Fok1 alleles: Methods and illustrative examples.

Supplemental Figure 2. Measurement of Poly-A alleles: Methods and illustrative examples.

Supplemental Figure 3. Relationship of BCC history to VDR Poly-A or Fok1 alleles.

Supplemental Table 1. Summary of demographics, 25(OH)D3 serum levels, and VDR allele data.

Supplemental Table 2. Fok1 and Poly-A alleles in our patient cohort, and comparison to the literature.

Supplemental Table 3. Original Fok1 + Poly-A combined score: Lack of correlation.

Supplemental Table 4. Comparison of 25(OH)D3 and 1,25(diOH)D3 serum levels in a patient subset.

Supplemental Table 5. Relationship between serum 25(OH)D3 levels and VDR alleles as a function of AK lesion burden.

NIHMS1913173-supplement-supplement.pdf (20.5MB, pdf)

Data Availability Statement

No genome-wide datasets were generated during this study.

RESOURCES