Abstract
“Are increasing MRI rates and increasing mastectomy rates related? At this point, there seems to be little remaining doubt, as sufficient data exist to support the fact that MRI is associated with a higher rate of mastectomy for early-stage breast cancer.”
Ever increasing amounts of data support the assertion that the use of breast MRI is associated with more invasive surgery, be it mastectomy versus breast conservation therapy or bilateral mastectomy versus unilateral mastectomy [1–3]. However, perhaps an equally important question is whether the observed increase in the rate of mastectomy related to the use of MRI is improving patient outcomes? Evidence supports the fact that more extensive breast cancer surgery is not necessarily better for women with respect to local recurrence (in cases of wider local excision), distant metastases or other more qualitative outcome measures, such as quality of life and body image. It is critical to determine whether preoperative MRIs are improving patient outcomes, given their impact on clinical care, the likely associated anxiety due to additional clinical findings and substantial costs associated with their use. These costs may be substantial, as not only are the MRIs themselves costly, but costs related to additional workup and surgical treatment are substantive as well.
Finding additional disease
There is no question that preoperative breast MRI detects mammographically and/or sonographically occult cancer in both the ipsilateral breast and in the contralateral breast [4,5]. MRI has superior sensitivity compared with mammography and ultrasound [6]. A meta-analysis by Houssami demonstrated that preoperative breast MRI detected additional areas of disease in the same breast in approximately 16% of women with newly diagnosed breast cancer [7]. A large study from 2007 revealed additional occult disease in the contralateral breast in 3% of newly diagnosed breast cancer patients who underwent preoperative breast MRI [8]. Although many of these additional foci were ‘satellite’ lesions in close proximity to the index lesion, multicentric disease (in a separate quadrant of the breast) was also not uncommonly diagnosed.
Rising rates of MRI use
In the context of increasing recognition that MRIs can detect clinically occult disease, the use of breast MRI has diffused rapidly. A recent study evaluated trends in MRI use by analyzing commercial insurance claims for over 52,000 women with newly diagnosed breast cancer aged 65 years and under between 2005 and 2008 [9], and showed that the use of breast MRI more than doubled from 20 to 50%. This year, our group demonstrated that among women in the USA over the age of 66 years covered by the Medicare program, the use of MRI also increased rapidly between the time period 2000 and 2009 from 0.8% of those newly diagnosed with breast cancer to 25.2% [3]. During the final year of this study even 10% of women aged 84–95 years had a preoperative breast MRI. The reasons for this rapid increase are not entirely clear, especially given the lack of benefit with regard to long-term outcomes for breast cancer when breast MRI is performed. This rapid increase is also concerning, as breast MRI is an expensive test and has a high false-positive rate.
Rising rates of mastectomy
In the years following 1990, when the NIH Consensus Development Panel declared that breast conservation therapy was the preferred method of primary surgical treatment for women with early-stage breast cancer [10], mastectomy rates began declining. By 1995, the rate of breast conserving therapy had nearly doubled, from 35% in 1989 to 60% among women with stage 1 disease, and for those with stage 2 disease comparable rates were 19–29% over the same time period [11,12]. However, several studies have documented a significant rise in mastectomy rates since that time. McGuire et al. published a single institution series in 2009 which documented a steady rise in the mastectomy rate, from just over 30% in the late 1990s to approximately 60% in 2007 [13] and Mahmood et al. reported increasing national rates among patients with early breast cancer [14]. Similarly, the rate of bilateral mastectomy has been increasing. An analysis of the Surveillance, Epidemiology, and End Results database by Tuttle et al. demonstrated that the use of contralateral prophylactic mastectomy among patients with unilateral breast cancer markedly increased from 1998 to 2003, going from 4.2% in 1998 to 11.0% in 2003 [15], and a study from Memorial Sloan—Kettering demonstrated that among nearly 3000 women presenting with unilateral breast cancer between 1997 and 2005, 13.8% chose contralateral prophylactic mastectomy [16].
“It is critical to determine whether preoperative MRIs are improving patient outcomes, given their impact on clinical care, the likely associated anxiety due to additional clinical findings and substantial costs associated with their use.”
Are increasing MRI rates and increasing mastectomy rates related? At this point, there seems to be little remaining doubt, as sufficient data exist to support the fact that MRI is associated with a higher rate of mastectomy for early-stage breast cancer. A large meta-analysis performed this year looked at nine eligible studies: two randomized controlled trials and seven comparative cohorts. The analysis included a total of 3112 patients, and demonstrated that MRI increases mastectomy rates [1]. Preoperative MRI increases the likelihood of contralateral prophylactic mastectomy as demonstrated by Sorbero et al. who reported a nearly two-times higher risk of contralateral prophylactic mastectomy when preoperative breast MRI was performed [17]. Although other factors play a role, preoperative breast MRI was the strongest factor predicting for the use of contralateral prophylactic mastectomy, with an odds ratio of over 3. Last, a recently reported German study of over 142,000 cases reported that breast MRI was independently associated with mastectomy (odds ratio: 1.42; 95% CI: 1.36–1.47) [18].
More mastectomies, more complications, higher costs
Is it not a good thing to find more cancer at an earlier stage with more advanced imaging? As others have pointed out, not if there is no improvement in clinical outcomes and no short-term benefits to patients. The National Surgical Adjuvant Breast and Bowel Project trials performed before the widespread use of preoperative breast MRI demonstrate 10-year rates of local recurrence consistently under 5%. Undoubtedly there were participants who had undiagnosed multicentric and contralateral cancers, which might have been diagnosed with preoperative MRI. It had been speculated that preoperative breast MRI would reduce the need for re-excision, reduce local recurrence, and possibly affect long-term survival from breast cancer. In reality, it is not clear that MRI has shown a difference with respect to any of these measures; however, long-term data will require many more years to be sure. What breast MRI does do in many cases is increase patient anxiety, increase the need for additional imaging and biopsy, drive-up costs, and delay definitive surgical treatment.
Financial & competing interests disclosure
CP Gross receives support from Medtonic Inc. and 21st Century Oncology. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.