Opportunities to reduce reoperations and to improve inter-facility profiling after initial breast-conserving surgery for cancer. A report from the NCDB.

BACKGROUND: Repeat operations after breast-conserving surgery (BCS) for cancer have been termed "epidemic." To aid improvement activities, we sought to identify those National Cancer Data Base (NCDB) characteristics that were associated with reoperations. METHODS: A retrospective cohort of patients with invasive breast cancer undergoing initial BCS in the NCDB from 2004 to 2015 were identified. Univariate, multivariate, ranking (effect size and R2), and time-trend methods were used to assess associations between patient, facility, tumor, treatment, and calendar-year characteristics with reoperation. RESULTS: In 1226 facilities, 84,462 (16.1%) of 524,594 patients underwent reoperations after BCS [range 0-75%; 10th/90th performance percentiles = 6.6%/25%]. Of 18 factors associated with reoperations, facility ID was the highest-ranked. Its estimated impact on the odds of reoperation was more than 10 times greater than any other factor considered, followed by tumor size, neo-adjuvant chemotherapy receipt, patient age, cancer histology, and nodal status. Reoperations after the year of the SSO-ASTRO margin guideline declined significantly compared with prior years. Significant inter-facility reoperation variability persisted after risk adjustment for more than a dozen distinct patient, facility, tumor, and treatment characteristics. CONCLUSION: In the NCDB, significant inter-facility variability exists regardless of case volume, case mix, and risk adjustment. There were fewer reoperations after the SSO-ASTRO guideline. An endorsed target rate of 10% was achieved by only 1 in 4 facilities. The most impactful determinant of reoperation was the facility itself. Thus, all stakeholders should consider participation in improvement activities. Such activities will benefit from risk-adjusted profiling; the relevant adjustors were identified.

A Reappraisal of the Comparative Effectiveness of Lumpectomy Versus Mastectomy on Breast Cancer Survival: A Propensity Score-Matched Update From the National Cancer Data Base (NCDB).

BACKGROUND: Recent observational studies are concerning because they document rising mastectomy rates coinciding with more than a dozen reports that lumpectomy has better overall survival (OS) than mastectomy. Our aim was to determine if there were differences in OS of matched breast cancer patients undergoing lumpectomy versus mastectomy in the National Cancer Database (NCDB). PATIENTS AND METHODS: A retrospective cohort of patients with stage I-III breast cancer in the NCDB (2004-2013) was identified. Propensity score matching (PSM), Kaplan-Meier, and multivariate Cox proportional hazards models were used to examine OS by type of surgery. RESULTS: Of 845,136 patients, 464,052 (54.9%) underwent lumpectomy and 381,084 (45.1%) underwent mastectomy. After PSM, the hazard ratio (HR) and confidence interval (CI) for OS in all patients comparing lumpectomy with mastectomy was 1.02 (CI, 1.00-1.04; P = .002). In patients with stage I, II, and III, they were HR 1.27 (CI, 1.23-1.36; P < .001), HR 0.98 (CI, 0.95-1.01; P = .21), and HR 0.83 (CI, 0.80-0.86; P < .001), respectively. In subgroup analyses of all patients by estrogen receptor (ER) status, they were HR 1.05 (CI, 1.03-1.07; P < .001) and HR 1.00 (CI, 0.96-1.03; P = .65) in ER+ and ER- patients. CONCLUSION: In our primary model of all stage I-III matched patients, using the most recent NCDB data and the largest observational sample size to date, the OS after mastectomy was not inferior to lumpectomy. This finding can be reassuring to patients and providers. In subgroup analyses, the association between type of surgery and OS differed by cancer stage and hormone receptor status.

A Strategy for Changing Adherence to National Guidelines for Decreasing Laboratory Testing for Early Breast Cancer Patients.

