Use of novel artificial intelligence computer-assisted detection (AI-CAD) for screening mammography: an analysis of 17,884 consecutive two-view full-field digital mammography screening exams

BACKGROUND

Novel artificial intelligence computer-assisted detection (AI-CAD) systems based on deep learning (DL) promise to support screen reading.

PURPOSE

To test a DL-AI-CAD system compared to human reading on consecutive screening mammograms.

MATERIAL AND METHODS

In this retrospective study, 17,884 consecutive anonymized screening mammograms, double-read from January to November 2018, were processed by the DL-AI-CAD system. AI-CAD reading was considered positive if the AI-CAD case scores exceeded 30 (range = 1-100) and the lesion was correctly marked. Likewise, human reading (R1 or R2, respectively) was considered positive if the lesion was correctly identified and called. Receiver operating characteristic (ROC) analysis was performed and accuracy data were calculated. Ground truth for benign lesions: absence of malignancy after cancer registry matching (2022); for malignancy: histopathologic proof; evaluation was patient-based.

RESULTS

In total, 114 screen-detected and 17 interval cancers (ICA) occurred. ROC analysis of screen-detected cancers yielded an AUC of 89% for AI-CAD. Sensitivity/specificity was 81.7%/80.2% for AI-CAD; 77.1%/91.7% for R1; 78.6/91.6% for R2. Combining each human reading with AI-CAD was as sensitive as human double-reading (all approximately 88%), but less specific (approximately 75%) compared to human double-reading (approximately 87%). These AI-CAD combinations required consensus readings for twice as many cases as the human combination. Four of 17 ICA exceeded a case score of 30; two of four CAD correctly marked the quadrant of the subsequent ICA.

CONCLUSION

Including ICA cases, this AI-CAD achieved comparable sensitivity to human reading at lower specificity. Combining human reading and AI-CAD allows increasing sensitivity compared to single-reading.

Image-guided breast biopsy and localisation: recommendations for information to women and referring physicians by the European Society of Breast Imaging

We summarise here the information to be provided to women and referring physicians about percutaneous breast biopsy and lesion localisation under imaging guidance. After explaining why a preoperative diagnosis with a percutaneous biopsy is preferred to surgical biopsy, we illustrate the criteria used by radiologists for choosing the most appropriate combination of device type for sampling and imaging technique for guidance. Then, we describe the commonly used devices, from fine-needle sampling to tissue biopsy with larger needles, namely core needle biopsy and vacuum-assisted biopsy, and how mammography, digital breast tomosynthesis, ultrasound, or magnetic resonance imaging work for targeting the lesion for sampling or localisation. The differences among the techniques available for localisation (carbon marking, metallic wire, radiotracer injection, radioactive seed, and magnetic seed localisation) are illustrated. Type and rate of possible complications are described and the issue of concomitant antiplatelet or anticoagulant therapy is also addressed. The importance of pathological-radiological correlation is highlighted: when evaluating the results of any needle sampling, the radiologist must check the concordance between the cytology/pathology report of the sample and the radiological appearance of the biopsied lesion. We recommend that special attention is paid to a proper and tactful approach when communicating to the woman the need for tissue sampling as well as the possibility of cancer diagnosis, repeat tissue sampling, and or even surgery when tissue sampling shows a lesion with uncertain malignant potential (also referred to as "high-risk" or B3 lesions). Finally, seven frequently asked questions are answered.

Consensus Meeting of Breast Imaging: BI-RADS® and Beyond

Organizers of medical educational courses are often confronted with questions that are clinically relevant yet trespassing the frontiers of scientifically proven, evidence-based medicine at the point of care. Therefore, since 2007 organizers of breast teaching courses in German language met biannually to find a consensus in clinically relevant questions that have not been definitely answered by science. The questions were prepared during the 3 months before the meeting according to a structured process and finally agreed upon the day before the consensus meeting. At the consensus meeting, the open questions concerning 2D/3D mammography, breast ultrasound, MR mammography, interventions as well as risk-based imaging of the breast were presented first for electronic anonymized voting, and then the results of the audience were separately displayed from the expert votes. Thereafter, an introductory statement of the moderator was followed by pros/cons of two experts, and subsequently the final voting was performed. With ≥75% of votes of the expert panel, an answer qualified as a consensus statement. Seventeen consensus statements were gained, addressing for instance the use of 2D/3D mammography, breast ultrasound in screening, MR mammography in women with intermediate breast cancer risk, markers for localization of pathologic axillary lymph nodes, and standards in risk-based imaging of the breast. After the evaluation, comments from the experts on each field were gathered supplementarily. Methodology, transparency, and soundness of statements achieve a unique yield for all course organizers and provide solid pathways for decision making in breast imaging.

Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System

Purpose To compare breast cancer detection performance of radiologists reading mammographic examinations unaided versus supported by an artificial intelligence (AI) system. Materials and Methods An enriched retrospective, fully crossed, multireader, multicase, HIPAA-compliant study was performed. Screening digital mammographic examinations from 240 women (median age, 62 years; range, 39-89 years) performed between 2013 and 2017 were included. The 240 examinations (100 showing cancers, 40 leading to false-positive recalls, 100 normal) were interpreted by 14 Mammography Quality Standards Act-qualified radiologists, once with and once without AI support. The readers provided a Breast Imaging Reporting and Data System score and probability of malignancy. AI support provided radiologists with interactive decision support (clicking on a breast region yields a local cancer likelihood score), traditional lesion markers for computer-detected abnormalities, and an examination-based cancer likelihood score. The area under the receiver operating characteristic curve (AUC), specificity and sensitivity, and reading time were compared between conditions by using mixed-models analysis dof variance and generalized linear models for multiple repeated measurements. Results On average, the AUC was higher with AI support than with unaided reading (0.89 vs 0.87, respectively; P = .002). Sensitivity increased with AI support (86% [86 of 100] vs 83% [83 of 100]; P = .046), whereas specificity trended toward improvement (79% [111 of 140]) vs 77% [108 of 140]; P = .06). Reading time per case was similar (unaided, 146 seconds; supported by AI, 149 seconds; P = .15). The AUC with the AI system alone was similar to the average AUC of the radiologists (0.89 vs 0.87). Conclusion Radiologists improved their cancer detection at mammography when using an artificial intelligence system for support, without requiring additional reading time. Published under a CC BY 4.0 license. See also the editorial by Bahl in this issue.

