Abstract P5-13-08: Association of tumor infiltrating lymphocytes and chemotherapy regimen in response to preoperative chemotherapy in underserved patients

Introduction

Preoperative chemotherapy can demonstrate an individual's response to the chemotherapeutic regimen by comparing the amount of cancer at presentation to the amount remaining after treatment. Multiple previous studies have demonstrated that the amount of residual cancer, or final pathologic stage, is a better indicator of prognosis than the initial stage at presentation.

Tumor infiltrating lymphocytes have been recognized in breast cancer, and when found concentrated in breast cancer specimens, have been associated with a good prognosis. Breast cancer is not only a heterogeneous disease, but also displays varied presentation and behavior in patients of different race/ethnicities.

This study was performed to evaluate factors which predict response to chemotherapy. The effectiveness of different chemotherapeutic regimens, the effect of breast cancer subtype, and tumor infiltrating lymphocytes (TILs) were evaluated in our racial/ethnic minority population.

Methods

All patients at the safety net institution in Phoenix, AZ who underwent preoperative chemotherapy from 2002 to 2017 and had tissue available for evaluation were included in the study. Response to chemotherapy regimen was recorded. Pathologic complete response (pCR) was defined as no invasive cancer in the breast and lymph nodes in the final pathologic specimen. Breast cancer subtypes were divided based on IHC and FISH testing. Luminal subtypes were classified based on Ki67 (>15%) and/or PR (<20%) for Luminal B. Her2 subtype was defined as Her2 IHC 3+ or Her2 FISH amplified. Triple negative breast cancer (TNBC) was defined as ER and PR (<5%) and Her2 negative. TILs concentration was determined from fixed formalin paraffin embedded (FFPE) core needle biopsy specimens.

Results

A total of 259 patients were included in the study. The mean age was 45 years. 80% of the population were racial/ethnic minorities. The vast majority (94%) were underinsured or uninsured, with 75% uninsured. The mean clinical tumor size at presentation was 6cm. 52% presented at clinical stage 2 while 48% presented at clinical stage 3. The overall pCR rate was 32%. pCR rate was impacted by breast cancer subtype with TNBC 52% and Her2 38% showing a better response to chemotherapy, while Luminal B was 16% and Luminal A 2% (p < 0.05). In TNBC, chemotherapy regimens with anthracycline and docetaxel may have improved efficacy with pCR of 56% (p = 0.05). In the subgroup available for TIL evaluation, breast cancer subtype appeared to show similar importance with pathologic complete response rates of TNBC 50%, Her2 44%, Luminal B 12%, and Luminal A 0%. TIL appeared to affect the likelihood of pCR. When TIL were less than 5% the pCR rate was 16% compared to when there were TIL of at least 5% or more the pCR rate was 41% (p < 0.05).

Conclusions

In our racial/ethnic minority population, breast cancer subtype and chemotherapy regimen did affect likelihood of pathologic complete response. Tumor infiltrating lymphocyte concentration as low as 5% may indicate a higher likelihood of pathologic complete response and could be used as an additional factor in the evaluation of patients for preoperative therapy.

Citation Format: Komenaka IK, Cocco D, Huliyar R, Hsu C-H, Martinez ME, Gago M, Nodora J, Mehta D, Caruso DM. Association of tumor infiltrating lymphocytes and chemotherapy regimen in response to preoperative chemotherapy in underserved patients [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-13-08.

Abstract P3-13-18: Technical skill of surgical residents may affect margin status of breast conserving operations

Background: Quality of patient care and surgical outcomes have come under increased scrutiny. Numerous other studies have examined outcomes at teaching hospitals compared to non-teaching hospitals. The most important factor to determine early success of breast conservation is the attainment of pathologically negative margins. Few studies, however, have examined the effect of teaching residents on margin status. The current study was performed to evaluate the effect teaching residents on margin status after lumpectomy.

Methods: Retrospective review of all patients from July 2006-June 2015 was evaluated. A resident was usually considered the primary surgeon. If the resident was unavailable or unprepared to operate, the attending surgeon was the primary surgeon. As part of the routine evaluation of surgical residents, technical ability was classified as satisfactory or unsatisfactory for level of training. All evaluations of the technical ability of the residents were completed prior to the collection of the current data. The effect of surgical residents' participation and their technical ability to perform lumpectomy was evaluated to determine if there was an effect on margin status. Logistic regression analysis was performed to adjust for clinical variables known to affect margin status.

