Automatic correction of performance drift under acquisition shift in medical image classification.

Image-based prediction models for disease detection are sensitive to changes in data acquisition such as the replacement of scanner hardware or updates to the image processing software. The resulting differences in image characteristics may lead to drifts in clinically relevant performance metrics which could cause harm in clinical decision making, even for models that generalise in terms of area under the receiver-operating characteristic curve. We propose Unsupervised Prediction Alignment, a generic automatic recalibration method that requires no ground truth annotations and only limited amounts of unlabelled example images from the shifted data distribution. We illustrate the effectiveness of the proposed method to detect and correct performance drift in mammography-based breast cancer screening and on publicly available histopathology data. We show that the proposed method can preserve the expected performance in terms of sensitivity/specificity under various realistic scenarios of image acquisition shift, thus offering an important safeguard for clinical deployment.

Performance of a Breast Cancer Detection AI Algorithm Using the Personal Performance in Mammographic Screening Scheme.

Background The Personal Performance in Mammographic Screening (PERFORMS) scheme is used to assess reader performance. Whether this scheme can assess the performance of artificial intelligence (AI) algorithms is unknown. Purpose To compare the performance of human readers and a commercially available AI algorithm interpreting PERFORMS test sets. Materials and Methods In this retrospective study, two PERFORMS test sets, each consisting of 60 challenging cases, were evaluated by human readers between May 2018 and March 2021 and were evaluated by an AI algorithm in 2022. AI considered each breast separately, assigning a suspicion of malignancy score to features detected. Performance was assessed using the highest score per breast. Performance metrics, including sensitivity, specificity, and area under the receiver operating characteristic curve (AUC), were calculated for AI and humans. The study was powered to detect a medium-sized effect (odds ratio, 3.5 or 0.29) for sensitivity. Results A total of 552 human readers interpreted both PERFORMS test sets, consisting of 161 normal breasts, 70 malignant breasts, and nine benign breasts. No difference was observed at the breast level between the AUC for AI and the AUC for human readers (0.93% and 0.88%, respectively; P = .15). When using the developer's suggested recall score threshold, no difference was observed for AI versus human reader sensitivity (84% and 90%, respectively; P = .34), but the specificity of AI was higher (89%) than that of the human readers (76%, P = .003). However, it was not possible to demonstrate equivalence due to the size of the test sets. When using recall thresholds to match mean human reader performance (90% sensitivity, 76% specificity), AI showed no differences inperformance, with a sensitivity of 91% (P =. 73) and a specificity of 77% (P = .85). Conclusion Diagnostic performance of AI was comparable with that of the average human reader when evaluating cases from two enriched test sets from the PERFORMS scheme. © RSNA, 2023 See also the editorial by Philpotts in this issue.

Comparing Prognostic Factors of Cancers Identified by Artificial Intelligence (AI) and Human Readers in Breast Cancer Screening.

Invasiveness status, histological grade, lymph node stage, and tumour size are important prognostic factors for breast cancer survival. This evaluation aims to compare these features for cancers detected by AI and human readers using digital mammography. Women diagnosed with breast cancer between 2009 and 2019 from three UK double-reading sites were included in this retrospective cohort evaluation. Differences in prognostic features of cancers detected by AI and the first human reader (R1) were assessed using chi-square tests, with significance at p < 0.05. From 1718 screen-detected cancers (SDCs) and 293 interval cancers (ICs), AI flagged 85.9% and 31.7%, respectively. R1 detected 90.8% of SDCs and 7.2% of ICs. Of the screen-detected cancers detected by the AI, 82.5% had an invasive component, compared to 81.1% for R1 (p-0.374). For the ICs, this was 91.5% and 93.8% for AI and R1, respectively (p = 0.829). For the invasive tumours, no differences were found for histological grade, tumour size, or lymph node stage. The AI detected more ICs. In summary, no differences in prognostic factors were found comparing SDC and ICs identified by AI or human readers. These findings support a potential role for AI in the double-reading workflow.

Preoperative diagnosis of ductal carcinoma in situ of the breast over a 24-year period.

