|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 23 (December 1), 2004: pp. 4859 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.04.160
Get the Facts StraightHarvard Medical School, Massachusetts General Hospital, Department of Radiology, Wang Ambulatory Care Center, Avon Foundation Comprehensive Breast Evaluation Center, Boston, MA To the Editor: The article by Demicheli et al1 concludes that premenopausal women should not be screened for breast cancer because they contend that the earlier treatment of breast cancers among these women hastens their demise. It would seem that the authors have a hypothesis that they wish to prove, and that they have misunderstood the facts, thinking that they fit the hypothesis. Peer review should have picked up the important, factually incorrect statements that seem to lend support to their hypothesis. Since their argument is based on factually incorrect information, and this could result in the unnecessary loss of life, the authors should provide corrections. Contrary to their assertions, the randomized controlled trials of mammography screening do not "all" show an early mortality excess for younger women. As with any such trials, the results in the early years reveal very small numbers. There is clearly statistical fluctuation that is to be expected since, even in the absence of screening, women are not expected to die from breast cancer at exactly the same time in the two groups of a trial. If these small numbers had any validity, then large trials would never be necessary. One need only look at the graphs of the survival data from the Malmo I trial2 and the Gothenberg trial,3 as well as the combined Swedish data4 to see this effect. The curves weave back and forth in the early years of follow-up, consistent with statistical fluctuation. Malmo II shows a clear benefit from the beginning for "younger" women.2 It has also been suggested that some deaths among the control group that were not counted may have been due to undiagnosed breast cancer. The authors also failed to take into account the fact that deaths among women in the screened group may also include women who refused, and were never screened. The analysis by Cox that is cited represents an artifact of incomplete analysis. Any adjustment for multiple testing would render the observation nonsignificant. The article provides two other importantly incorrect statements. There are, indeed, animal studies suggesting that, for some cancers, following the removal of the primary lesion, the growth rate of metastatic lesions does increase. What the authors neglected to point out is that this effect is transient, lasting for a few hours to, at most, a few days. They provide additional misleading information by suggesting that this results in earlier death. With the possible exception of their reference 15 from 1979, the other laboratory studies specifically state that this phenomenon did not lead to earlier deaths. Their own survival curves fail to show any earlier mortality among the premenopausal women. They do not provide sufficient data to permit an independent understanding of their analysis. None of their observations are statistically significant, with the exception of the women with very large cancers (generally not seen in screening programs). In fact, the survival curves suggest that there is no difference in survival for women with T1 lesions, and that the younger women have better survival, early on, for T2 and T3 cancers (contradicting their own interpretation). Furthermore, since their data come from the prescreening era, there were likely few of the T1a and T1b lesions that are now found by screening. Contrary to their contention, there is no scientific evidence that screening women younger than 50 years (or premenopausal women) leads to an excess of early breast cancer deaths. To the contrary, the randomized, controlled trials have demonstrated a benefit for women ages 40 to 49 years, and this has been confirmed in large studies of service screening in Sweden.5,6 Finally, since women in the screening trials who were allocated to screening and refused are still counted in the analyses as having been screened, the authors have no way of knowing whether the deaths in the screened groups (with the exception of the Canadian trial of volunteers) were among women who actually were screened. Author's Disclosures of Potential Conflicts of Interest The author indicated no potential conflicts of interest. REFERENCES
1. Demicheli R, Bonadonna G, Hrushesky WJM, et al: Menopausal status dependence of early mortality reduction due to diagnosis of smaller breast cancers (T1vT2-T3): Relevance to screening. J Clin Oncol 22:102-107, 2004 2. Andersson I, Janzon L: Reduced breast cancer mortality in women under 50: Update from the Malmo Mammographic Screening Program. J Natl Cancer Monogr 22:63-67, 1997 3. Bjurstam N, Bjorneld L, Duffy S, et al: The Gothenberg Breast Cancer Screening Trial: Preliminary results on breast cancer mortality for women ages 39-49. J Natl Cancer Monogr 22:53-55, 1997 4. Larson L-G, Andersson I, Bjurstam N, et al: Updated overview of the Swedish randomized trials on breast cancer screening with mammography: Age group 40-49 at randomization. Monogr Natl Cancer Inst 22:57-61, 1997 5. Tabar L, Vitak B, Tony HH, et al: Beyond randomized controlled trials: Organized mammographic screening substantially reduces breast carcinoma mortality. Cancer 91:1724-1731, 2001[CrossRef][Medline] 6. Duffy SW, Tabar L, Chen H, et al: The impact of organized mammography service screening on breast carcinoma mortality in seven Swedish counties. Cancer 95:458-469, 2002[CrossRef][Medline]
Related Article
Related Reply
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
