Abstract
Background and Objectives
More extensive surgical treatments for early stage breast cancer are increasing. The patterns of preoperative MRI overall and by stage for this trend has not been well established.
Methods
Using Breast Cancer Surveillance Consortium registry data from 2010 through 2014, we identified women with an incident non-metastatic breast cancer and determined use of preoperative MRI and initial surgical treatment (mastectomy, with or without contralateral prophylactic mastectomy (CPM), reconstruction, and breast conserving surgery +/− radiation). Clinical and sociodemographic covariates were included in multivariable logistic regression models to estimate adjusted odds ratios and 95% confidence intervals.
Results
Of the 13,097 women, 2,217 (16.9%) had a preoperative MRI. Among the women with MRI, results indicated 32% higher odds of unilateral mastectomy compared to breast conserving surgery and of mastectomy with CPM compared to unilateral mastectomy. Women with preoperative MRI also had 56% higher odds of reconstruction.
Conclusion
Preoperative MRI in women with DCIS and early stage invasive breast cancer is associated with more frequent mastectomy, CPM, and reconstruction surgical treatment. Use of more extensive surgical treatment and reconstruction among women with DCIS and early stage invasive cancer whom undergo MRI warrants further investigation.
Keywords: Breast cancer surgery, preoperative breast MRI, breast reconstruction
Introduction
The trend away from radical mastectomy to breast conserving surgery (BCS) for treatment of breast cancer in the 1990s [1] has recently been shown to be shifting back towards more extensive surgical treatment in the last decade including for ductal carcinoma in situ (DCIS) and early stage invasive breast cancer. (1) MRI may be employed in the preoperative setting due to its high sensitivity for detecting otherwise occult disease not identified by mammography, (2–4).
Prior studies report that more extensive surgeries are performed after preoperative MRI, including an extensive meta-analysis of surgical outcomes by Houssami et al in 2013, which focused on studies that integrated a control group. (5–7) Such reports raise concerns that women may be undergoing more aggressive surgery related to MRI examinations, the benefit of which is uncertain because evidence has indicated that mastectomy is not always the superior treatment strategy. (8) At the same time, some reports suggest with the greater sensitivity and negative predictive value of MRI compared with mammography, an MRI can confirm whether BCS is feasible and better define extent of disease to increase the likelihood of clear margins. (9)
In addition to mastectomy rates increasing from between 10–20% [1,15,19], women are increasingly opting for bilateral surgeries including women without risk factors that indicate a clear survival benefit to contralateral prophylactic mastectomy (CPM). (10, 11) A number of patient and cancer characteristics have been associated with selection of CPM, including younger age [4,6,13–16], Caucasian race [4,14,15], family history of breast cancer [4,13,15–17], private health insurance [14], DCIS histology [4], and lobular tumor histology. [15,16,18] Recent studies have suggested CPM may coincide with the use of preoperative MRI (12, 13) with one study indicating that women who underwent MRI were twice as likely to have a CPM, (14) while other studies report CPM is associated with breast reconstruction at the time of mastectomy. (15, 16) There are very few studies that include breast reconstruction when examining the influence of preoperative breast MRI on surgical management including CPM.
Although many studies examining the outcomes and associations of preoperative MRI are published [3,6,13,22,23], very few have reported the proportion of women receiving mastectomy versus BCS for Stage 0 (DCIS) and Stage I–III (invasive) cancer in relation to MRI. While either surgical approach may be appropriate, and will generally reflect women’s preferences, it is important to understand how the patterns of therapy in non-metastatic breast cancer may relate to the use of pre-operative MRI. The objective of this study was to assess the surgical management, including breast reconstruction, of Stage 0 (DCIS) and Stage I–III breast cancer with and without preoperative breast MRI in relation to clinical and women characteristics.
Materials and Methods
Data Source
The Breast Cancer Surveillance Consortium (BCSC) comprises regional breast imaging registries across the U.S., each consisting of multiple facilities. (17) Data from the registries are pooled at the BCSC Statistical Coordinating Center and include longitudinal data on breast imaging use and assessments, benign and malignant breast pathology, and linkage to state or regional Surveillance Epidemiology and End Results (SEER) cancer registries. Breast MRI data for this study were available from four BCSC registries: Group Health Cooperative (western Washington State), San Francisco, Vermont, and North Carolina. Each registry received approval for active or passive consenting processes or a waiver of consent by their respective institutional review boards. Data used for analysis were submitted to the BCSC from the four sites in March 2015 and included breast cancer diagnoses from January, 2010 through January, 2014. All procedures were HIPAA compliant and all sites hold U.S. Public Health Service Certificates of Confidentiality, which provide the highest degree of identity protection for participating women and health care providers. (18)
Study Population
The study cohort included women diagnosed with an incident non-metastatic unilateral breast cancer (stage 0 – III) and who received their initial surgical treatment within 6 months of their cancer diagnosis. Women who reported a personal history of breast cancer were excluded.