AIMS: The method of diagnosis of ductal carcinoma in situ (DCIS) has changed since the 1980s. The aim of this audit was to assess changes in the preoperative diagnosis of DCIS since the introduction of needle core biopsy, particularly the proportion with a preoperative biopsy diagnosis of DCIS. METHODS AND RESULTS: The preoperative diagnoses of patients with a final diagnosis of DCIS in the surgical specimen were reviewed (i) in 809 patients who presented through breast screening from 1997 to 2021, and (ii) in all patients in 5 individual years at 5-year intervals from 2000 to 2020 (254 in total). For screening-detected DCIS the proportion with a preoperative diagnosis of DCIS increased from 75% to 98% over the study period. In a detailed analysis of all cases of DCIS in 5 separate years the proportion with a preoperative diagnosis of DCIS increased from 68% in 2000 to 96% in 2020. For high-grade DCIS the proportion increased from 87% to 97%, and for low- or intermediate-grade DCIS from 48% to 93%. The proportion of women who had vacuum-assisted biopsy increased from 7% in 2000 to 58% in 2015. There was a small increase in the number of biopsies that had basal cytokeratin and oestrogen receptor immunohistochemistry to aid diagnosis. CONCLUSION: There has been an increase in the preoperative diagnosis of DCIS, particularly of low- or intermediate-grade, over the last two decades. The increasing use of vacuum-assisted biopsy is likely to be a major contributory factor to this increase.

Multi-vendor evaluation of artificial intelligence as an independent reader for double reading in breast cancer screening on 275,900 mammograms.

BACKGROUND: Double reading (DR) in screening mammography increases cancer detection and lowers recall rates, but has sustainability challenges due to workforce shortages. Artificial intelligence (AI) as an independent reader (IR) in DR may provide a cost-effective solution with the potential to improve screening performance. Evidence for AI to generalise across different patient populations, screening programmes and equipment vendors, however, is still lacking. METHODS: This retrospective study simulated DR with AI as an IR, using data representative of real-world deployments (275,900 cases, 177,882 participants) from four mammography equipment vendors, seven screening sites, and two countries. Non-inferiority and superiority were assessed for relevant screening metrics. RESULTS: DR with AI, compared with human DR, showed at least non-inferior recall rate, cancer detection rate, sensitivity, specificity and positive predictive value (PPV) for each mammography vendor and site, and superior recall rate, specificity, and PPV for some. The simulation indicates that using AI would have increased arbitration rate (3.3% to 12.3%), but could have reduced human workload by 30.0% to 44.8%. CONCLUSIONS: AI has potential as an IR in the DR workflow across different screening programmes, mammography equipment and geographies, substantially reducing human reader workload while maintaining or improving standard of care. TRIAL REGISTRATION: ISRCTN18056078 (20/03/2019; retrospectively registered).

Measuring reader fatigue in the interpretation of screening digital breast tomosynthesis (DBT).

OBJECTIVES: The interpretation of digital breast tomosynthesis (DBT) screening examinations is a complex task for an already overstretched workforce which has the potential to increase pressure on readers leading to fatigue and patient safety issues. Studies in non-medical and medical settings have suggested that changes in blink characteristics can reflect fatigue. The purpose of this study is to investigate the use of blink characteristics as an objective marker of fatigue in readers interpreting DBT breast screening examinations. METHODS: Twenty-six DBT readers involved in the UK PROSPECTS trial interpreted a test set of 40 DBT cases while being observed by an eye tracking device from November 2019 to February 2021. Raw data from the eye tracker were collected and automated processing software was used to produce eye blinking characteristics data which were analysed using multiple linear regression statistical models. RESULTS: Of the 26 DBT readers recruited, eye tracking data from 23 participants were analysed due to missing data rendering 3 participants' data uninterpretable. The mean reading time per DBT case was 2.81 min. There was a statistically significant increase in blinking duration of 0.38 ms/case as the reading session progressed (p < 0.0001). This was the result of a significant decrease in the number of ultra-short blinks lasting ≤50 ms (p = 0.0005) and a significant increase in longer blinks lasting 51-100 ms (p = 0.008). CONCLUSION: Changes in blinking characteristics could serve as objective measures of reader fatigue and may prove useful in the development of DBT reading protocols. ADVANCES IN KNOWLEDGE: Blink characteristics can be used as an objective measure of fatigue; however there is limited evidence of their use in radiological settings. Our study suggests that changes in blink duration and frequency could be used to monitor fatigue in DBT reading sessions.

