Application of a quantitative model to differentiate benign from malignant breast lesions detected by dynamic, gadolinium-enhanced MRI

Magnetic resonance imaging of breast cancer: clinical indications and breast MRI reporting system

Magnetic resonance imaging (MRI) is well suited to the investigation of breast cancer by virtue of its noninvasive nature and its multiplanar imaging abilities. MRI investigations showed high sensitivity but modest specificity for breast cancer detection and diagnosis. Most early studies tested the ability of MRI to evaluate and diagnose findings in the breast discovered by other imaging tests or by breast physical examination (1-4). When it was discovered that MRI identified small breast cancers undetected by mammography or breast ultrasound, MRI was used to estimate breast cancer extent in known cancer cases for surgical planning (5,6). These investigations led to the use of MRI in a multitude of breast imaging applications, raising further questions about the use of MRI in everyday practice: What are the indications for breast MRI in general practice? What is its role in light of other imaging tests? What are its benefits and limitations in each setting? How do I report these studies? The purpose of this article is to review the clinical background regarding indications for the use of MRI and relevant cases in which MRI can impact patient management in breast disease, and to describe new developments in reporting breast MRI studies. J. Magn. Reson. Imaging 2000;12:975-983.

Magnetic resonance imaging screening in women at genetic risk of breast cancer: imaging and analysis protocol for the UK multicentre study. UK MRI Breast Screening Study Advisory Group

The imaging and analysis protocol of the UK multicentre study of magnetic resonance imaging (MRI) as a method of screening for breast cancer in women at genetic risk is described. The study will compare the sensitivity and specificity of contrast-enhanced MRI with two-view x-ray mammography. Approximately 500 women below the age of 50 at high genetic risk of breast cancer will be recruited per year for three years, with annual MRI and x-ray mammography continuing for up to 5 years. A symptomatic cohort will be measured in the first year to ensure consistent reporting between centres. The MRI examination comprises a high-sensitivity three-dimensional contrast-enhanced assessment, followed by a high-specificity contrast-enhanced study in equivocal cases. Multiparametric analysis will encompass morphological assessment, the kinetics of contrast agent uptake and determination of quantitative pharmacokinetic parameters. Retrospective analysis will identify the most specific indicators of malignancy. Sensitivity and specificity, together with diagnostic performance, diagnostic impact and therapeutic impact will be assessed with reference to pathology, follow-up and changes in diagnostic certainty and therapeutic decisions. Mammography, lesion localisation, pathology and cytology will be performed in accordance with the UK NHS Breast Screening Programme quality assurance standards. Similar standards of quality assurance will be applied for MR measurements and evaluation.

Contrast-enhanced magnetic resonance imaging of the breast

Contrast-enhanced magnetic resonance (MR) imaging is increasingly used as a complementary diagnostic modality in breast imaging. The sensitivity of MR imaging of the breast for malignancy has consistently been reported to be excellent. The specificity has been rather variable. Study methods and imaging techniques are not standardized and there is still a great deal of uncertainty about MR imaging's place in clinical practice. Nevertheless, radiologists should be familiar with the current technique and the varying MR appearance of breast tumors to improve the accuracy of this method. This paper reviews the techniques for breast MR imaging, the pathopysiologic basis of contrast enhancement in breast tumors, and the current knowledge about detection and differentiation of breast tumors. In addition, future directions for breast MR imaging are discussed.

Dynamic contrast-enhanced MRI in the differentiation of breast tumors: user-defined versus semi-automated region-of-interest analysis

Dynamic contrast-enhanced MR mammography is an increasingly used method of evaluating breast pathology. The purpose of this study was to compare two semi-automated methods of region of interest (ROI) analysis with a user-defined method, in the discrimination of breast tumors using dynamic contrast-enhanced MRI. Results are presented from the retrospective analysis of 81 malignant and 36 benign breast lesions. The study demonstrates the importance of a consistent ROI strategy and also shows that semi-automated approaches offer a standardized method, which may improve the discrimination of primary breast tumors. J. Magn. Reson. Imaging 1999; 10:945-949.

