Amongst men in the United States, prostate cancer is the most common malignancy. While incidence rates of prostate cancer have decreased over the years [1
], studies have shown African American (AA) men to develop prostate cancer at a rate 1.5–1.9 times higher than their European American (EA) counterpart [2
]. These racial differences are further emphasized by the increased diagnosis of aggressive prostate cancer [4
]. Demographic characteristics, such as family history, socioeconomic status, access to medical care, other comorbidities, and diet and lifestyle have been shown to contribute to the increased burden of prostate cancer in AA men [2
]. Recently, however, studies have focused on differences in serum 25-hydroxyvitamin D (25(OH)D) concentrations as a source of the disparate trends seen in this disease.
Critical to overall health, 25(OH)D plays a role in bone mineralization, diabetes mellitus, and multiple sclerosis [9
]. The main source of 25(OH)D is derived from sunlight ultraviolet (UV)-B rays, accounting for over 90% of circulating levels [11
]. High melanin, commonly seen in ethnic groups with dark skin, such as AA men, reduces the amount of UVB radiation absorbed in the skin, thus decreasing the concentration of 25(OH)D and increasing susceptibility to developing vitamin D deficiencies [14
]. In the Health, Aging and Body Composition Study, comparison between AAs and EAs showed only 16% of older AA participants had serum 25(OH)D levels over 30 ng/mL, compared to 44% in EAs [17
]. Data from the Prostate Cancer Prevention Trial determined AA men with higher vitamin D levels see a reduced risk in high-grade disease [18
], while results in Afro-Caribbean men residing in the Caribbean indicate vitamin D insufficiency may contribute to increased prostate cancer risk [19
]. Moreover, molecular studies suggest deficiencies in vitamin D overtime may lead to progression from pre-clinical to clinically aggressive forms of prostate cancer [21
In addition to 25(OH)D, calcium intake is associated with risk of prostate cancer [22
], where high levels are associated with an increase in metastatic disease [24
]. In its association with vitamin D, results indicate that genes responsible for calcium absorption are regulated via the vitamin D receptor (VDR
]. In prostate cells that contain VDRs, response to the active form of vitamin D, 1-25 dihydroxyvitamin D (1,25(OH)2
D) increases differentiation and decreases proliferation [27
]. Conversely, high calcium has been shown to promote the proliferation of prostate cancer cells through calcium sensing receptors [28
]. Although studies have examined associations of calcium and 25(OH)D on prostate cancer separately, few studies have accounted for the effects of both in a population of AA men.
Given the emergence of 25(OH)D and calcium as modifiable risk factors in prostate cancer development in AA men, and the paucity of studies in this racial population, we utilized a case-only study to explore associations between serum 25(OH)D, calcium, VDR genetic variants, and aggressive prostate cancer in AA men.
In this case-only study of 58 AA men with non-aggressive prostate cancer and 46 AA men with aggressive prostate cancer, we investigated associations of serum 25(OH)D and total calcium intake with aggressive prostate cancer risk. Our data indicated deficient levels of 25(OH)D significantly increased the risk of aggressive prostate cancer, compared to non-aggressive cases. We did not find an association between calcium and aggressive prostate cancer, however total calcium levels above 800 mg modified the association between deficient 25(OH)D and aggressive disease, where we reported a significant interaction.
While laboratory and epidemiological studies have shown evidence that supports decreased risk of prostate cancer with higher levels of serum 25(OH)D, recent studies have shown contradictory evidence. In the Prostate Cancer Prevention Trial, logistic regression models estimated increased overall prostate cancer risk with higher levels of serum 25(OH)D (OR: 1.18, 95% CI: 0.91–1.53, p
-value = 0.08) [18
]. Amongst aggressive disease with Gleason score 8–10, the positive association disappeared (OR: 0.50, 95% CI: 0.20–1.22, p
-value = 0.22) [18
]. Results from the Alpha-Tocopheral Beta-Carotene study, a case-control study of Finnish men, showed positive associations between increased serum 25(OH)D and overall prostate cancer (ORQ5vs.Q1
: 1.36, p
-trend = 0.03) [32
]. Associations remained similar when stratified by aggressiveness (OR aggressive Q5vs.Q1
: 1.70, p
-trend = 0.02) [32
]. In a study done in the Health Professionals Follow-up Study, 25(OH)D deficiency showed inverse associations with prostate cancer risk (OR: 0.62, 95% CI: 0.43–0.91) [33
]. These studies, however, were conducted in populations that consisted of majority EA men. Amongst AA men, studies show multiple variables, including AA race and lack of vitamin D supplementation, increase the risk of 25(OH)D deficiency [9
], subsequently increasing the risk of prostate cancer [34
]. Interestingly, a recent study done in Jamaican men of African ancestry reported a positive association between high levels of 25(OH)D and total prostate cancer risk (ORQ3vs.Q1
: 2.47, 95% CI: 1.20–4.90, p
-value = 0.01) [35
]. Analysis of FFQ results demonstrate Jamaican men report higher levels of serum 25(OH)D, averaging ≥ 30 ng/mL, compared to mean levels among AA men in the United States. The population differences seen between American AA men and Jamaican men could be explained by the U-shaped associations between aggressive prostate cancer risk and serum 25(OH)D. Serum concentrations below 20 ng/mL, as well as above 30 ng/mL, in AA men may have similar effects on disease risk. Further studies must be conducted to examine these trends.
