SABCS 2014: Paclitaxel and Pathological Complete Response

Fiona WilmotBy Fiona Wilmot, MD, MPH, BCAction Program Consultant

This session addressed the use of paclitaxel in two forms – the regular formulation (brand name Taxol) and protein bound (brand name Abraxane). This study illustrates how the use of surrogate markers such as pathological complete response (pCR) may reduce the actual clinical applicability of research findings. Pathological complete response is typically defined as no invasive and no in situ cancer residuals in the breast and nodes. In clinical trials, a surrogate marker (or endpoint) such as pCR is a measure of the effect of a specific treatment that may correlate with overall survival but does not necessarily have a guaranteed relationship.

Michael Untch, MD, PhD presented data from The German Breast Group that compared the use of two preparations of paclitaxil – one, the usual preparation and one protein-bound in 1,200 patients (400 with Her 2+ tumors). The results showed an approximate 10% favorable difference in the protein-bound paclitaxel when measured by pCR, though patients treated with the protein-bound paclitaxel suffered a higher rate of sensory neuropathy. Of interest: protein-bound paclitaxel costs five times as much as non-bound paclitaxel.

It’s tempting to interpret these results as evidence supporting the use of protein bound paclitaxel:  pCR is a surrogate marker rather than the true endpoint of overall survival. As a result, we’re unable to determine if this result is clinically significant. And although in June 2012, the Food and Drug Administration endorsed using pCR as an endpoint in some instances, we have to remember that a change in a surrogate marker does not predict a change in survival. We need more meaningful endpoints before considering a change in clinical practice.

This example is a good illustration of both the promise and potential pitfalls of how we interpret research results. While we’re always on the lookout for good news, this study reminds us to be methodical in how we review evidence – and to resist the seduction of what “looks like” promising data. In this case, we’d prefer to wait for evidence that shows a survival benefit before considering changing from use of paclitaxel to the protein formulation of the drug.

Session: A randomized phase III trial comparing neoadjuvant chemotherapy with weekly nanoparticle-based paclitaxel with solvent-based paclitaxel followed by anthracyline/cyclophosphamide for patients with early breast cancer (GeparSepto) Untch M, Jackisch C, Schneeweiß A, et al.; GBG 69.

This entry was posted in BCA News.