Overcoming Mechanisms of Resistance to CDK4/6, PI3K Inhibition in Breast Cancer

Cynthia X. Ma, MD, PhD

Although survival outcomes have improved for patients with metastatic estrogen receptor (ER)–positive breast cancer with the addition of CDK4/6 and PI3K inhibitors to the treatment arsenal, many will experience disease recurrence, according to Cynthia X. Ma, MD, PhD, who added that overcoming resistance mechanisms to these agents is an important area of interest in the field.

Mechanisms of resistance to CDK4/6 inhibitors may be best studied in the neoadjuvant setting, according to Ma, as a biopsy can be done before and after treatment, which could reveal which pathways are still active following the receipt of a particular drug. However, more research is needed to further examine certain biomarkers of response or resistance to these agents.

The mechanisms of resistance to CDK4/6 inhibitors are pretty diverse,” said Ma. “The studies that have been done so far have mostly focused on using samples from patients with metastatic disease, specifically those who had metastatic disease and were then treated with CDK4/6 inhibitors. You can potentially correlate genomic alterations in the tumor sample with patient outcomes, [such as] progression-free survival [PFS], on a CDK4/6 inhibitor.” 

In an interview with OncLive® during the 2020 Institutional Perspectives in Cancer webinar on Breast Cancer, Ma, a professor in the Department of Medicine of the Division of Oncology at the Washington University School of Medicine, discussed several research efforts examining mechanisms of resistance to CDK4/6 and PI3K inhibitors in breast cancer and some of the novel combinations that may possess the potential to overcome this challenge.

OncLive®: What is some of the work that is being done to address resistance challenges faced in breast cancer? 

Ma: My particular interest is to develop therapeutics for patients with resistant breast cancer, especially for those with hormone receptor–positive disease. The latest advance made in managing [patients with] ER-positive breast cancer has been the introduction of CDK4/6 inhibitors and PI3K inhibitors. AKT inhibitors are also being tested in phase 3 [studies].

I also study patient-derived xenograft models to look at biomarkers that may predict response or resistance [to these agents], and [determine what] those resistance mechanisms are. Although CDK4/6 inhibitors have been shown to improve PFS, and even overall survival (OS), in some trials that were done in the metastatic setting, [findings in the] early-stage setting have not been consistent. For example, [results from] the monarchE study showed improved outcomes, but the PALLAS trial was negative. 

Although differences exist between the 2 trials in terms of their design—for example, high-risk patients were included in the monarchE trial—1 thing is clear: Some patients will not respond to CDK4/6 inhibitors or endocrine therapy. In both the PALLAS and monarchE trials, a significant number of patients relapsed really quickly.

Could you expand on what is currently known about mechanisms of resistance to CDK4/6 inhibitors?

Some studies have been done with preclinical models; these efforts used cell lines that developed acquired resistance later on. [We know that] several mechanisms potentially mediate that resistance, at least in the preclinical and metastatic settings. 

One [such mechanism] would be loss of function of RB mutations. Since RB is the target of CDK4/6, this is understandable; however, this mechanism is very uncommon. About 4% of patients in the metastatic setting may have this. In addition, some mechanisms may increase CDK6 levels, [while other] mechanisms may activate an alternative CDK, such as CDK2, by upstream growth factor receptor signaling. Also, FGFR amplification or expression could also induce resistance. However, none of these [mechanisms] are currently used [in the clinical setting] to select the right patients for these drugs. 

We cannot predict at the start whether a patient is going to be resistant [to these agents] and whether we should be treating them with chemotherapy [instead]. There is still a lot of interest to better understand what's causing the resistance mechanism. Neoadjuvant samples allow for more extensive genomic, as well as proteomic, studies so we can define the pathways better.

Shifting back to the monarchE and PALLAS trials and the discordant results yielded, do you feel that monarchE is practice changing or are more data needed to augment the positive findings that were observed?

