As CDK4/6 inhibitors solidify their role as the treatment of choice for patients with hormone receptor–positive, HER2-negative breast cancer, an ongoing clinical challenge remains: distinguishing the appropriate agent.
AS CDK4/6 INHIBITORS SOLIDIFY their role as the treatment of choice for patients with hormone receptor–positive, HER2-negative breast cancer, an ongoing clinical challenge remains: distinguishing the appropriate agent. In the absence of head-to-head trials, investigators are looking to predictive and prognostic factors in patient population data as well as safety profiles across ongoing studies to guide treatment selection. Three agents are approved by the FDA—palbociclib (Ibrance), ribociclib (Kisqali), and abemaciclib (Verzenio)—for use in first- and subsequent-line settings for pre-, peri-, and postmenopausal patients based on improvements in progression-free survival (PFS). Emerging data suggest that they also improve overall survival (OS).1,2
“CDK4/6 inhibitors have completely changed our view on how we manage patients with hormone receptor–positive disease,” panelist Massimo Cristofanilli, MD, said. “For the first time, we are starting to see an overall survival benefit. We never thought that this was going to be possible.”
During a recent OncLive Peer Exchange®, a panel of breast cancer experts examined molecular testing approaches and prognostic factors for patients with hormone receptor–positive metastatic breast cancer (mBC). Additionally, the panel put into context the clinical trial data for CDK4/6 inhibitors in this patient population, including how safety considerations may best determine the course of treatment, as well as whether the coronavirus disease 2019 (COVID-19) pandemic is impacting their prescribing practices.
Since molecular testing is taking on increasing importance across many solid tumors, including breast cancer, moderator Sara A. Hurvitz, MD, asked the participants about their approach to testing in patients with advanced hormone receptor–positive breast cancer, especially as targeted treatments and the genomic data regarding them rapidly evolve.
“Besides considering patients for clinical trials in the first-line setting of novel agents like PI3K inhibitors, all I really need to know to treat patients are ER [estrogen receptor], PR [progesterone receptor], and HER2 status,” Andrew D. Seidman, MD, said. Although germline testing for BRCA mutations may be considered because it can open the door to subsequent use of PARP inhibitors, Seidman said it is not a priority in the first-line setting.
Cristofanilli described a similar approach. “In de novo metastatic breast cancer, there is no mechanism of resistance that we’re going to target upfront,” he said. In contrast, once there is recurrence, it is important to determine the drivers of resistance, he explained. “When we biopsy in the metastatic setting, if there is tissue available from additional sites, we send those for genomic testing and NGS [next-generation sequencing] to find out if there are, for example, ESR1 mutations or [other] actionable mutations,” he said. “If there is bone disease, we can repeat NGS, so we probably look at liquid biopsies and wait to understand the molecular drivers for these patients,” he explained. He remarked that this approach helps identify which endocrine therapy backbone is most likely to benefit each individual patient. “We may even find F2 mutations that are actionable at this point. There are data suggesting these patients respond to tyrosine kinase inhibitors [TKIs] for HER2 [targeting]. But it’s very important to rule out immediately if an ESR1 mutation is the driver of the resistance,” he said.
Hurvitz added that there is no consensus on when to conduct molecular testing in patients with de novo hormone receptor–positive mBC. “Some people wait until the second- or third-line setting,” she said. “Certainly, for clinical trials, many of us are interested in querying what the tumor biology is for clinical trial eligibility.”
The discussion turned to prognostic factors in hormone receptor–positive mBC. Sara Tolaney, MD, MPH, said that when assessing disease prognosis for patients who have either recurrent or de novo disease in the first-line setting, physicians should consider the disease-free interval. “We know that patients who are recurrent on their endocrine therapy—particularly early on in their endocrine treatment, usually within the first 2 years of adjuvant therapy—are patients with endocrine-refractory disease. These are patients who we know are not going to be as sensitive to future endocrine therapy and generally have a poor prognosis. Someone who recurs 10 or 15 years later or who has de novo disease—those are patients with very different biology in their tumor and [they] generally have better prognostic features,” she explained.
