PS1-10: Spiritual Well-being and the Challenges of Living With an Ostomy: Resilience, Adaptation and Loss Among Colorectal Cancer Survivors

Background Spiritual well-being (SpWB), defined as hopefulness, inner peace, and sensing a reason to be alive, is integral to health related quality of life (HRQOL). It is only partially related to spirituality and religiosity. The challenges of colorectal cancer (CRC) and subsequent bodily changes can affect SpWB. We explored expressions of SpWB reported by CRC survivors with ostomies. Methods We recruited all adult CRC survivors (>=5 years) with permanent ostomies who were members of Kaiser Permanente Northwest, Northern California, or Hawai'i during 2000-2006 to complete a mailed survey (n=283). We asked participants to respond to an open-ended question about the greatest challenge they encountered in having an ostomy. Responses from the 62% (176 of 283) of patients who answered this question were analyzed based on the City of Hope HRQOL model for content related to SpWB, and to identify and categorize the SpWB themes found. Results The responses of 51% (90 of 176) of participants contained SpWB content. Seventeen SpWB themes were identified, reflecting positive, negative, and ambivalent dimensions. Some responses contained multiple themes but each theme was coded only once for each person. Fifty-three of 90 people (59%) expressed positive themes which included "positive attitude" "appreciate life more" "helping others helps me" "strength through religious faith" "leading an active life" and "I am fortunate." Negative themes included "struggling to cope" "not feeling 'normal'" and "loss" and were least common, expressed by only 28 of 90 people (31%). Ambivalent themes were most common (67%; 60 of 90 individuals) and included "learning acceptance" "ostomy is the price for survival" "reason to be around despite suffering" and "continuing to cope and function despite challenges." Discussion These CRC survivors with ostomies infrequently cited negative SpWB as a major challenge, though the greatest number of SpWB responses were ambivalent. More commonly, SpWB themes were mentioned as a source of resilience or part of the struggle to adapt to their altered body after cancer surgery. SpWB interventions should be broadly constructed to include the many types of SpWB, as opposed to narrowly defining SpWB only in terms of spirituality or religiosity.     

PS1-21: Long-term Rectal Cancer Survivors: Perceived Effects of Cancer on Work and Volunteering

Background/Aims Surgical treatments for rectal cancer include intestinal ostomy (externalization of the bowel to the abdominal wall) or, more frequently, an anastomosis (reconnection) of the rectum. These surgeries can give rise to long-term after-effects. We examined differences reported between survivors with ostomies (SO) versus anastomoses (SA) with respect to their work and volunteer activities, and how much they perceived their cancer affected these experiences. Methods We mailed our survey to 1,000 long-term (=5 years post-diagnosis) rectal cancer survivors in KP Northern California and KP Northwest during 2010-2011. Our overall response rate was 57.7% (577/1,000). The survey asked about current employment status and the impact of their cancer on labor force participation, demotions, job discrimination, forced retirement, and volunteer activities. Responses from 446 survivors (SA=292 (65%), SO=154 (35%)) were analyzed for differences in changes in work- and volunteer activities. Results Mean ages for both groups were significantly beyond the retirement age for most Americans (SA=72 yrs, SO=74 yrs). SAs were more likely to be currently working (FT+PT = 30% (89/292)) than SOs (FT+PT = 17% (25/151)), while SOs were more likely to be retired (67% (101/151) vs. 55% (160/292) (p<.05). Rates of permanent disability were similar between the groups-SOs=6.6% (10/151), SAs=5.5% (16/292). One-fifth of survivors (SA=22% (65/290), SA=18% (28/154)) reported volunteering on a regular basis an average of 11 (SOs) to 15 (SAs) hours/week. SAs reported more support from employers than SOs (64% (138/215) vs. 53% (60/113), p<.05), and from co-workers (68% (146/214) vs. 57% (64/112), p<.05). SAs and SOs did not differ in marital status at times of diagnosis or survey. The distribution of marital status shifted between diagnosis and survey: 72% (305/425) were married and 10% (43/425) were widowed at diagnosis; 62% (263/424) were married and 20% (84/424) were widowed at time of survey. No differences were reported by treatment about being forced into retirement. Discussion Compared to SOs, SAs were more likely to be currently employed, to receive higher rates of support from employers and coworkers, and to volunteer more hours. Perhaps interventions can be developed to allow ostomy survivors to participate more in these functions in their communities.     

