Search

New models of survivorship care are needed that improve outcomes for the growing number of cancer survivors, address the increasing complexity of their health needs, and deal with the shortage of clinicians and rising costs of this care. Technology can aid the delivery of personalized, stratified survivorship care pathways where the intensity of care, the care setting, and the providers required for that care vary with survivors’ needs. Building a cancer data ecosystem of connected data streams that supports and learns from each patient can be used to streamline care, enhance efficiency, reduce costs, and facilitate research. This manuscript describes the input, analytics, and output components of the cancer data ecosystem that must be built and connected and also provides a real-world use case of how such a system could transform care in a large US comprehensive cancer center.

Fifteen years ago, the Institute of Medicine (IOM) issued a report that defined Survivorship Care as a distinct phase of the cancer care continuum. The required domains to meet the health needs of cancer survivors were outlined in the report: cancer surveillance and screening, cancer prevention and lifestyle counseling, management of treatment related persistent or late effects, coordination of care, and psychosocial support services. In response to that report, The University of Texas MD Anderson Cancer Center implemented a tiered survivorship care model that is risk based. The core principle is that cancer survivors’ health needs will depend on the cancer treatment and disease-specific risks. We here describe this model for low-, intermediate-, and high-risk cancer survivors, in which comanagement between oncology and primary care providers is risk dependent. Our clinical model defines transition as appropriate when there is a minimal risk of primary cancer relapse, which is specific to each cancer type and disease stage. This model is embedded into disease-specific clinical practice algorithms, aligned with the IOM domains of care. Over the past 10 years, we have successfully transitioned nearly 25 000 patients to disease-specific survivorship clinics, providing care based on the IOM domains. We have learned from our process that expansion of survivorship care into established clinical settings requires engagement of champions and key clinical stakeholders. Future directions for survivorship care should explore the application and potential benefits of telemedicine as a care delivery system to meet the needs of cancer survivors.

Background This study reports long-term head and neck cancer (HNC) patient-reported symptoms using the MD Anderson Symptom Inventory Head and Neck Cancer Module (MDASI-HN) in a large cohort of HNC survivors. Methods MDASI-HN results were prospectively collected from an institutional survivorship database. Associations with clinicopathologic data were analyzed using χ2 , Mann-Whitney, and univariate regression. Results Nine hundred and twenty-eight patients were included. Forty-six percent had oropharyngeal primary tumors. Eighty-two percent had squamous cell carcinoma. Fifty-six percent of patients had ablative surgery and 81% had radiation therapy as a component of treatment. The most severe symptoms were xerostomia and dysphagia. Symptom scores were worst for hypopharynx and varied by subsite. Patients treated with chemoradiation or surgery followed by radiation ± chemotherapy reported the worst symptoms while patient treated with surgery plus radiation ± chemotherapy reported the worst interference. Conclusion HNC survivors describe their long-term symptom burden and inform efforts to improve care many years into survivorship.

271 Background: Patient reported outcomes (PRO) can be valuable clinical tools to embed the voice of patients into the clinical experience. The use of PROs in the context of cancer survivorship is relatively under-explored. Methods: Disease-specific modules of the MD Anderson Symptom Inventory (MDASI) PRO were integrated into the electronic health record (EHR), and distributed 3-7 days prior to survivorship clinic visits through the patients’ EHR portal. Patients with virtual visits were called 24 hours prior, instructed how to video link, and how to complete the MDASI PRO. For patients who completed the PRO electronically, results were reviewed during clinic visit. High symptom alerts (> = 7) for four key symptoms were established and triggered an automatic message to care teams for review. Results: From Sept 1, 2020 – August 31, 2021, 2,196 PRO surveys were sent to patients with 1,268 submitting results (57.74%) (see Table). Within specific clinics, the submission rates varied greatly between in-person and virtual appointments (33% vs. 83%). Submission rates varied by clinics, with responses ranging from 28.57% - 73.59%. Clinics with higher overall rates had a higher proportion of virtual visits. Conclusions: Electronic distribution of PROs can be successfully integrated into long term survivorship clinic visits. Virtual visits had overall much higher PRO response rates, due to pre-visit workflows which encouraged the completion of PROs. We conclude that patient education and real time support is needed to facilitate patients’ completion of PRO surveys through the electronic portal.[Table: see text]

Background: NCCN Guidelines recommend annual mammography for surveillance in asymptomatic women diagnosed with breast cancer. There is limited evidence to support which type of mammogram—screening or diagnostic—should be ordered for asymptomatic breast cancer survivors. Our objective was to assess the referral patterns for mammograms in 2 breast clinics: survivorship (SC) and oncology (OC). Methods: A retrospective analysis of institutional databases was conducted to identify a convenience sample of women with limited invasive breast cancer who were (1) alive 5 years post-treatment and (2) seen at a SC or OC visit scheduled from January 1 to December 31, 2015. The primary outcome was women who received a diagnostic or screening mammogram during the 2015 calendar year. Demographic, clinical, and mammogram characteristics were also analyzed. Simple descriptive statistics were used to aggregate and compare data. Chi-square analysis tested for statistical significance. Results: A convenience sample of 354 cases was identified. Of those, 247 met the eligibility criteria for this analysis, SC=147 and OC=100. In this cohort, the mean age of diagnosis was 50.42 (±11.12), the majority were white (74.1%), and most received a diagnostic mammogram (64.4%). A greater proportion of diagnostic mammograms were ordered for women seen in OC (40.1%) vs SC (24. 3%). This finding was statistically significant (P=.00). Overall, 91.9 % of the mammogram results were negative and with a low proportion of “positive” (2.0%) or “need for additional imaging testing” (6.1%) reported. Recurrence rates in this cohort were also found to be low (1.6%). Conclusions: In this cohort, we found a difference in the type of mammogram ordered by providers in dedicated survivorship or medical oncology clinics. Our findings suggest that despite the greater use of diagnostic mammograms among this cohort of long-term breast cancer survivors, the overall proportion of positive findings, further imaging tests, or recurrence rates was low. It is concerning that many asymptomatic breast cancer survivors continue to receive a diagnostic mammogram as part of their surveillance visit. Further studies are needed to identify the emotional, financial, and physical toxicities associated with diagnostic mammograms.

e24047 Background: After treatment, and at a time point of low risk of relapse, cancer patients at our institution transition to Survivorship Clinics (SC) that focus on wellness, late effects management, and psychosocial functioning in additional to cancer screening. We describe the self-reported symptoms of Head and Neck (HNC) survivors at the time of transition to SC, and the level to which these symptoms interfered with their function. Methods: Survivors reporting to their first in person SC appointment were assessed using the MD Anderson Symptom Inventory Head and Neck (MDASI–HN). This validated patient-reported outcome (PRO) instrument consists of 28 questions grouped into three subscales (core symptoms, disease specific symptoms, and interference items). Symptom severity is measured using a 0 to 10 scale (0 = not present, 10 = as bad as can be imagined). Descriptive statistics were used to characterize the sample. Percent of patients reporting moderate to severe symptoms (> = 4) are also noted. Results: This sub-analysis consisted of 377 patients seen for their initial transition visit to the survivorship clinic. Patients were mostly male (83%) and Caucasian (89%), with an average age of 63.8 years, and on average 41.5 months from diagnosis. Most common cancers were oropharynx (49%), oral cavity (33.7%), and larynx/hypopharynx (12.5%), nasopharynx (4.5%). Average symptom reports are presented in table. In rank order of prevalence (4% to 31% of patients), the top reported moderate to severe symptoms included dry mouth, swallowing/chewing, mucus, remembering, distress, and shortness of breath associated with moderated to severe interference in general activity in 8% of patients. Conclusions: HNC survivors report multiple persistent symptoms of moderate to high severity in a significant number, most commonly mouth dryness and swallowing and chewing difficulty. PROs can identify patients with high symptom burden, and lead in interventions to support the patients at their SC visit.[Table: see text]

