Started in August 2014 while working through PH207x. Later extended to record additional selenium and mortality work.
Revisiting the discussion in the PH207x Health in Numbers week 8 literature critique of selenium supplementation studies. Also see my notes for the class in ~/Documents/Education/edX - PH207X/PH207x Health in Numbers notes.doc
This is of special interest because of the Feb. 2014 media coverage emphasizing the negative consequences of selenium. See http://www.cancer.gov/newscenter/qa/2008/selectqa This media flurry was a reaction to the publication of Baseline Selenium Status and Effects of Selenium and Vitamin E Supplementation on Prostate Cancer Risk (A. R. Kristal et al. 2014)
The first week 8 critique of the literature by Lorelei Mucci focused on comparing the SELECT trial (which found essentially no benefit to selenium supplementation) to earlier epidemiological studies (which did find benefit). (S. M. Lippman et al. 2009)
The earlier papers:
Clark LC, Combs GF Jr, Turnbull BW, et al; Nutritional Prevention of Cancer Study Group. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin: a randomized controlled trial JAMA. 1996;276(24):1957-1963. PDF (L. C. Clark 1996)
Yoshizawa K, Willett WC, Morris SJ, Stampfer MJ, Spiegelman D, Rimm EB, et al. Study of prediagnostic selenium level in toenails and the risk of advanced prostate cancer. J Natl Cancer Inst. 1998; 90(16):12191224. [PubMed: 9719083] (referred to as “Epi study”?) (Yoshizawa et al. 1998)
Duffield-Lillico AJ, Dalkin BL, Reid ME, Turnbull BW, Slate EH, Jacobs ET, et al. Selenium supplementation, baseline plasma selenium status and incidence of prostate cancer: an analysis of the complete treatment period of the Nutritional Prevention of Cancer Trial. BJU Int. 2003; 91(7):608612. [PubMed: 12699469] [PDF](http://libgen.org/scimag/get.php?doi=10.1046%2Fj.1464-410x.2003.04167.x) (secondary analysis of 1996 Clark “NPC study”) (Duffield-Lillico et al. 2003)
Also see downloads in ~/Documents/Education/edX - PH207X/Handouts/Week 8
Unfortunately I was unable to post in this discussion because I took the class late and it was already closed. It is still possible to up/down vote though.
My primary interest in this is captured in this question:
CRITIQUE OF LITERATURE IV: QUESTION 4
Why did the results from the SELECT trial differ from the epidemiological studies? a) What might be plausible explanations for the differences? b) Think about this in the context of confounding, bias, as well as the measurement of the exposure and outcomes.
I think the key points are brought out in this post:
Q4. 1) Different outcomes, 2) High levels of selenium at baseline in the SELECT trial
Reproduced here in case the forum is closed later.
In the comparison of the studies (Q4) I was surprised not to see any discussion of how starting selenium levels might have impacted the SELECT RCT (there was some discussion of this in the question forum).
In the Clark study table 2 breaks the results down by tertile of baseline plasma Se level
| Tertile Se (ng/mL) | Se cases | Placebo cases | RR | P |
|---|---|---|---|---|
| <106.4 | 1 | 13 | 0.08 | 0.0002 |
| 106.4-121.2 | 4 | 13 | 0.30 | 0.03 |
| over 121.2 | 8 | 9 | 0.85 | 0.75 |
| Total | 13 | 35 | 0.37 | 0.002 |
These results would seem to indicate that baseline Se status is a key parameter for any study of Se supplementation in prostate cancer. From Table 1 the baseline Se levels had mean 115 and SD about 22 for both groups.
The Epi study compared the lowest and highest quintiles for toenail Se levels. It is unfortunate there is no blood test data to allow for some comparison with the other study populations (hopefully future studies will test both blood and toenails). Given the choice of the extreme quintiles I would have expected a smaller OR (based on the table above) and think both studies obtaining a result of 0.4 is best viewed as a coincidence (i.e. they are measuring different things so agreement is not meaningful, both being significantly below 1 is though).
Compare the Clark study serum Se levels of mean=115 SD=22 with the levels given in Table 2 of the SELECT study paper: median=137.6 interquartile range=[124.7-151.8]. It is clear that the subjects in the SELECT study had a starting serum Se level at the upper end of the Clark study population (and recall the 0.85 RR there). Given this, it is not surprising that the SELECT study would conclude that (my words) supplementing subjects with adequate Se status with Se does not reduce the risk of prostate cancer. Therefore I would say that the SELECT study does not contradict the Clark study. We would require information about serum Se levels in the Epi study to make any valid comparison there. One good feature about the SELECT study is they include before/after Se supplementation serum Se levels. Over two years the 200ug supplement raised serum Se levels roughly 100 ug/L. It would have been nice if the SELECT study had included an analysis based on a separation into quintiles by baseline Se level.
Note that the SELECT study paper did discuss this (on page 9, comparing to the NPC study). IIRC that subtlety failed to make it into the media coverage which resulted from the SELECT study.
I wonder if the Clark study being selected from an existing cancer population has any connection with the lower Se levels observed there. This is a potentially serious confounding factor.
