Calcium, Vitamin D, VDR Genotypes, and Epigenetic and Genetic Changes in Rectal Tumors

Abstract

Calcium, vitamin D, exposure to sunshine, and vitamin D receptor (VDR) genotypes have been associated rectal cancer. We used data from 750 rectal tumors and 1,205 population-based controls examine associations with TP53, KRAS2, and CpG Island methylator phenotype (CIMP) markers. Rectal tumors were associated with high levels of calcium overall and with TP53 tumor mutations specifically (OR = 0.6, 95% CI = 0.42–0.84). High levels of sunshine exposure had a borderline protective effect for TP53 tumor mutations (OR = 0.78, 95% CI = 0.59–1.03). A mutation at codon 248 was significantly associated with dietary calcium intake (OR = 0.26, 95% CI = 0.09–0.77); having the Ff/ff genotypes of the FOK1 VDR polymorphism significantly increased the odds of a mutation at codon 245 (OR = 4.74, 95% CI = 1.05–21.39); high levels of dietary vitamin D (OR = 3.42, 95% CI = 1.15– 10.17) and the Ff/ff genotypes of FOK1 (OR = 3.34, 95% CI = 1.11– 10.02) and the GA/AA genotypes of the CDX2 VDR polymorphism (OR = 2.85, 95% CI = 1.23–6.58) increased the odds of a TP53 mutation at codon 273. These data support an association between calcium and rectal tumors overall as well as specifically with TP53 mutations. However, given the number of comparisons, findings need to be confirmed in other studies.

INTRODUCTION

Calcium and vitamin D have been associated with both colon and rectal cancer (1–6). Some studies have suggested that associations with more distal colon and rectal tumors may be stronger than those observed for proximal colon tumors (5,7,8). The vitamin D receptor (VDR) has been shown to modify the associations between diet and colon cancer and colon adenomas (5,9–11). A study of colon cancer further suggested that the Fok1 and CDX2 VDR polymorphisms (rs10735810 and rs11568820) may be more strongly associated with KRAS2 and possibly CIMP tumors than with TP53 mutated tumors; these associations were modified by use of aspirin/NSAIDs (12) Associations between calcium, vitamin D, sunshine exposure, VDR, and specific mutations in rectal cancer have not been reported. However, given that associations for calcium and vitamin D have been reported as being stronger for more distal and rectal tumors in some studies, it is reasonable to determine if these factors are associated with specific rectal tumor mutations.

In this study, we determined CIMP and TP53 and KRAS2 mutations from 750 incident rectal cancer cases from a population-based study conducted in Northern California and Utah. We limit our analyses to rectal cancer since associations have been previously reported for colon cancer (12). We use a case-control study design utilizing data from 1,205 population-based controls to examine associations between calcium, vitamin D, sunshine exposure, and VDR genotypes.

METHODS

Participants in the study were from the Kaiser Permanente Medical Care Program of Northern California (KPMCP) and the state of Utah. All eligible cases within these defined areas were identified and recruited for the study. Cases with a first primary tumor in the rectosigmoid junction or rectum were identified between May 1997 and May 2001; tumor blocks were obtained between 2004 and 2007. Case eligibility was determined by the Surveillance Epidemiology and End Results Cancer Registries in Northern California and in Utah. To be eligible for the study, participants had to be between 30 and 79 yr of age at time of diagnosis, English speaking, mentally competent to complete the interview, could not have had previous colorectal cancer (13), and could not have known (as indicated on the pathology report) familial adenomatous polyposis, ulcerative colitis, or Crohn’s disease.

A total of 1,505 rectal cancer cases were identified; of these, 982 were interviewed. Reasons for nonresponse have been detailed (14). Block retrieval involved obtaining biopsy prior to treatment as well as paraffin embedded tissue from the resection. In some instances, because of radiation prior to resection, tissue was limited from the resection; therefore, biopsy specimens were used for making tumor DNA. In Utah, blocks were requested for all cases except those who refused release of blocks. For those who were not interviewed and had not signed a medical record release, the Utah Cancer Registry retrieved the blocks and released them to the study without key identifiers of name, address, and complete date of birth (year and month of birth were released). At the KPMCP, samples were retrieved from persons who signed a consent form or who had died. For the 1,495 eligible rectal cancer cases identified at both centers, 239 people identified with rectal cancer had not given consent to have the tissue released (15.9%); and for an additional 234 cases, either tumor tissue could not be obtained or DNA could not be extracted. Tumor DNA was extracted from 81.4% of all rectal cancer cases identified, of which 750 cases had interview data. Controls were randomly selected from membership lists at KPMCP, social security lists, and driver’s license list (people under 65 yr); 1,205 controls (68.8% of those selected) participated and are included in these analyses.