Breast ultrasound: recommendations for information to women and referring physicians by the European Society of Breast Imaging

This article summarises the information that should be provided to women and referring physicians about breast ultrasound (US). After explaining the physical principles, technical procedure and safety of US, information is given about its ability to make a correct diagnosis, depending on the setting in which it is applied. The following definite indications for breast US in female subjects are proposed: palpable lump; axillary adenopathy; first diagnostic approach for clinical abnormalities under 40 and in pregnant or lactating women; suspicious abnormalities at mammography or magnetic resonance imaging (MRI); suspicious nipple discharge; recent nipple inversion; skin retraction; breast inflammation; abnormalities in the area of the surgical scar after breast conserving surgery or mastectomy; abnormalities in the presence of breast implants; screening high-risk women, especially when MRI is not performed; loco-regional staging of a known breast cancer, when MRI is not performed; guidance for percutaneous interventions (needle biopsy, pre-surgical localisation, fluid collection drainage); monitoring patients with breast cancer receiving neo-adjuvant therapy, when MRI is not performed. Possible indications such as supplemental screening after mammography for women aged 40-74 with dense breasts are also listed. Moreover, inappropriate indications include screening for breast cancer as a stand-alone alternative to mammography. The structure and organisation of the breast US report and of classification systems such as the BI-RADS and consequent management recommendations are illustrated. Information about additional or new US technologies (colour-Doppler, elastography, and automated whole breast US) is also provided. Finally, five frequently asked questions are answered.

TEACHING POINTS

• US is an established tool for suspected cancers at all ages and also the method of choice under 40. • For US-visible suspicious lesions, US-guided biopsy is preferred, even for palpable findings. • High-risk women can be screened with US, especially when MRI cannot be performed. • Supplemental US increases cancer detection but also false positives, biopsy rate and follow-up exams. • Breast US is inappropriate as a stand-alone screening method.

Use of single-view digital breast tomosynthesis (DBT) and ultrasound vs. additional views and ultrasound for the assessment of screen-detected abnormalities: German multi-reader study

Background Data on the value of digital breast tomosynthesis (DBT) for further assessment of screen-detected lesions are still limited. Purpose To compare screening mammography, single-view DBT and ultrasound-information (TS) vs. screening mammography, additional views and ultrasound-information (AV) for assessment of screen-detected abnormalities. Material and Methods The use of wide-angle DBT for screen-detected, soft-tissue abnormalities requiring additional views was investigated: 241 cases (206 benign and 35 malignant lesions), verified by histology or two-year follow-up, were read by ten readers as TS and as AV sets, yielding 2410 diagnoses for each set. Readings were randomly sequenced. Results The mean interval between readings was nine weeks (random sequence). Evaluation was breast-based. Overall, in terms of area under receiver operating characteristic (AUC; varying degree of suspicion cutoff), TS and AV readings showed similar performance: for TS, AUC was 0.889 (95% confidence interval [CI] = 0.871-0.907) and for AV, AUC was 0.903 (95% CI = 0.886-0.921). TS readings had slightly higher sensitivity than AV readings (96.9% vs. 95.4%) but lower specificity (50% vs. 58.1%) and more variations between reader performance; absolute false negatives (FN) were reduced in 8/16 readers, equal in 5/16, and increased in only 3/16. Conclusion This study broadly confirms previous data showing equivalence of DBT to AV. However, bias against TS may have occurred since the region of interest was not indicated in the TS set as compared to its obvious identification on the AV set by the selected spot views. A key finding is that reader experience with DBT may be more important than so far reported.

Suppression of Unspecific Enhancement on Breast Magnetic Resonance Imaging (Mri) by Antiestrogen Medication

Aims and Background

The value of breast MRI may be impaired by unspecific enhancement. This may leave patients with difficult-to-assess breast tissue with an uncertain diagnosis. We examined whether this unspecific enhancement (which is mostly due to proliferative or hyperplastic changes of benign breast tissue) may be suppressed by antiestrogen medication.

Methods

In a trial of treatment, 10 peri- or postmenopausal patients who exhibited diffuse and/or focal enhancement on breast MRI before tamoxifen medication agreed to undergo a short-term tamoxifen treatment. MRI monitoring was performed 2, 4 and 8 weeks after onset of antiestrogen therapy (tamoxifen, 30 mg per day).

Results

Six patients showed a significant decrease of enhancement. Unchanged (n = 3) or increased (n = 1) enhancement was seen in 4 patients. One of the three patients with unchanged enhancement proved to have diffuse lobular carcinoma in situ.

Conclusions

Part of the unspecific enhancement seen on breast MRI can probably be suppressed by short-term antiestrogen medication.

Breast MRI: EUSOBI recommendations for women's information

This paper summarizes information about breast MRI to be provided to women and referring physicians. After listing contraindications, procedure details are described, stressing the need for correct scheduling and not moving during the examination. The structured report including BI-RADS® categories and further actions after a breast MRI examination are discussed. Breast MRI is a very sensitive modality, significantly improving screening in high-risk women. It also has a role in clinical diagnosis, problem solving, and staging, impacting on patient management. However, it is not a perfect test, and occasionally breast cancers can be missed. Therefore, clinical and other imaging findings (from mammography/ultrasound) should also be considered. Conversely, MRI may detect lesions not visible on other imaging modalities turning out to be benign (false positives). These risks should be discussed with women before a breast MRI is requested/performed. Because breast MRI drawbacks depend upon the indication for the examination, basic information for the most important breast MRI indications is presented. Seventeen notes and five frequently asked questions formulated for use as direct communication to women are provided. The text was reviewed by Europa Donna-The European Breast Cancer Coalition to ensure that it can be easily understood by women undergoing MRI.

