乌鸦传媒

Expert Critique

FROM THE ASCO Reading Room

Ashley Love Sumrall, MD, FACP Section Chief of Neuro-Oncology, Medical Oncologist Levine Cancer Institute

As data builds for optimal treatment of HR+/HER2- advanced breast cancer, the CDK4/6 inhibitors palbociclib, ribociclib, and abemaciclib have become standard options. This class of medications have lengthened PFS as first-line or salvage therapy for patients with HR+/HER2- breast cancer. We await OS data, and clinicians have questions, as we do not have biomarkers to predict response or guide therapy. Also, not all patients respond to this line of therapy initially. Efforts to improve our understanding of this have been ongoing, but slow.

Dr. Jennifer J. Gao and colleagues pooled raw patient-level data from five randomized controlled trials of CDK 4/6 inhibitors in first and second line (with hormonal blockade) 鈥� the PALOMA-2, PALOMA-3, MONALEESA-2, MONARCH-2, and MONARCH-3 studies. All subsets had similar improvement in PFS with the addition of a CDK 4/6 inhibitor. Understanding how beneficial this therapy is, investigators keep searching for potential markers of response to CDK4/6 inhibition, including looking at subtypes of disease. More data from the 2018 ASCO annual meeting from Dr. Richard S. Finn and co-investigators demonstrated that PFS is prolonged by adding palbociclib to hormonal blockade in the double-blind phase III PALOMA-2 (palbociclib + letrozole) and PALOMA-3 (palbociclib + fulvestrant) trials. STEPP (Subpopulation Treatment Effect Pattern Plot) analyses showed that patients with luminal A or B disease benefited from combining palbociclib and endocrine therapy. Benefit with palbociclib persisted regardless of the presence of visceral or nonvisceral metastases.

Lastly, the MONALEESA-2 trial compared PFS between letrozole +/- ribociclib for first-line treatment of HR+ metastatic breast cancer. Previous biomarker analyses using IHC for protein expression of pRb, p16, and Ki67, as well as gene expression of CDKN2A, CCND1, ESR1, next-generation sequencing of circulating tumor DNA for mutations in PIK3CA, TP53, CDG1, FGFR1/ZNF703 and receptor tyrosine kinase genes showed consistent benefit of ribociclib across biomarkers. Improved mean PFS was demonstrated across all gene expression signatures. In this trial, Dr. Gabriel Hortobagyi and co-researchers obtained mRNA expression data from baseline tumor samples of 391 of 668 patients. Unfortunately, samples were not available from all patients. Samples were classified into high or low mRNA expression subgroups using median expression as the cutoff, felt to be an arbitrary value. A trend toward greater benefit of ribociclib was observed with high versus low ESR1 expression (HR 0.39 versus 0.74), high versus low cell cycle-control gene expression, and low versus high RTK gene expression. It is unclear if use of mRNA expression will prove to be a biologically relevant cutoff.

Full Critique

The CDK4/6 inhibitors palbociclib, ribociclib, and abemaciclib have extended progression-free survival (PFS) when used as first-line or later therapy for patients with hormone receptor (HR)-positive/HER2-negative advanced breast cancer. Questions remain about incorporating CDK4/6 inhibitors into clinical practice, as about 20% of patients do not respond to CDK4/6 inhibitors initially, and resistance ultimately develops in all patients. Clinical markers and biomarkers that predict response would help to guide CDK4/6 inhibitor therapy.

At this year's ASCO annual meeting, several investigators examined potential markers of response to CDK4/6 inhibition and assessed whether use of these agents may be successfully expanded to patients with less common breast cancer subtypes. Offering her perspective as a discussant was Erika Hamilton, MD, director of the Breast Cancer Research Program of Sarah Cannon Research Institute in Nashville.

Breast Cancer Subtypes

Clinical data are limited with respect to the benefit and value of adding CDK 4/6 inhibitors to endocrine therapy in less common breast cancer subtypes, such as progesterone receptor (PR)-negative, de novo metastatic and lobular breast cancer.

Jennifer J. Gao, MD, of the Food and Drug Administration, and colleagues of CDK 4/6 inhibitors in the first and second line in combination with an aromatase inhibitor (AI) or fulvestrant -- the PALOMA-2, PALOMA-3, MONALEESA-2, MONARCH-2, and MONARCH-3 studies. All clinical breast cancer subsets had a similar improvement in PFS with the addition of a CDK 4/6 inhibitor, as follows:

• In patients with progesterone receptor (PR)-negative breast cancer (n=490), mean PFS improved from 7.4 to 16.5 months with the addition of a CDK4/6 inhibitor (HR 0.50, 95% CI 0.40-0.64)
• In patients with lobular breast cancer (n=264), mean PFS improved from 9.2 to 16.1 months (HR 0.58, 95% CI 0.42-0.80)
• In patients with bone-only metastases (n=875), mean PFS improved from 15.5 to 27.9 months (HR 0.55, 95% CI 0.45-0.67)
• In those with de novo metastatic disease (n=617), mean PFS improved from 16.8 to 27.8 months (HR 0.59, 95% CI 0.46-0.76)
• In those with a disease-free interval >12 months (n=929), mean PFS improved from 14.2 to 25.7 months (HR 0.55, 95% CI 0.46-0.67)

The benefits of CDK 4/6 inhibitors were seen in the first-line setting with an AI and in the second-line setting with fulvestrant. The researchers concluded that the natural history of disease and the magnitude of CDK 4/6 inhibitor benefit were similar for certain subsets that were previously believed to exhibit different behavior, such as lobular and PR-negative breast cancer.

