ATLANTA-Is there any value to yet another debate on screening mam-mography for women aged 40 to 49? Maybe so when one of the speakers is a breast cancer survivor whose disease was diagnosed mammographically at age 39. Especially since that speaker-Frances M. Visco, JD-spoke against universal mammographic screening for younger women.
ATLANTAIs there any value to yet another debate on screening mam-mography for women aged 40 to 49? Maybe so when one of the speakers is a breast cancer survivor whose disease was diagnosed mammographically at age 39. Especially since that speakerFrances M. Visco, JDspoke against universal mammographic screening for younger women.
Taking the pro side at the debate, held at the Perspectives in Breast Cancer symposium, Hiram S. Cody, III, MD, of Memorial Sloan-Kettering Cancer Center, noted that the controversy is one thats not going to go away.
Eight Randomized Trials
Dr. Cody reviewed the most recent metaanalysis of the eight randomized trials of mammography screening that included women in the 40- to 49-year-old age group. This report showed a gradual increase in risk reduction with longer follow-up, with an overall mortality reduction of 16%, not quite significant.
When the data were reanalyzed with the controversial Canadian trial excluded, the mortality reduction reached significance at 24%, comparable to the reduction seen in women aged 50 and over. Dr. Cody believes it is appropriate to exclude the Canadian data since the screened group in that trial included a number of women who already had breast cancer at the time of randomization.
It appears that in younger women, the mortality reduction is probably comparable to that in older women, but you need a longer period of follow-up to determine that, and its more difficult to prove, Dr. Cody said.
Researchers have estimated that an adequate trial of mammography screening in younger women would require enormous numbers of participants (at least 300,000 in each arm). In all eight of the randomized trials, we have just 180,000 subjects, Dr. Cody said, adding that in his opinion, there are not going to be other randomized trials of mammo-graphic screening. Its over.
In his own experience, Dr. Cody said, breast cancers are increasingly being diagnosed by mammographic screening. He reviewed 1,096 breast cancer patients seen at Memorial Sloan-Kettering from 1979 to 1993, a time during which mammog-raphy came into general use.
In the women under age 50, 29% of cancers were diagnosed mammographically in recent years; for women over age 50, that figure was 42%. The flip side of this is that a large fraction of breast cancers are still diagnosed on clinical exam, either by physicians or patients. So we cannot abandon self-exam or physician exam, he said.
The percentage of early stage (T1) cancers doubled over the study period in both age groups, as did the proportion of DCIS and microinvasive cancers. It appeared to me that the impact of mam-mography was just as great in younger women as in those over age 50, he said.
The rate of node-negative cancers also increased, but not to the same extent as that of T1 or DCIS disease. Mammography may not be reducing the fraction of node-positive cancers as much as we would like, he said. The implication is that perhaps the more aggressive cancers are less likely to be picked up mam-mographically.
In terms of the cost effectiveness of mammographic screening for younger women (assuming a 20% mortality reduction with screening), studies show a cost of $10,000 to $20,000 per year of life expectancy gained, which compares favorably with other widespread preventive interventions.
Mammography comes out in the same cost range as screening for cervical cancer with a Pap test, less expensive than treating hypertension or requiring seat belts and airbags in cars ($32,000 per year of life saved), and much less expensive than treating a patient for an elevated serum cholesterol level ($154,000 per year of life saved), Dr. Cody commented.
Ms. Visco, an attorney who heads the National Breast Cancer Coalition, started her talk with her conclusion: That the NIH Consensus Panels recommendations on mammography screening in women aged 40 to 49 were correct.
She summarized the panels conclusions as follows: The data available do not warrant universal recommendation for all women in their 40s; each woman should decide for herself in consultation with her physician; women should have access to the best possible relevant information to help make the decision; and for women in their 40s who choose mammography, the cost of the procedure should be reimbursed.
If a woman decides to have a mam-mogram in her 40s, she should be supported in every way possible; if she decides not to, she should not be demeaned for that decision, Ms. Visco said.
The real question, she believes, is one of public health. Is population screening by mammography of women aged 40 to 49 an appropriate public health measure? Just because screening mammography in these women may be cost effective compared with other widely accepted measures, she said, doesnt mean that it should be adopted as public policy.
In fact, she said, the annual cost of universal mammographic screening of younger women would be enormous$1.78 billionand that figure does not include the cost of subsequent diagnostic workups. If the goal is saving womens lives, she said, it would make more sense to use that money to buy health insurance for the women in this country who are underinsured and uninsured.
Ms. Visco expressed amazement that the oncology community continues to act as though this issue is the most important question in breast cancer, and also at the outrage that met the consensus panels recommendation. We should save our outrage for the fact that we dont know how to prevent or cure this disease, and that tens of thousands of women have no access to health care.
She also noted an air of condescension in those who opposed the panels statement on the grounds that women in their 40s will be confused if not given unequivocal advice on this issue. There is no outcry that men are going to be confused by the lack of a clear recommendation on the treatment of prostate cancer, for example, she said.
She stressed that the primary goal of screening is to decrease mortality from the disease, not to find the disease. And in her opinion, data from the eight randomized trials of mammography screening do not show a clear mortality reduction in younger women.
The best estimate is a 17% reduction, which is not sufficient in the context of public health pronouncements, she said. Further, these numbers have been massaged and re-analyzed so often that you have even less confidence in the result.
She focused on the point that none of the trials was designed to answer the question under debate: Whether screening mammography in younger women reduces mortality. A trial of that design would randomize women to mammog-raphy at age 40 or mammography at age 50, she said, adding that a trial with that design is underway now in the UK.
She also noted that no trial is perfect. Since all of the randomized trials had problems of one kind or another, she considers it inappropriate to toss out only the Canadian data.
Ms. Visco saved her most scathing remarks for the intrusion of politics into the medical debate. Shortly after the consensus panel findings, the Senate passed a resolution 98 to 0 that the National Cancer Advisory Board should request NCI to develop clear guidelines to recommend screening to younger women.
She noted with some irony that the National Breast Cancer Coalition has been lobbying Congress to provide insurance for women whose breast cancers are diagnosed through the CDC Breast and Cervical Cancer Screening Program.
These are under- and uninsured women who find they have breast cancer through a federally supported program but have nowhere to turn for service. We cant get the US Senate to support that issue. She also asserted that the congressional subcommittees that determine NCI funding have pressured the Institute to recommend screening.
We sometimes have to make public policy in the absence of clear science, she said, but we cant invent clarity where there is none. The clear message, get screened, is not the right message.