© 2005 Oxford University Press
IN THIS ISSUE
Using Microarrays in Predictive Models for Breast CancerData from microarray studies have been used to develop predictive models for treatment outcome in breast cancers. Reid et al. (p. 927) attempted to validate a predictive model for antiestrogen response after tamoxifen treatment that was based on the expression ratio of two genes. They studied an independent cohort of 58 patients with resectable estrogen receptor–positive breast cancer, measured the expression of HOXB13 and IL17BR genes, and assessed the association between their expression and outcome by use of four statistical tests. The authors also applied standard supervised methods to the original microarray dataset and to another independent dataset. They could not validate the performance of the two-gene predictor with their cohort of samples, nor could they develop a model with good predictive accuracy using the two microarray datasets. They conclude that treatment–response predictive models have poor performance with the sample sizes of patients and informative genes currently available.
In an editorial, Simon (p. 866) provides possible explanations for the inconsistencies in the results of Reid et al. and the original prediction model. He recommends that studies reporting the development of a genomic classifier be based on patients from clinical trials that address a specific therapeutic question. He also recommends that such developmental studies should require independent validation of the clinical benefit of the genomic classifier before its adoption into clinical practice.
EGFR, PKB/Akt, and Glioma Response to Erlotinib
Erlotinib, an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor, has shown promising response rates in patients with malignant gliomas. Haas-Kogan et al. (p. 880) investigated the association between expression of EGFR (and downstream signaling components) and the response of malignant gliomas to erlotinib by use of data and specimens from a phase I trial. They found that eight of 41 glioma patients responded to erlotinib treatment. This response was associated with EGFR expression and with EGFR amplification and was stronger among the 29 patients initially diagnosed with glioblastoma multiforme. None of the 22 tumors with high levels of phosphorylated protein kinase B (PKB)/Akt protein responded to erlotinib treatment, whereas eight of the 18 tumors with low levels of this protein responded to erlotinib treatment.
In an editorial, Cappuzzo (p. 868) states that erlotinib may be active in a small fraction of gliomas with EGFR expression and amplification. He points out that, because six of the 10 patients with EGFR amplification responded to treatment, other mechanisms may be involved. He concludes that these results should stimulate further prospective studies to identify predictive markers for sensitivity of gliomas to EGFR tyrosine kinase inhibitors.
Cost-effectiveness of HPV DNA Testing in European Countries
To determine the cost-effectiveness of adding human papillomavirus (HPV) DNA testing to the current cervical cancer screening programs in the United Kingdom, The Netherlands, France, and Italy, Kim et al. (p. 888) developed a computer model of cervical cancer carcinogenesis and compared each country's current screening policy with two strategies that would incorporate HPV DNA testing. Both new strategies, HPV triage for women with equivocal cytology results and HPV testing in combination with cytology for women more than 30 years of age, were more effective than the current policy in all four countries. HPV triage costs per year of life saved were less than $13,000 and costs for combination testing ranged from $9800 to $75,900, depending on screening interval. The authors conclude that including HPV DNA testing can potentially improve health benefits at a reasonable cost compared with the current screening policies in four European countries.
Age-Related Prevalence of Anal Cancer Precursors
Anal squamous intraepithelial lesions (ASIL) are anal cancer precursors that include low-grade SIL (LSIL) and high-grade SIL (HSIL). In a study of 1262 HIV-negative men who have sex with men (MSM), Chin-Hong et al. (p. 896) found that the overall prevalence of ASIL was 20% and that there was a similar prevalence of ASIL across all age groups. The risk of LSIL was associated with having an increased number of male receptive anal sex partners during the previous 6 months, any use of alkyl nitrites or the use of injection drugs two or more times per month during the previous 6 months, older age at first receptive anal intercourse, and infection with a greater number of human papillomavirus (HPV) types. The authors speculate that high prevalence of ASIL across all age groups may reflect ongoing sexual exposure to HPV in MSM.
Colorectal Cancer Risk and Meat and Fish Intake
To determine if colorectal cancer risk is associated with meat and fish intake, Norat et al. (p. 906) collected information on the diet and lifestyle of 478,040 men and women in 10 European countries who were free of cancer at enrollment between 1992 and 1998 and prospectively followed them for colorectal cancer incidence for an average of 4.8 years. When the authors examined the relationship between highest and lowest intakes of red and processed meat, poultry, and fish and colorectal cancer risk, they found that risk was directly associated with intake of red and processed meat, inversely associated with intake of fish, and not associated with intake of poultry. The authors calculated that the absolute risk of colorectal cancer within 10 years for a person aged 50 years was 1.71% for the highest category of red and processed meat intake versus 1.28% for the lowest and 1.28% for the highest category of fish intake and 1.86% for the lowest.
Iron Levels and Risk of Colorectal Adenoma in Women
Experimental data have raised the possibility that iron may be carcinogenic, but epidemiologic studies of such a role in humans have been inconclusive. It has been suggested that individuals who have iron overload associated with hereditary hemachromatosis may have an increased risk of colorectal cancer. In this issue, Chan et al. (p. 917) report the results of a prospective, nested case–control study within the Nurses' Health Study in which they evaluated the risk of colorectal adenoma in relation to dietary iron intake, biochemical markers of total body iron, and two mutations in the HFE gene that are associated with hereditary hemachromatosis. Although women with HFE mutations had increased total body iron stores, they did not have an increased risk of colorectal adenoma. In addition, neither dietary iron intake nor total body iron stores were associated with risk of adenoma.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||