Advances in Hormone Receptor Positive Metastatic Breast Cancer - Episode 9
An overview of key presentations from ESMO 2021 that continues studying the potential effect of novel-based treatment options for use in hormone receptor-positive metastatic breast cancer.
Gabriel N. Hortobagyi, MD, FACP: In addition to the CDK [cyclin-dependent kinase]4/6 inhibitors, there is much interest in exploring and understanding mechanisms of resistance to these agents. There is a long list of proposed potential mechanisms of resistance, including loss of RB [retinoblastoma protein], B16 abnormalities, and the development and overexpression of some cyclins in cyclin-dependent kinases. In the interest of developing agents that overcome, or perhaps prevent resistance through CDK4/6 inhibitors there is an extensive body of research to inhibit other cyclin-dependent kinases, including CDK2, CDK7, CDK9, and CDK12. Recent reports have suggested that a CDK7 inhibitor in combination with fulvestrant [Faslodex] has significant activity in a CDK4/6-resistant population. This is a first and important step in the development of this new field, which should expand the utilization of treatments that focus on the cell cycle. This has an independent role in the development and growth of malignant diseases, not only breast cancer but also other types of cancers. It is worth mentioning at this time that while the CDK4/6 inhibitors were developed in hormone receptor-positive HER2 [human epidermal growth factor receptor 2]-negative metastatic breast cancer, there is much reason to believe that they will have important roles to play in other types of breast cancers as well as other types of solid tumors and perhaps hematologic malignancies.
Aditya Bardia, MD, MPH: A common question that comes up in the clinic when a patient has disease progression on first-line aromatase inhibitor plus CDK4/6 inhibitor is, what should be the second-line therapy? I think in part they should be guided based on the mechanism of resistance. Based on the mechanism of resistance, we should choose sequential therapy and the sequential therapy should be directed against the mechanism that resulted in resistance to begin with. The common mechanisms of resistance to endocrine therapy plus CDK4/6 inhibitors include the development of ESR1 [estrogen receptor 1] mutation. ESR1 mutations are mutations in the ligand-binding domain of the estrogen receptor. When a patient is on an aromatase inhibitor for a prolonged period of time, aromatase inhibitors lower estrogen. One way by which the tumor can develop resistance to aromatase inhibitors is by developing a mutation of the estrogen receptor. When there’s a mutation of the estrogen receptor, the tumor becomes estrogen independent. You can have the best aromatase inhibitor in the world that would not work because the tumor is estrogen-independent, but it’s still dependent on the estrogen receptor. Drugs that target the estrogen receptor directly like SERDs [selective estrogen receptor degrader] would have activity in that setting. For a patient with ESR1 mutation, the utilization of a SERD, either as a single agent or in combination therapy, would be a very irrational strategy. In terms of other mechanisms of resistance, we’ve seen the development of RB1 mutation. CDK4/6 inhibitors block CDK4/6. CDK4/6, in general, interacts with RB and when it interacts with RB, it releases E2F, which allows the cell to move forward in the cell cycle. One of the mechanisms of resistance to CDK4/6 inhibitor is RB loss. You lose RB, and when you lose RB, CDK4/6 cannot interact with RB and thus CDK4/6 inhibitors would have limited efficacy. In that setting, the use of something that’s more downstream, maybe CDK2 or another downstream inhibitor would potentially have efficacy. The other thing we’ve seen is the upregulation of the receptor tyrosine kinase and HER2 PI3K [phosphatidylinositol 3-kinase] AKT [protein kinase B] pathway. If that pathway were the major driver of resistance, then the use of PI3K or mTOR [mechanistic target of rapamycin] inhibitors would potentially have efficacy in that setting. It just highlights based on the mechanism of resistance. One could potentially select rational therapy. That’s where liquid biopsies or plasma-based genotyping comes in because you are doing sequencing at the time of disease progression. You’re looking at the snapshot of a tumor at that time and if you see a mutation, say HER2 or PI3KCA, you can use a targeted therapy that’s directed against that mutation, which was likely the driver of resistance in that setting. Now finding terms of newer drugs or SERDs as we were reviewing earlier, and there are a number of oral SERDs in clinical development—amcenestrant, giredestrant, elacestrant. There are different companies making oral SERDs and they would particularly be valuable in the setting of resistance to AI [aromatase inhibitor] because they could target ESR1 mutations, but because these agents bind to estrogen receptors directly, they potentially would also have value in earlier lines, including [in the] adjuvant setting. There are ongoing studies evaluating these SERDs in the adjuvant setting.
Sara A. Hurvitz, MD: Targeting of the androgen receptors [ARs] has been something that many investigators have tried to do in clinical trials. Not so long ago we were looking at the use of AR targeted therapy for triple-negative breast cancer with modest results, but nothing leading to an FDA [Food and Drug Administration] approval. But we recently saw data relating to enobosarm, which is a selective androgen receptor modulator, an oral agent that was evaluated in ER+ [estrogen receptor-positive]/AR+ [androgen receptor-positive] metastatic breast cancer patients who are postmenopausal and had progressed on prior endocrine therapy. This phase 2 parallel-arm study evaluated enobosarm at 2 different doses, 9 milligrams and 18 milligrams, and enrolled about 136 heavily pretreated patients. In fact, patients had had a median of 4 prior lines of hormonally-targeted therapy, and over 90% had had a prior line of chemotherapy. We don’t expect to see amazing results in this very heavily pretreated setting. In this study, over three-quarters of patients had AR expression by central review of their tumor and the clinical benefit rate at 24 weeks, which was the primary end point of this study, was around 32% in the 9-milligram arm and 29% in the 18-milligram arm with progression-free survival being around 4 to 5 months for both of the treatment arms. Interestingly, quality of life didn’t appear to be diminished in patients treated with this agent, which is promising. Grade 3/4 adverse events tended to be an elevation of transaminases, hypercalcemia, some headache, and fatigue. But overall, the data from an efficacy standpoint and safety standpoint do support the evaluation of this agent in a larger clinical trial.
From a biological standpoint, it makes sense to target BCL-2 [B-cell lymphoma 2]. This is a protein that’s overexpressed in around 80% or greater of ER+ breast cancers. This is an antiapoptotic protein. It makes sense to look at targeted agents, and venetoclax [Venclexta] is 1 such agent. The VERONICA clinical trial was a phase 2 randomized study comparing fulvestrant plus venetoclax at 800 milligrams orally daily vs fulvestrant alone in patients who had metastatic breast cancer, that was hormone receptor-positive HER2–. Patients were allowed to participate if they’d received less than or equal to 2 lines of endocrine therapy, as well as the CDK4/6 inhibitor, but no chemotherapy. With about 103 patients in the intent-to-treat analysis of the clinical benefit rate for the 2 arms was no different. It was about 11.8% in the venetoclax arm vs 13.7% in the fulvestrant single-agent arm. Similarly, the progression-free survival was not improved with venetoclax, 2.69 months vs 1.94 months. When looking at the outcomes based on BCL-2 expression being either high or low, there were still no differences between the 2 arms that would indicate a benefit by using venetoclax. In addition, there were more adverse events noted by adding the BCL-2 inhibitor. The biomarker analysis is ongoing in this clinical trial. Overall survival is not yet mature, so I’m sure we’ll see more data relating to this, but at this point, the VERONICA study is a negative trial.
Transcript edited for clarity.