Objective

The objective of the present study is to evaluate the effect of comparison with multiple prior mammograms on the outcomes of screening mammography relative to comparison with a single prior mammogram.

Materials and methods

We retrospectively analyzed 46,288 consecutive screening mammograms performed at our institution for 22,792 women. We divided these examinations into three groups: those interpreted without comparison with prior mammograms, those interpreted in comparison with one prior examination, and those interpreted in comparison with two or more prior examinations. For each group, we determined the rate of examination recall. We also calculated the positive predictive value of recall (i.e., positive predictive value level 1 [PPV1]) and the cancer detection rate (CDR) for both the group of examinations compared with a single prior mammogram and the group compared with multiple prior mammograms. Generalized estimating equations with the logistic link function were used to determine the relative odds ratio of recall as a function of the number of comparisons, with adjustment made for age as a confounding variable. The Fisher exact test was performed to compare the PPV1 and the CDR in the different cohorts.

Results

The recall rate for mammograms interpreted without comparison with prior examinations was 16.6%, whereas that for mammograms compared with one prior examination was 7.8% and that for mammograms compared with two or more prior examinations was 6.3%. After adjustment was made for age, the odds ratio of recall for the group with multiple prior examinations relative to the group with a single prior examination was 0.864 (95% CI, 0.776-0.962; p = 0.0074). Statistically significant increases in the PPV1 of 0.05 (p = 0.0009) and in the CDR of 2.3 cases per 1000 examinations (p = 0.0481) were also noted for mammograms compared with multiple prior examinations relative to those compared with a single prior examination.

Conclusion

Comparison with two or more prior mammograms resulted in a statistically significant reduction in the screening mammography recall rate and increases in the CDR and PPV1 relative to comparison with a single prior mammogram.

Purpose

To assess the relationship between parameters measured on dynamic contrast material-enhanced (DCE) magnetic resonance (MR) imaging and fluorine 18 fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT) in primary invasive breast cancer.

Materials and methods

This HIPAA-compliant study was a retrospective review of medical records and therefore approved by the institutional review board without the requirement for informed consent. Patients with a diagnosis of invasive breast cancer from January 2005 through December 2009 who underwent both DCE MR imaging and FDG PET/CT before treatment initiation were retrospectively identified. Fractional volumes were measured for ranges of signal enhancement ratio (SER) values from DCE MR imaging data and compared with maximum standardized uptake values (SUVmax) from FDG PET/CT data. Linear regression analysis was performed to clarify the relationship between SER and SUVmax, adjusting for tumor size, pathologic grade, and receptor status.

Results

Analyzed were 117 invasive breast cancers in 117 patients. Overall, a higher percentage of high washout kinetics was positively associated with SUVmax (1.57% increase in SUVmax per 1% increase in high washout; P = .020), and a higher percentage of low plateau kinetics was negatively associated with SUVmax (1.19% decrease in SUVmax per 1% increase in low plateau; P = .003). These relationships were strongest among triple-negative (TN) tumors (4.34% increase in SUVmax per 1% increase in high washout and 2.65% decrease in SUVmax per 1% increase in low plateau; P = .018 and .004, respectively).

Conclusion

In invasive breast carcinoma, there is a positive relationship between the percentage of high washout and SUVmax and a negative relationship between the percentage of low plateau and SUVmax. These results are stronger in TN tumors.

Supplemental material

http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.13130058/-/DC1.

Purpose

To evaluate diffusion changes in the breast tumor-stromal boundary and adjacent tissue in response to neoadjuvant chemotherapy using high resolution diffusion-weighted imaging (HR-DWI).

Materials and methods

Seven patients with invasive breast cancer were imaged with HR-DWI before and early during treatment. The mean apparent diffusion coefficient (ADC) was plotted in 1-mm increments around the tumor boundary. Early change in ADC was measured for tumor, tumor boundary, and stromal regions, and the relationship to treatment response was evaluated using Spearman's correlation.

Results

Statistically significant correlations between treatment response and early changes in ADC were found for: (i) whole tumor (ρ = 0.93, 95% confidence interval [CI] = (0.58, 0.99), P = 0.003); (ii) tumor rim (ρ = 0.75, 95% CI = (-0.007, 0.96), P = 0.05); and (iii) boundary transition region (ρ = 0.86, 95% CI = (0.29, 0.98), P = 0.01). Early change in ADC of distal stroma had a marginally statistically significant positive correlation to treatment response (ρ = 0.71, 95% CI = (-0.084, 0.95), P = 0.07).

Conclusion

Proximity-dependent evaluation of HR-DWI data in the breast tumor-stromal boundary and adjacent tissue may provide information about response to therapy.

Purpose

To determine whether readout-segmented echo-planar diffusion imaging (RESOLVE) improves separation of malignant versus benign lesions compared to standard single-shot echo-planar imaging (ss-EPI) on BI-RADS 4/5 lesions detected on breast magnetic resonance imaging (MRI).

Materials and methods

Consecutive 3T breast MRI studies with BI-RADS 4/5 designation and subsequent biopsy or benign mastectomy were retrospectively identified. Freehand regions of interest (ROIs) were drawn on lesions and also on normal background fibroglandular tissue for comparison. Lesion-to-background contrast was evaluated by normalizing signal intensity of the lesion ROI by the normal background tissue ROI at b = 800. Statistical analysis used the Mann-Whitney/Wilcoxon rank-sum test for unpaired and Wilcoxon signed-rank for paired comparisons.