Another important prognostic consideration Tolaney noted is the site of disease. She explained that significant visceral involvement, especially liver involvement, is associated with a worse prognosis compared with bone metastases. Other poor prognostic factors include presence of higher-grade tumors and PR negativity. Some molecular tests may also help give an indication of prognosis, such as those evaluating circulating tumor cells (CTCs), she said, and noted that Cristofanilli has done considerable work in this area. For example, in an international expert consensus paper coauthored by Cristofanilli in 2019, CTC enumeration was demonstrated to be prognostic independent of clinical and molecular variables.3 Data from the study showed that patients with at least 5 CTCs/7.5 mL had a more aggressive disease course, whereas those with fewer than 5 CTCs/7.5 mL were found to have a more indolent disease course. In patients with hormone receptor–positive disease, this translated to a 26-month or higher OS benefit in the indolent group (44.0 vs 17.3 months; P < .0001).3
CDKs are multifunctional enzymes that play an important role in regulating the cell cycle. CDK4/6 inhibitors block cell-cycle progression during the G1 phase of the cell cycle, thereby preventing cancer progression.4 In patients with hormone receptor–positive, HER2negative breast cancer, CDK4/6 inhibitors are currently a National Comprehensive Cancer Network (NCCN) preferred first-line therapy in combination with aromatase inhibitors (AIs). They are also a preferred second- and subsequent-line treatment in combination with fulvestrant (Faslodex) in patients who have not previously received a CDK4/6 inhibitor.5 Although it is possible that the 3 FDA-approved agents (ie, palbociclib, ribociclib, abemaciclib) have similar efficacy, the panelists mentioned some key distinctions in the study populations in which these agents were studied, suggesting there may be some differences in benefit between agents.
“There are some studies where the patients were endocrine resistant. Some others included patients with de novo disease who had not been exposed to prior endocrine therapy,” Cristofanilli said, noting that sites of metastasis also varied between studies. “It seems that some of the agents would be more effective in those with endocrine resistance or visceral metastases, whereas others are particularly effective in patients with endocrine sensitivity and bone disease,” he said. Cristofanilli added that these distinctions between studies make it difficult to determine if efficacy is a function of the agents themselves or if it is characteristic of the patients enrolled in these studies. Nevertheless, he emphasized the importance of considering such distinctions when deciding between CDK4/6 inhibitors in clinical practice.
Another important consideration is the varied toxicity profiles of these agents (TABLE6-8). According to the participants, many toxicities that initially seemed difficult to manage have become less intimidating, such as neutropenia, which is a common adverse effect (AE) investigators observed across all 3 CDK4/6 inhibitors. “We feel pretty comfortable knowing how to manage that now by checking CBC [complete blood count] twice weekly in the first 2 months, perhaps at the first day of a couple of cycles, then based on discretion afterward. The same applies for potential liver function abnormalities,” Seidman said.
Investigators of the PALOMA-2 study (NCT01740427) assessed palbociclib plus letrozole (Femara) versus placebo plus letrozole in 666 treatment-naïve postmenopausal women with hormone receptor–positive, HER2-negative mBC.9 After a median follow-up of 23 months, the study showed a PFS of 24.8 months in the palbociclib arm compared with 14.5 months in the placebo arm (HR, 0.58; 95% CI, 0.46-0.72; P < .001). With extended follow-up of approximately 38 months, the PFS benefit increased further with the addition of palbociclib, reaching 27.6 months. PFS remained unchanged for patients treated with placebo plus letrozole (HR, 0.563; 95% CI, 0.461-0.687; P < .0001).10
“We’re still waiting for PALOMA-2 to read out in terms of survival,” Cristofanilli said. OS is generally considered the most important end point in clinical trials, but Seidman pointed out that “PFS still has primacy in the f irst-line setting for estrogen receptor–positive, HER2-negative disease.”
