Update in Breast Cancer Screening,Prevention,and Treatment

What we know

Meta-analyses of randomized controlled trials (RCTs) examining the impact of screening mammography have shown benefits to screening in terms of reduction in breast cancer incidence and mortality. ¹ But, because none of these trials included women over age 74, and with extremely limited data regarding women ages 70–74, whether these benefits continue with screening over age 70 (an age range in which we know a significant proportion of breast cancer deaths occur) remains unclear. As a consequence of this lack of high-quality data, existing guidelines have not reached consensus regarding when to stop screening, with some citing insufficient evidence regarding the balance of benefits and harms of screening over age 74, and others recommending continued screening until life expectancy is limited. ^2,3 Despite the lack of clarity in guidelines, many women in the United States continue to undergo screening into their 70s and beyond.

While additional prospective RCT data to address the unanswered questions about the potential benefits of screening in this age group are not expected, the push to harness other methodologies in pursuit of more clarity remains. The objective of this study was to estimate the effect of breast cancer screening on breast cancer mortality in women ages 70–84.

Study results

In this observational study, investigators used data from a random sample of Medicare beneficiaries aged 70–84 who enrolled between 1999 and 2008, had no previous history of breast cancer, had a life expectancy of at least 10 years, and who underwent screening mammography. For the 264,274 eligible women, they went on to model 2 screening strategies, referred to as “stop screening” and “continue screening,” to estimate the impact of screening in women ages 70–74 and those ages 75–84. Eight-year breast cancer mortality and incidence were estimated, adjusting for a number of potential confounding variables.

Overall, there were 1,533 breast cancer deaths under the “continue screening” strategy and 1,304 breast cancer deaths under the “stop screening” strategy. For women ages 70–74, the “continue screening” strategy was associated with a reduction in 8-year breast cancer mortality by 1 death per 1,000 women (hazard ratio [HR] 0.78, 95% confidence interval [CI] 0.63–0.95). This benefit was not observed with the “continued screening” strategy in the group of women ages 75–84 (HR 1.00, 95% CI 0.83–1.19). These findings remained consistent across a number of sensitivity analyses.

What this changes or adds

The ability to reach consensus regarding whether and how long mammography screening should be continued beyond age 70 has been hampered by lack of high-quality evidence to date. The authors of this study utilized population-based modeling in an attempt to add clarity to this very important question. Their findings of benefit in women ages 70–74 are consistent with and add support to those of the two previous RCTs that included women in this age group. While this study did not find benefit to continued screening beyond age 75, the practical implications of this finding remain unclear, in part, due to the use of a study period in which current screening technologies (digital mammography, tomosynthesis) were not standard. The use of shared decision-making regarding screening decisions in women 75 and older taking into account life expectancy, overall health, and individual goals and preferences remains advisable and is consistent with current guidelines. ^2,3

What we know

Women with dense breasts experience an increased risk for breast cancer, both because dense breast tissue is an independent risk factor for breast cancer and because screening mammography is less sensitive when performed on dense breasts. ^{4 –6} Whether supplemental imaging should be used as part of routine screening in this population remains unclear, in part, because reported improvements in cancer detection with supplemental imaging have been counterbalanced by increased false-positive rates, increased costs, and remaining questions regarding impact on breast cancer outcomes. As a result, no current U.S. guidelines recommend the use of supplemental imaging based solely on breast density. ² The objective of this study was to evaluate the effect of supplemental MRI on the incidence of interval cancers in women with extremely dense breasts.

Study results

In this multicenter, randomized controlled trial, investigators randomized participants in a 1:4 ratio to either mammography + supplemental MRI or mammography alone. A total of 40,373 women ages 50–75 who were participants in a population-based mammography screening program in the Netherlands were included. All women had extremely dense breast tissue and normal baseline screening mammography. Incidence of interval cancers during a 2-year screening period was assessed and compared between groups as the primary outcome. Interval cancer data were collected through linkage with the Netherlands Cancer Registry. Secondary outcomes included recall rate, MRI cancer-detection rate, false positive rate, and others.

Investigators found that the interval cancer rate was significantly lower in the MRI-invited group (2.5/1,000 screenings) compared to the mammogram-only group (5.0/1,000 screenings). Notably only 59% of the women in the MRI-invited group accepted the invitation, with most not declining or not responding to the invitation to participate. Among those women in the MRI-invited group who actually underwent MRI, cancer-detection rate was 16.5 per 1,000 screenings. In this same group, 95 per 1,000 were recalled for further testing, and the false positive rate was 79.8 per 1,000 screenings.