KEY POINTS

• Information on breast MRI concerns advantages/disadvantages and preparation to the examination • Claustrophobia, implantable devices, allergic predisposition, and renal function should be checked • Before menopause, scheduling on day 7-14 of the cycle is preferred • During the examination, it is highly important that the patient keeps still • Availability of prior examinations improves accuracy of breast MRI interpretation.

Magnetic resonance imaging: the evolution of breast imaging

INTRODUCTION AND AIMS

To provide an overview of the principle of current breast MRI, the available evidence concerning its indications and optimum use and future potentials.

METHODS AND RESULTS

To date sensitivities of 90-91% have been achieved with a specificity of 72-75%. MRI is the most sensitive method for detecting invasive carcinoma and comparable to mammography concerning detection of DCIS. The achievable specificity, false positive and biopsy rates, however, are much lower than for screening mammography thus do not allow its use for screening of the general population. Indications with proven advantages concern screening of women at high risk and special diagnostic problems that cannot be solved by conventional imaging and percutaneous biopsy: search for primary tumour in CUP syndrome, differentiation of nipple retraction, differentiation of scarring versus recurrence and selected difficult cases. There is no proven benefit for its general use for preoperative staging. One major problem may concern the imperfect interface between imaging and surgery. Further research is also needed for the use of MRI in women at intermediate risk. In women at low risk MRI screening is not recommended. Novel possibilities of MRI concern diffusion weighted imaging as well as MR spectroscopy. Their value for improved lesion differentiation is not yet fully established. Their main potential appears to concern an improved and earlier prediction of response to neoadjuvant therapy. Future developments might address development of more specific contrast agents, replacement of vascular enhancing agents by special MR techniques, testing of sodium MRI or image fusion with other imaging modalities.

DISCUSSION/CONCLUSION

MRI allows new patho-physiological information and thus can complement the information available by conventional methods. Present research should concentrate on improving specificity, improving the interface of imaging and surgery and has to include outcome analyses. Due to issues of specificity the responsible use of MRI should be limited to appropriate indications.

Advantages and Disadvantages of Mammography Screening

Mammography screening is the only method presently considered appropriate for mass screening of asymptomatic women. Its frequent use, however, warrants diligent analysis of potential side effects. Radiation risk is far below the natural yearly risk of breast cancer and should not be used as an argument against screening. False-positive calls lead to additional imaging or histopathological assessment, mainly percutaneous breast biopsy. These measures are tolerated and accepted fairly well. Their number is limited by strict quality assurance and constant training. Interval cancers represent a limitation of breast screening that should prompt further research for optimization. Evaluation of overdiagnosis is a highly debated topic in the literature. According to the probably most realistic available calculations, overdiagnosis is acceptable as it is compensated by the potential mortality reduction. Nonetheless, this potential side effect warrants optimal adjustment of therapy to the patient's individual risk. The mortality reduction seen in randomized studies was confirmed by results from national screening programs. A recent case referent study indicated that improvements in mortality reduction run parallel to improved mammographic techniques. Use of less aggressive therapies is another valuable effect of screening. Awareness of potential problems, strict quality assurance, and further research should help to further develop screening programs.

B3 Lesions: Radiological Assessment and Multi-Disciplinary Aspects

B3 lesions comprise different histopathological entities that are considered benign but 'of unknown biological potential'. These entities may act as risk indicators (for both breasts) or as non-obligatory precursors of malignancy. Being diagnosed at percutaneous breast biopsy, an additional risk of underestimate exists. Imaging appearances, histopathological appearance and risk of associated malignancy are presented. B3 lesions of high risk, which thus should usually be excised, include atypical ductal hyperplasia (ADH), pleomorphic or necrotic type of lobular neoplasia (LIN 3), and papillary lesions with atypias. Intermediate risk may be associated with classic lobular carcinoma in situ (LIN 2) or flat epithelial atypia (FEA), and low risk with radial sclerosing lesions (RSLs) and papillary lesions without atypias. LIN 1 is mostly an incidental finding acting as risk indicator. Follow-up is adequate if the initial diagnostic problem is solved. According to international guidelines, risk and subsequent recommendations should be discussed for each individual patient, taking into account biological risk, representative sampling, lesion size, lesion extent, percentage of lesion removal, other individual risks, and the possibility of surveillance. With vacuum-assisted breast biopsy (VABB), surgery may be avoided for more of the small lesions at low risk. Further data collection and diligent evaluation may help to better assess the individual risk, to better adapt treatment recommendations and avoid overtreatment.

Magnetic resonance imaging of the breast: recommendations from the EUSOMA working group

The use of breast magnetic resonance imaging (MRI) is rapidly increasing. EUSOMA organised a workshop in Milan on 20-21st October 2008 to evaluate the evidence currently available on clinical value and indications for breast MRI. Twenty-three experts from the disciplines involved in breast disease management - including epidemiologists, geneticists, oncologists, radiologists, radiation oncologists, and surgeons - discussed the evidence for the use of this technology in plenary and focused sessions. This paper presents the consensus reached by this working group. General recommendations, technical requirements, methodology, and interpretation were firstly considered. For the following ten indications, an overview of the evidence, a list of recommendations, and a number of research issues were defined: staging before treatment planning; screening of high-risk women; evaluation of response to neoadjuvant chemotherapy; patients with breast augmentation or reconstruction; occult primary breast cancer; breast cancer recurrence; nipple discharge; characterisation of equivocal findings at conventional imaging; inflammatory breast cancer; and male breast. The working group strongly suggests that all breast cancer specialists cooperate for an optimal clinical use of this emerging technology and for future research, focusing on patient outcome as primary end-point.

