Deep learning of mammogram images to reduce unnecessary breast biopsies: a preliminary study.

BACKGROUND: Patients with a Breast Imaging Reporting and Data System (BI-RADS) 4 mammogram are currently recommended for biopsy. However, 70-80% of the biopsies are negative/benign. In this study, we developed a deep learning classification algorithm on mammogram images to classify BI-RADS 4 suspicious lesions aiming to reduce unnecessary breast biopsies. MATERIALS AND METHODS: This retrospective study included 847 patients with a BI-RADS 4 breast lesion that underwent biopsy at a single institution and included 200 invasive breast cancers, 200 ductal carcinoma in-situ (DCIS), 198 pure atypias, 194 benign, and 55 atypias upstaged to malignancy after excisional biopsy. We employed convolutional neural networks to perform 4 binary classification tasks: (I) benign vs. all atypia + invasive + DCIS, aiming to identify the benign cases for whom biopsy may be avoided; (II) benign + pure atypia vs. atypia-upstaged + invasive + DCIS, aiming to reduce excision of atypia that is not upgraded to cancer at surgery; (III) benign vs. each of the other 3 classes individually (atypia, DCIS, invasive), aiming for a precise diagnosis; and (IV) pure atypia vs. atypia-upstaged, aiming to reduce unnecessary excisional biopsies on atypia patients. RESULTS: A 95% sensitivity for the "higher stage disease" class was ensured for all tasks. The specificity value was 33% in Task I, and 25% in Task II, respectively. In Task III, the respective specificity value was 30% (vs. atypia), 30% (vs. DCIS), and 46% (vs. invasive tumor). In Task IV, the specificity was 35%. The AUC values for the 4 tasks were 0.72, 0.67, 0.70/0.73/0.72, and 0.67, respectively. CONCLUSION: Deep learning of digital mammograms containing BI-RADS 4 findings can identify lesions that may not need breast biopsy, leading to potential reduction of unnecessary procedures and the attendant costs and stress.

Non-mass-type ductal carcinoma in situ of the breast on ultrasound: Features and pathological analysis.

AIMS: The aims of this study were to investigate the ultrasound features of non-mass-type ductal carcinoma in situ (DCIS) of the breast and conduct a pathological analysis. MATERIAL AND METHODS: Ultrasound images of 32 cases of non-mass-type DCIS of the breast, collected between September 2014 and June 2016, were analyzed. The characteristics of the lesions, including border, internal echogenicity, local glandular hyperplasia, micro-calcification, and intra-tumoral blood flow resistance index (RI), were analyzed, and a concurrent pathological analysis was conducted. RESULTS: Obvious local glandular hyperplasia was commonly observed in the 32 cases of non-mass-type DCIS of the breast. The internal echogenicity varied in intensity, exhibiting a "leopard pattern" or "zebra pattern." Color Doppler imaging revealed abundant blood flow signals within the lesion with an RI of >0.7. Isolated duct dilatation and micro-calcifications were occasionally observed within the lesions. High-grade DCIS was the predominant pathological type of non-mass-type DCIS. CONCLUSIONS: Non-mass-type DCIS of the breast often presents with obvious local glandular hyperplasia and varying internal echogenicity. High-grade DCIS is the frequent pathological type. Color Doppler imaging and RI measurement can assist in diagnosing non-mass-type DCIS of the breast.

Using the DNA Integrity Number to Analyze DNA Quality in Specimens Collected from Liquid-Based Cytology after Fine-Needle Aspiration of Breast Tumors and Lesions.

INTRODUCTION: Cancer genome analysis using next-generation sequencing requires adequate and high-quality DNA samples. Genomic analyses were conventionally performed using formalin-fixed paraffin-embedded sections rather than cytology samples such as cell block or smear specimens. Specimens collected from liquid-based cytology (LBC) have the potential to be sources of high-quality DNA suitable for genetic analysis even after long-term storage. METHODS: We collected breast tumor/lesion fractions from 92 residual LBC specimens using fine-needle aspiration (FNA) biopsy, including breast carcinoma (1 invasive carcinoma and 4 ductal carcinomas in situ), papillomatous lesion (5 intraductal papillomas), and fibroepithelial lesion (19 phyllodes tumors and 53 fibroadenomas) samples, and others (1 ductal adenoma, 1 hamartoma, 1 fibrocystic disease, and 7 unknown). DNA was extracted from all samples and subjected to DNA integrity number (DIN) score analysis. RESULTS: Average DIN score collected from 92 LBC specimens was significantly higher score. In addition, high-quality DNA with high DIN values (7.39 ± 0.80) was successfully extracted more than 12 months after storage of residual LBC specimens. CONCLUSION: Residual LBC specimens collected from FNA of the breast were verified to carry high-quality DNA and could serve as an alternate source for genetic analysis.

