1987 Physician's Weekly; A Whittle Communications Publication; reprinted with permission
YES. The lifesaving benefits of mammographic screening after 50 were well-documented by the Health Insurance Plan (HIP) of New York study and the Breast Cancer Detection Demonstration Project (BCDDP). But initially these studies offered limited evidence on whether screening asymptomatic women under 50 is worthwhile.
Now, however, a four-to-11-year BCDDP follow-up strongly indicates that screening gives significant benefits to younger women as well. More than a third of cancers in women 40 to 49 were detected by mammography alone; 46 percent were not palpable.
For women with invasive cancer, BCDDP's five-year and eight-year survival rates were 87 and 81 percent, respectively, versus the 74 and 65 percent reported by the National Cancer Institute's SEER program. And recent follow-up of HIP participants shows a 24 percent mortality reduction among women under 50 who entered the screening program 18 years earlier. Ongoing Swedish trials have begun to show a corresponding trend.
While younger women are admittedly more susceptible to radiation than older women, the low levels emitted by today's equipment pose little risk to those from 40 to 49.
Average dose for two views has dropped to 0.3 rads for low-dose film, and to 0.5 to 0.7 rads for Xerox imaging -- well below the 2 rads Harvard's Dr. John Bailar used to estimate the risk. Theoretically, 0.3 rads could induce one cancer per year for each million women screened. By comparison, about 130,000 U.S. women a year develop breast cancer.
Routine screening's benefits clearly outweigh its risks.
Gerald D. Dodd, MD
professor and head of the Division of Diagnostic Imaging, M.D. Anderson ospital and Tumor Institute, Houston; chairman, American College of radiology -- American Cancer Society Ad Hoc Committee on Screening Mammography
NO. Screening of asymptomatic younger women can be justified only by a demonstrable mortality benefit. But that's hard to prove from BCDDP data. BCDDP was never designed as a scientific study, so it has no control group. And participants, mainly white and affluent, hardly represent a valid cross-section of the population. Since they were self-selected, they tended not only to be aggressive about health care but to have better-than-average access to the system.
Comparisons of BCDDP case-survival rates with SEER rates, which represent a different population, are invalid.
The follow-up report allows for a one-year lead-time bias. But that's just a guess.
Lead time may actually be positively correlated with longevity. Those who survive only briefly after diagnosis may have shorter lead times than those who live longer.
This could make a big difference in the survival curves.
The new report may also be wrong in discounting the possibility of length bias.
Furthermore, younger women's denser, more glandular breasts are harder to image accurately and are more vulnerable to radiation. Starting the screening at 40 rather than 50 multiplies the risk by about four.
Our projections, based on NIH data, show that if a million women start screening at 40, and if two views are taken with the Xerox equipment now in wide use, about 670 will develop radiogenic breast cancer. Screening with low-dose film would cause about 100 radiogenic cancers.
About 90 percent of these projected patients would never have developed the disease if they hadn't been screened for it.
John K. Gohagan, MD
professor of engineering and applied science and associate professor of radiology, Washington University, St. Louis; past member, Cancer Screening Subcommittee and Scientific Working Committee on Breast Self-Examination, National Cancer Institute