Use of ultrasound-guided percutaneous vacuum-assisted breast biopsy for selected difficult indications

To assess ultrasound-guided vacuum-assisted biopsy (US-VAB) for selected problem cases and to report experiences with two different biopsy systems. Fifty-one lesions have been biopsied using the Mammotome (n = 24) or the Vacora (n = 27) system. Main indications: lesion in scarring (n = 5), complex cystic >or=8 mm (n = 7), increase in size (n = 10), architectural distortion (n = 4), uncharacteristic palpable abnormality (2), small size (n = 22), regional microcalcifications (n = 1). Results are verified by surgical excision (n = 10) or follow-up (n = 40). One patient was lost to follow-up. In four of the cases preceding core biopsy was inconclusive. four invasive carcinomas, two ductal carcinoma in situ (DCIS), three papillomas, six fibroadenomas, one adenosis tumor, one hamartoma, 10 complex cysts, 16 benign changes, three fat necroses, two granulomas, three unspecific inflammatory changes are verified. Surgery confirmed five malignancies, four benign changes, and converted one uncertain diagnosis (architectural distortion) from "inflammatory" to DCIS. Documented removal of all or most of the lesions correctly increased the level of confidence and open surgery could be avoided in 41/51 lesions. The two systems show different advantages and drawbacks. US-VAB may improve the level of confidence in selected difficult cases. Careful case selection and systematic retrospective correlation of imaging and histology remain crucial.

Proceedings of the international consensus conference on breast cancer risk, genetics, & risk management, April, 2007

A consensus conference including thirty experts was held in April, 2007, to discuss risk factors for breast cancer and their management. Four categories of risk were outlined, from breast cancer "average" through "very high" risk, the latter including individuals with high penetrance BRCA1/2 gene mutations. Guidelines for management of patients in each of these categories were discussed, with the major portion of the conference being devoted to individuals with BRCA1/2 mutations. Prevalence of these mutations in the general populations was estimated to be 1 in 250-500 individuals, with an increased prevalence in Ashkenazic Jews and other founder groups. Risk reduction strategies for these individuals include surveillance, with or without chemoprevention drugs, or surgical procedures to remove the organs at risk, i.e., bilateral mastectomy and/or bilateral salpingo-oophorectomy. These risk reduction strategies were evaluated fully, and recommendations were made for the care of patients in each of the risk categories. These guidelines for patient care were approved by the entire group of experts.

[Mammography screening. Concept, quality assurance and interdisciplinary cooperation]

In 2005/2006 the German National Mammography Screening Program was initiated and has now become established. The objective is to reduce breast cancer mortality and the early diagnosis and therapy of small cancers. The program follows the European guidelines and is controlled by over 30 parameters of quality. All trained members of the team document each step of the screening chain electronically. Histological assessment (HA) is recommended in up to 2% of examinations, 90% of HAs are performed by core needle biopsy (CNB) or by stereotactic vacuum-assisted biopsy (VABB). Open diagnostic biopsies are performed in <10% of all HAs and therapy is successful in some of the B3 lesions. Mammograms are interpreted by two independent readers. Recommendations of the regular interdisciplinary conferences, preoperative and postoperative, follow the European guidelines. About 45% of all breast cancers detected by screening are in-situ or less than 10 mm in size. The 17% alterations diagnosed by needle biopsy are B3 or B4 lesions and impose high demands on the pathologists and the interdisciplinary team. Due to the many early and discrete lesions counterchecking of representative biopsies is crucial. Problems may be caused by sampling error or partial volume effects. Interdisciplinary conferences and knowledge of the limitations of each discipline and method are needed to optimize diagnostic and therapeutic decisions.

Imaging studies for the early detection of breast cancer

INTRODUCTION

The effectiveness of mammographic screening has been proven at evidence level 1A. Mammography offers the best ratio of benefits to side effects of any screening method tested to date. In this literature review, we ask whether early detection might be improved still further by combining mammography with other imaging modalities.

METHODS

The authors performed a selective literature search for combined key words in the Medline and Cochrane Library databases from 1/2000 to 11/2007, screened all titles, and evaluated the full text of all original articles. We selected articles for further analysis according to systematic criteria (minimum numbers, avoidance of overlap) and also considered published guidelines.

RESULTS

No screening studies of comparable size to those for mammography are available for ultrasound or MRI. Smaller studies have indicated that the use of these two modalities might lead to the detection of additional cancers in selected subgroups. For mass screening an increase in the detection rate of 10% to 15% might become possible. This increase would probably be associated with a tripling of the breast biopsy rate, compared to mammography alone. The number of indeterminate cases in which short-term follow-up (i.e., at 6 months) would be recommended would increase roughly tenfold with MRI, and to an unknown extent with ultrasound. The related quality-assurance issues remain to be addressed.

DISCUSSION

Randomized controlled studies are needed for a realistic assessment of the achievable benefits and unavoidable side effects of combined screening. For women whose risk of breast cancer is not elevated, mammography remains the standard screening method.

[Experiences of the Bavarian mammography screening program]

The Bavarian Mammography Screening Program started in April 2003. A detailed analysis of the consistency of diagnosis in the evaluation of vacuum-assisted stereotactic or core needle breast biopsies is presented. A total of 32 pathologists participated in a blinded evaluation of the biopsies. Each case was evaluated independently by two participating pathologists. A total of 1,357 cases were reviewed. The histopathological reports of the biopsies made by the two consulting pathologists were compared. The concordance rate of the first and second consulting pathologist was 93% for the B-classification. In general, the level of diagnostic agreement was very high for well defined, benign and malignant lesions. Some of the discrepancies resulted from the incorrect application of the B-classification. Discrepancies in the reports were also due to divergent interpretation of benign and "borderline" lesions. The protocol for the blinded evaluation of breast biopsies in two rounds assured a high level of quality. In conclusion, prerequisites for the success of a mammography screening program are interdisciplinary consensus conferences and audit rounds involving pathologists.

MR-guided intervention in women with a family history of breast cancer

OBJECTIVE

A study was undertaken to assess the clinical value of magnetic resonance (MR) imaging-guided interventions in women with a family history, but no personal history of breast cancer.

METHODS AND PATIENTS

Retrospective review was performed on 63 consecutive women who had a family history, but no personal history of breast cancer. A total of 97 lesions were referred for an MR-guided intervention. Standardized MR examinations (1.0 T, T1-weighted 3D FLASH, 0.15 mmolGd-DTPA/kg body weight, prone position) were performed using a dedicated system which allows vacuum assisted breast biopsy or wire localization.

