CDK4/6 Inhibitors are Beneficial in HR+ MBC Harboring ESR1 Mutations and Fusions

CDK4/6 inhibition was shown to be effective in patients with hormone receptor-positive metastatic breast cancer (MBC) with ESR1 mutations and fusions, according to findings presented at the 2022 American Association for Cancer Research Annual Meeting.¹

This research was presented by Jamie Brett, fellow at Dana-Farber/Mass General Brigham Hematology/Oncology program, and evaluated patients with MBC who had previously started using a CDK4/6 inhibitor within 30 days of circulating tumor (ct)DNA testing. Patients were analyzed for time-to-next-treatment in order to evaluate CDK4/6 inhibitors in this patient population.

Experts used data from the GuardantInform database of commercial payer claims along with ctDNA tests from over 170,000 individuals. Patients were categorized as either ESR1 mutation or ESR1 wild type status. Patients with ESR1 fusions were detected by tissue RNA-sequencing and taken from a clinicopathologic database at an academic cancer center.

Findings revealed that there was no significant difference in time to next treatment between those with ESR1 mutations vs those with wild type status. Patients with ESR1 mutations demonstrated a shorter overall survival (HR, 0.58; 95% CI, 0.42 - 0.82, P =0.002) than those with wild type status.

Additional studies are expected to expand on the data from the ESR1 fusion cohort and to examine their CDK4/6 inhibitor responses.

In an interview with Targeted Oncology^TM, Jamie Brett and Seth Wander, MD, PhD, Massachusetts General Hospital, discussed recent findings regarding CDK4/6 for patients with hormone receptor-positive MBC with ESR1 mutations and fusions.

Can you explain your research on CDK4/6 inhibitors in hormone receptor-positive MBC with ESR1 mutations?

Brett: The project is on breast cancer that is hormone receptor-positive, and we are focusing on alterations in the estrogen receptor, ESR1 mutations, in 2 classes, ESR1 mutations or point mutations in the ligand binding domain, and then ESR1 fusions that cause endocrine resistance. These alterations are common, and they are important because endocrine therapy is one of the backbones of treatment for hormone receptor-positive breast cancer. Our project is looking at whether CDK4/6 inhibitors can be used to treat patients with these types of ESR1 alterations and there's some data from a retrospective trial analysis that CDK4/6 inhibitors might be able to work for patients with ESR1 mutations.

Then, there's a very small amount of preclinical data that CDK4/6 inhibitors might be able to work for patients with specific types of ESR1 fusions. But there is no data yet, that in the real world setting CDK4/6 inhibitors can work against patients with the ESR1 mutations, infusions, so that's what our work focused on.

For the mutations aspect, we collaborated with GuardantHealth, and used their GuardantInform database to gather real world data about patients with ESR1 mutations and fusions. We used our own MGH database to look for patients with ESR1 fusions. What we found is that in the real-world setting, CDK 4/6 inhibitors can be useful patients with ESR1 mutations. In fact, these patients do very similarly in terms of time to next treatment as patients with ESR1 wild type status. CDK 4/6 inhibitors also seem to be beneficial in each of the cases that we found of patients with ESR1 fusions.

The summary result is that in the real-world setting, CDK 4/6 inhibitors seem to be a useful strategy of therapy for patients with ESR1 alterations that are otherwise very hard to treat because of endocrine resistance. There's a lot of future directions that I think include expanding the ESR1 fusion case series to include a lot more patients. They're looking into the Guardant data in terms of which patients really do well with CDK 4/6 inhibitors and which patients do not do well, and why there's still some differences in survival long term with these patients despite similar outcomes in terms of time to next treatment on CDK 4/6 inhibitors.

Can you explain the similarities and differences between ESR1 mutations and fusions and why they are grouped together?

Brett: We group them together because it's the same gene and they have similar overall effects, but the whole space of ESR1 alterations is heterogeneous. ESR1 mutations are point mutations in the ligand binding domain of the estrogen receptor. They are known to have occurred about 20% to 40% of patients with metastatic breast cancer who have been treated with an aromatase inhibitor in the metastatic setting and do not really occur in other settings. These mutations make the estrogen receptor independent of estrogen and they create complete aromatase inhibition resistance and some relative resistance to other estrogen receptor targeted therapies.

Then there's ESR1 fusions that are less studied overall. They are thought to occur roughly in the range of 1% to 10% of new breast cancer or treated breast cancer, mostly a luminal B type. They are heterogeneous, so they cause different effects. The 2 main categories are that you can swap the ligand binding domain of the estrogen receptor for that of a different protein, and then you create sort of a hyperactive ligand independent estrogen receptor, or you can have the other protein, which is typically CCDC170, become altered by the fusion to the estrogen receptor.