Performance of Radiologists and Radiographers in Double Reading Mammograms: The UK National Health Service Breast Screening Program.

Background Double reading can be used in screening mammography, but it is labor intensive. There is limited evidence on whether trained radiographers (ie, technologists) may be used to provide double reading. Purpose To compare the performance of radiologists and radiographers double reading screening mammograms, considering reader experience level. Materials and Methods In this retrospective study, performance and experience data were obtained for radiologists and radiographer readers of all screening mammograms in England from April 2015 to March 2016. Cancer detection rate (CDR), recall rate (RR), and positive predictive value (PPV) of recall based on biopsy-proven findings were calculated for first readers. Performance metrics were analyzed according to reader professional group and years of reading experience using the analysis of variance test. P values less than .05 were considered to indicate statistically significant difference. Results During the study period, 401 readers (224 radiologists and 177 radiographers) double read 1 404 395 screening digital mammograms. There was no difference in CDR between radiologist and radiographer readers (mean, 7.84 vs 7.53 per 1000 examinations, respectively; P = .08) and no difference for readers with more than 10 years of experience compared with 5 years or fewer years of experience, regardless of professional group (mean, 7.75 vs 7.71 per 1000 examinations respectively, P = .87). No difference in the mean RR was observed between radiologists and radiographer readers (5.0% vs 5.2%, respectively, P = .63). A lower RR was seen for readers with more than 10 years of experience compared with 5 years or fewer, regardless of professional group (mean, 4.8% vs 5.8%, respectively; P = .001). No variation in PPV was observed between them (P = .42), with PPV values of 17.1% for radiologists versus 16.1% for radiographers. A higher PPV was seen for readers with more than 10 years of experience compared with 5 years or less, regardless of professional group (mean, 17.5% and 14.9%, respectively; P = .02). Conclusion No difference in performance was observed between radiographers and radiologists reading screening mammograms in a program that used double reading. Published under a CC BY 4.0 license Online supplemental material is available for this article. See also the editorial by Hooley and Durand in this issue.

Artificial Intelligence as Supporting Reader in Breast Screening: A Novel Workflow to Preserve Quality and Reduce Workload.

OBJECTIVE: To evaluate the effectiveness of a new strategy for using artificial intelligence (AI) as supporting reader for the detection of breast cancer in mammography-based double reading screening practice. METHODS: Large-scale multi-site, multi-vendor data were used to retrospectively evaluate a new paradigm of AI-supported reading. Here, the AI served as the second reader only if it agrees with the recall/no-recall decision of the first human reader. Otherwise, a second human reader made an assessment followed by the standard clinical workflow. The data included 280 594 cases from 180 542 female participants screened for breast cancer at seven screening sites in two countries and using equipment from four hardware vendors. The statistical analysis included non-inferiority and superiority testing of cancer screening performance and evaluation of the reduction in workload, measured as arbitration rate and number of cases requiring second human reading. RESULTS: Artificial intelligence as a supporting reader was found to be superior or noninferior on all screening metrics compared with human double reading while reducing the number of cases requiring second human reading by up to 87% (245 395/280 594). Compared with AI as an independent reader, the number of cases referred to arbitration was reduced from 13% (35 199/280 594) to 2% (5056/280 594). CONCLUSION: The simulation indicates that the proposed workflow retains screening performance of human double reading while substantially reducing the workload. Further research should study the impact on the second human reader because they would only assess cases in which the AI prediction and first human reader disagree.

The relationship between missed breast cancers on mammography in a test-set based assessment scheme and real-life performance in a National Breast Screening Programme.