Genetic Analysis

Tumor DNA obtained from paraffin-embedded tissue was characterized by their genetic profile that include sequence data for exons 5 through 8 or the hotspots of mutations of the TP53 gene; sequence data for KRAS2 codons 12 and 13; and 5 CIMP markers MINT1, MINT2, MINT31, p16, and MLH1. At this time there is no “consensus” as to the appropriate CIMP panel or method of detection. However, we have used our panel to demonstrate significant relationships between CIMP and numerous variables, including cigarette smoking and the BRAF V600E mutation, which were independent of microsatellite instability (15,16). This work has helped to support the legitimacy of the CIMP concept (17). Germline DNA was available from blood drawn at the time of the interview. We assessed three VDR markers including rs11568820 (CDX2), intron 8 Bsm I (rs154410), and the Fok1 (rs10735810) as previously described (10,18,19).

Diet and Lifestyle Data

Trained and certified interviewers collected diet and lifestyle data as previously outlined (20,21). The referent year for the study was the calendar year approximately 2 yr prior to date of diagnosis (cases) or selection (controls). Information was collected on demographic factors such as age, sex, and study center; physical activity as determined by a detailed physical activity questionnaire that obtained information on activity patterns 10 and 20 yr ago as well as activity during the referent year (22,23); body size, including usual adult height and weight 2 and 5 yr prior to diagnosis; cigarette smoking history; family history of colorectal cancer in first degree relatives; and medical and reproductive history including use of hormone replacement therapy. Regular use of aspirin and NSAIDs were obtained from the following question: “Before the referent date, did you ever take aspirin, excluding Tylenol, regularly? Some brand names for aspirin include Anacin, Arthritis Pain Formula, Ascriptin Tablets, Bayer, Buffrin, Empirin, Excedrin, and Vanquish. Before the referent date did you ever take other non-steroidal antiinflammatory drugs or arthritis medicines such as ibuprofen, Motrin, Clinoril, Naprosyn, or Feldene?” Regular was defined as at least 3 times/wk for 1 mo.

Dietary intake was ascertained using an adaptation of the CARDIA diet history (21,24,25). Participants were asked to recall foods eaten, the frequency at which they were eaten, serving size, and if fats were added in the preparation. Nutrient information was obtained by converting food intake data into nutrient data using the Minnesota Nutrition Coding Center (NCC) nutrient database. Additionally participants were asked about dietary supplements used. We determined dietary calcium and vitamin D using amount reported from the diet history questionnaire along with amount obtained from supplements. Assessment of various indicators of calcium and vitamin D showed that the diet plus dietary supplement amount was the most stable variable for both men and women when evaluating associations.

Statistical Analysis

Dietary variables were assessed by sex-specific tertile of intake based on the distribution of the controls for men and women separately. We also assessed differences in association by sex, any use of aspirin or ibuprofen-type drugs within the past 2 yr, and by VDR genotypes. Recent use, within the past 2 yr, of aspirin or ibuprofen-type drugs was assessed since recent use was shown to be a better predictor of rectal cancer risk than any or lifetime use. Previous results for colon cancer have shown that NSAIDs are import effect modifiers of VDR and other diet and lifestyle factors (12,26), thus, we have evaluated the potential effect modification of NSAIDs on the associations reported here.