These alterations are thought to cause complete endocrine resistance to aromatase inhibitors SERMs [selective estrogen receptor modulators] inserts. The main theme is that neither of these can receive an aromatase inhibitor. SERMs inserts are sort of controversial for the mutation group and can't be used for the fusion group. Then, targeted therapies are ongoing items in studies for both groups, including the CDK4/6 inhibitors and the most effective, safe, and tolerable treatments that we have and it needs better therapies to get around the mutations and estrogen receptors.

Wander: I would just underline the fact that we are learning a lot more about the molecular and genomic landscape of resistance for patients with metastatic hormone receptor-positive breast cancer. The more we sequence these patients both in the context of clinical trials as well as in the real-world setting, the more we appreciate that there are diverse mechanisms that lead to resistance both to estrogen directed therapies as well as to targeted therapy. As Jamie was just alluding to, in the group of patients who've received prior therapy with an estrogen-directed agent, upwards of 30% or more of these patients can develop alterations in the gene that encodes the estrogen receptor ESR1.

We are appreciating in the clinic how difficult it is to treat these patients. We are starting to learn a lot more about which specific therapeutic strategies are likely to be successful and which are likely to be less successful. I think the work that we presented at AACR highlights the different elements of that discussion. As Jamie was alluding to, one element is exploring the utility of CDK inhibitors in the real-world setting. We have some preliminary insight from prior clinical trials and translational studies.

The second is trying to take a bit of a deeper dive on these ESR1 fusions which are an entirely different class of mutations and can have different activities from the traditional point mutations that we're used to thinking about. I think the work highlights some of the key questions that need further elaboration in this field, both in terms of understanding which therapies can work in these patients as well as developing new strategies to deal with these fusions, which are an entirely different class of of mutations, even though it's related to the overall group.

Can you discuss the findings regarding ESR1 mutations that were discovered through your research?

Brett: We used the GuardantInform database, and we were collaborating with, especially Caroline M. Weipert, MS, Guardant Health, and her team. In that database that now has over 170,000 Guardant360 tests in it, there were 757 patients who had metastatic breast cancer with a Guardant360 test and were starting a CDK4/6 inhibitor within 30 days of that test. In about 20%, those patients, they had ESR1 mutations on that test. We were able to look at real world time to the next treatment as a proxy for progression-free survival, overall survival, and some other clinical factors. The main finding that we had was in the time to next treatment survival curves for patients with ESR1 mutations vs patients with ESR1 wild type disease.

What we found is that these curves are similar in terms of quantitative data and the shape of these curves, and there's no significant difference in terms of time to next treatment survival for mutation patients vs wild type patients. It is 99 days on the median for ESR1 mutation patients and 102 days for the wild type patients. When we adjust for other variables, like which CDK4/6 inhibitor was used, how many prior lines of treatment that the patients get, how old were the patients when they started a CDK4/6 inhibitor, the overall outcome is the same.

We found other interesting things, which is that if we split off the patients who with their CDK4/6 inhibitor got an aromatase inhibitor instead of fulvestrant, these patients did very poorly with about half as long time to next treatment as patients who had a wild type ESR1. This suggests that patients should not get it as an aromatase inhibitor of the ESR1 mutations, which is something that is known but very striking in the data that we have. We found that if we look at overall survival in these patients, we still see that the ESR1 mutations do worse. They have a median real-world overall survival of 2.2 years compared to 5.1 years for ESR1 wild type patients. Again, this difference persists significantly if we adjust for other variables, which is interesting, because there's no difference in terms of the time to next treatment, but there is a strong difference in terms of overall survival. It is known that the patients with ESR1 mutations do worse globally.

Overall, despite adjusting for all the other variables, we can suggest there's something else about other mutations that occur with these patients that are properties of the tumors or are things that we can't measure that give these patients an overall worse prognosis, which is something that we still have to address.

What were some of the conclusions found for ESR1 fusions?

For the ESR1 fusions, we had 4 cases within our database of patients with metastatic breast cancer who had ESR1 fusions. All the patients, because it is a heterogeneous group, had different fusions and none of the fusions had been molecularly characterized before, although some of them have features that allow us to predict function based on prior studies. All of these patients ended up getting a CDK4/6 inhibitor after the fusions were discovered. Then, all of them ended up doing well with that CDK4/6 inhibitor, including a patient who had done very poorly with just fulvestrant, which is very striking. Amongst all the heterogeneity, there is a theme of the CDK4/6 inhibitor is something that all these patients did well with.

Then for the mutation data with the GuardantInform database, we have stratified by different lines of treatment and show that there was no difference in the pattern, that the CDK4/6 inhibitor was effective, regardless of whether it was the first-line treatment, the second line treatment, the third-line treatment, or a fourth- or later-line treatment for that patient. Then, we also looked into different ESR1 mutant alleles, because there's a suggestion that different alleles create different degrees of resistance to SERDs [selective estrogen receptor degraders] and to CDK4/6 inhibitors. We found that in the real-world setting, there is no difference among the different alleles, which is different from some of the trial data and might be due to the heterogeneity of the population.