What this changes or adds

The findings of this study add to the evidence supporting increased sensitivity of breast MRI compared to mammography in women with dense breasts. As has been the case in other investigations of supplemental imaging, this increased sensitivity came with a significant recall rate. Furthermore, the trial design did not allow for assessment of impact of supplemental screening on breast cancer mortality, for evaluation of potential overdiagnosis or for assessment of the impact of ongoing screening with supplemental MRI over time. As a result, fully weighing the benefits versus harms of this screening strategy to inform practice is not possible, and we do not recommend use of supplemental MRI for screening based on breast density alone.

Comstock CE, et al. Comparison of abbreviated breast MRI vs digital breast tomosynthesis for breast cancer detection among women with dense breasts undergoing screening. JAMA 2020;323:746–756.

What we know

Breast MRI has shown benefits in terms of increased cancer detection when used as supplemental imaging in women with dense breasts. However, several limitations as described in the previous article summary have contributed to uncertainty about the use of MRI for screening in this population. A newer modality that consists of a shortened imaging protocol (<10 minutes), called abbreviated breast MRI, has been explored with a goal of reducing the resource requirements and complexity associated with standard MRI. Early data suggest that in addition to the listed advantages, this modality also has equivalent cancer detection rates to standard MRI. ⁷ The objective of this study was to compare the performance of abbreviated breast MRI with that of digital breast tomosynthesis (DBT), the screening modality that has increasingly become standard of care, in women with dense breasts.

Study results

In this cross-sectional multisite study in the United States and Germany, a total of 1,444 women with heterogeneously or extremely dense breasts ages 40–75 underwent screening by both DBT and abbreviated breast MRI in 2016–2017. Invasive cancer was detected in 17 women. While abbreviated MRI detected the invasive cancer in all 17 cases, DBT detected it in only 7 of the 17 women. Cancer detection rates were 11.8 per 1,000 for abbreviated MRI and 4.8 per 1,000 for DBT (p = 0.002 for difference in cancer detection). Notably, abbreviated MRI was associated with a higher false positive rate than DBT (187/1,407 vs. 36/1,407) with significantly lower specificity (86.7% vs. 97.4%). Additional imaging (either immediate callback or short-term follow-up) was recommended less frequently for abbreviated MRI (108/1,444 vs. 146/1,444), but the difference between modalities was not statistically significant.

What this changes or adds

The findings of favorable cancer detection rates for abbreviated MRI compared to DBT add to the collection of early data suggesting future promise for this imaging modality. Further study regarding impact on breast cancer morbidity and mortality, the potential for overdiagnosis and for false positives, and regarding cost-effectiveness will be needed to inform any future implications for screening strategies in women with dense breasts.

What we know

Aromatase inhibitors (AIs) have been shown to reduce breast cancer risk in at-risk women. ^8,9 Until now, long-term follow-up from the two large RCTs (MAP.3, IBIS-II) evaluating AI therapy has not been available. This study reports the findings of blinded long-term follow-up for the IBIS-II trial.

Study results

IBIS-II was a trial that compared 5 years of daily anastrozole 1 mg with placebo for risk reduction in postmenopausal women who were at high risk for breast cancer but had not yet developed breast cancer. After a median follow-up of 131 months, the overall risk reduction associated with anastrozole therapy was 49% (p < 0.0001), having declined over time from 61% at 5 years, when participants stopped the study drug, to 37% thereafter. As expected, risk reduction was greater for estrogen receptor (ER)-positive tumors than for ER-negative tumors. No significant differences were observed in overall or breast cancer-specific mortality between the two groups. Of note, the number of breast cancer deaths were low in both the treatment and placebo groups (two and three deaths, respectively). In terms of adverse effects, incidence of fractures and cardiovascular disease did not differ between the two groups.

What this changes or adds

This report of long-term follow-up from the IBIS-II trial is encouraging in that continued significant risk reduction without the advent of new adverse effects has now been documented for chemoprevention with anastrozole. The observed degree of risk reduction compares favorably to that reported from long-term follow-up of selective estrogen receptor modulator trials, as does the adverse effect profile. Providers should incorporate these updated findings into shared decision-making discussions with women at increased risk for breast cancer who are considering pharmacologic risk reduction. What remains to be seen is whether use of AIs for chemoprevention will be associated with a reduction in breast cancer mortality upon continued long-term follow-up.

What we know

The initial data from the Women's Health Initiative (WHI) study suggested that the combination hormone therapy (HT) with both estrogen and progestin (EPT) was associated with an increase in 8 cases of breast cancer per 10,000 women after 5 years of therapy; in contrast, 5 years of estrogen therapy (ET) alone revealed a reduction in breast cancer risk with 7 fewer cases of breast cancer per 10,000 women. ^10,11 Both of these results were maintained at 13 years of follow-up. ¹¹ In contrast, large and well done observational studies have suggested different results, especially with regard to the risk of ET alone and subsequent breast cancer. A meta-analysis of the data from Collaborative Group on Hormone Factors in Breast Cancer suggested that both EPT and ET were associated with increased breast cancer risk. ¹² In the Million Women Study, both EPT and ET were associated with a higher risk of mortality. ¹³ The objective of this study was to report updated data about breast cancer incidence and mortality for both EPT and ET after 20 years of follow-up.

