Application of deep learning on mammographies to discriminate between low and high-risk DCIS for patient participation in active surveillance trials.

BACKGROUND: Ductal Carcinoma In Situ (DCIS) can progress to invasive breast cancer, but most DCIS lesions never will. Therefore, four clinical trials (COMET, LORIS, LORETTA, AND LORD) test whether active surveillance for women with low-risk Ductal carcinoma In Situ is safe (E. S. Hwang et al., BMJ Open, 9: e026797, 2019, A. Francis et al., Eur J Cancer. 51: 2296-2303, 2015, Chizuko Kanbayashi et al. The international collaboration of active surveillance trials for low-risk DCIS (LORIS, LORD, COMET, LORETTA),  L. E. Elshof et al., Eur J Cancer, 51, 1497-510, 2015). Low-risk is defined as grade I or II DCIS. Because DCIS grade is a major eligibility criteria in these trials, it would be very helpful to assess DCIS grade on mammography, informed by grade assessed on DCIS histopathology in pre-surgery biopsies, since surgery will not be performed on a significant number of patients participating in these trials. OBJECTIVE: To assess the performance and clinical utility of a convolutional neural network (CNN) in discriminating high-risk (grade III) DCIS and/or Invasive Breast Cancer (IBC) from low-risk (grade I/II) DCIS based on mammographic features. We explored whether the CNN could be used as a decision support tool, from excluding high-risk patients for active surveillance. METHODS: In this single centre retrospective study, 464 patients diagnosed with DCIS based on pre-surgery biopsy between 2000 and 2014 were included. The collection of mammography images was partitioned on a patient-level into two subsets, one for training containing 80% of cases (371 cases, 681 images) and 20% (93 cases, 173 images) for testing. A deep learning model based on the U-Net CNN was trained and validated on 681 two-dimensional mammograms. Classification performance was assessed with the Area Under the Curve (AUC) receiver operating characteristic and predictive values on the test set for predicting high risk DCIS-and high-risk DCIS and/ or IBC from low-risk DCIS. RESULTS: When classifying DCIS as high-risk, the deep learning network achieved a Positive Predictive Value (PPV) of 0.40, Negative Predictive Value (NPV) of 0.91 and an AUC of 0.72 on the test dataset. For distinguishing high-risk and/or upstaged DCIS (occult invasive breast cancer) from low-risk DCIS a PPV of 0.80, a NPV of 0.84 and an AUC of 0.76 were achieved. CONCLUSION: For both scenarios (DCIS grade I/II vs. III, DCIS grade I/II vs. III and/or IBC) AUCs were high, 0.72 and 0.76, respectively, concluding that our convolutional neural network can discriminate low-grade from high-grade DCIS.

Clinician Burnout Associated With Sex, Clinician Type, Work Culture, and Use of Electronic Health Records.