In the PALOMA-3 study (NCT01942135), which assessed palbociclib plus fulvestrant versus placebo plus fulvestrant in the second-line setting, the addition of the CDK4/6 inhibitor resulted in an almost 7-month longer median OS after 44.8 months of follow-up in 521 women who relapsed or progressed on endocrine therapy.11 The median OS was 34.9 months with palbociclib plus fulvestrant compared with 28.0 months with placebo plus fulvestrant (HR for death, 0.81; 95% CI, 0.64-1.03; P = .09), but the finding did not reach statistical significance. When considering the subset of patients with sensitivity to previous endocrine therapy (n = 410), the median OS was 39.7 months in the palbociclib arm and 29.7 months in the placebo arm, showing an absolute difference of 10 months (HR, 0.72; 95% CI, 0.55-0.94).11
“If in the first line you have de novo and bone-only disease, there’s no question that using palbociclib in combination with an AI is absolutely a reasonable choice,” Cristofanilli said. The agent may be considered for patients with visceral disease, having shown a PFS benefit over placebo plus an AI and placebo plus fulvestrant combinations; nevertheless, a [different] CDK4/6 inhibitor may be better for such patients based on available data. For example, some data suggest the benefit of palbociclib is reduced in patients with visceral disease compared with those without. Specifically, in the second-line setting, the median OS was 27.6 months among the 311 patients with visceral metastatic disease treated with palbociclib plus fulvestrant versus 46.9 months in the 210 patients without visceral metastatic disease who were treated with this combination.11
Ribociclib has been assessed in various combinations and populations in the first- and second-line settings. In MONALEESA-2 (NCT01958021), investigators evaluated firstline ribociclib plus letrozole compared with placebo plus letrozole in 668 postmenopausal women, with a median follow-up of 26.4 months. Findings showed a 9.3-month PFS benefit in the ribociclib arm (25.3 vs 16.0 months; HR, 0.568; 95% CI, 0.457-0.0704; log-rank P = 9.63 x 10−8).12 The OS data were still immature, but 50 deaths in the ribociclib arm and 66 in the placebo arm had occurred.
MONALEESA-3 (NCT02422615) included treatment-naïve postmenopausal women as well as postmenopausal women who had received up to 1 line of endocrine therapy in the advanced setting. Investigators randomized patients 2:1 to either ribociclib plus fulvestrant (n = 484) or placebo plus fulvestrant (n = 242).13 Both treatment-naïve and previously treated women showed benefit with the addition of ribociclib, with a median PFS of 20.5 months with the combination versus 12.8 months with placebo (HR, 0.593; 95% CI, 0.480-0.732; P < .001).14
With an extended follow-up of 42 months, treatment-naïve patients had a PFS of 33.6 months in the ribociclib arm compared with 19.2 months in the placebo arm.14 The addition of ribociclib also showed significant OS benefit, with a 28% difference in the relative risk of death, which was consistent across most subgroups. The estimated OS rate was 57.8% for ribociclib versus 45.9% for placebo (HR, 0.72; 95% CI, 0.57-0.92; P = .00455).14
Similarly, data from an extended follow-up of 42 months in the MONALEESA-7 study (NCT02278120) showed an OS benefit in pre- and perimenopausal women treated with ribociclib plus endocrine therapy and goserelin (Zoladex) compared with placebo plus endocrine therapy and goserelin. The estimated OS rates were 70.2% and 46.0%, respectively (HR for death, 0.71; 95% CI, 0.54-0.95; P = .00973 by log-rank test).15
“If you have premenopausal patients who develop metastatic recurrence or have not been treated, [the use of] ribociclib is supported by the data,” Cristofanilli said. A unique concern with ribociclib is the potential for QT interval prolongation.7 Subsequently, “there’s still the need to do ECG [electrocardiogram] monitoring every 2 weeks during the first month of treatment to look for QT prolongation,” Seidman said.
