Duh, Mei Sheng

557 Background: Aromatase inhibitors (AIs) are a novel hormonal therapy for breast cancer. However, AIs can cause bone loss by blocking estrogen production. This study aims to assess the association between AIs and treatment-related bone loss in a large managed care population of women with breast cancer. Methods: Using medical and pharmacy claims data from over 5 million beneficiaries between 01/01/1998 and 01/31/2005, we identified 12,368 patients with ≥ 2 breast cancer claims in a 6-month period, who also had no bone metastasis and no prior osteoporosis or fracture claims. Patients who received anti-estrogen therapy were also excluded to remove the protective confounding effects. 1,354 patients receiving an AI (anastrozole, exemestane, letrozole) were compared to 11,014 controls who did not receive an AI with respect to their risk of bone loss. The observation start date for the AI and control groups was defined as the service date of the first AI claim and breast cancer claim, respectively. The bone loss endpoints analyzed were osteoporosis (including osteopenia) and clinical fractures. Results: The univariate analysis found that the prevalence of osteoporosis was 8.7% in the AI group vs. 7.1% in the control group, resulting in a statistically significant relative risk of 1.3 (95% CI=1.1–1.6, p=0.01). The prevalence of bone fracture was also statistically significantly elevated in the AI group compared to the controls (13.5% vs. 10.3%) with a relative risk of 1.4 (95% CI=1.2–1.6, p=0.001). Multivariate Cox proportional hazards regressions showed that after adjusting for age and comorbidities, the risk of bone loss remained statistically significantly higher in the AI group than the non-AI group, with 27% (95% CI=4%-55%, p=0.02) and 21% (95% CI=3%-43%, p=0.02) increase in the risk of osteoporosis and fractures, respectively. Conclusions: This retrospective longitudinal analysis of a large cohort of breast cancer patients corroborates previous findings from smaller clinical trials and demonstrates that AI therapies carry an increased risk of bone loss. Monitoring and treatment management strategies to alleviate bone loss risk are warranted in women receiving AI for breast cancer. [Table: see text]

Abstract Introduction Myelofibrosis (MF) is a chronic myeloid neoplasm in which 75% of patients have anemia and 25% are erythrocyte transfusion-dependent (TD) (Emanuel et. al. JCO 2012). MF patients receiving chronic transfusions are at risk of multiple co-morbid conditions and iron overload may develop with potential for end-organ damage. Iron chelating therapies (ICTs) help eliminate iron surplus by binding with plasma iron to form a non-toxic conjugate that can be safely excreted from the body. The objective of this study was to examine transfusion patterns and incidence of MF-related complications in TD MF patients treated with vs. without ICT from a US managed care perspective. Methods Two commercial claims databases, Truven Health Analytics MarketScan (2000-2012) and PharMetrics Integrated Database (2001-2012) were used to address the study objectives. Patients with ≥2 MF ICD-9 diagnosis codes (238.76, 289.83) ≥30 days apart, ≥18 years at the time of the first observed diagnosis for MF, and ≥6 months of continuous enrollment prior to the first observed evidence of transfusion dependency (index date), defined as ≥3 transfusion events within any 3-month period, were included (modified Gale et al. Criteria). A transfusion event was defined as a unique day when ≥1 procedure code for packed red blood cells, whole blood, or exchange transfusion was recorded. Frequency of transfusions and time from first evidence of TD to initiation of ICT were analyzed. Incidence of MF-related complications was assessed using incidence rates (IR) and compared between TD patients with vs. without ICT using adjusted incidence rate ratios (aIRR), adjusting for baseline comorbidities and complications. Results Characteristics of MF patients receiving ICT: Of the 571 TD MF patients who met the inclusion criteria, 103 (18%) received ICT and 468 (82%) did not. Mean (SD) age was similar between the two groups (with ICT: 67.2 (10.4) vs. without ICT: 66.6 (11.7), p=0.65), but the proportion of men was higher in the ICT group (with ICT: 70.9% vs. without ICT: 58.3%, p=0.02). The mean (SD) observation time was longer for patients with ICT than without ICT (months, 22.2 [13.9] vs. 12.6 [11.6], p<0.001). A greater proportion of TD patients without ICT had a history of essential thrombocythemia (with ICT: 11.7% vs. without ICT: 20.9%), whereas the proportion of patients with prior polycythemia vera was similar between the two groups (with ICT: 11.7% vs. without ICT: 13.7%). Differences in TD MF patients by utilization of ICT: Overall, patients without ICT were generally more ill than those with ICT (Charlson Comorbidity Index [CCI], with ICT: 1.8 vs. without ICT: 2.3, p=0.01). The mean (SD) number of transfusion events per year was similar between the two groups with 22.4 (19.5) events/year in the group with ICT compared to 22.2 (28.5) events/year in the group without ICT (p=0.94). Among patients receiving ICT, therapy was initiated after a median time of 5.6 months following the first evidence of TD. TD patients with ICT had lower rates of thrombocytopenia (aIRR: 0.54; 95% confidence interval [CI]: 0.40-0.74) and pancytopenia (aIRR: 0.53; 95% CI: 0.37-0.76), but higher rates of anemia (aIRR: 1.61; 95% CI: 1.27-2.02). The incidence of other MF-related complications considered was similar between the two groups. Conclusion In this first analysis of the utilization of ICT amongst MF patients whom are TD we identified several important differences between those receiving ICT compared to those who do not. Whether these difference are a result of ICT, or a reflection of whether physician practice leads to ICT utilization in certain subsets of MF patients (i.e. those with less thrombocytopenia/leukopenia) could not be assessed with the current data. Potential short term and long term benefits of ICT in MF need to be validated in prospective clinical trials. Disclosures Vekeman: Novartis Pharmaceuticals: Research Funding. Cheng:Novartis Pharmaceuticals: Research Funding. Sasane:Novartis Pharmaceuticals: Employment. Huynh:Novartis Pharmaceuticals: Research Funding. Kaminsky:Novartis Pharmaceuticals: Research Funding. Duh:Novartis Pharmaceuticals: Research Funding. Paley:Novartis Pharmaceuticals: Employment. Mesa:Novartis Pharmaceuticals: Research Funding; Incyte Corporation: Research Funding; Gilead Sciences: Research Funding; CTI: Research Funding; Celgene: Research Funding; Genentech: Research Funding; NS Pharma: Research Funding; Lilly: Research Funding.