All statistical analysis was done using SAS version 9.1 (SAS Institute, Cary, NC). Tumors were defined by specific mutations detected as any TP53 vs. no TP53 mutation, any KRAS2 mutations vs. no KRAS2 mutation, or CIMP positive vs. CIMP negative. CIMP positive was defined as at least two methylated markers. For TP53 and KRAS2 mutations, we also examined transversion and transition mutations since specific types of mutations were assessed because other studies have shown specific mutations to have etiologic associations (27,28). For TP53 mutations, we examined the more common point mutations since these have been shown to be uniquely related to dietary factors and colon cancer (27). Population-based controls were used to assess associations for the population overall while examining multiple outcomes defined by tumor status. Multiple logistic regression models were used to compare all interviewed cases, regardless of whether or not tumor tissue was obtained, to controls. A generalized estimating equation (GEE) method was used to assess associations for the population overall comparing specific types of mutations to controls as described by Kuss and McLerran (29). Cases could contribute one to three observations in the multinomial GEE models depending how an individual’s number of tumor mutations (CIMP, KRAS2, TP53). All models were adjusted for age at diagnosis or selection and sex along with other factors that have been shown to be related to colon cancer including body mass index (BMI) in kg/m², long-term vigorous physical activity, pack-years of cigarettes smoked, dietary calcium, and total energy intake. Additional adjustment for center did not alter results. Interaction was assessed by determining if the interaction term significantly improved the overall fit of the model by comparing a model with the interaction term as an ordered categorical variable to a model without the interaction term using the likelihood ratio test with 1 df. P for trend was assessed over ordered categories of variables; in the instance of genotypes, the trend P value was based on a model that included variant, heterozygote, and wild type, although in some instances, odds ratios are presented for the dominant model.

RESULTS

The distribution of tumor mutations within the study population is shown in Table 1. Eleven percent had a CIMP positive tumor, 28.9% had a KRAS2 mutation, and 48.3% had a TP53 mutation. The majority of both KRAS2 and TP53 mutations were transition mutations.

TABLE 1.

Description of study population

	Cases		Controls
	n	%	n	%
Total	951		1205
Tumor Mutations Data	750
CIMP	74	11.0
KRAS2	215	28.9
Overall     Transitions	140	21.0
Transversions	75	12.4
TP53	340	48.3
Overall     Transitions	248	40.5
Transversions	69	15.9
Codon 175	37	9.2
Codon 245	19	5.0
Codon 248	36	9.0
Codon 273	29	7.4
Codon 282	11	2.9
	Mean	SD	Mean	SD
Age	61.4	10.9	61.6	11.1
Calcium intake (g/day)	1, 343.4	767.4	1, 445.1	825.6
Vitamin D intake (g/day)	11.9	7.8	12.6	8.6
Sunshine exposure (h/wk)	16.8	14.2	17.4	14.2

Note. Abbreviation is as follows: CIMP, CpG Island methylator phenotype.

Dietary plus supplemental calcium reduced risk of rectal cancer overall as well as for all specific tumor mutations with a similar degree of risk reduction (Table 2). There were suggestions that hours of sunshine exposure reduced risk of TP53 mutations, although the association was not statistically significant. VDR was not associated with any specific mutation. None of the study variables were associated uniquely with either transition or transversion mutations (data not shown in Table 2).

TABLE 2.

Associations between calcium, vitamin D, sunshine exposure and VDR genotypes, and rectal cancer mutaitons

