Short partial doffs of release/relock sockets may effectively stabilize limb fluid volume in prosthesis users with transtibial amputation.

BACKGROUND: A challenge often faced by people with lower extremity amputation is management of prosthetic socket fit due to changes in fluid volume within their residual limb. Prior research suggests that intermittently doffing the prosthetic socket may help stabilize daily residual limb fluid volume. METHODS: To assess the effects of partial doff duration on residual limb fluid volume retention, participants with transtibial amputation were tested by walking on a treadmill in a controlled, laboratory setting under three conditions. An automated system to release the locking pin and enlarge the socket was used to produce the partial doffing. Percent limb fluid volume changes after partial doffing for 4 min (Short Rest) and for 10 min (Long Rest) were compared with no partial doffing (No Release). Limb fluid volume was monitored using bioimpedance analysis. FINDINGS: Mean percent fluid volume changes in the posterior region were  -1.2% for No Release, 2.7% for Short Rest, and 1.0% for Long Rest. Short and Long Rests had larger increases than No Release (P = 0.005 and 0.03, respectively); Short and Long Rests were not statistically different (P = 0.10). Eight of the thirteen participants experienced a higher percent fluid volume gain for both release protocols while four experienced a higher percent fluid volume gain for only one release protocol. INTERPRETATION: A partial doff duration as short as 4 min may be an effective strategy to stabilize limb fluid volume in prosthesis users with transtibial amputation. Trials in at-home settings should be pursued.

Nipple-Sparing Mastectomy versus Skin-Sparing Mastectomy: Does Saving the Nipple Impact Short- and Long-Term Patient Satisfaction?

BACKGROUND: Nipple-sparing mastectomy (NSM) is an oncologically safe alternative to skin-sparing mastectomy (SSM). This study evaluated whether NSM patients were more satisfied than SSM patients in short- and long-term follow-up. METHODS: Women who underwent NSM or SSM between 2009 and 2019 completed a postoperative BREAST-Q survey at least 1 year after surgery and patient characteristics were compared. Patient satisfaction at 1-5 years and 6-10 years after NSM and SSM were analyzed. RESULTS: Overall, 431 patients were included; 247 had NSM and 184 had SSM 1-10 years prior to BREAST-Q survey completion. SSM patients were older, had higher body mass index (BMI), larger breast weight, and more hypertension than NSM patients, but oncologic treatments were similar between groups. BREAST-Q Psychosocial Well-Being and Sexual Well-Being scores were significantly higher in NSM patients compared with SSM patients in the 1-5 years cohort; however, scores attenuated in the 6-10 years cohort. Satisfaction with breasts was nearly significantly higher in NSM patients compared with SSM patients in the 1-5 years cohort (p = 0.056), but no different in the 6-10 years cohort. Receipt of adjuvant chemotherapy, receipt of postmastectomy radiation therapy, and BMI ≥30 were independent risk factors for dissatisfaction with breasts. CONCLUSIONS: Women who are not candidates for NSM should be reassured that long-term qualify of life is not significantly different between SSM and NSM. Dissatisfaction with reconstructed breasts is linked with other factors (besides the nipple), which patients should be made aware of at the time of surgical decision making.

Performance of a novel protease-activated fluorescent imaging system for intraoperative detection of residual breast cancer during breast conserving surgery.

PURPOSE: Safe breast cancer lumpectomies require microscopically clear margins. Real-time margin assessment options are limited, and 20-40% of lumpectomies have positive margins requiring re-excision. The LUM Imaging System previously showed excellent sensitivity and specificity for tumor detection during lumpectomy surgery. We explored its impact on surgical workflow and performance across patient and tumor types. METHODS: We performed IRB-approved, prospective, non-randomized studies in breast cancer lumpectomy procedures. The LUM Imaging System uses LUM015, a protease-activated fluorescent imaging agent that identifies residual tumor in the surgical cavity walls. Fluorescent cavity images were collected in real-time and analyzed using system software. RESULTS: Cavity and specimen images were obtained in 55 patients injected with LUM015 at 0.5 or 1.0 mg/kg and in 5 patients who did not receive LUM015. All tumor types were distinguished from normal tissue, with mean tumor:normal (T:N) signal ratios of 3.81-5.69. T:N ratios were 4.45 in non-dense and 4.00 in dense breasts (p = 0.59) and 3.52 in premenopausal and 4.59 in postmenopausal women (p = 0.19). Histopathology and tumor receptor testing were not affected by LUM015. Falsely positive readings were more likely when tumor was present < 2 mm from the adjacent specimen margin. LUM015 signal was stable in vivo at least 6.5 h post injection, and ex vivo at least 4 h post excision. CONCLUSIONS: Intraoperative use of the LUM Imaging System detected all breast cancer subtypes with robust performance independent of menopausal status and breast density. There was no significant impact on histopathology or receptor evaluation.

