World J Urol DOI 10.1007/s00345-013-1201-5
ORIGINAL ARTICLE
The relationship between solar UV exposure, serum vitamin D levels and serum prostate-specific antigen levels, in men from New South Wales, Australia: the CHAMP study Visalini Nair-Shalliker • David P. Smith • Mark Clements • Vasikaran Naganathan Melisa Litchfield • Louise Waite • David Handelsman • Markus J. Seibel • Robert Cumming • Bruce K. Armstrong
•
Received: 19 May 2013 / Accepted: 23 October 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose We aim to determine the relationship between season, personal solar UV exposure, serum 25(OH)D and 1,25(OH)2D and serum prostate-specific antigen (PSA) levels. Methods Questionnaire data and blood samples were collected at baseline from participants of the Concord Health and Ageing in Men Project (n = 1,705), aged 70 and above. They were grouped as men ‘free of prostate disease’ for those with no record of having prostate cancer, benign prostatic hyperplasia, or prostatitis and with serum Electronic supplementary material The online version of this article (doi:10.1007/s00345-013-1201-5) contains supplementary material, which is available to authorized users. V. Nair-Shalliker (&) D. P. Smith Cancer Research Division, Cancer Council New South Wales, 153 Dowling Street, Wooloomooloo, Sydney, NSW 2011, Australia e-mail:
[email protected] D. P. Smith e-mail:
[email protected] V. Nair-Shalliker R. Cumming B. K. Armstrong Sydney School of Public Health, The University of Sydney, Sydney, Australia e-mail:
[email protected] B. K. Armstrong e-mail:
[email protected] D. P. Smith Griffith Health Institute, Griffith University, Gold Coast, QLD, Australia M. Clements Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden e-mail:
[email protected]
PSA levels below 20 ng/mL, and ‘with prostate disease’ for those with a record of either of these diseases or with serum PSA levels 20 ng/mL or above. Personal solar UV exposure (sUV) was estimated from recalled hours of outdoor exposure and weighted against ambient solar UV radiation. Sera were analysed to determine levels of PSA, 25(OH)D and 1,25(OH)2D, and analysed using multiple regression, adjusting for age, BMI and region of birth. Results The association between sUV and serum PSA levels was conditional upon season (pinteraction = 0.04). There was no direct association between serum PSA and 25(OH)D in both groups of men. There was a positive association between serum PSA and 1,25(OH)2D in men with prostate disease (mean = 110.6 pmol/L; pheterogeneity = 0.03), but there was no such association in V. Naganathan M. Litchfield L. Waite R. Cumming Centre for Education and Research on Ageing, Concord Hospital, University of Sydney, Sydney, NSW, Australia e-mail:
[email protected] M. Litchfield e-mail:
[email protected] L. Waite e-mail:
[email protected] D. Handelsman M. J. Seibel ANZAC Research Institute, The University of Sydney, Sydney, NSW, Australia e-mail:
[email protected] M. J. Seibel e-mail:
[email protected] M. J. Seibel Department of Endocrinology and Metabolism, Concord Hospital, Sydney, Australia
123
World J Urol
men free of prostate disease (mean = 109.3 pmol/L; pheterogeneity = 0.8). Conclusion The association between PSA and sUV may only be evident at low solar UV irradiance, and this effect may be independent of serum vitamin D levels. Keywords Season Solar UV 25(OH)D 1,25(OH)2D PSA Australia Abbreviations sUV Personal solar UV exposure NSW New South Wales CHAMP The Concord Health and Ageing in Men Project PSA Prostate-specific antigen PTH Parathyroid hormone ERSPC European Randomised study of Screening for Prostate Cancer
Introduction Prostate-specific antigen (PSA) is a biomarker for the detection of prostate diseases [1]. However, elevated PSA levels may also occur in non-cancer conditions such as benign prostatic hyperplasia (BPH), trauma, infections, including urinary infection, and prostatitis, and it may be elevated following ejaculation [1]. The Spanish arm of the European Randomised study of Screening for Prostate Cancer (ERSPC) showed PSA levels to be low in seasons when ambient temperatures were warm compared with seasons with high temperatures, thus coinciding with seasons when serum vitamin D levels are high and low, respectively, in cancer-free men [3]. Conversely, the French arm of the ERSPC reported that PSA levels were highest in months with the most hours of sunshine [4]. Furthermore, PSA doubling time in men with low-grade disease was longest in spring and summer, when vitamin D levels are highest, than in autumn/winter, suggesting a role for vitamin D in mediating serum PSA levels [2]. In vitro evidence supports a protective role for vitamin D in prostate cancer [5]. Thus, lower PSA levels when ambient solar radiation levels are high might indicate an effect of vitamin D in suppressing growth of early prostate cancer. There is, however, no direct evidence relating serum vitamin D levels to PSA levels in apparently healthy men [6–10]. The current study explored the relationship between season, sUV and serum PSA directly, and whether serum 25(OH)D and 1,25(OH)2D levels are related to serum PSA levels, in men with or without self-reported prostate disease. The role of parathyroid hormone (PTH) as a covariate in this relationship was also examined [11].
