Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster display

2960 - Relationship between progression-free survival and overall survival in advanced breast cancer: a novel approach using first-line treatment data for fulvestrant 500 mg and anastrozole


10 Oct 2016


Poster display


Mario Ouwens


Annals of Oncology (2016) 27 (6): 68-99. 10.1093/annonc/mdw365


M. Ouwens1, L.M. Grinsted2, C. Telford3

Author affiliations

  • 1 Global Medicines Development, AstraZeneca R&D, 431 83 - Molndal/SE
  • 2 Global Medicines Development, AstraZeneca, Cambridge/GB
  • 3 Global Payer Evidence Pricing, AstraZeneca, Gaithersburg/US


Abstract 2960


Overall survival (OS) is considered the gold standard for clinical benefit in oncology trials, but mature data are often unavailable. A relationship between progression-free survival (PFS) and OS in advanced cancer, including breast cancer (BC), is known: FDA (2007) suggests PFS may be a surrogate for OS; a NICE DSU report (Davis et al 2012) presents evidence of a relationship between PFS and OS in BC. This analysis further examined the relationship between PFS and OS with a view of using PFS as a predictor of OS using data from the Phase II FIRST study (n = 205; NCT00274469) of fulvestrant 500 mg vs. anastrozole as first-line treatment in hormone-receptor positive advanced BC.


In the interests of homogeneity, the relationship between PFS and OS was evaluated in endocrine-naïve patients in the FIRST study (n = 73 [72%] fulvestrant; n = 80 [78%] anastrozole) by substituting the linear expression into each other and using Weibull parametric fits and linear regression. PFS and OS data from a study of anastrozole in a similar population (Nabholtz et al 2000, 2003) were applied to validate the Weibull and linear regression model.


Using log cumulative hazard plots, a linear trend was shown for PFS and OS. From the Weibull model, relationships between OS and PFS were derived: for given S, number surviving, fulvestrant ln time_OS(S) = 0.82 + 0.80 ln time_PFS(S); and anastrozole ln time_OS(S) = 0.97 + 0.79 ln time_PFS(S). Based on linear regression of part of the PFS and OS curve (driven by apparent deviation from a linear relationship), the following relationships were derived: fulvestrant ln time_OS(S) = 0.77 + 0.63 ln time_PFS(S); and anastrozole ln time_OS(S) = 0.95 + 0.67 ln time_PFS(S). In all equations, time = days/1000. Applying both models to other clinical data for anastrozole showed a good fit and thus extends this relationship beyond the FIRST study.


This analysis shows a novel, validated approach by which the relationship between PFS and OS in patients receiving first-line treatment with fulvestrant or anastrozole can be modelled. These results add to the acceptance of PFS as a predictor of OS in this setting.

Clinical trial identification


Legal entity responsible for the study





M. Ouwens, L.M. Grinsted, C. Telford: Employment, stock or other ownership - AstraZeneca

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings