SAN ANTONIO--Screening mam-mography significantly improves breast cancer survival in women ages 40 to 49, compared with other methods of cancer detection, a Minnesota study suggests.
SAN ANTONIO--Screening mam-mography significantly improves breastcancer survival in women ages 40 to 49, compared with other methodsof cancer detection, a Minnesota study suggests.
Women whose cancers were detected by mammograms had a 94% five-yearsurvival, compared with 84% for breast self-examination (BSE),79% for patient incidental discovery, and 75% for clinical breastexamination.
Detection by clinical exam or incidental discovery doubled theodds for death within 5 years of discovery, Charles L. Murray,MD, reported at the San Antonio Breast Cancer Symposium. The studywas performed by Carol McPherson and colleagues at Methodist Hospital,Health-System Minnesota.
The study involved 969 breast cancer patients who were betweenthe ages of 40 and 49 at the time of diagnosis. The patients wereidentified through an upper midwestern tumor registry and camefrom 23 hospitals located in Minnesota, North Dakota, and SouthDakota.
Only invasive cancers were included in the analysis. Patientswere followed up to 8 years after treatment, with a mean follow-upof almost 3 years.
The largest number of patients (363) detected their breast cancersincidentally. Almost a third of the cancers were detected by screeningmammog-raphy. The fewest were identified via clinical examination.
Breast cancers detected by mammog-raphy were significantly smaller,averaging 1.85 cm, compared with 2.32 cm for clinical exam, 2.79cm for BSE, and 2.91 for incidental discovery. With mammog-raphy,the likelihood of detecting cancers while still localized wassignificantly greater, 75.8% versus 52% to 56% for the other detectionmodalities, said Dr. Murray, a medical oncologist at Park NicolletCancer Center, HealthSystem Minnesota, St. Louis Park.
Five-year survival correlated with tumor size at diagnosis. Amongpatients with tumors smaller than 1 cm, 5-year survival was 98%but dropped below 85% for tumors between 1.1 and 1.5 cm, and to60% for tumors larger than 5 cm.
Using the mammography group as the reference, Dr. Murray and hiscolleagues found that the relative risk of death at 5 years roseto 1.7 for BSE (not statistically significant). For incidentaldiscovery and clinical exam, the relative risk increased significantlyto 2.0 and 2.4, respectively.
Given the patient distribution among the detection methods, theinvestigators would have predicted 30 breast cancers in the mammographycohort, while the observed number was 12. The BSE group had thesame number of cancers as predicted, but the incidental discoveryand clinical exam groups had more cancers than expected.
In a multiple regression analysis, the survival difference betweenmammog-raphy and both clinical breast exam and incidental discoveryremained significant after controlling for tumor size and lymphnode status.
Dr. Murray acknowledged that the relatively short follow-up maylimit the applicability of the findings. He also acknowledgedthat some unidentified selection bias, such as a higher levelof health consciousness, might have favored mammography.
Despite the limitations, he concluded that "women with invasivebreast cancer found by mammograms had significantly smaller tumors,tumors that were significantly more likely to be localized, andapproximately half the mortality of women whose cancers were discoveredby clinical exam or incidentally. Smaller tumor size and lymphnode status did not fully account for the difference in survival."
In response to a question, Dr. Murray said that the investigatorsdid not know whether a substantial number of women in the self-examgroup might have had false-negative mammograms that created afalse sense of security.