123
Methods We used data from the Concord Health and Ageing in Men Project (CHAMP), which is a population-based longitudinal study investigating the health of older men [12, 13]. CHAMP was approved by the Sydney South West Area Health Service Human Research Ethics Committee— Concord Repatriation General Hospital Zone, and the current study was approved by Cancer Council New South Wales Ethics Committee. Study sample Men of 70 years and older were identified through the NSW electoral roll and recruited between January 2005 and June 2007. Of apparently eligible men, 47 % participated [12]. All participants were asked to complete a questionnaire on demographic characteristics, prostate health, and type, duration and frequency of physical activity, prior to their first clinic visit. Men were divided into two groups, based on their selfreported prostate disease and their baseline PSA measurement: (1) men ‘free of prostate disease’ for those with no record of having prostate cancer, BPH, or prostatitis and with serum PSA levels below 20 ng/mL and (2) men with a record of either of these diseases or with serum PSA levels 20 ng/mL or above. Serum analysis Fasting blood samples were collected from participants on the morning of their clinic visit. Sera were divided for PSA, vitamin D and PTH analyses. Total PSA analyses were carried out by the Central Sydney Area Health Service laboratory at Concord Repatriation General Hospital on the day of collection, using electrochemiluminescence immunoassay on the Modular Analytics E170 (Elecsys module) immunoassay analyser (Roche Diagnostics GmbH, D-68298 Mannheim). The intra-assay coefficient of variation was 2.73 % at 4 lg/L and 2.71 % at 37.6 lg/L. The standard reference range for the PSA assay was 0.0–4.0 ng/mL. Sera were stored frozen at -808C until the end of sample collection, after which remaining analyses were performed simultaneously, in duplicate. Serum 25OH and 1,25(OH)2D levels were measured by manual RIA using single batch reagents (DiaSorin Inc., Stillwater, MN). The assay for 25(OH)D has a sensitivity of\1.5 ng/mL with an intra-assay precision of 7.6 % and an inter-assay precision of 9.0 %. The assay for 1,25(OH)2D has a sensitivity of \2 pg/mL, an intra-assay precision of 7.7 % and an interassay precision of 12.3 %.
