UPDATED: Pfizer's flagship palbociclib stymies cancer progression but falls short on survival

Pfizer's closely watched cancer drug palbociclib doubled the average amount of progression-free survival (PFS) time among patients with advanced breast cancer. But in the first detailed glimpse of its impact on overall survival (OS)--a key feature to the future prospects of this flagship program--the therapy has failed to demonstrate a statistically significant improvement in extending patients' lives after an initial assessment.

Patients with hormone receptor-positive metastatic breast cancer enjoyed a median PFS rate of 20.2 months when taking a combination of palbociclib and the antiestrogen drug Femara (letrozole), compared to 10.2 months for letrozole alone, according to researchers presenting the latest data at the American Association for Cancer Research meeting in San Diego. The OS rate, though, was 37.5 months for the combo versus 33.3 months for the solo therapy. That was an improvement, but not a big enough one to qualify as significant enough to meet the secondary endpoint.

Pfizer's ($PFE) investigators were quick to note that there haven't been enough patient deaths in the study to firm up the OS numbers, happy at this stage to point to a trend in the drug's favor. But investors were left feeling somewhat deflated over the shortfall on OS, especially after seeing signs of a dramatic improvement in PFS earlier on. At the interim point, patients taking palbociclib saw a delay in disease progression of 26.1 months, compared to only 7.5 months in the control arm.

In the cooling-off period after the data were presented, Pfizer's shares slid about 3%, or close to $1, in early trading Monday. 

As Tim Anderson at Bernstein noted on Monday, the data are good. Analysts had expected the PFS rate over the letrozole arm to balance out a bit in the follow-up. And he expects that the results are sufficient for a regulatory application soon, which is what investors have been gambling on. Pfizer's execs, though, haven't come right out and said whether they're going for an early approval, so look for more ups and downs ahead as this story plays out.

Pfizer, meanwhile, has been pushing palbociclib into a much larger late-stage program. The cancer drug--granted the FDA's "breakthrough" drug status--offers Pfizer a rare shot at a blockbuster approval. Analysts have estimated peak potential sales ranging from $1 billion to more than $6 billion (ISI's Mark Schoenebaum). Anderson pegged 2020 sales at $2 billion, so you can see plenty of daylight between these projections.

At the end of last week, Schoenebaum noted that there's strong market sentiment favoring a shot at an early approval. But a lot depended on OS rates. "We think the stock could move up ~$1 if palbo shows a statistically significant improvement in OS along with a visually impressive Kaplan Meier curve. If PFE announces they plan to file an accelerated approval during their analyst meeting, PFE could move up ~$2. We see downside of ~$1 if there is no OS difference between the palbo and control arms due to the lack of mature data. We are currently modeling ~$6.3 B of probability-adjusted revenues in 2022."

Palbociclib is a CDK-4/6 inhibitor. And it's not alone. Last December, Novartis ($NVS) quickly pushed its rival program for LEE011 into Phase III, setting up a showdown over a market that could deliver billions in potential revenue. Both therapies target a pair of cyclin dependent kinases that play a role in cancer. Eli Lilly ($LLY) also has a contender in the pipeline, LY2835219. But the pharma giant is well behind the leaders in this race, and Lilly suffers from a reputation for carefully regimented development programs that often lag well behind those of rivals. 

Lilly's investigators reported at AACR that of the 47 patients with metastatic breast cancer in the study, 9, or 19%, had a partial response while 24, 51%, had stable disease. "Disease progressed despite treatment in 11 patients," according to a release from Lilly. "All of the nine patients who had a partial response, and 20 of the 24 patients who had stable disease, had HR-positive disease, which meant that the partial response and stable disease rates for patients with HR-positive disease were 25 percent and 55 percent, respectively."

Schoenebaum takes a positive view on Lilly's come-from-behind position in the race for an approval.

"Overall, these data seem to indicate that LLY's drug is clearly active," he wrote Sunday morning. "Whether it's actually MORE active than PFE's palbo remains an open question as you must be very careful when comparing across different trials with small patient numbers. As a reminder, LLY has argued that their drug might work better b/c it can be dosed continuously (vs palbo's 3 weeks on; 1 week off regimen) due to less severe neutropenia. LLY plans on announcing next steps later in the 1H:14--we believe it's possible that a phase 3 could begin by YE:14 or in early 2015. There is VERY LITTLE, IF ANYTHING, in LLY consensus estimates for this drug--so in theory, at least, it represents all upside."

The view from Pfizer's investigators remained upbeat, despite the setback on OS rates.

"This is an exciting data set that shows a major clinical benefit for patients who have hormone receptor-positive, metastatic breast cancer," said UCLA Professor Dennis Slamon. "The potential impact of this study could be huge. We are doing further Phase III work with the drug, but the current data are as exciting as the initial studies we were involved in when testing Herceptin for HER2-positive breast cancers."

- here's the release on the Pfizer data
- see the release on Lilly's program

Special Report: Top 10 experimental cancer drugs 2013 - Palbociclib