Do CDK4/6 inhibitors improve upon adjuvant endocrine therapy?


Two phase 3 trials have left physicians and patients in a quandary about the adjuvant use of CDK4/6 inhibitors in endocrine-sensitive breast cancer.

In the monarchE trial, adding abemaciclib to adjuvant endocrine therapy improved invasive disease-free survival (IDFS) and distant relapse-free survival (DRFS) in patients with hormone receptor–positive, HER2/neu oncogene–negative early breast cancer.

However, in the PALLAS trial, adding palbociclib to adjuvant endocrine therapy provided no benefit in such patients.

CDK4/6 inhibitors are highly valuable in women with metastatic hormone receptor–positive breast cancer, but these were the first reports of large, prospective trials of CDK4/6 inhibitors in the adjuvant setting.

Both trials were presented at the European Society for Medical Oncology Virtual Congress 2020.

Erica L. Mayer, MD, of Dana Farber Cancer Institute in Boston, presented results from PALLAS.

Stephen R.D. Johnston, PhD, of Royal Marsden Hospital NHS Foundation Trust in London, presented results from monarchE.

Dr. Stephen R.D. Johnston

Patients and treatment in monarchE

The phase 3 monarchE trial included 5,637 women with hormone receptor–positive, HER2/neu oncogene–negative early breast cancer. About 44% of patients were premenopausal, and 95% had received adjuvant or neoadjuvant chemotherapy.

The patients selected for the study were at high risk for an early relapse, by virtue of having at least four involved axillary nodes. Patients with one to three positive nodes were eligible if the primary tumor was at least 5 cm, had high histologic grade, or had a Ki-67 of 20 or more.

The patients were randomized to receive standard endocrine therapy alone (investigator’s choice) or standard endocrine therapy plus abemaciclib. Endocrine therapy was continued for 5-10 years (physician choice), and the abemaciclib was continued at 150 mg twice daily for the first 2 years.

Treatment commenced postoperatively and after radiation, neoadjuvant, or adjuvant chemotherapy (when indicated).

Credit: Sheheryar Kabraji, Sridhar Ramaswamy/National Cancer Institute

Results of monarchE

At a median follow-up of 15.5 months, abemaciclib plus endocrine therapy demonstrated a statistically significant improvement in IDFS, compared with endocrine therapy alone (hazard ratio, 0.747; P = .0096). This corresponded to a 25% relative reduction in the risk of an IDFS event.

The curves separated early and continuously. The 2-year absolute IDFS rates were 92% in the abemaciclib arm and 89% in the control arm.

A similar improvement was observed for DRFS (HR, 0.717; 3.3% absolute difference in distant recurrences). Most of the relapses prevented were in liver and bone.

In the abemaciclib arm, the most frequent adverse events (AEs) were diarrhea, neutropenia, and fatigue. In the control arm, the most frequent AEs were arthralgia, hot flashes, and fatigue. Interestingly, hot flashes and arthralgias were less common in the abemaciclib arm.

More patients receiving abemaciclib had the rare, serious toxicities of venous thromboembolism (2% vs. 0.5%) and interstitial lung disease (3% vs. 1%).

About 17% of patients discontinued abemaciclib prematurely because of toxicity. Of 463 patients who discontinued abemaciclib, 306 remained on endocrine therapy afterward.

Patients and treatment in PALLAS

The phase 3 PALLAS study enrolled 5,760 patients within 12 months of diagnosis and 6 months of initiating adjuvant endocrine therapy. About 82% of patients had stage IIB/III disease, and about 83% had received chemotherapy.

Like monarchE, PALLAS was relatively weighted toward a prognostically poor group of early breast cancer patients.

Patients were randomized to standard endocrine therapy with or without palbociclib at 125 mg daily for 21 days out of every 28 days. As in monarchE, the CDK4/6 inhibitor was continued for 2 years, and endocrine therapy was administered for 5-10 years (investigator choice).

The primary objective of PALLAS was to compare IDFS between treatment arms. Secondary objectives included other recurrence endpoints, safety, quality of life, and adherence. Biospecimens were required for translational science correlative studies.

Results of PALLAS

A prespecified futility boundary was crossed at the second interim analysis, at a median follow-up of 23.7 months, when 67% of events had occurred.

IDFS rates were 88% in the palbociclib arm and 89% in the control arm (HR, 0.93; P = .51). About three-quarters (76%) of events occurred in patients in high-risk clinical groups.

DRFS rates were 89% in the palbociclib arm and 91% in the control arm (HR, 1.00). Palbociclib did not reduce recurrence overall or in any subgroup.

