Magnetic Resonance Imaging: Pitfalls in Its Application to Breast Disease

Advances in diagnostic imaging have led to general improvement in treating patients for various maladies. Recently, advances in breast diagnostic imaging with magnetic resonance imaging (MRI) in unselected patient populations have exposed a number of pitfalls that threaten well-established, data-proven medical practice. Recognition of these pitfalls should lead to clinical trials to resolve the more controversial issues and result in practice algorithms to support the use of breast MRI in targeted patient populations.

Diagnostic imaging with mammography has identified many more women with cancerous tumors smaller than 2 cm. Decrease in stage at presentation of these smaller tumors has led to a survival benefit beyond that expected by simple stage migration alone for patients with screen-detected cancer versus those whose cancers are detected by clinical examination.¹ Decreases in mortality resulting from mammographic screening are estimated to be 46% and vary by the age of the cohort examined.² Still, mammography misses about 20% of all breast cancer, particularly in young women with dense breast parenchyma.³

Breast MRI

Advances in breast imaging with MRI have heightened sensitivity in breast cancer detection, though at increased cost: the sensitivity gained is plagued by significant variation in specificity, with most studies showing it to be inferior to that of screening mammography.⁴ Comparison of the specificity and sensitivity of MRI versus mammography in breast cancer screening has been studied primarily in women at high risk for developing breast cancer, such as BRCA gene carriers. Even in the groups where young age compromises the reliability of mammography, MRI has resulted in additional recall examinations in 10.7% of women, compared with 3.9% for mammography.⁵ In turn, these additional exams have led to a 3-fold increase in recommendations for biopsy (3.1% vs 1.3%). The increased cost of MRI, with its attendant increase in repeat exams and biopsies in high-risk populations, is likely to forestall its application to the routine screening of all women, where the putative marginal advantage of the sensitivity of MRI over that of digital mammography in the 50-year-or-older group may be slim.

The role of MRI in the management of patients newly diagnosed with breast cancer represents the area of controversy where the accumulated evidence of the past 30 years provides significant insight into the pitfalls associated with routine MRI use. It is here where MRI interpretation has the most significant impact, and a possibly deleterious effect on the care of the patient. To become fully aware of these pitfalls, we must review some of the remarkable progress made over the past 40 years in our understanding of the history of breast cancer and its treatment.

Evolution of breast cancer management

In 2002, reports by Fisher⁶ and Veronesi⁷ affirmed the long-term validity of breast-conserving surgery (BCS) for the management of early-stage breast cancer (ESBC). More than a decade earlier (in 1991), a National Cancer Institute Consensus Conference had endorsed BCS—margin-clear lumpectomy with postoperative whole-breast irradiation—over mastectomy as the preferred treatment of ESBC. This recommendation has been strengthened by the accumulating biological and clinical evidence of the natural history of this disease in patients with a single breast cancer documented by mammographic evaluation. When patients with ESBC are treated with BCS or mastectomy, local recurrence rates are the same at 5% to 10%.⁸ As discussed previously, the success of BCS has come from the increased use of screening mammography and the concomitant early diagnosis of smaller tumors.

Traditionally, BCS has been contraindicated in patients with multicentric disease (tumors in other quadrants of the breast) as documented by mammography at the time of diagnosis. The increasing use of MRI in patients with a known breast cancer has led to the detection of multiple lesions in patients who heretofore had only a single primary tumor identified by physical examination or mammography. The increased sensitivity of MRI in this setting leads to the identification of potential second cancers in the same breast in 13% to 30% of patients.^9,10 Curiously, the presence of multifocal tumors in patients with even the smallest of breast cancers was noted by Holland more than 20 years ago in detailed serial section histologic examination of mastectomy specimens.¹¹ In fact, Holland’s data suggested that as many as two-thirds of these patients had a second lesion within the same breast, with most of them mapping within 4 cm of the known primary. Holland‘s data remain timely because they were used then as an admonition about the validity of BCS. The rediscovery of these lesions by MRI is now revisited as a reason for the conversion from BCS to mastectomy in patients with ESBC.¹² Paradoxically, an analysis similar to Holland’s recently reported by Japanese investigators supports the use of accelerated partial-breast irradiation (APBI), which spares the breast tissue outside this 4-cm radius from any radiation in BCS. In this setting, none of the multifocal lesions outside the quadrant in question would have been subjected to radiation therapy, and thus a higher local failure rate would have been expected.¹³ Favorable 4-year recurrence rates of 1% have been published for APBI, with longer follow-up continuing.¹⁴