PS1-22 Studying Communication over the Cancer Care Continuum: A Feasibility Study

Background/Aims Effective communication is an essential component of high quality cancer care. Preventing breakdowns and mitigating their harmful effects requires a thorough understanding of the communication process, including the number and types of interactions that a patient experiences over time and the impact of these experiences on the patient. This pilot study investigated the feasibility of "communication logs" to collect primary data on the number, type, mode, and impact of patient-provider communications over the cancer care continuum. Methods English speaking patients, >21 years old, with recent cancer diagnoses were recruited from 4 CRN sites. Clinical staff identified patients meeting eligibility criteria. Patients who agreed to participate were interviewed about their previous communication experiences and asked to maintain a communication log for 4 months, noting all encounters with cancer care providers. Logs could be completed on paper, web-based, or telephone and prompted recording of who was present during the encounter, whether it was in-person, telephone, or email, key content, and global ratings of the communication experience. Participants who logged were asked to complete a post-log interview focused on the experience, timing, and perceived burden of logging. Results Three CRN sites recruited a total of 9 cancer patients. One site was unable to recruit any patients. Four types of cancer were represented (1 thyroid, 1 sarcoma, 1 bladder, 6 breast cancer). Patients' ages ranged from 41 to 69 years. Only 4 patients completed logs; 5 patients reported on their communication experiences in retrospective interviews. Five patients described 10 or more provider encounters in detail. All patients who completed logs chose the web-based option and reported no difficulty submitting logs online. Reports of communication experiences varied ("He [the provider] didn't seem interested. It was like he'd rather be somewhere else." vs. "He [the provider] never made me feel like I was annoying him with my questions."). Main factors contributing to a helpful communication experience included perceived provider sensitivity and sufficient information provided about diagnosis or treatment. Discussion Initial results revealed difficulty recruiting patients early in cancer care. Those who did participate provided valuable information on their experiences, and reported that logging was not burdensome.     

腹腔镜下直肠癌经肛门外翻切除吻合12例治疗体会

【目的】探讨腹腔镜直肠Dixon术治疗低位直肠癌的方法与疗效。【方法】对12例低位直肠癌患者行腹腔镜直肠Dixon术,术中将结肠和肛管外翻切除吻合,腹部不另作切口,常规放置肛管和经腹腔放置盆底引流管。【结果】手术时间平均150（120-260）min,术后1-2 d胃肠道功能恢复并下床活动,术后住院时间平均8.4（7-11）d。12例病人全部痊愈出院,无吻合口瘘,2例患者术后出现尿潴留,1例患者术后同时出现尿潴留并反复里急后重肛门刺痛感,经治疗后痊愈。【结论】该手术安全可行,且创伤小,疼痛轻,恢复快。术中不在腹腔断肠,减少了腹腔污染及种植转移危险;不在腔镜下腹腔内肠吻合,大大降低费用。     

PS1-61: Implementing CHESS eHealth Breast Cancer Support in Population-based Care

