To evaluate patient satisfaction with golimumab (GLM) and its auto-injector in rheumatoid arthritis (RA) patients switched from adalimumab (ADA) or etanercept (ETA) due to inadequate disease control. Data were from an interim analysis of 200 patients enrolled in GO-SAVE, a GLM multicenter, assessor-blinded, switch study of patients with active RA. Patients receiving methotrexate and having an inadequate response to current treatment with ADA or ETA entered the screening period at week -6 and continued current treatment. After re-screening at week 0, all eligible patients were actively switched to open-label GLM 50mg subcutaneous injections. Satisfaction prior to switching was assessed at week -2 for ADA or week -1 for ETA. Satisfaction with and preference for GLM were assessed at week 8. Mean (SD) age was 55.9 (11.3) years; 81.5% were female. Mean (SD) disease duration was 9.8 (9.7) years. Prior to switch, patients were treated with ADA pen (25.5%), ADA prefilled syringe (21.5%), ETA pen (27%), ETA prefilled syringe (24%), and ETA vial and syringe (2.0%). All 200 patients completed assessment at week -2 (ADA) or -1 (ETA); 170 completed the assessment at week 8 (GLM). At week 8, 82.9% were satisfied with the overall GLM experience, 80.0% were satisfied with injection frequency, and 75.3% were satisfied with injection device. Most patients experienced lower levels of injection related burning (80.0%), stinging (75.3%), discomfort (62.4%), redness (58.8%), and pain (57.6%) with the GLM auto-injector than with the previous injection device. Patients expressed greater preference for GLM over their previous medication (74.0%) and greater preference for the auto-injector (70.6%) over their previous injection device. A majority of RA patients switched to GLM from ADA or ETA were satisfied with their overall GLM experience, including preference for GLM and the auto-injector over previous medication and injection device.

e16507 Background: Two ESAs (darbepoetin alfa, epoetin alfa) are FDA-approved for reducing transfusions in patients (pts) with CIA. Clinical trial data have reported adverse safety signals in this population with ESA treatment targeting hemoglobin (Hb) > 12g/dL. In April 2007, UnitedHealthcare implemented a non-coverage policy for non-Medicare pts with Hb > 12g/dL. Previously, no Hb restrictions existed. The impact of this policy change on the proportion of ESA-treated pts with CIA requiring transfusion was evaluated. Methods: Retrospective claims of the i3 Innovus database from 7/1/2005 to 6/30/2008 were analyzed. Subjects were commercial enrollees ≥ 18 years with cancer and chemotherapy treatment, ≥ 2 ESA claims, and ≥ 1 qualifying treatment episode, which was defined as number of days from first to last ESA administration (no ESA claim for 90 days prior to first claim and 90 days after last claim). Subjects with claims for surgical procedures or dialysis were excluded. Proportion of pts transfused was evaluated comparing pts that initiated ESA treatment between 1/2006–9/2006 (prechange) and 7/2007–3/2008 (postchange). Results: 4,371 pts were identified; 2,775 and 1,709 had ≥ 1 ESA treatment episode in the pre-change and post- change periods, respectively. Age, gender, and tumor type were similar across both time periods. ESA treatment duration was longer in the prechange compared to postchange period (64.9 vs. 59.6 days; p < 0.001). In the prechange period, 11.5% of pts had ≥ 1 transfusions compared with 15.4% in postchange (p < 0.001). By subset analysis, the largest proportion of pts requiring transfusion was in those with lung cancer (19.8% pre, 29% post; p = 0.009). Conclusions: A small but significant increase in proportion of pts transfused and a small but significant decrease in ESA treatment duration was observed in the post-policy period. It is unknown if the increased transfusion rate was a result of policy implementation or the concurrent changes in the Centers for Medicare and Medicaid Services' coverage criteria for initiation of ESA therapy. Further study should evaluate the long-term impact of the policy on other clinical outcomes. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Centocor Ortho Biotech, United Healthcare Johnson & Johnson Johnson & Johnson