Results: Of 292 patients, 15% of patients had positive margins. The attending surgeon has positive margin rate of 10.7% vs 16% for Residents (p = 0.32).When technical skill evaluation was included, Residents with unsatisfactory technical skills had positive margin rate of 27% compared to 10.2% for residents with satisfactory skills (p = 0.002). In multivariate logistic regression analysis, operating surgeon remained significantly associated with positive margins. Operations done by residents with satisfactory technical skills or attending surgeon were less likely to have positive margins than those done by residents with unsatisfactory technical skills (OR 0.39, 95% CI 0.22-0.86; p = 0.03). Patients who underwent preoperative chemotherapy were also less like to have positive margins (OR 0.40, 95% CI 0.18-0.91); p = 0.04). In patients with at least 2 years of followup (mean follow up of 48 months) breast cancer specific survival was 94% and 2% had local recurrences as a first event.

Conclusions: Technically ability of residents may affect margin status after lumpectomy. With the increased use of surgical outcomes to measure quality of care in medicine, the importance of teaching surgical residents needs to be considered in future quality of care evaluation.

Citation Format: Komenaka IK, Djenic B, Hsu C-H, Nodora J, Winton L, Bouton M, Martinez ME. Technical skill of surgical residents may affect margin status of breast conserving operations [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-13-18.

Abstract P5-13-02: Weight gain after breast cancer diagnosis and patients' opinion on weight loss strategies

Background: The rate of obesity is increasing in many countries worldwide. Most populations are not aware of steady weight gain with age. In addition, obesity is increasingly being recognized as a risk factor for breast cancer. Some studies have also demonstrated that weight gain after diagnosis is associated with increased risk for recurrence. The current study was performed to evaluate weight gain in a population of patients seen at a safety net institution. Additionally, patients also gave their opinion on weight loss strategies.

Methods: A retrospective review of all breast cancer patients seen at the county, safety net institution from July 2001 to June 2014 who had at least 2 years of follow up were evaluated for change in weight. For the question on weight loss strategies, all patients from May 2014 to May 2015 were included. Sociodemographic, clinical, and treatment variables were evaluated.

Results: From July 2001 to June 2014, 225 breast cancer patients had follow up prior to January 2013. Of these patients 59% gained weight after their diagnosis of breast cancer. Overall these patients gained an average of 2 kg after their diagnosis.

Starting in January 2013, patients were given a simple message, “avoiding gaining weight” after their diagnosis. For 115 patients with follow up after January 2013, only 35% gained weight and on average, this group of patient lost 4 kg from diagnosis. Patients who initially gained weight were provided with basic weight loss strategies.

From May 2014 to May 2015, 1198 consecutive patients were seen. The average age was 45 years. 12% were non-Hispanic White. Only 30% of patients stated that their primary care provider discussed weight maintenance or weight loss as part of their routine health care. 40% of patients did feel that a permanent change was necessary to facilitate weight loss. However, fewer than half (44%) the patients felt that exercise was necessary. Dietary beliefs were varied, with the most commonly held belief was that vegetarian/vegan diet was necessary (28%). Patients rarely cited commonly recommended weight loss strategies: lower caloric intake (9%), eat frequently (0.3%), adequate fiber (0.1%), adequate protein (0.6%), enough sleep (0.1%), don't eat late (0.1%), no fast food (2%), avoid soda (1%), drink more water (1%). 6% of patients felt a low/no carbohydrate diet was important. By contrast, older strategies or popular ideas: low fat (8%), no flour/wheat/gluten (4%) were also felt to be important. 7.3% stated they did not know any strategy. 5% patients recommended stopping eating altogether to lose weight.

Conclusions: Similar to population wide data, breast cancer patients treated at a safety net institution tend to gain weight with follow up. Also similar to most populations, the population was not aware of recommended weight loss strategies and most do not feel that exercise is an important component of weight maintenance. As with the rest of the US population significant effort will be necessary to help patients avoid weight gain after diagnosis. Making patients aware of their weight may help patients avoid gaining weight.