In a more remarkable study, conducted by Sardanelli, MRI was performed in women who had elected to undergo mastectomy for a single known cancer.¹⁵ Subsequent pathologic step section of the entire breast specimen using 5-mm sections in the manner of Holland then attempted to localize the patient’s primary cancer as well as any other suspected cancers detected by the preoperative MRI. In 99 breast specimens, mammography and MRI missed 64 and 36 malignant tumors, 8 mm and 5 mm in median size, respectively. In fact, MRI failed to detect 19% of the malignant foci discovered by the pathologic evaluation of the breast. Although MRI was more sensitive than mammography, its predictive positive value was low (only 70%) for both modalities. Therefore, even if patients were to be selected for BCS by MRI, 19% would have a second unknown primary cancer in the breast. This figure is clearly much higher than any reported local recurrence rate in patients treated with BCS and appropriate systemic therapy. Not surprisingly, a recent report shows that BCS patients selected on the basis of unifocal disease by MRI have identical local recurrence rates (3%-4%) at 8-year follow-up when compared with a similarly treated cohort undergoing BCS based on unifocal disease established by routine mammography only.¹⁶

This compares favorably to the recent update by the Early Breast Cancer Trialists, which shows 10-year local failure rates of 8% for node-negative women undergoing a more deforming mastectomy and 27% for node-positive women similarly treated.¹⁷ Thus, with increasingly smaller cancers disclosed by effective mammographic screening, the added weight of 25 years of clinical experience, and the improvement in systemic therapies to date, it is difficult to see how the rediscovery of well-known subclinical tumors, which are well controlled by contemporary radiotherapy, should discourage physicians and patients alike from BCS. The resulting pitfalls in treatment recommendations resulting from indiscriminate use of breast MRI are shown by the Mayo Clinic report, in which preoperative MRI raised the odds of undergoing mastectomy by 60%, and in which having surgery in 2006 versus 2003 raised the odds by 70%.¹⁸ Bleicher just reported similar increases in mastectomy rates (19%-27%) in patients undergoing preoperative MRI.¹⁹ In Berg’s 2004 report, 12% of women chose an unneeded mastectomy without any biopsy to exclude malignancy of the questionable second ipsilateral lesion detected by MRI and which were determined at mastectomy not to be alignant.10 Thus, the adoption of MRI without any prospective data on its benefit in the management of ESBC undermines the long prospective and data-driven efficacy record of BCS compared with mastectomy. Indiscriminate use of breast MRI threatens to reverse breast conservation trends, to the detriment of women in the United States.

Examining contralateral prophylactic mastectomy

More alarming is the parallel increasing trend of contralateral prophylactic mastectomy (CPM)— removal of the opposite, unaffected breast. The introduction of MRI in breast diagnostics, coupled with inadequate risk assessment by clinicians and understandable apprehension on the part of patients, has led to unnecessary preventive CPM in patients with ESBC. Many breast cancer patients opt for CPM based on an MRI finding without histologic proof of cancer. Invariably, many who make this recommendation do not have the ability to perform MRI-guided core biopsy of lesions not detected by mammography or ultrasonography. The distressing report by Tuttle has shown an increase in CPM from 4.2% in 1998 to 11% in 2003.²⁰ In this report of 152,755 patients from the Surveillance, Epidemiology and End Results cancer registry, 8.3% were 60 or older at the time of CPM. Understandably, the figure was only 25% among women aged 18-49 years undergoing a unilateral mastectomy for the treatment of their breast cancer. There can be no question that this trend has been accelerated with the recent increased use of MRI, as Tuttle’s report also shows that the rate of CPM increased between 2000 and 2003.²⁰ It is of concern that the most common reason for CPM was a physician’s characterization of the risk of contralateral breast cancer, known since the days of the radical mastectomy to be negligible. This exemplifies how damaging informed consent can be if it is inadequate.