Accurate and rapid quantitative dynamic contrast-enhanced breast MR imaging using spoiled gradient-recalled echoes and bookend T(1) measurements

A method is presented that converts dynamic T(1)-weighted spoiled gradient-recalled echo (SGRE) image intensities into estimates of T(1) without the errors associated with imperfections in the slice profile and transmitter coil magnetic field (B(1)). The method involves T(1) measurements performed before and after a series of dynamic SGRE images. These measurements serve to calibrate and correct the SGRE signal strength equation used to estimate T(1). Simulations and phantom experiments were performed to test the method for slice-selective (two-dimensional) and slab-selective (three-dimensional) imaging, as well as for imaging performed with optimized and un-optimized B(1). For nearly all test conditions, T(1) was estimated accurately (within 10%) over a range of T(1) values expected in vivo ( approximately 1200 --> 300 msec). This method should be useful for quantifying dynamic SGRE imaging for many different applications including breast MR imaging. Magn Reson Med 42:746-753, 1999.

Cost-effectiveness of MR imaging and core-needle biopsy in the preoperative work-up of suspicious breast lesions

PURPOSE

To assess the clinical and economic consequences of the use of preoperative breast magnetic resonance (MR) imaging and core-needle biopsy (CNB) to avert excisional biopsy (EXB).

MATERIALS AND METHODS

A decision-analytic Markov model was constructed to compare MR imaging, CNB, and EXB without preoperative testing in a woman with a suspicious breast lesion. Stage-specific cancer prevalence, tumor recurrence, progression rates, and MR imaging and CNB sensitivity and specificity were obtained from the literature. Cost estimates were obtained from the literature and from the Medicare fee schedule.

RESULTS

EXB without preoperative testing was associated with the greatest quality-adjusted life expectancy, followed by MR imaging and CNB; life expectancies were 17.409, 17.405, and 17.398 years, respectively. EXB resulted in the greatest lifetime treatment cost ($31,438), followed by MR imaging ($29,072) and CNB ($28,573). Results were robust over a wide range of cancer prevalence, stage distribution, tumor progression rates, and procedure and treatment costs. Incremental cost-effectiveness ratios showed that preoperative testing was cost-effective, but the choice between MR imaging and CNB was highly dependent on the accuracy of each test and to patient preferences.

CONCLUSION

Preoperative testing of most suspicious breast lesions was cost-effective. More precise estimates of MR imaging and CNB test performance characteristics are needed. Until those are available, patient preferences should inform individual decisions regarding preoperative testing.

Heterogeneity analysis of Gd-DTPA uptake: improvement in breast lesion differentiation

PURPOSE

The aim of this study was to assess whether lesion heterogeneity through pixel-by-pixel analysis of contrast-enhanced MR parameter maps would improve breast lesion differentiation.

METHOD

Forty-nine female patients with primary tumors of the breast were imaged using contrast-enhanced MRI. The rate of transfer of contrast medium between the lesion and plasma was calculated via compartmental modeling over semiautomatically delineated regions of interest. The distributions of transfer rate values were subdivided into 10 segments for each lesion. The rate of change of the statistical moments of the transfer rate measured over the 10 segments allowed a quantitative measure of heterogeneity.

RESULTS

In this patient group of 25 malignant and 25 benign lesions, improved differentiation was achieved by using a measure of lesion heterogeneity. The sensitivity and specificity obtained were 88 and 88%, respectively, compared with 84 and 76% when using the mean value of the transfer rate coefficient.

CONCLUSION

Pixel-by-pixel analysis of contrast medium transfer rate maps allows the examination of lesion heterogeneity and offers improved lesion differentiation. The new method used to quantify lesion heterogeneity relies on the manner in which the exchange rate parameter varies when calculated over different segments of the distribution.

Accuracy of MR imaging in the work-up of suspicious breast lesions: a diagnostic meta-analysis

RATIONALE AND OBJECTIVES

The authors performed a systematic, critical review of the literature on magnetic resonance (MR) imaging for primary breast cancer detection in patients with suspicious breast lesions, analyzed MR test performance in the articles meeting study criteria, and used this information to examine the cost-effectiveness of preoperative MR imaging.