RESULTS

Histologic findings in 87 procedures revealed 9 (10%) invasive carcinomas, 12 (14%) ductal carcinomas in situ, 2 atypical ductal hyperplasias (2.5%) and 2 atypical lobular hyperplasias (2.5%). Sixty-two (71%) benign histologic results are verified by an MR-guided intervention, retrospective correlation of imaging and histology and by subsequent follow-up. In ten lesions the indication dropped since the enhancing lesion was no longer visible. Absent enhancement was confirmed by short-term re-imaging of the noncompressed breast and by follow-up.

CONCLUSION

Malignancy was found in 24%, high-risk lesions in 5% of successfully performed MR-guided biopsy procedures. A 57% of MR-detected malignancies were ductal carcinoma in situ. In 10% of the lesions the intervention was not performed, since no enhancing lesion could be reproduced at the date of anticipated intervention. Such problems may be avoided if the initial MRI is performed in the appropriate phase of the menstrual cycle and without hormonal replacement therapy.

Diagnostic Architectural and Dynamic Features at Breast MR Imaging: Multicenter Study

PURPOSE

To prospectively determine the prevalence and predictive value of three-dimensional (3D) and dynamic breast magnetic resonance (MR) imaging and contrast material kinetic features alone and as part of predictive diagnostic models.

MATERIALS AND METHODS

The study protocol was approved by the institutional review board or ethics committees of all participating institutions, and informed consent was obtained from all participants. Although study data collection was performed before HIPAA went into effect, standards that would be compliant with HIPAA were adhered to. Data from the International Breast MR Consortium trial 6883 were used in the analysis. Women underwent 3D (minimum spatial resolution, 0.7 x 1.4 x 3 mm; minimal temporal resolution, 4 minutes) and dynamic two-dimensional (temporal resolution, 15 seconds) MR imaging examinations. Readers rated enhancement shape, enhancement distribution, border architecture, enhancement intensity, presence of rim enhancement or internal septations, and the shape of the contrast material kinetic curve. Regression was performed for each feature individually and after adjustment for associated mammographic findings. Multivariate models were also constructed from multiple architectural and dynamic features. Areas under the receiver operating characteristic curve (Az values) were estimated for all models.

RESULTS

There were 995 lesions in 854 women (mean age, 53 years +/- 12 [standard deviation]; range, 18-80 years) for whom pathology data were available. The absence of enhancement was associated with an 88% negative predictive value for cancer. Qualitative characterization of the dynamic enhancement pattern was associated with an Az value of 0.66 across all lesion architectures. Focal mass margins (Az = 0.76) and signal intensity (Az = 0.70) were highly predictive imaging features. Multivariate models were constructed with an Az value of 0.880.

CONCLUSION

Architectural and dynamic features are important in breast MR imaging interpretation. Multivariate models involving feature assessment have a diagnostic accuracy superior to that of qualitative characterization of the dynamic enhancement pattern.

MRI detection of distinct incidental cancer in women with primary breast cancer studied in IBMC 6883

BACKGROUND

Prior single institution studies suggest MRI may improve the assessment of the extent of cancer within the breast, and thus reduce the risk of leaving macroscopic disease in the breast following breast conservation therapy. We report on the rate of MRI and mammography detection of foci of distinct incidental cancer in a prospective, multi center trial involving 426 women with confirmed breast cancer at 15 institutions in the US, Canada, and Germany.

METHODS

Women underwent mammography and MRI prior to biopsy of the suspicious index lesion. Additional incidental lesions (IL) greater than 2 cm from the index lesion that were detected by mammography and MRI were noted and characterized. Biopsy recommendations were associated with ILs given an assessment of suspicious or highly suspicous (BiRads 4 and 5). These assessments were considered a positive test.

RESULTS

MRI had a significantly higher yield of confirmed cancer ILs than mammography (0.18 (95%CI: 0.142-0.214) for MRI versus 0.072 (95%CI: 0.050-0.100) for mammography). The cancer ILs detected by MRI alone appeared to be similar to those detected by mammography with respect to size and histology. The percentage of biopsies of ILs that resulted in a cancer diagnosis was similar between the modalities (MRI 0.72(95%CI: 0.6-0.81); Mammography 0.85 (95%CI: 0.62-0.96)).

CONCLUSIONS

These results suggest that consideration needs to be given regarding the integration of breast MRI into the pretreatment evaluation of women seeking breast conservation therapy.

Clipmarkierung nach MRT-gesteuerter Vakuumbiopsie der Mamma

PURPOSE

The purpose of our study was to determine the placement accuracy, usefulness as a guide for wire localization, and long-term stability of tissue marker clips following MR-guided vacuum-assisted biopsy (VB) of breast lesions.

METHODS

During a 2-year period, MR-guided VB with an 11-gauge device was performed in 79 lesions. In 26 lesions a marker clip was placed at the biopsy site.

RESULTS

In 18 cases, the clip was shown to be closely adjacent to the lesion on post-interventional MR images. In seven cases in which minor bleeding occurred, the clip dislocated (< or =15 mm) in the direction of the needle pathway. In one case dislocation in the dorsal direction (< or =5 mm) was observed. In eight cases with a malignant or borderline histology according to the VB, the marker clips served as targets for mammographically guided wire localization. In all of those patients, histology results derived from open surgery confirmed those of VB. Eighteen patients with benign findings according to the VB were followed up 6 months later. Among these cases we found a significant displacement of the marker clip in one case (3 cm). The clip generally caused a round artifact (diameter of 9+/-2 mm). In two cases it was not possible to determine whether the lesion had been removed completely or was just behind the artifact caused by the clip.

CONCLUSION

Based on our results, clip marker placement following MR-guided vacuum biopsy should be called into question due to the possibility of masking the lesion by a metallic artifact and because of possible dislocation.