Background/Aims The woman who is newly diagnosed with breast cancer suddenly faces significant new needs. She must learn about her disease and its treatment, cope with emotional distress and interpersonal issues, and make decisions about her treatment, work and family obligations. To answer the need for information and support, an Interactive Cancer Communication Systems (ICCS) called CHESS has been well-researched and validated to provide these women access to integrated information for decision-making, behavior change, and emotional support. Methods This newly-funded dissemination and implementation project will offer CHESS to all newly diagnosed breast cancer patients as part of standard care in two Denver healthcare systems, Kaiser Permanente and Exempla St. Joseph's Hospital. This project is designed to evaluate the real-world implementation of this new technology by answering the "how much" and "why" questions of utilization at both the institutional and patient levels. We will assess how effectively CHESS is offered to patients at each institution as part of standard care, what proportion of patients utilize the CHESS system and how often, why some women use it and some do not, to what extent women use the system; how staff perceptions and actions affect patient perceptions and use, if CHESS use is related to patient satisfaction and quality of life, and how organizational strategies affect CHESS adoption and use. Discussion The data from this project will be relevant to decisions about whether and how to disseminate CHESS and other interactive cancer communication systems, especially in health care systems where the technology might be integrated with other aspects of patient care.     

PS1-15: Illustrating Changes in Cancer Variable Composition Over Time: Developing Documentation for Cancer Researchers

Background/Aims Cancer researchers often use variables that evolve over time due to changes in national standards, such as AJCC (American Joint Committee on Cancer) staging or Collaborative Staging versions. The evolution of these standards impacts the construction of composite variables such as Stage AJCC in the VDW, and Extent of Disease in the local Group Health data environment. Knowledge of how cancer data registry items evolve over time and how they map into composite variables will aid researchers and programmers design cancer research studies and maintain cancer data infrastructure. Methods We are experimenting with ways to document cancer variables that effectively illustrate their evolution over time and the relationships among them. Currently these consist of time lines that are delimited by the item number from NAACCR (North American Association of Central Cancer Registries) and annotated with other information. Time lines of items that are related under the same concept are drawn adjacent to each other. When appropriate, the diagrams demonstrate how Group Health has populated VDW and local composite variables through each period of changing standards. Results Early response from cancer researchers has been positive. The diagrams have been especially useful in assisting in the querying of cancer counter data via a complicated NAACCR ontology (on i2b2) and aiding discussions around how Group Health should build its VDW Tumor StageGen composite SEER Summary Staging variable. Discussion The documentation of the composition of classes of cancer variables addresses a need in the cancer research community. Continuing to explore ways to illustrate the cancer variables is a low cost and efficient way of improving understanding around these topics for a broad audience and improving data use and quality.     

Effectiveness of Digital Health on the Quality of Life of Long-Term Breast Cancer Survivors: A Systematic Review

ObjectivesTo identify, critically appraise, and synthesize the available evidence on the effectiveness of digital health interventions to improve the quality of life or any of its four dimensions (physical, psychological, social, and spiritual) in women survivors of breast cancer who are in the extended or permanent survival stage.Data SourcesSystematic review—Four databases were searched: PubMed, CINAHL, PsycINFO, and Web of Science.ConclusionThe clinical evidence shows a positive relationship or association between eHealth use and improved quality of life in breast cancer survivors at extended or permanent survival stage. However, the findings point to a deficit in the assessment of the social and spiritual domains that play a fundamental role in the quality of life of survivors.Implication for Nursing PracticeThe findings found reflect implications of great value for nursing practice because these professionals are the main users of digital health tools to provide them to patients. Using these digital tools contributes to improving evidence-based practice and providing greater efficiency and effectiveness in the care of long-term cancer survivors.     

PS1-17: Assessing Communication During Cancer Care: Multiple Perspectives, Multiple Considerations