INTRODUCTION: Past studies indicate delays in adoption of consensus-based guideline updates. In June 2016, the National Comprehensive Cancer Network changed its guidelines from routine testing to omission of ordering complete blood cell count (CBC) and liver function tests (LFT) in patients with early breast cancer. In response, we developed an implementation strategy to discontinue our historical practice of routine ordering of these tests in asymptomatic patients. METHODS: The ordering of CBC and LFT for clinical stage I-IIIA breast cancer patients was audited in 2016. In June 2016, we utilized the levers of the National Quality Strategy implementation methodology to enact a system-wide change to omit routine ordering. To measure the plan's effectiveness, guideline compliance for ordering was tracked continually. RESULTS: Of 92 patients with early stage cancer in 2016, the overall rate of compliance with guidelines for ordering a CBC and LFT was 82% (88/107) and 87% (93/107), respectively. Segregated by the pre- and post-guideline change time period, the compliance rates for ordering a CBC and LFT were 78% and 87% (P = 0.076). CONCLUSION: In contrast to historical reports of delays in adoption of new evidence-based guideline changes, we were able to quickly change provider practice during the transition from routine ordering to omission of ordering screening blood tests in newly diagnosed patients with early breast cancer.

Profiling Surgeon Performance for Breast Cancer Lumpectomy by Composite Measurement of Reoperations, Cosmetic Outcomes, and Patient Preferences.

BACKGROUND: Patients want information to search for destination of care for breast-conserving surgery (BCS). To inform patients wanting a lumpectomy, we aimed to develop a pilot project that communicated composite quality measure (QM) results using a '4-star' rating system. Two patient-centered QMs were included in the model-reoperation rate (ROR) and cosmetic outcome (COSM). METHODS: A prospective database was reviewed for stage 0-3 patients undergoing initial lumpectomy by three surgeons from 2010 to 2015. Self-reported COSM was assessed by survey. Multivariate analyses were used to test for interactions between surgeon and other variables known to influence RORs and COSMs. Models of surgeon profiling were developed that summed the ROR and COSM performance scores, then reported results using a Centers for Medicare and Medicaid Services (CMS) star-type system. Functionality for a patient to 'weight' the importance of the ratio of ROR:COSM before profiling was introduced. RESULTS: The unadjusted ROR for stage 1-3 patients for three surgeons was 9.5, 13.0, and 16.3%, respectively (p = 0.179) [overall rate 10.4% (38/366)]. After risk adjustment, differences between surgeons were observed for RORs, but not COSMs. Overall, patients reported excellent, good, fair, and poor COSMs of 55, 30, 11 and 4%, respectively. Composite star scores reflected differences in performance by surgeon, which could increase, or even disappear, dependent on the patient's weighting of the ROR:COSM ratio. CONCLUSION: Composite measures of performance can be developed that allow patients to input their weighted preferences and values into surgeon profiling before they consider a destination of care for BCS.

Does neoadjuvant chemotherapy affect morbidity, mortality, reoperations, or readmissions in patients undergoing lumpectomy or mastectomy for breast cancer?

BACKGROUND: The influence of neoadjuvant chemotherapy (NAC) prior to breast cancer surgery on postoperative complications is unclear. Our objective was to determine whether NAC was associated with postoperative outcomes in patients undergoing lumpectomy or mastectomy without reconstruction. METHODS: Patients meeting inclusion criteria were identified from the National Surgical Quality Improvement Program (NSQIP) database participant user files from 2005 through 2012, after which NSQIP discontinued the NAC variable. Primary outcome measures included a composite measure of morbidity and mortality (M&M) and reoperations and readmissions within 30 days of the index procedure. Rates of postoperative complications stratified by receipt of NAC were compared by χ2. A logistic regression model was then built that included confounding factors for M&M. RESULTS: There were 30,309 patients meeting inclusion criteria. NAC was not associated with any postoperative outcomes from 2005 through 2012, but it was associated with higher M&M in lumpectomy patients during 2011 to 2012 [P=0.011, odds ratio (OR) 2.579; 95% confidence interval (CI), 1.239-5.368]. CONCLUSIONS: The finding that NAC was associated with higher M&M in lumpectomy patients during 2011 to 2012 warrants further investigation. Therefore, we recommend that the NSQIP database reinstitute the NAC variable to allow monitoring during anticipated changes in chemotherapy agents and protocols.

