© 2005 Oxford University Press
IN THIS ISSUE
Skin Cancer Risk and In Vitro UVB SensitivityTo determine whether sensitivity to ultraviolet (UV)-induced DNA damage is associated with the risk of skin cancer in the general population, Wang et al. (p. 1822) compared the frequency of UVB-induced chromosomal breaks in the lymphocytes of 469 patients with skin cancer and of 329 control subjects in a hospital-based case–control study. They found that patients with basal cell and squamous cell carcinomas had a higher frequency of UVB-induced chromosomal breaks than patients with cutaneous malignant melanoma and control subjects. The authors conclude that sensitivity to UVB-induced DNA damage may be associated with susceptibility to non-melanoma skin cancer but not to cutaneous malignant melanoma.
Protein Expression, Antibody Concentration, and Prognosis
To determine whether the disparate results in the immunohistochemistry literature on the relationship between biomarker expression and patient outcome are influenced by subjective optimization of antibody concentration, McCabe et al. (p. 1808) used a quantitative automated system and various concentrations of antibodies against HER2, p53, and the estrogen receptor (ER) to assess expression of HER2, p53, and ER protein in a tissue microarray containing specimens from 250 breast cancer patients. When they used optimal cutpoints and a high concentration of antibodies against HER2 or p53, low expression was associated with poorer survival. However, when they used optimal cutpoints and a low concentration of these antibodies, high expression was associated with poorer survival. When they investigated ER protein expression, they found that ER expression was always associated with better survival. They conclude that biomarker antibody concentration appears to dramatically affect the apparent relationship between the expression of some biomarkers and outcome.
In an editorial, Henson (p. 1796) notes that changes in the association between biomarker expression and outcome as a function of the biomarker antibody concentration in the immunohistochemical procedure used to detect biomarker expression merit additional study with other antibodies, detection methods, and biomarkers. He concludes that these results warrant consideration before predictive biomarkers, evaluated through immunochemistry, are integrated into formal prognostic systems or used as indicators for specific therapy.
HPV-Induced Dysplasia in the Female Lower Genital Tract
Most lesions in the female lower genital tract are induced by infections with high-risk oncogenic human papillomaviruses (HR-HPVs). HR-HPV genomes frequently integrate into host cell chromosomes at random sites. Vinokurova et al. (p. 1816) analyzed viral integration sites in multiple metachronous lesions of the lower genital tract from seven women previously treated for HR-HPV–positive cervical dysplasia or cancer. They found identical HPV DNA integration loci in vaginal or vulvar and cervical samples of all lesions available for four of the five patients with a prior history of high-grade cervical dysplasia and for the two patients with a history of cervical cancer. The authors conclude that high-grade dysplastic lesions in the female lower genital tract may emerge primarily as clonal lesions from a transformed cell population derived from the uterine cervix.
In an editorial, Feng and Kiviat (p. 1798) discuss the possibility that vaginal cancers might have arisen from cells derived from the CIN3 lesions because the trauma from surgery on the vaginal epithelium predisposes it to accept implantation of transformed cells derived from the original CIN3 lesion. They also discuss the clinical implications of these findings with respect to vaginal cytologic screening recommendations.
Baseline Lung Cancer Detection in the PLCO Trial
The Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial was initiated to examine cause-specific mortality reduction from screening for these four cancers in men and women. Oken et al. (p. 1832) report baseline results from the lung cancer screening portion of the trial. In the initial screen, 8.9% of chest radiographs were suspicious for lung cancer (8.2% for women and 9.6% for men). Rates were highest for older age groups and for smokers. Among all participants, they detected 1.9 lung cancers per 1000 screens. High rates of lung cancer were found among current smokers (6.3 per 1000 screens) and former smokers who had smoked within the past 15 years (4.9 per 1000 screens). Among never-smokers, they detected 0.4 lung cancers per 1000 screens. Among all cancers detected, 44% were stage I nonsmall cell lung cancer. They conclude that, although there is a high rate of detection of lung cancer, it has not yet been determined whether this will result in an improvement in lung cancer mortality.
CXCR4 Expression and Metastasis
Because the chemokine and bone marrow-homing receptor CXCR4 has been implicated in metastatic dissemination of various cancers, Kaifi et al. (p. 1840) investigated the association between CXCR4 expression in esophageal cancer specimens and survival, lymph node microinvolvement, and bone marrow micrometastasis. They found that patients whose tumors expressed CXCR4 had statistically significantly higher lymph node microinvolvement and bone marrow micrometastasis than patients whose tumors did not express CXCR4. CXCR4 expression was also identified as the independent variable that was most strongly associated with reduced diseasespecific and overall survival. They conclude that CXCR4 expression was associated with poor clinical outcome in esophageal cancer patients and that it may have a role in early metastasis because of its association with micrometastasis to the lymph nodes and bone marrow.
DNA Levels in Stored Plasma
Levels of free circulating DNA have been found to be higher in the plasma of cancer patients than in that of healthy control subjects, and it has been suggested that this biomarker may be a useful component of lung cancer screening. For plasma DNA biomarkers to be useful in the clinic, however, measurements must be robust to variations in sampling, processing, and storage conditions. In this issue, Sozzi et al. (p. 1848) investigated the effect of prolonged storage of both plasma and isolated plasma DNA on the quantification of free circulating DNA. For both kinds of samples, levels declined substantially over time, with yearly decay rates of approximately 30%. The authors note that this observation is of importance for retrospective analyses of stored samples and should be considered in the development of clinical trials.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||