Fw: Imaging Breast Cancer: Specificity Still Eludes Investigators

[Guest guest]

From: ilena rose <ilena@...>

Sent: Monday, January 08, 2001 11:18 PM

Subject: Imaging Breast Cancer: Specificity Still Eludes Investigators

> Medscape posted 12/14/00

> http://www.medscape.com/medscape/cno/2000/RSNACME/Story.cfm?story_id=1845

>

> 86th Scientific Assembly and Annual Meeting of the Radiological Society of

> North America

>

> Imaging Breast Cancer: Specificity Still Eludes Investigators

>

> G. Bradley Jr, MD, PhD, FACR

>

> In a refresher course that focused on breast magnetic resonance imaging

> (MRI) held during the 86th Scientific Assembly and Annual Meeting of the

> RSNA, 3 experts addressed the topics of breast implants, breast cancer,

and

> equivocal mammograms.[1]

>

> Detecting Breast Cancer in Patients With Breast Implants

>

> Wendie A. Berg, MD, PhD, University of land, reviewed the types of

> breast implants (saline vs silicone, single lumen vs double lumen, and

> subpectoral vs subglandular), as well as the " linguini sign " of an

> intracapsular rupture. She noted that the inwardly collapsed shell

> producing " linguini " should consist of several redundant lines that must

be

> distinguished from radial folds, which are quite common. While droplets

can

> be seen in a silicone implant, this can either be a sign of failure of an

> outer saline lumen or it could be due to betadine, which is commonly

> injected into the implant at the time of surgery to adjust the size. Dr.

> Berg demonstrated several examples of intracapsular rupture, showing the

> shell or envelope pulling away from the fibrous capsule, which is formed

by

> the body around the implant. She demonstrated the Becker expandable

> implant-prosthesis, which is used in the setting of breast cancer where

the

> skin over the chest wall needs to be slowly stretched and expanded to

> accommodate the implant.

>

> Dr. Berg noted that mammograms are limited in their detection of breast

> cancer when implants are present. Specifically, visualization of the

breast

> is decreased by 30% when an implant is present (50% if a contracture is

> present). Even with " pushback views, " on average, 25% of the breast is

> still obscured by mammography. This number can be subdivided based on the

> location of the implants -- for submammary implants, 35% of the breast is

> still obscured on the pushback views and for subpectoral implants, 18% of

> the breast is obscured.[2] It should be noted that these percentages refer

> to the percentage of the breast visualized by mammography in the setting

of

> implants. When evaluating for visible breast cancer, only 66% of all

> carcinomas are visible with implants, increasing to 72% with pushback

views.

>

> One of the problems with extracapsular rupture is that silicone granulomas

> can calcify and mimic the calcification associated with certain

carcinomas.

> Thus, evaluation of the breast for carcinoma in the setting of implants is

> much better performed with magnetic resonance imaging (MRI) than with

> conventional mammography. Although there was interest a few years ago in

> Tc-99m sestamibi for the detection of breast cancer, this has largely

> abated because of its low sensitivity (72%).

>

> To summarize, Dr. Berg quoted Dr. Huch[3] in pronouncing MRI to be the

> " modality of choice " for the detection of breast cancer, however, she also

> noted that the relative expense compared to mammography would probably

> limit MR's general use.

>

> Use of MRI in the Evaluation of Breast Cancer

>

> , MD, Memorial Sloan-Kettering Cancer Center, New York,

NY,

> noted that MRI for breast cancer should be performed in a dedicated coil

> using intravenous gadolinium in scanning the entire breast with contiguous

> 2- to 3-mm (thick) slices. To improve visualization, fat can then be

> removed by either spectrographic fat suppression or subtraction. On her

> General Electric unit, she acquires a precontrast three-dimensional (3-D)

> T1-weighted gradient echo of the breast followed by 3 postcontrast

> identical acquisitions after injection of 0.1 mmol/kg of gadolinium

(single

> dose).

>

> For the preoperative evaluation of breast cancer, MRI was noted to be

> superior to mammography, particularly, for the extent of lobular

carcinoma.

> Specifically, MRI was able to demonstrate 85% of the extent of lobular

> carcinoma (vs pathology) compared with mammography, which only

demonstrated

> the extent of lobular carcinoma correctly 32% of the time.[4] The extent

of

> lobular carcinoma is much better demonstrated by MRI than either palpation

> or mammography. Even ductal carcinoma in situ (DCIS) can be demonstrated

by

> MRI due to tumor angiogenesis factor extending beyond the ducts. Although

> others have reported detecting a high percentage of DCIS (66% to 85%), it

> was noted that DCIS is still less well visualized than invasive

carcinomas.