Atypia Involving Fibroadenomas: Outcomes and Upgrade Rates.

OBJECTIVE: Fibroadenomas (FAs) involved by atypia are rare. Consensus guidelines for management of FAs involved by atypia when diagnosed on image-guided biopsy do not exist because of limited data reporting surgical upgrade rates to ductal carcinoma in situ (DCIS) or invasive malignancy. Therefore, these lesions commonly undergo surgical excision. METHODS: This single-institution retrospective study identified cases of FAs involved by atypical ductal hyperplasia (ADH), atypical lobular hyperplasia (ALH), and/or lobular carcinoma in situ (LCIS) diagnosed on image-guided biopsy between January 2014 and April 2023 to determine upgrade rates. Cases with incidental atypia adjacent to but not involving FAs were excluded. RESULTS: Among 1736 FAs diagnosed on image-guided biopsy, 32 cases (1.8%) were FAs involved by atypia including 43.8% (14/32) ALH, 28.1% (9/32) ADH, 18.8% (6/32) LCIS, 6.3% (2/32) LCIS + ALH, and 3.1% (1/32) unspecified atypia. The most common imaging finding was a mass. Most cases, 81.3% (26/32), underwent subsequent surgical excisional biopsy. A single case of ADH involving and adjacent to an FA was upgraded to FA involved by low-grade DCIS on excision for an overall surgical upgrade rate of 3.8%. There were no cases upgraded to invasive malignancy. For those omitting surgical excision, there was no subsequent malignancy diagnosis at the FA biopsy site over a mean follow-up of 73 months. CONCLUSION: Cases of radiologic-pathologic concordant FAs involved by atypia have a low upgrade rate of 3.8% and should undergo multidisciplinary review. Larger multi-institutional analysis is needed to determine whether guidelines for excision of atypia should apply to atypia involving FAs.

Male ductal carcinoma in situ: diagnosis and management of a rare disease in men.

Ductal carcinoma in situ is very rare in male patients, accounting for approximately 5%-7% of all male breast cancers. We present a case of a man in his early 70s who presented with bloody nipple discharge and gynaecomastia and was subsequently diagnosed with ductal carcinoma in situ (DCIS). We discuss his management with surgical resection and the consideration of adjuvant treatment. We also review the existing literature on the presentation, diagnosis and management of DCIS in men.

Superparamagnetic iron oxide (SPIO) for axillary mapping in patients with ductal carcinoma in situ undergoing mastectomy: single-institution experience.

PURPOSE: Unnecessary axillary surgery can potentially be avoided in patients with DCIS undergoing mastectomy. Current guidelines recommend upfront sentinel lymph node biopsy during the index operation due to the potential of upstaging to invasive cancer. This study reviews a single institution's experience with de-escalating axillary surgery using superparamagnetic iron oxide dye for axillary mapping in patients undergoing mastectomy for DCIS. METHODS: This is a retrospective single-institution cross-sectional study. All medical records of patients who underwent mastectomy for a diagnosis of DCIS from August 2021 to January 2023 were reviewed and patients who had SPIO injected at the time of the index mastectomy were included in the study. Descriptive statistics of demographics, clinical information, pathology results, and interval sentinel lymph node biopsy were performed. RESULTS: A total of 41 participants underwent 45 mastectomies for DCIS. The median age of the participants was 58 years (IQR = 17; range 25 to 76 years), and the majority of participants were female (97.8%). The most common indication for mastectomy was diffuse extent of disease (31.7%). On final pathology, 75.6% (34/45) of mastectomy specimens had DCIS without any type of invasion and 15.6% (7/45) had invasive cancer. Of the 7 cases with upgrade to invasive disease, 2 (28.6%) of them underwent interval sentinel lymph node biopsy. All sentinel lymph nodes biopsied were negative for cancer. CONCLUSION: The use of superparamagnetic iron oxide dye can prevent unnecessary axillary surgery in patients with DCIS undergoing mastectomy.

