Incorporation of a Comprehensive Geriatric Assessment (CGA) into a randomized phase III trial for metastatic breast cancer: The PELICAN Study

Abstract

9551 Background: The PELICAN trial is a randomized phase III trial evaluating efficacy and safety of pegylated liposomal doxorubicin (PLD) versus capecitabine as first-line therapy in patients (pts) with metastatic breast cancer who are not considered candidates for combination chemotherapy. Randomisation was stratified according to age (cut-off 65 years) and anthracycline pretreatment. CGA results and adverse events (AEs), dose reductions, and number of applied cycles were evaluated. Methods: CGA comprised data on activities of daily living (ADL), instrumental activities of daily living (IADL), ECOG and Karnofsky performance score, comorbidity (CIRS-G), and number of comedication. According to CGA, pts were classified into groups 1–3 (Balducci 2000). Associations of CGA groups and single CGA items with outcome measures were assessed using Fisher's exact test, Wilcoxon test, and logistic regression. Results: Results of 210 pts, mean age 61 years (SD 11; range 22–85) were analysed. According to CGA, 102 pts were considered fit (group 1, 72%), 21 compromised (group 2, 15%), and 19 frail (group 3, 13%). Age was associated neither with overall AEs, nor with dose reductions or number of applied cycles. However, the following significant associations were observed: any AE and number of comedication (p=0.026); less hand foot syndrome (HFS) and CGA group 2/3 (p=0.032); and more constitutional symptomes and age >=65 as well as CGA group 2/3 (p=0.002 and 0.015, respectively). Comedication was the variable most frequently associated with single AEs (dermatology other than HFS, p=0.038; gastrointestinal, p=0.035; constitutional symptomes, p=0.002; pain, p=0.001; pulmonary, p<0.0001; fatigue, p=0.01). Whereas no association between any item and dose reduction was found, CIRS-G grade 3–4 was associated with a reduced number of applied cycles (p=0.049). Conclusions: Incorporation of CGA into a phase III trial is feasible and yields information that would otherwise have been missed. Results of CGA should be validated prospectively in treatment algorithms for this population. [Table: see text]