World J Urol
Serum levels of intact PTH were determined by a twosite chemiluminescent ELISA on an Immulite 1,000 analyser (Diagnostic Products, Los Angeles, CA). The assay has a sensitivity of 1 pg/ml, an intra-assay precision of 5.5 % and an inter-assay precision of 7.9 %, and the laboratory reference range is 23–66 pg/ml. Data analysis Prostate-specific antigen levels were log-transformed and the following independent variables were derived accordingly: Personal weekly solar UV exposure (sUV) Summary measures were calculated using ambient solar UV radiation from 8 am to 5 pm (expressed as number of Standard Erythermal Doses (SED); 1 SED equals 10 mJ/ cm2 of erythemally effective UV) obtained from the Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) for Sydney for the date of the blood collection. Estimated ambient UV radiation for the 4 weeks before each man’s blood collection was combined with recalled number of hours spent outdoors each week, which was estimated from the following questions, to provide an estimate of weekly sUV: (1) Over the past 7 days, how often did you engage in light, moderate or strenuous sports or recreational activities? (2) What were those activities? (3) On average, how many hours per day did you engage in those activities? The answers to question (ii) were used to classify each physical activity as ‘entirely indoor’, ‘both indoor and outdoor’ or ‘entirely outdoor’ and given a score of 0.04, 1 and 2, respectively. Men who remained inside and those who did not perform any physical activity were identified as ‘entirely indoor’ and were scored 0.04 due to some incidental sun exposure they would have received. This score was then multiplied by the number of hours of activity per day, their frequency per week and the average daily ambient UV irradiance for the 4 weeks before the date of blood draw, to obtain an estimate of weekly sUV exposure in mJ/cm2, and divided into quartiles for statistical analysis: B109.9 mJ/cm2 (reference), 102–595.9 mJ/ cm2, 596–1,805.9 mJ/cm2 and 1,806 mJ/cm2 and more.
median as follows: deficient (25(OH)D \ 30 nmol/L, median = 26 nmol/L); insufficient (25(OH)D 30–49.9 nmol/L, median = 41 nmol/L); sufficient (25(OH)D 50–75 nmol/L, median = 61 nmol/L); and highly sufficient (25(OH)D [ 75 nmol/L, median = 87 nmol/L). Categories for serum calcitriol (1,25(OH)2D) were: low (1,25(OH)2D) \ 50 pmol/L, median = 35 pmol/L); moderate (1,25(OH)2D) 50–150 pmol/L, median = 91 pmol/L; and high (1,25(OH)2D) [ 150 pmol/L, median = 191.5 pmol/L). Statistical analysis We assessed variation in serum PSA over the course of an year using Stolwijk’s method [16], which incorporates the sine and cosine of 2p month/12, addressed the differences in monthly mean serum PSA levels. The model was initially tested for a single-harmonic mean with a 12 month period. However, as the graph of monthly mean PSA showed evidence of a second peak within the 1 year, we also incorporated a second harmonic with sine and cosine of 2p month/6, having a period of 6 months. We examined relationships between PSA and other sun exposure variables using multiple linear regression. We examined the relationship between quartiles of sUV and log-transformed serum PSA using a likelihood ratio test, where the lowest quartile of sUV was used as the reference category. Season is a surrogate for ambient UV exposure and, therefore, in a separate model, we examined seasonal variation in log-mean PSA levels during the course of an year. Season was tested for interaction with sUV. All models were adjusted for age, BMI and region of birth, which were categorical variables. The variation in levels of 25(OH)D and 1,25(OH)2D as functions of season and solar UV were examined separately, and their associations with log-transformed PSA were also examined. Regression parameters were expressed as per cent change in PSA levels with respect to a specified reference value. An increase in PSA levels was denoted by a plus (?) sign and a decrease by a minus (-) sign. All analyses were performed using SAS software version 9.1.
Season of blood draw
Results
Blood drawn from December to February was grouped as summer, March to May as autumn, June to August as winter (reference) and September to November as spring.