Background/Aims Patient-centered communication (PCC) between clinicians and patients is necessary for quality care. Measurement of communication over the cancer care continuum is an important step in describing current practices and remediating breakdowns. As a first step in developing a system to assess PCC longitudinally across the cancer care continuum, we interviewed patients, family members, physicians, health information technology (HIT) leaders, and health plan leaders. The purpose of the present study was to expand our understanding of possible approaches to create and implement such a system by conducting a symposium with experts and stakeholders. Methods We conducted a symposium with experts in the field of HIT, academic researchers, investigators from the 14 Cancer Research Network sites, information technology (IT) advisors from each CRN site, National Cancer Institute staff, and investigators and consultants involved with the project. The purpose of the symposium was to have experts in the field of HIT present possible approaches to computerizing data collection and feedback of patients' experiences with PCC and to engage in an active dialogue around the strengths and weaknesses of the different approaches proposed. Prior to the meeting, we distributed a list of questions for the IT experts to address in their presentations, based on issues, which have been identified during the interviews. The symposium was digitally recorded, professionally transcribed, and coded using grounded theory approaches. Results Themes identified during the symposium focused on contextual factors to be considered when creating a communication feedback system; data liquidity, using one system in multiple healthcare environments that are very diverse, and hat modality to use when there are a variety of preferences for communication and the technology is constantly evolving. The symposium also confirmed themes that emerged from the interviews; the system would need to be flexible and provide choices, the data obtained would need to be aggregated at a meaningful level, and feedback must be timely. Discussion Our multiple pronged approaches have provided us with unique insights that will be used in the development of a prototype data collection system, pilot testing of the system, and ultimately implementation of the system in multiple health plans.     

PS2-39: Key Challenges and Decisions in the Development of HMO Death Data

Background The mortality status of an HMO's members, whether a person is alive or dead, is critically important to health care research. The uses of mortality status include determination of causes of death, death rates, and for selection of study cohorts. Unlike other clinical activity and health statuses, mortality status cannot be determined solely with a health plan's internal data, even for well-defined populations, because health plan members often die outside of the plan's care delivery and claims payment systems. For such members, determination of mortality status and date of death requires linkage of health plan data to external sources of mortality data, such as state death certificate or federal Social Security data. This person-level statistical linkage is a complex, multi-step process that involves many decisions, assumptions, and choosing of priorities. Even when death data from internal systems is available, internal data sources can disagree, likewise requiring complex decisions to determine a member's "true" mortality status. Methods This presentation highlights key challenges and decisions points in the development of death data logic at Kaiser Permanente Northern California (KPNC). Results Key decisions in KPNC's death data logic include: handling internal membership data with conflicting personal identifiers, e.g., multiple Social Security Numbers associated with a single Medical Record Number; ranking, grading, and selecting from multiple matches between internal and external data; finding members with multiple MRNs; using utilization contact dates and membership activity dates to evaluate death records, judging reliability of internal health plan data, setting thresholds and criteria for confidence scoring, selecting between multiple possible dates of death, chart validation of final death data. Conclusions The development of death data logic at KPNC is a complex process that requires intricate decisions, careful analysis, and a higher degree of discretionary judgment than is typical of building other data areas. The quality of final results depends upon validation and iterative improvements to the process.     

PS1-25: Cervical Cancer Testing and Follow-up in Four Managed Care Plans, 1998-2007

Background/Aims Cervical cancer screening is performed to detect pre-cancerous cervical intraepithelial neoplasia or invasive cancerous cervical lesions prior to the onset of symptoms so they can be removed before the cancer has developed or spread. With the addition of high-risk human papillomavirus (HPV) testing to the long-established Papanicolaou (Pap) smear, national and health plan screening guidelines have been regularly updated in recent years. Guidelines on screening frequency and follow-up protocols have also changed over time and will likely continue to do so as HPV vaccination becomes more widespread. Here we describe patterns and results of cervical cancer testing and follow-up over a 10-year period within four geographically-dispersed U.S. managed care organizations. Methods Using data collected by the SEARCH: Screening Effectiveness And Research in Community-Based Healthcare project, we analyzed electronic medical record data on all women aged 20-65 during the period 1998-2007 across four HMORN sites. We created standardized files for Pap smear dates and results; cervical histology dates, types, and results; and HPV test dates and results. We also collected Virtual Data Warehouse data on HPV vaccinations, and selected diagnosis and procedure codes. We calculated rates of Pap testing, HPV testing, colposcopy, and cervical histology (biopsy and treatment). We also calculated rates for Pap testing we classified as "screening." Among women who had Pap tests in 2002 and 2007 and no abnormal test directly preceding the index test, we examined patterns of screening frequency. We also examined frequencies and trends in the results of "screening" Pap testing and cervical histology. Results Overall, annual Pap testing rates decreased and HPV testing rates dramatically increased over the 10 years, while rates of colposcopy, cervical histology, and cervical treatment did not display obvious patterns. Trends varied by age group and health plan. Pap screening frequency differed by health plan; overall, in 2007 a higher proportion of subjects had longer screening intervals (2 years or greater) than in 2002. Information on patterns of Pap and histology results will be presented. Discussion Evaluating trends in cervical cancer testing and follow-up may highlight opportunities to optimize cervical cancer screening delivery in community-based settings.     