Fewer Reoperations After Lumpectomy for Breast Cancer with Neoadjuvant Rather than Adjuvant Chemotherapy: A Report from the National Cancer Database.

BACKGROUND: Reoperations occur frequently after initial lumpectomy for breast cancer. The authors hypothesized that the receipt of neoadjuvant chemotherapy (NAC) is associated with fewer reoperations. METHODS: The association between timing of chemotherapy and reoperation rates (ROR) after lumpectomy was investigated for patients with stages 1-3 breast cancer in the National Cancer Database (NCDB) from 2010 to 2013 by multivariable logistic regression modeling. Then propensity score-matching was performed. RESULTS: The unadjusted ROR for 71,627 stages 1-3 patients was 11.4% for those who had NAC compared with 20.3% for those who had postoperative chemotherapy (p < 0.001) (odds ratio [OR] 0.53; 95% confidence interval [CI] 0.49-0.57; p < 0.001). The ORs for the reoperations performed for patients with stages 1, 2, and 3 cancers who received NAC were respectively 0.65 (95% CI 0.56-0.75), 0.50 (95% CI 0.45-0.56), and 0.27 (95% CI 0.19-0.38) The p values for all were lower than 0.001. CONCLUSION: For a population of patients receiving chemotherapy, the receipt of chemotherapy before instead of after surgery was associated with fewer reoperations after initial lumpectomy for breast cancer.

Extramammary findings in diagnostic breast magnetic resonance imaging among patients with known breast cancer: incidence and cost analysis.

BACKGROUND: Extramammary findings (EMFs) are common on breast magnetic resonance imaging (MRI). METHODS: A retrospective review of breast MRIs in breast cancer patients between January 2009 and December 2014 was performed to identify EMF occurrences, resultant evaluation, and added cost. RESULTS: EMFs were noted in 185 (59%) of 316 MRIs. Overall, 201 new EMFs were identified with 178 (89%) benign and 23 (11%) malignant. New malignant findings included 19 metastatic nodes (18 axillary, 1 internal mammary) and 4 primary malignancies (2 thyroid, 2 lung). New malignant nonaxillary EMFs occurred at a rate of 1.6% (5/316). EMFs resulted in 65 patients undergoing 98 imaging studies, 37 procedures, and 10 consultations with a median (range) total charge of $3,491 ($222 to $29,076] and out of pocket cost of $2,206 ($44 to $12,780) per patient. CONCLUSIONS: EMFs occurred in more than half of our patients, were usually benign, and frequently led to additional testing and costs.

Are Cure Rates for Breast Cancer Improved by Local and Regional Anesthesia?

BACKGROUND AND OBJECTIVES: Recent preclinical basic science studies suggest that patient tumor immunity is altered by general anesthesia (GA), potentially worsening cancer outcomes. A single retrospective review concluded that breast cancer patients receiving paravertebral block and GA had better cancer outcomes compared with patients receiving GA alone. This study has not been validated. We hypothesized that local or regional anesthesia (LRA) would be associated with better cancer outcomes compared with GA. METHODS: We retrospectively reviewed a prospectively collected database to identify all stage 0-III breast cancer patients undergoing surgery in a single center during a 9-year period ending January 1, 2010. Patients were divided into 2 groups: those who received only LRA and those who received GA. Overall survival (OS), disease-free survival (DFS), and local regional recurrence (LRR) were calculated using the Kaplan-Meier method with log-rank comparison before and after propensity score matching. RESULTS: Median age of the 1107 patients who met study criteria was 64 years (range, 24-97 years). Median and longest follow-up were 5.5 and 12.5 years, respectively. General anesthesia was used for 461 patients (42%), and 646 (58%) received LRA. The point estimates of cumulative OS, DFS, and LRR "free" rates at 5 years for the GA and LRA groups were 85.5% and 87.1%, 94.2% and 96.1%, and 96.3% and 95.8%, respectively. Cox regression showed no significant differences between the 2 groups (GA and LRA) for the 3 outcomes: OS (hazard ratio [HR], 0.81; 95% confidence interval [CI], 0.59-1.10; P = 0.17), DFS (HR, 0.91; 95% CI, 0.55-1.76; P = 0.87), and LRR (HR, 1.73; 95% CI, 0.83-3.63; P = 0.15). CONCLUSIONS: Breast cancer OS, DFS, and LRR were not affected by type of anesthesia in our institution. This result differs from that of the only prior published clinical report on this topic and does not provide clinical corroboration of the basic science studies that suggest oncologic benefits to LRA.