Citation Format: Komenaka IK, VanderVelde J, Hsu C-H, Nodora J, Winton L, Bouton M, Martinez ME. Weight gain after breast cancer diagnosis and patients' opinion on weight loss strategies [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P5-13-02.

Abstract P6-12-06: Preoperative chemotherapy regimens and breast cancer subtype in an underinsured Hispanic population

Background: Although the Hispanic population is among the fastest growing in the United States, however, less is known about them then other populations. Breast cancer in different racial/ethnic populations display different behaviors. The current study was performed to examine response to preoperative chemotherapy regimen and by breast cancer subtype in a Hispanic safety net population.

Methods: A retrospective review of Hispanic breast cancer patients who underwent preoperative chemotherapy from July 2001 to February May 2015 at a safety net hospital. Sociodemographic, clinical, and treatment variables were evaluated. Response to chemotherapy regimen was recorded. Breast cancer subtypes were divided based on IHC and FISH testing. Luminal B subtype was classified based on Ki67 (>15%) and PR (<20%).

Results: The average age of the 133 patients was 45 years. 93% of the patients were insured with Medicaid or uninsured (70%). The average size of the cancers at presentation was 5cm. Overall 86% of patients had a clinical response to preoperative chemotherapy and 35% had pathologic complete response (pCR). AC and TC regimens had the lowest rate of pCR at 16%. AC/T (every 3 week and weekly) had similar rates of pCR 23% and 29%, while dose dense regimens showed pCR 40%. Herceptin containing regimens had pCR 57%. 7 patients received TAC and 71% had pCR.

By subtype, Luminal A and B patients had low rates of pCR 9% and 14% respectively. Luminal B patients did benefit from preoperative chemotherapy as 86% of patients who were not candidates for breast conservation at presentation were able to undergo lumpectomy after preoperative chemotherapy. Her2 subtype patients who got Herceptin had pCR 57%. Triple negative patients had pCR 54%.

Conclusions: In this underinsured, Hispanic population who presented at advanced stages, differences in response to preoperative chemotherapy were seen based on breast cancer subtype. Differences were also seen based on chemotherapy regimen. TAC maybe a particularly effective regimen in triple negative Hispanic women.

Citation Format: Mehta D, Winton L, Walters J, Hsu C-H, Nodora J, Martinez ME, Bouton M, Komenaka IK. Preoperative chemotherapy regimens and breast cancer subtype in an underinsured Hispanic population. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-12-06.

Abstract P6-12-07: The benefit of preoperative chemotherapy in an underinsured Hispanic population with poor use of screening mammography

Background: Despite prospective clinical trials demonstrating the safety and effectiveness of preoperative chemotherapy for nearly 2 decades, it may still be underutilized in underserved, uninsured populations most likely to present with advanced cancers. The current study was performed to evaluate the effect of preoperative chemotherapy (PC) in a Hispanic safety net population.

Methods: A retrospective review of Hispanic breast cancer patients who presented at clinical stage 2 or higher and were treated from July 2001 to February May 2015 at a safety net hospital. Sociodemographic, clinical, and treatment variables were evaluated. Surgical outcomes were evaluated. Margin status was determined for those who underwent breast conservation.

Results: The average age of the 266 patients was 45 years. 93% of the patients were insured with Medicaid or uninsured (70%). Only 24% of patients underwent screening mammography. 133 underwent PC and 133 had primary operations. Patients who underwent PC presented with larger cancers (5cm vs 3.5cm, p < 0.001). 96% of patients in PC group were not candidates for lumpectomy at presentation. However, lumpectomy was performed more often in those who underwent PC (75% vs. 57%, p = 0.01). Re-excision for margins were necessary less often in those who underwent PC (10.3% vs 27%, p = 0.01). Patients who underwent PC were also less likely to require an ALND (33% vs 47%, p = 0.04). Despite presentation at higher average clinical stage in patients who underwent PC, at average follow up of 52 months, risk of IBTR and risk of regional recurrence were similar in the PC and no-PC groups (IBTR: 4.4% vs 3%, p = 0.99 and Regional: 1.5% vs. 1.5%, p = 0.99). Breast cancer specific survival was 86.5% in the PC group compared to 84% in the no-PC group (p = 0.68). Patients in the no-PC group were less likely to comply with recommended chemotherapy.Conclusions: In this underinsured, Hispanic population who did not use screening mammography, preoperative chemotherapy allowed many women to undergo breast conservation, undergo fewer operations, and were less likely to require ALND. Risk of local and regional recurrence is low in these patients and comparable to those who underwent primary operation.

