© Oxford University Press 2006.
IN THIS ISSUE
Nutritional Interventions and Cancer or Preinvasive LesionsDietary modifications and supplements are used widely as an adjunct to standard treatment by patients with cancer or preinvasive lesions. Davies et al. (p. 961) conducted a review of randomized controlled trials of nutritional interventions to examine benefits and harms in such patients. They identified 59 eligible trials; trial quality was generally low. Mortality, cancer recurrence, and other outcomes were not associated with weight loss, exercise, or a healthy diet—alone or with dietary supplements. There was thus no evidence that dietary changes by cancer patients improve survival or benefit disease prognosis.
There is also little evidence that vitamin or garlic supplements or Helicobacter pylori eradication impedes development of precancerous gastric lesions. You et al. (p. 974) conducted a randomized trial among adult residents of 13 Chinese villages to test the effects of one-time H. pylori eradication versus long-term use of vitamin or garlic supplements on these lesions. H. pylori treatment decreased the combined prevalence of several types of gastric lesions and gastric cancer and had favorable effects on precancerous gastric lesions. It did not reduce the combined prevalence of dysplasia and gastric cancer, but fewer subjects receiving H. pylori treatment developed gastric cancer. Long-term vitamin or garlic supplementation had no beneficial effects. Further data are needed to determine whether H. pylori treatment reduces gastric cancer incidence.
In an editorial, Baron (p. 945) suggests assessments of nutritional intervention trial quality in addition to those used by Davies et al.. He points out a few tantalizing findings from such trials, most notably nonstatistically significant reductions in all-cause and cancer mortality. Nutritional deficiencies in the subjects studied by You et al. reduce the generalizability of their results, and the high prevalence of some endpoints makes interpretation of relative risks complex. The studies "illustrate the contemporary status of chemoprevention: hard to summarize, many negative findings, but some nuggets of hopeful progress."
Gene Promoter Methylation and Bladder Cancer Detection
Aberrant promoter methylation (i.e., hypermethylation) silences tumor suppressor and other cancer-associated genes. Hoque et al. (p. 996) evaluated promoter hypermethylation of nine genes in primary tumor and urine sediment DNA from bladder cancer patients and control subjects. Four of the nine genes showed promoter hypermethylation in cancer patients but not in control subjects. A prediction model developed using the methylation levels of the remaining five genes and internally validated for subjects who were negative on the four-gene panel had an overall sensitivity of 82% and specificity of 96%. The authors conclude that testing hypermethylation of these genes in urine sediment DNA may be a promising noninvasive approach for bladder cancer detection.
Connective Tissue Growth Factor and Angiogenesis
To determine the role of connective tissue growth factor (CTGF) in tumor angiogenesis, Chang et al. (p. 984) assessed the growth and metastasis of human lung tumor cell xenografts in mice. They compared tumors that overexpressed CTGF and hypoxia-inducible factor (HIF)-1
with those with low levels of CTGF. Tumors with high CTGF levels grew more slowly, had reduced angiogenesis, formed fewer metastases, and expressed low levels of HIF-1 and vascular endothelial growth factor (VEGF)-A. Tumors with high levels of CTGF and HIF-1 expressed high levels of VEGF-A, and mice with these tumors had shorter survival than mice with tumors expressing high levels of CTGF and low levels of HIF-1. Thus, CTGF may act through VEGF-A to inhibit angiogenesis, and HIF-1 protein degradation may be involved in this process.
In an editorial, Tosetti et al. (p. 946) discuss the roles of CTGF, HIF-1, and VEGF in tumor angiogenesis. Future improvements in clinical outcomes may arise from targeting the tumor as well as its microenvironment.
DDT, DDE, and liver cancer
The insecticide DDT and its persistent metabolite DDE cause liver cancer in animals. To learn whether exposure to either compound is associated with liver cancer in humans, McGlynn et al. (p. 1005) carried out a nested case–control study in China, where liver cancer is common. The risk of developing liver cancer increased with higher serum DDT concentration; those in the highest quintile of serum DDT had nearly four times the risk of those in the lowest quintile. There was no association between serum DDE and liver cancer, but the association between DDT and liver cancer was stronger among individuals with lower DDE concentrations.
Hormone Receptor Status of Mouse Mammary Stem Cells
Different subtypes of breast cancer variably express estrogen receptor (ER), progesterone receptor (PR), and ErbB2. For example, the basal subtype is negative for ER, PR, and ErbB2. Asselin-Labat et al. (p. 1011) investigated expression of these markers in mouse mammary cells enriched for stem or luminal cells. In mice with intact ovaries, the cell population enriched with mouse mammary stem cells did not express ER, PR, or ErbB2 but did express epidermal growth factor receptor (EGFR), whereas the cells enriched with luminal cells expressed ER and PR but not ErbB2 or EGFR. Mouse mammary stem cells thus appear to share properties with poor-prognosis basal breast cancer.
In an editorial, Anderson and Matsuno (p. 948) note that animal, molecular, epidemiologic, and clinical studies point to at least two main types of breast cancer, classified by epithelial cellular origin and/or hormone responsiveness. One type peaks in younger women and is largely hormone dependent. The other peaks in older women and is dependent on accumulated hormonal and/or environmental exposures.
STK15 F31I Polymorphism and Breast Cancer Risk
STK15 may be a low-penetrance breast cancer susceptibility gene, and homozygosity for the STK15 F31I polymorphism has been associated with increased risk of breast cancer. Fletcher et al. (p. 1014) examined 507 patients with two primary breast cancers and 875 control subjects for the STK15 F31I polymorphism. The Ile/Ile homozygous genotype was not associated with an increased risk of breast cancer. A meta-analysis of this and other studies of the STK15 F31I polymorphism showed statistically significant heterogeneity among them that could reflect either population-specific inheritance patterns or artifacts such as population stratification or publication bias.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||