AEs were more common with palbociclib. In fact, 42% of patients discontinued palbociclib prematurely, the majority because of AEs. The rate of premature endocrine therapy discontinuation was 7%.

Toxicity from palbociclib included neutropenia (but not neutropenic fever), leukopenia, fatigue, thrombocytopenia, anemia, upper respiratory tract infection, and alopecia.

Explaining the disparate results

If the results of monarchE are true, abemaciclib is the first real advance in hormone receptor-positive breast cancer adjuvant treatment in many years and has the potential to save thousands of lives.

However, the juxtaposition of the monarchE presentation with the completely negative results of the PALLAS trial is sobering.

Discussant George Sledge Jr, MD , of Stanford (Calif.) University, offered a few possible explanations for the discrepant results.

First, he said, the hypothesis could be wrong. This seems unlikely since one of the studies was positive at the early time-point of 15.5 months of median follow-up. Long-term results will show whether the IDFS, DRFS, and – most importantly – overall survival curves separate continuously.

Dr. George Sledge Jr.
Another explanation could be that the study designs were different, precluding a uniform outcome across the trials. PALLAS included not only high-risk patients but also a lower-risk subset that was reflective of patients generally seen in clinical practice.

The monarchE trial, on the other hand, included only high-risk patients and, therefore, was weighted toward seeing more early events. Perhaps the monarchE patients had occult metastatic disease already, a setting in which CDK4/6 inhibitors would more likely show value.

Yet another explanation is that a difference between abemaciclib and palbociclib could contribute to the difference in outcomes. This seems unlikely since many trials of various CDK4/6 inhibitors have shown similar outcomes in stage IV breast cancer.

However, Dr. Sledge said that, among the CDK4/6 inhibitors, abemaciclib has relatively greater potency against CDK4 than CDK6. It is also possible that, since abemaciclib and palbociclib are administered according to different schedules (i.e., continuously versus intermittently), “schedule” is playing more of a role in the adjuvant environment than in metastatic disease.

Finally, the disparate results may come down to the play of chance, Dr. Sledge said.

Research continues

With an issue as important as preventing distant relapse in women with early breast cancer, it is inappropriate to be glib or dismissive about the divergent results of these two large, multi-institutional, well-conducted, phase 3 trials.

However, it is currently impossible to know whether regulatory agencies should approve, doctors should prescribe, and patients should receive adjuvant treatment with CDK4/6 inhibitors. 

The issue is crucial because one constant result between the trials was that CDK4/6 inhibitors contribute toxicity and significant expense to patient care.

Analysis of patient-reported outcomes, quality of life, treatment adherence, and clinical-pathologic correlation utilizing the more than 6,000 biospecimens collected as part of the TRANS-PALLAS study is awaited. 

Similarly, translational studies of tissue and plasma collected from monarchE patients will define whether there is a particular population for whom CDK4/6 inhibitor therapy can reverse primary endocrine resistance that was not defined by the clinical and pathologic subgroups defined in the interim efficacy analysis of monarchE.

Additional information about the fate of those patients who prematurely discontinued CDK4/6 inhibitor therapy and/or endocrine therapy, as well as long-term follow-up of all participants may tell the tale of whether this early snapshot will be representative of the long-term results of these two important trials.

Results of the ongoing NATALEE trial (NCT03701334)  of adjuvant ribociclib will also help to clarify the role CDK4/6 inhibitors should play in the adjuvant setting.

For now, oncologists are left with a dilemma and a complicated, lengthy discussion with patients.

Credit: Bruce Wetzel and Harry Schaefer (Photographers)/
National Cancer Institute

The PALLAS trial was funded by Pfizer and cosponsored by Alliance Foundation Trials, Austrian Breast & Colorectal Cancer Study Group, NSABP Foundation, PrECOG, and Breast International Group. The monarchE trial was sponsored by Eli Lilly and the NSABP Foundation. Dr. Mayer disclosed relationships with Eisai, Pfizer, Lilly, Novartis, Myriad, and Roche/Genentech. Dr. Johnston disclosed relationships with Eli Lilly, AstraZeneca, Novartis, Pfizer, Puma Biotechnology, Eisai, and Roche/Genentech. Dr. Sledge did not disclose any conflicts of interest.

Dr. Lyss was a community-based medical oncologist and clinical researcher for more than 35 years before his recent retirement. His clinical and research interests were focused on breast and lung cancers, as well as expanding clinical trial access to medically underserved populations. He is based in St. Louis. He has no conflicts of interest.

Dr. Alan P. Lyss