A Cancer Drug Class Hits Its Stride

The buzz last month at the annual meeting of the American Association for Cancer Research (AACR) was all about palbociclib, a Pfizer drug candidate that blocks two proteins called cyclin-dependent kinases. In women with estrogen-receptor-positive breast cancer, a group that represents 70% of the breast cancer patient population, adding the drug to conventional hormonal therapy doubled the amount of time the disease was halted.

“These results are strikingly positive and with a large potential impact to patients with ER-positive breast cancer,” José Baselga, physician-in-chief at Memorial Sloan Kettering Cancer Center, told the audience at an AACR plenary session highlighting the results. He was not involved with the trial, although he does consult for Eli Lilly & Co. and Novartis.

The results put the spotlight on palbociclib, but two similar compounds—Lilly’s bemaciclib and Novartis’s LEE 011—are also gaining traction in the clinic. Although Pfizer is ahead in the three-way race, differences in how the molecules might be used are starting to emerge, suggesting there’s room for all of them. As the drugs move toward commercialization, oncologists are trying to figure out who could benefit from them and what the best treatment combinations might be.

The encouraging data are the culmination of decades of work to understand how small molecules can be used to interrupt the cell cycle—the cascade of events that leads to DNA replication and cell division. Cyclin-dependent kinases, or CDKs, regulate the growth phase of the cell cycle, which is kicked off when growth factors spur the production of cyclin D, a protein that binds to CDK 4. The resulting complex triggers the first in a series of steps leading to DNA replication.

Most tumors have lost control of some player in the growth phase of the cell cycle—a mutation in CDK 4, overexpression of cyclin D, or loss of a natural inhibitor of CDK 4 called p16. Regardless of the culprit, the end result is overactivation of CDK 4 and a related protein, CDK 6. That aberrant activity gives cancer cells the upper hand over normal cell growth.

Scientists long ago realized it might be possible to stop tumor growth by targeting individual components of the cell cycle, but early drug discovery efforts failed. The first molecules hit the wrong members of the CDK family. For most, CDK 2 was the primary target; it was later found that CDK 4 compensated for the loss of CDK 2.

With palbociclib, Pfizer medicinal chemists were able to selectively block both CDK 4 and CDK 6, an approach that is finally bearing fruit in the clinic. In a Phase II study of postmenopausal women with ER-positive breast cancer, adding palbociclib to the hormonal therapy letrozole kept the disease from worsening for 20 months from just 10 months in patients who were given letrozole alone.

Although Baselga lauded the results as promising, he was also quick to point out the realities of cancer drug development. Many compounds that look good in Phase II go on to fail in larger Phase III trials.

Moreover, the palbociclib results came with caveats. One was the study design. The trial was open label, meaning the investigators knew which patients got palbociclib. Furthermore, the investigators themselves judged whether disease progressed. Both can bias the results

Most important, the data are not yet complete enough to prove that the drug helps patients actually live longer. Although overall survival improved by four-and-a-half months in the group receiving palbociclib, the trial didn’t have enough patients for the results to be statistically significant.

Pfizer is conducting a Phase III study of palbociclib as a breast cancer treatment, but the firm might be able to seek early Food & Drug Administration approval based on the Phase II results. Even without an early green light, palbociclib is closer to market than the drugs from Lilly and Novartis. With that leg up, stock analysts believe palbociclib sales could hit $2 billion annually by 2019.

Although they trail Pfizer in development, Lilly and Novartis are finding ways to distinguish their molecules. At this early stage, it’s impossible to know which of the three compounds is most effective, but differences are emerging in their activities and the molecules with which they might be able to be combined.

Lilly has been working on CDK inhibitors since the late 1990s, with the goal of finding a selective molecule that could stop the cell cycle at the growth phase, cross the blood-brain barrier, and be continuously dosed, says Richard B. Gaynor, vice president of clinical development and medical affairs for Lilly’s oncology business.

That elbow grease appears to have paid off. Of the three CDK 4/6 inhibitors, bemaciclib is so far the only one to have shown efficacy on its own. At AACR, Lilly presented data from a Phase I study of bemaciclib’s activity against a variety of tumors. Among the 36 women with hormone-receptor-positive breast cancer given the pill, 25% saw tumors shrink.

The result is impressive, particularly since these women had already been treated unsuccessfully with many other drugs, says Sara M. Tolaney, a clinical investigator in the Breast Oncology Center at the Dana-Farber Cancer Institute who treated several of the patients in the trial. Tolaney does not receive any industry funding.