Definitions
Initial surgical treatments for Stage 0 (DCIS) and early stage invasive breast cancer included mastectomy, CPM, BCS with and without radiation therapy (RT) and breast reconstruction (19). We defined the preoperative window as the time interval between the diagnosis and the initial surgical treatment. Preoperative MRI was the primary exposure of interest; the comparison group had no MRI after diagnosis and before surgery, but may have had other breast imaging. Mode of detection was defined as screen detected (a cancer preceded by a positive [BI-RADS 0,3,4, or 5] screening mammogram within 12 months prior to diagnosis), interval detected (a cancer preceded by a negative screening mammogram within 12 months of diagnosis or a negative diagnostic mammogram between 6–42 months prior to diagnosis) or clinically detected (a cancer with a prior mammogram >42 months or first mammogram or unknown prior mammogram).
Patient and Tumor Characteristics
Women’s age at diagnosis was categorized as (years): <40, 40–49, 50–59, 60–69, 70+. Women who reported at least one or more first-degree relative with a breast cancer diagnosis were considered as having a family history. Tumor characteristics included: invasive histology (ductal, lobular, or both), estrogen and progesterone receptor status (positive or negative), tumor size in cm (<1, 1 to <2, 2 to <5, 5+), grade (0, I, II, III), and nodal status (no positive nodes or positive nodes).
Statistical Analysis
We computed frequency distributions for MRI use and patient and tumor characteristics stratified by surgical treatments of mastectomy with and without CPM and BCS with and without RT. We compared frequency distributions for surgical treatments between women with and without MRI stratified by breast cancer stage and investigated the impact of MRI use on a stepwise progression across mastectomy, mastectomy plus CPM and mastectomy plus CPM and reconstruction. For the stepwise progression, we began by computing the proportion of women receiving mastectomy without CPM. Then among women with mastectomy, we computed the additional proportion of receiving CPM, and finally among women with CPM, the proportion of women with reconstruction.
For each of the 4 surgical treatment types we estimated separate unadjusted and multivariable adjusted logistic regression models to obtain odds ratios in relation to receipt of MRI. These models examined: 1) mastectomy vs. BCS estimated in the full study sample, 2) CPM vs. no CPM estimated among those with mastectomy, 3) reconstruction vs. no reconstruction estimated among women with mastectomy, and 4) RT vs. no RT among women with BCS. Model results are reported as crude and adjusted odds ratios (OR) with 95% confidence intervals (95% CI). Models were adjusted for patient and tumor characteristics including age at diagnosis, race, urban/rural status, family history of breast cancer, year of diagnosis, mode of detection, stage, histology, tumor size, grade, estrogen receptor (ER)/progesterone receptor (PR), nodal status and BCSC site. With regards to missing data, we performed sensitivity analyses, which resulted in the final models containing 71% of the non-missing data. For all analyses, we used SAS 9.4 SAS (SAS Institute Inc. 2015. SAS® 9.4 System Options: Reference, Fourth Edition. Cary, NC: SAS Institute Inc.). Statistical significance was determined based on a two-sided α = 0.05.
Results
Among the 13,097 women with stage 0–III breast cancer, 2,217 (16.9%) of women had preoperative MRI. (Table 1) For surgical treatment, 24.9% of women received a mastectomy without CPM; 8.9% received mastectomy with CPM and 66.2% received BCS. Of the women receiving BCS, 72.1% had BCS with RT and 27.9% had BCS without RT. Of women with a mastectomy, 30.4% underwent breast reconstruction. (Table 1).
Table 1.
Preoperative MRI, year of diagnosis, mode of detection, patient and tumor characteristics by surgical treatment among women diagnosed with non-metastatic breast cancer (N=13,097)