PURPOSE: This retrospective study determined whether a test-set based assessment scheme (PERFORMS) used in a national breast screening programme could be used to predict real-life performance by investigating if the number of cancers missed by mammography readers in real-life related to the number of cancers missed in the PERFORMS test-set and whether real-life reading volumes affected performance. METHOD: Data was obtained from consenting readers in the screening programme in England (NHSBSP) where double reading is standard. The rate of cancers missed by individual first readers but correctly identified by second readers was compared with the number of cancers missed in the PERFORMS test-set over a 3-year period. NHSBSP readers are required to interpret at least 1500 cases per year as a first reader, so results were compared between readers who exceeded this target and those that did not. Parametric and non-parametric correlations were calculated. RESULTS: Amongst the 536 readers, there was a highly significant positive correlation between the real-life and PERFORMS test-set missed cancer metrics (Pearson Correlation = 0.228, n = 536, p < .0001, Spearman's rho = 0.265, n = 536, p < .0001). There was no significant difference in rates of missed cancers between the 452 readers who exceeded the 1500 first read per year target and those who did not (t(94.2) = -1.87, p = .0643, r = 0.19). CONCLUSIONS: The use of a test-set based assessment scheme accurately reflects real-life mammography reading performance, indicating that it can be a useful tool in identifying poor reader performance.

Effectiveness of percutaneous vacuum-assisted excision (VAE) of breast lesions of uncertain malignant potential (B3 lesions) as an alternative to open surgical biopsy.

OBJECTIVES: Traditionally B3 breast lesions are treated surgically, but overtreatment is a concern, as the majority have a final benign diagnosis. A national screening program introduced vacuum-assisted excision (VAE) for managing B3 lesions in late 2016. This retrospective study aimed to assess the outcomes associated with this approach. METHODS: All B3 lesions diagnosed between 01/2017 and 12/2019 were identified at two centres. Information was obtained on the initial biopsy and final histology, and method of VAE image guidance, needle size and number of cores. Lesions were excluded if there was cancer elsewhere in the breast at the time of diagnosis; the lesion was not suitable for VAE due to position in the breast or had B3 pathology for which open biopsy was still required. The final decision to offer VAE was always made at a multidisciplinary meeting (MDM). Risk difference was used to test the significance at p ≤ .05. RESULTS: In total, 258 B3 lesions were diagnosed, 105 (40.7%) met the inclusion criteria and underwent VAE. VAE was performed under X-ray (89/105) or ultrasound guidance (16/105), taking an average of 18.5 cores with the 10-G needle or 10.8 cores with the 7-G needle. Nine cases (8.6%) were upgraded to a malignant diagnosis following VAE. Malignancy was found in 15.5% (9/58) of B3 lesions with epithelial atypia, but in none without atypia (0/47) (p = .004). No new lesions or malignancy has occurred at the site of the VAE with an average mammographic follow-up of 2.2 years. CONCLUSION: Upgrade to malignancy following VAE was uncommon (8.6%) and associated with atypia in the initial biopsy. VAE is an alternative approach to the management of B3 lesions, reducing open surgical procedures. KEY POINTS: • Upgrade to malignancy after a vacuum-assisted excision of a B3 breast lesion is uncommon with an 8.6% upgrade rate. • The risk of a malignant diagnosis after a vacuum-assisted excision was significantly higher for B3 lesions with atypia compared to those without (+15.5% difference, p = .004).

The Relationship between Mammography Readers' Real-Life Performance and Performance in a Test Set-based Assessment Scheme in a National Breast Screening Program.

Purpose: To compare an individual's Personal Performance in Mammographic Screening (PERFORMS) score with their Breast Screening Information System (BSIS) real-life performance data and determine which parameters in the PERFORMS scheme offer the best reflection of BSIS real-life performance metrics. Materials and Methods: In this retrospective study, the BSIS real-life performance metrics of individual readers (n = 452) in the National Health Service Breast Screening Program (NHSBSP) in England were compared with performance in the test set-based assessment scheme over a 3-year period from 2013 to 2016. Cancer detection rate (CDR), recall rate, and positive predictive value (PPV) were calculated for each reader, for both real-life screening and the PERFORMS test. For each metric, real-life and test set versions were compared using a Pearson correlation. The real-life CDR, recall rate, and PPV of outliers were compared against other readers (nonoutliers) using analysis of variance. Results: BSIS real-life CDRs, recall rates, and PPVs showed positive correlations with the equivalent PERFORMS measures (P < .001, P = .002, and P < .001, respectively). The mean real-life CDR of PERFORMS outliers was 7.2 per 1000 women screened and was significantly lower than other readers (nonoutliers) where the real-life CDR was 7.9 (P = .002). The mean real-life screening PPV of PERFORMS outliers was 0.14% and was significantly lower than the nonoutlier group who had a mean PPV of 0.17% (P = .006). Conclusion: The use of test set-based assessment schemes in a breast screening program has the potential to predict and identify poor performance in real life.© RSNA, 2020Keywords: Breast, ScreeningSee also the commentary by Thigpen and Rapelyea in this issue.