Characteristics of study participants
Serum vitamin D Serum vitamin D levels were grouped according to clinical relevance [14, 15], and each group was represented by its
Of the 1,705 participants, 67 were excluded due to missing data. There were 922 men free of prostate disease and 716 men with prostate disease in the final analysis (Supplementary Figure 1). There were no significant differences in the basic characteristics, between men free of prostate disease and
123
123
n
242
Spring
1.8
195
91
Geometric mean
p heterogeneity
191.5
1.6
138
497
35 (ref)
1,25 (OH)2D 1.6
1.9
154
87
1.5 1.7
93
297 330
1.6
Median PSAa (ng/mL)
41 61
p heterogeneity
a
2.4
2.3 2.7
2.7
26 (ref)
25(OH)D
Vitamin D
264 223
2.4
2.5
2.7
2.6
Mean PSAa (ng/mL)
Pherterogeneity = 0.16
193
Autumn Winter–reference
Q4 ([1,806 mJ/cm2)
Summer
216 230
Q3 (596 to 1,805.9 mJ/cm2)
n
223
Q2 (102 to 595.9 mJ/cm2)
Season
218
Q1(ref) (\109.9 mJ/cm2)
a
mean PSA (ng/mL)
Pheterogeneity = 0.11
n
Solar UV exposure
Free of prostate disease
0.8
7.2 (-13.5, 32.9)
2.3 (-14.9, 23.1)
0
0.9
3.7 (-19.8, 34.2)
1.3 (-19.4, 27.4) 3.5 (-17.6, 30.0)
0
% change (95% CI)
-9.7 (-24.4, 7.8)
-14.7 (-28.2, 1.4) 1.0
1.8 (-15.7,22.8)
% change (95% CI)
-8.3 (-23.8, 10.4)
-2.6 (-19.0, 17.2)
13.9 (-5.2, 36.9)
0
% change (95% CI)
70
153
384
115
120
240 265
n
5.6
3.5 3.8
3.8
Mean PSAa (ng/mL)
2.2
2.0
1.5
2.1
1.6 2.0
2.5
Median PSAa (ng/mL)
Pherterogeneity = 0.8
184
209 177
146
n
3.5
3.2
4.2
6.0
Mean PSAa (ng/mL)
Pheterogeneity = 0.50
176
182
173
174
n
With prostate disease
0.03
63.2 (12.1,137.7)
42.5 (3.2,97.0)
0
0.1
-5.6 (-41.1,51.4)
-24.7 (-50.8,15.3) 4.7 (-31.3,59.6)
0
% change (95% CI)
-10.7 (-35.2,23.1)
4.7 (-23.1,42.6) 0
-0.9 (-29.6,39.3)
% change (95% CI)
-20.1 (-42.7,11.3)
-16.5 (-39.8,15.9)
-20.0 (-42.4,11.2)
0
% change (95% CI)
Table 1 Association between personal solar UV, season, serum 25(OH)D and 1,25(OH)2D, and per cent change in serum PSA levels, after adjusting for age, region of birth and BMI
World J Urol
World J Urol Table 2 The association between UV(J/m2) and log serum PSA within each season: Per cent increase (?)/decrease (-) in log PSA levels with increasing quartiles of sun exposure, after adjusting for age, region of birth and BMI Free of prostate disease
With prostate disease
Season
Adjusted % change
(95 % CI)
p value
Adjusted % change
Summer
1.6
-9.8, 14.5
0.8
-13.1
Autumn
6.2
-4.9, 18.6
0.3
11.6
Winter Spring
-13.5 -13.3
-25.2, 0.04 -22.7, -2.9
0.05 0.01
p-interaction*
0.04
-5.5 -21.6
(95 % CI)
p value
-32.3, 11.7
0.3
-8.1, 35.4
0.3
-26.4, 21.4 -36.2, -3.6
0.7 0.02
0.20
* Interaction between quartiles of sun exposure and season of blood draw
men with prostate disease, except age when blood was drawn (Supplementary Table 1). The mean total serum PSA levels were higher in men with prostate disease compared with men with no prostate disease after adjusting for age, BMI and region of birth (data not shown). Associations between serum PSA and variables other than season, sun exposure and vitamin D metabolites are shown in Supplementary Table 2. In men free of prostate disease, there were associations between mean PSA levels and increasing age (p = 0.02), BMI (p = 0.05), region of birth (p = 0.0002) and in men with a family history of prostate cancer (p = 0.02). In men with prostate disease, only region of birth was significantly associated with mean PSA levels (p = 0.01). Personal solar UV exposure, season and with serum PSA There were 46 men with missing data for sun exposure. The annual monthly mean UV was lowest in June and July and highest in December and January, and its relationship with mean serum PSA levels is shown in Supplementary Figure 2. Assuming one sine/cosine cycle in a single year and adjusting for age, BMI and region of birth, no significant association was found in the monthly mean PSA levels (likelihood ratio test, p = 0.95) in either group of men [12]. There was, however, a suggestion of a possible second annual peak (in the warmer months), and the model was fitted for two annual cycles; the association between time in the year and PSA remained non-significant (likelihood ratio test for all four harmonics, p = 0.39). A regression analysis to test the relationship between serum PSA levels and quartiles of sUV showed no inverse association, after adjusting for age, BMI and region of birth, in both groups of men (Table 1). Sensitivity analyses were undertaken using median values for the quartiles and the log of sUV as an explanatory variable, with very similar conclusions (p value [0.05). Season is commonly used as a surrogate for UV exposure where summer and winter represent the seasons with