> Dr. also commented that the extent of the disease is more visible

> through MRI than mammography and thus can modify treatment. To be

specific,

> if mammography detects disease in 1 quadrant of the breast, a simple

> lumpectomy can be performed; however, if multifocal or multicentric

disease

> involving several quadrants is seen, the patient must have a

mastectomy.[5]

> MR is of value for preoperative planning for a single-stage resection. MR

> is particularly useful for posterior lesions, which are hard to see by

> mammography. It is also excellent for chest wall invasion, as extension of

> breast cancer between the ribs needs to be resected.

>

> Although MR is exceptionally sensitive for the detection of breast cancer,

> the lack of specificity may lead to problems. For example, if disease in

> noncontiguous quadrants is discovered by MRI, but not verified by needle

> localization or core biopsy, the patient may end up getting a mastectomy

> rather than a lumpectomy. Thus, the increased sensitivity of MRI may deny

> some patients more conservative therapy.

>

> Dr. also noted that resection of an invasive carcinoma in 1

location

> of the breast may cause DCIS in another part of the breast to regress

> without any treatment. Therefore, the finding of an infiltrating index

> lesion and DCIS in a noncontiguous quadrant need not necessarily result in

> mastectomy.

>

> Evaluating the contralateral breast for infiltrating lobular carcinoma is

> critical because carcinoma may be found in 4% of cases incidentally on the

> contralateral side. These lesions do need to be biopsied, however, because

> benign proliferative changes can enhance in a similar fashion to

carcinoma.

> At this time, needle localization by MRI with subsequent open biopsy is

> generally preferred to core biopsy, because the MR compatible core biopsy

> instruments are less mechanically robust than those that are not MR

> compatible.

>

> In the postoperative breast, MRI is definitely superior to mammography for

> detection of residual masses. It is normal to see a rim of enhancement

> around the operative site by MRI. Dr. noted that in some cases, the

> residual carcinoma could only be seen with MRI. Recurrent disease is

always

> invasive, but not DCIS.[6] Although MRI is excellent in detecting

recurrent

> tumor and distinguishing it from fibrosis, the next step (biopsy vs MRI)

is

> usually determined by cost. Dr. also noted that MRI was excellent

> for finding occult primary carcinoma in the setting of positive axillary

> adenopathy.

>

> MRI is useful to follow breast cancer being treated by chemotherapy.

> Specifically, if there is no decrease in the size of the lesion after 3

> cycles of chemotherapy, the prognosis is not good. Thus, regular follow-up

> with MRI will lead to earlier mastectomy and treatment of lesions that do

> not respond to specific chemotherapeutic regimens. Dr. noted that

> MRI was excellent for detecting recurrence following chemotherapy.

>

> In general, MRI is excellent for difficult cases such as local recurrence

> in a transverse rectus abdominis myocutaneous (TRAM) flap, for lesions

> behind implants, and in breasts that have had silicone injections.

>

> The question of whether MRI should be used for screening of high-risk

> patients was raised. Specifically, these are patients with the BRCA 1 and

2

> genes that have an 85% and 70% lifetime risk of developing breast cancer

> respectively. They also have a higher incidence of bilateral disease. In a

> study by Kuhl and colleagues,[7] 9 carcinomas were found in 192 carriers

of

> the BRCA 1 or 2 genes. Six of these 192 patients with cancer only had

> lesions documented by MRI -- not by other methods (eg, mammography or

> ultrasound). Whether there is a role for MRI in this high-risk population

> remains to be seen.

>

> MR-guided breast biopsy is best performed by needle localization. This is

> proven technology compared with the MR compatible coring devices, which

are

> not as mechanically robust. The importance of performing a needle

> localization is to prove that the tissue harvested is the same as the

> abnormality seen on MRI. Needle biopsies performed following

administration

> of gadolinium must be performed quickly (ie, within 15 minutes following

> administration of gadolinium, before it washes out of the lesion). Dr.

> emphasized the need for better MR-compatible biopsy systems, better

> standardization of techniques, and new clinical trials to demonstrate the

> efficacy of MRI in the management of breast cancer.

>

> MRI and the Equivocal Mammogram

>

> Carol Lee, MD, Yale University, New Haven, Connecticut addressed the use

of

> MRI in the evaluation of questionable mammography results. She performs

> mammography initially followed by ultrasound, and only uses MRI for

> equivocal lesions. A common indication for MRI is when a lesion can only

be

> seen on the medial-lateral oblique (MLO) view, but not the cranial-caudad

> (CC) view, a situation that will not allow biopsy. In some cases of

> questionable architectural distortion, MRI is also useful. She pointed out

> that scar enhances up to 6 months following a biopsy and up to 18 months

> following radiation therapy. Enhancement at the site of a previously

benign

> biopsy should still be addressed as tumor can be found where a benign

> lesion was previously noted.

