Proceedings of the 2013 annual meeting of the Netherlands Epidemiology Society
Volume 1 Issue S1 Abstract 18
X.A. Phi, University Medical Center Groningen, Groningen, the Netherlands
N. Houssami, Sydney Medical School, University of Sydney, Sydney, Australia
M. Leach, MARIBS, Institute of Cancer Research, Sutton, United Kingdom
F. Podo, HIBCRITS, National Institute of Health, Rome, Italy
C.C. Riedl, Medical University of Vienna, Vienna, Austria
F.Sardanelli, HIBCRITS, University of Milan School of Medicine
M.Tilanus-Linthorst, MRISC, Erasmus University Medical Centre, Rotterdam, The Netherlands
E. Warner, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Canada
I. Trop, Centre of Breast Imaging, Hospital of Montreal, Montreal, Quebec, Canada
G.H De Bock, University Medical Center Groningen, Groningen, the Netherlands
There is no consensus on the most effective strategy (mammography/MRI) for breast cancer screening in women with BRCA gene mutations. We aimed to evaluate the contribution of MRI to mammography in breast screening in BRCA gene mutation carriers (of all ages) using individual patient data (IPD) meta-analysis.
IPD were provided by investigators from six out of 13 relevant publications. Study-specific data was extracted and reported to the study’s investigators to confirm (or clarify) results, following which six databases were pooled and analysed. Only women with BRCA1 or BRCA2 mutations were included. A BI-RADS 3,4,5 were considered as positive imaging result. The standard reference was needle or excision biopsy, or confirmation of interval cancer. Pooled analysis was carried out under the assumption of study homogeneity. Sensitivity and specificity were calculated in cases where both MRI and mammography were performed and compared with mammography and the combination of MRI and mammography using McNemar’s test.
There were data on 2,045 women (mean age= 42.03, SE = 0.228) and 5,340 completed screening rounds. 163 tumors were detected including 16 interval cancers. Sensitivity and specificity of mammography were 35.58% and 94.60% respectively. Those of the combination of MRI and mammography were 87.12% and 83.34% correspondingly.
Overall, in BRCA gene mutation carriers, MRI is an effective adjunct to mammography. Our next analysis will assess whether MRI is similarly effective for BRCA1 as well for BRCA2 and for all age groups.
Published: 06 Jun, 2013