Importance: Electronic health records (EHRs) are considered a potentially significant contributor to clinician burnout. Objective: To describe the association of EHR usage, sex, and work culture with burnout for 3 types of clinicians at an academic medical institution. Design, Setting, and Participants: This cross-sectional study of 1310 clinicians at a large tertiary care academic medical center analyzed EHR usage metrics for the month of April 2019 with results from a well-being survey from May 2019. Participants included attending physicians, advanced practice providers (APPs), and house staff from various specialties. Data were analyzed between March 2020 and February 2021. Exposures: Clinician demographic characteristics, EHR metadata, and an institution-wide survey. Main Outcomes and Measures: Study metrics included clinician demographic data, burnout score, well-being measures, and EHR usage metadata. Results: Of the 1310 clinicians analyzed, 542 (41.4%) were men (mean [SD] age, 47.3 [11.6] years; 448 [82.7%] White clinicians, 52 [9.6%] Asian clinicians, and 21 [3.9%] Black clinicians) and 768 (58.6%) were women (mean [SD] age, 42.6 [10.3] years; 573 [74.6%] White clinicians, 105 [13.7%] Asian clinicians, and 50 [6.5%] Black clinicians). Women reported more burnout (survey score ≥50: women, 423 [52.0%] vs men, 258 [47.6%]; P = .008) overall. No significant differences in EHR usage were found by sex for multiple metrics of time in the EHR, metrics of volume of clinical encounters, or differences in products of clinical care. Multivariate analysis of burnout revealed that work culture domains were significantly associated with self-reported results for commitment (odds ratio [OR], 0.542; 95% CI, 0.427-0.688; P < .001) and work-life balance (OR, 0.643; 95% CI, 0.559-0.739; P < .001). Clinician sex significantly contributed to burnout, with women having a greater likelihood of burnout compared with men (OR, 1.33; 95% CI, 1.01-1.75; P = .04). An increased number of days spent using the EHR system was associated with less likelihood of burnout (OR, 0.966; 95% CI, 0.937-0.996; P = .03). Overall, EHR metrics accounted for 1.3% of model variance (P = .001) compared with work culture accounting for 17.6% of variance (P < .001). Conclusions and Relevance: In this cross-sectional study, sex-based differences in EHR usage and burnout were found in clinicians. These results also suggest that local work culture factors may contribute more to burnout than metrics of EHR usage.

Hymenolepis diminuta-based helminth therapy in C3(1)-TAg mice does not alter breast tumor onset or progression.

BACKGROUND AND OBJECTIVES: An individual's risk of breast cancer is profoundly affected by evolutionary mismatch. Mismatches in Western society known to increase the risk of breast cancer include a sedentary lifestyle and reproductive factors. Biota alteration, characterized by a loss of biodiversity from the ecosystem of the human body as a result of Western society, is a mismatch known to increase the risk of a variety of inflammation-related diseases, including colitis-associated colon cancer. However, the effect of biota alteration on breast cancer has not been evaluated. METHODOLOGY: In this study, we utilized the C3(1)-TAg mouse model of breast cancer to evaluate the role of biota alteration in the development of breast cancer. This model has been used to recapitulate the role of exercise and pregnancy in reducing the risk of breast cancer. C3(1)-TAg mice were treated with Hymenolepis diminuta, a benign helminth that has been shown to reverse the effects of biota alteration in animal models. RESULTS: No effect of the helminth H. diminuta was observed. Neither the latency nor tumor growth was affected by the therapy, and no significant effects on tumor transcriptome were observed based on RNAseq analysis. CONCLUSIONS AND IMPLICATIONS: These findings suggest that biota alteration, although known to affect a variety of Western-associated diseases, might not be a significant factor in the high rate of breast cancer observed in Western societies. LAY SUMMARY: An almost complete loss of intestinal worms in high-income countries has led to increases in allergic disorders, autoimmune conditions, and perhaps colon cancer. However, in this study, results using laboratory mice suggest that loss of intestinal worms might not be associated with breast cancer.

The impact of patient characteristics and lifestyle factors on the risk of an ipsilateral event after a primary DCIS: A systematic review.

OBJECTIVE: The majority of 'low-risk' (grade I/II) Ductal Carcinoma In Situ (DCIS) may not progress to invasive breast cancer during a women's lifetime. Therefore, the safety of active surveillance versus standard surgical treatment for DCIS is prospectively being evaluated in clinical trials. If proven safe and selectively implemented in clinical practice, a significant group of women with low-risk DCIS may forego surgery and radiotherapy in the future. Identification of modifiable and non-modifiable risk factors associated with prognosis after a primary DCIS would also enhance our care of women with low-risk DCIS. METHODS: To identify modifiable and non-modifiable risk factors for subsequent breast events after DCIS, we performed a systematic literature search in PUBMED, EMBASE and Scopus. RESULTS: Six out of the 3870 articles retrieved were included for final data extraction. These six studies included a total of 4950 patients with primary DCIS and 640 recorded subsequent breast events. There was moderate evidence for an association of a family history of breast cancer, premenopausal status, high BMI, and high breast density with a subsequent breast cancer or further DCIS. CONCLUSION: There is a limited number of recent studies published on the impact of modifiable and non-modifiable risk factors on subsequent events after DCIS. The available evidence is insufficient to identify potential targets for risk reduction strategies, reflecting the relatively small numbers and the lack of long-term follow-up in DCIS, a low-event condition.