| Total (N=13,097) |
Mastectomy without CPM (N=3,258; 24.9%) |
Mastectomy with CPM (N=1,166; 8.9%) |
BCS without RT (N=2,420; 18.5%) |
BCS with RT (N=6,253; 47.7%) |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Characteristics* | N | (Col %) | N | (Col %) | N | (Col %) | N | (Col %) | N | (Col %) |
| Preoperative MRI | ||||||||||
| No | 10,880 | 83.1 | 2,642 | 81.1 | 844 | 72.4 | 2,090 | 86.4 | 5,304 | 84.8 |
| Yes | 2,217 | 16.9 | 616 | 18.9 | 322 | 27.6 | 330 | 13.6 | 949 | 15.2 |
|
| ||||||||||
| Mode of Detection | ||||||||||
| Screen Detected | 4,976 | 38.0 | 948 | 29.1 | 361 | 31.0 | 1,087 | 44.9 | 2,580 | 41.3 |
| Interval Detected | 2,060 | 15.7 | 644 | 19.8 | 223 | 19.1 | 354 | 14.6 | 839 | 13.4 |
| Clinically Detected | 834 | 6.4 | 302 | 9.3 | 104 | 8.9 | 174 | 7.2 | 254 | 4.1 |
| Unknown | 5,227 | 39.9 | 1,364 | 41.9 | 478 | 41.0 | 805 | 33.3 | 2,580 | 41.3 |
|
| ||||||||||
| Age | ||||||||||
| <40 | 312 | 2.4 | 78 | 2.4 | 101 | 8.7 | 41 | 1.7 | 92 | 1.5 |
| 40–49 | 1,994 | 15.2 | 509 | 15.6 | 358 | 30.7 | 284 | 11.7 | 843 | 13.5 |
| 50–59 | 3,424 | 26.1 | 809 | 24.8 | 364 | 31.2 | 536 | 22.1 | 1,715 | 27.4 |
| 60–69 | 4,108 | 31.4 | 970 | 29.8 | 253 | 21.7 | 654 | 27.0 | 2,231 | 35.7 |
| 70+ | 3,259 | 24.9 | 892 | 27.4 | 90 | 7.7 | 905 | 37.4 | 1,372 | 21.9 |
|
| ||||||||||
| First-degree Family History of breast cancer | ||||||||||
| No | 9,465 | 79.3 | 2,370 | 80.4 | 752 | 71.5 | 1,767 | 79.0 | 4,576 | 80.2 |
| Yes | 2,477 | 20.7 | 579 | 19.6 | 300 | 28.5 | 470 | 21.0 | 1,128 | 19.8 |
|
| ||||||||||
| Year of diagnosis | ||||||||||
| 2010 | 3,023 | 23.1 | 838 | 25.7 | 258 | 22.1 | 429 | 17.7 | 1,498 | 24.0 |
| 2011 | 3,191 | 24.4 | 826 | 25.4 | 263 | 22.6 | 487 | 20.1 | 1,615 | 25.8 |
| 2012 | 3,212 | 24.5 | 759 | 23.3 | 321 | 27.5 | 581 | 24.0 | 1,551 | 24.8 |
| 2013 | 3,324 | 25.4 | 750 | 23.0 | 280 | 24.0 | 761 | 31.4 | 1,533 | 24.5 |
| 2014** | 347 | 2.7 | 85 | 2.6 | 44 | 3.8 | 162 | 6.7 | 56 | 0.9 |
|
| ||||||||||
| Stage*** | ||||||||||
| 0 | 2,437 | 18.6 | 475 | 14.6 | 203 | 17.4 | 561 | 23.2 | 1,198 | 19.2 |
| I | 5,780 | 44.1 | 993 | 30.5 | 375 | 32.2 | 1,094 | 45.2 | 3,318 | 53.1 |
| II | 3,979 | 30.4 | 1,264 | 38.8 | 442 | 37.9 | 668 | 27.6 | 1,605 | 25.7 |
| III | 901 | 6.9 | 526 | 16.1 | 146 | 12.5 | 97 | 4.0 | 132 | 2.1 |
|
| ||||||||||
| Invasive Histology | ||||||||||
| Invasive Ductal | 8,124 | 83.2 | 1,993 | 78.0 | 691 | 76.9 | 1,417 | 85.1 | 4,023 | 86.6 |
| Invasive Lobular | 1,032 | 10.6 | 367 | 14.4 | 150 | 16.7 | 139 | 8.3 | 376 | 8.1 |
| Invasive Lobular and Ductal | 605 | 6.2 | 194 | 7.6 | 57 | 6.3 | 110 | 6.6 | 244 | 5.3 |
|
| ||||||||||
| ER | ||||||||||
| Positive | 10,746 | 83.9 | 2,575 | 80.7 | 927 | 80.8 | 1,933 | 83.0 | 5,311 | 86.5 |
| Negative | 2,060 | 16.1 | 616 | 19.3 | 220 | 19.2 | 397 | 17.0 | 827 | 13.5 |
|
| ||||||||||
| PR | ||||||||||
| Positive | 9,321 | 73.3 | 2,186 | 69.0 | 818 | 71.8 | 1,690 | 72.9 | 4,627 | 76.1 |
| Negative | 3,391 | 26.7 | 984 | 31.0 | 321 | 28.2 | 629 | 27.1 | 1,457 | 23.9 |
|
| ||||||||||
| Tumor Size | ||||||||||
| < 1cm | 3,146 | 24.5 | 494 | 15.4 | 194 | 16.9 | 710 | 30.2 | 1,748 | 28.4 |
| 1 to <2cm | 4,851 | 37.7 | 905 | 28.2 | 364 | 31.8 | 922 | 39.2 | 2,660 | 43.2 |
| 2 to <5cm | 4,054 | 31.5 | 1,307 | 40.7 | 453 | 39.5 | 656 | 27.9 | 1,638 | 26.6 |
| 5+ cm | 815 | 6.3 | 502 | 15.6 | 135 | 11.8 | 63 | 2.7 | 115 | 1.9 |
|
| ||||||||||
| Tumor Grade | ||||||||||
| Grade I | 3,245 | 25.3 | 613 | 19.2 | 217 | 18.9 | 628 | 26.5 | 1,787 | 29.1 |
| Grade II | 5,591 | 43.5 | 1,441 | 45.0 | 504 | 43.9 | 1,018 | 42.9 | 2,628 | 42.8 |
| Grade III | 4,018 | 31.3 | 1,146 | 35.8 | 426 | 37.1 | 726 | 30.6 | 1,720 | 28.0 |
|
| ||||||||||
| Nodal Status | ||||||||||
| No positive nodes | 10,278 | 78.5 | 2,106 | 64.6 | 805 | 69.0 | 1,994 | 82.4 | 5,373 | 85.9 |
| Positive nodes | 2,819 | 21.5 | 1,152 | 35.4 | 361 | 31.0 | 426 | 17.6 | 880 | 14.1 |
Missing (N): family history (1,155), invasive histology (899), ER status (291), PR status (385), tumor size (231) and grade (243).
