Pfizer’s palbociclib data in ER+ breast cancer – hit or miss?
At AACR this weekend, Dr Richard Finn (UCLA) presented the much anticipated front-line phase II data for Pfizer’s CDK4/6 inhibitor, palbociclib (palbo) plus letrozole versus letrozole alone in ER+ HER2- breast cancer.
The PALOMA series of trials are designed to show that adding a specific CDK inhibitor to an aromatase inhibitor enhances efficacy and improves outcomes.
There are three metastatic breast cancer trials in all, with PALOMA–1 being the phase II study while PALOMA–2 and –3 are phase III randomised controlled registration studies aimed at confirming the initial phase II results in combination with letrozole and fulvestrant, respectively. In addition, palbociclib is also being evaluated in combination with standard endocrine therapy (PENELOPE-B) for certain early-stage breast cancers.
In short, an analysis demonstrated that the primary endpoint of progression free survival (PFS) was met, but the overall survival (OS) data was not significant at the time of the analysis.
What does this does this data mean and in what context should we look at the results?
Subscribers can log-in to read our latest insights or you can purchase access to BSB Premium Content.
This content is restricted to subscribers
9 Responses to “Pfizer’s palbociclib data in ER+ breast cancer – hit or miss?”
I believe Pfizer indicated that all drugs for ER+ metastatic breast cancer have been approved on the basis of PFS. See Femara label as an example.
Hi Patrick, yes Mace Rothenberg the Pfizer Head of Oncology R&D did say that at the end of the press briefing when I chatted with him. It’s broadly true, except that the FDA used the lack of confirmatory OS argument against Genentech with Avastin in breast cancer when two subsequent studies showed a marked reduction in the OS benefit compared with the original registration trial (4–>2 months change). The original approval was based on good results for both if my memory is correct.
The problem with this data is that while the PFS curve was stunning, the OS is immature and may make FDA hesitate given the issues in this indication.
I should add that their approval was a full approval and not an accelerated one, so ‘confirmatory’ trials weren’t really part of schema per se, but FDA can wield the sword either way according to the whim of the moment, as you well know.
I double checked this with the good folks at Gene as that was my memory too, but it turns out it’s not the case. They kindly confirmed that Avastin was approved for met breast cancer under accelerated approval based on PFS in the E2100 study. That means that OS in the confirmatory trials would be needed for full approval. It didn’t happen… and the rest is history.
They have done some basic biomarker stuff (IHC) but so far it is not panning out. We have some clues as to why, and a negative predictive marker we think, based off of preclinical stuff, but they have been keeping the samples close so we can’t look directly. 🙁
That’s interesting to note, thanks Angela. I’m kicking myself for not remembering to ask about it in the press briefing. A negative predictive marker is unusual!
But potentially very useful like Kras mutation in EGFR+ LC/CRC. 🙂
Ooooh now I’m really interested! That would be very interesting indeed. Any die what % of patients that might affect in breast cancer?
Issue is a bit complex – both expression and localization would matter. But at minimum about 40% have the marker. We would have to run the analysis to see whether defining 1 or 2 of the phenotypes as “negative predictive” gives the best separation. If 2 then, about 60% would be predicted less responsive (and more aggressive disease).
Comments are closed.