PS1-05: Disparities in Breast Cancer Survival by Molecular Subtype and Race/Ethnicity

Background/Aims Little information exists on breast cancer survival rates according to molecular profiles among different race/ethnic groups. We investigated the impact of molecular subtypes on breast cancer-specific survival by race/ethnicity in a large group of medically-insured women diagnosed with breast cancer. Methods The cohort included 20,749 diverse Kaiser Permanente Southern California (KPSC) women diagnosed with breast cancer (AJCC Stage 0-IV) between 1996 and 2007, and followed through 2009. The women's cancers were classified into four main molecular subtypes: luminal A (ER+ and/or PR+/HER2-); luminal B (ER+ and/or PR+/HER2+); basal-like ("triple negative", ER-/PR-/HER2-); and HER2+/ER-. The outcome was breast cancer mortality. Follow-up began on the date of surgery and ended on date of death, health plan disenrollment, or study's end. Hazard rate ratios (HR) and 95% confidence intervals (CI) were estimated using Cox proportional hazards models. We adjusted for age, tumor characteristics, cancer treatments, income, and comorbidity. Results Of the 20,749 women, 65% were white non-Hispanic (n=13,487); 13% Black (n=2,697); 12% Hispanic (n=2,490); and 10% Asian (n=2,075). We observed 2,019 deaths (10%) deaths due to breast cancer over 14 years of follow-up. In all race/ethnic groups combined, breast cancer mortality was higher in women with basal-like (HR 2.90, 2.38-3.53) and HER2+/ER- (HR 1.98, 1.55-2.54) tumors compared to women with luminal A subtype (referent group). In addition, we examined breast cancer mortality in each molecular subtype, stratified by race/ethnic group. Among women with luminal A tumors, Black women were more likely to die of breast cancer (HR 1.53, 1.02-2.29) than white women (referent group). In women with luminal B tumors, South Asians had a 10-fold increase in breast cancer mortality (HR=10.57, 1.42-78.64) compared to whites; however, numbers were small and the confidence interval was wide. Among women with basal-like tumors, Black women had a greater mortality risk (HR 1.36, 1.02-1.82) compared to white women. In women with HER2+/ER- tumors, breast cancer mortality was similar across the race/ethnic groups. No other comparisons were significant. Discussion Despite similar access to healthcare, we found survival disparities by race/ethnic groups within the luminal A, luminal B, and basal-like molecular subtypes of breast cancer.     