Similar rates of residual disease in patients with DCIS within 2 mm of lumpectomy margin regardless of the presence of invasive carcinoma.

PURPOSE: The 2014 Society of Surgical Oncology/American Society for Radiation Oncology (SSO/ASTRO) breast-conserving surgery (BCS) margin guidelines for invasive cancer recommended "no ink on tumor" as an adequate margin width. However, 2016 SSO/ASTRO margin guidelines for pure DCIS recommended a 2 mm margin. Thus, management of a margin with DCIS > 0 mm but < 2 mm differs based on presence or absence of invasive carcinoma. We compared rates of residual disease in patients with pure DCIS to patients with invasive cancer with DCIS. METHODS: BCS with complete shaved cavity margins (SCM) for invasive carcinoma or pure DCIS from 2004 to 2006 at our institution was reviewed. Margin width was measured on the main specimen and the presence of carcinoma in the SCM was used as a surrogate for residual disease in the cavity. Rates of residual disease were determined for varying margin widths of invasive carcinoma and DCIS. RESULTS: Of 329 BCS patients, 123 (37%) patients had pure DCIS and 206 (63%) had invasive cancer with DCIS. In the pure DCIS cohort, 61 patients had DCIS between 0 and 2 mm from the inked margin; 32 (52%) of which had residual disease in the SCM. In the invasive cancer plus DCIS cohort, 92 had DCIS between 0 and 2 mm from the inked margin; 39 (42%) of which had residual disease in the SCM (p = 0.221). CONCLUSION: Rates of residual disease are similar in patients treated with lumpectomy for pure DCIS and those with invasive carcinoma with DCIS when DCIS is found between 0 and 2 mm from the inked margin.

Exploiting Rules to Enhance Machine Learning in Extracting Information From Multi-Institutional Prostate Pathology Reports.

PURPOSE: Literature on clinical note mining has highlighted the superiority of machine learning (ML) over hand-crafted rules. Nevertheless, most studies assume the availability of large training sets, which is rarely the case. For this reason, in the clinical setting, rules are still common. We suggest 2 methods to leverage the knowledge encoded in pre-existing rules to inform ML decisions and obtain high performance, even with scarce annotations. METHODS: We collected 501 prostate pathology reports from 6 American hospitals. Reports were split into 2,711 core segments, annotated with 20 attributes describing the histology, grade, extension, and location of tumors. The data set was split by institutions to generate a cross-institutional evaluation setting. We assessed 4 systems, namely a rule-based approach, an ML model, and 2 hybrid systems integrating the previous methods: a Rule as Feature model and a Classifier Confidence model. Several ML algorithms were tested, including logistic regression (LR), support vector machine (SVM), and eXtreme gradient boosting (XGB). RESULTS: When training on data from a single institution, LR lags behind the rules by 3.5% (F1 score: 92.2% v 95.7%). Hybrid models, instead, obtain competitive results, with Classifier Confidence outperforming the rules by +0.5% (96.2%). When a larger amount of data from multiple institutions is used, LR improves by +1.5% over the rules (97.2%), whereas hybrid systems obtain +2.2% for Rule as Feature (97.7%) and +2.6% for Classifier Confidence (98.3%). Replacing LR with SVM or XGB yielded similar performance gains. CONCLUSION: We developed methods to use pre-existing handcrafted rules to inform ML algorithms. These hybrid systems obtain better performance than either rules or ML models alone, even when training data are limited.

Correction to: Feasibility Study of a Novel Protease-Activated Fluorescent Imaging System for Real-Time, Intraoperative Detection of Residual Breast Cancer in Breast Conserving Surgery.

The article Feasibility Study of a Novel Protease-Activated Fluorescent Imaging System for Real-Time, Intraoperative Detection of Residual Breast Cancer in Breast Conserving Surgery, written by Barbara L. Smith et al., was originally published electronically on the publisher's internet portal on January 2, 2020, without open access.

Feasibility Study of a Novel Protease-Activated Fluorescent Imaging System for Real-Time, Intraoperative Detection of Residual Breast Cancer in Breast Conserving Surgery.