The risk of contralateral disease is well established as 3% at 5 years.²¹ Furthermore, any risk of contralateral disease is decreased by 50% in patients using tomoxifen and by 20% in those treated with chemotherapy.¹⁷ Clearly, prophylactic removal of the unaffected breast can have no impact on the survival of a patient in whom the stage of the index cancer is the primary determinant of survival. Remarkably, even in high-risk breast cancer patients undergoing CPM, MRI missed 3 of 4 lesions (of these, 3/4 were ductal carcinoma in situ) noted at final pathology and treated at great additional cost.²²

Take-home message

There is no question that increased use of MRI by physicians not familiar with all its limitations, coupled with inordinate fear in patients and inadequate assessment of risk, has led to difficult conversations with patients at the time of informed consent. Many breast cancer specialists find that all this information is hard to convey during a single visit, as it may require discussion about BCS, fear of radiation therapy, surveillance strategies, and prevention strategies, as well as accurate genetic risk assessment. The pitfall here is that it is easier to recommend bilateral mastectomy than to adhere to data-driven practice patterns established over the past 30 years.²³ Thus, the current indiscriminate use of MRI in the screening and management of breast cancer is associated with a significant negative impact to the breast cancer patient in the absence of any evidence that it improves surgical care or prognosis.²⁴

Edibaldo Silva, MD, is attending surgical oncologist, Division of Surgical Oncology, University of Nebraska Medical Center’s Eppley Cancer Center, Omaha, Nebraska. He is also the program leader of the Melanoma & Soft Tissue Sarcoma service and manages breast cancer patients at the Olson Breast Center of the University of Nebraska Medical Center.

REFERENCES

1. Shen Y, Yang Y, Inoue LYT, et al. Role of detection method in predicting breast cancer survival: analysis of randomized screening trials. J Natl Cancer Inst. 2005;97(16):1195- 1203. Abstract available at: http://tinyurl.com/ye4hh6p.

2. Berry DA, Cronin KA, Plevritis SK, et al. Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med. 2005;353(17):1784-1792. Abstract available at: http://tinyurl.com/yc8a3zx.

3. Pisano ED, Gatsonis C, Hendrick E, et al. Diagnostic performance of digital versus film mammography for breast cancer screening. N Engl J Med. 2005;353(17):1773-1783. Abstract available at: http://tinyurl.com/yc7qmuz.

4. Bleicher RJ, Morrow M. MRI and breast cancer: role in detection, diagnosis, and staging. Oncology (Williston Park). 2007;21(12):1521-1533. Abstract available at: http://tinyurl.com/yh5k45x.

5. Kriege M, Brekelmans CT, Boetes C, et al. Effi cacy of MRI and mammography for breast cancer screening in women with a familial or genetic predisposition. N Engl J Med. 2004;351(5):427-434. Abstract available at: http://tinyurl.com/ygerym3.

6. Fisher B, Anderson S, Bryant, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med. 2002;347(16):1233-1241. Abstract available at: http://tinyurl.com/yf5yzcx.

7. Veronesi U, Cascinelli N, Mariani L, et al. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med. 2002;347(16):1227-1232. Abstract available at: http://tinyurl.com/yc3xvbq.

8. Montgomery DA, Krupa K, Jack WJL, et al. Changing pattern of locoregional relapse in breast cancer: the Edinburgh experience. Br J Cancer. 2007;96(12):1802-1807. Abstract available at: http://tinyurl.com/yae52kw.

9. Bluemke DA, Gatsonis CA, Chen MH, et al. MRI of the breast prior to biopsy. 9. JAMA. 2004;292(22):2735-2742. Abstract available at: http://tinyurl.com/yaa9ey7.

10. Berg WA, Gutierrez L, NessAiver MS, et al. Diagnostic accuracy of mammography, clinical examination, US and MRI imaging for preoperative assessment of breast cancer. Radiology. 2004;233(3):830-849. Abstract available at: http://tinyurl.com/yalpu3q.

11. Holland R, Veling SH, Mravunac M, Hendriks JH. Histologic multifocality of Tis, T1-2 breast carcinomas. Implications for clinical trials of breast-conserving surgery. Cancer. 1985;56(5):979-990. Abstract available at: http://tinyurl.com/yac5jqb.

12. Bedrosian I, Mick R, Orel SG, et al. Changes in the surgical management of patients with breast carcinoma based on preoperative magnetic resonance imaging. Cancer. 2003;98(3):468-473. Abstract available at: http://tinyurl.com/yc28jfr.

13. Imamura H, Haga S, Shimizu T, et al. Relationship between the morphological and biological characteristics of intraductal components accompanying invasive ductal breast carcinoma and patient age. Breast Cancer Res Treat. 2000;62(3):177-184. Abstract available at: http://tinyurl.com/yc5jx96.

14. Beitsh P,Vicini F, Zannis V, et al. Recurrence and survival in the American Society of Breast Surgeons (ASBS) MammoSite® RTS Registry Trial. Int J Radiat Oncol Biol Phys. 2008;72(suppl 1):S2-S3. No abstract available online.

15. Sardanelli F, Giuseppetti GM, Panizza P, et al. Sensitivity of MRI versus mammography for detecting foci of multifocal, multicentric breast cancer in fatty and dense breasts using the whole-breast pathologic examination as the gold standard. AJR Am J Roentgenol. 2004;183(4):1149-1157. Abstract available at: http://tinyurl.com/yffrq4s.

16. Solin LJ, Orel SG, Hwang WT, et al. Relationship of breast magnetic resonance imaging to outcome after breast-conservation treatment with radiation for women with early-stage invasive breast carcinoma or ductal carcinoma in situ. J Clin Oncol. 2008;26(3):386-391. Abstract available at: http://tinyurl.com/yggrarj.

17. Clarke M, Collins R, Darby S, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005;366(9503):2087-2106. Abstract available at: http://tinyurl.com/ye3qhf8.

18. Katipamula R, Degnim AC, Hoskin T, et al. Trends in mastectomy rates at the Mayo Clinic Rochester: effect of surgical year and preoperative magnetic resonance imaging. J Clin Oncol. 9009;27(25):4082-4088. Abstract available at: http://tinyurl.com/yc5zz7n.

19. Bleicher RJ, Ciocca RM, Egleston BL, et al. Association of routine pretreatment magnetic resonance imaging with time to surgery, mastectomy rate and margin status. J Am Coll Surg. 2009;209(2):180-187. Abstract available at: http://tinyurl.com/yalnktc.

20. Tuttle T, Haberman E, Grund E, et al. Increasing use of contralateral prophylactic mastectomy for breast cancer patients: a trend toward more aggressive surgical treatment. J Clin Oncol. 2007;25(33):5203-5209. Abstract available at: http://tinyurl.com/ydvlxku.

21. Gao X, Fisher SG, Emami B. Risk of second primary cancer in the contralateral breast in women treated for early-stage breast cancer: a population-based study. Int J Radiat Oncol Biol Phys. 2003;56(4):1038-1045. Abstract available at: http://tinyurl.com/y9xk9mn.

22. Black D, Specht M, Lee J, et al. Detecting occult malignancy in prophylactic mastectomy: preoperative MRI versus sentinel lymph node biopsy. Ann Surg Oncol. 2007;14(9):2477- 2484. Abstract available at: http://tinyurl.com/yc59ebx.

23. Schwartz M. Contralateral prophylactic mastectomy: efficacy, satisfaction, regrets. 23. J Clin Oncol. 2005;23(31):7777-7779. No abstract available online.

24. 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. Abstract available at: http://tinyurl.com/yzn8t8t.