335 Background: The recent pandemic led to an untypical demand for virtual care visits in our comprehensive cancer center. Our institution responded to this paradigm shift by offering virtual care to all cancer patients, including cancer survivors. In place of an in-person visit, patients were offered a live video or telephone (audio) appointment for their survivorship clinic visit. We conducted a quality improvement study to identify patterns in the type of virtual visits preferred by patients scheduled for appointments in eight dedicated survivorship clinics. Methods: We conducted a retrospective review of 4784 patients’ medical records who were seen in eight site-specific survivorship clinics from April 2020 – April 2021. Variables reviewed included type of visit, i.e. telephone, live video, or in-person, and patient characteristics such as age, gender, clinic, and distance from cancer center. Descriptive statistics were performed to assess for patterns of virtual care. The review was approved by the Quality Improvement Assessment Board. Results: Virtual care visits varied by clinic (0-86%). Five clinics ((breast, lymphoma, sarcoma, melanoma, and gynecological) performed < 3% of visits using virtual format. A sub-analysis of the 8 clinics performing virtual visits demonstrates 59.5% virtual activity (19.23% phone compared to 39.3% video). Older patients were more likely to use any form of virtual visit compared to in-person (58.9% for those ≥ 65 vs. 54.9% for 18-39). Older patients were more likely to prefer phone visits (25.5% for ≥ 65 vs. 11.2% for 18-39). Local patients were more likely to utilize virtual visits (63% for patients < 75 miles vs. 45% for patients > 500 miles). Patients who lived furthest from the cancer center were more likely to travel in for their appointment, though this population contained fewer patients than those within 75 miles. Use of virtual visits for survivorship care varied, depending on the types of clinical diagnostic tests or examinations required for an annual visit. Conclusions: Our results indicate the use of virtual care was optimal when only lab testing was ordered by their provider. Older adults, with no or limited access to and knowledge of digital technology (computers, smartphones, and others similar tools) may need additional support to engage in video visits. Our findings suggest virtual care can be successfully implemented in the care of survivors. Further research would help determine best-practice approaches to facilitating patient engagement in virtual visits since telemedicine continues to evolve as a standard method to deliver safe and optimal cancer care.

334 Background: Standard methods for implementation of electronic patient reported outcomes (ePROs) in the routine care of cancer survivors remain unclear. To address this gap, we assessed the technical and operational feasibility of implementing ePROs into the clinical workflow of four survivorship clinics, Genitourinary (GU), Head & Neck (HN), Thyroid (THY), and Stem Cell Transplant (SCT). Methods: This quality improvement study was conducted from 9/1/2019 to 4/30/2021. Benchmarks of ≥ 50% were established for two outcomes including: (1) Delivery Rate - percentage of patients who received the ePRO prior to their clinic visit and (2) Adherence Rate - percentage of patients who completed one ePRO. Data were collected at T0 (Baseline) (09/01/2019 to 12/31/2020) and T1 (1/1/2021 to 4/30/2021). Patients completed the survey 3-7 days prior to a clinic visit. The timing to capture ePROs’ data was tailored to the specific cancer site. Operational steps were: 1) questionnaire was ordered for scheduled survivorship appointment; 2) electronic medical record system (EMRS) pushed the questionnaire to the patient portal app with a message asking the patient to complete and submit the ePROs survey, 3) data was pushed back to the provider by the EMRS; and 4) results stored in the medical record were accessed by the provider prior to or on the day of the patient’s visit. Data were collected and stored using the EPIC Data Universes and reported through Microsoft Power BI. This institutional Quality Improvement Assessment Board approved this study. Results: The Table shows the improvement in the benchmark for sent and adherence rates between T0 and T1. A few clinics showed room for improvement. Patient engagement facilitators included being scheduled for a virtual visit and having staff sent prompts prior to visits. Challenges were: (1) variation in the time ePROs were available in the medical records, 2) inconsistent linking of scheduled visits and prompt to send the survey, and 3) patients’ not opening, completing, or submitting the survey even when prompted to do so. Opportunities included (1) developing a patient education flyer describing ePROs’ role in patient care, (2) providing instruction on the completion and completion of how to complete the ePRO, and (3) conducting staff education sessions on how to link scheduled visits with ePROs’ patient prompts. Conclusions: Our preliminary results indicate that integration ePROs can be considered an acceptable approach to improve follow-up care of cancer survivors but with room for improvement in some clinics. Future evaluation will further explore patients’ ePROs’ impact on clinical outcomes and patient engagement.[Table: see text]

Introduction: In 2009, our institution's leadership supported the creation of the Cancer Survivorship Program founded on core elements from the 2006 Institute of Medicine report “From Cancer Patient to Cancer Survivor: Lost in Transition.” These elements align with those recently posited in the 2019 Commission on Cancer (CoC) Standard 4.8. to establish and monitor a Survivorship Program. Here we describe our program model and present data collected over a decade to evaluate the program's outcomes. Procedures: Six key components guided the program's implementation and sustainability. These factors included: 1) inclusion of patients treated with curative intent, 2) recruitment of a transdisciplinary program staff, 3) delivery of clinical services based on 4 domains of survivorship care, 4) provision of consultative services available on-site or through referrals, 5) use of a quality improvement plan to monitor services provided, and 6) the creation and dissemination of treatment summaries and survivorship care plans. We reviewed electronic health records (EHR) of patients seen in the 12-survivorship clinics between 2009 and 2019. Data sources consisted of OneConnect, EPIC universes, and a departmental database. Data elements collected on arrived or missed appointments, no-show rate, and completion rates for the treatment summary and 30-day Passport (Survivorship Care Plans). Descriptive statistics summarized metrics per clinic and as a group. Results: Twelve site-specific cancer survivorship clinics scheduled appointments for survivors who had completed their curative treatment. A medical and operations director led the 6-member team who coordinated and monitored program activities. Patients had access to consultative services (i.e. cardiologists, nutritionists, social workers, etc.) provided by the institution or through community referrals. Table 1 summarizes data related to clinic visits for 10 years. In FY 2019, clinics with the highest number of arrived appointments included breast (4985), GU (1709), and head and/neck (1175). Clinics with the highest treatment summary compliance rates were lymphoma (97%), thyroid (77%), and stem cell transplant (72%). Each year, about 20% of patients scheduled are being seen for the first time in a survivorship clinic (rates vary by clinic). Conclusions: Results from our 10-year program experience suggest the newly proposed CoC survivorship program model can be practical and successful. Citation Format: Guadalupe Palos, Katherine R. Gilmore, Weiqi Bi, Patricia Hansberry Chapman, Kathy Carpenter, Maree Joy Pancho-Acorda, Maria Alma Rodriguez. A single center's ten-year experience in sustaining a cancer survivorship program [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5747.

e24073 Background: Head and neck cancer (HNC) survivors are known to experience debilitating disease-related symptoms and late effects. Patterns of symptoms among HNC survivors who have transitioned care to survivorship clinics could inform supportive care needs. We sought to assess patterns of self-reported symptoms in HNC survivors and to describe the level to which these symptoms interfered with their function. Methods: Symptoms and their interference were assessed using the MD Anderson Symptom Inventory Head and Neck (MDASI–HN). This validated patient-reported outcome (PRO) instrument consists of 28 questions grouped into 3 subscales; 1) 13 “core” or general symptoms, 2) 9 symptoms specific to HN and 3) 6 daily life interference items. Symptom severity is measured using a 0 to 10 scale, 0 = not present, 10 = as bad as can be imagined. Patients completed the MDASI-HN during a visit to the HNC Survivorship Clinic. Three databases a) MDASI HN, b) institutional, and c) the HNC survivorship provided data. We defined the primary outcome, symptom distress, as the combination of severity and interference rated ≥5 (0-10). Descriptive statistics were used to characterize the sample. Results: The sample included 1390 HNC survivors, majority were male (77%), Caucasian (83%) with a median age of 63 years (range 22- 95 years). The median years since diagnosis was five (range 2-37 years). The top disease sites were oropharynx (37%), oral cavity (35%), and larynx/hypopharynx (15%). The table summarizes the patterns of symptom distress. Conclusions: These results highlight the variation in symptom distress reported by long-term HNC survivors. Understanding the clinical implications of these patterns is important to ensure the well-being of those with head and neck cancers. [Table: see text]