Breast cancers from January 2014 included.
Derived AJCC stage group, 7th edition (2004+).
Overall, women’s median age at diagnosis (IQR) was 62 (53 – 69) years and 20.7 % of women had a first-degree family history of breast cancer (Table 1). A family history was more common among women with CPM (28.5%). A higher proportion of women diagnosed with an interval cancer received a mastectomy or mastectomy with CPM (19.8% and 19.1%, respectively) compared to those who received BCS with or without RT (14.6% and 13.4%, respectively). The overall distributions of tumor characteristics for the cohort were 83.9% ER positive, 73.3% PR positive, 37.8% tumor size > 2 cm, 25.3% grade I and 78.5% no positive nodes. Among women with invasive breast cancer, 16.8% had lobular or lobular and ductal carcinoma.
We compared surgical treatments stratified by preoperative MRI use. Among women with MRI, 27.8% underwent mastectomy without CPM, versus 24.3% of women without MRI. (Figure 1) Mastectomy with CPM was almost twice as frequent in women who received a MRI (14.5% v. 7.8%). The proportion of women receiving BCS with RT was lower among those with an MRI compared to those without an MRI (42.8% and 48.7%, respectively). The proportion of women with mastectomy who received reconstruction was higher among women receiving MRI (29.9% v. 20.9%); however, there was only a small difference between groups in the frequency of reconstruction among women with CPM (55.3% in the MRI group, 51.4% in the no MRI group (Figure 1).
Figure 1.
Distribution of surgery type, radiation and reconstruction by receipt of preoperative MRI among women with non-metastatic breast cancer (N=13,097)
Abbreviations: MRI = magnetic resonance imaging; CPM = contralateral prophylactic mastectomy; BCS = breast conserving surgery.
Mastectomy with CPM was nearly double for all stages among women with MRI compared vs. without MRI (stage 0: 14.1% vs. 7.4%; stage I: 9.8% vs. 5.9%; stage II: 18.4% vs.9.3%; and stage III: 22.9% v 14.6%; respectively) (Table 2). Mastectomies without CPM for women who received MRI compared to women without MRI were also more frequent for stage 0 – II.
Table 2.
Frequency of surgical treatment by preoperative MRI use, stratified by cancer stage (N = 13,097)
| MRI (N=2,217) | No MRI (N=10,880) | |||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Mastectomy without CPM N = 616 |
Mastectomy with CPM N = 332 |
BCS without RT N = 330 |
BCS with RT N = 949 |
Mastectomy without CPM N = 2,642 |
Mastectomy with CPM N = 844 |
BCS without RT N = 2,090 |
BCS with RT N = 5,304 |
|
|
| ||||||||
| Cancer Stage | N (row %) | N (row %) | N (row %) | N (row %) | N (row %) | N (row %) | N (row %) | N (row %) |
|
| ||||||||
| Stage 0 | 78 (22.0) | 50 (14.1) | 60 (17.0) | 166 (46.9) | 397 (19.1) | 153 (7.4) | 501 (24.0) | 1,032 (49.5) |
| N = 2,437 | ||||||||
|
| ||||||||
| Stage 1 | 189 (20.7) | 89 (9.8) | 141 (15.5) | 492 (54.0) | 804 (16.5) | 286 (5.9) | 953 (19.6) | 2,826 (58.0) |
| N =5,780 | ||||||||
|
| ||||||||
| Stage II | 256 (33.0) | 143 (18.4) | 113 (14.5) | 265 (34.1) | 1,008 (31.5) | 299 (9.3) | 555 (17.3) | 1,340 (41.9) |
| N =3,979 | ||||||||
|
| ||||||||
| Stage III | 93 (53.1) | 40 (22.9) | 16 (9.1) | 26 (14.9) | 433 (59.6) | 106 (14.6) | 81 (11.2) | 106 (14.6) |
| N =901 | ||||||||
Abbreviations: MRI = magnetic resonance imaging; CPM = contralateral prophylactic mastectomy; BCS = breast conserving surgery; RT= radiation therapy.