highest and lowest UV exposures, respectively. The recruitment of participants in this study was even across the seasons (Table 1). There was no significant seasonal variation in mean PSA levels in either group (p heterogeneity [ 0.05). There was evidence of a weak interaction between sUV and season (pinteraction = 0.04; Table 2), in men free of prostate disease. There were falls in PSA levels with increasing quartiles of sUV in winter (-13.5 %; p value = 0.05) and in spring (-13.3 %; p value = 0.01), but no association between the two in summer and autumn. In men with prostate disease, this fall was seen only in spring (-21.6 %; p value = 0.02). The relationship between sUV, season and serum vitamin D There were 23 men with missing data for 25(OH)D and 111 men with missing data for 1,25(OH)2D. Both 25(OH)D and 1,25(OH)2D were positively associated with sUV in both groups of men (Supplementary Table 3). The relationship between vitamin D and serum PSA There was no association between serum PSA and either 25(OH)D (p heterogeneity = 0.9) or 1,25(OH)2D (p heterogeneity = 0.8; Table 1) in men free of prostate disease. There was no association between serum levels of 25(OH)D and PSA in men with prostate disease, but mean PSA levels were positively associated with increasing 1,25(OH)2D levels (p value = 0.03), in this group.
Discussion This is the first study to address the relationship between personal sUV and serum PSA levels. Previous reports, from Spain and France, on the association between sun exposure and serum PSA levels, in cancer-free men, which were based on ambient measures of solar exposure, have
123
World J Urol
been contradictory [3, 4]. Despite higher levels of ambient sUV irradiance in Spain (40.4oN) than in France (48.9oN), the incidence of melanoma is higher in France than in Spain [17], suggesting not only genetic differences but also behavioural differences in the populations at risk of skin cancer. Thus, the use of ambient rather than personal estimates of sUV could explain the inconsistency in seasonal patterns of PSA variation between these previous studies. We found the relationship between sUV and serum PSA concentration may be conditional on ambient solar UV. Serum PSA levels were inversely associated with sUV in winter and spring, but not in summer or autumn, suggesting that this association may only be evident when ambient UV levels are low. This contradicts a previous Canadian report of a significant association between PSA and season, where PSA doubling times were greatest in summer and autumn in men with low-grade disease [2]. This pattern was attributed to a beneficial effect of vitamin D as higher doubling time occurred in seasons when serum 25(OH)D levels are high. This hypothesised beneficial effect of vitamin D has been supported by evidence from a small clinical trial, which also showed that 1,25(OH)2D was able to slow the rate of increase in PSA concentration in prostate cancer patients [7, 18]. Solar UVB exposure is essential for epidermal production of vitamin D [19]. Dietary contribution is small due to its low content in food [20], and intake of supplementation is only recommended for populations at risk of vitamin D deficiency; in our cohort, 13.9 % of the men reported taking vitamin D supplementation. Although we found no association between serum levels of either metabolite with total PSA levels, in men free of prostate disease, serum 1,25(OH)2D levels, which is the highly regulated active form, were positively associated with PSA levels, in men with prostate disease. This subtle difference may have biological relevance. Studies have shown that in concurrence with its tumour inhibitory properties, 1,25(OH)2D can also up-regulate androgen receptors in androgen-responsive cell lines, thereby enhancing the expression of androgen response genes, which include PSA [21–23]. Consequently, increasing 1,25(OH)2D levels will increase serum PSA levels [24, 25]. Thus, this positive association between 1,25(OH)2D and PSA in the present study may indicate the involvement of an androgen-responsive mechanism. Parathyroid hormone is also a systemic mediator of serum 25(OH)D levels, but showed no association with PSA in the current analysis, contrary to one other study [11]. There are limitations in the accuracy of the sUV measure, which we obtained by combining questions on outdoor activity. However, this measure was associated
123
positively with both serum 25(OH)D and 1,25(OH)2D, as would be expected if it measures personal exposure to solar UVR. Another potential weakness is the accuracy in selfreporting prostate problems especially in the presence of indolent prostate disease. Most error in self-reporting, however, is in the direction of over-reporting of prostate cancer incidence (Smith unpublished data), due to misinterpretation of high PSA as cancer. Our classification for men with prostate disease which includes men with BPH, prostatitis and prostate cancer, and in addition also include men with levels of 20 ng/mL or greater, may to some extent resolve this issue. As is common in aetiological epidemiology, we did not adjust for multiple comparisons [26]. Given that some of the associations were close to the 5 % critical p value, we suggest caution when interpreting these findings.