Recent insights on the clinical, pathological, and molecular features of intraductal carcinoma of the prostate.

Intraductal carcinoma of the prostate, a unique histopathologic entity that is often observed (especially in advanced prostate cancer), is characterized by the proliferation of malignant cells within normal acini or ducts surrounded by a basement membrane. Intraductal carcinoma of the prostate is almost invariably associated with an adjacent high-grade carcinoma and is occasionally observed as an isolated subtype. Intraductal carcinoma of the prostate has been demonstrated to be an independent poor prognostic factor for all stages of cancer, whether localized, de novo metastatic, or castration-resistant. It also has a characteristic genetic profile, including high genomic instability. Recognizing and differentiating it from other pathologies is therefore important in patient management, and morphological diagnostic criteria for intraductal carcinoma of the prostate have been established. This review summarizes and outlines the clinical and pathological features, differential diagnosis, molecular aspects, and management of intraductal carcinoma of the prostate, as described in previous studies. We also present a discussion and future perspectives regarding intraductal carcinoma of the prostate.

Granulomatous lobular mastitis co-existing with ductal carcinoma in situ: Report of three cases and review of the literature.

Granulomatous lobular mastitis (GLM) is a benign and infrequent chronic breast ailment. Although this lesion can be clinically and radiographically mistaken for early-onset breast cancer, it is a rare occurrence for the two to coexist. This report describes three such cases. In all three patients, the primary signs and symptoms were related to the formation of diffuse breast masses or abscesses. Breast ultrasound and MRI revealed glandular edema and dilated breast ducts. The biopsies of all lesions exhibited both granulomatous inflammation confined to the lobules of the breast, abundant interstitial inflammatory cell infiltrates, and apparently cancerous cells located in dilated ducts with intact basement membranes. The surgically excised specimens confirmed the diagnosis of GLM and ductal carcinoma in situ (DCIS) in all three patients who underwent breast mass resection. By clinical imaging and clinical manifestations, GLM may obscure a concurrent DCIS, as highlighted by the cases reported herein.

Nonclinical factors associated with the treatment of older women with newly diagnosed low-grade ductal carcinoma in situ.

BACKGROUND: Ductal carcinoma in situ (DCIS) is the most common form of noninvasive breast cancer and is associated with an excellent prognosis. As a result, there is concern about overdiagnosis and overtreatment of DCIS because most patients with DCIS are treated as though they have invasive breast cancer and undergo either breast-conserving surgery (BCS)-most commonly followed by radiation therapy (RT)-or mastectomy. Little research to date has focused on nonclinical factors influencing treatments for DCIS. METHODS: Population-based data were analyzed from five state cancer registries (California, Florida, New Jersey, New York, and Texas) on women aged 65 years and older newly diagnosed with DCIS during the years 2003 to 2014 using a retrospective cohort design and multinominal logistic modeling. The registry records with Medicare enrollment data and fee-for-service claims to obtain treatments (BCS alone, BCS with RT, or mastectomy) were merged. Surgeon practice structure was identified through physician surveys and internet searches. RESULTS: Patients of surgeons employed by cancer centers or health systems were less likely to receive BCS with RT or mastectomy than patients of surgeons in single specialty or multispecialty practices. There also was substantial geographic variation in treatments, with patients in New York, New Jersey, and California being less likely to receive BCS with RT or mastectomy than patients in Texas or Florida. CONCLUSIONS: These findings suggest nonclinical factors including the culture of the practice and/or financial incentives are significantly associated with the types of treatment received for DCIS. Increasing awareness and targeted efforts to educate physicians about DCIS management among older women with low-grade DCIS could reduce patient harm and yield substantial cost savings.

Atypical ductal hyperplasia on vacuum-assisted breast biopsy: a scoring system to predict the risk of upgrade to malignancy.