>

> Dr. Lee noted that there are 4 clinical settings in which MRI may be

useful

> in the evaluation of the equivocal mammogram and cited the incidence of

> malignancy found in each of these situations:

>

> Is there an abnormality? Yes, malignancy was found in 7% of cases. Where

> is the lesion? This resulted in 13% detection of malignancy. Is this a

> recurring lesion? Yes, in 7% of cases malignancy was detected. Scar vs

> tumor at the site of a previously benign biopsy? This scenario resulted in

> a 44% malignancy detection rate.

>

> Dr. Lee noted that diffuse parenchymal enhancement is found in 55% of

> breasts, but tends to be lowest during days 7 to 20 of the menstrual

cycle.

> The sensitivity of MRI overall is 80% to 100% for the detection of breast

> cancer. This represents a combination of 70% to 80% for DCIS and close to

> 100% for invasive carcinoma. Unfortunately, specificity is only 37% to

85%,

> depending on the specific series. The primary reasons for false-negative

> results are lesions that are slowly growing, have little tumor

> angiogenesis, or are in situ. Incidental lesions are found 30% of cases

> when MRI is performed for an equivocal mammogram. One of 30 of these

> lesions is malignant. Incidental lesions tend to be more common in younger

> premenopausal woman, and are also more common in dense breasts.

>

> Highly suspicious lesions seen on mammography should be biopsied to

confirm

> or exclude malignancy with a high index of confidence, and lesions that

> appear to be benign should be assessed by follow-up mammography. In cases

> in which mammography is equivocal, MRI is often useful in determining the

> optimal area to biopsy, which can be performed under MRI or ultrasound

> guidance. In instances where the lesion cannot be seen to be biopsied on

> ultrasound or if the stage of the patient's menstrual cycle prevents

> adequate visualization of the mass on MRI because of diffuse parenchymal

> enchancement, the clinician should repeat the MRI study at a different

> stage of the menstrual cycle, followed by MRI-guided biopsy. Dr. Lee

> employs a fast spoiled gradient-recalled echo (SPGR) technique (TR 19/TE

> 1.8/Flip 45 degrees) with a 14-cm field of view and a slice thickness of

> 3.4 mm. Dr. Berg concluded her presentation by noting that MRI is a

> wonderful problem-solving tool that should not be used ubiquitously or to

> replace mammography and ultrasound.

>

> Implications for Clinical Practice

>

> Gadolinium-enhanced MRI is an extremely sensitive technique for the

> detection of breast cancer, but should not be used to replace mammography

> or ultrasound. Its use should be reserved for problem solving -- ie,

> patients with implants, instances in which a mammogram is equivocal, and

> patients who have had previous biopsies or partial resection. The primary

> reason not to perform MRI of the breast in every patient is cost; however,

> MRI may also overdiagnose benign lesions, and it can be misleading at

> certain times during the menstrual cycle because of diffuse parenchymal

> enhancement. MRI should be used to follow patients with known breast

> cancer who have been treated with either surgery, radiation, or

> chemotherapy. The use of MRI to follow-up high-risk patients (ie, those

> with the BRCA 1 or 2 genes) may ultimately be indicated, however, there

are

> not sufficient data to support this use.

>

> References

>

> 1.Berg W, EA, Lee CH. Breast MR. Presented at the 86th

Scientific

> Assembly and Annual Meeting of the Radiological Society of North America;

> November 27, 2000;Chicago, Illinois. Refresher course

>

> 2.Handel N, Silverstein MJ, Gamagami P, Jensen JA, A. Factors

> affecting mammographic visualization of the breast after augmentation

> mammoplasty. JAMA. 1992;268:1913-1917.

>

> 3.Huch RA, Kunzi W, Debatin JF, Wiesner W, Krestin GP. MR imaging of the

> augmented breast. Eur Radiol. 1998;8:371-376.

>

> 4.Rodenko GN, Harms SE, Pruneda JM, et al. MR imaging in the management

> before surgery of lobular carcinoma of the breast: correlation with

> pathology. AJR Am J Roentgenol. 1996;167:1415-1419.

>

> 5.Esserman L, Hylton N, Yassa L, Barclay J, el S, Sickles E. Utility

> of magnetic resonance imaging in the management of breast cancer: evidence

> for improved preoperative staging. J Clin Oncol. 1999;17:110-119.

>

> 6.Orel SG, Troupin RH, EA, Fowble BL. Breast cancer recurrence

> after lumpectomy and irradiation: role of mammography in detection.

> Radiology. 1992;183:201-206.

>

> 7.Kuhl CK, Schmutzler RK, Leutner CC, et al. Breast MR imaging screening

in

> 192 women proved or suspected to be carriers of a breast cancer

> susceptibility gene: preliminary results. Radiology. 2000;215:267-279.

>

> Return To Conference Summaries for 86th Scientific Assembly and Annual

> Meeting of the Radiological Society of North America

>

>

>