e16559 Background: Cancer patients often experience CIA and may be treated with ESAs. Recent observational data on PT-Hb levels in ESA-treated patients with CIA have not been reported. This analysis evaluated PT-Hb levels for this population from RCTs and the DOSE registry, a prospective, observational study of real-world ESA treatment patterns and hematologic outcomes. Methods: Data from ESA-treated cancer patients with CIA were obtained from 3 RCTs (Waltzman/Oncologist 2005; Henry/Curr Med Res Opin 2006; Glaspy/Cancer 2009) and from the DOSE registry (Larholt/Clin Drug Inv 2008). Transfused patients with PT-Hb measurements were identified. PT-Hb levels were defined according to RCT protocols or as the most recent Hb value 7 days before the transfusion episode (DOSE registry). Descriptive statistics summarize the analyses. Results: Of the 834 ESA-treated CIA patients from RCTs, 150 received a total of 257 transfusions, of which 245 (95.3%) had PT-Hb levels available. Of 1,810 ESA-treated CIA registry patients (enrolled 12/2003-5/2009), 364 received a total of 596 transfusions, of which 508 (85.2%) had PT-Hb levels available. The PT-Hb was significantly lower in RCT patients compared to registry patients (mean±sd: RCT 8.0±1.0 g/dL, registry 8.9±1.1 g/dL, p<0.0001). The PT-Hb range was wide for both groups (RCT: 4.7-13.1 g/dL; registry: 5.1-12.9 g/dL). In RCT patients, 42.4% of transfusions occurred at PT-Hb levels <8.0 g/dL and 1.9% at ≥10.0 g/dL. In registry patients, 15.8% of transfusions occurred at PT-Hb levels of <8.0 g/dL and 13.1% at ≥10.0 g/dL. Conclusions: Wide variance of PT-Hb levels in ESA-treated patients with CIA was reported from RCT and observational data. Mean PT-Hb was significantly lower in the RCT compared to observational data. Patient characteristics or site-specific transfusion policies that may contribute to the latter represent an area of future research. Group Transfusions based on PT-Hb range (g/dL), n (%) <7.0 7.0-7.9 8.0-8.9 9.0-9.9 >10.0 Missing RCTs 21 (8.2) 88 (34.2) 105 (40.9) 26 (10.1) 5 (1.9) 12 (4.7) Registry 13 (2.2) 81 (13.6) 192 (32.2) 144 (24.2) 78 (13.1) 88 (14.8) Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Centocor Ortho Biotech Centocor Ortho Biotech Johnson & Johnson Centocor Ortho Biotech

Abstract Background: Some metastatic breast cancer (MBC) treatments are associated with cardiac toxicity. Real-world data regarding baseline cardiac comorbidities in patients with MBC prior to chemotherapy initiation are scarce. Such information would be useful in guiding treatment decisions. This study aims to describe the cardiac comorbidity profile of patients with MBC initiating doxorubicin-based (DOX) versus non-doxorubicin-based (Non-DOX) chemotherapy regimens from a managed-care perspective.Methods: Medical claims from the Ingenix Impact National Managed Care Database between 07/2002 and 06/2008 were analyzed. Patients included were aged ≥ 18 years, and had ≥ 2 diagnoses for breast cancer as well as ≥ 1 diagnosis for metastatic site within 12 months of chemotherapy initiation. Cardiac comorbidities [hypertension (HTN), cardiac arrhythmia (CAr), coronary artery disease (CAD), congestive heart failure (CHF), and myocardial infarction (MI)] were evaluated within 6 months prior to chemotherapy initiation. Patients were categorized based on receipt of DOX vs. Non-DOX-based chemotherapy.Results: A total of 12,345 patients (Non-DOX: 7,023, DOX: 5,322) formed the study population. Compared with the DOX group, the Non-DOX group was slightly older. Both groups reported a significant proportion of patients with cardiac comorbidities prior to chemotherapy, with significantly greater proportions reported in the Non-DOX group. Mean age (SD)HTNCArCADCHFMINon-DOX (n=7,023)56.9 (10.6)35.7%11.8%5.5%3.5%1.3%DOX (n=5,322)52.1 (9.1)30.8%8.1%3.3%2.0%0.5%P-value<.001<.001<.001<.001<.001<.001 Conclusions: Cardiac comorbidities were commonly reported in women with MBC prior to chemotherapy, with significantly greater proportions reported in the Non-DOX group. Such information is useful to healthcare professionals when considering potential interventions for patients with MBC. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 2078.