PS1-45: Evaluating the Quality of VDW Pharmacy Data

Background/Aims The HMO Research Network (HMORN) Virtual Data Warehouse (VDW) is a series of dataset standards and automated processes that aim to streamline the process of multi-site research. The Pharmacy file contains data on all outpatient dispensings captured within the HMORN. The objective was to assess the overall quality and completeness of the Pharmacy file. Methods The VDW Pharmacy Working Group created a data verification protocol to assess the overall quality and completeness of Pharmacy file data (e.g., to identify missing data or out-of-range values). A distributed SAS program was run at each HMORN site that maintains a VDW Pharmacy file (n=14 of 19 HMORN sites), and de-identified summary data were returned for analysis. Pharmacy file variables that were assessed included National Drug Code (NDC), days supplied, amount dispensed, and prescribing physician (which joins to the VDW Provider Specialty file). An "invalid" NDC was defined as any value not having exactly 11 digits or containing a non-numeric character. The days supplied variable was considered out-of-range if the value was < 0, = 500, or missing. Amount dispensed was defined as out-of-range for values < 0, = 1000, or missing. Results Fourteen HMORN sites had Pharmacy data from 2000-2009 and participated in this analysis; some sites had > 20 years of data. There were 93.4 million dispensings in 2009, with an average of 7.8 million dispensings per month among 3.1 million monthly users (average per user: 2.6 dispensings per month; range across sites 2.3-5.8). Across all sites from 2000-2009, 0.082% (712,131/870,182,026) and 0.072% (629,656/870,182,026) of dispensings had missing or "invalid" NDCs, respectively; and 0.083% (724,210/870,182,026) and 0.315% (2,736,756/870,182,026) had out-of-range days supplied and amount supplied, respectively. The prescribing physician was identified in 96% (1,038,725,660/1,079,370,265) of all dispensings. The prescriber's specialty was identified in 62% (539,944,688/870,182,078) of dispensings overall and improved from 61% (46,678,731/77,078,104) in 2000 to 63% (60,160,629/95,867,812) in 2009. Discussion The VDW Pharmacy file has excellent overall data quality results, and has improved in identification of prescriber specialty. It has measurably comprehensive and consistent outpatient dispensing data across 14 HMORN sites.     

Managing Chronic Pain in Cancer Survivors Prescribed Long-Term Opioid Therapy: A National Survey of Ambulatory Palliative Care Providers

Context

Chronic pain, or pain lasting more than three months, is common among cancer survivors, who are often prescribed long-term opioid therapy (LTOT).

PS1-24: Validation of AJCC TNM Staging for Breast Tumors Diagnosed Before 2004 in Cancer Registries

Background/Aims AJCC Tumor (T), Nodal (N) and Metastatic (M) staging is commonly used in clinical practice for treatment decisions. However before 2004, most cancer registries following the North American Association of Central Cancer Registries (NAACR) standards reported only SEER Summary Staging, including the Kaiser Permanente Northern California Cancer Registry (KPNCCR). Information on extent of disease (EOD; extent and size of primary tumor, presence of metastases, lymph node involvement) was recorded as required by NAACR and SEER. The KP Southern California Cancer Registry (KPSCCR) also recorded TNM staging since 1988, in accordance with the American College of Surgeons. We developed an algorithm to map EOD to AJCC TNM stages for female breast cancer cases diagnosed from 1988-2003 identified from the KPNCCR. This algorithm was validated using KPSCCR data from 1996-2003 (n=17,135 cases) with available EOD and TNM information. Methods : The SEER Comparative Staging Guide for Cancer (1993) and the AJCC Cancer Staging Manual 6th edition (2002) were used to derive TNM stages (AJCC Version 6) from KPNCCR EOD variables for breast cancer cases diagnosed from 1988-2003. This algorithm was then applied to KPSCCR EOD variables for breast cancer cases diagnosed from 1996-2003. T, N, M and stage from this algorithm were compared to T, N, M and stage ("gold standard") available in the KPSCCR. Test characteristics (percent agreement, Cohen's kappa, sensitivity, specificity) were calculated. KPNC and KPSC distributions of cancer diagnoses by stage from 1996-2003 were compared. Results Agreement between the derived and available T, N, M variables was excellent (percent agreement and Cohen's kappa): T 93% and 0.91; N 96% and 0.94; M 100% and 0.93. Consistent high sensitivity and specificity were observed except for somewhat lower sensitivity for Stage III (sensitivity, specificity): Stage 0 (0.99, 1.00), Stage I (0.97, 0.98), Stage II (0.91, 0.96), Stage III (0.69, 0.99), Stage IV (0.92, 1.00), Overall (0.92, 0.89). Distributions of derived TNM staging of KPNC and KPSC cases from 1996-2003 were similar. Discussion SEER EOD variables for breast tumors available in SEER-accredited cancer registries can be used to accurately map to AJCC TNM staging for conduct of epidemiologic studies using cancer registry data.     