What unmet needs still have to be filled within this space?

Brett: Starting with the fusion data, which is a very obvious, understudied type of ESR1 alteration, some things that the field is still working on are, what are the fusions doing molecularly? It seems that each new fusion study might be doing something different. There's a group of fusions that don't seem to have any molecular activity, that seem to be non-functional, and either acting because of loss of function of the partner or as a bystander. In particular for our cases, we had different mutations, 5 total fusions, and we did not know what any of them did based on the literature, and so kind of a back to the bench for what ESR1 fusions are doing and if we can actually target these specifically. There's some work regarding what the molecular function of the fusion is predicting, if it creates a hyperactive, constitutively active estrogen receptor based on a transcriptional signature that you can measure in solid tissue samples, and that's something that would be interesting to have.

While this currently doesn't distinguish between fusions and mutations and we know that there are differences in terms of the level of endocrine resistance that occurs, I think there's some work that can be done there. Unmet needs for the fusions are further characterization of the different types of fusions, being able to predict whether the fusions are functional or non-functional and if so, how functional? Then overall, trying to understand better what types of treatments these patients can get. We think CDK4/6 inhibitors based on our work.

There are other targeted treatments out there that might be effective like HER2 targeted therapies, T-inhibitors, based on some preclinical work that need to be tested in patients for the mutations. In terms of unmet needs, it's still not clear in the field overall how to determine which patients benefit from CDK4/6 inhibitors and which patients do not, especially in the later-line setting. Our work shows that patients with ESR1 mutations can get these inhibitors and do well, but we still aren't able to say which patients within these groups make up the heterogeneity that we see within our curves. Mutations preclude the CDK4/6 inhibitor use and which mutations are compatible with it, and what other clinical factors go with that and what other prior lines of treatment go with that. That's something that's being actively worked on, because of how often CDK4/6 inhibitors are used.

Wander: One of the themes that emerges is how important it is to explore some of these questions in the real-world setting. There are advantages to looking at this in more homogeneous clinical trial populations, but they tend to have smaller sample sizes, and they're not always reflective of what we're experiencing in the clinic where there's a lot more heterogeneity in terms of the type of patients that we're treating on the ground. When you think about unmet needs and future directions, you're really thinking about a broad group of questions in the scientific kind of translational research around, and then a group of questions in the clinical practice realm.

To me, some of the top important questions are that as we learn more about the role globally of ESR1 mutations in predicting response or resistance to specific targeted agents and hormonal agents, we need to start to dissect the contribution of individual mutations. Not all of these mutations are acting in the same manner. The larger sample sizes that we can assess via both clinical trials as well as in the real-world setting, the better we're going to be able to start to tease apart some of those differences. That's also true for the fusions. We are still learning a lot about the different fusion partners and how these are impacting response and resistance. Looking at this in the laboratory modeling these different mutations and their impact on drug response, trying to determine what some of the downstream effectors are, between the different mutations, is a really important area of research. In the clinic, learning how to approach these patients and what therapeutic opportunities are likely to be successful, is the most important unmet need and future direction here.

As we have a number of novel selective estrogen receptor degraders coming, we're starting to see phase 3 data and we hope that soon we will be able to start to use some of these drugs in the clinic. To me, one of the most interesting questions will be what is the right pairing between a specific ESR1 mutation and a novel SERD. These are provocative questions that will have to be answered because you are going to have to start to pair up different agents with different mutations. The more we can do to predict or anticipate where the success might be based on some of the molecular modeling in the lab, the better we're going to be able to treat these patients in the clinic.

What other recent research excites you?

Brett: Within breast cancer, all the novels that Seth mentioned are exciting to me, especially because of the different ways that they can be used, like in the adjuvant setting where fulvestrant isn't a great option in terms of mode of administration for patients. Then, there's some preclinical data that I'm interested in.

Wander: We're starting to see more data on some of these novel anti-estrogen agents that are moving through the various clinical pipelines. As it relates to some of our work, it's always interesting to see new emerging insights related to genomic predictors of response and resistance to CDK inhibitors and antiestrogens. In the broader breast cancer field in general, there's lots of other exciting areas that are upcoming, including the use of antibody drug conjugates in areas that we haven't yet had experience using. For example, in hormone receptor-positive disease as well as in HER2-low or negative disease. We are going to learn a lot more about that over the course of the next year, both in patient populations as well as in the translational and preclinical space.

Reference:

CDK4/6 inhibition (CDK4/6i) is effective in the real-world setting for hormone receptor-positive metastatic breast cancer (HR+ MBC) with ESR1 mutations and fusions. Cancer Res 2022;82 (12): 5248. doi:10.1158/1538-7445.AM2022-5248