MATERIALS AND METHODS

A structured, predefined MEDLINE search was conducted to identify potentially relevant, peer-reviewed, English-language references from January 1996 through August 1997 on the diagnostic accuracy of breast MR imaging. This information was supplemented by manually searching bibliographies of the retrieved articles for additional potentially relevant references. All studies were independently abstracted by two reviewers using a prospectively designed worksheet. Abstraction results were analyzed with the summary receiver operating characteristic (ROC) method.

RESULTS

Of 41 identified studies, 16 met the inclusion criteria. These studies reported sensitivities ranging from 63% to 100% and specificities ranging from 21% to 100%. Maximum joint sensitivity and specificity of the summary ROC curve was 89% (95% confidence interval [CI]: 82%, 93%). At a sensitivity of 95%, specificity was 67%. When test performance values were applied to a previous cost-effectiveness analysis, the cost-effectiveness of preoperative MR imaging relative to that of excisional biopsy was confirmed, but its cost-effectiveness relative to that of needle core biopsy varied widely.

CONCLUSION

For MR imaging to be a cost-effective alternative to excisional biopsy for diagnosis of suspicious breast lesions, its diagnostic test performance must be equal to or better than the best results in recently published studies.

Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions?

PURPOSE

To assess the relevance of the signal intensity time course for the differential diagnosis of enhancing lesions in dynamic magnetic resonance (MR) imaging of the breast.

MATERIALS AND METHODS

Two hundred sixty-six breast lesions were examined with a two-dimensional dynamic MR imaging series and subtraction postprocessing. Time-signal intensity curves of the lesions were obtained and classified according to their shapes as type I, which was steady enhancement; type II, plateau of signal intensity; or type III, washout of signal intensity. Enhancement rates and curve types of benign and malignant lesions were compared.

RESULTS

There were 101 malignant and 165 benign lesions. The distribution of curve types for breast cancers was type I, 8.9%; type II, 33.6%; and type III, 57.4%. The distribution of curve types for benign lesions was type I, 83.0%; type II, 11.5%; and type III, 5.5%. The distributions proved significantly different (chi 2 = 139.6; P < .001). The diagnostic indices for signal intensity time course were sensitivity, 91%; specificity, 83%; and diagnostic accuracy, 86%. The diagnostic indices for the enhancement rate were sensitivity, 91%; specificity, 37%; and diagnostic accuracy, 58%.

CONCLUSION

The shape of the time-signal intensity curve is an important criterion in differentiating benign and malignant enhancing lesions in dynamic breast MR imaging. A type III time course is a strong indicator of malignancy and is independent of other criteria.

Correlation of lesion appearance and histologic findings for the nodes of a breast MR imaging interpretation model

An interpretation model for evaluating magnetic resonance (MR) images of the breast was constructed that allowed differentiation of benign from malignant palpable or mammographically visible abnormalities. Architectural features define each node of the model. Investigation was subsequently made of the histologic findings in individuals within each node and of the frequency with which each histologic finding manifested as a particular architectural feature to determine whether nodal location and specific histologic findings are mutually predictive. The strongest associations were found between fibrocystic change and smooth masses, fibroadenoma and lobulated masses with nonenhancing internal septations, invasive ductal carcinoma (with or without ductal carcinoma in situ [DCIS]) and enhancing irregular or spiculated masses, invasive tubular carcinoma or radial scar and spiculated masses, medullary or colloid carcinoma and enhancing lobulated masses, invasive lobular carcinoma and the absence of a focal mass, DCIS and ductal enhancement, and DCIS (with or without invasive ductal carcinoma) and regional enhancement. Nodal location and histologic findings proved to be mutually predictive within the model; that is, the nodal location of MR imaging features within the model can be used to predict histologic findings and vice versa.

Intermediate outcomes: diagnostic and therapeutic impact

The conceptual basis for intermediate outcomes has been established, and scientifically sound methods are available for assessment of these short-term outcomes. While modest intermediate outcomes research experience assessing imaging technology has accumulated in the literature, much more remains to be done, for the benefit of both patients and physicians. I would recommend that intermediate outcomes research become more focused on assessment earlier in the diffusion phase of imaging technology development. These studies should be based on traditional diagnostic accuracy studies, but these should be extended to include well-designed, high scientific quality treatment impact studies. Furthermore, such studies need to be multidisciplinary and multiinstitutional/multisite and include patient utility assessment components.