11560 Background: Disease specific Survivorship Care Clinics (SCs) have been established within a comprehensive cancer center. Clinics are staffed by Advanced Practice Providers (APPs), Physician Assistants and Advanced Practice Registered Nurses, with experience in the management of each disease type. To determine the sustainability of this model of survivorship care, we analyzed the professional fees’ revenue generated by APPs’ billings for 6 clinics and then compared the APPs’ salaries across all clinics. Methods: A retrospective analysis was conducted of 6 survivorship clinic’s patient volumes and clinic days supported by APPs from 9/1/16-4/30/17. The full FTE salary of the APPs, including benefits were prorated to the time dedicated to each of the SCs. Institutional financial data was used to align professional fees to actual reimbursements received. Salary recovery percentage was calculated as the ratio of reimbursement received to prorated FTE salary. Results: Table shows variation in APPs’ salary commensurate to FTE proportion. Results also indicate there was an average of 99% professional fee recovery. Clinics with an FTE proportion > 0.5 had recovery higher than the anticipated prorated salary, suggesting there is a threshold to maximize efficacy and sustainability. Conclusions: APPs professional fees for care provided to cancer survivors are reimbursable, across disease types or payers, and proportionally supports their salaries. Our findings suggest delivery models based on APPs to manage care of long-term survivors can be self-supporting.[Table: see text]

Background: Despite numerous initiatives to improve the quality and safety of care for the growing number of cancer survivors, little empirical evidence is available about survivor satisfaction with post-treatment care or whether their information needs are met. Our specific aim was to conduct a comparative evaluation of 2 cohorts of survivors to determine their satisfaction with services provided in a breast survivorship clinic. Methods: We compared satisfaction in 2 separate cohorts of women who completed their curative treatment. Eligibility criteria for both groups included being a woman diagnosed with breast cancer and transitioned to “long-term survivor” status by their primary oncologist. Women were then grouped into no-visit, those who had never been to a breast survivorship clinic or completed-visit, survivors who had completed a visit to the survivorship clinic. Data was collected for the no-visit group in 2009, when the breast survivorship clinic was initially launched. Data collection for the comparative or completed-visit cohort is ongoing. The instrument consisted of 6 items reflecting 2 subscales: satisfaction and information needs. The first subscale focused on satisfaction with their visit, time spent with provider, and management of emotional distress. The second sub-scale focused on information received related to disease surveillance, late effects, health promotion, and social services. Survivors in both groups received the questionnaire upon their clinic arrival and completed it during their visit. Descriptive statistics were used to summarize all responses. All data were de-identified. Results: The no-visit cohort consisted of 81 breast cancer survivors who had never been seen in any type of survivorship clinic. Responses indicating that the survivors strongly agreed with an item were reported as follows: 79.5% satisfied with their visit, 77.3% information received, and 77.8% adequate time with provider. Of survivors who reported they were experiencing emotional distress, 50% strongly agreed that the clinicians in the clinic did all possible to address their distress, such as making referrals to the clinic social worker or to community-based mental health services. Bone health, cancer education, recurrence, and screening were the educational topics most requested. Final data collection and analysis for the comparative cohort, completed-visit, will be concluded by August 2013. Discussion: The no-visit cohort reported high satisfaction with survivorship care and information before they were seen in any type of institutional survivorship clinic. This finding could be a bias resulting from satisfaction with care received as a patient in active treatment. Results however, also suggest survivors of breast cancer who had not been seen in a survivorship clinic were aware of the type of specialized care or education needed to maintain their health after curative treatment. Further evaluation and research is needed to determine how satisfaction and unmet educational needs can be used to optimize outcomes for our growing population of cancer survivors. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-08-15.

This study's goal was to describe and begin to understand the experience of bereaved parents whose deceased child had received pediatric oncology services at a tertiary comprehensive cancer center. Focus groups were conducted with parents whose children were age 10 years and older at the time of death. Potential participants were contacted by mail and telephone. Sessions were audiotaped and transcribed verbatim. The ATLAS.ti qualitative software program was used to identify and analyze dominant themes. Fourteen parents identified four major themes: standards of care, emotional care, communication, and social support. Bereaved parents discussed the challenges associated with institutional procedures and interpersonal aspects of care in anticipation of and following their child's death. The results of these personal narratives may be used to guide care plans and deliver pediatric palliative and end-of-life interventions.

86 Background: Various models of care exist for the growing number of cancer survivors; yet little is known about the value or the difference in the models’ costs. We sought to assess the professional billing and reimbursements of 2 models of survivorship care; a) oncologist-led and b) advanced practice providers (APPs). Methods: Data were collected from 6 disease specific survivorship clinics, genitourinary, gynecology, head and neck, lymphoma, melanoma, and thyroid. Billing data was obtained to determine the professional billing and reimbursement amounts for these visits over one fiscal year. Comparisons were made between the APP and oncologists average salaries to determine costs per model. Clinic FTE salaries were based on the amount of time an APP devoted to a survivorship activities with 33% fringe benefits added to salaries. Oncologists’ salaries were based on the annual average faculty salary by department including fringe. Results: Billing data of 4,255 outpatient survivorship visits across the six clinics from 09/1/2016 - 8/31/2017 were abstracted. Table 1 summarizes the percentage (%) of professional reimbursement by model of care, oncologist versus APP and survivorship clinic. Conclusions: These data indicate that the advanced practice providers' model of survivorship care was financially more efficient than the oncologist model. Further research is warranted to compare the clinical effectiveness of these models and their impact on patients' outcomes related to their quality of life and satisfaction with care.[Table: see text]

e22090Background: Long-term head and neck cancer survivors are at risk to high symptom burden due to the late and lasting effects of the tumor and its treatment. Herein, we report on the results of...

16 Background: Opportunities for medical students and primary care providers to learn about evidence-based cancer survivorship care remains limited. To fill this void, we created an innovative curriculum using ECHO, a tele-mentoring approach, to teach faculty and residents in family medicine programs about the core principles and skills of survivorship care. Our objective was to explore participants' perceptions about their self-efficacy to deliver high quality care to survivors. Methods: Subjects were recruited from 3 academic centers with medical residency programs collaborating with our cancer center in a state-funded education grant to develop a tele-mentoring training program. The survey, which was developed by the research team, measured self-efficacy and satisfaction. Subjects were asked to rate their confidence in their abilities to offer evidence-based care after completing the program. Descriptive statistics were used to report results. REDCap, a web-based application, was used to manage the development and distribution of the survey. Results: A total of 116 surveys were sent out electronically with 59 returned (50.9%). Table 1 indicates most participants reported the ECHO program increased their knowledge, clinical skills, and expertise to provide comprehensive survivorship care. Conclusions: Overall, the participants’ perceptions of their expertise to offer optimal survivorship care were positive. There is room for improvement to maximize the learning experience for all participants.[Table: see text]

