A new NCI study on the benefits of X-ray computed tomography (CT),
...shows significant life extension from annual CT scans, albeit in a very high-risk group (heavy smokers age 55-74). Quoting the WSJ article:
After receiving three annual tests, study participants where then followed for up to five years with deaths from any documented cause, including lung cancer.
As of Oct. 20, a total of 354 deaths from lung cancer had occurred among participants in the CT arm of the study compared to 442 lung-cancer deaths among those in the X-ray group. The difference between the groups was a 20.3% reduction in lung-cancer deaths favoring the CT arm of the study.
The mechanism thought responsible is the simple one -- CT scans enable early detection, so better chances of treatment. Quoting the NYT article:
Lung cancer claims about 160,000 lives each year, more than the deaths from colorectal, breast, pancreatic and prostate cancers combined. In most patients, the disease is discovered too late for effective treatment, and 85 percent of those who are diagnosed with lung cancer die from it.
Until now, no screening method had proven to be effective at reducing mortality from the disease. Four randomized, controlled trials done during the 1970s showed that chest X-rays helped to catch cancers at an earlier stage, but had no effect on overall death rates. Since then, researchers have suggested that CT scans — which use coordinated X-rays to provide three-dimensional views of body tissue — could detect lung tumors at an even earlier stage than X-rays could, but no trial had shown conclusively that deaths could be averted.
Are preventative CT scans in the future? The helical CT chest scan in the study gives a radiation dose of about 8 mSv (0.8 rem) -- equal to more than a hundred chest x-rays. Annual CT scans would increase total radiation exposure from all sources by a factor of 3. The effects of low-dose radiation more than ever deserve serious research. We cannot continue to rely on untested, highly conservative risk estimates based on the LNT hypothesis; overestimating the hazards of radiation would needlessly impede useful medical imaging.
Some posts on this subject by Steve Packard (Depleted Cranium):