Associations between Health Behaviors,Health Self-Efficacy,and Long-Term Outcomes in Survivors of Childhood Cancer: A Cross-Sectional Study

ObjectivesMany survivors of childhood cancer experience multiple long-term chronic health conditions. Health behaviors are important because they contribute to chronic disease yet are highly modifiable. With growing pressure on cancer services, alternative models of care are required to address survivorship care needs. The authors sought to inform the development of a community-based model of cancer survivorship care for young people. This exploratory cross-sectional study aimed to assess the feasibility of study measures and processes, as well as investigate associations between various modifiable health behaviors, health self-efficacy, quality of life, and persistent symptoms.Data SourcesParticipants were recruited from a long-term follow-up clinic for childhood cancer survivors. A self-report survey was completed, and participants received an activity tracker. Bivariate regression analyses were used to explore the relationship between variables.ConclusionsThe study measures and processes were deemed feasible with >70% of eligible survivors enrolling in the study and completing >70% of study measures. Thirty participants (mean age 22 ± 4.4 years) were enrolled; 83.3% finished treatment ≥5 years previously and 36.7% were overweight or obese. Bivariate regression identified those with higher scores of health self-efficacy were more likely to meet physical activity guidelines, as were those who achieved more sleep and consumed greater servings of vegetables. Meeting the physical activity guidelines, was significantly positively associated with higher quality of life and self-efficacy.Implications for Nursing PracticeInterventions that target health self-efficacy have the potential to improve a range of health behaviors and long-term outcomes for survivors of childhood cancer. Nurses are ideally placed to use this knowledge to support patients with recommendations to optimise their recovery and rehabilitation.     

PS1-34: Overcoming Barriers for Tumor Block Retrieval and Testing in a Multi-site Cancer Research Network Project

Background/Aims KRAS testing is a pharmacogenomic test to help decide how to treat patients with metastatic colorectal cancer (mCRC). The CERGEN (Comparative Effectiveness Research in Genomics & Personalized Medicine for Colorectal Cancer) project, a collaboration among eight institutions of the NCI-funded Cancer Research Network (CRN) and four academic partners, utilized KRAS test results to evaluate overall survival in mCRC patients. As a part of this project, we needed to retrospectively test all patients that had not had the testing done as part of their clinical care. This project is a large, multi-site CRN project utilizing archived tumor tissue. Methods For cases that had not been previously tested for KRAS mutation status at the seven data collection sites, we tested archived pathology specimens from the primary tumor. Testing was planned to be performed at one central laboratory that was not affiliated with any of the institutions. Each site needed to identify study patients that had not been KRAS tested, retrieve blocks that could be utilized for research, section blocks per laboratory requirements, and return testing results to the lead site. Results All sites required varying degrees of approvals, paperwork, and sample processing. Despite many challenges, we successfully performed KRAS testing on tumor tissue from over 400 blocks. This poster will present the issues and challenges encountered at each of the seven sites in performing testing on tumor blocks, including: varying IRB requirements at each site; challenges around IRB ceding; site-specific biospecimen committees; timing around slides sent to the laboratory; staffing challenges in pathology; and varying material transfer and data sharing agreements. Additionally, the pathology department at one site required in-house testing only. We will also share our successes including: pilot testing the process before implementation; shipping blocks to another site for sectioning when there were pathology staffing shortages; working to standardize IRB applications; and working with the laboratory to ensure quality control in all of the samples. Discussion We were successfully able to standardize tumor block testing for research across seven HMORN sites. We hope that our "lessons learned" are helpful for other HMORN projects that wish to use archived tissue.     

