The JNCI study, led by Val Gebski from the University of Sydney, Australia, divided 38 comparisons from 36 trials of radiation therapy into 3 categories: optimal radiation therapy, defined by modern treatment guidelines; inadequate or excessive radiation therapy; or incomplete tissue coverage.
Overall, the study found a statistically significant 2.9 percent absolute increase in 5-year survival for all patients receiving optimal radiation therapy. This increase in survival rose to 6.4 percent at 10 years after treatment. In contrast, no significant increase in survival was found in patients receiving inadequate or excessive radiation therapy, or incomplete tissue coverage.
An increase in non-breast cancer death was noticed in the groups receiving any category of radiation therapy, but for optimal therapy, this was overshadowed by the survival benefit. The investigators applied the data on survival after optimal radiation therapy to hypothetical patients with different risks of recurrence and death. They determined that a high-risk patient receiving optimal radiation therapy has an 11 percent absolute reduction in risk of death from any cause.
"Consequently," state the authors, "we recommend that postmastectomy radiation therapy be considered as part of standard care for all women at high risk."
Two randomized clinical trials now under way in the United States and in Europe are investigating whether screening men for signs of prostate cancer saves lives. Many experts believe that only these large trials can answer the question of whether men should be screened, but the results are not expected until around 2009.
In the meantime, a study published last week found no evidence that screening reduced mortality among men in New England. The case-control study involved 1,002 men and looked at 2 commonly used screening methods, the prostate-specific antigen (PSA) test and a rectal examination. According to findings in the January 9 Archives of Internal Medicine, PSA testing, with or without rectal examinations, did not prevent deaths from prostate cancer among the men.
The researchers, led by Dr. John Concato of the Veteran Affairs Connecticut Healthcare System, acknowledge that more research is needed on the subject, and they urge physicians to obtain informed verbal consent from patients who undergo screening.
"This is a well-done study, but it does not really answer the question" of whether screening saves lives, comments Dr. Howard Parnes of NCI's Division of Cancer Prevention. "Until the results of the randomized trials are known, men should be encouraged to discuss the potential risks and benefits of screening with their physicians."
An editorial accompanying the study agrees with this message and notes that the long-awaited results are "now not that far away." The NCI-sponsored Prostate, Lung, Colorectal and Ovarian Screening Trial and the European Randomized Study of Screening for Prostate Cancer trial should be completed "after the next summer Olympics in Beijing," says Dr. Michael Barry of Massachusetts General Hospital.
Screening for colorectal cancer can significantly reduce mortality, particularly because precancerous polyps and adenomas can lead to blood in the stool, which is normally detected by a guaiac fecal occult blood test (FOBT). A study published in the January 6 online edition of Lancet Oncology shows that a more discriminating, immunochemical version of FOBT can be used to better determine which patients should be examined further by colonoscopy.
During an observational screening study in Scotland for colorectal cancer in men and women aged 50 to 69, researchers used the immunochemical FOBT test to collect 2 samples from each of 795 patients, all of whom had previously tested positive with guaiac FOBT. Patients then had clinical colonoscopies, and any disease detected was correlated with the results of the immunochemical FOBT tests. Patients who tested positive on both immunochemical FOBT tests were 7.57 times more likely to have an invasive cancer than were those who had either one or no positive tests. They were also 1.48 times more likely to have large adenomatous polyps and 3.11 times more likely to have more than three polyps of any size.
Dr. Callum G. Fraser, of the Ninewells Hospital and Medical School in Dundee, Scotland, and colleagues recommend that this "reflex" two-tier approach be used to distinguish between patients who would and would not benefit from colonoscopy. By eliminating guaiac FOBT-positive patients who are not double-positive on the immunochemical FOBT test, the need for colonoscopies would be reduced by about 30 percent. As reported, "The implications for national screening programs are important in terms of reducing costs, inconvenience, and associated morbidity, without compromising the effectiveness of screening."
A new meta-analysis indicates that the cholesterol-lowering drugs known as statins have no effect on cancer risk. The analysis was published in the January 4 Journal of the American Medical Association.
To conduct the study, Dr. Krista M. Dale and colleagues from the University of Connecticut School of Pharmacy and Hartford Hospital identified 26 randomized clinical trials of statins for cardiovascular risk reduction that met specific criteria, including the collection of information on cancer diagnosis or cancer death on more than 100 participants with follow-up of at least 1 year.
Statins have a "neutral effect on cancer and cancer death risk," the authors concluded.
The study's conclusion contradicts several recent retrospective, case-control studies, including a study published last May by researchers from Michigan and Israel, which found that people who took a statin for at least 5 years had a significantly lower risk of developing colorectal cancer than those not taking the drug.
The meta-analysis, although well-conducted, is not the final word on statins and cancer prevention, says Dr. Ernie Hawk, director of the NCI Office of Centers, Training, and Resources.
"The prior data from animal models treated with statins, the mechanistic data on possible ways they would fight cancer, and the observational studies suggesting reductions in cancer risk still make the question interesting," he says. "We still lack data from well-conceived phase II clinical trials focused on cancer-related mechanisms. These trials would help either to support or refute statins' potential use for cancer prevention."
NCI is sponsoring two such trials - one in patients at risk for colorectal cancer and the other in patients at risk for melanoma. Both trials are expected to begin early this year.
New databases will soon be available from NCI that allow calculation of cancer risks among diverse racial/ethnic groups in the United States. Using these databases, with data centered on the 1990 decennial census, NCI researchers report that 48 percent of Japanese American men will develop cancer in their lifetime, the highest risk of the 10 groups studied. White men follow at 47 percent. Among men, American Indians are the least likely to develop cancer, at 24 percent lifetime risk.
Women in all groups, except for Native Alaskans, have a lower lifetime risk than their male counterparts, according to the research published in the December 30, 2005, online edition of Cancer. The authors calculated lifetime and age-conditional cancer risks of 10 racial and ethnic groups; previous estimates provided figures only for whites and blacks. More recent databases with racial/ethnic information are also available, but are limited to broader classifications because of the lack of detailed population estimates needed for the calculations.
The type of risk estimate reported by the authors measures the population burden of cancer because it integrates the chance of developing cancer with the chance of dying of other causes. For example, although black men have significantly higher age-adjusted prostate cancer incidence rates than whites, the lifetime risks of developing cancer in the two groups are more closely comparable because fewer blacks live to older ages when prostate cancer predominantly occurs. The authors conclude that this type of cancer burden estimate can complement routinely reported incidence and mortality statistics.