26 Background: Research indicates survivors of head and neck cancers (HNC) are at greater risk for recurrences and adverse late effects. HNC survivorship algorithms are available to provide consistent and optimal care to this at-risk group. Here, we describe practitioners’ concordance with algorithms aimed at survivors diagnosed with cancers of the oral cavity (OC) oropharynx (OP), and larynx/hypopharynx (LH). Methods: We reviewed medical records, treatment summaries, and survivorship care plans of patients seen in the HNC Survivorship Clinic from 09/01/2011 to 08/31/2014. The primary outcome was providers’ rate of concordance (> 90%) with 3 HNC algorithms. Concordance rates (CRs) were measured and compared within and between groups. Results: 145 patients met the eligibility criteria. Demographics varied by cancer sub-type: 19.3% OC, 58.6% OP, and 22.1% LH. Most were white males with the mean age of 62.7 years at first visit during the study period. Table 1 indicates clinicians’ CRs ranged from 9.2% to 99.7%. Overall, CRs were highly consistent and met or exceeded the benchmark for H & P, chest x-ray, and dysphagia and lymphedema assessment. Body image assessment had the lowest CRS. Wide variation in CRs existed within and between sites for: T4/TSH, head/neck CTs, xerostomia assessments. Conclusions: Our findings indicate providers’ practice did deviate from recommendations listed in the HNC algorithms. Increasing clinicians' awareness and education may help achieve optimal adherence to survivorship algorithms. [Table: see text]

The paradigm shift to include survivorship care as part of the cancer care continuum underscores the critical need for a change in nursing practice. One way to ensure that change in practice is delivered in a safe and efficient manner is through the use of clinical decision tools (CDTs). Such tools can be used to increase relevant knowledge and skills of nurses and patients. Despite the widespread recognition of their value, CDTs to educate providers on cancer survivors’ care are limited and, when available, often are not used. Clinical practice algorithms were developed for disease-specific survivorship clinics in a cancer academic center. This article reviews the conceptual framework of the survivorship algorithms, describes the application of the algorithms in multidisciplinary disease-specific survivorship clinics, and discusses the implementation strategies used to promote clinicians’ adoption and implementation of the algorithms. At a Glance • The authors found that algorithms can be successfully used as clinical decision tools(CDTs) to deliver survivorship care. • Algorithms and other CDTs are powerful tools to enhance professional practice. • Additional studies are needed to assess their effect on clinical practice and survivor outcomes.

e21602 Background: We developed institutional algorithms for cancer survivors that are site specific. However, little data exists regarding concordance with these clinical management tools. Our aim was to evaluate concordance rates with our Head and Neck (H/N) survivorship algorithms. Methods: We reviewed electronic health records (EHR) of patients seen in the H/N Survivorship Clinic between 9/01/2011 and 8/31/2014. Eligibility included: diagnosis of primary head and neck cancer, ≥ 18 years of age, no active disease, and alive at time of abstraction. Data were collected at baseline (V1) and visits 9-15 months after prior visits (V2 and V3). Data were obtained from tumor registry, a departmental database and the EHR. Twelve data elements were grouped as: 1) lab/diagnostic tests, 2) physical and psychological late effects, and 3) history and physical exams. Concordance rate (%) for each element was defined as a clinical action or recommendation aligned to the standards outlined in the H/N algorithms. Descriptive statistics were used to summarize clinical, demographic, and concordance rates. Results: Included in this analysis were patients with primary H/N malignancies of oropharynx, oral cavity, larynx, and hypopharynx, and who had completed 3 visits (V1-3). 53 patients were evaluable for concordance analyses. Compliance ranged from 55% to 100%, with lowest rates for: CT scans (55.6%), distress screening (75.5%), T4/TSH level (78.4%) and laryngoscopy (79.2%). Highest rates were noted for full physical exam (100%), chest x-ray (97.5%), dental exam (95.9%), and dysphagia evaluation (95.1%). Assessments for lymphedema and xerostomia were similar (91.8%). Conclusions: High compliance ( > 90%) occurred when elements of the algorithm relied on the clinician’s assessment, whereas lower levels of compliance were seen with elements that required orders (CTs and labs), or reliance on others’ actions (i.e. distress screening by nursing). Concordance can be improved using best practice advisories in the EHR to remind providers of orders, and by fostering strong inter-professional communication on the elements of care.

93 Background: A core measure of the ASCO Quality Oncology Practice Initiative and Commission on Cancer Program Standard is adherence to use of written treatment summaries (TSs) and care plans (CPs) for cancer survivors. Previous data from ASCO’s quality improvement (QI) studies indicate inconsistent use and distribution of treatment summaries and care plans. Audits conducted of adherence rates among our survivorship providers indicated use varied across clinics. We developed a plan to improve use of TSs and CPs among providers and improve survivors’ outcomes. Methods: The Plan-Do-Study-Act QI model was used to identify gaps in performance and process failures that led to those gaps. The project used stratified purposive sampling to select clinics. Based on the disease site, staffing structure, and clinical infrastructure, the breast, genitourinary, and head/neck survivorship clinics were chosen. First, we identified primary elements of the processes detailed in clinical practice algorithms specific to each site. Next qualitative interviews were conducted with providers to identify strengths and barriers to the use of TSs and CPs. Interviews were audio-recorded, transcribed verbatim and analyzed to confirm themes. Descriptive statistics were used to summarize demographic and clinical data. Results: A total of 16 members of the 3 clinics were interviewed. Of these, 37.5% were providers, 31.3% administrators, 18.8% support staff, and 12.4% faculty champions. Major themes identified as hindering adherence rates centered on cumbersome scheduling processes i.e. “there’s so many people that have to “touch an order” to get it to happen”, fragmentation among electronic medical records systems, i.e. “the way to fix it would be if it was a smart automated document that took the information…there in the chart," limited support among other health care professionals about survivorship care, i.e. “when push comes to shove Survivorship gets pushed. That’s the nature of the beast.” Conclusions: The barriers identified can be used to develop feasible solutions to improve adherence rates among survivorship clinic providers. Findings from this QI project contribute to deeper understanding of the complex issues involved in improving outcomes for survivors.

94 Background: ASCO has recognized the need to expand its educational portfolio to focus on long-term care of cancer survivors. Academic programs across health care professions overlook this topic in their curricula. To address this gap, we developed a survivorship research internship for undergraduate students enrolled in public health or health education programs. Here we present the results of an evaluation to assess the level of satisfaction of trainees and improve the curriculum. Methods: All students who completed the internship were eligible to participate in the evaluation. The authors conducted the survey using an electronic data system. Respondents were asked to rate their level of satisfaction with the curriculum and specific assignments, i.e. conducting their own survivorship research project, taking classes about PubMed or EndNotes, and abstracting data from medical records to use in survivorship care plans. Other measures were whether a trainee had written/published an abstract/manuscript, pursued an advanced degree, or obtained a job in health care after completing the program. Descriptive statistics were used to summarize results. Results: An 85% response rate was achieved (34/40 respondents). The majority were female (82.4%), Caucasian (56.3%), and enrolled in public health programs (41.2%). Almost all respondents (90.9%) strongly agreed or agreed they had a better understanding of the process needed to conduct survivorship research. 50.0% reported being most satisfied with the internship experience. 36.4% reported data abstraction and participating in PubMed or EndNotes classes were least useful. 35.3% reported observing health care professionals in the clinical setting was the most useful. 88.2% strongly agreed or agreed that the training was useful in their current work position.18.8% have achieved advanced degrees in medicine, nursing, or public health. Two of the 34 trainees have published a paper in a peer-reviewed journal or presented a poster at a scientific conference. Conclusions: Results indicated the vast majority of trainees had a positive experience engaging in the survivorship research internship. This strategy is one method that can be used to build and enhance education in survivorship care and research.