Is the American College of Surgeons National Surgical Quality Improvement Program surgical risk calculator applicable for breast cancer patients undergoing breast-conserving surgery?

BACKGROUND: We aimed to analyze the applicability of the National Surgical Quality Improvement Program (NSQIP) calculator to patients undergoing breast-conserving surgery. METHODS: A total of 287 consecutive patients treated with breast-conserving surgery from 2010 to 2012 were identified retrospectively. The risk calculator was applied to each patient to generate an individual risk profile. Risk calculations were then compared with actual outcomes. The performance of the risk calculator was evaluated using 2 metrics: the Brier score and c statistic. RESULTS: The NSQIP calculator performed adequately for all complications, with Brier scores less than .05. However, 37 patients (12.9%) returned to the operating room for oncologic indications. Twenty-nine patients (10.1%) had positive margins, whereas 8 patients (2.8%) returned due to an upgrade in diagnosis. CONCLUSIONS: When considering return to the operating room for oncologic management, the observed rate of 13.9% is significantly higher than the NSQIP prediction. This deviation must be addressed when using the NSQIP risk calculator model during preoperative risk discussion.

Are Breast Cancer Outcomes Compromised by General Surgical Resident Participation in the Operation?

OBJECTIVE: The effect of surgery resident participation on breast cancer recurrence has not been previously reported. The objectives of this study were to determine if resident participation was associated with either immediate postoperative or long-term breast cancer outcomes. DESIGN: We retrospectively reviewed a prospectively collected breast center database to identify all patients with breast cancer undergoing surgery in a single center during a 9-year period ending 1 January 2010. Patients were divided into 2 groups based on whether surgery residents completed more than 50% of the critical portions of the case (Resident group) or not. The outcomes of operation length, reoperative rates, morbidity, and the long-term outcomes of cancer recurrence were compared by group. Comparisons of immediate postoperative outcomes were made with chi-square and Fisher exact tests. Comparisons of operation length were analyzed by Wilcoxon rank-sum testing. Survival analyses were calculated using the Kaplan-Meier method with log-rank comparison. Multivariate analysis with Cox regression was also performed. SETTING: The study occurred at a community-based hospital that has an accredited general surgery training program. PARTICIPANTS: In all, 1107 consecutive patients with stage 0-3 breast cancer undergoing breast cancer operations were included. RESULTS: Median age of patients was 64 years (range: 24-97). Median and longest follow-up were 5.5 and 12.5 years, respectively. Initial operation was breast conserving in 796 (72%) and mastectomy in 311 (28%). Of the 1107 patients, 887 (80.1%) had resident participation. The Resident group was associated with longer operative times. We identified no differences in operative morbidity, reoperations, overall survival, disease-free survival, or local-regional recurrence in the Resident and No Resident groups. CONCLUSIONS: Resident involvement in breast cancer operations was associated with longer operative times but did not affect any other perioperative or cancer outcome in our institution. This information can be used to reassure program directors, attending surgeons, and patients if they have questions or concerns about the safety or effectiveness of cancer surgery when there is surgical resident participation.

Institutional review of compliance with NCCN guidelines for breast cancer: lessons learned from real-time multidimensional synoptic reporting.