Investigators of the MONARCH 3 study (NCT02246621) compared abemaciclib plus endocrine therapy versus placebo plus endocrine therapy in 493 postmenopausal women who had received no prior systemic therapy in the advanced setting.16 At the planned interim analysis, which occurred after 189 events, median PFS was not reached in the abemaciclib arm versus 14.7 months in the placebo arm. The final PFS analysis, which occurred after a median follow-up of 26.73 months, showed a significantly longer PFS in the abemaciclib arm compared with the placebo arm (28.18 vs 14.76 months, respectively; HR, 0.540; 95% CI, 0.4180.698; P = .000002).17 “We’re still waiting for MONARCH3 to read out in terms of survival,” Cristofanilli said.
The MONARCH 2 study (NCT02107703) compared abemaciclib plus fulvestrant with placebo plus fulvestrant as a second-line treatment for 669 pre-, peri-, or postmenopausal women who had progressed on endocrine therapy.18 Follow-up to 47.7 months showed a significant PFS benefit with the addition of abemaciclib, with a median PFS of 16.9 months in the abemaciclib arm versus 9.3 months in the placebo arm (HR, 0.536; 95% CI, 0.445-0.645). Additionally, investigators observed an OS benefit in the abemaciclib arm (46.7 vs 37.3 months, respectively; HR, 0.757; 95% CI, 0.6060.945; P = .01).
“If you have a [patient with] primary- resistant disease, maybe with visceral metastases or liver disease, abemaciclib is something that may make them more comfortable,” Cristofanilli said. A challenge with abemaciclib is that it has a higher risk of gastrointestinal toxicity than the other CDK4/6 inhibitors. However, these toxicities can be managed. “With experience, time, and the use of antimotility agents, physicians have become comfortable using this agent,” he said. Cristofanilli also explained that in MONARCH-3 the likelihood of discontinuing abemaciclib because of diarrhea was low. In the final PFS analysis of the trial, only 6 abemaciclib-treated patients (1.8%) discontinued treatment because of diarrhea.17 Most cases of diarrhea (72.8%) were low grade, occurred early in the treatment course (69.1%), and were successfully managed with antidiarrheal medications and dose adjustments.
A concern with CDK4/6 inhibitors is that they may increase patients’ susceptibility to and the severity of COVID-19 because the inhibitors are associated with decreased neutrophil counts. Findings from a small study presented during the 2020 American Association for Cancer Research Virtual Meeting: COVID-19 and Cancer suggested that withdrawal or dose reduction of these agents may reduce the incidence of COVID-19; however, the findings did not reach statistical significance and the investigators noted a need for higher-quality data.19 Hurvitz asked the participants whether the pandemic has caused them to change their practices regarding CDK4/6 inhibitors.
“I’m not sure I consciously choose CDK4/6 inhibitors differentially because of the pandemic or make across-the-board dose reductions or tell patients to stay home and not come in for necessary monitoring,” Seidman said. “We can all be afraid of COVID-19, and we should be, but our patients fear the cancer that they have—the known rather than the unknown,” he added.
Cristofanilli agreed. “I have seen some cases of COVID-19 in our clinic, irrespective of CDK4/6 use. So, I would say that neutropenia probably doesn’t increase the risk of COVID-19; it’s an independent prior infection. Even the length of infection seems to be a relatively independent disease,” he said. Nevertheless, he noted that COVID-19 has had an impact on his patients and some of his treatment decisions. “We have seen patients who have comorbid conditions but are unable to get to the hospital because of COVID-19. Patients with inflammatory bowel disease were begging me not to give a drug that would produce diarrhea, so I didn’t. They responded, of course, to palbociclib, without having a problem. There are some other patients who were started on palbociclib and developed a neutropenia very quickly, so we dropped their dose to 75 mg,” he said. Although COVID-19 has led to some unique challenges, Cristofanilli emphasized it should not be a driver for preventing someone from using a CDK4/6 inhibitor.