Longer Follow-Up Is Imperative With Adjuvant CDK4/6 Inhibitors in ER+ Breast Cancer

Optimal patient selection and duration of therapy are the 2 biggest areas to tackle regarding CDK4/6 inhibition in the adjuvant setting of estrogen receptor (ER)–positive breast cancer, explained Ruth O’Regan, MBBCh, BAO.

Longer follow-up from the PENELOPE-B (NCT01864746), PALLAS (NCT02513394), and monarchE (NCT03155997) trials, and data from the ongoing NATALEE (NCT03701334) study, will likely assist the breast cancer community in answering these questions. Hopefully, she added, use of the agents in the adjuvant setting will help prevent the risk of late recurrence, as well.

“There are a lot of questions for sure, and obviously, this is a significant issue because there are a lot of recurrences in these ER-positive breast cancers, both early on, and later on, as well,” said O’Regan, chair of medicine and Charles A. Dewey Professor at the University of Rochester, physician-in-chief of Strong Memorial Hospital. “It’s the most common type of breast cancer, so having new agents that would help patients is of crucial importance.”

In an interview with OncLive® during an Institutional Perspectives in Cancer webinar on breast cancer, O’Regan, who is also associate director of Education and Mentoring at the Wilmot Cancer Institute at University of Rochester, discussed the preliminary data with the 4 studies of CDK4/6 inhibitors in the adjuvant setting and how to tackle the remaining questions with longer follow-up.

OncLive®: Trials with adjuvant CDK4/6 inhibitors have shown discordant results. In your opinion, how should “high-risk” be defined in breast cancer?

O’Regan: The question is: is it the anatomic stage that makes a patient high risk, or is it the biology of their cancer? There are data from studies suggesting that the biology of the cancer may be important. In the monarchE study, if you look at the patients whose tumors had Ki-67 expression at least 20% or higher, they had a greater benefit from abemaciclib [Verzenio] than the overall intent-to-treat population. Along with that, when you think about PENELOPE-B [NCT01864746],we know that luminal A cancers are less likely to achieve a pathologic complete response [to CDK4/6 inhibitors], so it's possible that those recruited to PENELOPE-B might have been enriched for luminal cancers and obviously, at 4 years, there was no benefit for palbociclib [Ibrance] in that group of patients.

In the metastatic setting, [researchers] made an attempt to try and look at whether certain cancers are hormone sensitive or hormone resistant, and whether that impacts the benefit of CDK4/6 inhibitors. It doesn't appear to, but the problem with the metastatic studies is that, in a lot of cases, they use primary tissue. [As such,] you don't really know in the metastatic setting what you're really dealing with, as far as the type of ER-positive breast cancer that you have. The bottom line is, based on monarchE, we would have to use anatomic stage to guide us; larger, node-positive tumors would be the most likely ones that we would treat.

Again, all that really means is that they probably have a higher recurrence rate, especially in the first 2 to 5 years. The biology may be more important, and one of the things that we've been really trying to look for is a robust biomarker to tell us which patients really do benefit from CDK4/6 inhibitors.

If a patient’s tumor has high Ki-67, they do get a greater benefit from abemaciclib. However, in cancers with low Ki-67, abemaciclib is also effective, so it's hard to know. Hopefully, we might get some intrinsic subtyping on these tumors to help us know whether it is luminal A or luminal B.

In the metastatic setting, what are the differences in efficacy between the CDK4/6 inhibitors in these patient populations?

What we have is the PENELOPE-B study, which was positive at year 2, positive at year 3, and then negative at year 4. monarchE is showing a positive signal at 2 years, but we don’t know if those curves will come together. Longer follow-up from monarchE is really key to make sure those curves stay apart. I certainly hope they do, because we definitely need treatment for patients with these high-risk ER positive breast cancers. At this point, we just don't know.

Is abemaciclib a better or more effective CDK4/6 inhibitor? Again, that's not supported for the metastatic setting because the hazard ratios of benefits of CDK4/6 inhibitors for all 3 of them in the first-line setting are remarkably similar. I don't think we have enough data to support that [one is better than another].

The other thing to keep in mind is adherence. Patients probably are more likely to be adherent in the metastatic setting, just based on their disease status.

Also, [in the adjuvant setting] particularly in PALLAS, there were very rigid guidelines in terms of when to hold a drug, which, in the real world, we're not doing that as much in the metastatic setting. There are a lot of components here, but we need longer follow-up from monarchE, the results from NATALEE, and to continue the PALLAS and PENELOPE-B studies, as well.

Were the stricter criteria in PALLAS why we saw greater adherence issues versus PENELOPE-B?

That is certainly a possibility. PENELOPE-B had pretty good adherence, and it was fairly comparable to monarchE, so the outlier here was PALLAS. You have to imagine that it was just the criteria of the trial in terms of managing neutropenia, etc. Honestly, what we know in the metastatic setting is that patients get neutropenic, but there really isn't a lot of consequences from that. They don't typically get febrile neutropenia. We do a lot of blood tests and make adjustments, but in truth, we don't know how important that is. We also know that if you do have to do dose reductions, these drugs are still effective.

Could the NATALEE trial also provide insight into the optimal duration of CDK4/6 inhibition in the adjuvant setting?

Absolutely. One of the things with PENELOPE-B is that if they had given the CDK4/6 inhibitor a for longer period, maybe the results would have held up longer. The nice thing with NATALEE is that it uses 3 years of ribociclib. Also, does it make sense to give the CDK4/6 inhibitor when you start endocrine therapy, or soon after? Or, you could imagine if you’re trying to prevent late recurrences—which we have no idea whether these drugs do or not—maybe you would want to start the CDK4/6 inhibitor later on. These are all questions we just don't know the answer to right now.

If so many of these recurrences occur after 5 years, should CDK4/6 inhibitors be given indefinitely, or just started later?

The likelihood that we give them indefinitely is probably not that high because they have toxicities, and they're very expensive. However, the exact timing from which we should start them [is still unclear]. That’s why following these trials is going to be very important. If we have longer follow-up, we can see if there's a signal in the late recurrence of whether the CDK4/6 inhibitors make a difference. If that is the case, maybe they need to get started much later.