A Study of Palbociclib in Addition to Standard Endocrine Treatment in Hormone Receptor Positive Her2 Normal Patients With Residual Disease After Neoadjuvant Chemotherapy and Surgery

Introduction:

Patients with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) primary breast cancer who receive chemotherapy as part of their primary treatment are at higher risk of relapse. About a third of the patients with residual invasive disease after neoadjuvant chemotherapy (NACT) will relapse despite adjuvant endocrine therapy (ET). The risk of relapse can be assessed more accurately using the clinical pathological staging-estrogen receptor grading (CPS-EG) scoring system. This involves the tumor stage before treatment start and at surgery, the estrogen receptor (ER) status, and the pathologic grading. A higher score indicates a higher risk of relapse. Patients with a score of three and higher had a disease-free survival of 70% at 5 years despite receiving ET. When combined with the group having a CPS-EG score of two and involved lymph nodes, the disease-free survival increased to 77%. This group comprises about 25% of all patients with residual disease after NACT.

Therapeutic inhibition of cyclin-dependent kinase 4 and 6 (CDK4/6) by palbociclib combined with ET demonstrated clinically relevant efficacy in metastatic HR+ and HER2- breast cancer irrespective of biomarker selection. After the pivotal PALOMA-1 trial, which led to accelerated approval of palbociclib in February 2015, further phase III trials demonstrated that the use of CDK4/6 inhibitors in pre- and postmenopausal patients with hormone-sensitive or resistant cancers, in first-line and pretreated metastatic breast cancer, improves progression-free and overall survival (OS).

Thus, we hypothesized that palbociclib may also be active in patients with residual disease after NACT who are at high risk of relapse. Based on the data from PALOMA-1, we designed the PENELOPE-B trial assessing the efficacy of one year palbociclib vs placebo added to any ET in HR+ and HER2- breast cancer patients with residual disease and high risk after NACT, based on the CPS-EG score.

Patient Selection and Study Design:

PENELOPE-B is a prospective, multicenter, multinational, randomized, double-blind, placebo controlled phase III study investigating the addition of palbociclib for one year to standard adjuvant ET for patients with high-risk residual disease according to the CPS-EG score after NACT for early HR+ and HER2- breast cancer.

The trial was sponsored by GBG Forschungs GmbH in collaboration with NSABP Foundation (plus I-SPY and CCTR), ABCSG, AGO-B, ANBCSG, BIG, Geicam, ICR-CTSU, JBCSG, and KCSG. Pfizer Inc funded the trial and provided drug. The trial was conducted according to ICH-GCP guidelines and the Declaration of Helsinki. The clinical trial Protocol (online only) was approved by the respective health authorities and ethics committees or institutional review boards. All patients provided written informed consent for trial participation, data transfer, and biomaterial collection. The trial was overseen by the International Steering Committee and the GBG Boards and supervised by an Independent Data Monitoring Committee (IDMC).

Eligible patients were randomly assigned in a 1:1 manner in permuted blocks of alternating size 4/6 stratified by risk status (CPS-EG sore ≥ 3 v 2/ypN+), nodal involvement after surgery (ypN0-1 v ypN2-3), Ki-67 (≤ 15% v > 15%), age (≤ 50 v > 50 years), and global region of participating site (Asian v non-Asian).

Treatment:

Patients received either palbociclib 125 mg orally once daily for 21 days followed by 1 week off treatment for a total duration of 13 4-week cycles or matching placebo in addition to standard adjuvant ET at the discretion of the investigator given for at least 5 years. Dose interruptions, reductions, or delays according to predefined toxicity management were acceptable in the case of significant treatment-related toxicity.

Objectives and End Points:

The primary objective of the study was to compare the invasive disease-free survival (iDFS) defined as the time in months between random assignment and first event (ipsilateral invasive in-breast or locoregional recurrence, distant recurrence, invasive contralateral breast cancer, second primary invasive cancer [nonbreast], or death because of any cause) for palbociclib versus placebo. Secondary end points included iDFS excluding second primary invasive nonbreast cancers, distant disease-free survival, OS, locoregional relapse-free interval (LRRFI), safety (which was reported as adverse events (AEs) irrespective of relatedness to study treatment based on National Cancer Institute Common Toxicity Criteria v4.0), and compliance.

All time-to-event end points were analyzed in the intent-to-treat population comprising all randomly assigned patients. Compliance and safety were analyzed in the safety analysis set including all randomly assigned patients who took study medication at least once. For the patients randomly assigned to placebo who received palbociclib at least once during the trial, the treatment group allocation for safety and compliance analyses was the palbociclib arm.

Sample Size and Interim Analyses:

In the initial sample size computation, 233 iDFS events were required to detect a hazard ratio for palbociclib/placebo of 0.67 (from 72% to 80% 3-year iDFS rate) corresponding to a clinically relevant risk reduction of 33% for invasive disease with a power of 85% using a two-sided stratified log-rank test and an overall two-sided significance level of 0.05. Eight hundred patients were planned to be enrolled. To accelerate recruitment after 68 patients had been enrolled, the population was expanded to patients with a CPS-EG score of two and ypN + who were also identified as high-risk patients but with a generally lower risk than the patients with CPS-EG three. Thereafter, the target hazard ratio for palbociclib/placebo was updated to 0.685 (from 77% to 83.6% 3-year iDFS rate), and the iDFS events were increased to 255 and sample size to 1,100 patients.

The study had an adaptive design with two interim efficacy analyses to monitor futility, to test for overwhelming efficacy, and to allow for sample size re-estimation. Safety was assessed at both interim analyses. O'Brien and Fleming type stopping boundaries based on the Lan-DeMets spending function were used in the interim analyses.

Statistical Analysis:

Final analysis of the primary end point iDFS was planned after 290 iDFS events with a nominal significance level of 0.0463 (two-sided). To address the concern of possible inflation of the type I error because of the sample size increase, statistical significance was determined using a weighted statistic of the stratified log-rank test (stratified by risk status, nodal involvement after surgery, Ki-67, age, but not global region of participating site, as prespecified in the Protocol) based on the method of Cui, Hung, and Wang (CHW) with CHW interim monitoring implemented in EAST version 6.5 (Cytel Inc). The 95% repeated CI was reported taking into account the adaptive sample size re-estimation and group sequential nature of the design.

For all other tests, the alpha was set to 0.05 (two-sided). Adjustment for multiple testing in the other tests was not planned. The Kaplan-Meier method was used to estimate the survival probability at specific time points together with a two-sided 95% CI. Univariable Cox-proportional hazards models stratified by the same factors as in the primary analysis were used for time-to-event end points (except LRRFI) to report hazard ratios with 95% CI. LRRFI was analyzed using the cumulative incidence function for rates at specific time points with stratified Gray's test for comparison and stratified univariate Fine-Gray model to report hazard ratio for treatment. Fisher's exact test was used to compare frequencies of AEs between arms.

All statistical analyses were performed using SAS Version 9.4 with SAS Enterprise Guide Version 7.1 on Microsoft Windows 10 Enterprise. Data cutoff date was August 24, 2020.