Hamilton's take was that analyses of more patients with relatively rare breast cancer subsets are needed to make definitive conclusions.

Potential Clinical Markers and Biomarkers

In the second study in the poster discussion session, Richard S. Finn, MD, of UCLA, and co-investigators examined the by prognostic and intrinsic cancer subtype in the double-blind phase III PALOMA-2 (palbociclib in combination with letrozole) and PALOMA-3 (palbociclib in combination with fulvestrant) trials in women with HR-positive, HER2-negative advanced breast cancer. In PALOMA-2, median PFS was 27.6 months with the combination of palbociclib and letrozole as first-line therapy, compared with 14.5 months with placebo plus letrozole (HR 0.56, P<0.000001). In PALOMA-3, median PFS was 11.2 months with palbociclib plus fulvestrant versus 4.6 months with placebo plus fulvestrant in patients whose breast cancer had progressed following endocrine therapy (HR 0.50, P<0.0001).

STEPP (Subpopulation Treatment Effect Pattern Plot) analyses demonstrated that neither treatment-free interval (TFI) nor disease-free interval (DFI) had an impact on the benefit of palbociclib. Patients with luminal subtype disease benefited from the combination of palbociclib and endocrine therapy regardless of luminal A or luminal B subtype. In addition, the consistent benefit with palbociclib persisted regardless of the presence of visceral or nonvisceral metastases, the researchers reported.

Their conclusion: "Together, these data strongly support palbociclib plus endocrine therapy as a standard of care for patients with HR-positive/HER2-negative advanced breast cancer, regardless of baseline TFI/DFI or intrinsic molecule subtype at the time of initial diagnosis or at the time of disease recurrence."

In the third study, the compared PFS between letrozole with or without ribociclib for the first-line treatment of HR-positive metastatic breast cancer. Previous biomarker analyses using immunohistochemistry for protein expression of pRb, p16, and Ki67, in addition to gene expression of CDKN2A, CCND1, ESR1, next-generation sequencing of circulating tumor DNA for mutations in PIK3CA, TP53, CDG1, FGFR1/ZNF703 and receptor tyrosine kinase genes showed consistent benefit of ribociclib across biomarkers.

Gabriel Hortobagyi, MD, of the University of Texas MD Anderson Cancer Center in Houston, and co-researchers obtained mRNA expression data from baseline tumor samples from 391 of 668 patients randomized into MONALEESA-2. The objective was to evaluate efficacy of ribociclib plus letrozole compared with letrozole alone by baseline gene expression. Patients were classified into high or low mRNA expression subgroups using the median expression as the cutoff.

The addition of ribociclib was associated with an improvement in mean PFS across all gene expression signatures. A trend toward greater benefit of ribociclib was observed with high versus low ESR1 expression (HR 0.39 versus 0.74), high versus low cell cycle-control gene expression and low versus high RTK gene expression.

"The shorter median PFS for placebo plus letrozole in patients with high versus low expression of cell-cycle-control genes suggests that high expression of these cell cycle-related genes may incur resistance to treatment with letrozole alone," the team concluded.

Limitations of the study included that mRNA data were obtained from only about 60% of the participants and that median mRNA expression was used as a cutoff, which may not represent a biologically relevant cutoff, said Hamilton.

She concluded from the Finn et al and Hortobagyi et al studies that clinical or tissue/blood biomarkers cannot currently be used to predict benefit from CDK 4/6 inhibition: "All HR-positive metastatic breast cancer patients should be offered CDK 4/6 inhibitors. The data strongly support use of CDK inhibitors regardless of gene-expression profile or molecular signature."

First or Second Line?

She noted that there is a suggestion from examining the PFS curves in the PALOMA-2 and PALOMA-3 trials that CDK 4/6 inhibition as first-line therapy may provide more benefit than if used in a later line of treatment, as evidenced by a 6.5-month longer PFS when used in the first line versus the second line (32.2 versus 25.7 months). "This difference seems clinically significant, but we really can't answer this question with simple addition games or manipulating PFS curves. We ultimately need overall survival data from these trials, patient-reported outcomes, and also cost-benefit analyses."

The more pertinent question, she said, is the choice of endocrine backbone therapy: "There are two populations that we should almost immediately be thinking of for fulvestrant upfront: those that have relapses while on an AI and those with known ESR1 mutations."