		Control		All Cases*			CIMP High			p53 Mutation			KRAS2 Mutation
		n	n	OR	(95% CI)	n	OR	(95% CI)	n	OR	(95% CI)	n	OR	(95% CI)
Calcium diet &   supplements (g)	T1	391	343	1.00		29	1		126	1		67	1.00
	T2	406	347	0.91	(0.73, 1.13)	27	0.95	(0.55, 1.65)	124	0.85	(0.64, 1.12)	92	1.26	(0.90, 1.76)
	T3	395	250	0.60	(0.46, 0.78)	17	0.73	(0.36, 1.47)	87	0.60	(0.42, 0.84)	53	0.71	(0.46, 1.10)
	P trend			<.01			0.41			<.01			0.20
Vitamin D diet &   supplements (g)	T1	391	316	1.00		28	1.00		110	1.00		64	1.00
	T2	407	346	1.09	(0.88, 1.36)	24	0.80	(0.45, 1.43)	127	1.11	(0.84, 1.47)	82	1.28	(0.91, 1.80)
	T3	394	278	0.93	(0.72, 1.20)	21	0.78	(0.42, 1.46)	100	0.95	(0.69, 1.31)	66	1.15	(0.76, 1.73)
	P trend			0.61			0.47			0.79			0.52
Average Sunlight Exposure (hr/day)	T1	395	350	1.00		27	1.00		136	1.00		72	1.00
	T2	413	291	0.84	(0.68, 1.04)	21	0.79	(0.45, 1.40)	103	0.77	(0.59, 1.01)83	1.30		(0.95, 1.79)
	T3	381	295	0.91	(0.73, 1.13)	25	1.06	(0.61, 1.87)	95	0.78	(0.59, 1.03)	56	0.91	(0.64, 1.29)
	P trend			0.37			0.85			0.08			0.68
VDR
BSM1	bb	366	307	1.00		26	1.00		121	1.00		66	1.00
	bB	461	341	0.92	(0.74, 1.14)	21	0.68	(0.39, 1.19)	117	0.82	(0.63, 1.08)	73	1.00	(0.71, 1.40)
	BB	141	112	0.97	(0.72, 1.31)	13	1.32	(0.69, 2.51)	42	0.87	(0.61, 1.26)	32	1.29	(0.86, 1.96)
	P trend			0.68			0.78			0.31			0.36
FOK1	FF	378	283	1.00		18	1.00		102	1.00		64	1.00
	Ff/ff	573	459	1.09	(0.89, 1.33)	41	1.46	(0.85, 2.52)	170	1.07	(0.83, 1.39)	105	1.08	(0.79, 1.47)
	P trend			0.52			0.15		0.38				0.90
CDX2	GG	613	451	1.00		34	1.00		161	1.00		103	1.00
	GA/AA	380	332	1.14	(0.94, 1.38)	26	1.15	(0.69, 1.92)	126	1.18	(0.92, 1.51)	75	1.06	(0.78, 1.43)
	P trend			0.40			0.98			0.21			0.60

^*Odds ratios (OR) and 95% confidence intervals (CI) adjusted forage, sex, BMI, long-term physical activity, pack-years of cigarettes, dietary calcium, and energy intake.

We observed unique associations with specific hotspots of the TP53 gene (Table 3). Assessment of specific mutations for TP53 showed that high levels of calcium had the most protection for point mutations at codon 248. On the other hand, high levels of vitamin D intake were associated with increased risk of having a point mutation at codon 273. Having either the Ff or ff genotype of the Fok1 VDR gene increased risk of a point mutation at codon 245 and 273; having the GA or AA genotype of CDX2 VDR polymorphism increased the odds of having a point mutation at codon 273.

TABLE 3.

Associations between calcium, vitamin D, sunshine exposure, VDR genotype, and specific p53 mutations

		Controls	Codon 248 Mut	p53 Codon 248 vs. Control		Codon 245 Mut	p53 Codon 245 vs. Control		Codon 273 Mut	p53 Codon 273 vs. Control
		n	n	OR	(95% Cl)	n	OR	(95% Cl)	n	OR	(95% Cl)
Calcium diet &   supplements (g)	T1	391	15	1.00		7	1.00		12	1.00
	T2	406	11	0.49	(0.21,1.13)	9	1.36	(0.46, 4.01)	9	0.91	(0.35, 2.34)
	T3	395	10	0.26	(0.09, 0.77)	3	0.44	(0.09, 2.22)	8	1.01	(0.33, 3.09)
	P trend			0.01			0.40			0.99
Vitamin D diet &   supplements (g)	T1	391	10	1.00		6	1.00		6	1.00
	T2	407	12	1.18	(0.49, 2.87)	10	1.99	(0.67, 5.96)	10	2.07	(0.72, 5.99)
	T3	394	14	1.41	(0.54, 3.68)	3	0.67	(0.14, 3.19)	13	3.42	(1.15, 10.17)
	P trend			0.48			0.77			0.02
Average sunlight exposure (hr/day)	T1	395	11	1.00		8	1.00		12	1.00
	T2	413	11	0.90	(0.38, 2.13)	6	0.99	(0.33, 2.98)	9	0.77	(0.31, 1.89)
	T3	381	12	0.97	(0.41, 2.29)	5	0.81	(0.25, 2.58)	7	0.69	(0.26, 1.85)
	P trend			0.95			0.73			0.45
VDR
BSM1	bb	366	9	1.00		9	1.00		10	1.00
	bB/BB	602	21	1.58	(0.70, 3.59)	7	0.54	(0.20, 1.49)	13	0.82	(0.35, 1.92)
	P trend			0.09			0.18			0.76
FOK1	FF	378	10	1.00		2	1.00		4	1.00
	Ff/ff	573	19	1.33	(0.60, 2.92)	13	4.74	(1.05, 21.39)	19	3.34	(1.11, 10.02)
	P trend			0.42			0.12			0.02
CDX2	GG	613	16	1.00		11	1.00		9	1.00
	GA/AA	380	16	1.43	(0.70, 2.93)	5	0.66	(0.22, 1.95)	16	2.85	(1.23, 6.58)
	P trend			0.26			0.48			0.03

Odds ratios (OR) and 95% confidence intervals (CI) adjusted for age, sex, BMI, long-term activity level, pack-years of cigarette smoking, dietary calcium, and energy intake.

