Null Results in Brief
Selenomethionine Treatment Does Not Alter Gene
Expression in Normal Squamous Esophageal
Mucosa in a High-Risk Chinese Population
Nina Joshi,1Laura Lee Johnson,3Wen-Qiang Wei,6Christian C. Abnet,2Zhi-Wei Dong,6
Philip R. Taylor,4Paul J. Limburg,7Sanford M. Dawsey,2Ernest T. Hawk,5
You-Lin Qiao,6and Ilan R. Kirsch1
1Genetics Branch, Center for Cancer Research, and
National Cancer Institute, Bethesda, Maryland;
Alternative Medicine, and
Training, and Resources, National Cancer Institute, NIH, Bethesda, Maryland;
Medical Sciences, Beijing, China; and
2Nutritional Epidemiology Branch, Division of Cancer Epidemiology and Genetics,
3Office of Clinical and Regulatory Affairs, National Center for Complementary and
4Genetic Epidemiology Branch, Division of Cancer Epidemiology and Genetics,
6Cancer Institute, Chinese Academy of
7Mayo Clinic College of Medicine, Rochester, Minnesota
5Office of Centers,
Selenium is a promising cancer chemoprevention agent. A
recent randomized controlled chemoprevention trial found
that selenomethionine (SeMet) supplementation for 10
months favorably effected a change in esophageal dysplasia
grade among participants who started the trial with mild
dysplasia. To further explore the role of SeMet in this trial,
we compared gene expression profiles by treatment group
using Affymetrix HU 133A chips in before/after supplemen-
tation paired normal esophageal biopsies from a subset of
29 trial participants, 16 who received SeMet, and 13 who
received placebo. Using P < 0.001 as a cutoff, 11 differentially
expressed genes were found in the SeMet supplementation
group but these genes did not include either known
selenoprotein genes or genes previously shown to be mo-
dulated by selenium treatment. Because the number of dif-
ferentially expressed genes (n = 11) was less than expected by
chance (n = 18), we concluded that SeMet supplementation
had no measurable effect on gene expression in the normal
squamous esophagus of these subjects with dysplasia.
(Cancer Epidemiol Biomarkers Prev 2006;15(5):1046–7)
Selenium compounds have been widely studied in the etiology
of cancer and as chemoprevention agents because of their
potential anticarcinogenic properties (1). Selenium deficiency
may play an important role in the etiology of esophageal
squamous cell carcinoma in the high-risk population of
Linxian, China, where this cancer is endemic and low serum
selenium concentrations are strongly associated with increased
risk of esophageal squamous cell carcinoma (2). The chemo-
preventative effects of selenomethionine (SeMet) and celecoxib
were recently assessed in a randomized, double-blind, placebo-
controlled, 2 ? 2 factorial chemoprevention trial conducted in
Linxian (3). This trial found that among patients with mild
esophageal squamous dysplasia, 10 months of daily treatment
with 200 Ag SeMet favorably effected a change in dysplasia
grade, such that there was less progression and moreregression
in SeMet recipients compared with participants who did not
receive SeMet (P = 0.02). The precise mechanisms underlying
the action of SeMet have not been defined. The present study
addressed this gap in knowledge by investigating potential
changes in gene expression in normal esophageal mucosa from
SeMet- and placebo-treated participants of this trial.
Materials and Methods
Details of the chemoprevention trial have been described
elsewhere (3). Briefly, asymptomatic adults with histologically
confirmed mild or moderate esophageal squamous dysplasia
were randomly assigned to one of four intervention groups
using a 2 ? 2 factorial design. Active treatments were SeMet
200 Ag once per day and/or celecoxib 200 mg twice per day.
Esophagogastroduodenoscopy examinations with Lugol’s io-
dine staining were conducted and biopsies were collected and
snap frozen before and after a 10-month intervention period.
We compared gene expression in paired histologically
normal biopsies from the same individual, one collected at
trial baseline (T0) and the second collected at the end of the
intervention period (T10), from each of 29 subjects, including
16 who received SeMet supplementation and 13 who received
placebo. Serum selenium concentrations were also examined
from blood samples collected at baseline and at the end of
supplementation using Inductively Coupled Plasma Mass
Spectroscopy (4). The study was approved by the Institutional
Review Boards of the Cancer Institute, Chinese Academy of
Medical Sciences and the U.S. National Cancer Institute.
RNA was extracted from frozen biopsies using standard
methods. The small sample RNA amplification protocol
described in detail elsewhere was used for this analysis (5).
Affymetrix HU 133A chip arrays, consisting of 18,400 tran-
scripts, including 14,500 known genes, were scanned in an
Affymetrix GCOS Argon-Ion Scanner at 488 nm. Analyses
included probe-level preprocessing using robust multiarray
analysis (justRMA) conducted with a 64-bit version of R 1.8.1
and Bioconductor 1.3 on the NIH/CIT Helix System and
S-Plus 6.1. To control for multiple comparisons, we considered
P < 0.001 as statistically significant. All tests were two sided.
Cancer Epidemiol Biomarkers Prev 2006;15(5). May 2006
on January 9, 2016. © 2006 American Association for Cancercebp.aacrjournals.org Downloaded from
Received 2/22/06; accepted 3/2/06.
Grant support: Intramural Research Program of the NIH and the National Cancer Institute.
The costs of publication of this article were defrayed in part by the payment of page charges.
This article must therefore be hereby marked advertisement in accordance with 18 U.S.C.
Section 1734 solely to indicate this fact.
Requests for reprints: Ilan R. Kirsch, Amgen, 1201 Amgen Court West, AW1-J 4144, Seattle,
WA 98119-3105. Phone: 206-265-7316; Fax: 206-216-5930. E-mail: email@example.com or
You-Lin Qiao, Cancer Institute, Chinese Academy of Medical Sciences, Panjiayian, Chaoyang
District, P.O. Box 2258, Beijing 100021, P.R. China. Phone: 86-10-6778-1331, ext. 8982;
Fax: 86-10-6771-3648. E-mail: firstname.lastname@example.org
Copyright D 2006 American Association for Cancer Research.