Lastly, the follow-up time may be important, but I wonder if the differing diagnostic criteria (advanced prostate cancer for the Epi study vs. any prostate cancer for SELECT) might be more important.
Now attempt to address this in a more formal fashion…
Look into other studies that have investigated the relationship of selenium levels to mortality.
Serum selenium and cancer mortality: a nested case-control study within an age- and sex-stratified sample of the Belgian adult population (Kornitzer et al. 2004) - downloaded as 1601754a.pdf
See pubmed references
Defining the Optimal Selenium Dose for Prostate Cancer Risk Reduction: Insights from the U-Shaped Relationship between Selenium Status, DNA Damage, and Apoptosis (Chiang et al. 2009) - downloaded as drp-08-285.pdf
Selenium for preventing cancer - 2014 (U Alehagen et al. 2015) - downloaded as nihms-684156.pdf
Relatively high mortality risk in elderly Swedish subjects with low selenium status (U Alehagen et al. 2015) - downloaded as ejcn201592a.pdf
Reduced Cardiovascular Mortality 10 Years after Supplementation with Selenium and Coenzyme Q10 for Four Years: Follow-Up Results of a Prospective Randomized Double-Blind Placebo-Controlled Trial in Elderly Citizens (Urban Alehagen, Aaseth, and Johansson 2015) - downloaded as pone.0141641.pdf
File originally created: Sunday, August 3, 2014
File knitted: Sun Jan 31 09:35:47 2016
Alehagen, U, P Johansson, M Björnstedt, A Ros\a’en, C Post, and J Aaseth. 2015. “Relatively High Mortality Risk in Elderly Swedish Subjects with Low Selenium Status.” European Journal of Clinical Nutrition 70 (1). Nature Publishing Group: 91–96. doi:10.1038/ejcn.2015.92.
Alehagen, Urban, Jan Aaseth, and Peter Johansson. 2015. “Reduced Cardiovascular Mortality 10 Years After Supplementation with Selenium and Coenzyme Q10 for Four Years: Follow-up Results of a Prospective Randomized Double-Blind Placebo-Controlled Trial in Elderly Citizens.” Edited by Harald HHW Schmidt. PLOS ONE 10 (12). Public Library of Science (PLoS): e0141641. doi:10.1371/journal.pone.0141641.
Chiang, Emily C., Shuren Shen, Seema S. Kengeri, Huiping Xu, Gerald F. Combs, J. Steven Morris, David G. Bostwick, and David J. Waters. 2009. “Defining the Optimal Selenium Dose for Prostate Cancer Risk Reduction: Insights from the U-Shaped Relationship Between Selenium Status, DNA Damage, and Apoptosis.” Dose-Response 8 (3). SAGE Publications: 285–300. doi:10.2203/dose-response.09-036.chiang.
Clark, Larry C. 1996. “Effects of Selenium Supplementation for Cancer Prevention in Patients with Carcinoma of the Skin.” JAMA 276 (24). American Medical Association (AMA): 1957. doi:10.1001/jama.1996.03540240035027.
Duffield-Lillico, A.J., B.L. Dalkin, M.E. Reid, B.W. Turnbull, E.H. Slate, E.T. Jacobs, J.R. Marshall, and L.C. Clark. 2003. “Selenium Supplementation, Baseline Plasma Selenium Status and Incidence of Prostate Cancer: An Analysis of the Complete Treatment Period of the Nutritional Prevention of Cancer Trial.” BJU International 91 (7). Wiley-Blackwell: 608–12. doi:10.1046/j.1464-410x.2003.04167.x.
Kornitzer, M, F Valente, D De Bacquer, J Neve, and G De Backer. 2004. “Serum Selenium and Cancer Mortality: A Nested Casecontrol Study Within an Age- and Sex-Stratified Sample of the Belgian Adult Population.” European Journal of Clinical Nutrition 58 (1). Nature Publishing Group: 98–104. doi:10.1038/sj.ejcn.1601754.
Kristal, A. R., A. K. Darke, J. S. Morris, C. M. Tangen, P. J. Goodman, I. M. Thompson, F. L. Meyskens, et al. 2014. “Baseline Selenium Status and Effects of Selenium and Vitamin E Supplementation on Prostate Cancer Risk.” JNCI Journal of the National Cancer Institute 106 (3). Oxford University Press (OUP): djt456–djt456. doi:10.1093/jnci/djt456.
Lippman, Scott M., Eric A. Klein, Phyllis J. Goodman, M. Scott Lucia, Ian M. Thompson, Leslie G. Ford, Howard L. Parnes, et al. 2009. “Effect of Selenium and Vitamin E on Risk of Prostate Cancer and Other Cancers.” JAMA 301 (1). American Medical Association (AMA): 39. doi:10.1001/jama.2008.864.
Yoshizawa, K., W. C. Willett, S. J. Morris, M. J. Stampfer, D. Spiegelman, E. B. Rimm, and E. Giovannucci. 1998. “Study of Prediagnostic Selenium Level in Toenails and the Risk of Advanced Prostate Cancer.” JNCI Journal of the National Cancer Institute 90 (16). Oxford University Press (OUP): 1219–24. doi:10.1093/jnci/90.16.1219.