In addition to being active on its own, she adds, bemaciclib is the only compound now in the clinic that penetrates the central nervous system. That feature could give it an edge in treating patients whose disease has metastasized to the brain.

Seamus Fernandez, a stock analyst at the investment firm Leerink Swann, notes that bemaciclib appears to cause milder side effects than palbociclib. He expects bemaciclib to bring in peak sales of $650 million per year. A Phase III trial is expected to start this year.

Novartis, meanwhile, is trying to close the development gap between palbociclib and LEE 011, its CDK 4/6 inhibitor. The company surprised the oncology and investment communities in November when it said it was jumping to Phase III tests of LEE 011 in breast cancer the following month; the drug candidate had previously been in Phase I studies.

Like Pfizer, Novartis set up a study comparing the benefit of adding its CDK 4/6 inhibitor to letrozole against ER-positive breast cancer. But Novartis is also staging Phase I drug combination studies that capitalize on its broad oncology pipeline.

This month, at the annual meeting of the American Society of Clinical Oncology (ASCO), data will be unveiled from a Phase I breast cancer study that combines LEE 011, letrozole, and BLY719, a Novartis compound that blocks PI3K-α, a protein in a signaling pathway critical to tumor survival.

Putting up a roadblock on this pathway on its own doesn’t seem to be enough to stop breast cancer progression, explains Pamela N. Munster, an oncologist who leads the developmental therapeutics unit at the University of California, San Francisco’s Helen Diller Family Comprehensive Cancer Center. But combining a CDK 4/6 inhibitor with BLY719 might cut off the alternate growth avenue, adds Munster, who will present the data at the ASCO meeting.

Oncologists see good reasons to explore other partners for the CDK 4/6 inhibitor class. Genomics studies suggest CDK 4/6 inhibitors could be combined with histone deacetylase inhibitors—which are also known to stop the cell cycle in the growth phase—for a synergistic effect. Munster also suggests that combinations of CDK 4/6 inhibitors with PD-1 immunotherapies, arguably the most exciting emerging drug class in oncology, could be worth testing.

Investigators are also trying to suss out the right biomarkers to guide the use of CDK 4/6 inhibitors. Pfizer’s Phase II study of palbociclib included an arm that explored whether elevated levels of cyclin D and loss of p16 could be used to predict the drug’s efficacy. But neither correlated with improved outcomes.

Such stumbles aside, oncologists say the arrival of the CDK inhibitor class is cause for excitement. “Today is a good day,” Baselga said at AACR. The early Lilly data and palbociclib results, if confirmed in Phase III studies, he said, “could represent a new standard of care” for patients with breast cancer.  

PIPELINES

Lost In The Merger Mix

As news emerged last month that Pfizer is pursuing a merger with AstraZeneca, many industry watchers voiced concern over the impact the union could have on the companies’ R&D operations. If realized, the acquisition is certain to bring job cuts to an already beleaguered drug research community. Moreover, observers argue that R&D productivity will suffer as the two sides try to integrate their vast activities.

If the winding commercialization path for palbociclib, a Pfizer cancer drug in late-stage studies, is any indication, there is reason to worry.

Palbociclib is considered one of Pfizer’s most important drug candidates; stock analysts project its sales could reach $2 billion annually by 2019. But the molecule languished for years in development purgatory. Scientists who worked on the project in its early days say a string of megamergers was in part to blame.

Palbociclib originated at Warner-Lambert’s Parke-Davis unit, where in the late 1990s, a team of scientists in Ann Arbor, Mich., began screening for CDK 4/6 inhibitors in collaboration with colleagues at Onyx Pharmaceuticals.

That screening effort yielded palbociclib in 2002. But Pfizer had acquired Warner-Lambert the year before, and the compound got lost in the R&D shuffle that so often accompanies a megamerger, says an ex-Pfizer scientist who helped discover it. The scientist, now employed elsewhere, asked not to be identified by name. The problem only worsened in 2003 when Pfizer continued its buying spree with the purchase of Pharmacia.

According to Clinicaltri​als.gov, the drug entered its first Phase I study in 2005. The results were reported two years later at the American Society of Clinical Oncology meeting, not long after Pfizer announced it would shut down the Ann Arbor site.

Work on palbociclib was transferred to Pfizer’s labs in La Jolla, Calif., where the compound was placed in a developmental queue behind other products in Pfizer’s pipeline—namely the kidney cancer treatment Sutent and the lung cancer drug Xalkori. Phase II work did not begin in earnest until 2010.

More than a decade after its discovery, palbociclib is finally heading toward commercialization. Industry watchers are now wondering which promising drug candidate might be the next to get lost in the mix.