BACKGROUND: Variations exist in compliance with NCCN Guidelines. Prior reports of adherence to NCCN Guidelines contain limitations because of lack of contemporary review and incomplete listing of reasons for noncompliance. PURPOSE: To assess institutional compliance and assist national quality improvement strategies through identifying valid reasons for noncompliance. METHODS: Compliance with NCCN Guidelines was recorded prospectively using electronic synoptic templates for patients with newly diagnosed breast cancer treated at a single institution between January 2010 and December 2011. Compliance with NCCN Guidelines was recorded. The accuracy of real-time synoptic auditing methods compared with retrospective chart review and reasons for noncompliance was assessed. SAS 9.3 software was used for data analysis. RESULTS: Compliance with NCCN Guidelines among 395 patients was 94% for initial staging evaluation, 97% for surgery, 91% for chemotherapy, 89% for hormone therapy, 91% for radiation therapy, 85% for follow-up, and 100% for determination of estrogen receptor/progesterone receptor and HER2 status. Age, comorbidities, and stage influenced guideline compliance. The most common reasons for noncompliance were patient refusal, patient choice after shared decision-making, and overuse of testing. Synoptic templated reporting was accurate in 97% patients. CONCLUSIONS: High compliance with NCCN Guidelines was demonstrated. Reasons for noncompliance were identifiable. Compliance and nonadherence can be evaluated quickly with electronic synoptic reporting. This allows real-time action plans to address quality concerns and aids national risk adjustment for comparison and benchmarking.

Core needle biopsy rate for new cancer diagnosis in an interdisciplinary breast center: evaluation of quality of care 2007-2008.

OBJECTIVE: The purpose of this study is to evaluate the rate of minimally invasive biopsy for diagnosis of breast cancer at an interdisciplinary breast center. BACKGROUND: Percutaneous core needle biopsy (CNB) is optimal for minimizing surgery for the diagnosis of benign and malignant lesions of the breast while preserving surgery for definitive resection. Core needle biopsy increases patient satisfaction and reduces the cost of diagnosis and treatment. Despite the endorsement of CNB by many professional organizations, the literature documents underutilization. METHODS: Institutional review board approval was obtained. An audit of a single institution's prospectively maintained cancer databases was performed for all breast cancers diagnosed in 2007 and 2008. Methods of diagnosis included image-guided and freehand-guided CNB, image-guided vacuum assisted needle biopsy, image-guided fine needle aspiration, punch biopsy, and open surgical biopsy. RESULTS: Three hundred sixty new breast cancers were diagnosed in 2007 and 2008. Malignancy was diagnosed by minimally invasive techniques in 350/360 (97%) cancers. CONCLUSION: A very high rate of accurate tissue diagnosis of breast cancer by minimally invasive techniques is achievable.

Contemporary breast imaging and concordance assessment: a surgical perspective.

Disease conditions of the breast are very common. Patients with such conditions often present to surgeons for both diagnostic evaluation and treatment. Nearly all of them will require breast imaging. This article summarizes the use, applicability, and concordance assessment of mammography, ultrasonography, and magnetic resonance imaging in patients who have breast complaints or abnormalities on clinical examination or imaging.

Does obesity impact lymph node retrieval in colon cancer surgery?

BACKGROUND: Evaluation of lymph nodes is important for the optimal treatment of colon adenocarcinoma. Few studies have assessed whether lymph node harvest is compromised by obesity. We hypothesized that lymph node retrieval in colon cancer resection would be reduced in obese patients. METHODS: Patients undergoing resection for colon adenocarcinoma diagnosed from 2000 to 2007 were reviewed retrospectively and stratified by body mass index (BMI). Lymph node harvest was evaluated. RESULTS: A total of 401 patients were included. Their mean age was 72.8 years, and 44% were men. Their mean BMI was 28.2 kg/m(2). Mean lymph node recovery among BMI groups was as follows: BMI less than 18.5 was 20.6; BMI of 18.5 to 24.9 was 25.1; BMI of 25 to 29.9 was 23.1; BMI of 30 to 34.9 was 22.4; BMI of 35 to 39.9 was 19.0; and BMI of 40 or greater was 21.1 nodes (P = .321). Surgical time increased with increasing BMI (P = .005). Adequacy of node harvest differed by stage (P = .007), left-sided versus right-sided resections (P = .001), and pathology technician (P = .001). CONCLUSIONS: Lymph node retrieval was not affected by BMI.

A quality review of the timeliness of breast cancer diagnosis and treatment in an integrated breast center.