56 Background: The Commission on Cancer’s recent mandate stated that accredited programs issue treatment summaries (TSs) with survivorship follow-up care plans (SCPs). ASCO’s model of survivorship care also supports the use of these documents. One of the primary purposes of the documents is to enhance coordination and communication between the oncology team and primary care providers. Here we describe the experience of a survivorship program in using electronic health records (EHR) to develop TSs and SCPs. Methods: An interdisciplinary team at an academic cancer center was appointed to develop clinical tools to facilitate the creation and dissemination of TSs and SCPs. Enhancements were made to an institutional off-the-shelf EHR system that automatically populated available treatment information to the TS. This system used SmartLinks to pull data from the primary source of entry (e.g. surgical history, chemotherapy administered, and cancer stage). Clinicians edited and added pertinent information not automatically generated using one of the 19 disease-specific templates that provided lists of common treatments for various cancers. Electronic routing functions existed to share TSs with external providers through the medical records department. Results: From March-Sept, 2016, 766 SCP were completed by 50 providers in 14 clinics. Reports were created in the EHR to track SCPs and TS metrics. Data was reported from the TS on both a patient level and aggregate level by provider and clinic. Patient level data allowed providers to track incomplete TSs and edit them directly from the report. Of the TSs completed, 528 (69%) have been shared with patients and 261 (34%) have been shared with their community-based providers. Conclusions: EHRs provide a mechanism to successfully create and share TSs and SCPs among team members and primary care providers. They promote patient-provider education and communication about follow-up care. Research is needed to determine how they enhance coordination and ultimately outcomes for long-term survivors.

Oncology nurses must become better prepared to conduct quality improvement projects that will optimize quality of care and patient safety for long-term cancer survivors. The growing interest in survivorship care has led to the availability of multiple versions of cancer survivorship care plans (SCPs). Despite the availability of SCPs, research is lacking evidence-based processes to evaluate whether providers comply with planning and issuing SCPs. In the current article, the authors describe exploratory efforts to monitor the providers' compliance rate in issuing SCPs in diverse disease-specific clinics.

Purpose: To assess providers’ concordance with surveillance and risk reduction recommendations for colorectal cancer (CRC) survivors after completion of their curative treatment. Patients and Methods: This was a longitudinal study of survivors who met the following eligibility criteria: diagnosed with a primary colon or rectal cancer before their first visit (V) to the CRC Survivorship Clinic, adult survivor ≥ 18 years old, no evidence of disease, alive at the time of data abstraction, and 1-3 clinic visits between 9/01/2011 and 8/31/2014. Data were collected at V1 scheduled between 9/1/2011 and 8/31/2012. V2 was scheduled 9-15 months after the first visit. V3 was also scheduled 9-15 months after V2. Data sources were survivorship care plans, electronic medical records, and CRC survivorship algorithms. For an annual visit, CRC algorithms recommended history/physical exams (H & PEs), carcinoembryonic antigen (CEA) testing when previously elevated, and colonoscopies for surveillance of cancer recurrence. Concordance rates (CR) were measured as the percent of yes/no responses to whether the providers followed minimum standards for the 3 procedures. Demographic and clinical characteristics were also collected. Descriptive statistics were used to summarize all data. Results: 81 of 117 CRC survivors who met all eligibility criteria were included in this sub-analysis. The number of survivors visits varied across time, V1 = 81, V2 = 56, and V3 = 36. Most survivors were male (51.9%) and Caucasian (66.7%). 67.9% reported being 5-8 years post-treatment. 58% were diagnosed with colon cancer and of those 61.7% were Stage IIIA-IV compared to 55.8% of rectal cancer survivors with advanced disease. Table 1 summarizes the percentage of CR rates across the 3 visits. Conclusion: CRs for H & PEs and colonoscopies remained high across the 3 visits. CRs were lowest for CEA recommendations. These low rates suggest further work is needed to determine barriers in clinical practice that limit use of CEA tumor marker monitoring in CRC survivors. Percent of Providers ConcordanceAlgorithm RecommendationVisit 1Visit 2Visit 3H & PE100100100CEA53.338.322.2Colonoscopy97.598.2100 Citation Format: Guadalupe Palos, Katherine R. Gilmore, Patricia Chapman, Paula Lewis-Patterson, Weiqi Bi, Maria Alma Rodriguez. Low concordance with CEA tumor marker monitoring in colorectal cancer survivors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3464.

e21571Background: Surveillance for patients of colon (CC) and rectal cancer (RC) is a core dimension of quality survivorship care. Minimum standards of care (MSoCs) were embedded in clinical algori...

239 Background: Despite the growing committment to improve survivorship care, measuring its quality remains difficult. In 2006, the American Society of Clinical Oncologists (ASCO) launched a practice-based system known as the Quality Oncology Practice Initiative (QOPI) to measure the quality of cancer care. Here, we demonstrate the use of QOPI procedures to compare compliance with survivorship care plans (SCPs) across different disease-specific clinics. Methods: Measures based on survivorship clinical practice guidelines were abstracted from SCPs issued to survivors scheduled for appointments in site-specific clinics. Following QOPI procedures, we identified the target condition as cancer, selected data collection methods, and determined core measures for abstraction. Procedures included; identifying the number of survivors arriving for appointments within the specified time period, standardizing denominators and numerators used per clinic, determining the number of full-time providers per clinic, and calculating the number of records to audit per clinic. All data were obtained from completed care plans, institutional electronic medical records, and scheduling systems. Descriptive statistics were used to conduct aggregate analyses of de-identified data. The same procedures were followed in each clinic. Results: From September 1, 2014, to August 31, 2015, we used a standard process to assess compliance for issuing survivorship care plans across 9 clinics. There were a total of 8864 arrived appointments; 46 providers completed a total of 7448 electronic survivorship care plans over the time period. Analysis of compliance rates indicated broad variation across clinics, including: breast (85.3%), gastrointestinal (80.2%), genitourinary (88.5%), gynecology (78.6%), head/neck (96.8%), lymphoma (99.3%), melanoma (62.2%), thyroid (75.4%) and thoracic (63.3%). Conclusions: Establishing uniform procedures, such as the QOPI process, to measure and compare compliance with SCPs will help achieve high quality standards of care for long-term cancer survivors. Further examination is warranted to determine longitudinal trends and factors contributing to variation in compliance rates.

161 Background: ASCO’s Clinical Oncology Clinical Expert Statement on Cancer Survivorship Planning addressed the importance of using survivorship care plans (SCPs) to improve survivors’ care. Evolving information technology strategies, such as web-based systems, offer real-time reporting tools to track providers’ adherence for completion of SCPs. Our abstract reports on the creation of a web-based system to monitor survivors’ care. Methods: Ongoing audits of SCPs indicated broad variation in compliance rates with SCPs. The institutional benchmark for compliance was set at 90%. Yet, FY 2014 rates ranged from 7-97%, X= 84%. To address low or non-compliance rates, our institution selected a commercially available system as the IT intervention to improve compliance with SCPs. Core metrics identified were arrived appointments, missed appointments, completed SCPs, and transitions to survivorship status. To collect data, we used existing institutional databases with information on medical health, scheduled appointments, and other clinical documentation tools were used to collect data. Descriptive statistics were used to analyze and present data. Results: The current design of the Survivorship Performance Scorecard provided a centralized system for reporting data from 11 survivorship clinics. An exploratory review found 18 Passport templates across the 11 clinics, 188 OrderSets, and 986 scheduling codes were used in planning survivorship clinical services. Procedures for patient transition to survivorship clinics were defined in OrderSets, which were reviewed to identify appropriate fields for inclusion in the report. To ascertain compliance, procedures were developed to link arrived appointments with completed SCPs. Final reports will display all metrics by clinic, provider, and other patient level data. After the system goes live in late 2015, the systems benefits and shortfalls will be analyzed. Conclusions: This evolving tool offers a practice-based method for collecting data elements to monitor compliance with SCPs’ completion standards. Future work will focus on designing an alert system to inform providers when compliance rates are below the 90% compliance benchmark.