Epidemiology, Biology, Treatment, and Prevention of Ductal Carcinoma In Situ (DCIS).

Ductal carcinoma in situ (DCIS) is a highly heterogeneous disease. It presents in a variety of ways and may or may not progress to invasive cancer, which poses challenges for both diagnosis and treatment. On May 15, 2017, the Dana-Farber/Harvard Cancer Center hosted a retreat for over 80 breast specialists including medical oncologists, surgical oncologists, radiation oncologists, radiologists, pathologists, physician assistants, nurses, nurse practitioners, researchers, and patient advocates to discuss the state of the science, treatment challenges, and key questions relating to DCIS. Speakers and attendees were encouraged to explore opportunities for future collaboration and research to improve our understanding and clinical management of this disease. Participants were from Dana-Farber Cancer Institute, Brigham and Women's Hospital, Massachusetts General Hospital, Beth Israel Deaconess Medical Center, Duke University Medical Center, and MD Anderson Cancer Center. The discussion focused on three main themes: epidemiology, detection, and pathology; state of the science including the biology of DCIS and potential novel treatment approaches; and risk perceptions, communication, and decision-making. Here we summarize the proceedings from this event.

Effect of Blood Collection Tube Type and Time to Processing on the Enumeration and High-Content Characterization of Circulating Tumor Cells Using the High-Definition Single-Cell Assay.

CONTEXT: - As circulating tumor cell (CTC) assays gain clinical relevance, it is essential to address preanalytic variability and to develop standard operating procedures for sample handling in order to successfully implement genomically informed, precision health care. OBJECTIVE: - To evaluate the effects of blood collection tube (BCT) type and time-to-assay (TTA) on the enumeration and high-content characterization of CTCs by using the high-definition single-cell assay (HD-SCA). DESIGN: - Blood samples of patients with early- and advanced-stage breast cancer were collected into cell-free DNA (CfDNA), EDTA, acid-citrate-dextrose solution, and heparin BCTs. Time-to-assay was evaluated at 24 and 72 hours, representing the fastest possible and more routine domestic shipping intervals, respectively. RESULTS: - We detected the highest CTC levels and the lowest levels of negative events in CfDNA BCT at 24 hours. At 72 hours in this BCT, all CTC subpopulations were decreased with the larger effect observed in high-definition CTCs and cytokeratin-positive cells smaller than white blood cells. Overall cell retention was also optimal in CfDNA BCT at 24 hours. Whole-genome copy number variation profiles were generated from single cells isolated from all BCT types and TTAs. Cells from CfDNA BCT at 24-hour TTA exhibited the least noise. CONCLUSIONS: - Circulating tumor cells can be identified and characterized under a variety of collection, handling, and processing conditions, but the highest quality can be achieved with optimized conditions. We quantified performance differences of the HD-SCA for specific preanalytic variables that may be used as a guide to develop best practices for implementation into patient care and/or research biorepository processes.

Intraoperative frozen section analysis of sentinel lymph nodes in breast cancer patients: a meta-analysis and single-institution experience.