Citation Format: Komenaka IK, Djenic B, Walters J, Hsu C-H, Nodora JN, Martinez ME, Bouton M, Mehta D. The benefit of preoperative chemotherapy in an underinsured Hispanic population with poor use of screening mammography. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-12-07.

Abstract P2-18-16: Patients’ impression of the expected appearance of breasts in 2013

Background: Concern exists about the increasing rate of bilateral mastectomy for patients with unilateral breast cancer. Patient concern for second primary, use of breast MRI, and increased genetic cancer risk assessment all may influence choice of operation. In recent years, cosmetic breast augmentation has also become more common. This study was performed to evaluate women's perception of the current expected appearance of breasts.

Methods: From April 2012 to May 2013, all patients seen at a Breast Clinic were shown two pictures (A “natural” and B “augmented”) and asked “What is the expected appearance of breasts in 2012(2013)?” Sociodemographic, clinical, and treatment variables were collected. Univariate analysis was performed to identify variables which were associated with picture choice. All of the variables with a p-value <0.10 were included in the multivariate analysis.

Results: 1,177 consecutive patients were seen with mean age 45 years. 93% of the patients were insured with Medicaid or uninsured and 19% were non-Hispanic White (NHW). The mean monthly income was only $1104. Overall 70% of patients felt that the augmented appearance is “expected in 2013.” Age was strongly associated with augmented appearance as younger patients were significantly more likely to choose picture B: age less than 40 years (85%), age 40-49 years (67%), age 50-59 years (65%), and age 60+ years (51%). Patients of all race/ethnic groups felt that the augmented appearance was expected. NHW 65%, AA 67%, and Hispanic 71%. Multivariate analysis found that more years of education (p = 0.001), younger age and lower BMI were strongly associated (p < 0.0001) with choosing the augmented appearance as “expected in 2013.”

Of the 1177 patients, 337 breast cancer patients were seen. Overall 62% felt that the augmented appearance was expected. Of those who underwent an operation, 178 (58%) patients underwent breast conservation and 62% felt the augmented appearance was expected. Of 128 patients who underwent mastectomy, 36 (28%) underwent reconstruction. 23 patients underwent bilateral mastectomy. Of those who underwent mastectomy alone (41/83 = 49%) only about half felt the augmented appearance was expected. By contrast those who underwent reconstruction (33/36 = 92%; p < 0.001) or bilateral mastectomy (17/23 = 74%; p = 0.057), were significantly more likely than those who underwent mastectomy alone to feel the augmented appearance was expected.

Conclusions: The current study suggests women's perception of the expected appearance of breasts is changing. Younger age was strongly associated with the perception that the “augmented” appearance is expected. This change may affect patients’ choice in surgical operations with more mastectomies and reconstructions in the future. In October 2012 from a common women's periodical, when asked about her choice of operation for a small area of unilateral DCIS, one woman responded, “Just take them off and give me implants. Everybody has implants; it's no big deal.”

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-18-16.

Abstract P1-09-15: Perception of breast cancer risk in an underinsured safety net population

Background: Patient perception of breast cancer risk can affect compliance with screening mammography. In an underinsured population, we have previously found that being insured and adequate health literacy were strong predictors of use of screening mammography. Previous studies in other populations have found that patients with heighted perceived risk also have increased use of screening. Previous studies have found an average perceived risk of 30%. “Risk” and risk assessment are difficult concepts for many people and this may be problematic in undereducated populations. The current study was performed to evaluate women's perception of their lifetime risk of breast cancer in a safety net population.

Methods: From May 2012 to May 2013, all patients seen at a safety net Breast Clinic were asked to estimate their lifetime risk of breast cancer. “If 0 = no chance and 100% = for sure, what are your chances of getting breast cancer in your life?” Sociodemographic, clinical, and treatment variables were collected. Univariate analysis was performed to identify variables which were associated with picture choice. All of the variables with a p-value <0.10 were included in the multivariate analysis.