86 Background: Patient-provider communication about survivorship care and care plans is a core tenet of quality cancer care. Our objective was to examine gaps in communication about maintaining health, importance of follow-up care, and unmet information needs in community-based survivors. Methods: A convenience sample of adult survivors attending a community-based cancer survivorship conference were eligible to take this survey. Consenting adult survivors completed a self-administered survey with selected questions from the NCI Follow-Up Care Use among Survivors Survey. Participants were informed that survey completion was voluntary and confidential. Descriptive statistics were used to summarize and analyze data. Results: 284 surveys were distributed and 169 surveys returned, resulting in a response rate of 59.5%. After excluding Spanish-language surveys and respondents not diagnosed with cancer, 109 surveys were included in this sub-analysis. Most survivors were female (77.7%), college-educated (86.1%), married (58.4%) and diagnosed with breast cancer (47.5%). Over half (55.5%) reported experiencing pain that interfered with their daily routine yet, 83.2% reported being in good to excellent health. The primary reason for receiving follow-up care was to check for recurrence (66.2%). Most survivors (78.5%) reported having no to some patient-provider discussions about the late effects associated their treatment. 47.6% reported not receiving information about a written care summary from their provider, 94.8% would have liked to have received a written summary, and 73.1% preferred more information about the benefits of a personalized care plan. Survivors also reported in the past 2 years, their providers did not discuss topics related to emotional concerns (35.3%), alcohol consumption (35.1%) and types of foods to eat (33.8%). Survivors wanted more information on self-management techniques such as how to improve energy (83.2%), manage fatigue (72.9%), or improve sleep (69.3%) Conclusions: Our findings highlight the need for further examination on the impact of inadequate patient-provider communication on survivors’ outcomes and providers’ practice.

245 Background: Clear patient-centered recommendations for follow-up and health maintenance is a primary goal of ASCO’s model for survivorship and palliative care. The Commission on Cancer also sought to improve the quality of survivorship care by requiring accredited programs to implement survivorship care plans (SCPs) and treatment summaries by 2015. Our aim is to describe the SCP's components and metrics collected in 12 survivorship clinics of an academic cancer center. Methods: A multidisciplinary team was convened to develop clinical practice tools to guide providers on transitioning patients to a survivorship clinic. These tools were based on 4 IOM's domains: surveillance, risk reduction, late effects management, and emotional function.Algorithms are supported with a Passport Plan for Health (Passport) specifically designed for each disease. These clinical tools allowed clinicians to provide recommendations for each survivor. Tools were electronically available to survivors and their primary care providers. Descriptive statistics were used to summarize metrics in this analysis. Results: To date, 44 cancer-site specific algorithms and 19 Passports have been launched and used by clinicians in 12 survivorship clinics.Metrics collected are: a) number of Passports completed, b) providers’ completion of SCPs, and c) dates with counseling recommendations outlined in the treatment pathways. Since FY08, 25,912 Passports were distributed to 13,105 survivors. In FY14, 6,160 Passports were issued (80.7% provider completion rate). Passport completion varied by provider (7%-97%). Survivors were counseled for recommendations related to colonoscopy (97% of the time), skin screening (98%), mammogram (97%), gynecological screening (95%), and prostate screening (74%). Conclusions: Algorithms and SCPs serve as clinical tools to help providers and survivors manage physical and emotional issues. These tools increase standardization in the delivery of survivorship care, safety, quality, and promote communication among survivors and providers across clinical settings.

236 Background: Maintaining health and well-being is a primary goal throughout the cancer experience. The dynamic nature of a survivor’s trajectory makes it difficult to tailor education and supportive care services to their actual needs. We sought to determine information and education needs preferred by survivors who had returned to their communities after completing their treatment. Methods: We partnered with survivors' advocacy groups to conduct a needs assessment with adult survivors attending a community survivorship conference. Once eligibility was confirmed, verbal consent was obtained, and instructions were given on survey completion. A modified version of the NCI Follow-Up Care Use among Survivors survey was used. Analyses focused on clinical/sociodemographic characteristics, preferences for follow-up care, information/education needs, and sources for receiving health information. Simple descriptives were used to report the results. Results: We distributed 284 questionnaires and 169 were returned, yielding a 59.5% response rate. Respondents were mostly female (79.9%), married (58.6%), college educated (81.7%) and had ver 10 post-treatment years (47.3%). 51.3% were diagnosed with breast cancer and 81.8% reported being in good to excellent health. 57.3% had follow-up cancer care, 30.8% did not recall having a discussion with their provider about late-effects, and 57.6% reported not receiving a summary care plan. 83.7% preferred to have the same provider follow them and 95.2% reported their oncologist would be their ideal doctor. Survivors preferred more information on ways to improve memory/concentration (83.9%), energy/exercise (83.3%), and maximizing the benefits of their survivorship care plan (74.6%). Preferred information sources were health care providers (78.0%) and the internet (39.3%). Conclusions: These survivors reported a high level of unmet information needs related to self-management of physical symptoms. Patient-provider communication about late-effects and benefits of survivorship care plans continues to be inadequate. Physicians and other providers are in ideal positions to initiate conversations about supportive survivorship care early in the patient experience.

242 Background: Late effects, such as pain and fatigue, increase a cancer survivors' risk to psychosocial distress. Assessment and identification of distress and other risk factors can be integrated into survivorship care. Yet, clinical tools and processes needed to coordinate these efforts are severely limited. Here we describe the processes and metrics used to assess and document distress in the survivorship care plans (SCPs) of survivors diagnosed with colon and rectal cancers (CC and RC). Methods: Eligibility criteria for this longitudinal analysis included being: ≥ 18 years, diagnosed with a primary colon/rectal cancer, no evidence of disease ≥ 5 years, seen between 09/01/2011 - 8/31/2014, and alive at time of data abstraction. 117 survivors met these criteria and the first 50 sequential cases were included in this sub-analysis. Data were collected at Visits 1, 2, and 3 within specific time parameters. Data sources included the SCP, electronic medical records (EMRs), and CC/RC algorithms. Variables included demographic/clinical characteristics and standards of care (SOC) elements embedded in the algorithm related to distress, fatigue, pain, and social support. Compliance with SOC were measured using a “yes/no” response format. Results: Of 50 survivors, 58% were diagnosed with CC and 42% with RC. 52% were male, 64% Caucasian, 54% age 65-89 years, and 68% were 5-9 years from diagnosis. 52% of CC survivors were Stage IIIA-C or IV and 57% of RC were Stage 1-IIB. Documentation and metrics for each element indicated a total of 111 visits over the course of the analysis, 59% were CC and 42% RC. Provider's compliance rates (PCRs) were: distress (65%), fatigue (32%), pain (68%), and social support (46%). The pain element was not included in the CC algorithm. PCRs for each element varied by site. Evidence that providers followed algorithms' recommdendations to assess survivors current risk of distress was found in the EMRs, SCPs. Conclusions: We found cancer survivorship clinical algorithms and care plans can guide assessment, documentation, and monitoring of psychosocial assessment. Screening for risk factors to distress in cancer survivors can be done in real time and help identify at-risk survivors during their visit.

e20596 Background: Advances in the treatment of colorectal carcinoma (CRC) have contributed to an increase in long-term survivors. Despite the advances made in this cancer, there is relatively litt...

243 Background: The utility of algorithms to guide clinical practice related to long-term survivorship care is unclear. In 2012, lymphoma survivorship algorithms for Hodgkin’s and Large B-Cell were developed as part of an institutional clinical practice initiative to enhance survivors’ outcomes. Here we describe providers’ concordance with minimum standards recommended in the survivorship lymphoma algorithm. Methods: Concordance, the primary outcome, was defined as whether minimum recommendations, summarized in the lymphoma algorithms, were provided during a survivorship visit. Concordance was measured using a “yes/no” response format assessing whether a provider’s recommendations were consistent with algorithms’ minimum standards including: 1) physical examination, 2) laboratory and imaging testing, 3) cancer screening procedures, and 4) health promotion education. Data were abstracted from electronic medical records and Passports (summary care plans). Secondary outcomes were number of survivors seen and Passports issued. All data included in these analyses were de-identified. Results: 63 survivors were seen in the Lymphoma Survivorship Clinic from 05/01 to 09/30/2012 and all received Passports. A provider tailored the plan to each survivor’s needs and counseled the individual on disease surveillance, life-style changes, screening recommendations, and specific procedures to assess for recurrence/late treatment effects. Results indicated 100% of survivors received a physical exam, including cardiovascular screening. Similar rates were also achieved for CBC testing, chest X-rays, and counseling survivors on screening recommendations relevant to: breast, cervical, colorectal and prostate cancers. Other counseling rates included 97% for skin exam recommendations and 100% for maintaining adult vaccinations. Finally, providers’ concordance with assessment of survivors’ psychosocial distress was 100%. Conclusions: Based on concordance results, providers met the minimum recommendations listed in the lymphoma algorithms. Future research is needed to assess the impact of providers’ use of algorithms on survivors’ outcomes and feasibility of replicating concordance analyses in other clinical care settings.