BACKGROUND: Accurate intraoperative pathologic examination of sentinel lymph nodes (SLNs) has been an important tool that can reduce the need for reoperations in patients with SLN-positive breast cancer. The objective of the current study was to determine the accuracy of intraoperative frozen section (IFS) of SLNs during breast cancer surgery. METHODS: The authors retrospectively reviewed the records of 326 patients with breast cancer who underwent IF analysis of SLNs at a single institution. Then, they conducted a meta-analysis that included 47 published studies of IFS of SLNs in patients with breast cancer. RESULTS: Hematoxylin and eosin (H&E) staining revealed metastasis in SLNs in 99 patients (30.4%), including 61 patients with macrometastasis (MAM) (>2 mm) (the MAM group) and 38 patients with micrometastasis (Mi) or isolated tumor cell (ITC) deposits (the Mi/ITC group). The overall sensitivity of the institutional series was 60.6% (60 of 99 patients), and overall specificity was 100% (227 of 227 true negatives). The sensitivity of IFS was significantly lower in the Mi/ITC group (28.9%) than in the MAM group (80.3%; P < .0001). According to the meta-analysis of published studies and data from the author's institution (47 studies, for a total of 13,062 patients who underwent SLN dissection with IFS of SLNs), the mean sensitivity was 73%, and the mean specificity was 100%. The mean sensitivity was 94% for the MAM group and 40% for the Mi/ITC group. CONCLUSIONS: IFS of SLNs was more reliable for detecting MAM than for detecting Mi/ITC deposits. It lacked sufficient accuracy to rule out Mi/ITC deposits.

Prognostic implications of positive nonsentinel lymph nodes removed during selective sentinel lymphadenectomy for breast cancer.

Nonsentinel lymph nodes (SLNs) are commonly removed at the time of selective sentinel lymphadenectomy (SSL). Their predictive value for the rest of the nodal basin is unknown. A retrospective review of 436 breast cancer patients who underwent SSL between 12/97 and 04/03 at a single institution. One-hundred nineteen patients had non-SLNs removed at SSL; eight were positive (6.7%). Positive non-SLNs predicted that SLNs would also be positive (p = 0.008). There was no difference in rates of additional positive nodes found on completion axillary node dissection between the non-SLN and SLN positive patients (p = 0.62). After adjustment for covariates, the presence of positive non-SLNs was not associated with poorer disease free survival (p = 0.24), time to systemic recurrence (p = 0.57), or overall survival (p = 0.70). Positive non-SLNs removed during SSL are not a significant risk factor for additional positive nodes on completion axillary nodal dissection (CALND) or for worse survival than positive SLNs.

Is it necessary to harvest additional lymph nodes after resection of the most radioactive sentinel lymph node in breast cancer?

BACKGROUND: No consensus exists about the number of sentinel lymph nodes (SLNs) that should be removed based on radioactivity counts in breast cancer, although the "10% rule" is often used. We hypothesized that the node with the highest radioactivity would have the strongest probability of being a positive SLN, and we sought to determine the lowest radioactive count of a node harboring cancer. STUDY DESIGN: We retrospectively studied 332 breast cancer patients who underwent lymphoscintigraphy by injection of technetium 99m-labeled thiosulfate colloid and sentinel lymphadenectomy (SL) between 1997 and 2006, with intraoperative determination of radioactive counts of nodes by a gamma probe. All SLNs were examined by permanent sections consisting of at least 3 levels of 40- to 100-mum intervals for hematoxylin and eosin evaluation, with or without immunohistochemical staining for cytokeratins. RESULTS: Seventy-four percent of patients had more than 1 SLN removed (mean 2.8 per patient); 23.5% had SLN metastasis. Of the node-positive patients, the hottest SLN was positive in 85.9% (67 of 78). Five of the 78 patients (6.4%) with positive nodes had counts less than 10% of those of the hottest node. The lowest radioactive count of a positive SLN was 4.2% of that of the hottest node. Lymphatic mapping based on the 10% rule could greatly improve the false-negative rates compared with removing only the hottest SLN (14.1% versus 6.4%). CONCLUSIONS: Most positive SLNs had the highest radioactivity. Our institutional experience indicates that to obtain an acceptable false-negative rate, nodes should be removed until the 10% rule is met.