1030P - Real-world comparative effectiveness analysis of second-line (2L) nab-paclitaxel (nab-P) vs eribulin (Erib) in patients (Pts) with metastatic breas...

Date 09 October 2016
Event ESMO 2016 Congress
Session Poster display
Topics Bioethics, Legal, and Economic Issues
Presenter Monika Parisi
Citation Annals of Oncology (2016) 27 (6): 351-358. 10.1093/annonc/mdw377
Authors M. Parisi1, S. Glück2, C. Pelletier1, Q. Ni1, F. Braiteh3
  • 1U.s. Health Economics And Outcomes Research, Celgene Corporation, 07901 - Summit/US
  • 2Pancreatic Cancer And Immuno-oncology, Celgene Corporation, 07901 - Summit/US
  • 3Oncology, University of Nevada School of Medicine and Comprehensive Cancer Centers of Nevada, 89169 - Las Vegas/US



nab-P and Erib are approved for ≥ 2L treatment (Tx) of MBC. However, no data are available on comparative effectiveness of nab-P vs Erib for 2L Tx of MBC in the real-world setting.


Fully de-identified data from a US electronic medical record platform of 1300 community (not university-based) oncologists were used in this retrospective study of women aged ≥ 18 yrs with MBC who started nab-P or Erib monotherapy as 2L Tx from 12/1/10 to 4/6/15 (≥ 1 cycle of nab-P or Erib required to be included). Time to Tx discontinuation (TTD, day 1 to last date + 7 days for 7-day cycle and day 1 to last date + 21 days for 21-day cycle) and time to next Tx (TTNT; day 1 of line 2 to day 1 of line 3) were the primary objectives. Adverse events (AEs) and supportive care were also examined. Subset analyses examined outcomes in pts with hormone receptor–positive (HR+)/human epidermal growth factor receptor 2–negative (HER2−) or triple-negative (TN) MBC.


This analysis included 176 pts (107 treated with nab-P and 69 with Erib). Baseline characteristics were similar between groups except that more pts treated with 2L nab-P had HR+ disease (P = 0.02) and were on combination Tx with a targeted agent (P = 0.03). More pts in the Erib group had used a taxane prior to MBC diagnosis (P = 0.01 for ≤ 1 year). Few pts with TN MBC were treated with 2L nab-P or Erib (Table). Overall, nab-P was associated with numerically longer TTD and TTNT vs Erib. AE rates were similar except that thrombocytopenia occurred more often with Erib. Antiemetics (IV) and G-CSF were used less often with nab-P. Steroids were used less often with Erib. Similar trends were reported in pts with HR + /HER2− or TN MBC.

nab-P n = 107 Eribulin n = 69 Unadjusted P Value Adjusted P Value
Overall population
Age, mean, years 60 61 0.688 -
Schedule, weekly, % 75 - - -
TTD, median, mos 4.0 3.1 0.074 0.507
TTNT, median, mos 6.0 4.7 0.538 0.759
Any grade AE in > 5% pts, %
Anemia 28.0 42.0 0.055 0.130
Neutropenia 18.7 23.2 0.470 0.376
Thrombocytopenia 7.5 18.8 0.023 0.107
Nausea + vomiting 11.2 5.8 0.222 0.299
Dehydration 8.4 2.9 0.205 -
Fatigue 2.8 7.2 0.266 -
Pain 2.8 7.2 0.266 -
Supportive Care doses/pt/100 days
Antimetic (IV Only) 5.77 6.92


In this US-based, real-world analysis, TTD and TTNT were numerically longer with nab-P compared with Erib, including in the HR + /HER2− and TN subsets. AEs were similar between the 2 groups. Pts receiving nab-P required less supportive care.

Clinical trial identification

This study is a retrospective analysis, therefore the protocol number and release data is not applicable.

Legal entity responsible for the study

Monika Parisi, MPH


Celgene Corporation


M. Parisi, S. Glück, C. Pelletier, Q. Ni: Mr: Employee of Celgene Corporation. F. Braiteh: AbbVie, ActiveBiotech, Amgen, AstraZeneca, Bayer, BMS, BIND, BioMarin, BioTheranostics, BN Therapeutics, BI, Caris, Cell Therapeutics, Celgene, Daiichi, Dendreon, Genomic Health, Gilead, GSK, Halozyme, Heron, Eli Lilly, Incyte, Insys, Novartis, Pfizer, etc.