RATIONALE AND OBJECTIVES: Our multicentric study analysed clinical, radiologic and pathologic features in patients with atypical ductal hyperplasia (ADH) diagnosed with vacuum-assisted biopsy (VAB), to identify factors associated with the risk of upgrade, to develop a scoring system to support decision making. MATERIALS AND METHODS: Patients with ADH on VAB under stereotactic/tomosynthesis guidance (2012-2022) were eligible. Inclusion criteria were availability of surgical histopathological examination of the entire lesion or radiologic follow-up (FUP) ≥ 24 months. VAB results were compared with surgical pathological results or with imaging FUP evolution to assess upgrade. A backward stepwise linear regression was used to identify predictors of upgrade. The discriminatory power of the model was calculated through the area under the receiver operating curve (ROC-AUC); the Hosmer-Lemeshow test was used to assess model calibration. The points system was developed based on the selected risk factors, and the probability of upgrade associated with each point total was determined. RESULTS: 112 ADH lesions were included: 91 (91/112, 81.3%) underwent surgical excision with 20 diagnosis of malignancy, while 21 (21/112, 18.7%) underwent imaging FUP with one interval change (mean FUP time 48 months). Overall upgrade rate was 18.7% (21/112). Age, menopausal status, concurrent breast cancer, BIRADS classification and number of foci of ADH were identified as risk factors for upgrade. Our model showed an AUC = 0.85 (95% CI 0.76-0.94). The points system showed that the risk of upgrade is < 2% when the total score is ≤ 1. CONCLUSION: Our scoring system seemed a promising easy-to-use decision support tool for management of ADH, decreasing unnecessary surgeries, reducing patients' overtreatment and healthcare costs.

Outcomes of High-Risk Breast MRI Screening in Women Without Prior History of Breast Cancer: Effectiveness Data from a Tertiary Care Center.

OBJECTIVE: To evaluate the diagnostic performance outcomes of a breast MRI screening program in high-risk women without prior history of breast cancer. METHODS: Retrospective cohort study of 1 405 consecutive screening breast MRI examinations in 681 asymptomatic women with high risk of breast cancer without prior history of breast cancer from January 1, 2015, to December 31, 2019. Outcomes (sensitivity, specificity, positive predictive value, negative predictive value, false-negative rate [FNR], cancer detection rate [CDR]) and characteristics of cancers were determined based on histopathology or 12-month follow-up. MRI examinations performed, BI-RADS assessments, pathology outcomes, and CDRs were analyzed overall and by age decade. Results in incidence screening round (MRI in last 18 months) and nonincidence round were compared. RESULTS: Breast MRI achieved CDR 20/1000, sensitivity 93.3% (28/30), and specificity 83.4% (1 147/1375). Twenty-eight (28/1 405, CDR 20/1000) screen-detected cancers were identified: 18 (64.3%, 18/28) invasive and 10 (35.7%, 10/28) ductal carcinoma in situ. Overall, 92.9% (26/28) of all cancers were stage 0 or 1 and 89.3% (25/28) were node negative. All 14 incidence screening round malignancies were stage 0 or 1 with N0 disease. Median size for invasive carcinoma was 8.0 mm and for ductal carcinoma in situ was 9.0 mm. There were two false-negative exams for an FNR 0.1% (2/1 405). CONCLUSION: High-risk screening breast MRI was effective at detecting early breast cancer and associated with favorable outcomes.

Cytomorphological Assessment in Aspirates of Ductal Carcinoma in situ: Correlations with Histopathologic Grade, Architectural Pattern, and Invasion.

INTRODUCTION: Fine-needle aspiration biopsy (FNAB) of the breast is an effective and widely adopted diagnostic technique. Histopathologic grading of ductal carcinoma in situ (DCIS) has prognostic significance. In this current study, FNAB of DCIS was reviewed to identify parameters that predict grading, histopathologic architecture, and presence of invasion in DCIS. METHODS: Aspirates from histopathology-proven cases of DCIS were retrieved and reviewed for cytomorphologic parameters including cellularity, composition, epithelial fragment architecture cellular/nuclear features. RESULTS: In total 104 aspirates were reviewed. Cytopathologic cellular features - large nuclear size (p = 0.005), prominent nucleoli (p = 0.011), increased nuclear membrane irregularity (p = 0.043), high variation in nuclear size (p = 0.025), and presence of apoptotic figures in epithelial structures (p < 0.001); and background debris (p = 0.033) correlated with a high-grade diagnosis. Cytoplasmic vacuolation (p = 0.034) was seen exclusively in non-high-grade aspirates. Epithelial fragment architecture did not correlate with grading. A predominance (≥50%) of solid aggregates and papillary fragments on FNAB correlated with histopathologically solid (p = 0.039, p = 0.005) and papillary (p = 0.029, < p = 0.001) patterns. No parameter showed correlation with invasion. CONCLUSION: FNAB is effective in predicting DCIS grading. Epithelial fragment architecture assessment is limited to papillary or solid types, and FNAB cannot predict focal invasion in DCIS.