Accuracy of GE digital breast tomosynthesis vs supplementary mammographic views for diagnosis of screen-detected soft-tissue breast lesions.

OBJECTIVE: To compare the accuracy of standard supplementary views and GE digital breast tomosynthesis (DBT) for assessment of soft-tissue mammographic abnormalities. METHODS: Women recalled for further assessment of soft-tissue abnormalities were recruited and received standard supplementary views (typically spot compression views) and two-view GE DBT. The added value of DBT in the assessment process was determined by analysing data collected prospectively by radiologists working up the cases. Following anonymization of cases, there was also a retrospective multireader review. The readers first read bilateral standard two-view digital mammography (DM) together with the supplementary mammographic views and gave a combined score for suspicion of malignancy on a five-point scale. The same readers then read bilateral standard two-view DM together with two-view DBT. Pathology data were obtained. Differences were assessed using receiver operating characteristic analysis. RESULTS: The study population was 342 lesions in 322 patients. The final diagnosis was malignant in 113 cases (33%) and benign/normal in 229 cases (67%). In the prospective analysis, the performance of two-view DM plus DBT was at least equivalent to the performance of two-view DM and standard mammographic supplementary views-the area under the curve (AUC) was 0.946 and 0.922, respectively, which did not reach statistical significance. Similar results were obtained for the retrospective review-AUC was 0.900 (DBT) and 0.873 (supplementary views), which did not reach statistical significance. CONCLUSION: The accuracy of GE DBT in the assessment of screen detected soft-tissue abnormalities is equivalent to the use of standard supplementary mammographic views. ADVANCES IN KNOWLEDGE: The vast majority of evidence relating to the use of DBT has been gathered from research using Hologic equipment. This study provides evidence for the use of the commercially available GE DBT system demonstrating that it is at least equivalent to supplementary mammographic views in the assessment of soft-tissue screen-detected abnormalities.

The use of lower resolution viewing devices for mammographic interpretation: implications for education and training.

AIMS: To establish whether lower resolution, lower cost viewing devices have the potential to deliver mammographic interpretation training. METHODS: On three occasions over eight months, fourteen consultant radiologists and reporting radiographers read forty challenging digital mammography screening cases on three different displays: a digital mammography workstation, a standard LCD monitor, and a smartphone. Standard image manipulation software was available for use on all three devices. Receiver operating characteristic (ROC) analysis and ANOVA (Analysis of Variance) were used to determine the significance of differences in performance between the viewing devices with/without the application of image manipulation software. The effect of reader's experience was also assessed. RESULTS: Performance was significantly higher (p < .05) on the mammography workstation compared to the other two viewing devices. When image manipulation software was applied to images viewed on the standard LCD monitor, performance improved to mirror levels seen on the mammography workstation with no significant difference between the two. Image interpretation on the smartphone was uniformly poor. Film reader experience had no significant effect on performance across all three viewing devices. CONCLUSION: Lower resolution standard LCD monitors combined with appropriate image manipulation software are capable of displaying mammographic pathology, and are potentially suitable for delivering mammographic interpretation training. KEY POINTS: • This study investigates potential devices for training in mammography interpretation. • Lower resolution standard LCD monitors are potentially suitable for mammographic interpretation training. • The effect of image manipulation tools on mammography workstation viewing is insignificant. • Reader experience had no significant effect on performance in all viewing devices. • Smart phones are not suitable for displaying mammograms.

Mammographic features of breast cancers at single reading with computer-aided detection and at double reading in a large multicenter prospective trial of computer-aided detection: CADET II.

