CDK4/6 Inhibitors Firmly Established in HR+ Breast Cancer, OS Data Awaited

Daphne B. Stewart, MD, highlights the pivotal trials evaluating CDK4/6 inhibitors and what unanswered questions remain with this class of agents.

Daphne B. Stewart, MD

The improvement in progression-free survival (PFS) that has been observed with CDK4/6 inhibitors in patients with metastatic hormone receptor (HR)—positive, HER2-negative breast cancer has led to changes in clinical practice. However, physicians are still awaiting further overall survival (OS) findings that will solidify the use of these agents, said Daphne B. Stewart, MD.

Thus far, one set of data demonstrated a benefit in OS with this class of drugs. In an analysis of the phase III PALOMA-3 study, the combination of the CDK4/6 inhibitor palbociclib (Ibrance) and fulvestrant (Faslodex) was associated with a 10.0-month improvement in OS in patients with HR-positive, HER2-negative advanced disease with sensitivity to previous endocrine therapy.

Moreover, while all 3 FDA-approved CDK4/6 inhibitors—palbociclib, ribociclib (Kisqali), and abemaciclib (Verzenio)—have comparable outcomes in PFS, they have differing adverse event (AE) profiles. Physicians are also anticipating answers as to what the optimal sequence is, and whether there’s a benefit to the agents they’re paired with—aromatase inhibitors (AIs) or fulvestrant.

“The standard of care in the management of estrogen receptor—positive metastatic breast cancer is that combination endocrine therapy is superior to endocrine monotherapy, [which comprises] aromatase inhibitors or selective estrogen receptor degraders, in terms of improvements in PFS,” said Stewart, an associate professor of clinical medicine at City of Hope. “It approximately doubles PFS versus endocrine monotherapy. However, to date, there has been no OS benefit shown.”

In an interview with Stewart during the 2019 OncLive® State of the Science Summit™ on Breast Cancer, she highlighted the pivotal trials evaluating CDK4/6 inhibitors and what unanswered questions remain with this class of agents.

OncLive: Could you provide a brief history of the recent trials and where we stand today?

Stewart: The story began with the PALOMA-1 trial, which was a study that looked at women who previously had to demonstrate that they were sensitive to endocrine therapy. They were patients who had prior adjuvant therapy, but did not have endocrine therapy resistance. They developed metastatic estrogen receptor (ER)-positive breast cancer, and were randomized to letrozole or letrozole in combination with palbociclib. That trial showed a significant improvement in PFS; the combination therapy group had a median PFS of 20 months compared with 10 months in the control arm, and this was highly statistically significant.

That was a phase II clinical trial and that led to the phase III PALOMA-2 trial, which essentially showed the same benefit. The really interesting part, which was unexpected, was in a subgroup analysis and showed that virtually every subgroup enjoyed the same benefit. This meant that there were no reliable clinical biomarkers that could indicate who was going to benefit from the therapy—it was a benefit shown by all. Again, this looked at people who were sensitive to endocrine therapy, not those who failed prior aromatase inhibitor therapy. But, it was very encouraging and well supported.

That was then followed by the PALOMA-3 trial, which looked at patients who had progressed on prior AI therapy and underwent randomization to fulvestrant alone versus in combination with palbociclib. It was very similar with a near doubling in PFS. These were patients who failed their prior endocrine therapy, so their PFS was not as long as the responders in the PALOMA-1 and PALOMA-2 trials, but it was still very significant and benefited all subgroups.

Competing agents, ribociclib and abemaciclib, [were studied in the] MONALEESA and MONARCH trials. These trials are very consistent across the board, the hazard ratios that were shown in all of the treatment groups are virtually the same—somewhere around 0.55, so very encouraging. The one trial that is distinctive is the MONALEESA-7 trial, which looked at premenopausal and perimenopausal women with ER-positive metastatic breast cancer. They were randomized to ovarian cancer suppression with the combination of tamoxifen or an AI plus ribociclib, and again showed a similar hazard ratio and similar prolongation of survival, and similar benefit across subgroups.

While the PFS benefit is similar, what differences exist between the 3 CDK4/6 inhibitors?

The important consideration is [to determine] what kind of patient you are dealing with. Is this a patient who, for example, will be less willing to have an intramuscular injection every 30 days? Would they prefer to take a pill? That would lead you to fewer visits, the cost of care is less if you give an AI as your pair partner with a CDK4/6 inhibitor, and then you’re making choices between CDK4/6 inhibitors.

Given that they seem to have equal efficacy—fairly equal potency across the drugs—then you want to focus on the AE profiles. Palbociclib and ribociclib tend to be very similar—they predominantly result in neutropenia. Ribociclib seems to be associated with more QT prolongation than its competitors, and monitoring with EKG is required. Abemaciclib has a more distinct profile; it is a more potent inhibitor of CDK4, which leads to slightly less neutropenia, but unfortunately that is matched by a slight increase of diarrhea. Therefore, gastrointestinal toxicities, specifically diarrhea, can be high grade and is a problem with abemaciclib.

Upon failure on a CDK4/6 inhibitor, could a patient be given another one or can it be combined with a different pair partner?

That is a big question. For example, if you have been treating a patient with a CDK4/6 inhibitor plus an AI, does it make sense to continue treating them with a CDK4/6 inhibitor, and add that to fulvestrant? No one knows. The optimal sequencing has not been determined; for the time being, we would say, “No, you would move on to the next line.” I would advocate that the next line of therapy would be a combination of endocrine therapy and an mTOR inhibitor, such as everolimus (Afinitor).

At that point, do you recommend testing for PIK3CA mutations?

PI3K inhibitors are not yet commercially available in the clinic, and phase III clinical trials have been completed and show a considerable increase in PFS—but only in patients with ER-positive metastatic breast cancer with PIK3CA mutations. That is sometimes as many as 60% of patients, so it’s appropriate to do the mutational testing. Is that standard of care yet? I would assume that once alpelisib comes to market, there will be a matched diagnostic test to go along with that, but it’s clearly primed for use in the third-line setting, at least for patients with those mutations who predict that sensitivity.

Moving forward, what research would you like to see in this space?

Optimal sequencing is essential. We have so many drugs, so many options, they’re very expensive, and I’m very hesitant to discontinue endocrine therapy in favor of another. It’s clear that when you have mutations, such as the ESR1 mutation, you have estrogen independent receptor stimulations, so continuing AIs doesn’t make a lot of sense. Although, if you don’t have that, what is the right way to sequence? Would you just continue the same AI as opposed to switching to fulvestrant? There are so many different drugs that could be combined in so many different ways, it would be great to know what the optimal sequencing is.

Another area of research is women who present with metastatic disease that is quite bulky; you perceive they require a very urgent response in order to control the disease. Really, the CDK4/6 inhibitors with endocrine therapy hasn’t been studied in that setting, so can it be used upfront? Or, should we be pushing chemotherapy upfront in order to optimally debulk tumors and then follow up with a CDK4/6 inhibitor with endocrine therapy maintenance? It’s just not known.

The future is much more promising, given we know now CDK4/6 inhibitors are so well tolerated. The most important thing that needs to be published is, “Are these associated with an OS benefit?” That would make us much more confident in the kinds of investments we are making, and the toxicities we are managing—if we can prove that they are also associated with improved OS.

Turner NC, Slamon DJ, Ro, I, et al. Overall survival with palbociclib and fulvestrant in advanced breast cancer. N Engl J Med. 2018;379(20):1926-1936. doi: 10.1056/NEJMoa1810527.