Results: 1,089 consecutive patients were seen. After exclusion of men and patients known to have breast cancer, 838 patients were included. The mean age was 43 years. 93% of the patients were uninsured or insured with Medicaid. Average education 10 years and 17% were non-Hispanic White (NHW). Overall the average perceived lifetime risk for the population was 29.7%. For all patients age 35 years+, the average Gail model Risk Assessment scores were 1.1% (5 year) and 9% (LT). Only 17% of patients would be considered at increased risk according to Gail model 5 year risk (> 1.6%).

In univariate analysis, continuous variables associated with higher perceived risk were higher BMI (p = 0.003), more years of education (p = 0.0009), and higher 5 year Gail model score (p = 0.0005). Categorical variables associated with higher perceived lifetime risk were being insured (34% vs 27%; p = 0.004) and NonHispanic patients (34% vs 27%; p = 0.003). Adequate health literacy (38% vs 27%; p = 0.0001) and family history of breast cancer in a first degree relative (FDR; 43% vs 27%; p = 0.0001) were the variables with the strongest association with perceived risk. In multivariate analysis after adjustment for the above factors, only family history in FDR remained significant (p = 0.003).

Over a similar time period, being insured (OR 1.52; 95% CI 1.18 – 1.95; p = 0.001) and adequate health literacy (OR 3.75; 95% CI 2.71 – 5.19; p < 0.0001) were the only variables significantly associated with the use of screening mammography.

Conclusions: Underinsured, minority women overestimate their lifetime risk of breast cancer (30%) at a magnitude similar to other populations. Family history has a strong influence on an underinsured populations’ perception of breast cancer risk. Patients who were insured or had adequate health literacy perceived a higher lifetime risk of breast cancer and were significantly more likely to use screening mammography.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-09-15.

Abstract PD08-04: Factors which affect surgical management in an underinsured, county hospital population

Background: Significant variation exists between institutions in the use of lumpectomy, mastectomy, and reconstruction. Much less is known about minorities and populations outside the large academic institutions. The current study was performed to evaluate variables that affect patient choice in surgical management in a county hospital population.

Methods: A retrospective review of all patients seen at the county, safety net institution with breast cancer from January 2010 to May 2012. Sociodemographic, clinical, and treatment variables were evaluated. Univariate analysis was performed to identify variables which were associated with type of operation. All of the variables with a p-value <0.10 were included in the multivariate analysis.

Results: 403 patients were seen with mean age 53 years. 92% of the patients were insured with Medicaid or uninsured and 29% were non-Hispanic White. Only 20% of patients underwent screening mammography and therefore presentation with palpable, Stage 2A/B cancer was most common (46%). 54 patients presented with T4 tumors and 13 (24%) were found to have metastases. Only 2 of 340 (0.6%) patients who presented at Stage 3A or earlier presented with metastatic disease. Patients with operable cancer underwent lumpectomy in 65%, mastectomy in 26%, and 9% mastectomy with reconstruction. With respect to breast conservation vs mastectomy, in adjusted analysis, married patients (OR 2.59, p = 0.003) and patients with larger tumors (p = 0.003) were more likely to undergo mastectomy, while patients who were Hispanic (OR 0.38, p = 0.004), underwent preoperative chemotherapy (OR 0.25, p = 0.002), or had their operation by breast surgical oncologist (OR 0.30, p = 0.005) were more likely to undergo breast conservation. When patients who underwent mastectomy alone were compared to those who underwent reconstruction, unadjusted analysis suggested that reconstruction patients were more likely to speak English, have adequate health literacy (HL), lower clinical stage, and be seen by a breast surgical oncologist. Adjusted analysis demonstrated that having been seen by a breast surgical oncologist (OR 18.4, p = 0.007), younger age (p = 0.05) and adequate HL (OR 3.13, p = 0.06) were associated with likelihood of reconstruction compared to mastectomy alone.

Conclusions: Breast conservation and mastectomy with reconstruction can be achieved in a significant proportion (74%) of underscreened and underinsured patients. Patients who underwent preoperative chemotherapy were more likely to undergo breast conservation. Younger patients and patients with adequate HL were more likely to choose reconstruction after mastectomy. Patients treated by breast surgical oncologists were more likely to have breast conservation or mastectomy with reconstruction. Even in an underscreened population, presentation with metastatic disease is uncommon in patients with operable breast cancer.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr PD08-04.