PS1-31: Processes of Cardiovascular Care Relevant to Complex Patients

Background/Aims Clinicians commonly face decisions about prevention of coronary heart disease following a new diagnosis of cancer. Such decisions must take into account cancer prognosis, cardiovascular risk status, overall burden of morbidity, and the patient's goals, preferences and values. In cases of favorable cancer prognosis and elevated CHD risk, continued or intensified treatment of risk factors is warranted. In cases of poor cancer prognosis and/or high morbidity burden, a less intensive approach may be appropriate. Methods Study design: Historical cohort analysis. Population: 10,313 persons with a cancer diagnosis between January, 2001 and December, 2008 and a range of SEER 5-year survival categories. We assessed LDL cholesterol (LDL) control, blood pressure control, and hemoglobin A1c control each as a function of morbidity burden (using Charlson score), cancer 5-year survival, and the interaction between them in specific at-risk sub-populations. We used linear mixed models with random intercept for each patient and splines at cancer diagnosis and 6- and 12- months post cancer diagnosis to model each outcome occurring less than 2 years prior and 5 years post cancer diagnosis. Results Better LDL goal attainment was associated with higher morbidity in sub-populations characterized by need for cholesterol treatment. For persons with a diagnosis of hypertension, better blood pressure control was associated with lower morbidity burden. Neither LDL nor blood pressure goal attainment was associated with cancer stage at diagnosis. Both blood pressure and LDL control improved in the 6 months after cancer diagnosis. In persons with diabetes, attainment of hemoglobin A1c goal was not associated with either morbidity burden or cancer stage. Discussion In these sub-cohorts characterized by indications for cardiovascular risk factor control, control of risk factors was a function of morbidity, but not cancer stage. This raises the possibility that guideline-driven care of cardiovascular risk factors is neither intensified in the face of good cancer prognoses nor decreased in the face of poor cancer prognoses. These findings should be further explored with attention to detailed patient and clinician factors in order to make appropriate recommendations for patient-centered care delivery in complex patient populations.     

PS1-16: Patterns of Surgical Care for Ductal Carcinoma In-situ

Background/Aims Overall prognosis for ductal carcinoma in-situ (DCIS), a non-invasive breast malignancy, is generally quite positive with a mortality rate of < 2%. DCIS may progress to invasive breast cancer if left undetected or inadequately treated. Age, family history and individual preference are factors influencing quality and extent of surgical treatment of DCIS. Breast conserving surgery with or without radiation is commonly used to treat limited DCIS, with consideration for adjuvant endocrine therapy. Re-excision is not uncommon for DCIS, as it may be difficult to initially obtain a negative margin and also there may be suspicion of co-existing invasive breast cancer. Sentinel lymph node biopsy and prophylactic mastectomy have been considered over-treatment for DCIS, except in cases where DCIS is high grade, diffuse, large size, or has components of microinvasion. Our study aim was to evaluate patient characteristics and impact of features of DCIS on surgical patterns of care for DCIS. Methods Data relating to initial surgical treatment of stage 0-III breast cancer was collected both electronically and by manual chart abstraction to develop the multicenter Breast Cancer and Surgical Outcomes database. Study sites included the University of Vermont, Marshfield Clinic (Wisconsin), Kaiser Permanente (Colorado) and Group Health (Seattle). Women with a pre-operative diagnosis of DCIS (n=943) between 2003-2008 were included in this analysis. Results Of 943 women with a pre-operative diagnosis of DCIS, 932 (99.0%) had post-operative final diagnosis of DCIS. Planned analyses include determination of variability in surgical treatment of DCIS, including initial mastectomy and sentinel node biopsy, and incidence of positive margins following initial surgical excisions. Variability will also be examined based on factors unique to this dataset such as surgeon-specific and regional variation patterns of surgical care. Patient characteristics, DCIS features, MRI use and prophylactic contralateral mastectomy rate will also be reported. Discussion Understanding the variability of surgical treatment for DCIS may lead to improved quality of care and better outcome for patients with DCIS.