Breast fibroadenoma: mapping of pathophysiologic features with three-time-point, contrast-enhanced MR imaging--pilot study

The capability of three-time-point, gadolinium-enhanced magnetic resonance imaging to depict vascular permeability and extracellular volume fraction of breast fibroadenoma was evaluated with histopathologic correlation. This method demonstrated an even distribution of high extracellular volume fraction and low to moderate microvascular permeability in these common breast lesions, providing a nonsurgical means of improving the accuracy of diagnosis of fibroadenoma.

Temporal sampling requirements for the tracer kinetics modeling of breast disease

The physiological parameters measured in the tracer kinetics modeling of data from a dynamic contrast-enhanced magnetic resonance (MR) breast exam (blood flow-extraction fraction product [FE], volume of the extracellular extravascular space [Ve], and blood volume [Vb]) may enable non-invasive diagnosis of breast cancer. One of the factors that compromises the accuracy and precision of the parameter estimates, and therefore their diagnostic potential, is the temporal resolution of the MR scans used to measure contrast agent (gadolinium-diethylenetriamine pentaacetic acid [Gd-DTPA]) concentration in an artery (arterial input function [AIF]) and in the tissue (tissue residue function [TRF]). Using computer simulations, we have examined, for several AIF widths, the errors introduced into estimates of tracer kinetic parameters in breast tissue due to insufficient temporal sampling. Temporal sampling errors can be viewed as uncertainties and biases in the parameter estimates introduced by the uncertainty in the relative alignments of the AIF, TRF, and sampling grid. These effects arise from the model's inherent sensitivity to error in either the AIF or TRF, which is dependent on the values of the tracer kinetic parameters and increases with AIF width. Based on the results of the simulations, to ensure that the error in FE and Ve will be under 10% of their true values, we recommend a rapid bolus injection of contrast agent (approximately 10 s), that the AIF be sampled every second, and that the TRF be sampled every 16 s or less. An accurate measurement of Vb requires that the TRF be sampled at least every 4 s. The results of these investigations can be used to set minimum dynamic imaging rates for tracer kinetics modeling of the breast.

Quantification of the extraction fraction for gadopentetate across breast cancer capillaries

To quantify the extraction fraction, E, for gadopentetate across tumor capillaries, R3230 adenocarcinomas were implanted in the mammary fat pads of seven rats. The value of E was determined by using a two-compartment tissue model in which the endothelial transfer coefficient, K(PS) (ml x min(-1) x cc(-1) of tissue), was estimated from the model fitted to changes in R1 relaxation time (deltaR1; s(-1)) measured by dynamic three-dimensional spoiled gradient recalled magnetic resonance imaging after injection of 0.1 mmol x kg(-1) of gadopentetate dimeglumine. The plasma flow rate through the tumor capillaries, Fp, (ml x min(-1) x g(-1) of tissue), was independently measured with fluorescent microspheres. E could be calculated by the relationship, E = K(PS)/Fp. The mean E for gadopentetate in the R3230 tumor was 0.197 +/- 0.118 with a range of 0.123-0.454. The relatively small mean value of E for gadopentetate allows a fair approximation of the permeability surface area product by K(PS) in this R3230 tumor model.

Correlation of dynamic contrast-enhanced MR imaging with histologic tumor grade: comparison of macromolecular and small-molecular contrast media

OBJECTIVE

The endothelial integrity of microvessels is disrupted in malignant tumors. Quantitative assays of tumor microvascular characteristics based on dynamic MR imaging were correlated with histopathologic grade in mammary soft-tissue tumors.

MATERIALS AND METHODS

A spectrum of tumors, benign through highly malignant, was induced in 33 female rats by administration of N-ethyl-N-nitrosourea, a potent carcinogen. Dynamic contrast-enhanced MR imaging was performed using a small-molecular contrast medium (gadopentetate, molecular weight = 0.5 kDa) and a macromolecular contrast medium (albumin-(Gd-DTPA)30, molecular weight = 92 kDa) at an interval of 1-2 days. Permeability surface area product (PS), as estimated by the corresponding endothelial transfer coefficient (K(PS)), and fractional plasma volume (fPV) were calculated for each tumor and each contrast agent using a two-compartment bidirectional kinetic model. MR imaging microvascular characteristics were correlated with histopathologic tumor grade.