Study results

A total of 27,347 postmenopausal women were randomized in the WHI to receive EPT, ET, or placebo. Mortality information was available for more than 98% of these women at 20 years of follow-up. Among the 16,608 women randomized to EPT compared with placebo, there was a significantly higher incidence of breast cancer with 584 cases in the treatment group compared to 447 cases in placebo users with a HR of 1.28 (p < 0.001). There was no significant mortality difference between EPT users and placebo users with 71 deaths in the EPT users and 53 deaths in the placebo users (p = 0.11). For the 10,739 women randomized to ET compared with placebo, there was a statistically significant lower breast cancer incidence with 238 cases in the ET compared to 396 cases in the placebo users with a HR of 0.78, p = 0.005. There was also lower mortality in the ET users with 30 deaths compared to 46 deaths in the placebo (HR 0.60, p = 0.04).

What this changes or adds

This study offers long-term follow-up in over 29,000 women enrolled in two randomized, placebo controlled trials of either EPT versus placebo or ET versus placebo. The results of the previous analyses of the WHI studies regarding the impact of HT on breast cancer incidence and mortality stand. EPT is associated with a small but real and sustained risk of breast cancer incidence, but even at 20 years no increase in breast cancer mortality. With regard to ET alone, these results remain reassuring in that there is both a decreased incidence of breast cancer and breast cancer mortality in women treated with ET alone after 20 years of follow-up. It is also important to note that the absolute risk of all outcomes remains extremely small.

What we know

Breast cancer screening programs have resulted in a marked increase in the diagnosis of ductal carcinoma in situ (DCIS), and now this diagnosis accounts for ∼20% of all breast cancer diagnoses from screening. ¹⁴ The long-term risks of invasive breast cancer and death from breast cancer remain controversial, including how these risks evolve over time. It has been suggested that DCIS is being overdiagnosed and overtreated. ¹⁵ The objective of this study was to evaluate the long-term risks of invasive breast cancer and death from breast cancer after a DCIS diagnosis from breast cancer screening.

Study results

All 35,024 women enrolled in the National Health Service Screening Program in England with a diagnosis of DCIS were followed for the development of incident breast cancer and death from breast cancer until December 2014. A total of 13,606 women were followed for up to 5 years; 10,998 for 5–9 years; 6,861 women were followed for 10–14 years; 2,620 women were followed for 15–19 years; and 939 women had been followed for at least 20 years.

A total of 2,076 women developed invasive breast cancer for an incidence rate of 8.8 per 1,000 women per year (95% CI 8.45–9.21). This was significantly higher than the incidence rates of 2.52 per year (95% CI 2.41–2.63) expected from national cancer incidence rates in the overall population. This increase started in the second year after the diagnosis of DCIS and continued until the end of follow-up. In this same cohort, 310 died of breast cancer, corresponding to a death rate of 1.26 per 1,000 women. There was no significant increase in death rate from breast cancer compared to that expected in the population for years 1–5 follow-up, but this rate increased for the remainder of the follow-up. During years 5–9, it increased to 1.98 per 1,000 women and continued to increase, rising to a rate of 2.99 per 1,000 women in years 10–14.

A total of 29,044 women with unilateral DCIS underwent surgery. Those with larger final surgical margins had lower rates of invasive breast cancer. Compared with women whose final margin distances were at least 5 mm, women with a final margin distance of 1–2 mm had an incidence rate of 1.74. The adjusted rate ratio for women with involved margins was even higher, for a rate of 3.73. Increased benefit was also seen with those receiving more intensive treatment (mastectomy, radiotherapy for women with breast conserving surgery, and endocrine treatment in estrogen receptor positive disease). By 15 years after diagnosis of DCIS, the cumulative rate of ipsilateral invasive breast cancer in women with estrogen positive disease who received estrogen endocrine treatment was 4.7%, compared to 7.3% in women who did not receive it.

What this changes or adds

This 20-year follow-up study of women diagnosed with DCIS provides robust data to help clinicians evaluate both the incidence of invasive ductal carcinoma and mortality from breast cancer after this diagnosis. The rates of both invasive breast cancer and death from breast cancer were doubled in this cohort of women. It also speaks to the benefits of aggressive therapy, including wide surgical margins, radiotherapy, and endocrine therapy, when appropriate compared to those treated more conservatively. While there may be women for whom less aggressive therapy would be safe, at this time there is no way to identify which women do not need more intensive treatment.