Comparison of surgical treatments by stage showed higher frequency at all stages of the stepwise progression from mastectomy to mastectomy plus CPM to mastectomy plus CPM and reconstruction for women with preoperative MRI. The differential in frequencies at all stages of the stepwise progression from mastectomy to mastectomy plus CPM to mastectomy plus CPM and reconstruction for women with MRI compared to those without MRI was greatest for those with stage 0 disease (MRI: 22.0% to 36.1% to 76.0% vs. No MRI: 19.1% to 26.5% to 69.9%). Similar but attenuated differences were found for stage I and II (Figure 2). Women with stage III cancer and MRI had a lower proportion of mastectomy than women without MRI. However, mastectomy plus CPM, and mastectomy plus CPM and reconstruction was higher for women with MRI. (Figure 2).
Figure 2.
Stepwise progression of women receiving mastectomy and among those mastectomy plus CPM, and among those mastectomy plus CPM and reconstruction with and without MRI by breast cancer stage (N=4,424)
Abbreviations: MRI = magnetic resonance imaging; CPM = contralateral prophylactic mastectomy.
We examined the odds of each surgical treatment in relation to preoperative MRI. In adjusted logistic regression models, women with MRI were more likely to have a mastectomy (OR = 1.32; 95% CI 1.16–1.50) compared to women without MRI. Among women receiving mastectomy, women with MRI were also more likely to receive CPM (OR = 1.32 95% CI 1.05–1.65) and have breast reconstruction (OR = 1.56; 95% CI 1.24–1.97). Among women treated with BCS, MRI was not associated with use of radiation therapy (OR = 0.84; 95% CI 0.70–1.00) (Table 3).
Table 3.
Unadjusted and adjusted1 odds ratios (OR) and 95% confidence intervals (95% CI) for the association between MRI and surgical treatments among women diagnosed with non-metastatic breast cancer, Stages 0–III (N=13,097)
| Mastectomy vs. BCS (N=13,097) |
Mastectomy with CPM vs. Mastectomy without CPM (N=4,424) |
Reconstruction vs. no Reconstruction (N=4,424) |
BCS with RT vs. BCS without RT (N=8,673) |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MRI | OR | 95% CI | OR | 95% CI | OR | 95% CI | OR | 95% CI | ||||
| Unadjusted | 1.55 | 1.42 | 1.71 | 1.64 | 1.40 | 1.91 | 1.60 | 1.37 | 1.86 | 1.13 | 0.99 | 1.30 |
| Adjusted1 | 1.32 | 1.16 | 1.50 | 1.32 | 1.05 | 1.65 | 1.56 | 1.24 | 1.97 | 0.84 | 0.70 | 1.00 |
Adjusted for age at diagnosis, race, urban/rural status, family history of breast cancer, year of diagnosis, mode of detection, stage, histology, tumor size, grade, ER/PR/nodal status and BCSC site.
Abbreviations: MRI = magnetic resonance imaging; BCS = breast conserving surgery; RT = radiation therapy; CPM = contralateral prophylactic mastectomy; BCSC = Breast Cancer Surveillance Consortium.
Discussion
These national community-based data provide contemporary evidence on the association between preoperative MRI and surgical treatment for breast cancer in the U.S. MRI receipt was associated with a greater likelihood of mastectomy and of CPM and of reconstruction, compared to women with no MRI. Even after adjusting for important clinical and sociodemographic characteristics, MRI was associated with 32% increased odds of mastectomy alone, and 32% increased odds of CPM, and 56% higher odds of reconstruction. These effects were most notable for the earliest stages – particularly DCIS and stage I and II invasive breast cancers. These findings add to the growing evidence that the observed national trends towards more aggressive surgical treatment of early stage breast cancer are associated with preoperative MRI. However, it remains unknown whether this relationship between MRI and more extensive surgery is due to MRI findings or instead patient and/or provider preferences.
National trends away from BCS have been noted since the early-mid-2000s in the U.S., despite the randomized-controlled-trial evidence showing equivalent recurrence and survival rates between BCS and mastectomy, and some morbidity benefits with BCS (20–23). A study of 1.2 million women (mean age in years: 59.6 for mastectomy, 61.6 for BCS) using the National Cancer Database showed a 34% increase in the adjusted odds of having a mastectomy among BCS-eligible women from 2004–2011, but more than a two-fold increase among women with DCIS. (16) At the same time, they found that rates of CPM rose from about 2% to 11%. (16) Within a National Comprehensive Cancer Network (NCCN) study of 10,249 women, no differences were seen in mastectomy relative to BCS from 2000–2009, but there was a significant increase in bilateral mastectomy (from 13% to 30%) among women with mastectomy (23%). (24) A similar increase in CPM was reported in an analysis of SEER data between 2002 and 2012, with a tripling in use of CPM (3.9% to 12.7%) during that time. (11) A rise in reconstruction rates mirrors these trends, and is thought that greater reconstruction options and availability contributory to the increase in use of CPM. (10, 11, 16, 24, 25) These consistent trends from very large, population-based data sources support a pattern of more extensive surgical treatment of breast cancer in the U.S. Our distribution of surgical treatments in our study is consistent with these recent reports.