Abstract P4-10-12: Wire Guided Breast Procedures with Intraoperative Ultrasound

Background: Wire guided breast procedures (WGBP) are the most commonly used breast conserving operation for non-palpable cancers. Suboptimal needle placement can result in a larger amount of tissue removed and increased difficulty in removal of the lesion. The use of intraoperative ultrasound can improve the execution of wire guided breast procedures by more precisely demonstrating the direction and depth of the wire as well as the location of the wire hook and radiographic clip. These benefits allow a more appropriately located incision and smaller volume of tissue removed.

Methods: A retrospective review of all patients who underwent the WGBP at a county hospital with associated surgical residency program. 28 patients underwent the procedure with intraoperative ultrasound from June 2009 to March 2010.

Results: Breast cancer patients who underwent a wire-guided lumpectomy with intraoperative ultrasound had a lower rate of positive margins (9% vs. 26%, p = 0.28) and a smaller volume of tissue removed (126 cm3 vs. 146 cm3, p = 0.57) compared to patients who underwent wire guided lumpectomy alone. In patients who underwent wire guided excisional biopsy for diagnostic purposes, the volume of tissue removed was smaller in the intraoperative ultrasound group (30 cm3 vs. 44 cm3, p = 0.17) and the targeted area was more likely to be removed in one specimen (1.1 vs. 1.5, p = 0.03) compared to those who underwent the procedure without intraoperative ultrasound. Two illustrative cases with figures are presented. The first demonstrates the benefit when the distance from skin entry point of the wire to the target area is great. The hook can be localized with ultrasound for optimal placement of the incision.

The second demonstrates potential benefit when wire placement is suboptimal and the wire hook is distant from the target. The surgeon may use the wire location to identify the approximate area of the lesion and then the exact location of the radiographic clip can be localized with intraoperative ultrasound.

Conclusions: Intraoperative ultrasound can improve surgical outcomes of the wire guided breast procedure.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P4-10-12.

Prolonged Injection Site Mass Can Occur with Methylene Blue but Not Lymphazurin Blue after the Sentinel Node Procedure

Objectives: The purpose of this study is to evaluate the differences in outcomes and complications of the sentinel node dissection for breast cancer using Methylene blue and Lymphazurin blue.Methods: From 2004-2008, 160 patients underwent the sentinel node procedure at a county hospital, teaching institution. The sentinel node procedure was performed using the combined technique of blue dye and radioactive isotope. From 2004 to June 2006, Lymphazurin blue (LB) was used. From July 2006 to 2008, Methylene blue (MB) was used. Axillary dissection was performed if the sentinel node procedure revealed positive lymph nodes or failed.Results: Fifty patients underwent the sentinel node procedure with LB and 110 underwent the procedure with MB. The patients were of similar age and stage of presentation. More patients in the MB group were Hispanic and underwent preoperative chemotherapy. The sentinel node identification rate was 100% in the LB group and 99% in the MB group (P = 1.0). The mean numbers of sentinel lymph nodes were similar (LB 2.2 vs. MB 2.7). In the LB group 34% of patients were lymph node positive compared to 40% in the MB group.There were no allergic reactions in either group. The infection rate in the breast was similar in both groups LB (2%) and MB (1.8%). Previous studies have reported skin manifestations of necrosis, ulceration, tattooing, erythema, rash, or telangiectasia/vascular change at the site of injection with methylene blue. These findings were not seen in any of the patients in this study. In ten of the 110 patients (9%) who underwent the procedure with MB, however, a local inflammatory reaction occurred which resulted in a palpable mass at the site of injection. No injection site reactions were noted in LB group. In the MB group, the masses were clearly more prominent than the surrounding breast tissue and identified either at the one week postoperative check (6) or at the first 6-month followup (4). Ultrasound of the injection sites did reveal a hypoechoic mass in one patient. Biopsies were performed on the first four patients, but revealed only inflammation and non-atypical hyperplasia. Nearly all of the masses resolved by one year, however, one resolved after 18 months.Conclusion: Methylene blue is a reasonable alternative to lymphazurin blue in the sentinel node procedure for breast cancer. Awareness of injection site reactions to MB are important, as a new dominant mass is always reason for concern in breast cancer patients. Palpable mass at the site of MB injection can occur but typically resolve over a one year period.