Magnetic resonance imaging of the breast prior to biopsy

CONTEXT

Breast magnetic resonance imaging (MRI) has been shown to have high sensitivity for cancer detection and is increasingly used following mammography to evaluate suspicious breast lesions.

OBJECTIVE

To determine the accuracy of breast MRI in conjunction with mammography for the detection of breast cancer in patients with suspicious mammographic or clinical findings.

DESIGN, SETTING, AND PATIENTS

Prospective multicenter investigation of the International Breast MR Consortium conducted at 14 university hospitals in North America and Europe from June 2, 1998, through October 31, 2001, of 821 patients referred for breast biopsy for American College of Radiology category 4 or 5 mammographic assessment or suspicious clinical or ultrasound finding.

INTERVENTIONS

MRI examinations performed prior to breast biopsy; MRI results were interpreted at each site, which were blinded to pathological results.

MAIN OUTCOME MEASURES

Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of breast MRI.

RESULTS

Among the 821 patients, there were 404 malignant index lesions, of which 63 were ductal carcinoma in situ (DCIS) and 341 were invasive carcinoma. Of the 417 nonmalignant index lesions, 366 were benign, 47 showed atypical histology, and 4 were lobular carcinoma in situ. The AUC pooled over all institutions was 0.88 (95% confidence interval [CI], 0.86-0.91). MRI correctly detected cancer in 356 of 404 cancer cases (DCIS or invasive cancer), resulting in a sensitivity of 88.1% (95% CI, 84.6%-91.1%), and correctly identified as negative for cancer 281 of 417 cases without cancer, resulting in a specificity of 67.7% (95% CI, 62.7%-71.9%). MRI performance was not significantly affected by mammographic breast density, tumor histology, or menopausal status. The positive predictive values for 356 of 492 patients was 72.4% (95% CI, 68.2%-76.3%) and of mammography for 367 of 695 patients was 52.8% (95% CI, 49.0%-56.6%) (P<.005). Dynamic MRI did not improve the AUC compared with 3-dimensional MRI alone, but the specificity of a washout pattern for 123 of 136 patients without cancer was 90.4% (95% CI, 84%-95%).

CONCLUSIONS

Breast MRI has high sensitivity but only moderate specificity independent of breast density, tumor type, and menopausal status. Although the positive predictive value of MRI is greater than mammography, MRI does not obviate the need for subsequent tissue sampling in this setting.

Stereotactic vacuum-assisted breast biopsy in 2874 patients

BACKGROUND

Vacuum-assisted breast biopsy (VAB) can replace surgical biopsy for the diagnosis of breast carcinoma. The authors evaluated the accuracy and clinical utility of VAB in a multicenter setting using a strict quality assurance protocol.

METHODS

In the current study, VABs were performed successfully for 2874 patients at 5 sites. Benign lesions were verified by follow-up. Surgery was recommended for malignant and borderline lesions. VAB was performed on patients with lesions rated as highly suspicious (6%), intermediate to suspicious (85%), or probably benign (9%). Fifty-eight percent of the lesions were < 10 mm and 70% had microcalcifications.

RESULTS

The authors identified 7% of patients with invasive carcinomas, 15% with ductal carcinomas in situ (DCIS), 5% with atypical ductal hyperplasias (ADH), and 0.6% with lobular carcinomas in situ. The results of the VAB necessitated an upgrade of 24% of patients with ADH to DCIS or DCIS and invasive carcinoma. Twelve percent of patients with DCIS proved to have invasive carcinoma. Seventy-three percent of the patients had benign lesions. Only 1 false-negative result was encountered (negative predictive value, 99.95%). Minor side effects were reported to occur in 1.4% of patients and 0.1% of patients required a subsequent intervention. Scarring relevant for mammography was rare among patients (i.e., 0.3% of patients had relevant scarring).

CONCLUSIONS

Quality-assured VAB was found to be highly reliable. VAB effectively identified patients with benign lesions and assisted therapeutic decisions. Most important, only a single case of malignancy was missed. A close interdisciplinary approach assured optimal results.

Success of sentinel lymph node mapping after breast cancer ablation with focused microwave phased array thermotherapy

BACKGROUND

Breast cancer tumor ablation as part of a multimodality approach in the treatment of breast cancer is the subject of recent interest. This study was conducted to determine if the ability to perform sentinel node biopsy was impaired after thermal-induced ablation of breast cancer.

METHODS

We studied patients who had sentinel node biopsy after preoperative focused microwave phased array for breast cancer ablation.

RESULTS

Twenty-one patients with T1-T2 breast cancer and clinically negative axilla underwent wide local excision and sentinel node biopsy guided by blue dye and sulfur colloid. Surgery was done an average of 17 days after microwave ablation. Fifteen of 22 patients (68%) had histologic evidence of tumor necrosis. Sentinel lymph node mapping was successful in 19 of 21 patients (91%). Axillary metastases were detected in 42% of cases.

CONCLUSIONS

This study documents successful sentinel lymph node mapping for patients treated with antecedent local tumor ablation using focused microwave phased array ablation.

Interdisciplinary consensus on the use and technique of vacuum-assisted stereotactic breast biopsy

PURPOSE

Quality assurance of stereotactic vacuum-assisted breast biopsy (VB).

METHOD

A consensus was achieved based on the existing literature and the experience of VB users (Ethicon Endosurgery, Norderstedt).

RESULTS

The imaging work-up must be completed according to existing standards before an indication for stereotactic VB is established. Indications include microcalcifications and small non-palpable masses; for the time being lesions very close to the skin and architectural distortions (radial scar) are considered less suitable. Acquisition of >20 cores (11 Gauge) should be routinely attempted (goals: as complete a removal of small lesions as possible, thereby increasing diagnostic confidence and reducing so-called 'underestimates'). The pre/post-fire and post-biopsy stereotactic images and a post-biopsy orthogonal mammogram must be documented. All cases with no or uncertain histopathological correlation require discussion in a regular interdisciplinary conference and a documented consensus concerning further work-up or therapy. Standardised documentation of the primary findings and follow-up mammography after approximately 6 months is requested.

