© Oxford University Press 2006.
IN THIS ISSUE
Overdiagnosis, Extended Follow-up of Mayo Lung ProjectOverdiagnosis is detecting disease that would never have been diagnosed without screening. The Mayo Lung Project was a large randomized controlled trial of lung cancer screening among male cigarette smokers in the 1970s and early 1980s. Its results indicated that overdiagnosis of lung cancer was associated with screening. At the end of follow-up in 1983, the screened and usual care arms did not differ in lung cancer mortality, but more cases were found in the screened arm than the usual-care arm. Because these extra cases could have been due to short follow-up, Marcus et al. (p. 748) extended the follow-up by 16 years. They found that 585 lung cancers were diagnosed in the screened arm and 500 in the usual-care arm. They conclude that the persistent excess lung cancers in the intervention arm even after extended follow-up supports overdiagnosis associated with lung cancer screening.
In an editorial, Patz (p. 724) notes that the Mayo Lung Project's critics argued that the original follow-up was too short to confirm that overdiagnosis was responsible for the observed differences, which should equalize with time. However, after the additional follow-up, increasing divergence—not convergence—was observed in the numbers of lung cancers. He concludes that the benefits of screening must be rigorously validated in clinical trials before introduction into routine practice.
Urokinase Acts on Tumors in Breast Cancer Mouse Model
The understanding of proteases in tumor biology is evolving. To determine the effects of tissue plasminogen activator (tPA) and urokinase (uPA) overexpression on tumor progression, Merchan et al. (p. 756) implanted mouse mammary cancer cells that overexpress tPA and uPA and control cancer cells into mouse mammary glands. Tumors derived from tPA- or uPA-overexpressing cells grew more slowly and produced fewer metastases than those derived from control cells, and mice carrying tPA- or uPA-overexpressing tumors lived longer. In addition, tumors from cells that overexpressed proteolytically inactive uPA grew faster than those overexpressing wild-type uPA. The authors conclude that, in this mouse model, uPA overexpression has tumor-delaying effects that are mediated by its protease activity, possibly through antiangiogenic and antiproliferative mechanisms. They also note that these results challenge the current dogma that proteases are only tumor promoting.
In an editorial, Narazaki and Tosato (p. 726) discuss the roles of proteases and their inhibitors in tumor invasion and metastasis. They compare the mechanisms of the mouse model with potential mechanisms of proteases in human breast cancer.
Randomized Trial of HPV Screening Techniques
To compare cervical cancer screening methods, Ronco et al. (p. 765) performed a multicenter randomized controlled trial of more than 30,000 women ages 35 to 60. Women in the control arm were screened using conventional cytology and were referred to colposcopy if atypical cells of undetermined significance or higher cytology was found. Women in the experimental arm were screened using liquid-based cytology and tested for high-risk human papillomavirus (HPV) types; they were referred to colposcopy if atypical cells of undetermined significance or higher cytology was found or if the HPV test was positive. The authors found that screening with liquid-based cytology and HPV testing was more sensitive (more high-grade precancerous lesions were detected) than conventional cytology but provided more false positive results. They also found that HPV testing alone, using a cutoff of 2 pg/mL, provided a similar gain in sensitivity with a much lower false-positive rate than combined screening.
Cost Effectiveness of Screening Mammography
Many guidelines recommend a screening mammogram every 1–2 years for women older than 40. Stout et al. (p. 774) used a model of breast cancer epidemiology to compare costs and quality-adjusted life-years (QALYs)—a measure that combines the quantity and quality of life—associated with observed U.S. screening mammography patterns for women age 40 or older from 1990 to 2000. They found that actual screening patterns cost $166 billion and accrued 947.5 million QALYs over the womens' lifetimes, resulting in an additional $62.5 billion and a gain of 1.7 million QALYs compared with no screening. Many alternative scenarios generated more QALYs for less cost than the actual patterns. The authors conclude that the costs of increasing screening and the acute quality-of-life effects of mammography should guide future screening recommendations.
Targeting PGFR on Drug-Resistant Prostate Cancer Cells
Inhibiting phosphorylation of platelet-derived growth factor receptor (PDGFR) by treatment with the PDGFR kinase inhibitor imatinib and the chemotherapeutic agent paclitaxel reduce the incidence and size of human prostate cancer bone tumors in nude mice. Kim et al. (p. 783) investigated the primary target for imatinib using mice with tumors derived from multidrug-resistant human prostate cancer cells. After treatment with imatinib and paclitaxel, mice implanted with these cells had decreased incidence of bone tumors and lymph node metastasis, reduced median tumor weight, and fewer bone lesions. Treatment with imatinib alone had similar effects. Imatinib also inhibited phosphorylation of PDGFR on tumor cells and tumor-associated endothelial cells and increased apoptosis of endothelial cells but not tumor cells. Treatment with imatinib, or with imatinib and paclitaxel, decreased mean vessel density, which was followed by tumor cell apoptosis. The authors conclude that imatinib targets tumor-associated endothelial cells in prostate cancer bone metastasis.
COX-2-Independent Anticancer Effects of Cox-2 Inhibitors
Nonsteroidal antiinflammatory drugs (NSAIDs) appear to reduce the risk of developing cancer by inhibiting the activity of cyclooxygenase 2 (COX-2), which is overexpressed in various cancer tissues. However, there may also be COX-independent mechanisms of NSAIDs. Grösch et al. (p. 736) review the COX-independent molecular targets of NSAIDs that selectively inhibit COX-2 activity because these targets may also be involved in the drugs' anticarcinogenic activities. They point out that the gastrointestinal and cardiovascular side effects associated with some COX-2 inhibitors may not be attributable solely to COX-2 inhibition. It will be important to identify the COX-2–independent targets that participate in the drugs' anticancer effects so that new drugs with fewer side effects can be designed for these targets.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||