Over the past decade, there has been a well-documented increase in use of preoperative MRI, (26) which may be a potential influence on surgical treatment trends. (7, 26) There is limited and conflicting evidence on the potential role of MRI in increasing mastectomy rates. A 2002–2009 study of 414 women with stages 0–III breast cancer found an 80% increase in odds of mastectomy compared to BCS for women with MRI (median age 51 years) (OR 1.8; 95% CI 1.1–3.2) (27). However a meta-analysis of 9 studies, focused only on DCIS did not find a significant difference in mastectomy in relation to MRI, which may be due to measuring all mastectomies, rather than mastectomy as the index surgical treatment. (28) Our results in a national sample of over 13,000 women show a 32% increase in the adjusted odds of mastectomy compared with BCS with MRI. Only a few small, single-institution studies have explicitly examined the potential association of preoperative breast MRI in CPM use. These studies have shown between a 70–82% increase from the approximately 3% CPM rate, with MRI use. (13, 29, 30) We report a lower increase in CPM, but significantly increased among women who receive a MRI. The prior studies were based on several hundred women, but included both academic (13, 30) and community-based (29) hospitals, and covered a similar time period to our study and women of similar ages (with CPM: 48.8–50.4 years and without CPM: 56.8–58.9 years). One of these studies (30) examined the association of breast MRI with reconstruction, from a single institution with 446 breast cancer patients from 2002–2012. They found that a significantly higher proportion of women who had undergone mastectomy and preoperative MRI vs. no MRI also underwent reconstruction (83.9% v. 63.6%). Our results add to this evidence given our national sample and large study size and our estimates adjusted for clinical and sociodemographic characteristics that could impact choice of reconstruction.
We were able to examine the association between preoperative MRI and surgical treatment by stage. Given that risk of second ipsilateral cancer events, subsequent contralateral breast cancer, and death are all related to stage, (19) in addition to other factors, the clinical rationale for the most aggressive surgical course is expected to vary by stage. DCIS poses a particular challenge, given the very low risk of breast cancer mortality (via subsequent invasive cancer), and the concomitant greater likelihood of harm from possible overtreatment. (31) We found the greatest relative increase in CPM frequency, comparing the MRI group to the no MRI group among those diagnosed with DCIS. A retrospective chart review of 214 patients with DCIS at an academic medical center found that preoperative MRI did not lead to change in the planned surgical management, but that it was associated with increased mastectomy and CPM. (32) Costs, complications, reoperation, and morbidity have also been noted as concerns with aggressive surgery for DCIS. (1, 12)
In addition to the strength of our large, national sample comprised mostly of community-based settings, recent years of data, and BCSC data which includes women characteristics, this study also has limitations. We were unable to include MRI outcomes in this study, and thus could not determine whether MRI was causative in surgical management plans. We also did not have information on eligibility for BCS, margin status, or insurance information. Although we included many covariates, we could not identify women who were not clinical candidates for a particular surgical approach. We accounted for some established breast cancer risk factors, but did not have BRCA status. Finally, and importantly – we could not account for women’s or providers’ preferences in use of MRI or in use of surgical treatments.
Conclusion
Overall, our study reinforces previously reported trends of increasing mastectomy, CPM and reconstruction in relation to preoperative MRI, and provides new evidence related to the relationship of MRI and stage on surgical treatment and reconstruction for breast cancer. Given that more aggressive treatment of early stage breast cancer, is occurring, some component may be “overtreatment” relative to impact on medical outcomes including survival.
Synopsis.
Extensive surgical treatment for early stage breast cancer has been increasing however, patterns of preoperative MRI in relation to this trend have not been well established. Our study found that preoperative MRI in women with DCIS and early stage invasive breast cancer is associated with more frequent mastectomy, CPM, and reconstruction.
Acknowledgments
We thank the BCSC investigators, participating women, mammography facilities, and radiologists for the data they have provided for this study. A list of the BCSC investigators and procedures for requesting BCSC data for research purposes are provided at: http://www.bcsc-research.org/". The collection of cancer and vital status data used in this study was supported in part by several state public health departments and cancer registries throughout the US. For a full description of these sources, please see: http://breastscreening.cancer.gov/work/acknowledgement.html
We thank Steven Balch of Group Health Research Institute for his assistance with data aquisition and creation of an analytic file for this manuscript.
This work was supported by the National Cancer Institute (R01CA149365), (P01CA154292) and the Breast Cancer Surveillance Consortium (HHSN261201100031C). Vermont Breast Cancer Surveillance System data collection was also supported by U54CA163303.
Footnotes
Disclosures and funding: Author, Sally Herschorn, has stock ownership and is an unpaid investigator for the for-profit healthcare company, Hologic. All remaining authors confirm that they do not have any financial disclosures or conflicts of interest.