111 Background: A fundamental tenet of survivorship care is to identify and deliver a wide range of supportive care services. Survivors rely on clinicians to provide integrated care personalized to their own supportive care needs. Yet there is lack of evidence demonstrating which method is “best practice” when assessing survivor’s needs. The aim of our Plan-Do-Study-Act quality improvement project was to determine the best method to identify survivors’ supportive care needs and coordinate services. Methods: We used the appointment list of the Breast Survivorship Clinic to identify 114 adult survivors seen for their initial survivorship visit between January-March 2014. Survivors were asked to complete a questionnaire assessing their needs related to: 1) energy-balance, 2) complementary medicine, 3) lymphedema, 4) psychosocial concerns, 5) sexual health, and 6) tobacco use. Survivors were randomly assigned into 3 groups to identify the best method to assess their health care needs: 1) Mail, 2) Secure Electronic Messaging, and 3) Navigation Visit with Health Educator (HE). Questionnaires for the mail and electronic messaging groups were distributed before being seen in the clinic. Navigation participants completed the questionnaire during the visit with HE. Descriptive statistics were used to describe and compare group characteristics. Results: A total of 114 patients were randomized to complete the assessment by either mail (N=28), secured electronic message (N=50), or during the navigation visit (N=36). The in-person, navigation visit had the highest completion rate (78%) compared to secured message (22%), and mail (7%). An overwhelming majority of survivors’ in the navigation group (93%) indicated they had needs in in at least one of the 6 domains. The top two needs across all groups were nutrition and physical activity. Conclusions: This evaluation suggests the use of a health navigator to guide the survivor through the needs assessment was the most successful way to assess needs of long-term breast cancer survivors. Enhanced navigation during a cancer survivor’s visit can be used to tailor integrative health services, which can improve their well-being and mental health outcomes.

112 Background: The Institute of Medicine noted cancer survivors’ care may be improved by providing each with a cancer survivorship care plan (CSCP). Recently, numerous versions of CSCPs have emerged; however, evidence-based processes to evaluate their use and quality in providing personalized survivorship care have yet to be implemented. Here, we describe our efforts to use the approach recommended by the American Society of Clinical Oncology (ASCO) Quality Oncology Practice Initiative (QOPI) to track and report the compliance rate for using electronic CSCPs in our survivorship clinics. Methods: The criteria used to conduct our audit evolved from the QOPI template, which included selection of survivors seen in 7 different out-patient survivorship clinics within a specific 6-month period, identification of denominators and numerators unique to each clinic, and determination of the number of records to audit based on the number of full-time providers in each clinic. Next, we established the target condition, method(s) used for data collection, appropriate measures, and other criteria used in the QOPI. All data were obtained from our CSCPs’ mined data, institutional electronic medical records, and scheduling systems. Results: From December 1, 2011, to May 31, 2012, we conducted an electronic medical records audit to determine the compliance rate for issuing survivorship care plans in 7 clinics. The survivorship clinics included breast, colorectal, endocrine (thyroid), genitourinary, gynecology, head/neck, and melanoma. A total of 3,274 electronic medical records were reviewed. 40 providers issued a total of 2,761 electronic CSCPs within 30 days of the arrived appointment. The compliance rate per clinic ranged from 64.5% to 94.7% (X=84.5%). Conclusions: We found the QOPI process to be user-friendly and applicable to our compliance monitoring effort. All data were extracted from electronic tools, thus ensuring the processes used were reliable, reproducible, and reasonably consistent across clinics. Further examination is warranted to determine characteristics of clinics with high- vs. low-compliance rates and their impact on survivors’ outcomes as related to specific CSCP recommendations.

9081 Background: We explored the association between activation of inflammatory networks and development of severe symptoms during mobilization and the acute phase of autologous hematopoietic stem cell transplantation (ASCT) for multiple myeloma (MM). Methods: From mobilization through the first 3 months after ASCT, multiple symptoms were measured repeatedly via the myeloma module of the M. D. Anderson Symptom Inventory (MDASI-MM). Serum levels of interleukin (IL)-6, IL-10, tumor necrosis factor (TNF)-a, soluble TNF receptors 1 and 2 (sTNF-R1, sTNF-R2), IL-1 receptor antagonist (IL-1RA), vascular endothelial growth factor (VEGF), macrophage inflammatory protein-1 (MIP-1a), monocyte chemoattractant protein (MCP)-1 and C-reactive protein (CRP) were collected up to 11 times per patient and assayed by Luminex. Ordinal regression modeling was used to analyze repeated-measures data. Results: Among50 MM patients, the most severe symptoms reported by werefatigue, pain, muscle weakness, disturbed sleep, drowsiness, numbness, poor appetite, and bone aches. Symptoms worsened rapidly from conditioning therapy and peaked at WBC nadir. Over time, controlling for age, sex, and MM stage, increased serum IL-6 (P=.0007) and MCP-1 (P=.0005) were significantly associated with worsening of the most severe symptoms; MIP-1a (P=.005) and VEGF (P=.002) were inversely associated with these symptoms. Increased CRP was significantly associated with worsening pain (P=.01), fatigue (P=.007), and bone aches (P=.006). Conclusions: This longitudinal study indicates a strong association between dynamic changes in circulating inflammatory molecules and the most severe symptoms in MM patients during the acute phase of ASCT. This observation, similar to IL-6-related multisymptom development around WBC nadir of allogeneic SCT (Wang et al, 2008), provides evidence of potential inflammation-induced behavioral changes in humans. Testing of anti-inflammatory interventions is warranted to further confirm the role of inflammation in the development of symptoms and identify effective mechanism-driven symptom management during ASCT.

8091 Background: Macrophage inflammatory protein-1α(MIP-1α) is a growth factor for human multiple myeloma (MM) cells. As an osteoclastic factor, its presence provides pathophysiological evidence of the development of lytic bone lesions in MM. C-reactive protein (CRP) indicates systemic inflammation. The relationship between disease-driven inflammatory markers and both patient-experienced symptoms and tumor response to induction therapy is unknown. Methods: MM patients (N=39) who were either newly diagnosed or had received fewer than 2 cycles of chemotherapy, and who also were to receive induction therapy, were enrolled. To test concentrations of MIP-1α and CRP, serum samples were collected before and after induction and assayed by Luminex. Multiple symptoms were measured twice a week via the M. D. Anderson Symptom Inventory MM module (MDASI-MM) from -8 to +112 days of induction. The MM-specific items of the MDASI-MM are bone aches, constipation, muscle weakness, diarrhea, sore mouth or throat, rash, and difficulty paying attention. Correlation between symptom severity and inflammatory markers at baseline was examined by linear regression modeling. Kruskal-Wallis significance test and Wilcoxon test were used to examine association between MIP-1α and tumor response. Results: Patients received either bortezomib-based (89%) or lenalidomide-based induction therapy. Baseline MIP-1α and CRP were significantly inversely related to the mean severity component score of the 5 most-severe symptoms (fatigue, pain, bone aches, poor sleep, drowsiness) (p=.03; p=.02), and significantly inversely related to the severity of a component score of the module-specific symptoms (p=0.04; p=.002). Change over time in MIP-1α differed significantly by tumor response category (p=.04), with the partial response group having a higher median score than the complete response group (1.483 vs. 0.016, p=.01). Conclusions: Our data suggest that higher baseline levels of serum MIP-1α and CRP predict effective chemotherapy-induced reduction of disease-related symptoms in MM patients. Higher serum MIP-1α expression after induction therapy was related to less-ideal tumor response.