Mode of Detection of Second Breast Cancers in Patients Undergoing Surveillance After Treatment of Ductal Carcinoma in Situ.

BACKGROUND: For patients undergoing posttreatment surveillance after ductal carcinoma in situ (DCIS), the NCCN Guidelines for Breast Cancer recommend annual breast imaging and physical examination every 6 to 12 months for 5 years, and then annually. The aim of our study was to evaluate the modes of detection (imaging, patient reported, or physical examination) of second cancers in a cohort of patients undergoing surveillance after primary DCIS treatment to better inform surveillance recommendations. METHODS: We performed a retrospective cohort study of patients with DCIS treated between January 1, 2008, and December 31, 2011, within a large integrated health care system. Information on patient demographics, index DCIS treatment, tumor characteristics, and mode of detection of second breast cancer was obtained from the electronic health record or chart review. RESULTS: Our study cohort consisted of 1,550 women, with a median age of 59 years at diagnosis. Surgical treatment of DCIS included lumpectomy (75.0%; n=1,162), unilateral mastectomy (21.1%; n=327), or bilateral mastectomy (3.9%; n=61), with or without sentinel lymph node biopsy. Additionally, 44.4% (n=688) and 28.3% (n=438) received radiation and endocrine therapies, respectively. Median follow-up was 10 years, during which 179 (11.5%) women were diagnosed with a second breast cancer. Of the second cancers, 43.0% (n=77) were ipsilateral and 54.8% (n=98) contralateral, and 2.2% (n=4) presented with distant metastases; 61.5% (n=110) were invasive, 36.3% (n=65) were DCIS, and 2.2% (n=4) were Paget's disease. Second breast cancers were imaging-detected in 74.3% (n=133) of cases, patient-detected in 20.1% (n=36), physician-detected in 2.2% (n=4), and detected incidentally on imaging or pathology from procedures unrelated to oncologic care in 3.4% (n=6). CONCLUSIONS: In our cohort of patients undergoing surveillance following diagnosis and treatment of DCIS, 2% of second breast cancers were detected by a clinical breast examination. This suggests that survivorship care should prioritize mammography and patient education regarding breast self-examination and symptoms that warrant evaluation to detect second breast cancers.

KRT15 in early breast cancer screening and correlation with HER2 positivity, pathological grade and N stage.

Objective: This study was designed to explore KRT15 dysregulation and its correlation with clinical characteristics among ductal carcinoma in situ (DCIS), DCIS with microinvasion (DCIS-MI) and invasive breast cancer (IBC) patients. Methods: KRT15 from lesion samples of 50 DCIS patients, 48 DCIS-MI patients and 50 IBC patients was detected by immunohistochemistry. Results: KRT15 discriminated IBC patients from DCIS patients (area under the curve [AUC] = 0.895; 95% CI = 0.836-0.954) and DCIS-MI patients (AUC = 0.707; 95% CI = 0.606-0.808). In DCIS patients, KRT15 was negatively correlated with pathological grade (p = 0.015). In DCIS-MI patients, KRT15 was positively related to estrogen receptor positivity but negatively associated with Ki-67 (both p < 0.05). In IBC patients, KRT15 was negatively linked to HER2 positivity, histological grade, N stage and tumor node metastasis stage (all p < 0.05). Conclusion: KRT15 assessment may help with early breast cancer screening.

Recruiting women with ductal carcinoma in situ to a randomised controlled trial: lessons from the LORIS study.