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 3115.

Understanding of breast cancer concepts in an undereducated county hospital population

Abstract #2102

Objectives: The purpose of this study was to determine whether patients at a county institution are sufficiently knowledgeable about breast cancer concepts to make informed decisions about their care.
 Methods: An 11 question multiple choice examination was devised at a Flesch-Kincaid grade level of 5.8. This was administered to patients prior to undergoing their definitive cancer operation to which they had previously consented. The completed test was reviewed with the patient prior to surgery, and all incorrect answers were thoroughly discussed.
 Results: Forty-one of 42 consecutive women agreed to participate in the study. Seventy-eight percent self identified as Latin American, 15% Caucasian, and 7% African American. Fifty-nine percent of respondents chose to receive their test in Spanish and 41% chose the English version. The educational backgrounds of participants were: 64% less than high school, 12% finished high school, 24% had some college education.
 These undereducated patients demonstrated some understanding of their operative procedure. For the question “Which surgery gives me a better chance to live?” 51% said mastectomy and lumpectomy are equal, 33% thought mastectomy was superior, and 16% felt lumpectomy was superior. Of those who felt mastectomy was better, 38% still chose to undergo a lumpectomy. In response to the question which we expected to be most difficult, “What is a sentinel lymph node?” 63% selected the correct answer.
 There was a significant correlation between highest level of education with test score (p = 0.001) and between employment status with test score (p = 0.001). Age, insurance status, household income, and ethnicity did not correlate with test score. The average score for those with less than a high school education was significantly lower than those who at least completed high school (mean 5.5 vs. 8.3, p = 0.002). The patients who chose the exam in Spanish scored lower than those who chose the exam in English (mean 6 vs. 7.8, p = 0.04).
 Of interest, 88% of respondents understood that mammograms are important because “they find cancers when they are small and save lives.” Sixty-two percent of the patients, however, did not get screening mammograms. The most common reasons patients reported for lack of compliance with screening recommendations was the cost of mammograms (33%), they had no family history (24%), and they had “no problem” with their breasts (19%).
 Conclusion: Although patients felt subjectively that they had enough information to sign an informed consent prior to their operation, this examination revealed deficits in the understanding of this undereducated population. Higher level of education and employment status did correlate with improved score on the test. Despite their low level of education, the patients recognize the importance of mammography. Their reasons for not undergoing mammograms, however, indicate that they may not understand the concept of screening nor how to access this important tool.

Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 2102.

Erythromycin and position facilitated placement of postpyloric feeding tubes in burned patients

BACKGROUND

One of the main difficulties encountered in enteral nutrition in critically ill patients is the impaired gastric emptying: while the small bowel is ready the stomach is still 'lazy'. This paper describes a simple bedside method for postpyloric feeding tube placement, using a gastric prokinetic agent, erythromycin, and patient positioning.

METHODS

In eight critically ill, burned patients a gastric feeding tube was placed in a reverse Trendelenburg, right lateral decubitus position. A dose of 250 mg of erythromycin was administered intravenously.

RESULTS

In 13 out of 14 attempts, the tube passed into the duodenum, a success rate of 92%.

CONCLUSIONS

Combining the prokinetic effects of erythromycin with proper patient positioning allows a rapid bedside transpyloric placement of feeding tubes.

Vertical banded gastroplasty staple-line dehiscence after blunt trauma to the abdomen

An extensive Medline search revealed no documentation in the literature regarding staple-line disruption of a vertical banded gastroplasty after blunt abdominal trauma. A 41-year-old woman with a past surgical history significant for a vertical banded gastroplasty 13 months ago was admitted to our hospital one week after an episode of blunt trauma to the abdomen. The patient presented with complaints of nausea, vomiting, and anorexia after the incident. The suspected diagnoses were stenosis of the stoma due to hematoma after the blunt trauma as well as staple-line dehiscence. A partial disruption of the vertical staple-line was identified by esophagogastroduodenoscopy and barium swallow.