Results

Of 38 lesions in 32 patients, 10 were malignant. Lesion-to-background contrast was higher on RESOLVE than ss-EPI (1.80 ± 0.71 vs. 1.62 ± 0.63, P = 0.03). Mean apparent diffusion coefficient (ADC) was the same or lower on RESOLVE than ss-EPI, and this effect was largest in malignant lesions (RESOLVE 0.90 ± 0.13; ss-EPI 1.00 ± 0.13; median difference -0.10 (95% confidence interval [CI]: -0.17, -0.02) × 10(-3) mm(2) /sec; P = 0.014). By either diffusion method, there was a statistically significant difference between benign and malignant mean ADC (P < 0.001).

Conclusion

Increased lesion-to-background contrast and improved separation of benign from malignant lesions by RESOLVE compared to standard diffusion suggests that RESOLVE may show promise as an adjunct to clinical breast MRI.

Rationale and objectives

The aim of this work was to compare a high-resolution diffusion-weighted imaging (HR-DWI) acquisition (voxel size = 4.8 mm(3)) to a standard diffusion-weighted imaging (STD-DWI) acquisition (voxel size = 29.3 mm(3)) for monitoring neoadjuvant therapy-induced changes in breast tumors.

Materials and methods

Nine women with locally advanced breast cancer were imaged with both HR-DWI and STD-DWI before and after 3 weeks (early treatment) of neoadjuvant taxane-based treatment. Tumor apparent diffusion coefficient (ADC) metrics (mean and histogram percentiles) from both DWI methods were calculated, and their relationship to tumor volume change after 12 weeks of treatment (posttreatment) measured by dynamic contrast enhanced magnetic resonance imaging was evaluated with a Spearman's rank correlation.

Results

The HR-DWI pretreatment 15th percentile tumor ADC (P = .03) and early treatment 15th, 25th, and 50th percentile tumor ADCs (P = .008, .010, .04, respectively) were significantly lower than the corresponding STD-DWI percentile ADCs. The mean tumor HR-ADC was significantly lower than STD-ADC at the early treatment time point (P = .02), but not at the pretreatment time point (P = .07). A significant early treatment increase in tumor ADC was found with both methods (P < .05). Correlations between HR-DWI tumor ADC and posttreatment tumor volume change were higher than the STD-DWI correlations at both time points and the lower percentile ADCs had the strongest correlations.

Conclusion

These initial results suggest that the HR-DWI technique has potential for improving characterization of low tumor ADC values over STD-DWI and that HR-DWI may be of value in evaluating tumor change with treatment.

In this proof-of-concept work, we have developed a 3D-CNN architecture that is guided by the tumor mask for classifying several patient-outcomes in breast cancer from the respective 3D dynamic contrast-enhanced MRI (DCE-MRI) images. The tumor masks on DCE-MRI images were generated using pre- and post-contrast images and validated by experienced radiologists. We show that our proposed mask-guided classification has a higher accuracy than that from either the full image without tumor masks (including background) or the masked voxels only. We have used two patient outcomes for this study: (1) recurrence of cancer after 5 years of imaging and (2) HER2 status, for comparing accuracies of different models. By looking at the activation maps, we conclude that an image-based prediction model using 3D-CNN could be improved by even a conservatively generated mask, rather than overly trusting an unguided, blind 3D-CNN. A blind CNN may classify accurately enough, while its attention may really be focused on a remote region within 3D images. On the other hand, only using a conservatively segmented region may not be as good for classification as using full images but forcing the model's attention toward the known regions of interest.

Objective

We hypothesize that radiologists' estimated percentage likelihood assessments for the presence of ductal carcinoma in situ (DCIS) and invasive cancer may predict histologic outcomes.

Materials and methods

Two hundred fifty cases categorized as BI-RADS category 4 or 5 at four University of California Medical Centers were retrospectively reviewed by 10 academic radiologists with a range of 1-39 years in practice. Readers assigned BI-RADS category (1, 2, 3, 4a, 4b, 4c, or 5), estimated percentage likelihood of DCIS or invasive cancer (0-100%), and confidence rating (1 = low, 5 = high) after reviewing screening and diagnostic mammograms and ultrasound images. ROC curves were generated.

Results

Sixty-two percent (156/250) of lesions were benign and 38% (94/250) were malignant. There were 26 (10%) DCIS, 20 (8%) invasive cancers, and 48 (19%) cases of DCIS and invasive cancer. AUC values were 0.830-0.907 for invasive cancer and 0.731-0.837 for DCIS alone. Sensitivity of 82% (56/68), specificity of 84% (153/182), positive predictive value (PPV) of 66% (56/85), negative predictive value (NPV) of 93% (153/165), and accuracy of 84% ([56 + 153]/250) were calculated using an estimated percentage likelihood of 20% or higher as the prediction threshold for invasive cancer for the radiologist with the highest AUC (0.907; 95% CI, 0.864-0.951). Every 20% increase in the estimated percentage likelihood of invasive cancer increased the odds of invasive cancer by approximately two times (odds ratio, 2.4). For DCIS, using a threshold of 40% or higher, sensitivity of 81% (21/26), specificity of 79% (178/224), PPV of 31% (21/67), NPV of 97% (178/183), and accuracy of 80% ([21 + 178]/250) were calculated. Similarly, these values were calculated at thresholds of 2% or higher (BI-RADS category 4) and 95% or higher (BI-RADS category 5) to predict the presence of malignancy.

Conclusion

Using likelihood estimates, radiologists may predict the presence of invasive cancer with fairly high accuracy. Radiologist-assigned estimated percentage likelihood can predict the presence of DCIS, albeit with lower accuracy than that for invasive cancer.