A136 Abstracts itabine 14%, vinorelbine 11% or gefitinib 9%). Approximately one-quarter of patients were treated in third line and approxi- mately 10% received fourth line treatment. The vast majority of third and fourth line treatments involved single agents. Toxici- ties associated with drug therapy were consistent with those that have been previously reported elsewhere. CONCLUSION: In the first line of therapy, patients received care largely reflecting the existing NSCLC evidence base from controlled trial data avail- able during the 2001–2003 period. Treatment patterns and out- comes of patients in community-based practices represent a potential rich source of data to complement controlled trial data. To improve the availability of real-world practice data, further work is necessary to overcome limitations of claims-based oncol- ogy data, to enhance the development of analyzable electronic health records, and to establish treatment registries. PCN41 DRUG COST CONSIDERATIONS FOR ERYTHROPOIETIC STIMULATING THERAPIES (ESTS) AGENTS IN PATIENTS INITIATED AT FDA-APPROVED DOSING: RESULTS FROM PRACTICE PATTERNS IN A PROSPECTIVE OBSERVATIONAL STUDY Pashos CL 1 ,Wang Q 2 , Larholt K 1 , Bookhart B 3 , Mck enzie RS 3 , Piech CT 3 1 Abt Associates Inc, Lexington, MA, USA, 2 Abt Associates, Inc, Bethesda, MD, USA, 3 Ortho Biotech Clinical Affairs, LLC, Bridgewater, NJ, USA OBJECTIVES: Two ESTs received FDA-approval for fixed initial dosing in cancer patients with chemotherapy-induced anemia: 40,000 units for epoetin alfa (EPO) and 500 mcg for darbepo- etin alfa (DARB). Understanding cost considerations, data was analyzed from the Dosing and Outcomes Study of Erythro- poiesis-Stimulating Therapies (D.O.S.E.) Registry, an ongoing, prospective registry collecting data on real-world practice pat- terns. METHODS: Data from 18 U.S. hospital and community- based outpatient practices were assessed from January 2006–December 2006. Chemotherapy-treated adult patients ini- tiated on either EPO 40,000 Units or DARB 500 mcg were eval- uated. Outcomes assessed included: mean administered dose, mean treatment duration, dosing patterns, and mean cumulative administered dose. EST cost was based on dose and 9/2006 wholesale acquisition cost (EPO $12.17/1000 Units; DARB $4.446/mcg) with sensitivity analysis based on 4Q06 ASP + 6%. RESULTS: A total of patients (145 EPO, 23 DARB) were eligi- ble for analysis. Patient groups were similar with regard to base- line age, gender, tumor type, and Karnofsky score. The predominant dosing pattern was QW for EPO and Q3W for DARB. The mean administered dose was EPO 42,879 Units and 497 mcg in DARB group, corresponding to an EST cost of $522 and $2210 per injection. Treatment duration and number of office visits was similar between groups. Mean cumulative administered dose was 305,241 Units for EPO and 1665 mcg for DARB. The corresponding EST costs were $3715 for EPO and $7404 for DARB (p < 0.0001) with similar findings based on sensitivity analysis. CONCLUSION: Practice pattern data from this observational study of cancer patients initiated at FDA- approved fixed dosing reported significantly lower costs in the EPO group compared to the DARB group. Mean cumulative drug cost was $3689 less (50% less) in the EPO group compared to the DARB group. Findings provide further understanding of anemia management costs for health care professionals, hospital systems, and patients. PCN42 A MULTI-PART EVALUATION OF A CANCER SYMPTOM MANAGEMENT INFORMATION TECHNOLOGY SYSTEM Mark TL 1 , Fortner B 2 , Johnson G 2 1 Thomson Medstat, Washington, DC, USA, 2 Supportive Oncology Services, Inc., Accelerated Community Oncology Research Network, Inc, Memphis, TN, USA OBJECTIVES: The Patient Assessment, Care & Education (PACE) SystemTM was designed to address the persistent problem of under-identification and treatment of chemotherapy- related symptoms. The PACE SystemTM uses a pen-based e/Tablet that operates off a wireless network. Cancer Support NetworkTM on the e/Tablet provides educational materials to patients in text, video, audio, and graphic format. The PACE System TM also administers the Patient Care MonitorTM, a psy- chometrically validated, patient-reported symptom severity screening scale that generates a real-time, point-of-care report for the provider. The aim of the study was determined provider and patient opinions of the PACE System TM and to determine whether symptom assessment rates increased after the PACE System was implemented METHODS: Ninety-two providers (i.e., physicians, nurse practitioners, physician assistants) at 16 community oncology clinics were surveyed about their experi- ences with the PACE System TM In addition, 100 patients at two community oncology clinics were surveyed about their percep- tions of the PACE System™. At two oncology clinics 100 patient charts were abstracted in the year prior to implementation of the PACE system and in the year after the implementation the PACE system to determine symptom assessment rates. RESULTS: The majority of patients reported that they were generally satisfied with the PCM (55%). Slightly more than half indicated that it helped them to remember symptoms, although only 44% said it encourage them to discuss their symptoms. 91% of respondents said the e/tablet was easy to use. The majority of providers thought that the PCM increased the frequency with which symp- toms were identified and treated. The results from the chart review show statistically significant increases in the assessment rates for depression, pain, and fatigue after the PACE system was implemented. CONCLUSION: The PACE system appears to be a promising approach to addressing the widespread problem of under-identification and under-treatment of symptoms in patients undergoing cancer treatment. PCN43 ADHERENCE TO GUIDELINES FOR USE OF ERYTHROPOIESIS STIMULATING PROTEINS IN PATIENTS WITH CHEMOTHERAPY-INDUCED ANEMIA:TRENDS FROM ELECTRONIC MEDICAL RECORDS Luo W 1 , Nordstrom B 1 , Ranganathan G 1 , Linz H 1 , Stokes M 1 , Ross SD 1 , Knopf K 2 1 United BioSource Corporation, Medford, MA, USA, 2 Pacific Hematology/Oncology Associates, San Francsico, CA, USA OBJECTIVES: Two national evidence-based guidelines recom- mend initiation of erythropoiesis stimulating proteins (ESPs) in chemotherapy-induced anemia (CIA) at hemoglobin (Hb) levels of °Ü 10 g/dL per ASH/ASCO guideline (2002) and <11 g/dL per NCCN guideline (1998), and maintenance of Hb near but not over 12 g/dL. The extent to which these guidelines are followed in actual practice is unknown. This retrospective study examined the patterns of ESP use in cancer patients with CIA. METHODS: The Varian Medical Oncology database of electronic medical records (EMRs) from 17 outpatient oncology practices in the US was utilized. Adults with a malignant tumor diagnosis between January 1, 2004 and July 27, 2006 who received at least one cycle of chemotherapy were studied. The proportion of patients