PURPOSE: To evaluate the mammographic features of breast cancer that favor lesion detection with single reading and computer-aided detection (CAD) or with double reading. MATERIALS AND METHODS: The Computer Aided Detection Evaluation Trial II study was approved by the ethics committee, and all participants provided written informed consent. A total of 31,057 women were recruited from three screening centers between September 2006 and August 2007. They were randomly allocated to the double reading group, the single reading with CAD group, or the double reading and single reading with CAD group at a ratio of 1:1:28, respectively. In this study, cancers in the women whose mammograms were read with both single reading with CAD and double reading were retrospectively reviewed. The original mammograms were obtained for each case and reviewed by two of three experienced breast radiologists in consensus. The method of detection was noted. The size and predominant mammographic feature of the cancer were recorded, as was the breast density. CAD marking data were reviewed to determine if the cancer had been correctly marked. RESULTS: A total of 227 cancers were detected in 28,204 women. A total of 170 cases were recalled with both reading regimens. Lesion types were masses (66%), microcalcifications (25%), parenchymal deformities (6%), and asymmetric densities (3%). The ability of the reading regimens to correctly prompt the reader to recall cases varied significantly by lesion type (P < .001). More parenchymal deformities were recalled with double reading, whereas more asymmetric densities were recalled with single reading with CAD. There was no difference in the ability of either reading regimen to prompt the reader to correctly recall masses or microcalcifications. CAD correctly prompted 100% of microcalcifications, 87% of mass lesions, 80% of asymmetric densities, and 50% of parenchymal deformities. CAD correctly marked 93% of spiculated masses compared with 80% of ill-defined masses (P = .054). There was a significant trend for cancers detected with double reading to occur only in women with a denser mammographic background pattern (P = .02). Size had no effect on lesion detection. CONCLUSION: Readers using either single reading with CAD or double reading need to be aware of the strengths and weaknesses of reading regimens to avoid missing the more challenging cancer cases.

Increasing participant recruitment into large-scale screening trials: experience from the CADET II study.

OBJECTIVES: Multicentre randomized trials frequently encounter difficulties in meeting their recruitment targets, resulting in extension of the trial and delays in implementation of the findings. We report on recruitment strategies implemented in a randomized evaluation of computer-aided detection in women attending routine screening in the UK Breast Screening Programme. SETTING: The target population for the trial was identified from an existing NHS database of women aged 50-70 invited for routine mammography in Coventry, Manchester and Nottingham, UK. Women were asked to consent to their mammograms being randomly allocated (in a ratio of 28:1:1) to one of three film-reading protocols. Trial information was mailed to women, along with the invitation to attend screening, and informed consent was obtained at the mammography appointment. Several strategies were implemented to increase recruitment rates. RESULTS: Recruitment rate increased significantly over time in the study (P < 0.0010 in all centres) with an overall acceptance rate of 46% of those attending screening. Mailing of the trial information sheet separate from the screening invitation in Coventry and Nottingham increased the recruitment rate, even after adjustment for the trend over time and for socioeconomic status of the attendees (P < 0.001). Extension of recruitment to mobile screening units in Nottingham, and the presence of an additional member of staff also improved recruitment (P < 0.001). Simplification of the trial information sheet had little effect. Increases in recruitment rate were not attributable to socioeconomic status of the attendees. CONCLUSIONS: In multicentre trials, monitoring of local recruitment protocols is required to ensure that each centre can maximize accrual targets.

Breast carcinoma with basal phenotype: mammographic findings.