CONCLUSION

This consensus includes protocols for the establishment of an indication, performance indicators, interdisciplinary interpretation and therapeutic recommendation, documentation and follow-up. It does not replace official recommendations for percutaneous biopsy.

Follow-up of breast lesions detected by MRI not biopsied due to absent enhancement of contrast medium

Our objective was to follow-up patients in whom scheduled MR-guided vacuum biopsies for suspicious lesions were aborted due to absent enhancement of contrast medium. Thirty-seven of 291 scheduled MR-guided vacuum biopsies were aborted. Six cases were lost to follow-up. Two could be unequivocally identified and were nevertheless biopsied. In 25 of 29 patients absent enhancement was confirmed on subsequent studies without compression. Varying hormonal or inflammatory changes between initial MRI and MR-guided vacuum biopsy most probably explain the findings. Enhancement re-appeared on short-term follow-up <6 months without compression in 4 of the 29 patients. Too strong compression during MR-guided vacuum biopsy explains the absence of enhancement in these patients. Of note, on histology, three of these cases proved malignant. We conclude that short-term follow-up without compression is necessary and recommended for all lesions not visible during scheduled MR-guided vacuum biopsy.

European quadricentric evaluation of a breast MR biopsy and localization device: technical improvements based on phase-I evaluation

Our purpose was to report about technical success, problems and solutions, as experienced in a first multicentre study on MR-guided localisation or vacuum biopsy of breast lesions. The study was carried out at four European sites using a dedicated prototype breast biopsy device. Experiences with 49 scheduled localisation procedures and 188 vacuum biopsies are reported. Apart from 35 dropped indications, one localisation procedure and 9 vacuum biopsies were not possible (3 times space problems due to obesity, 2 times too strong compression, 3 times impaired access from medially, 2 times impaired access due to a metal bar). Problems due to too strong compression were recognised by repeat MR without compression. During the procedure problems leading to an uncertain result occurred in eight vacuum biopsies, two related to the procedure: one limited access, and one strong post-biopsy enhancement. Improvements after phase-I study concerned removal of the metal bar, development of an improved medial access, of a profile imitating the biopsy gun, optimisation of compression plates and improved software support. The partners agreed that the improvements answered all important technical problems.

Axillary lymph node metastases: a statistical analysis of various parameters in MRI with USPIO

PURPOSE

To assess the value of plain vs. iron oxide-enhanced MRI vs. the combined study (plain + postcontrast) based on qualitative and quantitative parameters of three pulse sequences.

MATERIALS AND METHODS

Data from two sites were acquired using the same technique; therefore, this data could be pooled. T1W-SE, T2W-FSE, and 3D-PSIF were used before and 24-36 hours after MRI with ultra small particles of iron oxide (USPIO) was performed. A total of 52 lymph nodes (LNs) in nine patients (25 benign, 27 malignant) were evaluated by two readers who were visually and quantitatively blinded to the histology. Combinations of the following diagnostic parameters were compared using logistic regression analysis: the short-axis diameter of the LN, the signal distribution of the LN on postcontrast agent MRI (homogeneous or heterogeneous), and qualitatively and quantitatively determined signal changes of the LN following administration of contrast agent in the three evaluated sequences.

RESULTS

Using pre- and postcontrast data, the optimized accuracy based on the statistically most significant parameters (LN diameter > 6 mm, visual assessment of signal change on T2W-SE) was 87% (81% sensitivity, 92% specificity). Precontrast data alone yielded 75% accuracy (63% sensitivity, 86% specificity). Postcontrast data alone yielded 75% accuracy (56% sensitivity, 96% specificity).

CONCLUSION

Based on our results, USPIO-MRI improved the diagnosis of metastatic axillary LNs compared with precontrast MRI alone. Both pre- and postcontrast studies are needed. T1W-SE and T2W-PSIF did not yield significant additional information. This study may help to further improve the technique of USPIO imaging.

Multicenter study for the evaluation of a dedicated biopsy device for MR-guided vacuum biopsy of the breast

The purpose of this multicenter study was to determine the accuracy and clinical value of a dedicated breast biopsy system which allows for MR-guided vacuum biopsy (VB) of contrast-enhancing lesions. In five European centers, MR-guided 11-gauge VB was performed on 341 lesions. In 7 cases VB was unsuccessful. This was immediately realized on postinterventional images or direct follow-up combined with histopathology-imaging correlation; thus, a false-negative diagnosis was avoided. Histology of 334 successful biopsies yielded 84 (25%) malignancies, 17 (5%) atypical ductal hyperplasias, and 233 (70%) benign entities. Verification of malignant or borderline lesions included reexcision of the biopsy cavity. Benign histologic biopsy results were verified by retrospective correlation with the pre- and postinterventional MRI and by subsequent follow-up. Our results indicate that MR-guided VB, in combination with the dedicated biopsy coil, offers the possibility to accurately diagnose even very small lesions that can only be visualized or localized by MRI.

[Do tissue marker clips after sonographically or stereotactically guided breast biopsy improve follow-up of small breast lesions and localisation of breast cancer after chemotherapy?]

PURPOSE

We wanted to determine if tissue marker clips after sonographically or stereotactically guided breast biopsy improve the follow-up of small breast lesions classified BI-RADS 4/5 and the localisation of breast cancer (TNM stage 2 or 3) after neoadjuvant chemotherapy.

MATERIAL AND METHODS

Prospective analysis was performed of 108 breast lesions 1 cm or smaller mammographically classified as BI-RADS 4/5 and 14 breast lesions larger than 2 cm mammographically classified as BI-RADS 5. 33 of the 108 breast lesions 1 cm or smaller underwent sonographic core cut breast biopsy (group 1) and 75 stereotactic vacuum-assisted breast biopsy (group 2). All 14 lesions greater than 2 cm were stereotactically vacuum-assisted breast biopsied (group 3). The centre of the lesion was marked by a clip after the biopsy. Mammographies were performed in all patients of group 1 and 2 with a histologically benign finding (n = 31, n = 69, respectively) and in all patients of group 3 directly after clip placement and after 6 and 12 months. Clip localisation and possible divergence from the original position were verified by a grid.

