|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2008.21.2498 on February 9 2009 © 2009 American Society of Clinical Oncology.
In ReplyDivision of General Surgery, Mayo Clinic, Rochester, MN
Division of Medical Oncology, Mayo Clinic, Rochester, MN
Division of Biostatistics, Mayo Clinic, Rochester, MN Amir and Freedman raise important points about potential explanations for the poor performance of the Gail model1,2 in women with atypical hyperplasia (atypia), and the need for alternative risk prediction tools, such as the Tyrer-Cuzick model,3 which incorporates extensive family history data. To our knowledge, our study4 was the first assessment of the accuracy of the Gail model in a cohort of women with atypia, whose tissue was reviewed centrally by a breast pathologist, and who had long-term follow-up for cancer events. Much of the poor performance of the Gail model in women with atypia may be explained by the fact that atypia was not considered in the initial development of the Gail model. It was added to the original model in a post-hoc fashion using an estimate of prevalence of atypia of 7.8%, and an atypia-associated relative risk of breast cancer of 1.96. Larger studies of atypia using the current criteria of Page et al5 have suggested that the prevalence of atypia is closer to 4%, with a relative risk of approximately 4. Other possible explanations are suggested by our results listed in Table 1 of our article,4 where the risk factors included in the Gail model did not significantly stratify risk in atypia. It is possible that the risk associated with these factors (eg, family history) is already reflected in the tissue phenotype of atypia, although larger studies are required to confirm this possibility. The Tyrer-Cuzick model indeed may perform better than the Gail model in estimating the risk of breast cancer in women with atypia. It incorporates an estimated relative risk for atypia of the same magnitude seen in more recent studies. Risk estimates are also obtained against the backdrop of extensive family history data. Use of family history allows the model to anchor a woman's estimate to her likelihood of carrying a mutation in the BRCA1 or BRCA2 genes, and the respective penetrance of each of these genes. Unfortunately, it is not likely that mutations are the predominant cause of the incidence of breast cancer observed in our cohort.4 In our 331 patients with atypia, only 4% had two or more first-degree relatives with breast cancer, and the majority (60%) had no family history of breast cancer. One of the major limitations of statistical models is found in the very nature of models for risk prediction. These models are constructed using regression-based approaches. These approaches tend to develop models that are well behaved on average. Therefore, regression-based models tend to perform well across the full range of study groups that are reflective of the data used in model development. However, in specific subgroups, the models may not perform as well, particularly when these subgroups are dissimilar to the larger study groups. Taken to the extreme, if models were used for risk prediction in an individual, only the most robust and comprehensive model would be expected to perform adequately. Indeed, although current risk prediction models can provide a rough stratification of the risk of breast cancer, no model exists capable of providing an estimate of the risk of breast cancer for an individual woman as reliably as would be desired for the perfect application of individualized risk management or mitigation strategies. Currently, the Gail model remains the most widely used clinical tool by physicians for breast cancer risk assessment and counseling. It is hoped that our article4 raises physicians' awareness of limitations of the Gail model, particularly in the high-risk group of women with atypia. Additional model development is required for more refined stratification of breast cancer risk. The identification of novel markers of breast cancer risk—whether genetic, tissue based, or otherwise—is likely to provide the best chance for improving our ability to assess the risk of breast cancer at the individual level. We have begun to search for such markers, and have found that the number of foci of atypia6 and the level of COX-2 expression7 are associated with the risk of breast cancer development in women with atypia. We feel that the future of risk prediction is likely to be enhanced by the incorporation of additional biomarkers measured from biopsy-derived tissue factors. It is important that efforts to evaluate existing models, and to develop improved risk prediction tools, continue. As our ability to accurately predict the risk of breast cancer development improves, our ability to better tailor prevention strategies to individual women will likewise increase. AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. ACKNOWLEDGMENT Supported by Department of Defense Center of Excellence Grant No. FEDDAMD17-02-1-0473-1, National Institutes of Health Grant No. R01 CA46332, Susan G. Komen Breast Cancer Foundation Grant No. BCTR99-3152, the Regis Foundation for Breast Cancer Research, and the Fred C. and Katherine B. Andersen Foundation. REFERENCES
1. Gail MH, Brinton LA, Byar DP, et al: Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst 81:1879–1886, 1989. 2. Costantino JP, Gail MH, Pee D, et al: Validation studies for models projecting the risk of invasive and total breast cancer incidence. J Natl Cancer Inst 91:1541–1548, 1999. 3. Tyrer J, Duffy SW, Cuzick J: A breast cancer prediction model incorporating familial and personal risk factors. Stat Med 23:1111–1130, 2004.[CrossRef][Medline] 4. Pankratz VS, Hartmann LC, Degnim AC, et al: Assessment of the accuracy of the Gail model in women with atypical hyperplasia. J Clin Oncol 26:5374–5379, 2008. 5. Page DL, Dupont WD, Rogers LW, et al: Atypical hyperplastic lesions of the female breast: A long-term follow-up study. Cancer 55:2698–2708, 1985.[CrossRef][Medline] 6. Degnim AC, Visscher DW, Berman HK, et al: Stratification of breast cancer risk in women with atypia: A Mayo cohort study. J Clin Oncol 25:2671–2677, 2007. 7. Visscher DW, Pankratz VS, Santisteban M, et al: Association between cyclooxygenase-2 expression in atypical hyperplasia and risk of breast cancer. J Natl Cancer Inst 100:421–427, 2008.
Related Article
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2009 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