Conclusion An inverse association between sUV and serum PSA concentration is evident only in seasons of low UV. The positive association observed between serum levels of 1,25(OH)2D and PSA, in men with prostate disease, suggests altered vitamin D metabolism in the presence of prostate disease. Acknowledgments This study was funded by the Cancer Council New South Wales project Grant (512513). Conflict of interest authors.
There is no conflict of interest for any of the
References 1. Schroder FH (2009) Review of diagnostic markers for prostate cancer. Recent Results Cancer Res 181:173–182 2. Vieth R, Choo R, Deboer L, Danjoux C, Morton G, Klotz L (2006) Rise in prostate-specific antigen in men with untreated low-grade prostate cancer is slower during spring-summer. Am J Ther 13:394–399 3. Lujan GM, Pascual MC, Rodriguez GN, Garcia Mediero JM, Pascual DT, Paez BA, Berenguer SA (2006) Impact of the weather on the serum levels of prostatic specific antigen (PSA). Arch Esp Urol 59:247–252 4. Salama G, Noirot O, Bataille V, Malavaud S, Rebillard X, Villers A et al (2007) Seasonality of serum prostate-specific antigen levels: a population-based study. Eur Urol 52(3):708–714 5. Beer TM, Garzotto M, Park B, Mori M, Myrthue A, Janeba N, Sauer D, Eilers K (2006) Effect of calcitriol on prostate-specific antigen in vitro and in humans. Clin Cancer Res 12:2812–2816 6. Ahn J, Peters U, Albanes D, Purdue MP, Abnet CC, Chatterjee N, Horst RL, Hollis BW, Huang WY, Shikany JM, Hayes RB (2008) Serum vitamin D concentration and prostate cancer risk: a nested case-control study. J Natl Cancer Inst 100:796–804
World J Urol 7. Gross C, Stamey T, Hancock S, Feldman D (1998) Treatment of early recurrent prostate cancer with 1,25-dihydroxyvitamin D3 (calcitriol). J Urol 159:2035–2039 8. Platz EA, Leitzmann MF, Hollis BW, Willett WC, Giovannucci E (2004) Plasma 1,25-dihydroxy- and 25-hydroxyvitamin D and subsequent risk of prostate cancer. Cancer Causes Control 15:255–265 9. Travis RC, Crowe FL, Allen NE, Appleby PN, Roddam AW, Tjonneland A, Olsen A, Linseisen J, Kaaks R, Boeing H, Kroger J, Trichopoulou A, Dilis V, Trichopoulos D, Vineis P, Palli D, Tumino R, Sieri S, Bueno-de-Mesquita HB, van Duijnhoven FJ, Chirlaque MD, Barricarte A, Larranaga N, Gonzalez CA, Arguelles MV, Sanchez MJ, Stattin P, Hallmans G, Khaw KT, Bingham S, Rinaldi S, Slimani N, Jenab M, Riboli E, Key TJ (2009) Serum vitamin D and risk of prostate cancer in a case-control analysis nested within the European Prospective Investigation into Cancer and Nutrition (EPIC). Am J Epidemiol 169:1223–1232 10. Woo TC, Choo R, Jamieson M, Chander S, Vieth R (2005) Pilot study: potential role of vitamin D (Cholecalciferol) in patients with PSA relapse after definitive therapy. Nutr Cancer 51:32–36 11. Skinner HG, Schwartz GG (2009) The relation of serum