BACKGROUND: The National Consortium of Breast Centers defines "quality" of breast cancer care as "accurate evaluation and appropriate services ... in a timely manner." We sought to determine timeliness of care and relationship to patient satisfaction. STUDY DESIGN: The electronic medical records of breast cancer patients seen at a breast center from 2004 through 2007 were retrospectively reviewed. Dates of patient service were audited. A postal survey was then conducted to determine patient satisfaction with timeliness. RESULTS: Median time interval in business days from abnormal screening mammogram to diagnostic evaluation and core needle biopsy was 6 days. Median time intervals from core needle biopsy to core needle biopsy pathology report and then subsequent surgical consultation and breast cancer operation were 1, 3, and 7 days, respectively. Breast MRI, systemic imaging, plastic surgery consultation, type of surgery, and patient choice prolonged time to treatment. More than 90% of breast cancer patients who responded to our postal survey had their expectations met or exceeded for the dates of service provided for diagnostic evaluation and treatment. CONCLUSIONS: Evaluation of timeliness as a quality indicator for breast cancer care is complex and requires an assessment of whether patient expectations were met for dates of service. Factors that prolong time to treatment, such as additional imaging, should be considered for risk adjustment for peer performance comparison and compliance with published timeliness target goals.

Proteomic analysis of neuroblastoma subtypes in response to mitogen-activated protein kinase inhibition: profiling multiple targets of cancer kinase signaling.

INTRODUCTION: Survival for high-risk neuroblastoma (NB) remains poor despite aggressive therapy. Novel therapies are vital for improving prognosis. We previously showed differential NB subtype sensitivity to p42/44 mitogen-activated protein kinase (ERK/MAPK) pathway inhibition. In this study, we investigated proteomic changes associated with resistance or sensitivity to MAPK kinase (MEK) inhibition in NB subtypes. MATERIALS AND METHODS: SH-SY5Y (N-type), BE(2)-C (I-type), and SK-N-AS (S-type) were treated with MEK inhibitor U0126 (10 microM) for 1 and 24 h. Proteins were extracted from untreated and treated cells and analyzed for differential expression by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). Selected polypeptides were extracted from the gel and identified by liquid chromatography-linked tandem mass spectrometry (LC-MS/MS). RESULTS: We identified 15 proteins that were decreased by 2.5-fold between untreated and 1 h treated cells and subsequently up-regulated 5-fold after 24 h drug treatment. N-type NB (MEK-resistant) showed the least altered proteomic profile whereas the I-type (MEK-sensitive) and S-type NB (MEK-intermediate) generated significant protein changes. The majority of proteins identified were induced by stress. CONCLUSIONS: Protein differences exist between MEK inhibitor-treated NB subtypes. Identified polypeptides all have roles in mediating cellular stress. These data suggest that inhibition of the ERK/MAPK in NB subtypes leads to an intracellular stress response. The most resistant NB cell line to MEK inhibitor treatment generated the least protective protein profile, whereas the intermediate and most sensitive NB cells produced the most stress response. These findings suggest stress related protein expression may be targeted in assessing a response to ERK/MAPK therapeutics.

Identification and characterization of noncalcemic, tissue-selective, nonsecosteroidal vitamin D receptor modulators.

Vitamin D receptor (VDR) ligands are therapeutic agents for the treatment of psoriasis, osteoporosis, and secondary hyperparathyroidism. VDR ligands also show immense potential as therapeutic agents for autoimmune diseases and cancers of skin, prostate, colon, and breast as well as leukemia. However, the major side effect of VDR ligands that limits their expanded use and clinical development is hypercalcemia that develops as a result of the action of these compounds mainly on intestine. In order to discover VDR ligands with less hypercalcemia liability, we sought to identify tissue-selective VDR modulators (VDRMs) that act as agonists in some cell types and lack activity in others. Here, we describe LY2108491 and LY2109866 as nonsecosteroidal VDRMs that function as potent agonists in keratinocytes, osteoblasts, and peripheral blood mononuclear cells but show poor activity in intestinal cells. Finally, these nonsecosteroidal VDRMs were less calcemic in vivo, and LY2108491 exhibited more than 270-fold improved therapeutic index over the naturally occurring VDR ligand 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3] in an in vivo preclinical surrogate model of psoriasis.