RESULTS

Two patients of group 1 and 6 patients of group 2 had breast conservative surgery (BET) because of the histological diagnosis of a ductal carcinoma in situ or invasive breast cancer. The tissue marker clips of the remaining 31 patients of group 1 and 69 patients of group 2 diverged with a mean value of 0.4 cm (standard deviation +/- 0.23 cm; range 0.1 cm to 0.9 cm) from their placement position after 6 months. After 12 months the marker clips deviated with a mean value of 0.4 cm (standard deviation +/- 0.21 cm; range 0.1 cm to 0.9 cm) in 94 patients and 0.8 cm (standard deviation +/- 0.25 cm; range 0.1 cm to 0.9 cm) in 6 patients from their original location. No tumour progression of the benign lesions in group 1 and 2 was diagnosed in follow-up mammograms. In all patients of group 3 the tissue marker clips were the only possibility to localize the tumour after neoadjuvant chemotherapy as all other diagnostic methods showed inconsistent results.

CONCLUSION

Positioning a tissue marker clip in the tumour centre seems to be reasonable after interventional biopsy of breast lesions of 1.0 cm or smaller and before neoadjuvant chemotherapy.

MR-guided interventional breast procedures considering vacuum biopsy in particular

Histologic work-up of just MR-detected breast lesions has become essential with increasing use of contrast-enhanced MR imaging. In the present article an overview is given about the different MR-guided breast interventions, performed since 1990. Presently, for reasons of costs and image quality closed magnets are most widely used. The following approaches have been described: MR-guided freehand localization in supine position, stereotaxic localization in supine position and most frequently used localization in the prone position by means of a compression device that immobilises the breast to prevent tissue shift during intervention. Only limited experience exists with interventions on open magnets. MR-guided wire localization is a well-established procedure. Recently, percutaneous vacuum biopsy of enhancing breast lesions has become possible under MR guidance. The new system allows accurate and safe access to lesions in any location of the breast and direct check-up of representative excision by visualisation of the cavity. Thus reliable histologic evaluation of lesions smaller than 10 mm is possible with this approach.

Stereotactic breast biopsies: analysis of needle deviation based on stereotactic views

Stereotaxy is widely used for breast biopsy and needle localization of mammographically detected lesions. If a lesion shift occurs during stereotaxy, corrections (even though possible with vacuum biopsy) is difficult due to the difficult assessment of the exact 3D shift. In this study we investigated the correlation between lesion shift (in up to three dimensions) and its visualisation on the stereotactic images (0 degree, -15 degrees, +15 degrees). The study was performed on a Fischer prone table (Fischer Imaging Europe, Vejle, Denmark) using a 3.8-mm steel ball (as lesion) and a 20-G needle. The 17 major malpositions of the ball with respect to the needle were imitated and imaged. A simple rule is suggested as to how the deviation in all three dimensions can be detected from the scout and the two stereotactic views. The rule proved to be a valuable tool to correctly assess lesion shifts.

CT-guided obturator nerve block for diagnosis and treatment of painful conditions of the hip

Obturator nerve blocks (ONB) have been performed by anaesthesiologists mainly to eliminate the obturator reflex during transurethral resections. An effect on hip pain has also been described. However, being a time-consuming and operator-dependent procedure if performed manually, it has not been widely used for chronic hip pain. The purpose of this pilot study was to check whether CT guidance could improve reproducibility of the block (= immediate effect) and to test its potential value for treatment of chronic hip pain. Fifteen chronically ill patients with osteoarthritis underwent a single ONB. Sixteen millilitres of Lidocaine 1% mixed with 2 ml Iopramide was injected into the obturator canal. The patients were followed up to 9 months after the intervention. With a single injection pain relief was achieved for 1-8 weeks in 7 of 15 patients. Excellent pain relief for 3-11 months was achieved in another 4 patients. Reasons for a mid-term or even long-term effect based on a single injection of local anaesthetic are not exactly known. The CT-guided ONB is a fast, easy and safe procedure that may be useful for mid-term (weeks) and sometimes even long-term (months) treatment of hip pain.

MR-Guided vacuum biopsy of 206 contrast-enhancing breast lesions

PURPOSE

To determine the accuracy and clinical use of MR-guided vacuum biopsy (VB) of enhancing breast lesions.

MATERIAL AND METHODS

254 lesions were referred to MR-guided vacuum-assisted breast biopsy. In 43 (16 %) patients the indication was dropped because the lesions could not be identified at the time VB was scheduled. This was due to hormonal influences (n = 37), to too strong compression (n = 3) or to misinterpretation of the initial diagnostic MRI (n = 3). In 5 cases (2 %) VB was not performed due to obesity (n = 2); problems of access (n = 2) or a defect of the MR-unit (n = 1). VB was performed on altogether 206 lesions. In 4 cases (2 %) VB was unsuccessful. This was immediately realized on the post-interventional images. Thus a false negative diagnosis was avoided. Verification included excision of the cavity in cases with proven malignancy or atypical ductal hyperplasia (ADH) and (for benign lesions) retrospective correlation of VB-histology with pre-and postinterventional MRI and subsequent follow-up.

RESULTS

51/202 successful biopsies proved malignancy. In 7 cases ADH and in 144 cases a benign lesion was diagnosed. One DCIS was underestimated as ADH. All other benign or malignant diagnoses proved to be correct.

CONCLUSION

MR-guided VB allows reliable histological work-up of contrast-enhancing small lesions which are not visible by any other modality.

Preservation of histopathologic assessment by use of a special technique for tissue inactivation by radiofrequency

Thermoinactivation is under investigation as a new method of cancer treatment. Prior histopathologic verification has been indispensable. Our experiments show that thermoinactivation is in principle possible without destroying histopathologic morphology, if the temperature stays within certain limits. It could thus become applicable already before diagnostic biopsy, which might be used to avoid any potential spread during biopsy and before definitive surgery.