References
- 1.Mahmood U, Hanlon AL, Koshy M, Buras R, Chumsri S, Tkaczuk KH, et al. Increasing national mastectomy rates for the treatment of early stage breast cancer. Ann Surg Oncol. 2013;20(5):1436–43. doi: 10.1245/s10434-012-2732-5. [DOI] [PubMed] [Google Scholar]
- 2.Berg WA, Gutierrez L, NessAiver MS, Carter WB, Bhargavan M, Lewis RS, et al. Diagnostic accuracy of mammography, clinical examination, US, and MR imaging in preoperative assessment of breast cancer. Radiology. 2004;233(3):830–49. doi: 10.1148/radiol.2333031484. [DOI] [PubMed] [Google Scholar]
- 3.Esserman L, Hylton N, Yassa L, Barclay J, Frankel S, Sickles E. Utility of magnetic resonance imaging in the management of breast cancer: evidence for improved preoperative staging. J Clin Oncol. 1999;17(1):110–9. doi: 10.1200/JCO.1999.17.1.110. [DOI] [PubMed] [Google Scholar]
- 4.Fischer U, Kopka L, Brinck U, Korabiowska M, Schauer A, Grabbe E. Prognostic value of contrast-enhanced MR mammography in patients with breast cancer. Eur Radiol. 1997;7(7):1002–5. doi: 10.1007/s003300050240. [DOI] [PubMed] [Google Scholar]
- 5.Houssami N, Hayes DF. Review of preoperative magnetic resonance imaging (MRI) in breast cancer: should MRI be performed on all women with newly diagnosed, early stage breast cancer? CA Cancer J Clin. 2009;59(5):290–302. doi: 10.3322/caac.20028. [DOI] [PubMed] [Google Scholar]
- 6.Houssami N, Turner R, Morrow M. Preoperative magnetic resonance imaging in breast cancer: meta-analysis of surgical outcomes. Ann Surg. 2013;257(2):249–55. doi: 10.1097/SLA.0b013e31827a8d17. [DOI] [PubMed] [Google Scholar]
- 7.Killelea BK, Long JB, Chagpar AB, Ma X, Soulos PR, Ross JS, et al. Trends and clinical implications of preoperative breast MRI in Medicare beneficiaries with breast cancer. Breast Cancer Res Treat. 2013;141(1):155–63. doi: 10.1007/s10549-013-2656-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bloom S, Morrow M. A clinical oncologic perspective on breast magnetic resonance imaging. Magn Reson Imaging Clin N Am. 2010;18(2):277–94. ix. doi: 10.1016/j.mric.2010.02.007. [DOI] [PubMed] [Google Scholar]
- 9.Kuhl CK, Schrading S, Bieling HB, Wardelmann E, Leutner CC, Koenig R, et al. MRI for diagnosis of pure ductal carcinoma in situ: a prospective observational study. Lancet. 2007;370(9586):485–92. doi: 10.1016/S0140-6736(07)61232-X. [DOI] [PubMed] [Google Scholar]
- 10.Albornoz CR, Matros E, Lee CN, Hudis CA, Pusic AL, Elkin E, et al. Bilateral Mastectomy versus Breast-Conserving Surgery for Early-Stage Breast Cancer: The Role of Breast Reconstruction. Plast Reconstr Surg. 2015;135(6):1518–26. doi: 10.1097/PRS.0000000000001276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wong SM, Freedman RA, Sagara Y, Aydogan F, Barry WT, Golshan M. Growing Use of Contralateral Prophylactic Mastectomy Despite no Improvement in Long-term Survival for Invasive Breast Cancer. Ann Surg. 2016 doi: 10.1097/SLA.0000000000001698. [DOI] [PubMed] [Google Scholar]
- 12.Rahbar H, Hanna LG, Gatsonis C, Mahoney MC, Schnall MD, DeMartini WB, et al. Contralateral prophylactic mastectomy in the American College of Radiology Imaging Network 6667 trial: effect of breast MR imaging assessments and patient characteristics. Radiology. 2014;273(1):53–60. doi: 10.1148/radiol.14132029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Xia C, Schroeder MC, Weigel RJ, Sugg SL, Thomas A. Rate of contralateral prophylactic mastectomy is influenced by preoperative MRI recommendations. Ann Surg Oncol. 2014;21(13):4133–8. doi: 10.1245/s10434-014-3852-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sorbero ME, Dick AW, Beckjord EB, Ahrendt G. Diagnostic breast magnetic resonance imaging and contralateral prophylactic mastectomy. Ann Surg Oncol. 2009;16(6):1597–605. doi: 10.1245/s10434-009-0362-3. [DOI] [PubMed] [Google Scholar]
- 15.Cemal Y, Albornoz CR, Disa JJ, McCarthy CM, Mehrara BJ, Pusic AL, et al. A paradigm shift in U.S. breast reconstruction: Part 2. The influence of changing mastectomy patterns on reconstructive rate and method. Plast Reconstr Surg. 2013;131(3):320e–6e. doi: 10.1097/PRS.0b013e31827cf576. [DOI] [PubMed] [Google Scholar]
- 16.Kummerow KL, Du L, Penson DF, Shyr Y, Hooks MA. Nationwide trends in mastectomy for early-stage breast cancer. JAMA Surg. 2015;150(1):9–16. doi: 10.1001/jamasurg.2014.2895. [DOI] [PubMed] [Google Scholar]
- 17.Ballard-Barbash R, Taplin SH, Yankaskas BC, Ernster VL, Rosenberg RD, Carney PA, et al. Breast Cancer Surveillance Consortium: a national mammography screening and outcomes database. AJR Am J Roentgenol. 1997;169(4):1001–8. doi: 10.2214/ajr.169.4.9308451. [DOI] [PubMed] [Google Scholar]
- 18.Carney PA, Geller BM, Moffett H, Ganger M, Sewell M, Barlow WE, et al. Current medicolegal and confidentiality issues in large, multicenter research programs. Am J Epidemiol. 2000;152(4):371–8. doi: 10.1093/aje/152.4.371. [DOI] [PubMed] [Google Scholar]
- 19.http://www.nccn.org/professionals/physician_gls/pdf/breast_blocks.pdf (Accessed February 11, 2016).
