A Phase II Study of Abemaciclib for Patients with Retinoblastoma-Positive, Metastatic Triple-Negative Breast Cancer

Abstract

Purpose: Cyclin-dependent kinase 4/6 inhibitors can significantly extend survival when given in combination with endocrine therapy in patients with hormone receptor-positive metastatic breast cancer. However, their activity has been relatively underexplored in patients with metastatic triple-negative breast cancer (mTNBC).

Patients and methods: We conducted a single-arm phase II study of abemaciclib monotherapy in patients with Rb-positive mTNBC. Patients were treated with abemaciclib 200 mg orally twice daily until disease progression, unacceptable toxicity, or withdrawal of consent. The primary endpoint was the objective response rate; secondary endpoints included progression-free survival (PFS), overall survival (OS), clinical benefit rate, disease control rate, and safety and tolerability.

Results: A total of 27 patients were enrolled before the trial was closed early because of slow accrual. Patients had received a median of two lines of systemic therapy in the metastatic setting prior to enrollment. After a median follow-up of 28.5 months, the objective response rate was 0%, the clinical benefit rate was 14.8%, and the disease control rate was 22.2%. The median PFS was 1.94 months (95% confidence interval, 1.84-11.47), and the median OS was 8.44 months (95% confidence interval, 4.57-15.57). Median PFS and OS did not differ significantly based on androgen receptor and PD-L1 status. Pretreatment gene expression profiling of tumor tissue provided some hypothesis-generating insights into biological features associated with clinical benefit in this study. The most common treatment-related adverse events of grade 2 or higher were diarrhea (40.7%), neutropenia (40.7%), anemia (29.6%), and nausea (29.6%).

Conclusions: Abemaciclib monotherapy did not show clinical activity in patients with pretreated Rb-positive mTNBC.

Figure 1.

Trial diagram. Diagram depicting the selection criteria, treatment, and sample collection of patients with TNBC. C, cycle; cfDNA, circulating free DNA; D, day; MBC, metastatic breast cancer.

Figure 2.

Best response, PFS, and OS by treatment arm and pathologic response. A, Waterfall plot colored by best response. B, Kaplan–Meier plot represents PFS probability for all patients with advanced TNBC. Numbers of patients at risk over time are depicted under graph. [Median survival, 95% CI, 1.94 (1.84–11.47)]. Median follow-up time was 14.3 months. C, Kaplan–Meier plot represents OS probability for all patients with advanced TNBC. Numbers of patients at risk over time are depicted under graph [median survival, 95% CI, 8.44 (4.57–15.57)]. Median follow-up time was 28.5 months. PD, progressive disease; PR, partial response; SD, stable disease.

Figure 3.

Clinical response. Swimmer plot demonstrating clinical response and number of treatment days for individual patients. PD, progressive disease; PR, partial response; SD, stable disease.

Figure 4.

Best response and survival probabilities based on AR and PD-L1. A, Waterfall plot colored by best response based on AR status. B, Kaplan–Meier plot represents PFS probability of patients with advanced TNBC that are positive (blue) or negative (orange) for the AR. Numbers of patients at risk over time are depicted under graph [median survival, 95% CI, 1.46 (1.05–11.47) for AR-negative; 6.69 (2.03–NR) for AR-positive]. No difference was detected between the two strata (P = 0.063). C, Kaplan–Meier plot represents OS probability of patients with advanced TNBC that are positive (blue) or negative (orange) for PD-L1. Numbers of patients at risk over time are depicted under graph [median survival, 95% CI, 4.81 (3.58–NR) for AR-negative; 12.41 (11.17–NR) for AR-positive]. No difference was detected between the two strata (P = 0.3). D, Waterfall plot colored by best response based on PD-L1 status. E, Kaplan–Meier plot represents PFS probability of patients with advanced TNBC that are positive (blue) or negative (orange) for PD-L1. Numbers of patients at risk over time are depicted under graph [median survival, 95% CI, 2.04 (1.54–NR) for PD-L1–negative; 1.84 (1.38–NR) for PD-L1–positive]. No difference was detected between the two strata (P = 0.48). F, Kaplan–Meier plot represents OS probability of patients with advanced TNBC that are positive (blue) or negative (orange) for PD-L1. Numbers of patients at risk over time are depicted under graph [median survival, 95% CI, 8.44 (1.54–NR) for PD-L1–negative; 11.50 (6.54–NR) for PD-L1–positive]. No difference was detected between the two strata (P = 0.12). NR, not reached.

Figure 5.

Correlative analysis of tumor gene expression. A, Normalized gene expression [log₂ (FPKM + 1)] of pretreatment tumor biopsy RNA-seq baseline samples between patients without or with clinical benefit for genes of interest with violin plots of density estimates. B, Differential gene expression of baseline samples between patients without or with clinical benefit. Genes with an FDR < 0.10 have their names displayed in black, and genes of interest are shown in blue if down and orange if up in clinical benefit patients. C, Gene set variation analysis enrichment scores of genes in the RBsig expression signature for baseline samples. Boxplots show RBsig scores in tumors from patients who experience clinical benefit or not. Points have been horizontally jittered to avoid overplotting. D, Normalized enrichment scores (NES) and FDR q values from GSEAPreranked run on baseline samples for Hallmark gene sets using log₂ fold change preranked gene lists from testing a contrast comparing patients without and with clinical benefit. Gene sets with an FDR < 0.25 are shown. FPKM, fragments per kilobase of transcript per million; GSEA, Gene Set Enrichment Analysis.