The Case for Continued Breast Cancer Screening in Older Women

In her article "Breast Cancer Screening: At What Age to Stop?" Dr Mara Schonberg does an excellent job of presenting the pros and cons of continued breast cancer screening in elderly women and of explaining why the current guidelines are vague.¹ My purpose here is not to take issue with anything that she presented, but to make some additional points in favor of continuing breast cancer screening into a woman's "golden" years.

In an article published nearly 10 years ago, I argued that given the higher breast cancer incidence and mortality in older women, as well as the increased life expectancy with little or no activity limitation seen among today's elderly population, consideration should be given to timely breast cancer screening of elderly women.² In addition, since surgical and adjuvant therapies for breast cancer in older women have fewer complications than therapies for other cancers in the elderly, the cost-benefit ratio for breast cancer screening in this age-group may prove to be more promising.

I believe these factors are just as relevant today as they were back then. In fact, Dr Schonberg herself points out that breast cancer–specific mortality rates have risen for women 80 years and older and suggests that these trends may result from underscreening and/or undertreatment of these women. Furthermore, life expectancy continues to increase: the latest figures are 9.8 years for women aged 80 to 81 and 7.2 years for women aged 85 to 86 years.³

BREAST CANCER SCREENING AND MORTALITY

A review of the literature on routine screening mammography in women aged 75 to 84 years identified three studies that found the risk of breast cancer mortality to be approximately two-fold higher among women who did not receive screening compared with women who did.⁴ Two of these studies were cited by Dr Schonberg in her article.^5,6

McCarthy and associates⁵ performed a retrospective cohort study of 9767 American women who had received a first primary diagnosis of breast cancer. These women were classified according to their mammography use during the two years preceding diagnosis: regular users had had two screening mammograms and nonusers had no screening mammograms within this period. A consistently higher risk of death from breast cancer was observed among nonusers compared with regular users, which was statistically significant in the 75- to 84-year age-group (hazard ratio [HR], 2.47; 95% confidence interval [CI], 1.70-3.58) but not in the 85+ age-group (HR, 1.45; 95% CI, 0.63-3.32). The nonsignificant result in the oldest age-group could have been a result of relatively fewer women in this age category (n = 437).

This study also found that previous mammography was strongly associated with earlier stage at diagnosis in each age-group, even after adjusting for factors that are associated with late-stage disease at diagnosis. The results were stronger as women aged (adjusted odds ratios [OR] for stage II or higher disease at diagnosis in nonusers versus regular users were 3.64 [95% CI, 2.96-4.48] for those aged 75 to 84 years and 6.84 [95% CI, 3.97-11.90] for those 85 years or older).

There were certainly problems with this study, including difficulties in distinguishing between screening and diagnostic mammography in the administrative medical code data and perhaps some lead time bias. However, the conclusions of the researchers seem appropriate: that older women who undergo regular mammography receive a diagnosis at an earlier stage and are less likely to die of their disease than those who do not, and that the data support regular mammography in elderly women and suggest that such screening can reduce breast cancer mortality even in those 85 and older.

McPherson and colleagues⁶ performed a retrospective cohort study that included 5186 breast cancer patients aged 65 to 101 years. They compared survival between women whose tumors were diagnosed by screening mammography and those whose tumors were diagnosed clinically (ie, clinical breast examination, self breast examination, incidental finding). Stratification was done not only by age categories but also by level of comorbidity to determine an upper age limit or a quantifiable level of comorbidity that would render mammography screening ineffectual in decreasing mortality.

Women with mammographically detected tumors and no comorbidity had significantly lower relative risks (RRs) of all-cause death in all age-groups (age 75 to 79: RR, 0.36 [95% CI, 0.26-0.49]; age 80 to 84: RR, 0.66 [95% CI, 0.52-0.83]) than women whose tumors were diagnosed clinically. Similar results were seen in women with moderate comorbidity (age 75 to 79: RR, 0.47 [95% CI, 0.25-0.88]; age 80 to 84: RR, 0.52 [95% CI, 0.33-0.80]) and in women with severe comorbidity (age 75 to 79: RR, 0.53 [95% CI, 0.20-1.36]; age 80 to 84: RR, 0.64 [95% CI, 0.30-1.87]), but the latter results did not achieve statistical significance. The investigators concluded that older age, even when coupled with mild to moderate comorbidity, is not a sufficient reason to withhold screening mammography and that early detection might still benefit older women with severe comorbidity by limiting additional morbidity caused by metastases.

Badgwell and associates⁷ used the linked Surveillance, Epidemiology, and End Results (SEER)–Medicare database to evaluate 12,358 women 80 years and older who received a diagnosis of breast cancer between 1996 and 2002. Patients were grouped according to the number of mammograms they had during the 60 months before diagnosis: nonusers (no mammograms), irregular users (one or two mammograms), and regular users (three or more mammograms). On multivariate analysis, patients were 0.37 times less likely to present with late-stage cancer for each mammogram obtained (OR, 0.63; 95% CI, 0.63-0.67). Breast cancer–specific, five-year survival was 82% among nonusers, 88% among irregular users, and 94% among regular users.

Although survival from causes other than breast cancer was also associated with mammography use, which suggests a bias for healthier patients to undergo mammography, the authors concluded that the results do at least point to some potential benefit in older women, specifically a greater likelihood of diagnosing early-stage disease in patients who obtained at least three mammograms within the five-year period preceding the diagnosis. The results also suggest that few women who receive a diagnosis of breast cancer after age 80 are obtaining screening mammograms regularly. This is consistent with another study that found extremely low biennial mammography rates, ranging from 24% to 37% for women aged 80 to 84 and 14% to 23% for women aged 85 to 89 from 1991 to 2001.⁸

RECOMMENDATIONS FOR CLINICAL PRACTICE

I certainly agree with Dr Schonberg that some women 75 years and older in good health with excellent functional status would likely benefit from mammography screening, but I am not as comfortable with her conclusion that others in poor health and with short life expectancies would probably not benefit. I suggest that physicians concentrate their efforts on discussing the potential benefits of screening mammography with their older patients and not be distracted by efforts to discuss stopping screening with older women who have shorter life expectancies. By taking this approach, physicians can significantly increase the number of older women who regularly obtain screening mammograms.