Assessment of interaction between VDR genotype and calcium and vitamin D for TP53 and KRAS2 mutations showed nonsignificant interaction (data not shown in tables). Likewise, interactions between calcium, vitamin D, and VDR genotypes and recent use of aspirin/NSAIDs were not statistically significant. Associations were similar for men and women.

DISCUSSION

Our data support a significant inverse association between calcium and rectal tumors overall as well as for TP53 tumor mutations specifically. We also observed a trend toward lower risk of a TP53 mutation with increased hours of sunshine exposure. Specific point mutations of the TP53 gene were associated with the Fok1 and CDX2 VDR genotypes.

Many epidemiological studies have shown an association between high levels of dietary calcium intake and reduced risk of colorectal cancer, with stronger effects for more distal colon and rectal cancers and adenomas reported for some but not all studies (7,8,30–33). Fewer studies have assessed vitamin D and sunshine exposure with rectal cancer. We have previously reported that both calcium and vitamin D have been associated with reduced risk of rectal cancer and that the association may be modified by VDR genotype. Others have failed to observe a similar association (8). A review of over 60 observation studies likewise did not observe an association between vitamin D and rectal cancer (7). However, studies have not examined associations with tumor markers, and our findings suggest that associations differ by marker examined. However, vitamin D has been shown to reduce rectal cell proliferation (34). A study of Miller and colleagues (35) showed that both calcium and vitamin D were associated with apoptotic score, providing additional evidence for the involvement of calcium and vitamin D in CRC carcinogenesis.

The strongest associations were with TP53 in our study. TP53 is one of the more commonly mutated genes in rectal tumors (36) and is involved in important cellular functions including apoptosis, DNA repair, cell-cycle control, and differentiation (37). Furthermore, specific point mutations such as an arginine residue at codon 248 in the L3 loop of the core domain of the gene is thought to play a critical role in DNA binding; a glycine at codon 245 also in the L3 loop allows the L3 loop to assume unique conformations; arginine at codon 273 in the loop-sheethelix motif (38). Furthermore, mutations in these specific sites have been linked with specific exposures that alter cancer risk. For instance, polycyclic aromatic hydrocarbons and benzo[a] pyrene has been associated with G>T transversions and TP53 mutations at codons 157, 248, and 273 (37,39); aflatoxin has been associated with mutations at codon 249; sunlight exposure has been associated with CC to TT transition mutations (37). Whereas we observed a significant inverse association with TP53 for calcium and sunshine exposure overall, point mutation at codons 245 and 273 were associated with an increased risk of these factors.

This is the largest study to date to report on specific rectal tumor markers and calcium, vitamin D, sunshine exposure, and 3 VDR polymorphisms as they relate to tumor markers. This is 1 of the largest studies of rectal cancer conducted to date, and we were therefore able to look specifically at rectal vs. colorectal cancer. We were able to examine KRAS2, TP53, and CIMP in these rectal tumors in combination with diet and genetic factors. However, although we examined the hot spots of both the KRAS2 and TP53 genes, accounting for most mutations, it is possible that mutations were missed. Likewise, our markers for CIMP have been shown to be associated with cigarette smoking and other factors (27,40,41), but other markers could have expanded our ability to detect associations. There are limitations to our study, including that we have made many comparisons and therefore some findings may be spurious in nature. Additionally, it is important that others confirm these findings in rectal cancer and not just in colon cancer.

Our data add further support for the role of calcium, vitamin D, and the VDR gene in the etiology of rectal cancer. Through our examination of tumor markers, it appears that the pathway associated with calcium, vitamin D, and VDR in rectal tumors is most strongly associated with TP53 mutations. Although these findings are intriguing, it is important that other confirmatory studies be conducted.