- 20.Fisher B, Redmond C, Fisher ER. The contribution of recent NSABP clinical trials of primary breast cancer therapy to an understanding of tumor biology--an overview of findings. Cancer. 1980;46(4 Suppl):1009–25. doi: 10.1002/1097-0142(19800815)46:4+<1009::aid-cncr2820461326>3.0.co;2-h. [DOI] [PubMed] [Google Scholar]
- 21.Jatoi I, Proschan MA. Randomized trials of breast-conserving therapy versus mastectomy for primary breast cancer: a pooled analysis of updated results. Am J Clin Oncol. 2005;28(3):289–94. doi: 10.1097/01.coc.0000156922.58631.d7. [DOI] [PubMed] [Google Scholar]
- 22.Lippman ME. Selecting initial therapy for invasive breast cancer. Cancer. 1987;60(8 Suppl):2050–3. doi: 10.1002/1097-0142(19901015)60:8+<2050::aid-cncr2820601517>3.0.co;2-f. [DOI] [PubMed] [Google Scholar]
- 23.Veronesi U, Banfi A, Salvadori B, Luini A, Saccozzi R, Zucali R, et al. Breast conservation is the treatment of choice in small breast cancer: long-term results of a randomized trial. Eur J Cancer. 1990;26(6):668–70. doi: 10.1016/0277-5379(90)90113-8. [DOI] [PubMed] [Google Scholar]
- 24.Vaz-Luis I, Hughes ME, Cronin A, Rugo HS, Edge SB, Moy B, et al. Trends in the use of mastectomy in women with small node-negative breast cancer treated at US academic centers. Breast Cancer Res Treat. 2016;155(3):569–78. doi: 10.1007/s10549-016-3707-1. [DOI] [PubMed] [Google Scholar]
- 25.Lang JE, Summers DE, Cui H, Carey JN, Viscusi RK, Hurst CA, et al. Trends in post-mastectomy reconstruction: a SEER database analysis. J Surg Oncol. 2013;108(3):163–8. doi: 10.1002/jso.23365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.McLaughlin S, Mittendorf EA, Bleicher RJ, McCready DR, King TA. The 2013 Society of Surgical Oncology Susan G. Komen for the Cure Symposium: MRI in breast cancer: where are we now? Ann Surg Oncol. 2014;21(1):28–36. doi: 10.1245/s10434-013-3307-9. [DOI] [PubMed] [Google Scholar]
- 27.Miller BT, Abbott AM, Tuttle TM. The influence of preoperative MRI on breast cancer treatment. Ann Surg Oncol. 2012;19(2):536–40. doi: 10.1245/s10434-011-1932-8. [DOI] [PubMed] [Google Scholar]
- 28.Fancellu A, Turner RM, Dixon JM, Pinna A, Cottu P, Houssami N. Meta-analysis of the effect of preoperative breast MRI on the surgical management of ductal carcinoma in situ. Br J Surg. 2015;102(8):883–93. doi: 10.1002/bjs.9797. [DOI] [PubMed] [Google Scholar]
- 29.Chandwani S, George PA, Azu M, Bandera EV, Ambrosone CB, Rhoads GG, et al. Role of preoperative magnetic resonance imaging in the surgical management of early-stage breast cancer. Ann Surg Oncol. 2014;21(11):3473–80. doi: 10.1245/s10434-014-3748-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Pinell-White XA, Kolegraff K, Carlson GW. Predictors of Contralateral Prophylactic Mastectomy and the Impact on Breast Reconstruction. Annals of Plastic Surgery. 2014;72(6):S153–S7. doi: 10.1097/SAP.0000000000000099. [DOI] [PubMed] [Google Scholar]
- 31.Esserman L, Shieh Y, Thompson I. Rethinking screening for breast cancer and prostate cancer. JAMA. 2009;302(15):1685–92. doi: 10.1001/jama.2009.1498. [DOI] [PubMed] [Google Scholar]
- 32.Davis KL, Barth RJ, Jr, Gui J, Dann E, Eisenberg B, Rosenkranz K. Use of MRI in preoperative planning for women with newly diagnosed DCIS: risk or benefit? Ann Surg Oncol. 2012;19(10):3270–4. doi: 10.1245/s10434-012-2548-3. [DOI] [PMC free article] [PubMed] [Google Scholar]