370 Background: The symptoms reported by patients being treated for advanced pancreatic cancer (PC) could be driven by disease and treatment. Few clinical studies have addressed the interaction between disease, treatment, and symptom burden. We conducted a longitudinal study tracking multiple symptoms to provide an evidence-based symptom profile for patients with advanced PC receiving gemcitabine-based chemotherapy. Methods: From start of the first chemotherapy cycle (baseline), patients rated symptoms weekly during chemotherapy via the M. D. Anderson Symptom Inventory (MDASI), then continued symptom assessment every two weeks for up to 40 weeks. Longitudinal symptom responses were analyzed by mixed-effect modeling. Results: Among 100 PC patients enrolled, 87% were treatment-naïve. The median age was 65 yrs old (range 35-85). Half of the sample was female; 32% received opioids for severe pain. The three most severe MDASI symptoms at baseline were fatigue, lack of appetite, and pain. Poor performance status and opioid use were significantly related to severity of these baseline symptoms (all P Conclusions: Our results suggest that standard gemcitabine-based chemotherapy plus active palliative care for advanced PC was associated with significant improvement in appetite. However, worsening fatigue suggests a need to provide active intervention to minimize this symptom for PC patients. Patients suffering from greater symptom burden at baseline and overtime, seems consistent with greater disease burden, may be expected to discontinue treatment earlier.

Abstract 677FN2 Background: It is well recognized that patients with multiple myeloma (MM) who undergo induction chemotherapy experience treatment-induced numbness and tingling, indicating the development of peripheral neuropathy (PN) that can be a dose-limiting side effect. There is need to identify which patients are more at risk for this side effect, to understand its association with the development of other symptoms, and to explore possible biomarkers associated with its development. Inflammation has been proposed as one mechanism underlying PN. Methods: Patients with MM who were scheduled to receive bortezomib-based chemotherapy with or without lenalidomide before evaluation for autologous stem cell transplantation were enrolled on study. Enrollment generally occurred before induction therapy, but no later than two cycles of induction. Patients with pre-existing PN were excluded, but patients with diabetes were not. Patients repeatedly rated numbness/tingling, pain, and other symptoms via the M. D. Anderson Symptom Inventory (MDASI) during induction therapy (twice a week for 12 weeks, then weekly up to 16 weeks). Patients contributed a serum sample before start of every chemotherapy cycle. A panel of cytokines was examined by Luminex method. Mixed regression analysis was used to describe the trajectory of numbness/tingling over time and to examine the relationship of numbness to other symptoms during this period. Group-based trajectory modeling was used to identify a group of patients who developed more-severe numbness during induction, to determine baseline risk factors for numbness development, and to compare serum cytokine profiles between high-numbness and low-numbness groups. Results: Sixty-four patients were enrolled. During induction, numbness/tingling persistently worsened from baseline (P=.011). A diagnosis of diabetes without PN (10/64) was the only baseline clinical predictor of high numbness development (P=.01). Increasing numbness over time was significantly associated with pain (P=.005), disturbed sleep (P Discussion: This prospective, longitudinal study evidenced that in patients with newly diagnosed MM receiving induction therapy, 42% suffer from a rapid increase in numbness/tingling. Diabetes and high symptom levels at baseline were predictive of membership in the high-numbness group. An increase in numbness was associated with increases in other symptoms, including pain, disturbed sleep, and muscle weakness. We also observed a proinflammatory response (IL-1, IL-6, and TNF-alpha) during induction therapy that was associated with the development of more-severe numbness, warranting further translational studies to confirm the status of these cytokines as biomarker(s) related to PN and, ultimately, to develop mechanism-driven, effective patient care during standard induction therapy. Funding: This study was funded by NIH grant P01 CA124787 Disclosures: No relevant conflicts of interest to declare.

Abstract 4920 Background: Patients with multiple myeloma (MM) undergoing induction therapy experience both disease- and therapy-related symptoms. This longitudinal study investigated the association between the trajectory of symptom severity and changes in levels of inflammatory markers. Methods: Patients with MM repeatedly rated symptoms via the M. D. Anderson Symptom Inventory (MDASI) during induction therapy. Patients contributed serum samples before start of every cycle of chemotherapy. A panel of cytokines, including interleukin (IL)-6, soluble IL-6 receptor (sIL-6R), soluble IL-1 receptor type 1 (sIL-1R1), soluble tumor necrosis factor receptor type 2 (sTNF-R2), monocyte chemotactic protein-1 (MCP-1), macrophage inflammatory protein-1 alpha (MIP-1A), and IL-10, was examined by Luminex method. Ordinal regression analyses were used to describe the interaction of cytokines and symptom outcomes across time, adjusted for patient and clinical factors of age, sex, diabetes diagnosis, anemia, body mass index, co-morbidities, tumor stage, Eastern Cooperative Oncology Group performance status (ECOG PS), prior treatment status, tumor response, opioid use, and chemotherapy regimen. Results: Sixty-two patients were enrolled on study; of these, 89% received bortezomib-based induction therapy. During induction, the most severe patient-reported symptoms were (in order of severity): fatigue, muscle weakness, disturbed sleep, pain, drowsiness, bone aches, and numbness. Fatigue was persistently the most severe symptom, while therapy-induced neuropathy (MDASI numbness item) increased significantly from baseline (P=.01). We observed significant longitudinal associations between sIL-1R1 and distress and sadness (both P=.02); between sIL-6R and disturbed sleep (P=.001), poor appetite (P=.04), and sore mouth (P=.006); between IL-6 and pain, fatigue, nausea, and sore mouth (all P Discussion: Frequent assessment documented the longitudinal course of multiple symptoms during the induction period, providing an opportunity to evaluate potential changes in inflammation as one source of symptom distress. Our preliminary study suggests a network of pro- and anti-inflammatory cytokines relevant to symptom expression during induction. The results warrant further basic and translational studies to confirm the role of inflammation in the development of symptoms during treatment. Such information may suggest methods of effective symptom management and identify potential biomarkers for patients at high risk of symptom burden during treatment for multiple myeloma. Disclosures: No relevant conflicts of interest to declare.

9094 Background: Oxaliplatin-based chemotherapy (FOLFOX or FOLFIRI), a standard first-line chemotherapy for patients with colorectal cancer (CRC), is not only associated with dose-limiting toxicity...

e19627 Background: Patients with multiple myeloma (MM) undergoing induction therapy experience both disease- and therapy-related symptoms. This study selectively focused on the trajectory of pain, ...

e16594 Background: The aim of this study is to define the content domain for a patient-reported outcome (PRO) measure of gastrointestinal stromal tumor (GIST) symptom burden. Symptom burden, defined as the combined impact of symptoms and treatment on daily functioning, can be an effective endpoint in clinical trials to establish treatment benefits. Methods: This is Part 1 of a study to develop a PRO GIST symptom-burden measure. Twenty patients with GIST, recruited from a clinic at MD Anderson Cancer Center, described symptoms over their disease courses in single qualitative interviews. After the interviews, patients reported symptoms experienced in the last 24 hours on the core MD Anderson Symptom Inventory (MDASI) to provide context for the analysis. Content analysis was used to identify symptoms and define content domain. Results: Mean patient age was 57.9 (standard deviation [SD] = 13.1) years, 55% were male, 40% were employed, 85% were receiving tyrosine kinase inhibitor therapy, 25% had surgery withi...