Present in approximately two-thirds of estrogen receptor-positive breast cancers, the role of ESR1 mutations in disease metastasis is poorly defined.
Present in approximately two-thirds of estrogen receptor (ER)-positive breast cancers, the role of ESR1 mutations in disease metastasis is poorly defined.1 “It is an important research area because it is so prevalent and we [must] identify drugs we can [use] to target these ESR1-mutant [tumors],” said Steffi Oesterreich, PhD. “There is some evidence now, from some trials, such as the SOPHIA trial [NCT02492711], that some specific hormone therapy might still have some efficacy, in contrast with aromatase inhibitors, for example. [Arriving at] the therapy [regimen] where we combine novel [agents] with drugs targeting some of these metastatic phenotypes, that is the goal. That is where we would like to get.”
In an interview with OncologyLive®, Oesterreich, a professor in the Department of Pharmacology and Chemical Biology at the University of Pittsburgh Medical Center (UPMC), coleader of the Cancer Biology Program at UPMC Hillman Cancer Center, and codirector and director of education at the Women’s Cancer Research Center, Magee-Womens Research Institute in Pennsylvania, discussed the findings of an analysis she coauthored, which examined the role of ESR1 mutations in the development of metastatic phenotypes in patients with breast cancer.
Oesterreich: ESR1 was the focus [of our analysis] because ER-positive breast cancer is very prevalent. Two-thirds of all breast cancers are ER-positive, and [most] deaths from breast cancer are [associated with] ER-positive disease. Over the past few years, [several] teams, including our own, have shown that approximately 30% of tumors [that] become resistant to endocrine therapy have mutations in the gene coding for the estrogen receptor ESR1.
This is an important problem; this is thousands and thousands of women each year. The reason for looking more into this was to identify druggable dependencies. Basically, [we are trying to] find something in these tumors that they depend on for growth and metastasis—the Achilles heel, so to speak. Because it is so prevalent, we thought we need to study this more.
We used [a few] different approaches, starting with models in the lab. We used genome engineering, where we take a mutation, put it into the endogenous genome of breast cancer cell lines, then look at how these cells behave. We also use clinical specimens from patients with ER-positive, metastatic disease. We compare the tumors, which have wild-type ER with [samples] from patients where the ER is mutant. It is a large collaboration. [The authors of] these papers include cancer biologists, pathologists, surgeons, medical oncologists, [and more]. It is a very large multidisciplinary study, using approaches from different expertise.
Over the past few years, it has been shown that these ER mutations cause drug resistance, they cause resistance to hormonal therapy. We [observed this] as soon as after the mutation and local recurrences in the breast [happened]—when the tumor becomes resistant to hormonal therapy—and we compared this with the currents of the mutations in distant metastases. The difference is, in 1 case, the tumor metastasized and in the other, it did not. Both [tumors] are endocrine resistant and both are drug resistant, but 1 metastasized and the other remained in the breast.
We did not find mutations in the endocrine-resistant tumor in the breast, but we found mutations in the metastatic sites in the bone, liver, and brain. This suggests mutations play a role not only in resistance to therapy, but they [also] make the tumor metastasize. That was unexpected.
[We found] several phenotypes. You could almost say the ER is a master regulator of several metastatic phenotypes. One [we found] particularly interesting made the tumor cells stick together. Usually, you think of the tumor as becoming more metastatic when they become small individual tumor cells and they can move better. But we found that, [in this particular phenotype], the cells stick together better.
We then collaborated with other investigators to look at whether that potentially results in circulating tumor cell clusters. In the blood of patients with metastatic disease, there are tumor cells floating around and we can measure them. In patients with wild-type or ESR1-mutant tumors, we looked at whether there are more single tumor cells floating around or more clusters. In patients with ESR1-mutant disease, we found more circulating tumor cell clusters, again suggesting they stick more together.
The other very important and somewhat surprising finding was that the different mutations—although very close and localized in the ER genes—they have quite different phenotypes. Depending on the mutation, the phenotype is different. That is very important for clinical trial design. If you notice drugs targeting some of these pathways, which are activated, you [must] look and test the drugs in patients with 1 but not the other ER mutation.
Finally, the expression of these basal phenotypes resulting in altered immune infiltration was clearly an interesting finding, as well, [with] potential clinical relevance.
We would like to follow up on the ultimate question: What is the therapy for patients’ ER mutations when regular endocrine therapy does not work any longer? We would like to follow up on these pathways we identified that drive metastasis in young mutant tumors.
For example, we just talked about the stickiness [of the tumor cells], and there are some drugs—not in breast cancer but in other diseases—that can target stickiness. Maybe we could test this in additional models, [and] if this is validated, there could be clinical trial testing.
We identified [several] other [implicated] pathways. Again, [we are] thinking whether these provide opportunities for the dependencies of these tumors. [We must] make sure the tumor depends on it for survival because sometimes you have upregulation of a pathway [but] that does not mean it is a high efficacy drug target.
[Another] idea is [regarding] other immune infiltration. Can it be used? This is challenging, because the analysis of the role of the immune system needs special models—you must have the entire tumor microenvironment. That is not so easy, but that is a phenotype we are following up.