BACKGROUND: Obtaining tumor-free margins is critical to prevent recurrence after lumpectomy for breast cancer. Unfortunately, current approaches leave positive margins that require second surgeries in 20-40% of patients. We assessed the LUM Imaging System for real-time, intraoperative detection of residual tumor. METHODS: Breast lumpectomy cavity walls and excised specimens were assessed with the LUM Imaging System after 1 mg/kg intravenous LUM015, a protease-activatable fluorescent agent. Fluorescence at potential sites of residual tumor in lumpectomy cavity walls was evaluated intraoperatively with a sterile hand-held probe, with real-time predictive results displayed on a monitor intraoperatively, and later correlated with histopathology. RESULTS: In vivo lumpectomy cavities and excised specimens were imaged after LUM015 injection in 45 women undergoing breast cancer surgery. Invasive ductal and lobular cancers and intraductal cancer (DCIS) were included. A total of 570 cavity margin surfaces in 40 patients were used for algorithm development. Image analysis and display took approximately 1 s per 2.6-cm-diameter circular margin surface. All breast cancer subtypes could be distinguished from adjacent normal tissue. For all imaged cavity surfaces, sensitivity for tumor detection was 84%. Among 8 patients with positive margins after standard surgery, sensitivity for residual tumor detection was 100%; 2 of 8 were spared second surgeries because additional tissue was excised at sites of LUM015 signal. Specificity was 73%, with some benign tissues showing elevated fluorescent signal. CONCLUSIONS: The LUM015 agent and LUM Imaging System allow rapid identification of residual tumor in the lumpectomy cavity of breast cancer patients and may reduce rates of positive margins.

Long-Term Outcomes of Multiple-Wire Localizations for More Extensive Breast Cancer: Multiple-Wire Excision Does Not Increase Recurrence, Unplanned Imaging, or Biopsies.

BACKGROUND: We previously reported that breast conservation was feasible for women with large or irregularly shaped breast cancers when tumor resection was guided by multiple localizing wires. We now report long-term outcomes of multiple-wire versus single-wire localized lumpectomies for breast cancer. PATIENTS AND METHODS: We retrospectively reviewed wire-localized lumpectomies at our institution from May 2000 to November 2006. Rates of ipsilateral in-breast tumor recurrence, metastasis, and subsequent unplanned diagnostic imaging and biopsy were compared between multiple-wire and single-wire cohorts. RESULTS: We identified 112 multiple-wire and 160 single-wire breast cancer lumpectomies that achieved clear margins. Median age was 64 years in the multiple-wire cohort and 57 years in the single-wire cohort. Mean lumpectomy volume was 75 mL in multiple-wire patients and 49 mL in single-wire patients (P = .003). Invasive tumor size, axillary node status, and use of radiation and systemic therapy were similar, but the multiple-wire group had more patients with ductal carcinoma-in-situ only (38% vs. 28%). At 108 months' median follow-up, there was no significant difference in local or distant recurrence rates between multiple-wire and single-wire cohorts. Six (5%) multiple-wire patients and 6 (4%) single-wire patients had local recurrences and 3 (3%) multiple-wire and 5 (3%) single-wire patients developed metastatic disease. Unplanned diagnostic imaging was required for 53 (47%) multiple-wire and 65 (41%) single-wire patients. Subsequent ipsilateral biopsy occurred in 15 (13%) multiple-wire and 19 (12%) single-wire patients. CONCLUSION: Breast-conserving surgery with multiple localizing wires is a safe alternative to mastectomy for breast cancer patients with large mammographic lesions.

Do Neural Information Extraction Algorithms Generalize Across Institutions?