RESULTS

Tumor permeability to macromolecular contrast medium, characterized by K(PS), showed a highly positive correlation with tumor grade (r2 = .76, p < 10(-10)). K(PS) values were zero for all benign and some low-grade carcinomas, greater than zero in all other carcinomas, and increased in magnitude with higher tumor grade. A considerably smaller but significantly positive correlation was found between fPV and tumor grade using macromolecular contrast medium (r2 = .25, p < .003). No correlation between K(PS) or fPV values and tumor grade was found using gadopentetate (r2 = .01, p > .95 and r2 = .03, p > .15, respectively).

CONCLUSION

Quantitative tumor microvascular permeability assays generated with macromolecular MR imaging contrast medium correlate closely with histologic tumor grade. No significant correlation is found using small-molecular gadopentetate.

Intraoperative dynamic MRI: localization of sites of brain tumor recurrence after high-dose radiotherapy

In patients with malignant astrocytomas or metastatic brain disease treated with high-dose radiotherapy, conventional imaging methods may not adequately distinguish recurrent tumor from radiation change. We used a fast spoiled gradient refocusing technique in the open-configuration intraoperative MR system to assess the rate of regional enhancement of the treated tumor bed and to localize specific sites for pathologic sampling to determine whether gadolinium uptake correlated with histologic data. Twenty-four patients were studied. Fourteen of 15 patients with areas of early enhancement had recurrent tumor present in histologic samples, and 8 of the remaining 9 patients had only reactive changes. Dynamic MRI was predictive of recurrent tumor (P < .0005, Fisher exact test and P < .002, Student t test). We conclude that dynamic MRI in the open-bore magnet is a promising method for localizing potential sites of active tumor growth in patients treated for malignant astrocytomas and metastatic brain lesions.

Multifeature analysis of Gd-enhanced MR images of breast lesions

The objective of this study was to determine whether linear discriminant analysis of different independent features of MR images of breast lesions can increase the sensitivity and specificity of this technique. For MR images of 23 benign and 20 malignant breast lesions, three independent classes of features, including characteristics of Gd-DTPA-uptake curve, boundary, and texture were evaluated. The three classes included five, four and eight features each, respectively. Discriminant analysis was applied both within and across the three classes, to find the best combination of features yielding the highest classification accuracy. The highest specificity and sensitivity of the different classes considered independently were as follows: Gd-uptake curves, 83% and 70%; boundary features, 86% and 70%; and texture, 70% and 75%, respectively. A combination of one feature each from the first two classes and age yielded a specificity of 79% and sensitivity of 90%, whereas highest figures of 93% and 95%, respectively, were obtained when a total of 10 features were combined across different classes. Statistical analysis of different independent classes of features in MR images of breast lesions can improve the classification accuracy of this technique significantly.

Mapping pathophysiological features of breast tumors by MRI at high spatial resolution

Magnetic resonance imaging (MRI) is a noninvasive method that reveals anatomical details in vivo and detects lesions for diagnosis. Although standard breast MRI cannot clearly delineate breast cancer, contrast-enhanced MRI enables the detection of breast masses with high sensitivity. Dynamic studies demonstrated that malignant lesions were characterized by a faster signal enhancement rate than benign ones. Dynamic MRI of human breast cancer in mice revealed high heterogeneity in the distribution of contrast-enhanced curves and derived pathophysiological features, indicating the importance of high spatial resolution. With clinical MRI, it is difficult to achieve simultaneously high spatial and temporal resolution. In previous dynamic studies, the emphasis was on high temporal resolution and mainly empiric analyses. We describe here a new model-based method that optimizes spatial resolution by using only three time points, and yet characterizes tumor heterogeneity in terms of microvascular permeability and extracellular fraction. Mapping these pathophysiological features may aid diagnosis and prognosis assessment, while the high spatial resolution may improve the capacity to detect smaller lesions. The method was tested in human breast tumors implanted in mice and in a limited number of benign and malignant breast lesions of patients.