It’s certainly exciting to now have another agent that could potentially reduce early recurrence for our patients with ER-positive breast cancer. However, longer follow-up will be needed because we do not know what the optimal duration of these drugs is and how patients will do later on when they come off the CDK4/6 inhibitors. I certainly am very excited about the data, but we do need longer follow up to find the right patients [for this approach].

Could you expand on some of the efforts you are involved in that are examining resistance mechanisms in the neoadjuvant setting specifically?

We have a neoadjuvant study where we [examined] the CDK4/6 inhibitor palbociclib [Ibrance], in combination with endocrine therapy in patients with endocrine [therapy]–resistant disease. [Through this research], we can study potential resistance mechanisms and compare the [use of palbociclib] in sensitive versus resistant cases. We would like to better understand the biology of resistant tumors.

Is it easier to examine those mechanisms of resistance in the neoadjuvant setting?

In the neoadjuvant setting, because there's a tumor in the breast, you can actually [perform] a biopsy before you start treatment and then later on in the treatment journey. You can study which pathways are still active when you give a particular drug and you can potentially do more omic studies with the frozen samples and fresh biopsies available from these trials. 

The challenge in patients with metastatic cancer when doing these biomarker studies is that sometimes the tumors are not easily accessible. You have to do CT- or ultrasound-guided biopsies, which for internal organ lesions could sometimes be difficult and may not be safe. 

The neoadjuvant [setting] really offers plenty of [room for] tumor assessment. One disadvantage with neoadjuvant studies is that we use biomarkers, or a clinical response, as a surrogate for relapsed outcomes in the long term. We need more follow-up from [patients who received] neoadjuvant [treatment so we can] correlate their biomarker response or imaging response in the short term, with long-term relapse outcomes. 

Are there any challenges to enrolling patients to these neoadjuvant studies?

Outcomes are very good [for those] in the earlier-stage setting; we cure the majority of patients with breast cancer. However, about 30% of patients with this disease experience recurrence over the years to come. Patients know that. 

These [neoadjuvant] trials are not done for [those with] stage I [disease]; they are done for those with locally advanced disease. These patients could have had positive lymph nodes or a bigger tumor to begin with. As such, we have pretty good success in enrolling patients to these studies. Patients are fairly motivated to do all that they can to help. They understand that sometimes these biopsies may not provide any benefit for them specifically, but they contribute to the research and that's helpful.

Are there any other ongoing neoadjuvant trials being done that you would like to spotlight?

What we're trying to do to overcome CDK4/6 inhibitor resistance or to delay this resistance is to add on agents that would inhibit, for example, the PI3K pathway. We have a National Cancer Institute study that is looking at a triplet regimen comprised of the PI3K inhibitor copanlisib [Aliqopa] plus fulvestrant [Faslodex] and abemaciclib [Verzenio]. The goal is that by adding the PI3K inhibitor, we will be able to delay progression on the combination. 

The interesting concept here is that copanlisib is intravenously administered and is given once a week on day 1, day 8, and day 15 every 24 days. This drug differs from other PI3K inhibitors that are oral medications that need to be taken every day. The intravenous administration allows [us to deliver] a higher dosage and thus, more complete inhibition of the pathway. The hope is that the triplet will behave like cytotoxic drugs at that time point so that we can prevent the development of resistance. 

These triplets [have been found to be] very potent in the preclinical studies and in patient-derived xenograft models. These [regimens are] still [being examined] in the metastatic setting. Whether future studies will bring some of these ideas to the early-stage setting would be interesting to see, although we would have to worry about toxicity.

As you mentioned, these agents are not without toxicity. Were any management strategies in place to combat the adverse effects associated with the triplet regimen in that study?

We did not specify any prophylactic measures, but because copanlisib can cause hyperglycemia, the protocol called for fasting or having a low hyperglycemic diet. Patients could not receive the infusion right after a meal. We also monitor glucose after the infusion for 2 hours. In addition, if the patient experiences diarrhea—both drugs could have overlapping toxicity in that sense—we do advise that physicians prescribe prophylactic antidiarrheal agents. We try to give a very good education to the physicians and coordinators who are participating on this trial so that we can manage toxicities better.