parathyroid hormone and serum calcium to serum levels of prostatespecific antigen: a population-based study. Cancer Epidemiol Biomarkers Prev 18:2869–2873 12. Cumming RG, Handelsman D, Seibel MJ, Creasey H, Sambrook P, Waite L, Naganathan V, Le CD, Litchfield M (2009) Cohort profile: the Concord health and ageing in men project (CHAMP). Int J Epidemiol 38:374–378 13. Litchfield MJ, Cumming RG, Smith DP, Naganathan V, Le Couteur DG, Waite LM, Blyth FM, Handelsman DJ (2012) Prostate-specific antigen levels in men aged 70 years and over: findings from the CHAMP study. Med J Aust 196:395–398 14. Need AG, Kemp A, Giles N, Morris HA, Horowitz M, Nordin BE (2002) Relationships between intestinal calcium absorption, serum vitamin D metabolites and smoking in postmenopausal women. Osteoporos Int 13:83–88 15. Nowson CA, McGrath JJ, Ebeling PR, Haikerwal A, Daly RM, Sanders KM, Seibel MJ, Mason RS (2012) Vitamin D and health
16.
17.
18.
19. 20. 21. 22.
23.
24.
25.
26.
in adults in Australia and New Zealand: a position statement. Med J Aust 196:686–687 Stolwijk AM, Straatman H, Zielhuis GA (1999) Studying seasonality by using sine and cosine functions in regression analysis. J Epidemiol Community Health 53:235–238 Ferlay J, Parkin DM, Curado MP et al. (2010) Cancer incidence in five continents, volumes I to IX: IARC. Lyon, France: International agency for research on cancer. Available from: http://ci5. iarc.fr Wagner D, Trudel D, Van der Kwast T, Nonn L, Giangreco AA, Li D, Dias A, Cardoza M, Laszlo S, Hersey K, Klotz L, Finelli A, Fleshner N, Vieth R (2013) Randomized clinical trial of vitamin D3 doses on prostatic vitamin d metabolite levels and Ki67 labeling in prostate cancer patients. J Clin Endocrinol Metab 98:1498–1507 Lips P (2006) Vitamin D physiology. Prog Biophys Mol Biol 92:4–8 http://www.foodstandards.gov.au/scienceandeducation/monitoring andsurveillance/ Accessed date: 2007. Ref Type: Generic Krishnan AV, Peehl DM, Feldman D (2003) The role of vitamin D in prostate cancer. Recent Results Cancer Res 164:205–221 Zhao XY, Ly LH, Peehl DM, Feldman D (1999) Induction of androgen receptor by 1alpha, 25-dihydroxyvitamin D3 and 9-cis retinoic acid in LNCaP human prostate cancer cells. Endocrinology 140:1205–1212 Zhao XY, Peehl DM, Navone NM, Feldman D (2000) 1 alpha, 25-dihydroxyvitamin D3 inhibits prostate cancer cell growth by androgen-dependent and androgen-independent mechanisms. Endocrinology 141:2548–2556 Krishnan AV, Peehl DM, Feldman D (2003) Inhibition of prostate cancer growth by vitamin D: regulation of target gene expression. J Cell Biochem 88:363–371 Peehl DM, Krishnan AV, Feldman D (2003) Pathways mediating the growth-inhibitory actions of vitamin D in prostate cancer. J Nutr 133:2461S–2469S Rothman KJ (1990) No adjustments are needed for multiple comparisons. Epidemiology 1:43–46
123