PURPOSE: Natural language processing (NLP) techniques have been adopted to reduce the curation costs of electronic health records. However, studies have questioned whether such techniques can be applied to data from previously unseen institutions. We investigated the performance of a common neural NLP algorithm on data from both known and heldout (ie, institutions whose data were withheld from the training set and only used for testing) hospitals. We also explored how diversity in the training data affects the system's generalization ability. METHODS: We collected 24,881 breast pathology reports from seven hospitals and manually annotated them with nine key attributes that describe types of atypia and cancer. We trained a convolutional neural network (CNN) on annotations from either only one (CNN1), only two (CNN2), or only four (CNN4) hospitals. The trained systems were tested on data from five organizations, including both known and heldout ones. For every setting, we provide the accuracy scores as well as the learning curves that show how much data are necessary to achieve good performance and generalizability. RESULTS: The system achieved a cross-institutional accuracy of 93.87% when trained on reports from only one hospital (CNN1). Performance improved to 95.7% and 96%, respectively, when the system was trained on reports from two (CNN2) and four (CNN4) hospitals. The introduction of diversity during training did not lead to improvements on the known institutions, but it boosted performance on the heldout institutions. When tested on reports from heldout hospitals, CNN4 outperformed CNN1 and CNN2 by 2.13% and 0.3%, respectively. CONCLUSION: Real-world scenarios require that neural NLP approaches scale to data from previously unseen institutions. We show that a common neural NLP algorithm for information extraction can achieve this goal, especially when diverse data are used during training.

Nipple Discharge After Nipple-Sparing Mastectomy With and Without Associated Pregnancy.

BACKGROUND: Nipple-sparing mastectomies (NSMs) preserve the intact nipple, including nipple duct orifices. Retained orifices might remain patent and communicate with the underlying reconstruction. We report the incidence and outcomes of nipple discharge after NSM in pregnant and nonpregnant women. PATIENTS AND METHODS: Retrospective review of all NSMs at our institution from June 2007 to June 2018 was performed. Subsequent pregnancies and nipple discharge were documented. Patient demographic, operative, histopathology, and cancer treatment data were collected. Descriptive analysis was performed for patients who developed nipple discharge. RESULTS: From June 2007 to June 2018, 2778 NSM procedures were performed in 1620 patients, with a mean age of 48 (range, 20-80) years. Fifteen hundred sixty-eight NSMs were therapeutic and 1210 were for risk reduction. Thirty-three subsequent pregnancies were observed in 27 patients, with a mean age of 33 (range, 26-42) years at NSM. Bilateral or unilateral discharge occurred in 6 of 27 (22%) postpartum patients and resolved spontaneously. At 54 months mean follow-up after NSM (range, 16-98 months) and 23 (range, 1-61) months after delivery, no local-regional recurrences were observed. In 1593 patients without subsequent pregnancy, there were 4 patients (0.25%) treated with bilateral NSM with subsequent unilateral watery nipple discharge. There was no evidence of associated malignancy on physical exam, imaging, or cytology, and with 55 to 110 months follow-up, no new or recurrent cancers have been observed. CONCLUSION: Despite extensive removal of nipple and subareolar duct tissue during NSM, milky nipple discharge is possible postpartum. Watery, acellular discharge occurs rarely in nonpregnant patients. To date, no patient with discharge has developed a local recurrence or new breast cancer.

Real-time, intraoperative detection of residual breast cancer in lumpectomy cavity walls using a novel cathepsin-activated fluorescent imaging system.

PURPOSE: Obtaining tumor-free surgical margins is critical to prevent recurrence in breast-conserving surgery but it remains challenging. We assessed the LUM Imaging System for real-time, intraoperative detection of residual tumor. METHODS: Lumpectomy cavity walls and excised specimens of breast cancer lumpectomy patients were assessed with the LUM Imaging System (Lumicell, Inc., Wellesley MA) with and without intravenous LUM015, a cathepsin-activatable fluorescent agent. Fluorescence at potential sites of residual tumor was evaluated with a sterile hand-held probe, displayed on a monitor and correlated with histopathology. RESULTS: Background autofluorescence was assessed in excised specimens from 9 patients who did not receive LUM015. In vivo lumpectomy cavities and excised specimens were then imaged in 15 women undergoing breast cancer surgery who received no LUM015, 0.5, or 1 mg/kg LUM015 (5 women per dose). Among these, 11 patients had invasive carcinoma with ductal carcinoma in situ (DCIS) and 4 had only DCIS. Image acquisition took 1 s for each 2.6-cm-diameter surface. No significant background normal breast fluorescence was identified. Elevated fluorescent signal was seen from invasive cancers and DCIS. Mean tumor-to-normal signal ratios were 4.70 ± 1.23 at 0.5 mg/kg and 4.22 ± 0.9 at 1.0 mg/kg (p = 0.54). Tumor was distinguished from normal tissue in pre-and postmenopausal women and readings were not affected by breast density. Some benign tissues produced fluorescent signal with LUM015. CONCLUSION: The LUM Imaging System allows rapid identification of residual tumor in the lumpectomy cavity of breast cancer patients and may reduce rates of positive margins.