OBJECTIVE: Basal phenotype has been found to be an independent poor prognostic factor for breast cancer. The aim of this study was to assess the mammographic appearance of screening-detected breast carcinoma with the basal phenotype. MATERIALS AND METHODS: A series of 1,944 consecutively enrolled patients with operable invasive breast cancer underwent immunohistochemical analysis with cytokeratin 5/6 and cytokeratin 14 markers to identify tumors exhibiting basal phenotype characteristics. Among those patients, 356 women with breast cancer were common to a prospectively collected database of screening-detected cases of breast cancer. The predominant mammographic appearance and any associated features were reported by experienced image readers blinded to phenotype status. A chi-square test was used to assess difference between the mammographic appearances of a group of tumors with the basal phenotype and those of a group with the nonbasal phenotype. RESULTS: Forty-one (12%) of the screening-detected tumors had basal phenotypic expression, and these were compared with 309 (88%) nonbasal tumors. Basal-phenotype tumors were significantly more likely to manifest as an ill-defined mass (basal phenotype, 25 [61%] of 41 tumors; nonbasal phenotype, 75 [24%] of 309 tumors; p < 0.001) or with comedo calcification (basal phenotype, nine [22%] of 41 tumors; nonbasal phenotype, 30 [10%] of 309 tumors; p = 0.019). Nonbasal-phenotype tumors were more likely to manifest as a spiculated mass (nonbasal phenotype, 150 [49%] of 309 tumors; basal phenotype, eight [20%] of 41 tumors; p < 0.001). The low rate of spiculation in basal tumors was independent of histologic grade. CONCLUSION: Screening-detected breast tumors with a basal phenotype have a mammographic appearance different from that of nonbasal tumors. This finding may explain the good prognostic value of mammographic spiculation reported in previous studies.

Influence of mammographic parenchymal pattern in screening-detected and interval invasive breast cancers on pathologic features, mammographic features, and patient survival.

OBJECTIVE: The aim of our study was to assess the effect of mammographic parenchymal pattern on patient survival, mammographic features, and pathologic features of breast cancer in a screened population. MATERIALS AND METHODS: We classified the parenchymal pattern (according to BI-RADS) of 759 screened women who presented with a screening-detected (n = 455) or interval (n = 304) invasive breast cancer. Pathologic details (tumor size, histologic grade, lymph node stage, vascular invasion, and histologic type) and mammographic appearances were recorded. Breast cancer-specific survival was ascertained, with a median follow-up of 9.0 years. RESULTS: An excess of interval cancers was seen in women with dense breasts (p < 0.0001). Screening-detected (but not interval) tumors were significantly smaller in fatty breasts (p = 0.014). Tumor grade, lymph node stage, vascular invasion, and histologic type did not vary significantly with mammographic parenchymal pattern in screening-detected or interval cancers. Screening-detected cancers in fatty breasts were more likely to appear as indistinct (p = 0.003) or spiculated (p = 0.002) masses in contrast to cancers in dense breasts, which more commonly appeared as architectural distortions (p < 0.0001). No significant breast cancer-specific survival difference was seen by mammographic parenchymal pattern for screening-detected cancers (p = 0.75), interval cancers (p = 0.82), or both groups combined (p = 0.12). CONCLUSION: The prognosis of screened women presenting with breast cancer is unrelated to dense mammographic parenchymal pattern despite an excess of interval cancers and larger screening-detected tumors in this group. These data support the mammographic screening of women with dense parenchymal patterns.

Is mammographic spiculation an independent, good prognostic factor in screening-detected invasive breast cancer?

OBJECTIVE: The aim of this study was to review the prognostic significance of pathologic and radiologic factors for screening-detected invasive breast cancers of any size. MATERIALS AND METHODS: The patient group was a consecutive series of 470 screening-detected invasive breast cancers that were diagnosed between 1988 and 1998. Data regarding tumor type, grade, maximum invasive diameter, lymph node status, and the presence or absence of vascular invasion were recorded, as were the mammographic features of the lesion. Survival was ascertained from hospital records and a cancer registry. Differences in survival were assessed using Kaplan-Meier survival curves with a log-rank test for difference. The significance of any correlations was assessed using the chi-square test and the chi-square test for trend. Multivariate analysis used a Cox proportional hazards model. RESULTS: At univariate analysis, large invasive size, the presence of definite vascular invasion, high histologic grade, and nodal involvement were associated with poorer breast-cancer-specific survival. Mammographic spiculation (the presence of either a spiculated mass or distortion) was associated with more prolonged breast-cancer-specific survival. The presence or absence of mammographic comedo calcification did not influence breast-cancer-specific survival. In a Cox multivariate analysis that included those factors significant in univariate analysis, size, grade, nodal stage, and mammographic spiculation maintained their prognostic significance. CONCLUSION: Mammographic spiculation is an independent, good prognostic factor for screening-detected invasive breast cancer. The mechanism of how mammographic spiculation confers a beneficial prognostic effect is not clear.