Calcium intake has also played a significant role in prostate cancer development. Results from the California Collaborative Prostate Cancer Study show a 54% decrease in risk of aggressive disease amongst AA men with low levels of calcium intake [36
]; EA men showed a 35% decrease in risk. A study in the San Francisco Bay and LA County areas showed positive associations between high dietary, total, and supplemental calcium intake and prostate cancer risk, in both total and aggressive disease [37
]. The authors postulated the association between prostate cancer risk and high calcium may be an effect of rs11568820
]. Studies have recognized the importance of the vitamin D receptor in disease development. As calcium mediates the effects of vitamin D in the body, variants of the VDR transcription factor binding sites have been shown to play an important role in the expression of genes that regulate prostate cancer development [38
, and Fok1
, have been examined closely for their effects on prostate cancer risk [36
]. Similar to other studies [39
], we did not find associations between rs1544410
, and prostate cancer risk [39
]. It is rs11568820
, however, which has shown interesting effects.
Our study found high total calcium intake modified the association between deficient 25(OH)D levels and aggressive prostate cancer, increasing the risk of disease. Results from the Alpha-Tocopherol, Beta-Carotene (ATBC) Study contrast our results. Here, associations between high levels of serum 25(OH)D and total prostate cancer were stronger among men with higher intakes of total calcium (OR:Q5vsQ1
low calcium: 1.15, 95% CI: 0.75–1.75; ORQ5vsQ1
high calcium: 1.82, 95% CI: 1.20–2.76, p
-interaction = 0.06) [32
]. Moreover, our results are also in contrast with the report from Steck et al., which showed an inverse association between the highest tertile of serum 25(OH)D and aggressive prostate cancer amongst men with high levels of calcium [41
]. As laboratory studies have shown that calcium promotes growth of prostate cancer cells [42
], the results of our study are consistent with these findings. The differences we see with results from the ATBC Study could be due to race, but this does not explain the differences we see with Steck et al., as that study was stratified by race. A possible explanation for the modification by high total calcium may be related to genetic variation of the VDR
that plays a role in calcium absorption. Located in the 5′ regulatory region of the VDR
has been postulated to affect the CDX2
transcription factor affinity for the VDR
]. These effects may induce changes to VDR
expression in prostate cells and, thus, development of aggressive prostate cancer. rs11568820
increased the risk of aggressive prostate cancer amongst 25(OH)D deficient levels, consistent with findings presented by Rowland et al. [37
]. Here, the dominant A allele of rs11568820
was associated with an increased risk of aggressive prostate cancer [37
]. Studies, such as the Health Professionals Follow-up Study, show results in total prostate cancer only, where men with the rs11568820
variant A allele who are deficient in 25(OH)D have a significant reduction in total prostate cancer risk (OR: 0.41, 95% CI: 0.21–0.82) [33
]. Similar trends were seen among men with aggressive prostate cancer (OR: 0.18, 95% CI: 0.05–0.63) [33
]. These differences in trends could be due to racial disparities, as our study was conducted in a population of all AA men, while the Health Professionals Follow-up Study consists of majority EA men. Moreover, these studies focus more on total prostate cancer, which is different from our unique focus on aggressive disease.
Interestingly, differences in calcium intake between AA and EA men also encompass differences in the rs11568820
A allele expression. It is known that the Cdx2 A
allele is more strongly associated with African ancestry [37
]. Upon binding to the A allele on the VDR
, the rs11568820
transcription factor induces the expression of genes associated with calcium absorption, including calbindin. Increased calcium binds to calcium-sensing receptors on prostate cells, regulating proliferation and differentiation and, subsequently, the development of prostate cancer [44
]. Given this pathway, further studies are needed to clarify the role increased calcium absorption plays within different races.
Our results showed similar effect sizes between two of our analyses: aggressive prostate cancer and 25(OH)D deficiency when rs11568820 was added to the regression model, and aggressive disease and 25(OH)D deficiency without rs11568820 in the model. Together, this data indicates a need for additional investigations into the association between aggressive prostate cancer and VDR SNPs, such as rs11568820.
Several limitations of our study should be noted. As a case-only study, our sample size of men was very small, with only 46 men with aggressive disease and 58 men with non-aggressive disease. Given the small number of cases being assessed, the results from our study must be cautiously interpreted, as we acknowledge this study is underpowered. Although other studies have assessed these outcomes in similar cohorts of similar sizes, this speaks to the need for studies that oversample minority and underrepresented participants, to increase the power to detect differences if they in fact exist. Our measurements of serum 25(OH)D and total calcium intake relied upon a single time point, which may have created an inadequate reflection of true levels. Another possible limitation is the use of the Block FFQ, which limits our accuracy of assessing dietary intake and may have introduced recall bias in the reporting. However, the Block FFQ has been validated for use in our study, and has been used in multiple epidemiologic studies assessing diet and disease associations. Strengths of our study include the oversampling of AA men with a focus on aggressive prostate cancer. Additionally, few epidemiologic studies have examined associations between serum 25(OH)D, calcium intake, and aggressive prostate cancer risk in a population of AA men.