Abstract
Abstract 3137
Older 5-hydroxytryptamine3 receptor antagonists (5-HT-3 RA), e.g., ondansetron and the more recent palonosetron are indicated for prevention of chemotherapy (CT) induced nausea and vomiting (CINV). Step therapy is a policy that encourages the use of generic 5-HT3 RAs, reserving prophylaxis with palonosetron as second-line therapy as means to containing costs.
To evaluate potential impact of step therapy policy on risk of severe CINV events among patients (pts) with lymphoma who used ondansetron, granisetron, or dolasetron and later switched to palonosetron versus pts who were initiated and maintained on palonosetron throughout multiple CT cycles.
Pts initially diagnosed with lymphoma during 2005–2009 were selected from PharMetrics claims dataset if enrolled for ≥6 months before first lymphoma diagnosis and received cyclophosphamide based CT and 5-HT-3 RA prophylaxis. Pts were followed (FUP) for 6 months from 1st CT date. Based on 5-HT-3 RA used at FUP, pts were grouped into those initiated and maintained on palonosetron throughout versus those initiated with an earlier 5-HT-3 RA and later switched to palonosetron. Outcomes included risk of severe CINV (using ICD-9 codes) associated with hospitalizations or ER admissions, as well as days with severe CINV. Logistic and Poisson regression models were used to compare risk and days with severe CINV, adjusting for cyclophosphamide dose received during the 6-month FUP, age, gender, Charlson Comorbidity Index (CCI), and the year of lymphoma diagnosis.
A total of 953 patients who used palonosetron throughout the FUP (palonosetron group) and 113 patients who switched from an older 5-HT-3 RA to palonosetron (switched group) were analyzed. The average (mean±SD) age at lymphoma diagnosis was 60.3±13.8 years, CCI was 0.47±0.97, number of cyclophosphamide treatment days was 5.6±1.93, and cyclophosphamide dose (mg/m2) per CT cycle was 930±676. Palonosetron group was significantly older (60.8 vs. 55.9 years, p=0.0006). There were no significant differences (NS) between the two groups in gender, CCI, or year of lymphoma diagnosis. Palonosetron group had significantly fewer 5-HT-3 RA prescription claims (4.81 vs. 7.72, p<0.0001). There was NS difference between the two groups in cyclophosphamide dose [palonosetron: 937 vs. switched: 871], p>0.05), although palonosetron group had significantly fewer CT treatment days (5.5 vs. 6.3, p=0.0135). Proportion of patients with ≥1 severe CINV events was significantly higher in the switched group (15.9% vs. 6.3%, p=0.0002). Switched group also had significantly more claims and days with severe CINV (2.4 vs. 0.6 in CINV claims, p=0.0001; 0.4 vs. 0.1 in CINV days, p=0.0002). Logistic regression controlling for covariates showed the palonosetron group to experience significantly lower risk of approximately 70% for hospital/ER related CINV than the switched group (Odds Ratio=0.311, p=0.0001), while older age significantly increased such an outcome. Poisson regression with number of days with severe CINV as the outcome variable found that palonosetron group had nearly 75% fewer hospital/ER related CINV days than the switched group (p=0.0024).
Patients with lymphoma initiated and maintained on palonosetron throughout the CT cycles were at significantly lower risk and experienced fewer days with severe CINV as compared to those who were initiated on an older 5-HT-3 RA and later on switched to palonosetron. Further studies of the impact of step therapy policy with regard to first 5-HT3-RA therapy in patients with lymphoma are needed in order to assess the related economic and humanistic burden.
Hatoum:Eisai: Consultancy, Research Funding. Lin:Eisai: Consultancy. Balu:Eisai: Employment.
Author notes
Asterisk with author names denotes non-ASH members.