BACKGROUND: The LOw RISk DCIS (LORIS) study was set up to compare conventional surgical treatment with active monitoring in women with ductal carcinoma in situ (DCIS). Recruitment to trials with a surveillance arm is known to be challenging, so strategies to maximise patient recruitment, aimed at both patients and recruiting centres, were implemented. METHODS: Women aged ≥ 46 years with a histologically confirmed diagnosis of non-high-grade DCIS were eligible for 1:1 randomisation to either surgery or active monitoring. Prior to randomisation, all eligible women were invited to complete: (1) the Clinical Trials Questionnaire (CTQ) examining reasons for or against participation, and (2) interviews exploring in depth opinions about the study information sheets and film. Women agreeing to randomisation completed validated questionnaires assessing health status, physical and mental health, and anxiety levels. Hospital site staff were invited to communication workshops and refresher site initiation visits to support recruitment. Their perspectives on LORIS recruitment were collected via surveys and interviews. RESULTS: Eighty percent (181/227) of eligible women agreed to be randomised. Over 40% of participants had high anxiety levels at baseline. On the CTQ, the most frequent most important reasons for accepting randomisation were altruism and belief that the trial offered the best treatment, whilst worries about randomisation and the influences of others were the most frequent most important reasons for declining. Most women found the study information provided clear and useful. Communication workshops for site staff improved knowledge and confidence but only about half said they themselves would join LORIS if eligible. The most common recruitment barriers identified by staff were low numbers of eligible patients and patient preference. CONCLUSIONS: Recruitment to LORIS was challenging despite strategies aimed at both patients and site staff. Ensuring that recruiting staff support the study could improve recruitment in similar future trials. TRIAL REGISTRATION: ISRCTN27544579, prospectively registered on 22 May 2014.

Clinical presentations and outcomes of young women aged <40 yrs with Ductal carcinoma in situ.

INTRODUCTION: Ductal carcinoma in situ (DCIS) in young women is rare and not well studied. Since they do not qualify for mammographic screening, they are more likely to present with symptoms. Young women have also been associated with poorer outcomes, but it is unknown whether presentation mode affects outcome. We aimed to compare characteristics of DCIS patients <40 years of age presenting with symptoms versus those without, and determine whether presentation mode affects recurrence. METHODS: Pure DCIS patients aged <40 years were retrospectively analyzed. Clinical presentation, pathology and recurrence data were collected. Statistical analysis was performed to investigate the correlation of presentation mode with outcomes. RESULTS: 40 patients with 41 cases were included. The mean age at diagnosis was 32.3 years (range 17-39). 73.2% and 26.8% presented with symptoms or abnormal imaging respectively. Of the cases who presented with symptoms, 86.7%, 10.0% and 3.3% had palpable lump, nipple discharge or breast pain, respectively. The average tumor size was 22.0 mm (range 2.0-86.9) and 12.2 mm (range 3-25) for patients who presented with symptoms and non-symptomatic group, respectively. Cases presenting with symptoms were statistically associated with higher grade (p = 0.0090). On median follow-up of 85 months, there were 3 (7.3%) recurrences, which were not statistically associated with presentation mode. CONCLUSION: Young women with DCIS tend to present with symptoms, with breast lump as the commonest symptom. Symptomatic patients tend to be associated with grade III tumours, compared to non-symptomatic patients. On long-term follow-up, mode of presentation was not statistically associated with recurrence.

Incidental DCIS within a fibroadenoma.

A female in her early 20s was referred to the breast-endocrine surgeons with a self-detected tender left breast lump on the background of a family history of breast cancer. A physical examination revealed a rubbery and mobile mass in the left upper breast. Ultrasound demonstrated a solid hypoechoic mass with a likely differential diagnosis of fibroadenoma, with a subsequent core needle biopsy (CNB) confirming a fibroadenoma. Given the size and tenderness of the lump, an excisional biopsy was performed. Histology revealed a fibroadenoma with components of low-grade ductal carcinoma in situ, contained within the fibroadenoma and excised with clear margins.Following surgical excision, a multidisciplinary review determined that no further local therapy was required and recommended a genetics referral. This case was interesting as it raised important questions, including what the best surveillance strategies are for female patients with breast cancer within fibroadenoma and determining the risk and probability of missing epithelial atypia via CNB.